Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14914
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14720
### Context
`VersionSet::LogAndApply` loads table handlers before the `MANIFEST` update is durable. If that `MANIFEST` update later fails, newly added files can be discarded by `VersionBuilder` without ever being installed in a `Version`. `VersionBuilder::UnrefFile` released the `FileMetaData` `pinned_reader` handle, but releasing that handle only dropped the reference; it did not erase the cache key, so an orphaned table-cache entry could survive for a file that is neither live nor quarantined. When metadata read fault injection prevents the obsolete-file scan from cleaning up the orphan, DB close can trip `TEST_VerifyNoObsoleteFilesCached` with File N is not live nor quarantined.
### Fix
After releasing a pinned_reader for a `FileMetaData` whose refcount reaches zero, explicitly evict that file number from the table cache. The new `VersionBuilderDBTest.FailedLogAndApplyEvictsTableCacheEntry` regression test exercises the realistic production path: it creates a real SST, calls `LogAndApply` through the DB's own `VersionSet`, injects a MANIFEST sync failure via the `AfterSyncManifest` sync point, and asserts that the table cache no longer contains the file after `ProcessManifestWrites` destroys the `VersionBuilder`. With the eviction removed, the test fails deterministically with `leaked_cache_entry=true`.
Reviewed By: hx235
Differential Revision: D99759696
fbshipit-source-id: 0ebb32b1544ad1982d23a6c06569984be97a0a88
Summary:
Fixes three unrelated nightly/CI failures.
Makefile: 'make build_folly' was tripping the build-signature change check, causing the folly-flavored jobs (e.g. nightly clang-21 ASAN+UBSAN+folly, which runs 'make build_folly' then 'USE_FOLLY=1 ... make check') to abort with 'Build parameters changed since the last build'. build_folly only builds the external folly library under third-party/folly and never compiles RocksDB objects into OBJ_DIR, so it should not record or check a build signature; add it to BUILD_SIG_NONBUILD_GOALS alongside checkout_folly.
util/rate_limiter_test.cc: RateLimiterTest.Rate's minimum-rate assertion is timing-sensitive and flakes on the small/loaded ARM CI runners, which (unlike x86_64) can't reliably sustain the target rate. Extend the existing CI exemption with an ARM-only GITHUB_ACTIONS branch, preserving the check on x86_64.
tools/ldb_test.py: the list_live_files_metadata output includes the full DB path, and the unescaped-dot regex '\d+.sst' could match a <digit><char>sst run inside the random tempfile.mkdtemp suffix (drawn from [a-z0-9_]) that precedes the real filename, collapsing testMap to a single bogus key such as '8asst'. Require a literal '.sst'; temp-dir characters can never contain a dot, so only the real filename matches.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14910
Test Plan:
- build_folly: confirmed 'make build_folly' no longer writes .build_signature and that BUILD_SIG_DO_BUILD resolves empty for that goal.
- rate_limiter_test: confirmed via the C preprocessor that x86_64 keeps the SANDCASTLE-only exemption while aarch64 selects the new GITHUB_ACTIONS branch.
- ldb_test: reproduced the collapse with a poisoned temp suffix (produces {'8asst': ['0', 'mycol2']}) and confirmed the escaped-dot regex yields the correct 4-entry map.
Reviewed By: mszeszko-meta
Differential Revision: D110405164
Pulled By: pdillinger
fbshipit-source-id: a9a73b7a717e726fdf79d7795f085f3dbede940e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14907
`RandomAccessFileReader::ReadAsync` took its "already aligned" fast path when the request offset and length were sector-aligned, forwarding the caller's `req` (including its `scratch` pointer) straight to the FileSystem. A null `scratch` trivially satisfied the alignment check (`uintptr_t(nullptr) & (alignment - 1) == 0`), so when a caller submitted an aligned request with `scratch == nullptr` and an `aligned_buf` out-parameter for the reader to allocate into, the null buffer was submitted to the async read. For direct IO with io_uring this becomes a null `iovec` base, which the kernel rejects with `EFAULT` (surfaced as `Req failed: Unknown error -14`).
This is exactly how `IODispatcher` submits MultiScan async reads for plain direct IO (`scratch == nullptr`, `aligned_buf` provided), so the failure manifested intermittently as `db_stress` iterator divergence ("Iterator diverged from control iterator") in crash tests with `use_direct_reads=1 --use_multiscan=1 --multiscan_use_async_io=1`. It is intermittent because it only triggers when the coalesced read's offset and length both happen to be sector-aligned. Task T267030385.
The fix excludes the aligned fast path when the caller provided no `scratch` but did provide an `aligned_buf`, forcing the allocating path so a valid backing buffer is always handed to the async read. This matches the existing behavior of the synchronous `Read`, which already guards its aligned fast path with `scratch != nullptr`.
Reviewed By: xingbowang
Differential Revision: D110361301
fbshipit-source-id: 3884f7ddbc746cf51b2566d74918e2fb97233dcb
Summary:
Experimental embedded ("same-file") blob SST support (https://github.com/facebook/rocksdb/issues/14851) overloads blob file number 0 as both kInvalidBlobFileNumber and the same-file sentinel kCurrentFileBlobIndexFileNumber. Same-file references must be resolved by EmbeddedBlobResolvingIterator before they reach any generic file-metadata or integrated-BlobDB path, all of which reject file number 0 as invalid.
When resolving a same-file blob hit an error (e.g. a db_stress-injected blob-region read fault), EmbeddedBlobResolvingIterator::value() fell back to returning the RAW, unresolved same-file value, surfacing the error only via status()/Valid(). Compaction consumed the raw value before consulting status, leaking an unresolved same-file reference into the output stream. Depending on the value type this produced two different crash-test failures, both from this one root cause:
* T277566778 -- wide-column entity variant. The raw same-file entity (kTypeWideColumnEntity) reached FileMetaData::UpdateBoundaries, whose blob-ref scan (correctly) rejected file number 0: Flush failed: Corruption: Invalid blob file number (also observed from a background compaction). The tripwire fired on the leak, so the real injected read error was masked behind a misleading corruption status.
* T277310719 -- whole-value BlobIndex variant, and more dangerous because it escapes that tripwire. ResolveKeyType() rewrites the key type kTypeBlobIndex -> kTypeValue with no I/O (so it always succeeds); on the masked error the emitted entry is therefore {kTypeValue, raw BlobIndex bytes}. UpdateBoundaries only scans kTypeBlobIndex/entity types, so it does NOT reject this record: the corruption is silently written to the compaction output and persists. A later point lookup reads those raw BlobIndex bytes back as the value, which db_stress detects as db_stress: expected_value.cc:102: Assertion `ExpectedValue::IsValueBaseValid(value_base)' failed from TestGet. In the captured crash the aborting db_stress process had injected no faults itself -- it read corruption persisted by an earlier faults-on run (blackbox reuses the DB directory), confirming the persistence route.
Fix: EmbeddedBlobResolvingIterator now resolves eagerly for callers that do not opt into unprepared values (allow_unprepared_value=false, e.g. compaction), via a new EagerEmbeddedBlobResolvingIterator (the caller is selected in BlockBasedTable::NewIterator). Eager callers resolve the value during positioning, so a resolution error (blob-region I/O or corruption) is observable through status()/Valid() BEFORE value() is consumed, and value() never exposes an unresolved same-file BlobIndex. Lazy callers (allow_unprepared_value=true, user iteration) keep value laziness but must honor PrepareValue()'s result. This keeps the "callers never see an unresolved same-file blob" invariant structural, even on the error path. Using a template for EmbeddedBlobResolvingIterator minimizes unnecessary overheads.
Also hardens and documents the integrity tripwires that caught the leak, so they are not "fixed" by weakening them -- doing so would mask real corruption and could persist unresolvable same-file references into ordinary (non-embedded) SSTs, turning a transient error into permanent data loss: FileMetaData::UpdateBoundaries (write/output path), Version::GetBlob and Version::MultiGetBlob (integrated-BlobDB read path), plus the contract note in blob_constants.h.
Adds a test-only sync point
"BlockBasedTable::MaybeResolveEmbeddedValue:InjectError" (compiled out under NDEBUG) to simulate a blob-region resolution fault deterministically.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14906
Test Plan:
New unit tests in db_blob_index_test, both of which fail without the fix and pass with it:
* DBBlobIndexTest.EmbeddedBlobResolveErrorDuringCompactionNotMasked -- entity variant. Without the fix CompactRange fails with the masked "Invalid blob file number"; with the fix it fails with the injected IOError.
* DBBlobIndexTest.EmbeddedBlobResolveErrorWholeValueDuringCompactionNotMasked -- whole-value variant. Without the fix CompactRange SUCCEEDS (silently masking the error and persisting {kTypeValue, raw BlobIndex}); with the fix it fails with the injected IOError.
Confirmed the tests exercise the real root cause by temporarily neutralizing the eager resolution (MaybeEagerlyMaterialize): both tests then fail in their respective pre-fix modes (entity -> "Invalid blob file number"; whole-value -> compaction succeeds), then pass again once restored.
End-to-end: the db_stress reproducer for T277566778 (blackbox crash test with embedded-blob ingestion and fault injection) reproduced the "Invalid blob file number" crash before the fix and no longer reproduces with it.
Reviewed By: anand1976
Differential Revision: D110361228
Pulled By: pdillinger
fbshipit-source-id: 1faaa9a9ead726f17b611ff150418fa59ed75fd4
Summary:
This PR adds flush observability for debugging write stalls and flush bottlenecks.
Changes include:
- Add flush reason tickers for physical flush jobs:
- rocksdb.flush.reason.write_buffer_full
- rocksdb.flush.reason.write_buffer_manager
- rocksdb.flush.reason.memtable_max_range_deletions
- Add flush memtable size histograms:
- rocksdb.flush.memtable.memory.bytes
- rocksdb.flush.memtable.total.data.size.bytes
- rocksdb.flush.write_buffer_full.memtable.memory.bytes
- rocksdb.flush.write_buffer_manager.memtable.memory.bytes
- Add FlushReason::kMemtableMaxRangeDeletions and propagate it from memtables when memtable_max_range_deletions triggers a flush.
- Preserve atomic flush as a single request-level reason. For atomic flush, the reason is derived only from CFs that scheduled the flush, not every CF selected into the atomic flush cut.
- Add atomic flush request-level counters:
- rocksdb.atomic_flush.request.reason.write_buffer_full
- rocksdb.atomic_flush.request.reason.write_buffer_manager
- rocksdb.atomic_flush.request.reason.memtable_max_range_deletions
- rocksdb.atomic_flush.request.reason.other
- Keep existing per-CF flush job reason counters unchanged. Under atomic flush, rocksdb.flush.reason.* still counts physical CF flush jobs, while rocksdb.atomic_flush.request.reason.* counts one logical atomic flush request.
- Wire the new flush reason, ticker, and histogram types through Java and JNI bindings.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14898
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D110214337
Pulled By: xingbowang
fbshipit-source-id: 4a445c4175d2e09ff5edbe66fde81fa00bb9e8da
Summary:
D110140135/https://github.com/facebook/rocksdb/issues/14896 worked around a db_stress failure by disabling test_backward_scan whenever embedded-blob SSTs are ingested. The real limitation was in EmbeddedBlobResolvingIterator: DBIter requires the underlying iterator's value to be pinned for backward iteration (Prev/SeekForPrev), and rebuilt wide-column entity values (with same-file blob columns) were stored in an iterator-owned std::string that was cleared on every reposition, so IsValuePinned() returned false and DBIter returned NotSupported (or tripped the IsValuePinned() assertion in FindValueForCurrentKeyUsingSeek). Whole-value same-file blobs were already pinned and worked.
This change makes the rebuilt wide-column value pinnable like a whole-value blob: it is moved into a heap buffer and pinned into resolved_pinned_value_, whose cleanup ResetState() hands to the PinnedIteratorsManager across repositioning so the value survives backward iteration. The now-redundant resolved_value_ member is removed and value()/IsValuePinned() simplified. The Get/MultiGet resolution path (MaybeResolveEmbeddedValue) is unchanged. The db_stress test_backward_scan workaround is removed now that backward iteration over embedded blobs (including wide-column entities) is supported.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14902
Test Plan:
Added DBBlobIndexTest.EmbeddedBlobBackwardIteration: ingests an embedded-blob SST with whole-value blobs plus a mixed wide-column entity (inline default column + embedded same-file blob column), then iterates backward via SeekToLast/Prev and SeekForPrev, verifying values and columns. Verified it fails before the fix (NotSupported at the entity) and passes after.
Full db_blob_index_test and sst_file_reader_test suites pass. make format-auto clean. Direct db_stress run with
ingest_external_file_with_embedded_blobs=1, use_put_entity_one_in=3, test_backward_scan=1, iterpercent>0 ran with no divergence/NotSupported/ assert.
Reviewed By: anand1976
Differential Revision: D110271453
Pulled By: pdillinger
fbshipit-source-id: 501a50d3c97bac9a217b604c43c8748abbb563c9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14867
Adds an opt-in --randomize_stress_flags mode to the CPU corruption injection runner.
The fixed DB_STRESS_PRESET keeps an injected corruption isolated and observable, but every run then exercises the same stress test flags (default values for non DB_STRESS_PRESET flags). To widen what a runner covers, we want every other db_stress knob to vary run to run.
Therefore this PR makes each run's db_stress flags are a fresh db_crashtest.py roll (its full blackbox parameter space) with DB_STRESS_PRESET pinned on top. A small CLOSURE gate set stops db_crashtest's finalize_and_sanitize() from un-pinning preset flags, and an assertion fails the run fast if any preset flag is changed. This is considered as a work-around until we can refactor the flag generation of db_crashtest.py into a isolated function to take a set of flags to honor. This PR also adds some randomization into `max_keys` and local vs remote compaction as they are subjected to some special constraint unique to the cpu corruption injection framework.
The per-run roll is seeded by the run's seed, so randomized runs' flags are reproducible with --seed (though the actual db stress test behavior may still have some difference under the same flags).
**Test plan:**
Functional test
- build_runs(..., randomize=True) produces a full randomized flag set (~286 flags vs ~49 for the fixed preset) and the preset survives the db_crashtest roll (the _assert_preset_survived check never fires).
- randomized runs are reproducible: same base_seed -> same flags; run i seeded base_seed+i.
Integration test
```
for op in write flush compaction; do
python3 tools/cpu_corruption_injector/runner.py --op $op --runs 30 --randomize_stress_flags --parallel 8 --stress_cmd "$BIN" --report_dir /tmp/icc-rand/$op
done
```
- every run still carries the pinned DB_STRESS_PRESET values (preset-survived assertion never fires across the campaign).
- the workload genuinely varies run to run (e.g. max_keys, different compaction_style / memtablerep / *_one_in values across runs), confirming randomization is active.
- SDC still appears across write / flush / compaction -- randomization widens coverage without losing the SDC signal; NO_INJECTION and ERROR stay ~0.
**Runs' Outcomes (`summary.json`):**
| SDC | CORRUPTION | CRASH | NO_EFFECT | NO_INJECTION | ERROR |
| --- | --- | --- | --- | --- | --- |
| 2 | 7 | 7 | 43 | 1 | 0 |
Local vs remote compaction. With `--randomize_stress_flags` a compaction run may flip to the remote compaction service (`remote_compaction_worker_threads=1`), so the injector breaks on `CompactionServiceCompactionJob::Run` instead of the local `CompactionJob::Run`. Splitting the runs by injection entry_fn:
| where | runs | SDC | CORRUPTION | CRASH | NO_EFFECT | NO_INJECTION |
| --- | --- | --- | --- | --- | --- | --- |
| local | 25 | 1 | 1 | 3 | 20 | 0 |
| remote | 35 | 1 | 6 | 4 | 23 | 1 |
Reviewed By: pdillinger
Differential Revision: D108551605
fbshipit-source-id: d3dc5687e70ab57779ad6e4f87cac7ed0086da34
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14866
Orchestration layer of the CPU corruption injector -- the glue between detection (https://github.com/facebook/rocksdb/pull/14852) and injection (https://github.com/facebook/rocksdb/pull/14858
).
One CPU-corruption injection says little on its own. What matters is the OUTCOME DISTRIBUTION over many injections -- how often a corruption is silently absorbed (NO_EFFECT), crashes the process (CRASH), is caught by an integrity check (CORRUPTION), or more importantly slips through as a silent data corruption (SDC) and the paths frequently leading to those outcomes. A trustworthy distribution needs a (somewhat) repeatable and reproducible harness as well as a db_stress configuration in which an injected corruption is both reachable and attributable to the chosen stress test op (i.e, write, foreground compaction or flush).
Therefore this PR implements a runner that can launch N independent runs for a chosen op type (i.e, write, foreground compaction or flush). Each run picks where to inject, runs db_stress under gdb via the `injector,py` (https://github.com/facebook/rocksdb/pull/14858), and is classified into one outcome bucket (https://github.com/facebook/rocksdb/pull/14852).
The runner has DB_STRESS_PRESET -- the pinned db_stress config that isolates a single injected corruption (single-threaded, integrity checks on, other fault injection off, auto-compaction off). The runner also does gdb preflight that fails fast when the build or gdb cannot support injection because, for example, the hard-coded `target_fn` has changed its name, provides parallel launching of many runs and one summary.json per campaign (the outcome distribution plus each run's record). The whole run set is reproducible from one logged base_seed (run i uses base_seed + i).
**Test plan:**
Build: `make DEBUG_LEVEL=0 EXTRA_CXXFLAGS="-g -fno-inline" db_stress`
# 1. Preflight (`verify_injection_site`) catches a build that can't support injection
Before doing any work, the runner has gdb confirm — on this exact binary — that every injection-site function resolves and gdb can read its source line. A good build logs:
```
INFO gdb check OK for op=compaction: rocksdb::CompactionJob::Run, rocksdb::CompactionIterator::NextFromInput, rocksdb::BlockBuilder::Add
```
A build that cannot support injection (functions renamed, fully inlined, or absent) fails fast with exit 2 before any run — forced here by pointing `--stress_cmd` at a non-db_stress binary:
```
$ python3 tools/cpu_corruption_injector/runner.py --op compaction --runs 1 --stress_cmd /bin/ls --report_dir /tmp/icc/preflight_demo
ERROR gdb could not set a breakpoint on these functions in ls (renamed, fully inlined, or not in this build?): rocksdb::CompactionJob::Run, rocksdb::CompactionIterator::NextFromInput, rocksdb::BlockBuilder::Add
Function "rocksdb::CompactionJob::Run" not defined.
Function "rocksdb::CompactionIterator::NextFromInput" not defined.
# exit code 2
```
So a broken/inlined build is rejected up front instead of silently producing `NO_INJECTION` runs.
# 2. Compaction op -- 100 runs
```
$ python3 tools/cpu_corruption_injector/runner.py --op compaction --runs 100 --stress_cmd /data/users/huixiao/rocksdb/db_stress --report_dir /tmp/icc/preflight_demo
```
**Runs' outcomes (`summary.json`):**
| SDC | CORRUPTION | CRASH | NO_EFFECT | NO_INJECTION | ERROR |
| --- | --- | --- | --- | --- | --- |
| 9 | 5 | 6 | 79 | 1 | 0 |
**Spread:**
- target_fn x outcome: `NextFromInput` {SDC 9, CORRUPTION 3, CRASH 2, NO_EFFECT 34, NO_INJECTION 1}; `BlockBuilder::Add` {CORRUPTION 2, CRASH 4, NO_EFFECT 45}.
- corruption_type x outcome: `bit_flip` {SDC 8, CORRUPTION 3, CRASH 6, NO_EFFECT 32}; `flag_flip` {SDC 1, CORRUPTION 2, NO_EFFECT 42}; `lane_bit_flip` {NO_EFFECT 5}.
**Analysis:** all 9 SDCs land on the read/iterate path (`NextFromInput`); corrupting the output writer (`BlockBuilder::Add`) never produced an SDC (its blocks are checksummed -- corruption there is caught or inert). The 5 detected `CORRUPTION`s are compaction's key-order and record-count cross-checks firing (both CompactRange and CompactFiles origins appear), correctly bucketed by the fix.
### A representative compaction SDC: `run_00000`
What we corrupted (`inject.json`):
```json
{"op":"compaction","op_index":17,"entry_fn":"rocksdb::CompactionJob::Run","target_fn":"rocksdb::CompactionIterator::NextFromInput","injection_result":"injected","db_stress_crash_signal":null,
"corruptions":[{"instruction":"mov %rdx,0x160(%r12)","register":"rdx","corruption_type":"bit_flip","before":"0x10","after":"0x18",
"details":{"source":"rocksdb::CompactionIterator::NextFromInput @ db/compaction/compaction_iterator.cc:719"}}]}
```
The recorded silent corruption (`data_corruption.<tid>.json`):
```json
{"kind":"wrong-value","cf":0,"key":70,"value_from_db":"010000000504070609080B0A0D0C0F0E070F105E78787878","value_from_expected":"010000000504070609080B0A0D0C0F0E","op_status":"Get: OK"}
```
**Walkthrough:** a single bit flip on `rdx` (`0x10 -> 0x18`, 16 -> 24) at `CompactionIterator::NextFromInput` (`compaction_iterator.cc:719`) -- `rdx` holds the value LENGTH stored into the iterator's record (offset `0x160`). The length reads 24 instead of 16, so compaction copies a value 8 bytes too long into the output SST, absorbing adjacent bytes. The internal key is untouched (`ParseInternalKey` passes); the over-long value is the single Slice fed to both the paranoid validator and the SST builder, so the file is self-consistent and every checksum agrees. On read-back `Get(key=70)` returns OK with the wrong bytes -- `value_from_db` is the expected value (`...0F0E`) **plus 8 trailing bytes** (`070F105E78787878`). Silent: read OK, all checks pass, the value visibly grew. `classify()` routes `kind=wrong-value` to the SDC bucket.
### A representative compaction CORRUPTION (detected): `run_00007`
What we corrupted (`inject.json`):
```json
{"op":"compaction","op_index":41,"entry_fn":"rocksdb::CompactionJob::Run","target_fn":"rocksdb::CompactionIterator::NextFromInput","injection_result":"injected","db_stress_crash_signal":"SIGABRT",
"corruptions":[{"instruction":"mov (%rbx),%rdi","register":"rdi","corruption_type":"bit_flip","before":"0x7fffeee8c1c0","after":"0x7fffeee8c1c2",
"details":{"source":"rocksdb::IterKey::SetKeyImpl @ ./db/dbformat.h:941","call_chain":["rocksdb::IterKey::SetKeyImpl @ ./db/dbformat.h:941","rocksdb::CompactionIterator::NextFromInput @ db/compaction/compaction_iterator.cc:781"]}}]}
```
The recorded detection (`data_corruption.<tid>.json`):
```json
{"kind":"detected-corruption","cf":-1,"key":-1,"value_from_db":"","value_from_expected":"","op_status":"compactfiles: Corruption: Compaction sees out-of-order keys."}
```
**Walkthrough:** a bit flip on `rdi` (`...c1c0 -> ...c1c2`, a key pointer) at `IterKey::SetKeyImpl` (`dbformat.h:941`), reached from `NextFromInput`, mis-sets the iterator's key so the next emitted key is out of order. Compaction's key-order check catches it and returns `compactfiles: Corruption: Compaction sees out-of-order keys`. The op then takes `SIGABRT`, but `classify()` reads the recorded `data_corruption` result before the crash signal, so the run is correctly bucketed `CORRUPTION` (the bucketization fix; pre-fix this surfaced as `CRASH`). `classify()` routes `kind=detected-corruption` to the `CORRUPTION` bucket.
# 3. Flush op -- 100 runs
```
$ python3 tools/cpu_corruption_injector/runner.py --op flush --runs 100 --stress_cmd /data/users/huixiao/rocksdb/db_stress --report_dir /tmp/icc/preflight_demo
```
**Runs' outcomes (`summary.json`):**
| SDC | CORRUPTION | CRASH | NO_EFFECT | NO_INJECTION | ERROR |
| --- | --- | --- | --- | --- | --- |
| 2 | 12 | 11 | 74 | 1 | 0 |
**Spread:**
- target_fn x outcome: `BlockBuilder::Add` {SDC 1, CORRUPTION 5, CRASH 5, NO_EFFECT 49}; `NextFromInput` {SDC 1, CORRUPTION 7, CRASH 6, NO_EFFECT 25, NO_INJECTION 1}.
- corruption_type x outcome: `bit_flip` {SDC 2, CORRUPTION 9, CRASH 9, NO_EFFECT 32}; `flag_flip` {CORRUPTION 3, CRASH 2, NO_EFFECT 42}.
**Analysis:** flush mirrors compaction's mechanisms (shared iterator/builder). The 2 SDCs are a value/key-pointer corruption that slips past the checksums; the 12 corruptions are caught by the flush-time key-order / key-size integrity checks.
### A representative flush SDC: `run_00027`
What we corrupted (`inject.json`):
```json
{"op":"flush","op_index":16,"entry_fn":"rocksdb::FlushJob::Run","target_fn":"rocksdb::BlockBuilder::Add","injection_result":"injected","db_stress_crash_signal":null,
"corruptions":[{"instruction":"mov (%rdi),%rax","register":"rax","corruption_type":"bit_flip","before":"0x7fffef059400","after":"0x7fffef059440",
"details":{"source":"rocksdb::Slice::data @ ./include/rocksdb/slice.h:58","call_chain":["rocksdb::Slice::data @ ./include/rocksdb/slice.h:58","rocksdb::BlockBuilder::AddWithLastKeyImpl @ table/block_based/block_builder.cc:351"]}}]}
```
The recorded silent corruption (`data_corruption.<tid>.json`):
```json
{"kind":"lost","cf":0,"key":763,"value_from_db":"","value_from_expected":"010000000504070609080B0A0D0C0F0E","op_status":"Get: NotFound"}
```
**Walkthrough:** a bit flip on `rax` (a key/value data pointer, `...9400 -> ...9440`) at `Slice::data` (`slice.h:58`), reached from `BlockBuilder::AddWithLastKeyImpl` while the flush builds the output block, makes the builder read key bytes from the wrong address, so key 763's entry is written wrong and the key is dropped from the flushed SST. On read-back `Get(key=763)` returns `NotFound` for a committed key -- silent. `classify()` routes `kind=lost` to the SDC bucket.
### A representative flush CORRUPTION (detected): `run_00047`
What we corrupted (`inject.json`):
```json
{"op":"flush","op_index":7,"entry_fn":"rocksdb::FlushJob::Run","target_fn":"rocksdb::CompactionIterator::NextFromInput","injection_result":"injected","db_stress_crash_signal":"SIGABRT",
"corruptions":[{"instruction":"cmp $0x7,%rax","register":"eflags","corruption_type":"flag_flip","before":"0x216","after":"0x256",
"details":{"source":"rocksdb::ParseInternalKey @ ./db/dbformat.h:523","call_chain":["rocksdb::ParseInternalKey @ ./db/dbformat.h:523","rocksdb::CompactionIterator::NextFromInput @ db/compaction/compaction_iterator.cc:731"]}}]}
```
The recorded detection (`data_corruption.<tid>.json`):
```json
{"kind":"detected-corruption","cf":-1,"key":-1,"value_from_db":"","value_from_expected":"","op_status":"flush: Corruption: Corrupted Key: Internal Key too small. Size=16. "}
```
**Walkthrough:** a flag flip (`eflags 0x216 -> 0x256`) on a `cmp $0x7,%rax` branch in `ParseInternalKey` (`dbformat.h:523`), reached from `NextFromInput`, makes the parser mis-judge the internal-key size, so the flush emits a malformed key and the key-size integrity check returns `flush: Corruption: Corrupted Key: Internal Key too small`. The op takes `SIGABRT`; `classify()` reads the recorded `data_corruption` before the signal and buckets `CORRUPTION` (bucketization fix). `classify()` routes `kind=detected-corruption` to the `CORRUPTION` bucket.
# 4. Write op (`MemTable::Add`) -- two key spaces
```
$ python3 tools/cpu_corruption_injector/runner.py --op write --runs 100 --stress_cmd /data/users/huixiao/rocksdb/db_stress --report_dir /tmp/icc/preflight_demo
```
A write injection corrupts a single `MemTable::Add` (a Put/Delete/DeleteRange). The corruption is reachable and attributable, but whether it surfaces as a *silent* write SDC depends heavily on the key space. A silent write SDC needs the affected/mispositioned key to have other live versions to fall through to -- which only happens in a dense, multi-version memtable. We therefore run two write campaigns: the default `max_key=1000`, then a small `max_key=8`. The contrast is what motivates randomizing `max_key` (see PR https://github.com/facebook/rocksdb/pull/14867 for `--randomize_stress_flags`).
### 4a. Default `max_key=1000` -- 100 runs (no silent write SDC)
**Runs' outcomes (`summary.json`):**
| SDC | CORRUPTION | CRASH | NO_EFFECT | NO_INJECTION | ERROR |
| --- | --- | --- | --- | --- | --- |
| 0 | 31 | 13 | 56 | 0 | 0 |
With a 1000-key space almost every write touches a distinct key, so a corrupted entry has no older live version to mask it: a value/key byte flip is caught at write by the per-key checksum (`VerifyEncodedEntry` -> `CORRUPTION`), and a structural flip tends to crash (`CRASH`) rather than silently mis-read. `ERROR=0`, `NO_INJECTION=0`. No write op silently corrupted data -- every reachable corruption was caught or crashed.
### 4b. Small `max_key=8` -- 100 runs (surfaces 2 silent write SDCs)
**Runs' outcomes (`summary.json`):**
| SDC | CORRUPTION | CRASH | NO_EFFECT | NO_INJECTION | ERROR |
| --- | --- | --- | --- | --- | --- |
| 2 | 33 | 8 | 57 | 0 | 0 |
Shrinking the key space makes each key hold ~125 versions (`ops_per_thread` / `max_key`), so a misplaced entry can fall through to an older version of the *same* key and be returned silently -- the per-key checksum (bytes intact) and on-seek verify cannot see a pure link-position error.
### A representative write SDC: `run_00028` (skiplist misposition -> silent stale read, flush catches)
What we corrupted (`inject.json`):
```json
{"op":"write","op_index":317,"entry_fn":"rocksdb::MemTable::Add","target_fn":"rocksdb::MemTable::Add","injection_result":"injected","db_stress_crash_signal":null,
"corruptions":[{"instruction":"cmp %rbx,-0xb8(%rbp)","register":"eflags","corruption_type":"flag_flip","before":"0x216","after":"0x217",
"details":{"source":"rocksdb::MemTable::Add @ db/memtable.cc:1319",
"call_chain":["rocksdb::MemTable::Add @ db/memtable.cc:1319"]}}]}
```
The recorded silent corruption (`data_corruption.<tid>.json`):
```json
{"kind":"resurrected","cf":0,"key":1,"value_from_db":"110000001514171619181B1A1D1C1F1E0100030205040706","value_from_expected":"","op_status":"Get: OK"}
```
**Walkthrough:** a flag flip (CF, `eflags 0x216 -> 0x217`) on the `cmp` that produces `KeyIsAfterNode` inside `InlineSkipList::Insert` (`inlineskiplist.h:1253`; the `@ memtable.cc:1319` in the record is inlining line-drift) inverts the key comparison, so the Delete tombstone for key 1 is linked at the wrong position. The stored bytes and per-key checksum are intact, so neither the checksum nor on-seek verify sees anything wrong -- on read-back `Get(key=1)` returns OK with key 1's live value for a key that was Deleted (`kind=resurrected`, silent). A follow-up `Flush()` in unit test repro *does* catch it: the full-scan order check returns `Corruption: Out-of-order keys found in skiplist` -- caught only after the silent read, not during it.
### A representative write CORRUPTION (detected) `max_key=1000 or 8` : `run_00018`
Where `run_00028`'s pure link-position error is invisible to the per-key checksum, this run shows a byte-level corruption that the checksum *catches* at write time. What we corrupted (`inject.json`):
```json
{"op":"write","op_index":106,"entry_fn":"rocksdb::MemTable::Add","target_fn":"rocksdb::MemTable::Add","injection_result":"injected","db_stress_crash_signal":null,
"corruptions":[{"instruction":"mov %rsi,(%rdi)","register":"rsi","corruption_type":"bit_flip","before":"0x7fffeec2a21c","after":"0x7fffeec2a25c",
"details":{"source":"rocksdb::Slice::Slice @ ./include/rocksdb/slice.h:39",
"call_chain":["rocksdb::Slice::Slice @ ./include/rocksdb/slice.h:39","rocksdb::GetVarint32 @ ./util/coding.h:280","rocksdb::MemTable::VerifyEncodedEntry @ db/memtable.cc:1102","rocksdb::MemTable::Add @ db/memtable.cc:1189"]}}]}
```
The recorded detection (`data_corruption.<tid>.json`):
```json
{"kind":"detected-corruption","cf":-1,"key":-1,"value_from_db":"","value_from_expected":"","op_status":"put: Corruption: ProtectionInfo mismatch"}
```
**Walkthrough:** a bit flip on `rsi` (`0x7fffeec2a21c -> 0x7fffeec2a25c`) at `Slice::Slice` (`slice.h:39`) while `MemTable::Add` re-parses the just-encoded entry through `VerifyEncodedEntry` (`memtable.cc:1102`) corrupts the Slice the verifier reads, so the recomputed per-key protection info no longer matches and the put returns `Corruption: ProtectionInfo mismatch`.
Reviewed By: pdillinger
Differential Revision: D108367345
fbshipit-source-id: 250b069c8e9d63a2dc257f6de86f1d9040284dd6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14858
This PR is the injection layer of the CPU corruption injector, runs inside gdb and randomly corrupts a register by bit flip in exactly one db_stress op (i.e, write, foreground compaction and flush) per stress test run. Detection layer is at db_stress (https://github.com/facebook/rocksdb/pull/14852); orchestration layer is coming up.
__How one run works__
- The orchestration layer, coming up, randomly picks which stress test `op` instance (so corruption can land at different points in the LSM shape journey) and which `target_fn` of that `op` (so to cap instructions to step under a reasonable limit; `injector.py` in this PR randomly picks which instruction within the `target_fn` to inject (so corruption can land at different points of a `target_fn`).
- Attach: gdb starts with injector.py's parameters passed via -iex and the db_stress command after --args, so db_stress runs unmodified. Example:
```
gdb --batch --nx \
-iex "py import sys; sys.argv=['injector.py','--op','write','--op_index','42','--entry_fn','rocksdb::MemTable::Add','--target_fn','rocksdb::MemTable::Add','--corruptions_per_op','1','--seed','7','--dir','<rundir>']" \
-x tools/cpu_corruption_injector/injector.py \
--args <db_stress> --threads=1 --verify_cpu_corruption_dir=<rundir> ...
```
- Navigate: The orchestration layer will pick op_index. `entry_fn` is called exactly once per stress test run's op so the op_index-th op is its op_index-th call. `injector_navigate.py` breaks on `entry_fn` and set a gdb ignore-count of op_index-1 to fast-forward to op_index-th one. It also breaks at the first `target_fn` within that `entry_fn`.
- Warm up: `injector_critical_instruction.py` will choose "critical instruction" (those that move key/value bytes with general-purpose or vector registers or set a branch flag) uniformly within the chosen `target_fn` by the orchestration layer. In order to do that, it needs to approximate how many such instructions within the `target_fn`. Hence we have this warm-up phase. It single-steps the instruction within the first encoutering of `target_fn` to count and draw the critical instruction index, then corrupt that index at a later call.
- Corrupt: on a later call of `target_fn`, `injector_critical_instruction.py` single-step to the m-th critical instruction and bit-flip the register through `injector_register_corruption.py`. The way to corrupt register depends on what instruction it is. If the current call of `target_fn`'s m-th instruction is not a critical instruction, we will try next `target_fn` till running out of `target_fn`.
- Record: `injector_telemetry.py` provides telemetry to capture the corruption for later analysis.
**Test plan:**
1. Isolated tests (real gdb-captured x/i fixtures): test_inject_critical_instruction
2. E2E test on navigation, inject, telemetry will be done in the later orchestration PR. Below is inject.json from such run
```
{
"injection_result": "injected",
"db_stress_crash_signal": null,
"op": "write",
"op_index": 279,
"entry_fn": "rocksdb::MemTable::Add",
"target_fn": "rocksdb::MemTable::Add",
"critical_instruction_index": 37,
"corruptions": [
{
"instruction": "mov %rsi,0x8c8(%rbx)",
"register": "rsi",
"corruption_type": "bit_flip",
"before": "0x7fffee4c64d8",
"after": "0x7fffee4c64c8",
"details": {
"source": "rocksdb::Arena::AllocateAligned @ ./fbcode/internal_repo_rocksdb/repo/memory/arena.cc:135",
"call_chain": [
"rocksdb::Arena::AllocateAligned @ ./fbcode/internal_repo_rocksdb/repo/memory/arena.cc:135",
"rocksdb::ConcurrentArena::AllocateAligned(unsigned long, unsigned long, rocksdb::Logger*)::{lambda()#1}::operator()() const @ fbcode/internal_repo_rocksdb/repo/memory/concurrent_arena.h:65",
"rocksdb::ConcurrentArena::AllocateImpl<rocksdb::ConcurrentArena::AllocateAligned(unsigned long, unsigned long, rocksdb::Logger*)::{lambda()#1}>(unsigned long, bool, rocksdb::ConcurrentArena::AllocateAligned(unsigned long, unsigned long, rocksdb::Logger*)::{lambda()#1} const&) @ fbcode/internal_repo_rocksdb/repo/memory/concurrent_arena.h:145",
"rocksdb::ConcurrentArena::AllocateAligned @ fbcode/internal_repo_rocksdb/repo/memory/concurrent_arena.h:63",
"rocksdb::InlineSkipList<rocksdb::MemTableRep::KeyComparator const&>::AllocateNode @ fbcode/internal_repo_rocksdb/repo/memtable/inlineskiplist.h:868",
"rocksdb::InlineSkipList<rocksdb::MemTableRep::KeyComparator const&>::AllocateKey @ fbcode/internal_repo_rocksdb/repo/memtable/inlineskiplist.h:855",
"rocksdb::(anonymous namespace)::SkipListRep::Allocate @ ./fbcode/internal_repo_rocksdb/repo/memtable/skiplistrep.cc:36",
"rocksdb::MemTable::Add @ ./fbcode/internal_repo_rocksdb/repo/db/memtable.cc:1157"
]
}
}
],
"ops_seen": 279,
"critical_instructions_seen": 38
}
```
Reviewed By: pdillinger
Differential Revision: D107999835
fbshipit-source-id: 0863715e5d596d69abff8b8e4d9a252fa8c24b6c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14852
Detection layer of the CPU corruption injector (https://github.com/facebook/rocksdb/pull/14858). With `--verify_cpu_corruption_dir=<dir>`, db_stress reads back the full keyspace after every write/manual flush/manual compaction op and compares it to the expected-values model, classifying any mismatch by `kind`: `lost` / `resurrected` / `wrong-value` (silent data corruption) or `detected-corruption` (a status/checksum-caught error). Each finding is written to `<dir>/data_corruption.<tid>.json` ({kind, cf, key, value_from_db, value_from_expected, op_status}) and routed through db_stress's standard `VerificationAbort` for a clean exit-1. A startup guard requires `--threads=1` and all fault injection off so the read-back is single-writer and the only corruption present is the injected one
Bonus: a minor refactoring into the surrounding error handling code in these ops
**Test plan:**
1.Startup guard rejects misconfiguration:
```
--threads=2 -> exit 1: "--verify_cpu_corruption_dir requires --threads=1"
--read_fault_one_in=5 -> exit 1: "requires all fault injection off"
```
2.No false positive (clean CORE preset run, no injection):
```
$ db_stress --verify_cpu_corruption_dir=<dir> --threads=1 (full protections, all *_fault_one_in=0) ...
exit 0; no data_corruption.<tid>.json produced; "Verification successful"
```
3.Write-path cpu corruption injection (coming up, e.g, gdb flips a register inside MemTable::Add), then the immediate post-op read-back catches it. Real `<dir>/data_corruption.<tid>.json`:
silent data corruption -- write returned OK but the key is gone on read-back:
```
{"kind":"lost","cf":0,"key":9814,"value_from_db":"","value_from_expected":"010000000504070609080B0A0D0C0F0E","op_status":"Get: NotFound"}
```
detected corruption -- read-back Get returns Corruption via the memtable per-key checksum:
```
{"kind":"detected-corruption","cf":0,"key":139,"value_from_db":"","value_from_expected":"","op_status":"Get: Corruption: Corrupted memtable entry, per key-value checksum verification failed."
```
4.See PR https://github.com/facebook/rocksdb/pull/14866 test plan's spread in the outcome for verification of detection
Reviewed By: pdillinger
Differential Revision: D107999834
fbshipit-source-id: 18decbf51dc56eec62b735136e2dfb4e1175b773
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14896
Replace the earlier mitigation that disabled test_backward_scan when
separate_key_value_in_data_block is set (D109837793). That gate was incorrect:
separate_key_value_in_data_block does not affect value pinning -- it was only
coincidentally correlated with the failing crash-test runs (the crash test
biases it on ~2/3 of the time).
Root cause: embedded blob SSTs are read through EmbeddedBlobResolvingIterator,
which resolves same-file blob references (and wide-column entities with same-file
blob columns) into an iterator-owned buffer. Those resolved values are not
pinnable, but DBIter requires pinned values for backward iteration
(Prev/SeekForPrev), returning NotSupported (surfaced as "Iterator diverged from
control iterator") or hitting the IsValuePinned() assertion in
FindValueForCurrentKeyUsingSeek (db_iter.cc).
Gate test_backward_scan on ingest_external_file_with_embedded_blobs instead.
This is a stress-test mitigation only; the underlying limitation (backward
iteration over unpinnable resolved values) should be fixed separately by making
the resolved value pinnable via the PinnedIteratorsManager.
___
Differential Revision: D110140135
fbshipit-source-id: 5288dab67ab753cbba22806e09fb8fec1e6f5a01
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14889
When `separate_key_value_in_data_block=true`, the data block iterator stores keys and values in separate sections. Under certain conditions during backward iteration (Prev/SeekForPrev), `DBIter::FindValueForCurrentKey` encounters entries whose values cannot be pinned, causing a `NotSupported` error: "Backward iteration not supported if underlying iterator's value cannot be pinned." This manifests in the crash test as `TestIterateAgainstExpected` failures when the random flag combination includes `separate_key_value_in_data_block=1` (added in 47cfeb656ff3) with `test_backward_scan=1` (the default). The fix automatically disables `test_backward_scan` when `separate_key_value_in_data_block` is enabled, matching the pattern used for other option incompatibilities in the stress tool.
Reviewed By: xingbowang
Differential Revision: D109837793
fbshipit-source-id: 04230e6c0af42e98fb82d0ddcc3d857419e879e3
Summary:
Add rate limiter statistics that make it easier to diagnose whether RocksDB background IO is being throttled by the GenericRateLimiter.
New ticker counters:
- rocksdb.rate.limiter.bytes.read/write: bytes granted by the limiter, split by operation type.
- rocksdb.rate.limiter.requests.read/write: requests granted by the limiter, split by operation type.
- rocksdb.rate.limiter.delayed.requests.read/write: requests that actually had to wait for a future refill because available tokens were exhausted.
- rocksdb.rate.limiter.total.wait.micros.read/write: cumulative wait time for delayed requests.
New histograms:
- rocksdb.rate.limiter.wait.micros.read/write: per-request wait latency for delayed requests, enabling percentile debugging.
GenericRateLimiter now overrides the OpType-aware Request() path so these stats preserve the configured limiter mode. The legacy three-argument Request() path is treated as write IO for compatibility. Existing NUMBER_RATE_LIMITER_DRAINS is left intact, but the new delayed request counters are the clearer signal for out-of-token throttling because one delayed request can observe multiple drain intervals.
Also update the Java ticker and histogram mappings so Java consumers can access the new statistics.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14884
Test Plan:
- make format-auto
- make -j16 rate_limiter_test statistics_test
- ./rate_limiter_test
- ./statistics_test
- make check-sources
Reviewed By: anand1976
Differential Revision: D109694030
Pulled By: xingbowang
fbshipit-source-id: e92756788ba629f25655958c7beb91099184cda4
Summary:
Local Make builds silently reuse object files even when build parameters (DEBUG_LEVEL, sanitizers, ASSERT_STATUS_CHECKED, RTTI, etc.) change, since all object files share the same paths. This leads to confusing linker errors, sanitizer false negatives, ODR violations, and "phantom" bugs that are easy to miss -- a pitfall for humans and especially for AI agents driving builds non-interactively.
This change improves the Make experience on three fronts:
1. Build-parameter change detection in the Makefile. After flags are fully resolved, we hash the effective compile/link inputs (CC|CXX|CFLAGS|CXXFLAGS|LDFLAGS|EXEC_LDFLAGS) into a per-OBJ_DIR .build_signature stamp and compare against the previous build:
- Default: stop with a clear error when parameters changed. This is optimized for the expert who might not intend a full rebuild, and for avoiding CI inefficiency.
- AUTO_CLEAN=1: automatically run clean-rocks and proceed.
- ALLOW_BUILD_PARAMETER_CHANGE=1: skip the check (e.g. intentionally mixing DEBUG_LEVEL=1 and DEBUG_LEVEL=2 objects).
The check is skipped for dry runs (-n) and for non-building / self-cleaning targets (clean*, format, check-*, tags, gen-pc, check-progress, watch-log, release, coverage, the asan_*/ubsan_* variants, etc.). make_config.mk is removed only by the top-level `clean` target, not by clean-rocks, so the auto-clean does not delete the make_config.mk already included by the running make. The signature is removed last in clean-rocks so an interrupted clean keeps change detection meaningful.
3. New build_tools/rockstest.sh and build_tools/rocksptest.sh helpers that build and run unit tests in one step. They set AUTO_CLEAN=1 and build with -j<NCORES> (computed like the Makefile). rocksptest.sh runs one or more binaries under the checked-in gtest-parallel (sharing one parallel worker pool); rockstest.sh runs a single binary directly for a small number of cases. Both reject a leading-option first argument; `rockstest.sh install` writes ~/bin/rockstest and ~/bin/rocksptest shims that defer to the in-tree scripts (with a friendly message when run outside a source root). This is a big win for AI-agent workflows: one command instead of a separate `make` then test-run invocation, removing the overhead of monitoring two long commands, while avoiding serial builds and serial test execution.
4. CLAUDE.md guidance updated to recommend AUTO_CLEAN=1 for manual make invocations and the rocks(p)test.sh helpers for running tests
Although there might be drive to consolidate around the BUCK build for internal use, I'll note that last I checked it was around ~20 minutes to run all the unit tests under buck and ~2 minutes under `make check`.
Bonus: the first revision of this had copyright/license mistakes, and this change corrects similar errors elsewhere, with CLAUDE.md advice updated.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14883
Test Plan:
Manual verification of the Makefile change detection:
- Confirmed the resolved-flags hash is deterministic across runs for the same goal, and differs between configs (e.g. default/shared vs static_lib, and with/without ASSERT_STATUS_CHECKED).
- Unchanged parameters: rebuild does not error.
- Changed parameters: errors by default; AUTO_CLEAN=1 runs clean-rocks and proceeds; ALLOW_BUILD_PARAMETER_CHANGE=1 bypasses the check.
- Dry runs (make -n) and exempt targets (e.g. `make check-progress`, list_all_tests, gen-pc) do not trip the check and leave the stamp untouched.
- Reproduced the ASSERT_STATUS_CHECKED=1 auto-clean path end to end and confirmed it no longer fails with "No rule to make target 'make_config.mk'"; verified `make clean` removes make_config.mk while `clean-rocks` does not, and that the signature is deleted as the final clean-rocks step.
Manual verification of the helper scripts:
- shellcheck-clean and `bash -n` pass for both scripts.
- Usage/guard behavior: missing argument and leading-option first argument are rejected with a clear message.
- `rockstest.sh install` (validated against a temp HOME) generates both shims; running a shim outside a source root prints "(Not in a rocksdb source root directory?)" and exits non-zero.
- rocksptest.sh argument splitting verified for single binary, multiple binaries, and binaries followed by gtest-parallel/test args.
- .gitignore patterns verified with `git check-ignore` for .build_signature (root and jl/jls) and build_tools/__pycache__/.
Reviewed By: xingbowang
Differential Revision: D109713512
Pulled By: pdillinger
fbshipit-source-id: a801d29afa91f53a2dce717db9fe5e5d485cc154
Summary:
### Embedded Blob Resolving Iterator Fix
#### Use-After-Free Bug Fix
This diff fixes a use-after-free bug in the `EmbeddedBlobResolvingIterator` class. The bug occurred when the `key()` method was called before the `value()` method, causing the `resolved_internal_key_` buffer to be freed prematurely.
#### Changes
The fix involves modifying the `EmbeddedBlobResolvingIterator` class to ensure that the `resolved_key` is not moved into `resolved_internal_key_` if it has already been resolved (`key_resolved_` flag is set to `true`). This prevents the `resolved_internal_key_` buffer from being overwritten and freed when `MaterializeValue()` is called.
Reviewed By: xingbowang
Differential Revision: D109796428
fbshipit-source-id: d89fb36428ba62e8d0be53c2acfa3cdb9bda76f4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14887
This reverts two changes to the range-lock endpoint comparator and restores the prior, direction-agnostic behavior:
- https://github.com/facebook/rocksdb/pull/14831 ("Fix range lock manager crash with reverse comparator"), which made RangeTreeLockManager::CompareDbtEndpoints reverse-aware, and
- https://github.com/facebook/rocksdb/pull/14880 (a follow-up that only corrected the equal-length / point-range branch).
Why: #14831 taught the endpoint comparator about reverse-ordered column families by flipping the suffix tie-break based on comparator direction. However, callers that use range locking on a reverse-ordered CF already flip the start/end endpoints themselves before calling lock_range() -- for example, MyRocks does this in ha_rocksdb::set_range_lock() (see its "RangeFlagsShouldBeFlippedForRevCF" comment). The two corrections stack and produce left > right again, which:
(a) trips paranoid_invariant(compare(left, right) <= 0) in toku::locktree::try_acquire_lock() and aborts on equality/point locks, and
(b) for range scans, mis-orders the endpoints and produces an incorrect lock range, leading to silent correctness issues under concurrency.
#14880 only addressed (a) (the equal-length / point-range branch), not (b), so it was an incomplete fix at the wrong layer.
Decision: keep CompareDbtEndpoints direction-agnostic and make it an explicit caller requirement to pass a flipped (swapped start/end, with negated infimum/supremum flags) range for range locking to work on a reverse-ordered column family. A function-level comment documenting this contract is added so reverse-awareness is not reintroduced here.
Removes the RangeLockWithReverseComparator unit test added by #14831 (it asserts the reverted direction-flipped ordering). Keeps a regression test that a point range lock [infimum(K), supremum(K)] works on a reverse-comparator column family.
Reviewed By: xingbowang
Differential Revision: D109713989
fbshipit-source-id: 4ea0999c92645e1adc9cd1114d924cad142f2001
Summary:
Refresh the fbcode platform010 toolchain and library pins to current third-party2 versions, and remove the now-unused Intel TBB dependency.
Toolchain / deps:
- Fix update_dependencies.sh: the gcc and binutils trees moved from centos8-native to centos9-native, so the old paths resolved to nothing and a re-run emitted empty GCC_BASE/BINUTILS_BASE. Point them at centos9-native.
- Bump clang 15 -> 21; binutils 2.37 -> 2.43; libunwind 1.4 -> 1.8; valgrind 3.19 -> 3.22; plus hash-only refreshes for the LATEST-tracked libs. Regenerate dependencies_platform010.sh.
- Keep GCC pinned at 11.x. The only newer GCC in third-party2 (13.x) is built for glibc >= 2.35 (its libgcc_s needs _dl_find_object@GLIBC_2.35), but platform010 ships glibc 2.34, so GCC 13 will not link/run here.
- zlib stays 1.2.8 (the only version with an x86_64 platform010 build).
- Document why GCC/libgcc/glibc/zlib remain pinned in update_dependencies.sh.
Remove TBB (no longer used) from the build system: the get_lib_base entry, both fbcode_config*.sh, build_detect_platform detection, the CMake WITH_TBB option / find_dependency, and cmake/modules/FindTBB.cmake. Dockerfiles and the HISTORY.md changelog are left untouched. (TBB was used by the old clock cache, long ago removed.)
Although this change was originally motivated by upgrading gcc for its libasan not to hit process lifetime thread limits, upgrading gcc proved impractical under platform010.
Bonus: fix USBAN+gcc build by making GetParam() valid by the time it is called in several test class constructors.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14882
Test Plan:
a variety of local builds (gcc, clang; various sanitizers) using fbcode tooling
No production code changes
Reviewed By: xingbowang
Differential Revision: D109631787
Pulled By: pdillinger
fbshipit-source-id: 9c466c59b039c2a67ee0318c0ccbac02e19f537b
Summary:
Add EXPERIMENTAL embedded blob SST support for SstFileWriter through
OpenWithEmbeddedBlobs(). Eligible large values are written as same-file blob
records inline in a block-based SST as values are added (interleaved with data
blocks), while table entries store same-file BlobIndex references that readers
resolve for Get, MultiGet, and iteration, including mixed embedded and
non-embedded wide-column values.
Embedded-blob handling is folded directly into BlockBasedTableBuilder rather than
living in SstFileWriter or a separate table-builder wrapper: SstFileWriter only
selects the mode (via TableBuilderOptions::embedded_blob_options), and the
builder writes blob records inline using its own file writer and running offset.
This is enabled by disabling index-value delta encoding for these SSTs — delta
encoding reconstructs a data block's offset from the previous block and so
requires byte-contiguous data blocks, which interleaved blob records would break.
With full (non-delta) index values, blob records can sit between data blocks, so
no entry buffering/replay is needed. To keep inline blob appends correctly
ordered with data-block writes, these SSTs use single-threaded (non-parallel)
compression. The mode is the only entry point today but the placement keeps it
open to generalization beyond SstFileWriter; regardless, this experimental
feature is expected only to have niche applications.
(Previous revisions of this change allowed delta-encoded index blocks by
putting all blobs at the beginning of the file, but that was a more awkward
and memory-hungry implementation due to buffering all the data blocks before
writing.)
The on-disk record format (SimpleGen2Blob: payload bytes followed by a 5-byte
trailer of a compression marker plus a builtin checksum that is context-modified
by the record's absolute file offset) lives in db/blob/blob_gen2_format.{h,cc},
which now owns both the read (ReadAndVerifySimpleGen2BlobRecord) and write
(WriteSimpleGen2BlobRecord) sides so the format is defined in one place. This is
expected to be reused for upcoming "blog file" support.
Readers need no record-layout metadata: same-file blob resolution is purely
absolute-offset keyed, and the per-record offset-modified checksum (plus a cheap
"record fits within the file" bound) is the corruption guard. The reader's only
embedded-blob metadata is presence: a best-effort auxiliary table property
(blob count and payload-byte totals, for diagnostics) whose mere presence signals
that the SST contains embedded blobs.
Reads route through the column family's BlobSource when a DB is attached, so
embedded payloads are served from / inserted into the blob value cache and
recorded in BLOB_DB_* statistics; the cache key is derived from the
SimpleGen2Blob offset scheme (the same scheme block-based SST blocks use), so
embedded blob records stay collision-free with data blocks even when the blob
cache and block cache are the same cache. Non-DB openers (SstFileReader,
sst_dump, repair, ingestion prevalidation) have no BlobSource and fall back to a
direct, uncached read.
Same-file BlobIndex references use blob file number 0 as the marker. That value
also serves as the invalid blob-file-number sentinel in broader metadata code,
but the meanings do not conflict when used carefully: only the embedded-SST
reader/writer path interprets 0 as same-file, while generic file-metadata paths
continue to reject it as invalid. Using 1 would be worse because legacy
"stackable" BlobDB can use low blob file numbers, including 1, so reserving it
would collide with real blob files.
Compression options remain in the public API as placeholders, but embedded blob
compression support is deferred. Integrating compression with
BlockBasedTableBuilder while avoiding copied CompressAndVerifyBlock-style logic
is tricky enough to deserve a separate, focused PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14851
Test Plan:
- Added basic feature coverage to the crash test.
- Added BlobIndexTest.SameFileBlobIndex and BlobGarbageMeterTest.SameFileBlobIndex
coverage for same-file BlobIndex encoding, display, recognition, and ignoring
same-file references in blob-garbage accounting.
- Extended FileMetaDataTest.UpdateBoundariesBlobIndex to preserve the generic
zero-file-number corruption check while keeping same-file embedded blob
semantics at the table-reader/writer layers.
- SstFileReader embedded blob coverage: round-trip Get, MultiGet, and iterator
reads; format_version gating; ignored placeholder compression options; the
2048-byte default min_blob_size; wide-column mixed embedded/non-embedded
values; and early append-error surfacing.
- Added an interleaved-layout test (small block_size with alternating
small/large values) asserting the SST property index_value_is_delta_encoded==0,
more than one data block, and that blob records are interspersed with data
blocks (not a strict front prefix), with all values read back correctly via
Get and iteration; replaces the old "ignored bytes before/after the blob record
prefix" test.
- Added an embedded-record corruption test: flipping a byte inside a blob
record's payload yields Corruption on read with verify_checksums (the
offset-keyed record checksum is the guard now that the range pre-check is gone).
- Exercised the cached read path through BlobSource, including blob-cache
hit/miss behavior and the shared blob_cache == block_cache configuration, in
db_blob_index_test.
- Normal-path CPU regression check: release (DEBUG_LEVEL=0) db_bench on a
non-embedded DB comparing this change vs upstream main, 3 interleaved reps of
fillseq, fillrandom, readrandom, and readseq (5M keys, value_size=100,
compression none, DB on /dev/shm). All deltas were within run-to-run noise
(~1%), i.e. no measurable regression from adding the embedded-mode branch to
the builder hot path.
Reviewed By: xingbowang
Differential Revision: D108564468
Pulled By: pdillinger
fbshipit-source-id: 5f01ffb1d40c6fd5b82d2451ec3342abb5040ca6
Summary:
- Current 3rd party jdk8 installation fails due to trust issue in homebrew.
- Detect and reuse an existing Jdk installation on macOS runners before installing a new cask.
- Still target on jdk 8 binary build.
- Stop using 3rd party jdk installation,
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14885
Test Plan: - CI
Reviewed By: jaykorean
Differential Revision: D109696828
Pulled By: xingbowang
fbshipit-source-id: cc71958a05bd58acaac5f05fe3349ffd3e57880f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14881
Read-only DB instances can share a directory with a live writer, so their view of live files is only a snapshot from the time they opened. After read-only DB open began setting `opened_successfully_` for its intended open-success semantics, that state had an unintended side effect: the common close path treated a successful read-only open like a read-write open and ran obsolete-file cleanup. If the writer created or made files live after the read-only handle opened, that cleanup could use the stale read-only live set and delete files still needed by the writer.
This change records whether a `DBImpl` was opened read-only and skips close-time obsolete-file cleanup for read-only DBs instead of simply keeping `opened_successfully_=false`, which could be confusing and easily mistaken in the future again.
For completeness I also updated other callsites of `opened_successfully_`, but those should not be real bugs as ReadOnly DBs do not run flushes/compactions or write to WALs.
Reviewed By: xingbowang
Differential Revision: D109622445
fbshipit-source-id: d7be4b20fce86ccb218a63ac6f5b707b316aac79
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14880
The RocksDB 11.5.0 reverse-comparator range-lock fix (the HISTORY.md entry "Fixed a bug where the range lock manager would crash with an assertion failure when using a reverse comparator column family") made `RangeTreeLockManager::CompareDbtEndpoints` reverse-aware by wrapping the endpoint-suffix tie-break results in an `is_reverse` flip. The length-disparity (different user key) branches need that flip, but the equal-length / equal-user-key branch was flipped too, and that is incorrect.
Range-lock endpoints are encoded as `[suffix_byte][user_key]` where the suffix is `SUFFIX_INFIMUM` (0x0) or `SUFFIX_SUPREMUM` (0x1). When two endpoints share the same user key and differ only in the suffix byte, the suffix is a positional boundary marker for that key, not key content, so the infimum endpoint must sort before the supremum endpoint in any total order, regardless of the column family sort direction. Flipping it for a reverse comparator makes `m_cmp(infimum(K), supremum(K)) > 0`.
MyRocks issues an equality lock (e.g. `UPDATE t1 SET a=123 WHERE a=35`) as the range `[infimum(K), supremum(K)]` over the same user key K. On a reverse (`rev:`) column family — which uses the stock `ReverseBytewiseComparator()`, so `is_reverse` is true — the flipped comparator reports `left > right`, tripping `paranoid_invariant(m_cmp(left_key, right_key) <= 0)` in `toku::locktree::try_acquire_lock` and aborting the server with SIGABRT. This broke many MyRocks `---range_locking` MTR tests (partial_index*, bypass_select_range_sk*, alter_table_online_debug, allow_no_primary_key_with_sk) after RocksDB was bumped to 11.5.0 in D108936924.
Fix: in the equal-length / equal-user-key branch of `CompareDbtEndpoints`, compare the suffix bytes without the `is_reverse` flip so infimum always precedes supremum. The length-disparity branches keep the reverse-aware flip (still covered by `RangeLockWithReverseComparator`).
Note: this lands in `internal_repo_rocksdb/repo`, the RocksDB development tree; it reaches `fbcode/rocksdb/src` (what MyRocks/ZippyDB link) through the RocksDB release sync. Tracked in T276980728.
Reviewed By: xingbowang
Differential Revision: D109591908
fbshipit-source-id: 078b16e6e51f4f365b4a8d539ac5b968a7461729
Summary:
This continues and finishes **https://github.com/facebook/rocksdb/issues/14572** ("Add semi-automated code generation for RocksDB C API bindings") by xingbowang. The original author is unavailable to finish it, so I've taken it over. **All 13 of the original commits are preserved** (this branch was created from the PR head and builds on top of it — `git log` shows the original `Xingbo Wang` authorship intact); my follow-up work is in the commits prefixed `C API codegen:`.
The underlying design is unchanged and is the original author's: hand-written source templates (`tools/c_api_gen/c_base.h` / `c_base.cc`) plus two generators (auto-discovery from the C++ headers + a spec-driven generator) are inlined into a single, self-contained, `generated` `include/rocksdb/c.h` and `db/c.cc`. This grows the public C API by **668 functions** while keeping `c.h` a single includable header with no `-I` requirement (so `bindgen` and other FFI tools keep working unchanged).
This branch reconciles the PR with ~4 months of `main` and addresses the outstanding review feedback (clang-tidy bot, the automated code review, the `c.h` self-containedness discussion, and pdillinger's points about `include/rocksdb` hygiene and `generated` marking).
## What changed on top of the original PR
### Reconciled with current `main`
- Merged current `main` (conflicts were confined to the generated/test files) and regenerated. Reconciled the 14 C API functions `main` added since the merge-base (e.g. `rocksdb_set_db_options`, the backup-engine rate limiters, `memtable_batch_lookup_optimization`, `optimize_multiget_for_io`, …) and restored 5 enum constants that upstream had added by hand (`rocksdb_txndb_write_policy_*`, `..._index_block_search_type_auto`, `rocksdb_blob_cache_read_byte`).
### Maintainer feedback (pdillinger)
- **No non-user-includable files in `include/rocksdb`.** Moved the hand-written templates out of `include/rocksdb/` and `db/` to `tools/c_api_gen/c_base.{h,cc}`. They were `#include`-ing generated fragments, which broke `make check-headers` and was shipped by `make install`. `include/rocksdb/` now contains only the user-facing, self-contained, `generated` `c.h`.
- `c.h` / `c.cc` carry the `// generated` marker.
### Backward compatibility (zero ABI break)
- The generator derived each wrapper's C type purely from the C++ field, which had silently changed **5 already-shipped signatures** (e.g. `rocksdb_writeoptions_disable_WAL` `int` → `unsigned char`). Added an ABI type-pinning layer (`tools/c_api_gen/abi_type_overrides.json`) so already-shipped functions keep their historical C signature (the body still casts to the real field type). A repo-wide diff against the merge-base now reports **0 ABI drift**.
- New `check_api_compatibility.py` gate (wired into CI + `make`) fails on any removed/changed public function **or** removed enum/typedef symbol, vs a reference revision. Intentional changes go in an allowlist with a reason.
### Correctness (from the automated review)
- Restored 5 option setters that were declared in `c.h` but **defined nowhere** (link failure for downstream bindings such as `rust-rocksdb`). Added `check_api_completeness.py` (dependency-free; runs in CI + `make`) asserting every declared function has exactly one definition — this is the gate that would have caught it.
- `CopyStringVector` now null-checks `malloc`; the WAL filter `std::move`s the `WriteBatch`; the backup exclude-files callback captures by value instead of the wrapper pointer.
### Build / CI robustness
- Removed the dead `C_API_CODEGEN_STAMP` Makefile prerequisite (it was a silent no-op).
- The `make check` staleness check is now opt-out-able (behind `SKIP_FORMAT_BUCK_CHECKS`) and skips gracefully when `clang++` is unavailable, so `make check` works without the codegen toolchain. CI remains the authoritative gate.
- Pinned `clang-format` consistently through `regen_all.py` / `verify_generated_up_to_date.py` (CI uses clang-format-21) so regeneration is byte-reproducible across environments.
- Cleared all 20 `clang-tidy` warnings the bot reported on `db/c.cc` changed lines (fixed in the `c_base.cc` template, not the generated output).
- Updated the internal Buck `c_test_bin` wrapper to expose generated `c_api_gen/*.inc` fragments as headers, so sandboxed Buck builds can compile `db/c_test.c` after the generated round-trip tests are included.
### Test coverage
- Added `gen_roundtrip_tests.py`, which derives **462 set→get→assert round-trip checks across 25 option objects** from the same generated fragments and wires them into `db/c_test.c`. Coverage now tracks the generated surface automatically.
### Docs
- Added the `unreleased_history/public_api_changes` note and fixed `claude_md/add_public_api.md`, which still told contributors to hand-edit the now-`generated` `c.h`/`c.cc`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14868
Test Plan:
- `make c_test && ./c_test` — **passes**, including the 462 generated round-trip assertions (a successful link also confirms the API is complete).
- `python3 tools/c_api_gen/check_api_completeness.py` — all 1737 declared functions defined exactly once.
- `python3 tools/c_api_gen/check_api_compatibility.py --ref <release>` — 1070 reference functions + 229 enum/typedef symbols preserved, 0 removed/changed.
- `python3 tools/c_api_gen/verify_generated_up_to_date.py` — generated output is stable.
- `include/rocksdb/c.h` confirmed self-contained (only `<stdbool.h>`, `<stddef.h>`, `<stdint.h>`).
cc xingbowang
- `buck2 build --flagfile fbcode//mode/dev fbcode//internal_repo_rocksdb/repo:c_test_bin` — passes.
- `buck2 build --flagfile fbcode//mode/dev --config fbcode.arch=aarch64 fbcode//internal_repo_rocksdb/repo:c_test_bin` — passes.
Reviewed By: pdillinger
Differential Revision: D109149150
Pulled By: xingbowang
fbshipit-source-id: 3417375345f360a4c78bdfe27e9850b89d0a226a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14877
The rollback-after-failed-commit retry loop in `db_stress_test_base.cc` (added in b0b52e1d14f3) counted every loop iteration — including iterations where `Resume()` returned Busy — against a single 100-count limit. With 32 concurrent threads, when write fault injection causes all prepared transaction commits and rollbacks to fail simultaneously, `Resume()` returns Busy (recovery in progress from another thread) for the entire 100-iteration budget (100 × 10ms = 1 second). Recovery itself keeps failing due to ongoing fault injection causing repeated background flush errors, so no thread ever gets a chance to perform an actual rollback.
The fix separates two budgets:
- `kMaxRollbackRetries` (100): counts only actual rollback attempts after `Resume()` returns OK
- `kMaxResumeBusyWaits` (3000): allows up to 30 seconds of busy-waiting for recovery to complete on another thread
Key triggering options: `write_fault_one_in=1000` + `txn_write_policy=1` (write-prepared) + `two_write_queues=1` + 32 threads + `error_recovery_with_no_fault_injection=1`.
Reviewed By: joshkang97
Differential Revision: D108561314
fbshipit-source-id: 03f4578ea33289685ad982208451f0584bf0535f
Summary:
- Add a blog post explaining FIFO KV-ratio compaction for BlobDB-backed TTL workloads.
- Describe why the existing FIFO intra-L0 picker can repeatedly compact small BlobDB SST files and how the new KV-ratio target plus tiered merging strategy works.
- Add diagrams for the old picker behavior and the KV-ratio tiering flow, along with the required blog author metadata.
## Testing
- `git diff --check upstream/main...HEAD`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14871
Reviewed By: joshkang97
Differential Revision: D109243987
Pulled By: xingbowang
fbshipit-source-id: 7ad07953f38c55c77ffb6510824ba8500db78c0e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14874
### Problem
RocksDB [alert](https://www.internalfb.com/monitoring/alert/?alert_id=twshared38314.04.ldc2%40%23%24host_health.crashlooping_key.tw_agent.tw_agent.task_failure.app.count.60%40%23%24uhc%3A%20tupperware%20task%20crashlooping) fired for `tsp_ldc/RocksDB/rocks_on_ws_stress_test` after the job started crashlooping with `ModuleNotFoundError: No module named fault_injection_log_parser`.
### RCA
D101973626 added `import fault_injection_log_parser` to `db_crashtest.py`, but `rocksdb_db_stress_internal_repo` only packaged `crashtest.py` and `rocks_db_stress`, so the `release_test` fbpkg rolled to TW without the new helper module.
### Fix
Export `tools/fault_injection_log_parser.py` and include it in the `rocksdb_db_stress_internal_repo` fbpkg `path_actions` so it is installed next to `crashtest.py`; also add the BUCKLINT-required `network_access = network_access_utils.none()` on the existing `db_stress_auto_tuner_test` target.
Reviewed By: xingbowang
Differential Revision: D109346224
fbshipit-source-id: 3f770a7ad306d246f279641bf0f19142d05560c4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14862
Add blog post for range tombstone conversion.
Reviewed By: xingbowang
Differential Revision: D108946953
fbshipit-source-id: f2ded24a5452d2f56a12928b979a8f44930dde03
Summary:
Update Windows CI to stop hardcoding Visual Studio 2022. The Windows build action now detects the installed Visual Studio C++ toolchain with vswhere, derives the matching CMake generator, and passes the corresponding version range to microsoft/setup-msbuild@v2.
Also update the Windows build steps to use the preinstalled JDK instead of installing Liberica JDK8 through Chocolatey, and build Snappy/RocksDB through cmake --build, instead of assuming fixed .sln names. Rename Windows CI jobs from VS2022-specific names to MSVC-neutral names, and move the nightly AVX2 Windows job to windows-8-core so PR and nightly CI use the same runner/toolchain path.
## Testing
- CI
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14869
Reviewed By: jaykorean
Differential Revision: D109176291
Pulled By: xingbowang
fbshipit-source-id: 848c5964922cf6e09b727b1ed96cbf475b325142
Summary:
# Summary
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14864
With `two_write_queues`=`true`, `WRITE_UNPREPARED` can allocate sequence numbers through `FetchAddLastAllocatedSequence()` before those numbers are published through `SetLastSequence()`. Error recovery can then create new memtable/WAL boundaries from a stale `LastSequence()`, which can later surface as "sequence number going backwards" corruption. Fix this by syncing `LastSequence()` to `LastAllocatedSequence()` at the recovery entry point after draining both write queues, and again at the recovery flush fence after `FlushAllColumnFamilies()` has released/reacquired the DB mutex and before `SwitchMemtable()` consumes `LastSequence()`. Apply the same fence to atomic recovery flushes.
# Perf comparative results
TLDR; no visible regression.
I ran a deterministic local recovery benchmark to check whether the new two-write-queue recovery fence adds measurable recovery latency.
## Workload (P2385173320)
- Optimized build.
- `TransactionDB` with `WRITE_UNPREPARED`.
- `two_write_queues=true`.
- Each iteration dirties the memtable with WUP transactions, forces a retryable flush IO error through `FaultInjectionTestFS`, re-enables the filesystem, then measures `DB::Resume()`.
- 50 warmup recoveries + 500 measured recoveries per run.
## Why this is relevant
This targets the changed recovery path directly rather than measuring generic write throughput. The change only runs during recovery, so the useful signal is recovery latency around `ResumeImpl()` and the recovery flush fence.
## Code path exercised
- `DB::Resume()`
- `ErrorHandler::RecoverFromBGError(true)`
- `DBImpl::ResumeImpl()`
- new `two_write_queues_` recovery queue fence
- non-atomic recovery `FlushMemTable()` fence before `SwitchMemtable()`
This does not cover atomic flush recovery or a writer-backlog stress case.
Comparative results:
| Revision | Attempt | resume_p50_us | resume_p95_us | Retryable BG errors |
| Parent | | 1555.75 | 2092.22 | 550 |
| Fix | 1 | 1537.68 | 1905.65 | 550 |
| Fix | 2 | 1475.10 | 1883.26 | 550 |
| Fix | 3 | 1617.52 | 1993.97 | 550 |
Reviewed By: xingbowang, anand1976
Differential Revision: D108946310
fbshipit-source-id: 75fcb4bf0c9a932b5fbed7e934c7f14d981a1bf2
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14872
### Motivation
The `build-windows-vs2022` job in `pr-jobs.yml` was running on the `windows-8-core` runner, whose image no longer provides a Visual Studio 2022 installation. The job sets `CMAKE_GENERATOR: Visual Studio 17 2022`, so CMake's generator probes for an installed VS 2022 instance and aborts the Snappy dependency configure with:
```
CMake Error: Generator Visual Studio 17 2022 could not find any instance of Visual Studio.
```
The failure surfaces during the thirdparty Snappy build, before any RocksDB code is compiled, so it blocks the entire Windows PR signal.
### Changes
Switch the `build-windows-vs2022` job from `runs-on: windows-8-core` to `runs-on: windows-2022`. The GitHub-hosted `windows-2022` image ships VS 2022 with the Desktop C++ workload preinstalled, so the `Visual Studio 17 2022` generator resolves out of the box and the `CMAKE_GENERATOR` value stays consistent with the runner's installed toolchain. This also brings `pr-jobs.yml` in line with `nightly.yml`, whose `build-windows-vs2022-avx2` job already runs on `windows-2022`.
NOTE: `windows-2022` is the standard 2-core hosted runner (vs. the larger `windows-8-core` pool), so Windows builds will be somewhat slower; ccache keeps the compile cost mostly in cache hits.
Reviewed By: pdillinger
Differential Revision: D109229258
fbshipit-source-id: b9f6d9ef8b4080b1b1129ed0ae6a9b1da2b7b14c
Summary:
- switch fault injection error recording from an in-memory ring buffer to per-run fixed-record binary logs under `TEST_TMPDIR/fault_injection_logs` (or `/tmp/fault_injection_logs`) so crash paths survive DB reopen cleanup
- keep the raw and decoded fault logs for external artifact collection/cleanup, and make `db_crashtest` print consistent blackbox/whitebox summaries after decoding
- make expected-state tracing fail fast on trace write failures and document offline trace inspection via `trace_analyzer`
- add coverage for binary log persistence/decoding/truncated-tail handling and keep info logs excluded from fault injection
Reviewed By: hx235
Differential Revision: D101973626
fbshipit-source-id: fdcb5b6370cf92a046e09b8d3391e80eecb66c23
Summary:
db_bench worker operations historically responded to a fatal error (failed Get/Put/Merge/Delete, checksum failure, etc.) by calling db_bench_exit() -> exit() in place. When that happened on a worker thread while sibling workers and background (compaction/flush) threads were still running, exit() began static/global destruction underneath those live threads. A thread mid-call into a user callback (e.g. a custom Env/ FileSystem) could then touch a global/singleton that had already been destroyed -> cross-thread use-after-free. This is the same class of bug fixed for db_stress in https://github.com/facebook/rocksdb/issues/14850, but it also matters for db_bench's secondary role as an *integration validation* tool: we want production-like processes to shut down cleanly after an error (so a custom FileSystem can log the adverse condition and run its teardown) rather than crash or _exit at the first sign of trouble.
This change makes fatal errors on benchmark worker threads unwind instead of exiting in place, and performs an ordered shutdown on the main thread:
* SharedState gains a cooperative fatal flag + first-error Status/message and SetFatal(). A worker that hits a fatal error records it and returns out of its operation (releasing its DbUseGuard) instead of exiting.
* Op loops stop promptly via Duration: the loop-termination helper now consults the run's fatal flag at its existing throttled check boundary (~every FLAGS_ops_between_duration_checks ops), so peers wind down with no added per-key cost.
* After RunBenchmark joins all workers, the (guard-free) main thread runs the ordered shutdown -- DeleteDBs() closes all DBs, joining background threads and running user Env/FileSystem teardown -- and only then exits nonzero. No static destructors run while RocksDB threads or user callbacks are still live.
* The worker fatal sites that previously called db_bench_exit() (BGWriter, DoDelete, RandomWithVerify, UpdateRandom, XORUpdateRandom, MergeRandom, ReadRandomMergeRandom, VerifyChecksum, VerifyFileChecksums, RandomReplaceKeys, Replay, DoDeterministicCompact) now SetFatal + return. Flush (main-thread) routes to ErrorExit() which closes DBs first.
Supporting changes:
* Duration is now a nested member type of SharedState, constructible only via SharedState::MakeDuration() (movable, non-copyable). This binds every Duration to its run's fatal flag by construction and removes the previous mutable global (g_cooperative_abort_flag) and its set/clear lifecycle. A `using Duration = SharedState::Duration;` alias keeps call sites tidy.
* New test-only flag --fault_injection_read_after_reads: wraps the FileSystem so random-access reads start returning a non-retryable IOError after N reads, to exercise the fatal/shutdown path on demand.
* New flag --fatal_shutdown_watchdog_seconds (default 180, 0 disables): arms a one-shot detached watchdog on the first fatal error (and before ErrorExit's cleanup) that forces port::ImmediateExit() if the careful shutdown itself hangs (e.g. a background thread or misbehaving Env that never returns). This is a backstop, not the path under test.
* Added an "Error handling background" comment documenting the rationale and the machinery, and an invariant comment at db_bench_exit covering pre-Open vs worker vs main-thread exit rules.
Existing _Exit()/abort() sites are intentionally left unchanged.
Future work:
* Worker ops that still abort() on error (ReadRandom, ReadRandomFast, ReadToRowCache, MultiReadRandom, MixGraph, AppendRandom, UpdateRandom's read path, BGScan, RandomTransaction, CreateNewCf) are not the use-after-free class (abort() skips static destruction) but are not graceful either; converting them would extend clean-shutdown coverage. RandomTransaction needs a rollback decision; MultiReadRandom needs a structured break-out.
* Post-Open main-thread db_bench_exit() sites in Open()/Run()/ VerifyDBFromDB() are the harder, https://github.com/facebook/rocksdb/issues/14850-class case: Open() is reachable both under and outside a DbStateMutationGuard, so it cannot unconditionally call ErrorExit() without deadlock. Needs guard-aware handling or restructuring Open() to return Status.
* Pre-Open setup exits (option/flag parsing, cache setup) correctly stay plain exit() -- no live threads -- and are out of scope.
* Fault injection currently covers read faults only; verifying graceful shutdown for write/open/transaction error paths would need write-fault, open-fault, and conflict injection.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14855
Test Plan:
Built debug db_bench (make db_bench) and validated manually:
* Correctness of loop termination (unchanged by the Duration refactor):
- Count-bounded is exact: `db_bench --benchmarks=fillrandom --num=300` reports "300 operations".
- Time-bounded honors duration and is not capped by max_ops: `db_bench --benchmarks=readrandom --use_existing_db --duration=5 --reads=3000` ran 5.002 s / ~2.0-2.7M operations.
* Graceful shutdown on a fatal error (multi-threaded, peers live): `db_bench --benchmarks=xorupdaterandom --use_existing_db --threads=8 --duration=30 --fault_injection_read_after_reads=5000` -> exit code 1, ZERO "Received signal", prints 'Fatal error during "xorupdaterandom"' after the ordered shutdown.
* Confirmed this is a real fix, not just masking: rebuilt pristine main + only the fault-injection flag (graceful-shutdown changes reverted) and ran the same command -> SIGSEGV (rc=139), crash stack in __run_exit_handlers -> jemalloc free, triggered from XORUpdateRandom -> db_bench_exit() -> exit() with sibling threads still live. With the fix in place the same workload is graceful (above).
* Watchdog:
- Default (180 s): graceful runs above complete with no watchdog message (no spurious firing).
- --fatal_shutdown_watchdog_seconds=0 disables it (still graceful).
- Positive: temporarily inserted a 60 s sleep before DeleteDBs in the fatal path, ran with --fatal_shutdown_watchdog_seconds=3 -> process exited at ~3 s (not 60 s) with the watchdog message and rc=1; reverted the temporary sleep and rebuilt clean.
* No benchmark performance impact (normal, error-free runs):
- Structural: the cooperative-abort check adds no per-key work. It is evaluated only at Duration's pre-existing throttled boundary (~every FLAGS_ops_between_duration_checks ops, same site as the existing time check); there is no atomic load on the per-op hot path. Fatal recording (SetFatal) lives only in cold error branches. A Duration is constructed once per benchmark via the factory (returned by value, elided), not per op.
- Observed: error-free fillrandom / fillseq / readrandom throughput (e.g. fillrandom ~300K ops/s, fillseq ~800K-870K ops/s) was in line with pre-change runs; no regression observed.
* make format-auto applied; debug build is clean.
Reviewed By: anand1976
Differential Revision: D108690137
Pulled By: pdillinger
fbshipit-source-id: 7ffdf4abc665b4c2172b718670c5c2fa1da58500
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14854
External SST ingestion can already skip the open-and-scan metadata path when ingesting files produced by `SstFileWriter`. This extends that fast path to live SST files produced by a DB, so applications that move DB-generated files between DBs can request prepared metadata directly from the source DB and pass it through ingestion.
The API is intentionally limited to live table files whose metadata is already available from the DB and cached table reader. Callers remain responsible for keeping the source file alive and unchanged until ingestion completes.
Reviewed By: xingbowang
Differential Revision: D108440584
fbshipit-source-id: e732278ea194aac71f046b54234238b35c6fe0e5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14853
`IngestExternalFileOptions` now exposes `file_opening_threads` so external file ingestion can open table readers for newly committed SSTs in parallel. The requested thread count is carried from each ingestion job into commit and then into manifest `LogAndApply`, where `VersionBuilder::LoadTableHandlers()` uses it while installing the ingested files.
This is intended for workloads that ingest many external SSTs in one call, where serial table-reader opens can dominate commit latency. The ingestexternalfile benchmark now has a flag for this option, and `db_stress` now creates multiple data SSTs per ingest operation so stress runs cover multi-file ingestion, file-info ingestion, and the parallel file-opening path together.
Benchmarks
Ran `db_bench --benchmarks=ingestexternalfile --use_existing_db=0 --num=2200000 --compression_type=none --statistics --use_direct_reads=true --ingest_external_file_batch_size=200 --ingest_external_file_num_batches=1 --ingest_external_file_use_file_info=true --ingest_external_file_fill_cache=false` while varying `--ingest_external_file_file_opening_threads`. The generated SST files were about 256 MiB each.
| `file_opening_threads` | `rocksdb.ingest.external.file.prepare.micros` | `rocksdb.ingest.external.file.run.micros` | File opens |
| 1 | 16,150 us | 327,136 us | 200 |
| 32 | 16,665 us | 38,349 us | 200 |
At the same ~256 MiB SST size, 2,000 files is roughly 500 GiB of SST data. A straight 10x extrapolation of the 200-file ingest histograms puts prepare+run at about 3.43s with `file_opening_threads=1` and 0.55s with `file_opening_threads=32`; run alone is about 3.27s vs. 0.38s.
Reviewed By: pdillinger
Differential Revision: D108347952
fbshipit-source-id: 49efa48524df64cace010b881030f63f771cbb95
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14837
External file ingestion (`DB::IngestExternalFile[s]`) re-opens every SST file and scans it -- footer, properties, index, filter, and the first/last data blocks -- to recompute the boundary keys, sequence-number bounds, and table properties before committing. For cold, I/O-bound files this scan dominates ingest latency, even when the file is moved/linked rather than copied.
This change lets a caller skip that work when it already has the file's metadata. `SstFileWriter::Finish()` now returns a `PreparedFileInfo` with the file size, table properties, and prepared internal `smallest`/`largest` bounds, and a new `IngestExternalFileArg::file_infos` field carries one `PreparedFileInfo` per file into `IngestExternalFiles()`. When set, ingestion reuses that metadata instead of re-opening and scanning each file. The file is still copied/linked, and the checksum is still verified when `verify_checksums_before_ingest` is set (the fast path opens the file only for that). Point-key and range-deletion bounds are folded into the same prepared bound pair, and user-defined-timestamp files (including the "UDT in Memtables only" format, whose boundary keys carry no timestamp) are handled.
Internally, ingestion-job metadata acquisition was split into `GetIngestedFileInfoFromFileInfo` (reuse caller metadata) and `GetIngestedFileInfoFromFile` (open + scan). Prepared boundary updates use RocksDB comparators, and the timestamp-stripping path pads prepared bounds back to the internal timestamp shape before installing the file.
The `ingestexternalfile` `db_bench` benchmark now also exposes `--ingest_external_file_fill_cache` so ingestion-read block-cache effects can be controlled independently while comparing the file-info fast path against the normal open-and-scan path.
A future PR will support creating `PreparedFileInfo` from a existing DB-generated file, so this optimization is not just limited to SstFileWriter.
## Benchmark Results
Existing `db_bench ingestexternalfile` benchmark, release build, using an XFS flash-backed filesystem. Files are linked (`move_files`) so the measurement isolates the ingest path rather than file-copy throughput. These numbers used direct reads and `fill_cache=false` to avoid warming the block cache while comparing the file-info fast path against the normal open-and-scan path.
Each run used 1M keys/SST, one ingest call, and `db_bench` reported a 62.9 MB estimated file size. The comparison varied the number of files in that ingest call across 10, 30, and 50 files.
| Files/batch | Config | Prepare P50 | Run P50 | Run P50/file | Total ingestion P50 | Total P50 drop |
| --- | --- | --- | --- | --- | --- | --- |
| 10 | Baseline | 10.909 ms | 7.593 ms | 0.759 ms/file | 18.502 ms | -- |
| 10 | `file_info=true` | 1.127 ms | 7.541 ms | 0.754 ms/file | 8.668 ms | 53.2% |
| 30 | Baseline | 29.049 ms | 23.161 ms | 0.772 ms/file | 52.210 ms | -- |
| 30 | `file_info=true` | 2.734 ms | 23.384 ms | 0.779 ms/file | 26.118 ms | 50.0% |
| 50 | Baseline | 49.036 ms | 39.224 ms | 0.784 ms/file | 88.260 ms | -- |
| 50 | `file_info=true` | 4.133 ms | 39.417 ms | 0.788 ms/file | 43.550 ms | 50.7% |
With `fill_cache=false` and direct reads, the metadata fast path cuts `Prepare()` by roughly 90-92%. `Run()` is roughly 0.75-0.79 ms per file across these runs. End-to-end `db_bench` micros/op is almost unchanged because this benchmark includes SST generation and compaction work; use `rocksdb.ingest.external.file.prepare.micros` and `rocksdb.ingest.external.file.run.micros` to isolate ingestion.
Benchmark args: `--benchmarks=ingestexternalfile --use_existing_db=0 --num=1000000 --compression_type=none --statistics --use_direct_reads=true --ingest_external_file_batch_size=<10|30|50> --ingest_external_file_num_batches=1 --ingest_external_file_use_file_info=<false|true> --ingest_external_file_fill_cache=false`.
Reviewed By: xingbowang
Differential Revision: D107721261
fbshipit-source-id: b06ecb7b35f260ec02acf837e7986057bc23cdbf
Summary:
# Context
The C API lacks a SetDBOptions wrapper, so callers using the C bindings cannot dynamically reconfigure DB-level options at runtime without dropping down to the C++ API.
# Changes
- Add `rocksdb_set_db_options` to the C API as a wrapper around `DB::SetDBOptions`
- Adds new tests in `c_test.c` covering both valid and invalid option updates
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14615
Reviewed By: joshkang97
Differential Revision: D101010593
Pulled By: xingbowang
fbshipit-source-id: 3feac94bd0e3b7d6a4913ce9bc014e1233292309
Summary:
We have seen multiple coding error with the usage of WideColumn API. Hence, improving the documentation to reduce the chance of this happening again.
Clarify that `WideColumn` and `WideColumns` are non-owning views over caller-managed memory.
Update the public `PutEntity()` API documentation to state that column name and value storage must remain valid until the call returns.
Add safe and unsafe usage examples in `wide_columns.h` to make lifetime requirements explicit.
## Testing
Not run. Documentation-only change.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14626
Reviewed By: archang19
Differential Revision: D101280295
Pulled By: xingbowang
fbshipit-source-id: 076103321f1d54103dd1c2772aa9e84affdfd2ad
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14849
This change splits ingestion into two public calls: `PrepareFileIngestion()` performs all of the off-mutex work and returns an opaque `FileIngestionHandle`, and `CommitFileIngestionHandle()` makes the prepared files visible under the mutex.
Even though the "Prepare" phase was off the DB mutex, the application layer may still have application logic that depends on the completion of `IngestExternalFile`.
`IngestExternalFiles(args)` is also just `PrepareFileIngestion(args)` followed by `CommitFileIngestionHandle()`, so its behavior is unchanged.
`CommitFileIngestionHandles()` also supports committing multiple handles atomically. However, it is possible that a CF may be present in multiple handles. In this case, we merge the ingestion jobs together. This allows applications to prepare file ingestions at different times, but still provide a single atomic commit.
Reviewed By: xingbowang
Differential Revision: D108225105
fbshipit-source-id: 082e149f13983a4b14a8fd711d2113728485421f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14839
Wrap the filesystem passed to external table readers when statistics or file IO listeners are configured so reads performed by implementations such as Nimble update the same RocksDB metrics as regular SST reads. The wrapper records `SST_READ_MICROS`, activity-specific `FILE_READ_*` histograms, last/non-last-level and temperature read tickers, and forwards `OnFileReadFinish`/`OnIOError` to configured listeners so ZippyDB file-read counters observe external-table IO. `MultiRead` records the elapsed read histogram once and applies byte/listener accounting to each completed request.
Add a table test that makes a dummy external table reader read through `ExternalTableOptions::fs` and verifies both `Statistics` counters/histograms and file-read listener updates. Add an unreleased-history entry for the new external table read metrics coverage. This revision also applies the RocksDB `make check-format` clang-format output for `external_table.cc`.
Reviewed By: pdillinger
Differential Revision: D108012449
fbshipit-source-id: 6b6c993c0aa99479e6e043171124bd48f0416bec
Summary:
- Add `tools/gtest_parallel_repro.py` to reproduce gtest flaky tests that depend on process-level CPU contention and scheduler delays.
- Run many fresh gtest processes concurrently, isolate each process with its own `TEST_TMPDIR`, and optionally pin the workload to a smaller CPU set with `taskset`.
- Record per-run logs, write `failures.jsonl`, summarize failure keys, support stopping on first failed batch, and optionally rebuild with `COERCE_CONTEXT_SWITCH=1`.
- Document when to use the runner in `CLAUDE.md` for CI-style flaky tests that single-process `--gtest_repeat` or coerce mode alone does not reproduce.
- Using the new runner, we found and fixed one flaky test and improved c_test harness
- Flaky test: EnvPosixTest.ReadAsyncQueueFull simulated a full io_uring submission queue by overwriting the SQE pointer after io_uring_get_sqe() had already consumed a submission slot. That left stale state in the thread-local ring, so later AbortIO tests could process an unexpected completion and crash. Add a skip_io_uring_get_sqe syncpoint before the io_uring_get_sqe() call and use it from the test so the Busy path is exercised without mutating the ring.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14827
Test Plan: - CI
Reviewed By: pdillinger
Differential Revision: D107758097
Pulled By: xingbowang
fbshipit-source-id: 3256612ae1ae98b66901593371743b3de7e5bb53
Summary:
When `FinishInitDb()` or `Open()` calls `exit(1)` after the DB has been opened, background compaction/flush threads are still running. `exit()` triggers static object destruction (including the `KillPoint` singleton and its `rocksdb_kill_exclude_prefixes` vector) while those threads are still accessing them via `TestKillRandom()`, causing a heap-use-after-free detected by ASAN.
This became more likely to trigger after the multi-DB support commit (3d0d60101e7f) which runs `RunStressTestImpl` on worker threads, making the race window larger when one DB fails initialization while other DBs background threads are active.
The fix replaces `exit(1)` in all error paths that fire after the DB has been opened with a wrapper around `_exit(1)`. `_exit()` terminates immediately without running atexit handlers or destroying static objects, avoiding the race with background threads.
Also updated CLAUDE.md to help get cross-platform compatibility right the first time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14850
Reviewed By: hx235
Differential Revision: D108298839
Pulled By: pdillinger
fbshipit-source-id: 8e87fcb259c273e2be5eb26b4eaf6009f5b998f1
Summary:
kLZ4Compression and kLZ4HCCompression share the same on-disk format and decompressor, but historically kLZ4Compression only honored negative (acceleration) levels while kLZ4HCCompression only honored positive levels. This unifies them so `compression_opts.level` alone selects the variant: level <= 0 uses LZ4 fast (acceleration = -level) and level >= 1 uses LZ4HC (1..12), regardless of which of the two types is configured.
The configured type now only determines the default compression level (LZ4: acceleration 1, equivalent to level -1; LZ4HC: level 9).
For code simplicity, the recorded per-block type comes from the compression type derived from the level, which could differ from the configured type. To preserve the originally configured choice for debugging/tracking, it is recorded as a `_type=<decimal>` pseudo-option in the rocksdb.compression_options SST table property.
Out-of-range non-default levels are clamped to the nearest effective value (LZ4 acceleration capped at 65537, which also avoids signed overflow negating INT_MIN; LZ4HC level capped at 12). The cost-aware (auto-tune) compressor's LZ4 level grid is changed to negative accelerations so it actually exercises fast LZ4 (positive levels now route to LZ4HC).
Related inclusion: The ZSTD library has a discontinuity at level=0, which maps to level 3, which is more aggressive than levels 1 and 2, which are more aggressive than levels -1, -2, etc. For better friendliness to auto-tuning (etc.), we now map level 0 to be the same as level -1, so that increasing compression level numbers have non-decreasing aggressiveness.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14819
Test Plan:
New unit tests in compression_test.cc:
- UnifiedLZ4LZ4HCLevels: for a representative set of non-default levels, both configured types produce identical output and the same recorded type (selected by the level), levels that clamp to the same effective parameter compress identically, and each round-trips; plus per-type default-level behavior.
- ConfiguredCompressionTypeRecordedInProperties: the `_type=` pseudo- option appears in the SST table property for each configured type.
- ZSTDLevelZeroMapsToMinusOne: level 0 behaves like -1, not like 3.
Reviewed By: joshkang97
Differential Revision: D107536580
Pulled By: pdillinger
fbshipit-source-id: a1281956f70a75d0620cb73d0bfb9ad76c52cca3
Summary:
Fixed a bug where the range lock manager would crash with an assertion failure when using a reverse comparator column family. The CompareDbtEndpoints function used bytewise ordering for length-disparity comparisons, ignoring the user comparator direction.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14831
Test Plan:
- Added unit test RangeLockWithReverseComparator
- Ran make check (4552 tests, 0 failures)
- Verified range_locking_test passes (9/9)
Reviewed By: xingbowang
Differential Revision: D107880089
Pulled By: laurynas-biveinis
fbshipit-source-id: 77f2f2f58197ab5104d4e42f005b0aab1132f1bb
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14772
Updated the iterator creation scheme to happen lazily (on request) as oppsed to eagerly. this allows us to prune the iterator tree structure at the time of requesting iterator preparation as opposed to creation, and allows pruning to become an implementation detail. Version now skips non-overlapping SST levels and files before adding children to the iterator tree, returns direct table iterators when a level has a single matching file, and uses pruned LevelIterator instances when multiple files in one non-L0 level match. The overload no longer prepares iterators during creation; callers that need prepared multiscan execution still call Prepare explicitly after construction, and MultiScan does that itself.
Benchmark: ran `db_bench` in opt mode for the base revision and this diff, with `fillseq,compact,levelstats,multiscanrandom`, `--num=1000000`, `--reads=10000000`, single thread, fixed seeds, `--multiscan_use_async_io=false`, and `--use_multiscan=true`. Both A and B had exactly one SST file and no memtable/L0 data (`L0: 0 files`, `L1: 1 file, 61 MB`). `multiscanrandom` creates `MultiScanArgs` and calls `NewMultiScan(...)`, which reaches the new `NewIterator(..., scan_opts)` pruning path in this diff.
```
seed base A pruning B delta
424242 21824.333 17693.333 -18.9%
424243 24042.014 19424.056 -19.2%
424244 22424.974 17636.910 -21.4%
424245 22404.213 18612.840 -16.9%
```
Average: base `22673.9 us/op`, pruning `18341.8 us/op`, about `19.1%` faster.
Reviewed By: xingbowang
Differential Revision: D104904298
fbshipit-source-id: a742106a1d5813fb795a39eeeb35f8cddc02e886
Summary:
- Add a controlled async filesystem regression test for released direct I/O async reads, verifying released handles are aborted before a later `Poll()` can complete them.
- Add a MultiScan-style regression test for skipping a pending range remainder before seeking into a later async read.
- Extend db_stress/crashtest coverage so MultiScan can stop early after `--num_iterations` results and randomize `--multiscan_use_async_io` again.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14844
Test Plan: CI
Reviewed By: archang19, anand1976
Differential Revision: D108191370
Pulled By: xingbowang
fbshipit-source-id: 5b56099644cf918ea11071521875d0d059f3a69e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14842
### Problem
A process using RocksDB can hit an ASAN `heap-use-after-free` in `rocksdb::InstrumentedMutex::Lock()` when a `DBImpl` is closed while obsolete-file purge work is still in flight. This can happen during close storms with `avoid_unnecessary_blocking_io=true` on a shared Env, where a queued or late handoff purge can run after `~DBImpl` has destroyed `mutex_`.
### Root Cause
`CloseHelper()` has several mutex-unlocked windows late in shutdown. A dropped column-family handle or SuperVersion cleanup can run in one of those windows and start obsolete-file purge work. `FindObsoleteFiles()` marks that work by incrementing `pending_purge_obsolete_files_`, but `PurgeObsoleteFiles(..., true)` can then spend time outside the DB mutex before transferring the work to `bg_purge_scheduled_` via `SchedulePurge()`. During that handoff, `bg_purge_scheduled_` is still zero even though purge work is pending.
### Fix
Make the final `CloseHelper()` drain wait for both `pending_purge_obsolete_files_` and `bg_purge_scheduled_` before destroying `versions_` / `DBImpl`. This covers both the pending handoff and the scheduled `BGWorkPurge` state. Also move `TEST_VerifyNoObsoleteFilesCached()` and `table_cache_->EraseUnRefEntries()` after the final drain, so debug/ASAN table-cache verification runs only after close-time purge work has settled. The regression test parks a dropped-CF cleanup in the pending handoff window and verifies `CloseHelper()` blocks in the final drain until that handoff is released.
Reviewed By: xingbowang
Differential Revision: D107920881
fbshipit-source-id: f73cd79afe50fc30e0f5d14889dd4285bf350a58
Summary:
Use Compressor::GetRecommendedParallelThreads() overrides to disable parallel compression for Snappy, LZ4 (accelerated, not LZ4HC), and for accelerated levels of ZSTD (level < 0). These do not generally benefit from parallel compression.
Also add --verify_compression option to sst_dump for some basic "recompress" benchmarking with that option.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14841
Test Plan:
unit test updated.
In planning the scope of this change, I manually tested some production SST files with release build sst_dump --command=recompress and various settings for --compression_parallel_threads, --compression_types, and --compression_level, while I had the CPUs mostly busy using cache_bench in the background. Here's an example, before the change to override parallel_threads:
```
$ for PT in 1 8; do /usr/bin/time ./sst_dump --command=recompress --compression_parallel_threads=$PT --block_size=16384 --compression_types=kLZ4Compression --verify_compression=1 test.sst; done
...
Compression: kLZ4Compression Block Size: 16384 Threads: 1
Cx level: 32767 Cx size: 168501634 Uncx size: 791721664 Ratio: 4.698599 Write usec: 2345894 Read usec: 429022 Cx count: 48661 (100.0%) Not cx for ratio: 0 ( 0.0%) Not cx otherwise: 0 ( 0.0%)
2.54user 0.29system 0:02.84elapsed 99%CPU (0avgtext+0avgdata 460816maxresident)k
...
Compression: kLZ4Compression Block Size: 16384 Threads: 8
Cx level: 32767 Cx size: 168501634 Uncx size: 791721664 Ratio: 4.698599 Write usec: 2476459 Read usec: 439464 Cx count: 48661 (100.0%) Not cx for ratio: 0 ( 0.0%) Not cx otherwise: 0 ( 0.0%)
3.95user 0.33system 0:02.98elapsed 143%CPU (0avgtext+0avgdata 455504maxresident)k
```
Here as in many of the cases I'm changing, it actually takes longer to compress with parallel, despite the added parallel opportunity of verify_compression. And overall CPU is much higher from 2.54 `CPU*s` to 3.95 `CPU*s`.
The difference disappears with the change, because both use single-threaded SST construction.
Reviewed By: joshkang97
Differential Revision: D108075938
Pulled By: pdillinger
fbshipit-source-id: 5d1fc77ddbccf9f3b24a4f0b20b2b3c43074e89d
Summary:
Adds a new `DBOption use_direct_io_for_compaction_reads` (default false). When on, compaction-input SST files are opened with `O_DIRECT` so the sequential read-once data from compaction doesn't pollute the OS page cache and evict the hot user-read working set. User reads keep going through the buffered fast path. This protects user-read tail latency on write-heavy workloads without forcing user reads onto the existing global `use_direct_reads` knob (which pays in throughput and P50 — see the bench below).
The interesting bit is that just flipping the FileOptions returned by `FileSystem::OptimizeForCompactionTableRead` doesn't actually trigger `O_DIRECT` at the kernel level. The TableCache (and `FileMetaData::pinned_reader`) is already holding buffered handles opened at flush time or at `DB::Open` via `LoadTableHandlers`. When compaction asks for an iterator, it gets back the cached buffered handle and the kernel never sees the `O_DIRECT` flag.
So this PR also adds a small bypass path:
- `TableCache::FindTable` / `NewIterator` learn a `open_ephemeral_table_reader` mode. When set, the pinned-reader fast path and the shared cache are skipped, `GetTableReader` is called directly with the caller's FileOptions, and ownership of the freshly opened TableReader is handed back via a `unique_ptr`. The iterator takes ownership via `RegisterCleanup` and frees the reader on destruction.
- `VersionSet::MakeInputIterator` and `LevelIterator` plumb the flag through both L0 and L1+ compaction-input paths.
- `CompactionJob::ProcessKeyValueCompaction` turns the bypass on when `use_direct_io_for_compaction_reads` is set, the global `use_direct_reads` is off, and `OptimizeForCompactionTableRead` produced `use_direct_reads=true` in the compaction-read FileOptions.
The option is opt-in: when off, nothing changes for existing users. When on, only the compaction-input opens take the bypass path; user reads keep hitting the TableCache and the buffered fast path normally.
There's also a small db_bench helper in the same PR: a new `--bgwriter_num` flag that lets the writer thread in `readwhilewriting` (and the other "while writing" variants) spread its puts across `[0, bgwriter_num)` instead of `[0, num)`. Without this the readers and writer share a key range and you can't have both a hot read subset and meaningful compaction work — this lets you have both.
### Benchmark
Setup: Ubuntu 24.04 (kernel 7.0.5, OrbStack Linux VM on Apple Silicon), 14 vCPUs, virtio-blk disk, btrfs. MGLRU disabled (`echo 0 > /sys/kernel/mm/lru_gen/enabled`) so the kernel uses the classic active/inactive LRU. 14 GB DB (3.5M keys × 4 KB values), no compression. Each measurement run is pinned to a 1 GB cgroup via `systemd-run --scope -p MemoryMax=1G -p MemorySwapMax=0`. Page cache is dropped between configs. db_bench is Release build.
Workload: `readwhilewriting` for 120s. 4 reader threads doing random reads over a hot key subset, plus 1 writer thread spreading overwrites across the full 3.5M-key keyspace (via `--bgwriter_num=3500000`) throttled at 200 MB/s, so there's continuous compaction running while the readers go.
The size of the hot reader subset relative to available page cache controls how visible the optimization is. The Cassandra blog ([Lightfoot 2026](https://lightfoot.dev/direct-i-o-for-cassandra-compaction-cutting-p99-read-latency-by-5x/)) documented the same thing: biggest wins when the hot set is big enough to actually compete for cache, smaller wins when the hot set trivially fits, neutral when the hot set is way bigger than cache. So I ran two hot-set sizes.
#### Small hot set: ~30 MB (~3% of the 1 GB cgroup) — N=5 iterations, mean (CV)
`--num=7500`. The hot set is small enough that the page cache holds it without much trouble even under compaction, so the wins here are real but on the modest side.
| Config | Throughput (ops/s) | Read P50 (µs) | Read P99 (µs) | Read P99.9 (µs) | Read P99.99 (µs) |
|---|---|---|---|---|---|
| buffered (default) | 233,477 (8.2%) | 16.09 | 82.24 | 721.0 | 2,102.5 |
| direct_compaction_writes_only (existing knob alone) | 287,405 (2.8%) — **+23.1%** | 13.00 (−19.2%) | **66.77 (−18.8%)** | 553.9 (−23.2%) | 1,787.6 (−15.0%) |
| direct_compaction_read_only (new knob alone) | 250,669 (2.4%) — +7.4% | 14.16 (−12.0%) | 102.99 (+25.2%) | 689.8 (−4.3%) | 1,801.3 (−14.3%) |
| direct_compaction_read_write (new + existing, recommended) | 277,920 (3.3%) — **+19.0%** | **12.99 (−19.3%)** | 84.23 (+2.4%) | 613.4 (−14.9%) | **1,738.2 (−17.3%)** |
| use_direct_reads=true (existing global) + write-side | 249,014 (2.5%) — +6.7% | 15.95 (−0.9%) | 68.78 (−16.4%) | **450.8 (−37.5%)** | 1,814.5 (−13.7%) |
CV is 2.4–3.3% on the optimized configs (8.2% on buffered), so the deltas are real. With a hot set this small, the existing `use_direct_io_for_flush_and_compaction` knob is already doing most of the work — the new flag's main extra contribution here is P99.99 (combined wins it by ~2 points vs writes-only-alone). Worth noting: the new flag *alone* (without the existing write-side flag) improves P99.99 but regresses P99 by 25% on this small-hot-set workload, because direct compaction reads lose kernel readahead and compaction-output writes are still hitting the page cache. That regression goes away once you combine with the existing write-side flag, or once the hot set is bigger (see next table). So if you're using just one knob, use the existing one. If you're using this PR's flag, pair it with `use_direct_io_for_flush_and_compaction=true`.
#### Larger hot set: ~400 MB (~40% of cache) — N=5 iterations, mean (CV)
`--num=100000`. This is the case the Cassandra blog calls out — hot set big enough to actually fight compaction for cache. Their analogous setup (1M hot partitions, ~33% hot/cache) reported 1.93× p99 improvement. Numbers here are the headline:
| Config | Throughput (ops/s) | Read P50 (µs) | Read P99 (µs) | Read P99.9 (µs) | Read P99.99 (µs) |
|---|---|---|---|---|---|
| buffered (default) | 68,959 (7.7%) | 44.81 | 541.22 | 2,225.2 | 11,334.5 |
| direct_compaction_writes_only (existing knob alone) | 73,973 (10.3%) — +7.3% | 42.22 (−5.8%) | 456.27 (−15.7%) | 2,016.9 (−9.4%) | 9,190.0 (−18.9%) |
| direct_compaction_read_only (new knob alone) | 84,337 (2.3%) — +22.3% | 38.66 (−13.7%) | 386.97 (−28.5%) | 1,644.8 (−26.1%) | 4,837.9 (−57.3%, 2.34×) |
| direct_compaction_read_write (new + existing, recommended) | **104,923 (8.4%) — +52.2%** | **34.26 (−23.5%)** | **290.97 (−46.2%)** | **1,143.4 (−48.6%)** | **3,080.3 (−72.8%, 3.68×)** |
| use_direct_reads=true (existing global) + write-side | 71,598 (9.1%) — +3.8% | 51.33 (+14.5%) | 297.91 (−45.0%) | 1,663.6 (−25.2%) | 6,530.0 (−42.4%) |
Combined config gets a 3.68× p99.99 win, 1.86× p99, p50 down 23%, throughput up 52%. Same shape as the Cassandra blog's 1.93× p99 result — the improvement just lands at deeper percentiles for us because RocksDB's baseline data path is roughly 40× faster than Cassandra's (their buffered p99 was 35 ms, ours is 0.54 ms), so the cache-miss tail is further out.
A few things worth calling out from this table:
- The new flag is doing real work on top of the existing write-side flag here, not just shifting things around. Combined throughput is +42% over `direct_compaction_writes_only` alone, and combined p99.99 is 3× better. The existing knob alone gives a fairly modest +7% throughput / -19% p99.99 in this case — there's a clear gap that the new flag fills.
- The new flag *alone* (no existing write-side flag) is also a real improvement here: +22% throughput, p99.99 down 57%. The P99 regression we saw in the small-hot-set case is gone, because the cache-protection effect now dominates the lost-readahead cost.
- `use_direct_reads=true` (the existing global flag) actually regresses P50 by 14.5% in this workload — taking user reads off the page cache hurts you when the hot data could have been cached. It also gets the worst throughput of any direct config. It's not an equivalent way to get these gains.
### `compaction_readahead_size` matters when this flag is on
Direct I/O bypasses kernel readahead, so RocksDB's own `DBOptions::compaction_readahead_size` becomes the only prefetch the iterator has. The default of 2 MB is enough and real users will get it automatically. **But `db_bench`'s `--compaction_readahead_size` CLI default is 0**, which defeats prefetch and makes direct compaction look slower than it actually is. If you're reproducing the numbers above, pass `--compaction_readahead_size=2097152` (or larger).
- Recommended production config is `use_direct_io_for_compaction_reads=true` + `use_direct_io_for_flush_and_compaction=true`. Strongest configuration at every percentile and throughput in both benches.
- The new flag is the read-side counterpart to `use_direct_io_for_flush_and_compaction`, which handles compaction-write cache pollution. They address different sources of pollution and compose. The gap between "combined" and "writes-only-alone" is 17 percentage points on p99.99 in the small-hot-set bench and 54 points in the larger one, so the new flag is contributing real value, especially as the hot set grows.
- The new flag alone is also a real improvement when the hot set is big enough to compete with cache (+22% throughput, 2.34× p99.99 in the larger-hot-set bench). On a very small hot set it improves p99.99 but regresses p99, so pairing with the existing write-side flag is safer.
- The benefit is workload-dependent. Small hot sets get modest tail-latency wins. Hot sets sized to actually compete for cache get the big multi-percentile wins shown above. Hot sets bigger than cache (not benched here but covered in the Cassandra blog) see no change either way — every read misses regardless.
### Reproducing
Any Linux host (or a Linux VM on macOS via OrbStack / Multipass / lima):
```bash
sudo apt-get install -y build-essential clang cmake git pkg-config \
libgflags-dev libsnappy-dev zlib1g-dev libbz2-dev liblz4-dev libzstd-dev
cmake -DCMAKE_BUILD_TYPE=Release -DPORTABLE=1 -DWITH_GFLAGS=1 -DWITH_TESTS=0 ..
make -j db_bench
echo 0 | sudo tee /sys/kernel/mm/lru_gen/enabled
```
Build the source DB once, unrestricted memory:
```bash
./db_bench --benchmarks=fillrandom,compact,waitforcompaction,stats \
--db=/path/to/source_db --num=3500000 --key_size=16 --value_size=4096 \
--write_buffer_size=16777216 --target_file_size_base=16777216 \
--max_background_jobs=4 --compression_type=none --cache_size=4194304 \
--max_bytes_for_level_base=67108864 --disable_wal=1 --sync=0
```
For each config, copy `source_db -> scratch_db`, run `sync && echo 3 > /proc/sys/vm/drop_caches`, then:
```bash
sudo systemd-run --scope -p MemoryMax=1G -p MemorySwapMax=0 \
./db_bench --use_existing_db=1 \
--benchmarks=readwhilewriting,stats --db=/path/to/scratch_db \
--threads=5 --duration=120 --statistics=true --histogram=1 \
--num=7500 --bgwriter_num=3500000 \
--key_size=16 --value_size=4096 \
--write_buffer_size=16777216 --target_file_size_base=16777216 \
--max_background_jobs=4 --compression_type=none \
--cache_size=4194304 --open_files=200 \
--skip_stats_update_on_db_open=true \
--max_bytes_for_level_base=67108864 \
--benchmark_write_rate_limit=209715200 \
--compaction_readahead_size=2097152 \
--rate_limiter_bytes_per_sec=0 \
--use_direct_reads={true|false} \
--use_direct_io_for_compaction_reads={true|false} \
--use_direct_io_for_flush_and_compaction={true|false}
```
For the larger hot-set table, change `--num=7500` to `--num=100000`.
The five configs in the tables:
- `buffered`: all three flags false.
- `direct_compaction_writes_only`: `use_direct_io_for_flush_and_compaction=true`, the other two false. This is what users have today without this PR.
- `direct_compaction_read_only`: `use_direct_io_for_compaction_reads=true`, the other two false.
- `direct_compaction_read_write`: `use_direct_io_for_compaction_reads=true`, `use_direct_io_for_flush_and_compaction=true`, `use_direct_reads=false`. **Recommended.**
- `direct_all`: `use_direct_reads=true`, `use_direct_io_for_flush_and_compaction=true`, `use_direct_io_for_compaction_reads=false`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14743
Reviewed By: pdillinger
Differential Revision: D108017601
Pulled By: xingbowang
fbshipit-source-id: 4039d490d7e77b476db7a477a2f3d24738db6336
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14838
Remote compaction can pre-filter whole input files when universal compaction sees standalone range tombstones. In that case the worker's iterator count no longer describes the full original input set, so returning `has_accurate_num_input_records=true` can trigger a false-positive input-record verification on the primary. Mark the remote count inaccurate whenever the reconstructed compaction filtered input files before iteration, and add a regression test covering that path.
Reviewed By: jaykorean
Differential Revision: D107961493
fbshipit-source-id: 0c302081964ab1d894b34bddd2c55f5a4f06752f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14836
Adds a public `Statistics` histogram, `INGEST_EXTERNAL_FILE_TIME` (`"rocksdb.ingest.external.file.micros"`), recording the end-to-end latency in microseconds of each `IngestExternalFile(s)` call. Ingestion timing was previously only available through per-thread `perf_context` counters, which require setting a `PerfLevel` and are not aggregated, so there was no process-wide latency distribution (p50/p99/max) for dashboards.
It is recorded with an RAII `StopWatch` at the top of `DBImpl::IngestExternalFiles` -- one sample per call (not per column family), covering all return paths. It is null-safe and self-gating on the stats level, so there is no cost when statistics are off, and ingestion is not a hot path. Java bindings are kept in sync per the `statistics.h` requirement; the C API needs no change.
Reviewed By: xingbowang
Differential Revision: D107721260
fbshipit-source-id: 0705f9e0f7392329a7bcbdbd9f3afd34594d20bb
Summary:
Add read-scoped block buffers for scan reads. Introduce an experimental read-scoped block buffer provider API, configured through ReadOptions::read_scoped_block_buffer_provider, so supported block-based table iterator scans and MultiScan data-block reads can use caller-provided read-scoped storage for final data-block contents.
When configured, supported provider-backed scan data-block reads bypass the data-block cache while preserving normal index/filter block-cache behavior. Known-uncompressed reads can attach provider cleanup to provider-backed read buffers without copying. Compressed reads decompress directly into provider-backed output, while maybe-compressed reads that turn out to be uncompressed copy once into provider-backed final contents. mmap reads ignore the provider.
Extend AlignedBuffer and RandomAccessFileReader direct-I/O paths to support external aligned allocations, then use that support for read-scoped iterator, async I/O, and MultiRead scratch buffers. Centralize read-scoped I/O policy, keep coalesced async reads safe when blocks are released before completion, and validate provider lease contracts.
Add focused coverage for read-scoped ownership, compressed and uncompressed blocks, direct I/O, data-block cache bypass behavior, invalid provider leases, async release handling, and stress-test provider invariants. Add public API release notes for read-scoped block buffers.
Bonus change: Fixed a flaky test in ReserveThread
## Testing
- CI
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14806
Reviewed By: anand1976
Differential Revision: D106999951
Pulled By: xingbowang
fbshipit-source-id: b1f23d4bab6318b6373ba2ca99a5c4d6a842dc5a
Summary:
Extends RocksDB's C API with two `BackupEngine` capabilities needed by
language bindings (e.g. Rust via librocksdb-sys) that consume the C API:
- **StopBackup**: Add `rocksdb_backup_engine_stop_backup()` to allow cancelling
an in-progress backup.
- **Rate limiters**: Add
`rocksdb_backup_engine_options_set_backup_rate_limiter()` and
`rocksdb_backup_engine_options_set_restore_rate_limiter()` to expose the
`shared_ptr<RateLimiter>` fields on `BackupEngineOptions`. The existing
`uint64_t` setters only throttle writes; these expose the richer `RateLimiter`
object that supports read+write throttling (e.g. `kAllIo` mode).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14722
Test Plan:
- [x] New tests in `db/c_test.c` cover `StopBackup` and rate limiter
setter/getter roundtrips, plus opening a real backup engine with rate
limiters set and running a backup end-to-end
- [x] `make check` passes with no regressions
Reviewed By: pdillinger
Differential Revision: D107654882
Pulled By: xingbowang
fbshipit-source-id: f50c3989779e6a099113fec203231d47b9480cb9
Summary:
`TransactionDB` supports different write policies (e.g. `WritePrepared`), but this functionality is not currently accessible via the C API. Creating a C shim to expose the functionality for setting the write policy.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14810
Reviewed By: pdillinger
Differential Revision: D107654915
Pulled By: xingbowang
fbshipit-source-id: b0d915a3420057de5236fe9f6cb47d291294788c
Summary:
PR https://github.com/facebook/rocksdb/issues/14585 added FileOpenContract enforcement to FaultInjectionTestFS. The no-reopen-for-write check recorded contracts by path but did not clear a stale contract after the file was deleted. A later SST create that reused the same path could be rejected as a forbidden reopen, causing DBWALTest.WALWithChecksumHandoff to fail with "NewWritableFile violates no-reopen-for-write contract". If the ASSERT exited the test early, the local Env stack object could also be destroyed before DB fixture teardown closed the DB, producing the follow-on TSAN heap-use-after-free.
When opening a file for write, drop stale no-reopen tracking if the target file no longer exists, while still rejecting writes through reopened handles. Also close the DB before the WAL test's local Env unwinds on assertion failure.
A separate TSAN report in DBBlobBasicIOErrorTest.GetEntityMergeWithBlobBaseIOError exposed unsynchronized access to FaultInjectionTestEnv::error_: SetFilesystemActive() updated the stored error under mutex_, while background purge work read it through GetError() without that mutex. Copy injected errors under the same mutex and read the no-space state in GetFreeSpace() while protected. Apply the same synchronization to FaultInjectionTestFS.
Bonus fix: Fix another TSAN data race in FaultInjectionTestEnv GetError not synchronized under mutex_
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14822
Test Plan: CI
Reviewed By: mszeszko-meta
Differential Revision: D107689662
Pulled By: xingbowang
fbshipit-source-id: 6d4f8fdc8b898d3ddcad7816e385f3b7f20c4727
Summary:
- Rename the misspelled `memtable_veirfy_per_key_checksum_on_seek` option and related flags/config keys to `memtable_verify_per_key_checksum_on_seek`.
- Update memtable option plumbing, options serialization/logging, db_bench/db_stress/crash-test flags, tests, and the option-addition guide to use the corrected name.
- Keep the checksum-on-seek behavior unchanged.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14811
Test Plan: CI
Reviewed By: pdillinger
Differential Revision: D107382276
Pulled By: xingbowang
fbshipit-source-id: 7621dc718d61503b982a7e3f65cc9293a1ad085b
Summary:
- Add a contract-boundaries section to the RocksDB code review checklist in `CLAUDE.md`.
- Document common review prompts for keeping caller-specific policy out of reusable lower layers.
- Add "Contract Boundary Leaks" to the common review feedback patterns.
## Testing
CI
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14821
Reviewed By: pdillinger
Differential Revision: D107652555
Pulled By: xingbowang
fbshipit-source-id: 383bd78fec6e3ecc251c752d30f9de3ea44a10de
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14585
Adds typed file-open contracts and open-file size handling needed for blob direct-write files on remote or SHM-backed file systems.
This diff:
- Adds typed `FileOpenContract` semantics to RocksDB `FileOptions`, including `kNoReopenForWrite`, `kNoReadersWhileOpenForWrite`, bitwise helpers, and constructor propagation from `EnvOptions` to `FileOptions`.
- Marks SST/table output creation paths with both `kNoReopenForWrite` and `kNoReadersWhileOpenForWrite`, including builder output, compaction output, and external `SstFileWriter` output.
- Marks blob write paths with the appropriate contracts: regular blob file builders use both `kNoReopenForWrite` and `kNoReadersWhileOpenForWrite`, while blob direct-write partition output uses `kNoReopenForWrite` so active readers can still observe SHM-backed data through the read path.
- Adds `GetFileSizeFromOpenFileOrPath` so readers can prefer an already-open `FSRandomAccessFile::GetFileSize` result and fall back to path-level `FileSystem::GetFileSize` according to the caller's fallback policy.
- Updates blob file reader and table footer reader size checks to use the new open-file-or-path size helper, preserving malformed-file validation while supporting file systems where path-level size is not the only valid source.
- Extends `FaultInjectionTestFS` to track file-open contracts, reject readers while a no-readers writer is open, reject reopened data writes for `kNoReopenForWrite`, allow `SyncFile` to use a reopened handle, and preserve contract state across create, reopen, reuse, rename, link, reset, and delete paths.
- Keeps `IOStatus` return paths simple by returning status locals directly in the file-size helper and fault-injection contract validation paths.
- Adds clang-format-compliant tests covering file-open contract enforcement, `SyncFile` being allowed while reopened data writes are rejected, blob direct-write behavior, blob reader behavior, and the new file-size fallback helper behavior.
- Leaves WAL and MANIFEST contract assignment for follow-up work because live WAL/MANIFEST readers and `reuse_manifest_on_open` need separate handling before RocksDB can safely claim stronger file-open contracts for those file classes.
Reviewed By: anand1976, pdillinger
Differential Revision: D100002686
fbshipit-source-id: 26b4021bc13333ede1077f0ff8cbd71335c5f294
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14739
Adds public path-level file sync APIs to RocksDB so callers can ask `Env` or `FileSystem` to sync a named file without hand-rolled reopen logic.
This diff:
- Adds public `Env::SyncFile` and `FileSystem::SyncFile` APIs for syncing or fsyncing a file by name without requiring callers to reopen it directly.
- Documents the `SyncFile` contract: filesystems may override it as a no-op when flush/close already provides durability, and RocksDB callers should use it instead of hand-rolled `ReopenWritableFile()+Sync()/Fsync()` so filesystems can reject post-close data-write reopen paths.
- Implements the default `Env` and `FileSystem` behavior by reopening the file as writable, calling `Sync` or `Fsync`, closing the handle, returning the sync/fsync error ahead of a close error when both fail, and explicitly consuming the close status on the sync-failure path.
- Wires `SyncFile` through the Env/FileSystem bridge and wrapper layers, including `EnvWrapper`, `FileSystemWrapper`, `CompositeEnvWrapper`, `EncryptedFileSystem`, `RemapFileSystem`, read-only file systems, mock file systems, and fault-injection wrappers.
- Keeps fault-injection `SyncFile` overrides on the base default implementation so the operation still exercises the wrapper's `ReopenWritableFile`, wrapped-file `Sync`/`Fsync`, and `Close` hooks instead of bypassing them through target forwarding.
- Updates external SST ingestion to call `FileSystem::SyncFile` instead of manually reopening an ingested file as writable, while preserving the `NotSupported` skip behavior and failure logging.
- Adds release-note coverage for the new public API and unit coverage for default `Env`/`FileSystem` success, reopen failure, close failure, sync-versus-close failure precedence, and external SST sync behavior.
There was an earlier version of this API which got reverted (PR #13987) due to internal change broken. Re-apply the change. The internal issue will be resolved in next release.
Reviewed By: pdillinger
Differential Revision: D104918547
fbshipit-source-id: 3f8d2d127fc962b68b423bbb90de551fe6706224
Summary:
The historical default block compression `kSnappyCompression` dates to when Snappy was the obvious fast/cheap choice. On modern server CPUs LZ4 matches or beats Snappy on compression ratio while decompressing substantially cheaper, so it is a better default. This changes the default `ColumnFamilyOptions::compression` to `kLZ4Compression`, with a runtime fallback of LZ4 -> Snappy -> none depending on what is compiled into the binary (new `GetDefaultCompressionType()` in util/compression.h). Only column families that do not explicitly set `compression` are affected (including compaction output when
`CompactionOptions::compression == kDisableCompressionOption`), and only newly written SST files; existing data is read as before. Doc comments, the Java bindings, and the sorted_run_builder example are updated to describe the new default.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14818
Test Plan:
Adjusted two unit tests that implicitly depended on the old Snappy default to pin `compression = kSnappyCompression`: db_iterator_test's ReadAhead (readahead byte thresholds assume Snappy-sized files) and compaction_service_test's CustomFileChecksum (LSM shape after auto-compaction determined whether a manual CompactRange had work to do).
This change is primarily validated by performance testing. Added a `compressreject` db_bench benchmark (output buffer sized just below the predicted compressed size) to measure the cost of attempting then declining compression, alongside `compress`/`uncompress`. Both db_bench and sst_dump compress/decompress a modest block at a time, as in a real workload.
sst_dump on SST files from production workloads (4 files, 16KB blocks, single thread) on recent server-class AMD, Intel, and ARM CPUs, LZ4 vs Snappy:
- Compression ratio: comparable; LZ4 is slightly smaller on the more compressible files (up to ~8%) and within ~2% on the rest.
- Compression (write) CPU: a wash, within ~2% either direction.
- Decompression (read) CPU: the clear win -- Snappy costs ~1.2x-1.5x as much as LZ4, i.e. LZ4 saves ~25-30% read CPU, consistently across AMD, Intel, and ARM.
db_bench synthetic workload (100-byte values), at 1 / 12 / 160 threads, LZ4 vs Snappy:
- Compression throughput: LZ4 ~10-30% higher.
- Decompression throughput: LZ4 much higher, and the advantage grows with core count -- from ~+12% at 12 threads to ~+45-50% at 160 threads, i.e. better multi-core scaling.
- Rejection (insufficient ratio) path: comparable; LZ4 ~15-18% faster on compressible-but-rejected blocks and Snappy within ~10% on barely-compressible blocks. No meaningful regression, confirming incompressible data is still efficiently detected and stored raw.
Reviewed By: xingbowang
Differential Revision: D107536490
Pulled By: pdillinger
fbshipit-source-id: f8abaee630d782674778338148e36ed0c84e3661
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14798
Add cross-version format compatibility testing for remote compaction to `check_format_compatible.sh` as primary and worker RocksDB instances can run different versions.
Two new ldb commands coordinate via local files:
- `remote_compaction_primary`: opens an existing DB and runs `CompactRange()` through `LocalFileCompactionService`, which writes `input.bin` via `Schedule()` and polls for `result.bin` via `Wait()`.
- `remote_compaction_worker`: polls for `input.bin`, calls `OpenAndCompact()`, writes `result.bin`.
The test script creates a DB using `generate_random_db.sh` with the primary's ldb binary (new optional 3rd argument) so the OPTIONS file matches the primary's version. An overlap key is written to ensure `CompactRange` triggers a real compaction (not a trivial move). For each old ref in `db_forward_with_options_refs`, the script tests both directions -- current primary + old worker and old primary + current worker -- to catch wire-format incompatibilities in `CompactionServiceInput`/`CompactionServiceResult`. Old refs lacking the commands are skipped gracefully.
Reviewed By: pdillinger
Differential Revision: D106321150
fbshipit-source-id: e0341b57c1b12e1fa5296609f0463f77484c1a6e
Summary:
When searching/grepping through unit tests, it's convenient to use test.cc suffix to match all unit tests, or to exclude test.cc when excluding unit tests from a code search. (I've even seen my AI assistant grep through `*test.cc`.) So I'm renaming db_test2.cc (as planned in https://github.com/facebook/rocksdb/issues/14076)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14218
Test Plan: existing tests + CI
Reviewed By: xingbowang
Differential Revision: D90140624
Pulled By: pdillinger
fbshipit-source-id: 78aa099290670bbb4093b2c7c02bc47ab3bf5f8e
Summary:
Fix a false-positive compaction corruption error in the remote compaction path. Remote workers can mark `CompactionJobStats::num_input_records` as unreliable when input iteration uses seek/skip behavior, but the remote result serialization was dropping `has_accurate_num_input_records`. The primary then deserialized the flag as its default `true`, trusted an unreliable zero input-record count, and raised `Compaction number of input keys does not match number of keys processed`.
This change serializes the accuracy flag with the rest of `CompactionJobStats` so the primary preserves the remote worker's "do not verify this count" signal. It also adds a targeted regression test and a release note for the bug fix.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14808
Test Plan: New regression test that triggers corruption error without the fix by modifying `has_accurate_num_input_records=false`
Reviewed By: xingbowang
Differential Revision: D107128357
Pulled By: joshkang97
fbshipit-source-id: 3243c9c2ab18534652737f7410fe3c51724db549
Summary:
Async WAL precreation in https://github.com/facebook/rocksdb/pull/14738 / D105020559 was motivated by slow file creation time on remote storage. MANIFEST does not need the same precreation treatment as WAL because most MANIFEST writes come from background flush and compaction work, but user-facing metadata operations can still pay MANIFEST rotation file creation latency inline. File ingestion performance is a particular concern to some Meta users.
Relax the effective MANIFEST rotation limit by 25% for MANIFEST write batches containing any foreground VersionEdit, while keeping background-only flush/compaction batches on the configured or auto-tuned limit. This covers column family manipulation, external file ingestion and import, and DeleteFilesInRange(s). SetOptions remains expected to avoid MANIFEST writes; the test keeps a regression guard for that behavior.
The relaxation is intentionally bounded. It reduces the chance that foreground metadata operations create a new MANIFEST inline, while still allowing foreground operations to rotate once the current MANIFEST is beyond the relaxed threshold. Heavier blocking operations like manual Flush or CompactRange already trigger additional file creation and do not get this treatment here, though that could be reconsidered later.
This should reduce a potential latency hazard of manifest file size auto-tuning: more frequent MANIFEST rotations. With this change, rotation latency is shifted toward background-only MANIFEST batches when possible.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14797
Test Plan:
Expanded DBEtc3Test.AutoTuneManifestSize to cover the foreground threshold behavior and the original auto-tuning behavior in separate phases:
- verifies foreground-only CreateColumnFamily writes get only bounded 25% headroom by asserting the first four large-CF additions do not rotate and the fifth does;
- verifies auto-tuned background thresholds still prevent excessive rotation;
- verifies foreground operations stay below the relaxed threshold for CreateColumnFamily, IngestExternalFile, CreateColumnFamilyWithImport, and DeleteFilesInRanges;
- verifies SetOptions still does not write to MANIFEST;
- verifies a following background flush still rotates at the normal threshold;
- preserves the persisted compacted manifest size close/reopen coverage.
Reviewed By: xingbowang
Differential Revision: D106578771
Pulled By: pdillinger
fbshipit-source-id: f8e274032cd9e7f50e95b685c949242f95351498
Summary:
StringToMap previously stripped the outer braces from nested values,
making single-entry round-trip via `key=value;` (e.g. SetOptions)
silently corrupt values that contain ';'. For example, a filter_policy
with sub-options serialized as
filter_policy={id=ribbonfilter:10:-1;bloom_before_level=-1;}
parsed to map[filter_policy] = "id=ribbonfilter:10:-1;bloom_before_level=-1;",
and embedding that map entry directly into another `key=value;` string
re-exposed the inner ';' as a top-level delimiter.
This change is a step toward the "essential view": the outer `{...}` of
a nested value is part of the value's identity. StringToMap preserves it, so
each map entry is in self-contained form (a simple value or a single
balanced `{...}` block) and can be embedded directly without further
escaping by the caller. However some permissiveness is kept for
compatibility: list-style typed parsers (ParseVector, ParseArray,
kStringMap, the listener parser) and scalar dispatch in
OptionTypeInfo::Parse accept either the bare or the wrapped form via a
new OptionTypeInfo::StripOuterBraces helper that peels at most one
outer-pair-matched layer. So braced scalar input like `key={42};` and
`key={true};` still parses, and `key={};` denotes an empty list,
distinct from `key={{}};` which denotes a one-element list with an
empty element.
A new public MapToString is added to convenience.h as the symmetric
inverse of StringToMap: a trivial `key=value;` joiner that relies on
each value already being self-contained (which StringToMap guarantees).
Also fixed and refactored our trim() function because it would do the wrong
thing for a single space. This problem was detected by expanding the
StringToMapRandomTest based on code review feedback, and should only
improve existing callers to trim().
The OPTIONS-file serializer code is untouched, so the on-disk format
is byte-for-byte identical and format-compatibility is preserved.
Bonus: fixes / clarifications to CLAUDE.md's build-system note.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14805
Test Plan:
db_bloom_filter_test extends MutableFilterPolicy with the OPTIONS-file
format for filter_policy to confirm that form round-trips through
SetOptions.
options_test updates StringToMapTest expectations for the new
brace-preserving behavior, and adds:
- MapToStringTest covering the new joiner.
- StringToMapMapToStringRoundTripTest including a single-entry
pull-and-embed scenario that previously corrupted on round-trip.
- FullOptionsStringToMapRoundTripTest that drives a populated
DBOptions and CFOptions through GetStringFromXxx -> StringToMap ->
per-entry single round-trip -> MapToString -> GetXxxFromString ->
VerifyXxxOptions, catching any custom serializer that emits a
value with ';' without enclosing it in '{}'.
- EmptyBracedVectorAndBracedScalarTest covering `key={};` (empty
list) and braced scalar input like `key={42};`, `key={true};`.
- Expanded StringToMapRandomTest with round-tripping.
Some explicit trim() tests added to string_util_test.cc
Format compatibility verified with
SHORT_TEST=1 tools/check_format_compatible.sh
Other existing tests in options_test, customizable_test, options_settable_test,
listener_test, options_file_test, options_util_test, db_options_test,
and compaction_service_test have extensive coverage of serialization /
deserialization logic.
Reviewed By: hx235, xingbowang
Differential Revision: D106855466
Pulled By: pdillinger
fbshipit-source-id: 872aac8d819c7b90c807b92bab85569b48fa2aa0
Summary:
- AbortAllCompactions can race with an already-scheduled automatic compaction before the background worker calls PickCompactionFromQueue(). MaybeScheduleFlushOrCompaction() has already consumed one unscheduled_compactions_ credit when it scheduled the worker, but the CF is still present in compaction_queue_ with queued_for_compaction=true. If the worker returns kCompactionAborted before popping the CF, ResumeAllCompactions() can see unscheduled_compactions_ == 0 and fail to schedule the still-queued work, leaving compaction permanently stalled until DB restart.
- Restore the unscheduled compaction credit in the non-prepicked abort path, matching the existing BG-work-stopped handling for the same scheduled-before-pick state. Prepicked/manual compactions are not adjusted because they do not represent an unpopped automatic compaction_queue_ entry whose scheduling credit was consumed.
- Add AbortScheduledAutomaticCompactionBeforePick to deterministically reproduce the lost-credit race with a sync point at BackgroundCallCompaction:0. The test verifies that ResumeAllCompactions() schedules the still-queued automatic compaction by checking COMPACT_WRITE_BYTES advances and L0 file count drops.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14800
Test Plan: - Without the implementation fix, `DBCompactionAbortTest.AbortScheduledAutomaticCompactionBeforePick` fails because `COMPACT_WRITE_BYTES` remains unchanged after `ResumeAllCompactions()`.
Reviewed By: pdillinger
Differential Revision: D106684155
Pulled By: xingbowang
fbshipit-source-id: 6dacea473ef481eb3122eb15e296e5b69635413f
Summary:
- Propagate validated, sorted MultiScan range state from `DBIter::Prepare()` through `MultiScanArgs`.
- Mark block-based table IO jobs as already sorted when the public MultiScan ranges have been validated.
- Keep `IODispatcher::SubmitJob()` as the normalization boundary for unsorted callers, while allowing sorted callers to skip the defensive block-handle sort.
- Update private dispatcher coalescing helpers to consume sorted block indices and add debug assertions for that precondition.
## Testing
CI, new unit test
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14783
Reviewed By: anand1976
Differential Revision: D106301516
Pulled By: xingbowang
fbshipit-source-id: 99b7ffcaecbf27cb79f15feb4af8680ff1e422d9
Summary:
Fixes an off-by-one bug in how sequence numbers are handled when splitting range tombstones across output levels in per-key-placement (tiered) compaction. The bug was either introduced or propagated in https://github.com/facebook/rocksdb/issues/13256.
Point entries move to proximal output only when their sequence number is strictly greater than `proximal_after_seqno_`, but range tombstones were previously split using an inclusive lower bound at that same seqno. That allowed a range tombstone at the boundary to be emitted to the proximal level while point keys at the same seqno stayed in the last level, which could create overlapping files in the proximal level (caught by `force_consistency_checks` as `L<n> has overlapping ranges`).
This matters when `proximal_output_range_type_` is `kNonLastRange`: the compaction only owns the selected proximal-level input range, so existing last-level data at the split boundary must stay in the last level. In `kFullRange`, the compaction owns the relevant proximal-level range, so newer last-level data can be safely emitted to proximal output.
The fix splits range tombstones at `proximal_after_seqno_ + 1` (saturating at `kMaxSequenceNumber`), so the half-open `[lower, upper)` tombstone filter lands on the same boundary as the strict `seqno > proximal_after_seqno_` rule used for point keys.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14795
Test Plan:
New regression test `PrecludeLastLevelTestBase.RangeDelAtProximalSeqnoBoundaryStaysInLastLevel` uses a targeted manual compaction to exercise the `kNonLastRange` case directly, verifying a boundary range tombstone stays in the last level instead of widening the proximal output range.
Worked example:
```text
Initial state:
L6 (last level): Put(Key 2)s1, Put(Key 12)s2, RangeDel[Key 2, Key 12)s3
L5 (proximal): file A [Key 0 .. Key 4]s4-s5 file B [Key 5 .. Key 9]s6-s7
preclude_last_level_min_seqno is forced to 0 via sync point.
Manual CompactFiles selects only L5 file B + the L6 file (output level 6).
Only part of the proximal level is selected, so this is the kNonLastRange case:
max_last_level_seqno = 3
proximal_after_seqno_ = max(0, 3) = 3
Point keys Key 5 (s6) and Key 9 (s7): both seqno > 3 -> proximal output (L5) OK
Range tombstone s3:
Old (buggy): proximal keep range [3, MAX) includes s3, so the tombstone is
emitted to the proximal output. The output is built from L5 input
file B [Key 5 .. Key 9], but the tombstone covers [Key 2, Key 12),
so the proximal output file starts at Key 2 and spills past Key 5
into the existing, untouched L5 file A [Key 0 .. Key 4].
Two overlapping files in L5 -> Corruption.
New (fixed): split at proximal_after_seqno_ + 1 = 4.
proximal keep [4, MAX) excludes s3; last-level keep [0, 4) includes
s3 -> tombstone stays in the last level (L6). OK
```
Verification (debug build, `make -j64 tiered_compaction_test`):
- Without the fix, the test fails at the `CompactFiles` call:
```
tiered_compaction_test.cc:2940: Failure
Corruption: force_consistency_checks(DEBUG): VersionBuilder: L5 has overlapping ranges:
file https://github.com/facebook/rocksdb/issues/11 largest key: Key(4) seq:5, type:1 (Put) vs.
file https://github.com/facebook/rocksdb/issues/17 smallest key: Key(2) seq:3, type:15 (range deletion)
```
- With the fix, the test passes (range deletions absent from L5, still present in L6).
Reviewed By: pdillinger
Differential Revision: D106528471
Pulled By: joshkang97
fbshipit-source-id: a5f99b426d3a7a6253bc1972cf8cb60d1cb85089
Summary:
- Use `IndexValue::first_internal_key` from `kBinarySearchWithFirstKey` index entries to decide whether the final bounded MultiScan candidate block starts at or beyond the scan limit.
- Skip that block when its first user key compares greater than or equal to the range limit with `CompareWithoutTimestamp`.
- Preserve existing conservative behavior for unbounded ranges, index entries without first-key metadata, and normal `kBinarySearch` indexes.
- Add parameterized coverage for boundary limits, in-block limits, bytewise and reverse comparators, and stripped/persisted user-defined timestamp modes.
## Testing
CI
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14784
Reviewed By: pdillinger
Differential Revision: D106302205
Pulled By: xingbowang
fbshipit-source-id: 1acebdf48bf7c18d35a781ca41c7bfd5c4ab8f47
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14799
When `CompactionService::Wait()` returns `kSuccess` but `CompactionServiceResult::Read()` fails before the primary renames any remote output file from `CompactionServiceResult::output_path` into the DB directory, fall back to local compaction for the same job and notify the service with `OnInstallation(..., kUseLocal)`.
At that point the remote SSTs are still in the service-managed output directory recorded in `CompactionServiceResult::output_path`, and the primary has not installed any of them into the DB yet.
Update `CompactionServiceTest.InvalidResultFallsBackToLocal` to verify the fallback completes successfully, preserves the data, and invokes `OnInstallation()` exactly once with `kUseLocal`.
Reviewed By: jaykorean
Differential Revision: D106321319
fbshipit-source-id: 39d9206f0e3f62612a52c03462bd1bee69020b80
Summary:
- parse folly getdeps manifests with bare package entries so fallback prefetching actually runs
- validate and remove bad cached/downloaded archives before trying fallback mirrors
- download through temporary files and include libiberty in the GNU toolchain fallback set
Context:
Nightly test failed with dependency download failure in folly.
```
Assessing autoconf...
Download with https://ftpmirror.gnu.org/gnu/autoconf/autoconf-2.69.tar.gz -> /tmp/fbcode_builder_getdeps-Z__wZrocksdbZrocksdbZthird-partyZfollyZbuildZfbcode_builder-root/downloads/autoconf-autoconf-2.69.tar.gz ...
[Complete in 136.616022 seconds]
raise Exception(
Exception: https://ftpmirror.gnu.org/gnu/autoconf/autoconf-2.69.tar.gz: expected sha256 954bd69b391edc12d6a4a51a2dd1476543da5c6bbf05a95b59dc0dd6fd4c2969 but got e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855
make: *** [folly.mk:152: build_folly] Error 1
##[error]Process completed with exit code 2.
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14763
Test Plan:
1. Connection failure:
- Forced first mirror to http://127.0.0.1:1/...
- It logged connection refused.
- It then tried https://mirrors.kernel.org/gnu/...
- Download succeeded, size 1927468, SHA matched 954bd69b...
2. Empty file / bad hash:
- Ran a local HTTP server returning a zero-byte autoconf-2.69.tar.gz
- Script logged mismatch with actual=e3b0c442... size=0
- It removed the bad download and fell back to mirrors.kernel.org
- Download succeeded with the expected SHA.
3. Existing zero-byte cache:
- Seeded cache with an empty tarball.
- Script removed invalid cache and downloaded a verified copy.
Reviewed By: mszeszko-meta
Differential Revision: D105859558
Pulled By: xingbowang
fbshipit-source-id: ff1f20f87debad561610271ce99b8b8de2d4264f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14749
Add `--num_dbs` flag to run N independent DB instances in parallel. Each `StressTest` instance has its own DB with isolated fault injection (from D104959945). `db_crashtest.py` defaults to `num_dbs=1`.
For `num_dbs=1`: `--db` and `--expected_values_dir` are paths used as-is.
For `num_dbs>1`: they are parent directories; C++ creates `db_0/`, `db_1/`, ... subdirs underneath.
Path ownership: C++ owns DB and secondary dir creation (supports remote env). Python owns EV dir creation (always local). C++ also creates EV dirs as fallback for direct CLI usage. `DestroyAllDbs` cleans up subdirs and the parent dir.
Per-DB: `threads`, `max_key`, `ops_per_thread`, `reopen`, `column_families`, and all DB options.
Shared: background env threads (compaction, flush pool), `block_cache`, `write_buffer_manager`, `compressed_secondary_cache`, `rate_limiter`, `compaction_thread_pool_adjust_interval`.
Reviewed By: anand1976
Differential Revision: D104959942
fbshipit-source-id: 3d0d60101e7f2e600306e5a9c4018686bf649658
Summary:
- Restore prepared transactions to `PREPARED` state when writing the commit marker fails, so callers can still roll them back.
- Preserve `PREPARED` state when rollback of a prepared transaction hits a retryable write error, allowing rollback to be retried after `DB::Resume()`.
- Update `db_stress` to clean up prepared transactions after failed commits and report detailed rollback cleanup failure diagnostics.
- Add a WritePrepared regression test covering retryable commit write failure, retryable rollback write failure, successful rollback retry, and DB reopen.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14778
Test Plan: CI
Reviewed By: pdillinger
Differential Revision: D106202437
Pulled By: xingbowang
fbshipit-source-id: b0b52e1d14f39b023b9692dd8fc44060fa35c446
Summary:
- When `reuse_manifest_on_open` reuses the current MANIFEST, `DB::Open` recovery can flush WAL data into a new L0 SST and append the corresponding `VersionEdit` to that already-current MANIFEST.
- If open later fails and the process crashes, the MANIFEST edit can be durable while the recovered SST directory entry is not, leaving the DB pointing at a missing SST.
- Fsync the recovered SST's data directory before adding the file to the recovery edit when appending to a reused MANIFEST.
- Add a regression test that injects failure after MANIFEST sync, simulates crash cleanup of files created after the last directory sync, and verifies the recovered key remains readable.
## Task
- T272584339
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14780
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D106201774
Pulled By: xingbowang
fbshipit-source-id: a44a7d1263d5bc1d82b995c90eef1a825eab4182
Summary:
- Fix secondary catch-up WAL discovery to retain existing WAL readers until MANIFEST replay advances min_log_number_to_keep past them.
- Do not treat a higher-number WAL appearing in the directory as proof that a lower-number current WAL is obsolete; async WAL precreation can expose that shape while the lower-number WAL is still growing.
- Continue scanning from the smallest retained reader so later appends to the current WAL are replayed, and add a regression test that precreates an empty future WAL before secondary catch-up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14781
Test Plan:
- make -j128 db_secondary_test
- ./db_secondary_test --gtest_filter=DBSecondaryTest.CatchUpTailsCurrentWalWhenFutureWalExists
- ./db_secondary_test
- make check-sources
Reviewed By: hx235
Differential Revision: D106296411
Pulled By: xingbowang
fbshipit-source-id: acc850a177c02968372981d1407721540bc164f5
Summary:
`AdvancedColumnFamilyOptions::memtable_batch_lookup_optimization` ([advanced_options.h](https://github.com/facebook/rocksdb/blob/main/include/rocksdb/advanced_options.h)) gates the skip-list memtable's batch-lookup optimization for `MultiGet`. When enabled, the search path is cached between consecutive keys, reducing per-key cost from `O(log N)` to `O(log d)` where `d` is the distance between consecutive keys.
The C++ field exists; the C API setter does not. This PR adds the missing pair, mirroring the existing `rocksdb_options_{set,get}_memtable_huge_page_size` shape exactly — the closest sibling on both axes:
- C API: adjacent memtable knob, same `rocksdb_options_t*` receiver.
- C++: same `AdvancedColumnFamilyOptions` parent struct, same immutability semantics.
## Motivation
Without this setter, C API consumers and downstream bindings cannot opt into the batch-lookup optimization. Non-skip-list memtable implementations fall back to per-key lookups, so the flag is a no-op for them.
The field is immutable on the C++ side, so calling the setter on options that are already in use by an open DB has no effect on that DB — same constraint as the underlying C++ field. This matches the behavior of every other immutable-options setter in the C API.
No change to the C++ API.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14776
Reviewed By: joshkang97
Differential Revision: D106364224
Pulled By: xingbowang
fbshipit-source-id: 90946af498fba51581a1e7d493c9e5c9b98472a2
Summary:
The block-based table format gained an "auto" index-block search mode ([table.h](https://github.com/facebook/rocksdb/blob/main/include/rocksdb/table.h)) that selects binary vs interpolation search per index block based on key uniformity. The C++ surface exposes this as two coupled knobs:
- `BlockSearchType::kAuto = 0x02` selects the per-block adaptive search at read time.
- `BlockBasedTableOptions::uniform_cv_threshold` (default `-1`, i.e. disabled) is the coefficient-of-variation threshold checked on the write path to set the per-block `is_uniform` footer bit that `kAuto` reads.
This PR adds the missing C API coverage for both:
1. **`rocksdb_block_based_table_index_block_search_type_auto = 2`** enum constant. The existing setter `rocksdb_block_based_options_set_index_block_search_type` already does `static_cast<BlockSearchType>(v)`, so `kAuto = 2` was reachable today by passing the raw int — only the named constant was missing.
2. **`rocksdb_block_based_options_set_uniform_cv_threshold(...)` setter**. The field had no C wrapper, so C/binding users could select `kAuto` but the `is_uniform` bit was never set on the write path, making `kAuto` degenerate to `kBinary`.
No getter is added: the surrounding `rocksdb_block_based_options_set_*` functions in `c.h` do not expose getters either, so adding one only here would be inconsistent with the local style.
## Motivation
Without both pieces, `kAuto` is effectively unreachable from C. This matters for binding consumers (Rust, Go, Java-via-JNI shim, etc.) who want to opt index-block search into the per-block adaptive mode.
The setter mirrors the existing `rocksdb_block_based_options_set_data_block_hash_ratio` shape exactly (same struct, same `double` payload, same naming pattern), so review surface is minimal.
No change to the C++ API.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14775
Reviewed By: joshkang97
Differential Revision: D106364288
Pulled By: xingbowang
fbshipit-source-id: bb532eac6d4c04d032a7235f25ac29ab74f636f2
Summary:
- Suppress retryable `io_uring_wait_cqe()` errors in `PosixFileSystem::Poll()` and `AbortIO()` before they reach stderr during crash-test SIGTERM timeout handling.
- Keep terminal `wait_cqe` failures logged and fatal.
- Extend the `db_crashtest.py` SIGTERM stderr filter only for retryable `Poll`/`AbortIO` wait_cqe errors and add regression coverage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14779
Test Plan:
CI
## Related
- T272682963
Reviewed By: hx235
Differential Revision: D106201579
Pulled By: xingbowang
fbshipit-source-id: 2331d5cdca064ad901f6af7341b4e8f15b418663
Summary:
The async MultiGet support introduced two `ReadOptions` flags: `async_io` and `optimize_multiget_for_io`. The setter/getter for `async_io` was exposed in the C API in [ff04fb154](https://github.com/facebook/rocksdb/commit/ff04fb154bd74fd0681baa83b478095207e2719d); the corresponding pair for `optimize_multiget_for_io` was not.
This PR adds `rocksdb_readoptions_set_optimize_multiget_for_io` and `rocksdb_readoptions_get_optimize_multiget_for_io`, mirroring the existing `async_io` pattern exactly.
## Motivation
The flag is consulted in `db/version_set.cc` only inside the `#if USE_COROUTINES` guard, so this setter has no behavioral effect in non-coroutine builds. It matters for:
1. **API parity** with the C++ surface and with the existing `async_io` C API. The two flags were introduced together as part of the async MultiGet feature; only exposing one is an oversight.
2. **CPU/latency tuning in `USE_COROUTINES` builds.** Per the [Asynchronous IO in RocksDB blog post](https://rocksdb.org/blog/2022/10/07/asynchronous-io-in-rocksdb.html), `async_io=true` with `optimize_multiget_for_io=false` (single-level parallel reads) ran 775 μs/op vs 508 μs/op with `optimize_multiget_for_io=true` (multi-level), with the latter incurring additional CPU overhead from coroutine scheduling. Without this setter, coroutine-enabled builds cannot reach the single-level configuration from C.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14752
Reviewed By: mszeszko-meta
Differential Revision: D106080144
Pulled By: xingbowang
fbshipit-source-id: 28f12f07f29660392ba6ef7840b22804dab3567b
Summary:
* Fix Makefile default target (was ordered after a folly target)
* Improved the speed of `make clean` by using just one `find` and by pruning "hidden" .* and third-party directories that should not be modified anyway.
* Reduce excessive output from `make clean`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14767
Test Plan: manual
Reviewed By: mszeszko-meta
Differential Revision: D105972958
Pulled By: pdillinger
fbshipit-source-id: 2c0f6097c74c3129b815450f23c19ef07bfbe656
Summary:
Fix flaky test failures in EnvPosixTestWithParam.AllocateTest and DBWALTest.TruncateLastLogAfterRecoverWithFlush that occur when running on filesystems where preallocated space is not reliably reflected in st_blocks.
The tests use stat() to check st_blocks and verify that fallocate with FALLOC_FL_KEEP_SIZE actually preallocates disk space. However, on certain filesystems, this check is unreliable:
1. btrfs and zfs: Copy-on-write filesystems where preallocated extents are not reliably reflected in st_blocks, especially under load.
2. tmpfs: Memory filesystem that may not report preallocated blocks in st_blocks.
3. overlayfs: Union filesystem common in containers that may not pass through fallocate properly or report preallocated space.
4. Any filesystem where FALLOC_FL_KEEP_SIZE is not supported: The preallocation will fail silently (error ignored with PermitUncheckedError), leaving only the written data in st_blocks.
Changes made:
env/env_test.cc:
- Added filesystem magic number definitions for TMPFS_MAGIC, OVERLAYFS_SUPER_MAGIC, and ZFS_SUPER_MAGIC
- Extended the AllocateTest to skip block count checks on zfs, tmpfs, and overlayfs in addition to btrfs
- Added runtime fallback: if st_blocks is less than expected, print a warning and skip the check instead of failing. This handles unknown filesystems or configurations where preallocation isn't supported.
db/db_wal_test.cc:
- Added includes and filesystem magic number definitions
- Added ShouldSkipAllocationCheck() helper function to detect problematic filesystems
- Modified TruncateLastLogAfterRecoverWithoutFlush, TruncateLastLogAfterRecoverWithFlush, TruncateLastLogAfterRecoverWALEmpty, and ReadOnlyRecoveryNoTruncate tests to skip allocation checks on problematic filesystems
- Added runtime fallback checks similar to env_test.cc
These changes make the tests robust against filesystem differences while still validating preallocation behavior on filesystems where it works correctly (ext4, xfs).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14744
Test Plan: Many local 'make -j100 check' runs that would previously fail with good probability.
Reviewed By: hx235
Differential Revision: D105331256
Pulled By: pdillinger
fbshipit-source-id: 862a1512da1466cb037af15342404939b677c02a
Summary:
### Notify listeners before DB shutdown begins
db_stress listener bookkeeping correlates compaction callbacks with file deletion callbacks. DBImpl intentionally skips some compaction listener callbacks once shutdown starts, but file deletion callbacks can still be delivered. That mismatch is normally tolerable for production listeners, but db_stress uses listener-local tracking to detect callback consistency and can report a false positive when shutdown interrupts a compaction callback sequence.
The failed-open case is the important gap. DB::Open() can create a DBImpl, recover or flush files, schedule background compaction, and then fail during late open work such as persisting OPTIONS or waiting for open-time compaction under fault injection. At that point DB::Open() tears down the internal DBImpl before returning an error to StressTest::Open(). If OnCompactionBegin already recorded an input file or job in DbStressListener, DBImpl shutdown can suppress the later OnCompactionPreCommit/OnCompactionCompleted callbacks that would normally clear that state. Since control has not returned to the stress harness yet, StressTest::CleanUp()/Reopen() cannot notify the listener in time. A later shutdown-time callback, or the next open retry reusing the same listener, can then observe stale tracking and abort even though RocksDB did not compact the same SST concurrently.
Add EventListener::OnDBShutdownBegin and fire it once from DBImpl::CancelAllBackgroundWork() before publishing shutting_down_. The callback also covers cleanup of a failed DB::Open() attempt, where the DB pointer refers to the internal DBImpl that was never returned to the caller. Track shutdown_notification_sent_ separately from shutting_down_ because listeners are invoked with mutex_ released, and a concurrent or reentrant cancellation must not deliver the callback twice.
Update DbStressListener to consume the DBImpl-driven shutdown notification instead of relying on StressTest::CleanUp()/Reopen() to manually notify it. This lets db_stress mark itself as shutting down before DBImpl starts skipping shutdown-sensitive compaction notifications, including during failed-open cleanup.
### Also keep listener state scoped to one db_stress open attempt.
Initialize listeners at the top of each non-transactional open retry before enabling open fault injection, so retry attempts get fresh listener state without widening open fault injection to listener construction. Factor the open fault setup into a small helper to keep the retry loop readable. The transaction open path still initializes listeners once because it does not use this open-fault retry loop.
### Also fix multi-ops transaction listener checks during shutdown
A TSAN race was reported where MultiOpsTxnsStressListener::OnCompactionCompleted called VerifyPkSkFast on a background compaction thread while the main thread was destroying the transaction DB wrapper in StressTest::CleanUp(). The reported read was a virtual call through the DB object and the write was the WritePreparedTxnDB/WriteUnpreparedTxnDB destructor updating the vptr.
The vulnerable ordering is that DBImpl can already be inside NotifyOnCompactionCompleted before shutdown is requested. It unlocks db mutex while iterating listeners; another listener such as DbStressListener can spend time in its callback, giving the main thread time to enter CleanUp()/Close(). The DB object is still shutting down, but MultiOpsTxnsStressListener may be invoked later in the same callback iteration and call VerifyPkSkFast through stress_test_->db_aptr_, racing with DB wrapper destruction.
With DBImpl-owned EventListener::OnDBShutdownBegin callback, we have MultiOpsTxnsStressListener consume that callback directly. Once DBImpl begins shutdown, the listener skips both flush-completed and compaction-completed verification callbacks, avoiding DB access during teardown. This also covers failed-open cleanup without name-based downcasts in StressTest.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14769
Test Plan: CI
Reviewed By: pdillinger
Differential Revision: D106011452
Pulled By: xingbowang
fbshipit-source-id: 768838ddcd9910de5d1b5204c990a4d88dbc850c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14757
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14750
Prerequisite for multi-DB stress test support where each StressTest instance owns one DB. Moves fault injection from a single global to per-StressTest instance so each DB gets isolated fault injection; errors injected into one DB do not leak to others.
FS/Env architecture change:
Before (upstream):
raw_fs → FaultInjectionTestFS (global) → DbStressFSWrapper → db_stress_env → DB::Open
After (this diff):
Global: raw_env (no wrappers)
Used outside StressTest: non-FS ops (threads, time, sleep), test framework FS setup (dirs), DB destruction
Used inside StressTest: cleanup ops that must succeed (external file delete)
Per StressTest:
raw_fs → DbStressFSWrapper (db_stress_fs_, always)
→ FaultInjectionTestFS (db_fault_injection_fs_, optional)
→ CompositeEnvWrapper (db_env_, always) → DB::Open
Expected values state: Env::Default() (always local PosixEnv even when raw_env is remote)
FS layer order swapped (DbStressFSWrapper now innermost). Safe because:
- Error injection returns early; inner wrapper never executes (same behavior)
- DbStressFSWrapper assertions do not modify data (checksum validation, IOActivity checks)
- MANIFEST rename tracking slightly better for crash simulation in new order
Stored members (per StressTest):
- db_stress_fs_: DbStressFSWrapper. Always active regardless of fault injection flags.
- db_fault_injection_fs_: FaultInjectionTestFS. Only when fault injection flags set. Direct access for Enable/Disable/SetThreadLocal.
- db_env_: CompositeEnvWrapper. Always present. THE env for all DB I/O (options_.env).
Eliminated globals: db_stress_env → renamed to raw_env (no wrappers, DbStressFSWrapper moved to per-StressTest); db_stress_listener_env, db_stress_raw_fs removed; fault_fs_guard, fault_env_guard moved to per-StressTest.
Other changes:
- CleanupOutputDirectory simplified (uses raw_env, no disable/enable needed)
- SstFileManager recreated with per-StressTest env in Open()
- Remote compaction override env uses options.env directly (fixes pre-existing silent bug)
- Comments added: Env::Default() always local, DbStressDestroyDb MANIFEST explanation, fault injection log path in TEST_TMPDIR
- TestFSWritableFile::Close() now mirrors the production FSWritableFile close boundary. After the first Close() attempt, later explicit or destructor Close() calls are wrapper-level no-ops, while FaultInjectionTestFS still records the first close attempt for crash/recovery simulation.
- Added targeted fault_injection_fs_test coverage for injected metadata-close failures to ensure FaultInjectionTestFS does not retry the inner Close() path.
Reviewed By: anand1976
Differential Revision: D104959945
fbshipit-source-id: 7cf9bb494dec2b372528d5f119c023b6d392ffca
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14754
**Problem**
When db_bench shuts down (fatal IO error, unknown benchmark name, etc.), ErrorExit() or ~Benchmark() destroys db_/multi_dbs_ while worker threads may still be inside benchmark methods holding raw DBWithColumnFamilies pointers returned by SelectDBWithCfh(). In single-DB mode, db_.db becomes nullptr and multi_dbs_ is empty, so workers evaluate rand_int % 0 and SIGFPE. The cascade across all workers masks whatever originally triggered the shutdown. Observed in the reliable_volumes crash-fault test at ~5 SIGFPE events per hour.
**Approach**
Guard DB lifetime at the worker-thread boundary with a reader-writer lock (port::RWMutex): workers acquire a read lock for the duration of their benchmark method; mutators acquire the write lock (which waits for active readers to drain) before destroying DBs. This ensures raw pointers returned by SelectDBWithCfh() remain valid for their entire usage scope.
**Fix**
- Workers hold a DbUseGuard (RAII wrapper over a read lock on db_lifecycle_rwlock_) for the entire benchmark method in ThreadBody. Cost: one rdlock/unlock pair per worker lifetime, zero hot-path overhead.
- Every DB mutation path (ErrorExit, ~Benchmark, fresh-DB reopen) takes DbStateMutationGuard, which acquires the write lock and then stops the secondary update thread before mutation may proceed.
- ErrorExit() routes to std::_Exit(1) when called from any thread that cannot safely run the mutation cleanup path: a DbUseGuard holder (would self-wait on the write lock) or the secondary update thread (would self-join via StopSecondaryUpdateThread). Tracked via two thread-locals: holds_db_use_guard_ and is_secondary_update_thread_. Main thread takes DbStateMutationGuard, cleans up, dispatches through db_bench_exit() / ToolHooks::Exit.
- StopSecondaryUpdateThread() resets secondary_update_stopped_ to 0 after joining, so a replacement thread created by a subsequent Open() is not immediately killed.
- Protocol is mechanically enforced in debug builds:
* SelectDBWithCfh asserts holds_db_use_guard_ (read-side ownership)
* DeleteDBs asserts holds_db_state_mutation_guard_ (write-side ownership)
* DbUseGuard and DbStateMutationGuard ctor/dtor assert correct imbalance (no nested or stray release).
* DbStateMutationGuard ctor additionally asserts !holds_db_use_guard_ and !is_secondary_update_thread_, catching the new deadlock modes the single-lock design makes reachable (WriteLock while holding ReadLock; secondary thread self-joining via StopSecondaryUpdateThread).
These convert "trust me" invariants into "the assert fires if you break it." Direct field accesses (e.g. db_.db->NewIterator inside a benchmark method, multi_dbs_.clear() in the reopen branch) are protected by the enclosing guard scope but are not per-site asserted.
**Caveat**
port::RWMutex is pthread_rwlock_t with default attrs -> reader-preferred on Linux/glibc. The current call graph has no concurrent worker spawn during mutator wait (mutator paths run only from the main thread, not concurrently with RunBenchmark spawning workers), so writer starvation is not reachable. Documented as a constraint; revisit if that invariant changes. The port layer doesn't expose pthread_rwlockattr_setkind_np cross-platform.
**Alternatives considered**
- std::_Exit(1) on every shutdown path, skip cleanup entirely. Loses ToolHooks::Exit dispatch, flushed traces, and end-of-run stats -- those matter for the RV crash-fault test image which consumes db_bench output. Rejected.
- shared_ptr<DBWithColumnFamilies> from SelectDBWithCfh, let DB lifetime extend naturally to the last reader. Adds a per-call atomic refcount bump in the hot path; the RWMutex approach is per-method-call instead of per-op, making the hot-path cost zero. Rejected for hot-path neutrality.
Reviewed By: xingbowang
Differential Revision: D104974784
fbshipit-source-id: 3a04d9e1c0b5042436d690f573cf369de6b4c9df
Summary:
- Rebuild db_stress event listeners through a shared `InitializeListenersForOpen()` helper.
- Reinitialize listeners before retrying `DB::Open()` after injected open or open-compaction failures.
## Context
- A stress test failed with "Concurrent compaction of SST file detected".
- Root cause: StressTest::Open() built DbStressListener once before its DB::Open() retry loop. With open fault injection, a DB::Open() attempt can create a DBImpl, schedule background compaction, and then fail during late open work such as persisting OPTIONS. During teardown of that failed DBImpl, DBImpl's shutdown flag can suppress later compaction callbacks, leaving listener-local compaction bookkeeping stale.
- Diagnosis: Sandcastle DB LOGs showed file 16821 flushed, then a failed open attempt with an injected read error and a background compaction picking 16821. The crash was in DbStressListener::OnCompactionBegin, so this was stale db_stress listener state across open attempts rather than DBImpl allowing a real concurrent compaction of the same SST.
- Fix: factor listener construction into InitializeListenersForOpen() and call it before each DB::Open() attempt, including the retry path after open/open-compaction failure. Each DBImpl open attempt now gets fresh listener state.
- Verification: make clean; make db_stress -j192; make check-sources; git diff --check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14765
Test Plan:
- `make clean`
- `make db_stress -j192`
- `make check-sources`
- `git diff --check`
Reviewed By: pdillinger
Differential Revision: D105969381
Pulled By: xingbowang
fbshipit-source-id: 759e2be7e1215a498ed449ab36f13e8c7975f4a4
Summary:
- Clean DBTestBase's fixture-owned alternate WAL and db_log_dir paths during setup and teardown.
- Prevent kDBLogDir option tests from leaving dbname_/db_log_dir behind and polluting later DBTest cases in the same gtest shard.
- Preserve production DestroyDB() behavior while making the test fixture cleanup complete.
Context:
- The ARM nightly failure was exposed by the 32-shard db_test layout running DBTest.GetPicksCorrectFile before DBTest.PurgeInfoLogs in the same process.
- GetPicksCorrectFile can use kDBLogDir, which creates logs under dbname_/db_log_dir. DestroyDB() intentionally ignores DeleteDir() failures when unknown children remain, so the next fixture could observe dbname_ still existing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14764
Test Plan:
- make -j14 db_test
- TEST_TMPDIR=/tmp/rocksdb_arm_fix_test ./db_test --gtest_filter=DBTest.GetPicksCorrectFile:DBTest.PurgeInfoLogs
- TEST_TMPDIR=/tmp/rocksdb_arm_fix_test_shard GTEST_TOTAL_SHARDS=32 GTEST_SHARD_INDEX=18 ./db_test
Reviewed By: mszeszko-meta
Differential Revision: D105860476
Pulled By: xingbowang
fbshipit-source-id: b2685063ca6c4eedec589697b439fe4aee4eda1a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14756
Pure mechanical refactor: replace all direct FLAGS_db / FLAGS_expected_values_dir / FLAGS_secondaries_base reads with accessor methods on StressTest. No new flags, no parameters, no behavior change. Prepares for multi-DB stress test where each StressTest instance has its own DB.
Changes:
- GetDbPath(), GetExpectedValuesDir(), GetSecondariesBase() accessors return the corresponding FLAGS values directly
- Replace ~15 FLAGS_db references with GetDbPath() in db_stress_test_base.cc
- Move SharedState constructor from .h to .cc (needs full StressTest type for GetExpectedValuesDir())
- Move DbStressListener constructor from .h to .cc (same reason)
- Replace FLAGS_db / FLAGS_expected_values_dir in db_stress_driver.cc with accessor calls
- NO changes to db_crashtest.py or db_stress_gflags.cc
Reviewed By: anand1976
Differential Revision: D104959943
fbshipit-source-id: d7ef6a39d4c2ed467b2960417629c09f3988faf5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14742
This change enables testing of the two new MANIFEST optimization options introduced in D103568447:
1. optimize_manifest_for_recovery - Skips unnecessary MANIFEST edits during recovery
2. reuse_manifest_on_open - Reuses existing MANIFEST file on DB open
Changes:
- tools/db_crashtest.py: Add both options with 20% probability to default_params
- db_stress_tool/db_stress_test_base.h: Add ManifestVerifyMode enum and member variables for tracking MANIFEST state
- db_stress_tool/db_stress_test_base.cc: Implement RecordManifestStateBeforeReopen() and VerifyManifestNotRewritten() methods to validate MANIFEST reuse on DB reopen
The verification logic handles 4 combinations:
- Both disabled: No verification (baseline)
- Only optimize enabled: No verification (hard to measure without sync points)
- Only reuse enabled: Verify MANIFEST file is reused
- Both enabled: Verify MANIFEST reused AND CURRENT unchanged
Verification is warning-only (not fatal) to account for legitimate fallback cases like corruption or size limits.
Reviewed By: hx235
Differential Revision: D105224068
fbshipit-source-id: 5baa65680fdd639674d87ff1e9187b743e691bc1
Summary:
seen several times in the build-linux-mini-crashtest job. Raise ulimit in container spec.
Task: T271298423
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14755
Test Plan: look at reported ulimits, watch CI
Reviewed By: hx235
Differential Revision: D105664569
Pulled By: pdillinger
fbshipit-source-id: 6f7f3976bd73ecc86509ac955d5e190316e98ba3
Summary:
NotifyOnCompactionPreCommit initially omitted the shutting_down_ guard to avoid a false positive abort in the db_stress OnTableFileDeleted check, where stale compacting_files_ entries from skipped notifications would be mistaken for a bug. The better fix is to add the shutting_down_ guard for consistency with Begin and Completed, and instead make OnTableFileDeleted tolerate stale tracking during shutdown by checking a new atomic bool in the listener intended to track DBImpl's shutting_down_.
Also fix lint: release mutex before RandomSleep() in PreCommit; use char literal for find_last_of; avoid unnecessary string copy.
Document WART in listener.h: all three compaction callbacks are skipped during DB shutdown, so a committed compaction may go unobserved.
Bonus: update CLAUDE.md with instructions on avoiding non-ASCII characters
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14753
Test Plan: manually trigger many crash test runs
Reviewed By: hx235
Differential Revision: D105591913
Pulled By: pdillinger
fbshipit-source-id: 32029fea4c2571d88f645eb325db2e25a94e0d26
Summary:
BackgroundJobPressure test was flaky because Phase 3 used TEST_WaitForCompact() which waits for bg_compaction_scheduled_ but NOT bg_pressure_callback_in_progress_. The final "healthy" pressure callback could still be in-flight when the test checked snapshots.back(), causing compaction_scheduled=1 instead of 0.
Fix: replace TEST_WaitForCompact() with TEST_WaitForBackgroundWork() which explicitly checks bg_pressure_callback_in_progress_ (db_impl.cc:478-484). Also add TEST_WaitForBackgroundWork() before Phase 1 and Phase 2 snapshot checks for consistency, ensuring all pressure callbacks are delivered before assertions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14708
Reviewed By: jaykorean
Differential Revision: D103784101
fbshipit-source-id: 275802b5bb70094af62486bde26b599a292e71fa
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14751
`regression_test.sh` already cleans the current run's `TEST_PATH` on normal success and on the early-exit path when a recent `db_bench` is still running. But a hard benchmark failure goes through `exit_on_error`, which exits before the end-of-`main()` cleanup runs.
This change adds an `EXIT` trap and a single-shot finalizer so the current invocation still cleans its own `TEST_PATH` on hard failure. It does not reintroduce any sibling-directory cleanup, and it preserves the existing success-path cleanup, early-exit cleanup, and debug preservation behavior.
| Scenario | Previous code | New code |
| --- | --- | --- |
| Normal success | Cleans current `TEST_PATH` at end of `main()` | Still cleans current `TEST_PATH` |
| Hard benchmark failure via `exit_on_error` | Can leave current `TEST_PATH` behind | `EXIT` trap cleans current `TEST_PATH` |
| Early exit because recent `db_bench` exists | Cleans current `TEST_PATH`, exits `2` | Same behavior |
| Debug mode / `DELETE_TEST_PATH=0` | Preserves artifacts | Same behavior |
Reviewed By: jaykorean
Differential Revision: D105411220
fbshipit-source-id: 37a335b87faaeee86d44ef2e24bebf1b7b9626d6
Summary:
Reduce wall-clock time of parallel 'make check' by improving the scheduling and granularity of slow test binaries.
Three Makefile changes:
1) Refresh slow_test_regexp with current observed bottlenecks. Adds
binaries (point_lock_manager_stress_test, compaction_service_test,
corruption_test, comparator_db_test, external_sst_file_basic_test,
rate_limiter_test, db_compaction_test, db_merge_operator_test,
db_dynamic_level_test, db_bloom_filter_test, error_handler_fs_test,
merge_helper_test, db_kv_checksum_test, inlineskiplist_test) whose
shards take >=15s but were not being front-loaded for early
queueing. Also drops stale FIXME comments that no longer apply
and adds tier annotations + a maintenance recipe.
2) Add SHARD_SIZE_OVERRIDES, a per-binary override of GTEST_SHARD_SIZE,
so binaries with slow individual tests (e.g.
point_lock_manager_stress_test where each test is ~10s) can be
chopped into more, smaller shards. The default of 10 stays for
everything else. Each shard's effective size is reported in the
'Generating ... shards for ...' line.
3) Add 'make suggest-slow-tests' to print a per-binary aggregation of
the most recent LOG (max single-shard time, total time, shard
count) for any binary worth attention. Used to maintain the regex
and override list above.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14745
Test Plan:
Two runs each of 'make -j166 check', before and after this change (all compilation already finished):
Before: 197s and 198s
After: 123s and 125s
Reduction: 37%
Reviewed By: xingbowang
Differential Revision: D105332444
Pulled By: pdillinger
fbshipit-source-id: 1d1c2f89a32647e6651e2ffeb72da9d51bcc004f
Summary:
In PR https://github.com/facebook/rocksdb/issues/13531, we added a `saved_key_.SetUserKey(ikey.user_key)` call in `FindValueForCurrentKeyUsingSeek` to fix unprepared-value reverse iteration. The default `copy`=`true` parameter unconditionally copies the key into the internal buffer, breaking is-key-pinned when `pin_data`=`true`. This path triggers only when a key has more versions than `max_sequential_skip_in_iterations` (default 8 ), making the bug rare and hard to repro deterministically.
**Fix:** pass `!pin_thru_lifetime_ || !iter_.iter()->IsKeyPinned()` as the copy parameter, matching every other `SetUserKey` call site in `DBIter`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14746
Test Plan: New test `SeekForPrevKeyPinnedWithManyVersions`: writes 20 versions of the same key, confirms `FindValueForCurrentKeyUsingSeek` is taken via `NUMBER_OF_RESEEKS_IN_ITERATION`, asserts is-key-pinned == "1" after `SeekForPrev`. Fails without fix, passes with.
Reviewed By: xingbowang
Differential Revision: D105397906
Pulled By: mszeszko-meta
fbshipit-source-id: 5581105e9d929bb4c582c52dd6a7ae3e8dd9da72
Summary:
Adds a new `EventListener::OnCompactionPreCommit` callback that fires after a compaction job's output files are written but *before* the manifest write commits the new Version. At that point input files still have `FileMetaData::being_compacted == true`, so listeners that maintain bookkeeping of "files currently being compacted" can clean up that state without racing the compaction picker.
A check implemented by a Meta-internal RocksDB user crashes when the same file appears as input to two concurrent compactions. Tracking that set in `OnCompactionBegin` / `OnCompactionCompleted` produces false positives because `Compaction::ReleaseCompactionFiles()` flips `being_compacted` back to false before `OnCompactionCompleted` fires, so another thread can pick the same file and trigger `OnCompactionBegin` before the previous compaction's `Completed` callback runs. Doing the cleanup in `OnCompactionPreCommit` closes that race. Default implementation is a no-op, so no API break.
A trivial refactoring to split PerformTrivialMove ensures data is populated for the new callback, while calling back before the trivial move compaction is committed.
Bonus: `CLAUDE.md` update for when to call `make clean` as I've recently had it get thoroughly confused TWICE mixing build modes.
Task: T269479969
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14740
Test Plan:
- New unit test in `db/listener_test.cc` checks that `OnCompactionPreCommit` fires strictly between `OnCompactionBegin` and `OnCompactionCompleted` for the same compaction, and that input files still have `being_compacted == true` at the time it fires.
- Crash test: adds a concurrent-compaction sanity check to `db_stress`'s listener resembling the Meta-internal intended usage: tracks input file numbers from `OnCompactionBegin` to `OnCompactionPreCommit`, aborting if the same file appears as input to two concurrent compactions. Also checks other ordering constraints and checks for "leaks" from possible failure to call OnCompactionPreCommit(). Exercises all compaction styles and trivial-move/FIFO paths under load.
Reviewed By: hx235
Differential Revision: D105065326
Pulled By: pdillinger
fbshipit-source-id: 8f606a7c0fac899574e7340c600b71fed902394a
Summary:
- Add experimental immutable `DBOptions::async_wal_precreate` to reserve and open one future WAL on a background HIGH-priority task, with sanitization that disables the optimization when WAL recycling is configured.
- Split WAL creation into open/preallocate and start phases so `SwitchMemtable()` can consume a prepared WAL after writing normal WAL metadata, wait for in-flight precreation, fall back to synchronous creation, and delete an unstarted prepared WAL on start failure.
- Keep WAL numbering, close, recovery, and read-only open safe for empty future WAL files left by async precreation; `error_if_wal_file_exists=true` now rejects non-empty WALs while tolerating empty WALs.
- Add public option plumbing for the C API, options parsing/stringification, random option testing, `db_bench`, `db_stress`, and crash-test configuration.
- Add WAL precreate statistics counters plus Java `TickerType`/JNI mappings, and update C++, C, and Java read-only-open documentation for the empty-WAL behavior.
- Add focused WAL/option/C/Java tests for async precreate ready/wait/failure/recovery paths, read-only WAL detection, option sanitization, and API plumbing, plus write-flow docs and unreleased history entries for the new feature and behavior change.
PR https://github.com/facebook/rocksdb/pull/14738
Reviewed By: pdillinger
Differential Revision: D105020559
fbshipit-source-id: 5059b424702e021abb8de65ceeb6d3b975280ffc
Summary:
**Summary:**
Adds a --listener_uri flag to db_stress that creates an EventListener via ObjectLibrary from the given URI and attaches it to the DB options.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14741
Test Plan:
- Compilation
- e2e test will be done next internally with a customized listener
Reviewed By: anand1976
Differential Revision: D104750476
Pulled By: hx235
fbshipit-source-id: cdc00191de6b7434e4b373db30769ff34d99b80d
Summary:
last_compacted_manifest_file_size_ drives TuneMaxManifestFileSize() to compute the manifest rotation threshold, but it started at 0 on every DB::Open and was only populated after the first manifest rotation. This is really only a problem with reuse_manifest_on_open, because no fresh manifest is created on open.
Add a new forward-compatible (safe-to-ignore) MANIFEST tag kLastCompactedManifestFileSize that records the approximate compacted manifest size at the end of WriteCurrentStateToManifest. During recovery, the value is loaded and used to immediately tune the rotation threshold.
The record includes a rough estimate of its own overhead (~15 bytes) and must be the last record written by WriteCurrentStateToManifest for accurate estimation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14725
Test Plan:
Extended AutoTuneManifestSize in db_etc3_test to close and reopen with reuse_manifest_on_open after establishing a known auto-tuning state. Verifies that the manifest file number is preserved (no spurious rotation) and that subsequent CF additions don't trigger rotation -- proving the persisted compacted size keeps the tuned threshold correct. Verified the test fails when the recovery loading is disabled.
Relax a fragile Java test that was dependent on the exact size of the manifest file.
SHORT_TEST=1 ./tools/check_format_compatible.sh
Reviewed By: anand1976
Differential Revision: D104464522
Pulled By: pdillinger
fbshipit-source-id: 4f5d22d2e149bd40a523ee11780e5e3344803c19
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14726
Add SstFileReader::ParseTableIteratorKey() so callers of NewTableIterator() have a public way to decode raw table keys without duplicating RocksDB internal-key layout. The implementation delegates to the existing public ParsedEntryInfo parser using the reader comparator.
Reviewed By: pdillinger
Differential Revision: D104584393
fbshipit-source-id: 98e21c4d6676fbba69e533376b3da67539dd8fad
Summary:
Previously, calling `DB::CreateColumnFamily(opts, "", &handle)` returned `Status::OK()` with a usable handle, but the column family was not persisted in the manifest. Any data written to the empty-named CF was silently lost on DB reopen, and `ListColumnFamilies` would not show it.
The empty string is also reserved as a sentinel meaning "no/unknown column family" in various RocksDB APIs and serialization formats (e.g. `TablePropertiesCollectorFactory::Context::kUnknownColumnFamily` and table properties), so allowing it as a real CF name is ambiguous in addition to being broken.
This change rejects an empty CF name with `Status::InvalidArgument` at the top of `DBImpl::CreateColumnFamilyImpl`, which covers the single-CF `CreateColumnFamily` API as well as both `CreateColumnFamilies` overloads (by-names and by-descriptors).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14732
Test Plan: * Added `ColumnFamilyTest.EmptyNameRejected` covering all three Create entry points; verifies `IsInvalidArgument()` and that no spurious handles are returned.
Reviewed By: hx235
Differential Revision: D104753911
Pulled By: pdillinger
fbshipit-source-id: e52d792830b965484a618f4e55981eee4eb6f515
Summary:
- propagate lower-level read and merge failures through `GetContext` via `read_status`, so `Get` and `GetEntity` preserve the original error instead of synthesizing `Corruption` when blob-backed reads or merge resolution fail
- teach `GetMergeOperands` to resolve blob-backed default columns from wide-column entities, covering both the direct base-value path and the merge-plus-base path
- add regression coverage for blob-read IO errors during `Get`/`GetEntity` merge resolution and for `GetMergeOperands` on blob-backed wide-column entities
- fix the `DBFlushTest.MemPurgeCorrectLogNumberAndSSTFileCreation` test race by waiting for flush callbacks and cleaning up sync points
## Testing
- `make db_blob_basic_test -j14`
- `/usr/bin/perl -e 'alarm shift; exec ARGV' 60 ./db_blob_basic_test --gtest_filter='DBBlobBasicTest/DBBlobBasicIOErrorTest.GetBlob_IOError/*:DBBlobBasicTest/DBBlobBasicIOErrorTest.GetEntityMergeWithBlobBaseIOError/*'`
## Task
T265824017, T265415808
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14640
Reviewed By: anand1976
Differential Revision: D101690700
Pulled By: xingbowang
fbshipit-source-id: 2b6fc357b37a01efa72a2d54dcff55be8992f42a
Summary:
### Motivation
Claude's auto review workflow classifies the PR as complex and sets `MAX_THINKING_TOKENS`=**32000** in `.github/workflows/ai-review-analysis.yml:534`. The Claude action then sends a request where `thinking.budget_tokens` is **32000**, but the request `max_tokens` is not greater than that, so Anthropic rejects it before the review starts ([example](https://github.com/facebook/rocksdb/actions/runs/25691112276/job/75427435133)).
### Fix
Reduce the "complex" thinking budget from **32000** to **24000** tokens and clamps manual overrides to the same ceiling, preventing the thinking budget from exceeding or equalling the action's effective `max_tokens`. 24k is still generous — enough for deep reasoning on complex PRs while guaranteeing the model can emit the formatted review.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14731
Reviewed By: xingbowang
Differential Revision: D104749426
Pulled By: mszeszko-meta
fbshipit-source-id: bf94d0f50e7c2c9bc9e4d4cdffd61087d739018a
Summary:
The non-ASCII character check in check-sources.sh used git grep -P (Perl regex), which requires git compiled with PCRE support. On systems without it, the command fails with exit code 128, which is != 1 (no match), so the check always reported a violation -- effectively dead.
Even in CI where git has PCRE2 support, the check was silently broken: git grep -P uses PCRE2 in UTF mode by default, which interprets [\x80-\xFF] as a Unicode codepoint range (U+0080 to U+00FF). Characters like em-dash (U+2014), arrows (U+2192), and math symbols (U+2248, etc.) fall outside that range and were not detected. Only Latin-1 Supplement characters (U+0080-U+00FF) would have been caught.
Replace with LC_ALL=C git grep using bash $'[\x80-\xff]' literal byte range, which works with basic regex in the C locale, and replace all non-ASCII characters in non-excluded source files:
- em-dash to --
- arrow to ->
- math symbols to ASCII equivalents (~=, <=, >=)
- box-drawing characters to ASCII art
Also exclude .github/ from the check, as scripts there can use non-ascii without disrupting RocksDB builds on non-UTF-8 systems.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14729
Test Plan: manual / CI (make check-sources passes clean)
Reviewed By: hx235
Differential Revision: D104692574
Pulled By: pdillinger
fbshipit-source-id: 1d884c21056dcd83558b825a04b867f1c08e3f45
Summary:
Previously, `Classify PR complexity (Codex)` ran under `bash -e`, so any `codex exec` failure aborted the entire Codex review before the real review step could run. The classifier only selects the review budget, so on failure we now log the classifier output tail, default to the `complex` review budget, and continue. This keeps actual Codex review failures visible through the existing review exit-code/log handling while preventing the auxiliary classifier from blocking review generation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14721
Reviewed By: xingbowang
Differential Revision: D104326081
Pulled By: mszeszko-meta
fbshipit-source-id: c17388bbf576f71ff85ded320e0740f89072f8c1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14723
### Context
`GetCreationTimeOfOldestFile()` assumed `max_open_files` = -1 meant every live SST had a pinned table reader. That is not true with `open_files_async`: recovery intentionally skips loading table files, `DB::Open()` returns, and `BGWorkAsyncFileOpen()` pins readers later. Any caller — e.g. fb_rocksdb's daily report at `FbRocksDb.cpp:1025` — invoking the API in the window between `DB::Open` returning and the background opener completing trips a debug assert. Reported in https://fb.workplace.com/groups/rocksdb/permalink/31668956482726231/.
Removing the assert alone is insufficient. For legacy DBs whose manifest does not carry `file_creation_time`, `FileMetaData::TryGetFileCreationTime()` falls back to the pinned reader; with no reader, it returns `kUnknownFileCreationTime` and the function silently returns 0 (the "info unavailable" sentinel). The caller cannot distinguish "no info" from "raced with async open."
### Changes
- Remove the invalid debug assert in `Version::GetCreationTimeOfOldestFile`.
- Add a private helper `DBImpl::WaitForAsyncFileOpen()` that blocks on `bg_cv_` while `bg_async_file_open_state_ == kScheduled`. The synchronization machinery (`bg_async_file_open_state_` + `bg_cv_`) already exists — the destructor wait loop in `db_impl.cc:674-687` uses the same pattern. The helper is a no-op when `open_files_async = false`, and bails on `shutting_down_` so `DB::Close()` is not blocked by an in-flight caller.
- Call `WaitForAsyncFileOpen()` at the top of `DBImpl::GetCreationTimeOfOldestFile()` (inside the `max_open_files == -1` branch).
- Document the blocking behavior in `include/rocksdb/db.h`.
- Replace the regression test with one that uses a `DBImpl::WaitForAsyncFileOpen::BeforeWait` sync point: spawn a thread that calls `GetCreationTimeOfOldestFile`, deterministically confirm it blocks inside the wait, release async open, confirm the caller wakes with the real value.
### Potential Followups (not included here)
- Apply the same wait to `GetLiveFilesMetaData` and `GetColumnFamilyMetaData` — both zero out `oldest_ancester_time` / `file_creation_time` in the SST metadata they return during the async-open window (`db/version_set.cc:7877-7878`, `2090-2091`, `2172-2173`).
- Address compaction-picker effects: TTL/periodic file selection (`db/version_set.cc:4039`, `4092-4094`), bottommost over-marking (`db/version_set.cc:4700-4701`), FIFO TTL/temperature pickers (`db/compaction/compaction_picker_fifo.cc:105-107`, `167-170`, `401-411`), and tiered-compaction output time inheritance (`db/compaction/compaction.cc:981`, `1000`).
- Harden `FbRocksDb.cpp:1025` to check `status.ok()` instead of `status.code() != kNotSupported`.
Reviewed By: mszeszko-meta
Differential Revision: D104285992
fbshipit-source-id: ea46375ea1b3ba77fe6b548071aee1101ac0da77
Summary:
- Use the liburing TSAN suppressions in `tools/tsan_suppressions.txt` instead of defining the process-wide `__tsan_default_suppressions()` hook, avoiding conflicts with downstream applications.
- Wire RocksDB TSAN make and crash-test flows to use that suppressions file by default without overriding caller-provided `TSAN_OPTIONS`.
- Cover direct `db_crashtest.py` launches by passing the default suppressions to `db_stress` subprocesses.
- Fix the GCC 16 unity-build warning in `CacheItemHelper` by directly initializing the no-secondary-cache helper fields instead of delegating with `this`.
Imported from D101303486.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14710
Test Plan:
- `COMPILE_WITH_TSAN=1 make -j128 env_test`
- `timeout 60s ./env_test --gtest_filter=EnvPosixTest.IOUringAddressReuseNoTsanFalsePositive`
- `python3 tools/db_crashtest_test.py`
- `python3 -m py_compile tools/db_crashtest.py tools/db_crashtest_test.py`
- CI: `build-linux-unity-and-headers` passed after the `CacheItemHelper` fix
Reviewed By: hx235
Differential Revision: D104103800
Pulled By: xingbowang
fbshipit-source-id: 6066d9abe02a3c44d75f9ce449889468c927ce56
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14704
Add an immutable DBOption `reuse_manifest_on_open` (default false). When enabled, `DB::Open` can keep using the recovered MANIFEST for the first post-open metadata update instead of rebuilding a fresh MANIFEST, which can reduce warm-open latency for DBs whose MANIFEST is expensive to regenerate.
Reuse is still best-effort. If RocksDB cannot safely resume appending to the recovered MANIFEST, it falls back to the existing fresh-MANIFEST path. The option is also disabled under `best_efforts_recovery`.
This diff also teaches the reopened MANIFEST writer to adopt the existing file size before appending, documents the small-`max_manifest_file_size` caveat for the reused path, and keeps the full warm-reopen composition working with `optimize_manifest_for_recovery`.
Reviewed By: hx235, pdillinger
Differential Revision: D103568447
fbshipit-source-id: f4f5c35ea3ef0b80a0d52d94be40c6bd11505999
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14703
Extend `optimize_manifest_for_recovery` so a clean `DB::Close` can persist up-to-date WAL recovery markers when that can be done safely. Combined with the recovery-side optimization in the previous diff, a clean close/reopen can avoid recovery-time MANIFEST appends.
This remains best-effort: if the close-time write is disabled, skipped, or fails, RocksDB falls back to the standard recovery path on the next open. The option stays mutable so it can be turned off before close to suppress the optimization without restarting the DB.
The close-time path respects the existing recovery constraints for 2PC, non-empty column families, dropped column families, and WAL tracking, and preserves the existing file-number invariants.
Reviewed By: pdillinger, hx235
Differential Revision: D103568449
fbshipit-source-id: ae62867507a8a87640a2c140bea852b7c608cb66
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14702
Add a mutable DBOption `optimize_manifest_for_recovery` (default false) as a temporary rollout / kill switch for warm-reopen MANIFEST optimizations.
In this diff, enabling the option lets recovery skip MANIFEST updates during `DB::Open` when the recovered state is already reflected on disk, which reduces metadata appends after a clean shutdown and can lower warm-reopen latency on storage where MANIFEST appends are expensive.
If the option is disabled, RocksDB follows the existing recovery path unchanged. The optimization is disabled under `best_efforts_recovery`, where recovery intentionally rewrites metadata as part of salvage, and the option is mutable so later diffs in this stack can share the same rollout knob.
Reviewed By: pdillinger, hx235
Differential Revision: D103568448
fbshipit-source-id: 9ec930343e434f1bee6130bcdbd7738dddd92b6d
Summary:
This fixes a GCC 16 unity-build failure in `Cache::CacheItemHelper`. The no-secondary-cache constructor delegated to the full constructor while passing this as `without_secondary_compat`. GCC 16 reports that pattern as `-Wmaybe-uninitialized` because the delegated constructor receives a pointer to the object under construction and its debug assertions dereference it. Since RocksDB builds with `-Werror`, the warning is promoted to a build error. We see it now because the unity job uses `gcc:latest`, which now has `GCC 16.1.0`. The code pattern itself dates back to PR https://github.com/facebook/rocksdb/issues/11299 / ccaa3225b from 2023.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14713
Reviewed By: xingbowang
Differential Revision: D104162382
Pulled By: mszeszko-meta
fbshipit-source-id: d6cb579e37bed0d833c6427c3b0541c857425d42
Summary:
Add MLIR-compatible PrefixVarint32/64 helpers in util/prefix_varint.h. Document the format, split decode API, and hot paths, and cover the codec in coding_test with round-trip, disk-read, overflow, and truncation cases.
I'm intending to use this in the new blog file format because you only need to read the first byte to know how many varint bytes to read total, and it might be more CPU efficient in other uses as well (to be determined in follow-up work).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14692
Test Plan: Unit tests included
Reviewed By: joshkang97
Differential Revision: D103245079
Pulled By: pdillinger
fbshipit-source-id: dca435beccb6e666d864a31d8683ad19e2121c34
Summary:
Reduces per-step heap allocations in the external table iterator's hot path
by replacing the `InternalKey` member of `ExternalTableIteratorAdapter` with
`IterKey`. `InternalKey` stores its bytes in a `std::string`, which
heap-allocates whenever a key exceeds the small-string threshold. `IterKey`
keeps a 39-byte inline buffer and only spills to the heap for larger keys.
`UpdateKey` runs on every `Next` / `Prev` / `Seek*`, so this swap removes a
recurring per-step allocation for typical key sizes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14695
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D103440780
Pulled By: joshkang97
fbshipit-source-id: deb0a8ea04110e3dff0a1bcad767e7136192fdc6
Summary:
See SC crash test failure 2729181374202803470.
Flush/compaction retention was protecting only explicit snapshots, while readers without an explicit snapshot were still bounded by GetLastPublishedSequence(). That mismatch becomes observable in WRITE_COMMITTED when commit_bypass_memtable runs: a newer version can already be installed into WBWI-backed immutable memtables before SetLastSequence() publishes it. This can happen with two-write-queue enabled OR disabled.
Example setup:
- K@P = "old_value" is the current published version of key K.
- K@U = "new_value" is written later with U > P.
- commit_bypass_memtable ingests K@U into immutable memtables, but publication is still at P.
- flush starts in that window and builds its snapshot context.
Before this change, if there was no explicit snapshot at P, flush/compaction was free to collapse K@P under K@U. A reader at the published boundary P then saw neither version: K@P had been discarded, while K@U was still too new, so Get(K) returned NotFound.
A simple solution is to add a "fake" snapshot boundary for pessimistic write committed txns.
NOTE: A behavioral side effect is that this will prevent a merge operand at snapshot seqno from being filtered via compaction filter because of the new managed snapshot. But this keeps it aligned with the behavior that write prepared and write unprepared have.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14681
Test Plan: Regression test that fails without fix, specifically K@P gets deleted and a snapshot read returns NotFound, when it should exist.
Reviewed By: hx235
Differential Revision: D102856943
Pulled By: joshkang97
fbshipit-source-id: 46cb3936e9d4e771966d39fd0d15436335bf6ccf
Summary:
This commit fixes several potential issues introduced by cc8d9ea04 (CI: AI review workflow improvements) that broke the agent review jobs in GitHub CI.
1. Codex workflows fail when OPENAI_API_KEY is not configured:
- Added Codex review workflows fail with exit code 1 if the API key secret is missing
- Added script-level checks to skip Codex steps gracefully when OPENAI_API_KEY is unset, avoiding CI noise
2. Missing pull-requests: read permission in manual-review job:
- The manual-review job calls github.rest.pulls.get() but lacked the required permission
- Added 'pull-requests: read' to the manual-review job permissions
3. Potential crash when headSha is undefined:
- parse-claude-review.js and parse-codex-review.js accessed meta.headSha.substring() without null check
- Added defensive checks to handle undefined headSha gracefully
4. Claude model claude-opus-4-7 incompatible with thinking API:
- The commit upgraded default model to claude-opus-4-7, but this model doesn't support the 'thinking.type.enabled' API used by the Claude Code action
- Error: 'thinking.type.enabled' is not supported for this model. Use 'thinking.type.adaptive' and 'output_config.effort' to control thinking behavior
- Reverted default model to claude-opus-4-6 and removed claude-opus-4-7 from options
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14696
Test Plan: Created a draft PR with this change and draft production code changes, including a branch on facebook/rocksdb. Go to https://github.com/facebook/rocksdb/actions/workflows/claude-review.yml and choose "Run workflow" with that branch and PR. (Changes in the PR do are not picked up for the agent code review workflows; the only way to test what is outside of main is with a branch on facebook/rocksdb)
Reviewed By: joshkang97
Differential Revision: D103453631
Pulled By: pdillinger
fbshipit-source-id: 15a82257b5762e9e0b1b393dc45c4b343c866f7a
Summary:
Two CPU-side optimizations to `DBIter`'s hot iteration path.
1. `PrepareValueInternal` always re-parses the internal key after `iter_.PrepareValue()`, even though most inner iterators report their value was already prepared and cannot have moved `iter_.key()`.
2. `ResetValueAndColumns` and `ResetBlobData` unconditionally clear wide-column / blob state at the top of every `Next()`, even on plain-value scans where neither has been populated since the last reset.
Both wins are pure CPU on hot, fully-cached workloads. The dirty-flag bookkeeping introduced for (2) is extracted into a small reusable `DirtyTracked<T>` utility so the same pattern can be reused for any T whose `Reset()` is non-trivial.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14684
Test Plan:
**Benchmark — `db_bench seekrandom`.** Each op = 1 Seek + 100 Next on a fully cached 10M-key DB (16 B key, 100 B value, no compression).
| | ops/sec |
|---|---:|
| BEFORE | 42,131 |
| AFTER | 45,269 |
| **Δ** | **+3,138 (+7.4%)** |
Reviewed By: xingbowang
Differential Revision: D103070805
Pulled By: joshkang97
fbshipit-source-id: 6b184a1aa559649c575d95be28ab43bbee9ffe64
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14685
D100235293 and D102823090 migrated the generated RocksDB BUCK file from
`//folly/experimental/coro:*` to `//folly/coro:*`, but did not update
the buckifier script that generates it in internal_repo_rocksdb. The next release would revert the change.
Update buckify_rocksdb.py to match, so the generated BUCK file stays consistent with the folly coro migration.
Reviewed By: nmk70
Differential Revision: D103096688
fbshipit-source-id: 9055769ed5e9893397c7504ada22e21980f59dd2
Summary:
- Upgrade the Claude review workflow models and add complexity-based thinking budget selection for the main review run.
- Keep AI review comments tied to the reviewed commit and restructure long reviews so high-severity findings stay visible while details live in a collapsible section.
- Add early auto-trigger gating plus Codex review/comment workflows, including shared comment-building and parsing helpers.
## Testing
- Not run. The branch refresh was a no-op rebase onto current `upstream/main`, and no new code changes were made during this task.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14659
Reviewed By: pdillinger
Differential Revision: D102224743
Pulled By: xingbowang
fbshipit-source-id: 39c25494062cbcd52d3e85f8a859b5a3907c758e
Summary: Migrate coro references across the repo to the non shim version.
Differential Revision: D102823090
fbshipit-source-id: 3bec59aed1f2645139a1ad9c18363e6af462e61f
Summary:
The previous pre-push hook auto-formatted, committed, and re-pushed on behalf of the user. This was fragile: it's not clear whether the format fix should amend the current commit or create a new one. Adding a commit breaks populating the summary for creating a new PR. Also, the hook's internal re-push would drop flags like --set-upstream from the original command. Replace with a simple check-only approach that blocks the push and tells the user what to fix.
Also add a new check for untracked source files (.cc, .h, .py, etc.) in tracked directories (excluding third-party/). These typically indicate files that were forgotten in the commit, which would cause the pushed code to fail to build.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14680
Test Plan: manual
Reviewed By: xingbowang
Differential Revision: D102855227
Pulled By: pdillinger
fbshipit-source-id: 7dc2c4e7a2b2c392bf8da74d7ea43883c8c075a9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14676
When fast_sst_open is disabled, RocksDB was still passing previously-persisted
file_open_metadata from the MANIFEST to NewRandomAccessFile. This could cause
failures when the metadata becomes stale (e.g. expired filesystem credentials).
This change gates the consumption of file_open_metadata in
TableCache::GetTableReader on the fast_sst_open option. When fast_sst_open is
false, previously persisted metadata is ignored and not passed to the filesystem
via FileOptions::file_metadata.
The fast_sst_open flag is threaded from MutableDBOptions through VersionSet ->
ColumnFamilySet -> ColumnFamilyData -> TableCache at construction time, ensuring
the gate is active before any table readers are opened during recovery. Dynamic
changes via SetDBOptions are also propagated to all existing TableCache instances.
Reviewed By: mszeszko-meta, xingbowang
Differential Revision: D102735581
fbshipit-source-id: 9a2c4dc0644a2f65c36b2468605df57779e127cd
Summary:
Fix false internal warnings about skipped tests (seen on D102718613).
ROCKSDB_GTEST_SKIP signals a gap in test coverage due to the build/execution
environment, while ROCKSDB_GTEST_BYPASS is for intentionally and permanently
omitting certain parameterizations. These six tests skip based on the test
parameterization (forward vs reverse, primary vs secondary mode), not due to
any environment limitation, so BYPASS is the correct macro.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14679
Test Plan: these are tests, look at internal signals
Reviewed By: joshkang97
Differential Revision: D102833346
Pulled By: pdillinger
fbshipit-source-id: e57bd1e217ca63aaf418b68c788563e67bbb07e3
Summary:
- External Table is only PUTs with no seqno, so we can route it to the simpler more efficient `GetContext::SaveValue`.
- `SstFileReader::Get` was allocating a string to append key footer, instead use LookUpKey to allocate on stack for short keys.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14673
Test Plan: CI passes
Reviewed By: anand1976
Differential Revision: D102659533
Pulled By: joshkang97
fbshipit-source-id: e55127548c9e4b7bdeb63c46acfdb8eb5883db14
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14672
This fixes the forward path introduced in the original pull request: https://www.internalfb.com/diff/D102409405
That change tried to avoid using user-owned `iterate_upper_bound_` memory by reading `iter_.key()` when the internal iterator was still valid and using it as the exclusive end key for the synthetic range tombstone. That is not safe: after the upper-bound break, `iter_.key()` is only the current merged child position, not a key the bounded scan proved safe to read or use as a range endpoint.
Example: run a prefix-bounded scan over prefix `b` with tombstones `ba`, `bb`, `bc`, `iterate_upper_bound = "bd"`, live key `be` in an SST, and later same-prefix key `bz` in the memtable. The SST child can drop out when it notices `be >= bd`, while the memtable child is not upper-bound aware and can still surface `bz`. If cleanup uses `iter_.key() == "bz"`, it inserts `[ba, bz)` and covers the live `be`, even though that key was never safe for the bounded scan to reason about.
The fix is to always end the forward range at `saved_key_`, which is the last tombstone user key the iterator actually proved deleted. That is conservative but safe. The regression test builds the mixed SST/memtable example above and asserts that the live `be` remains visible. The diff also re-enables randomized `min_tombstones_for_range_conversion` in `db_crashtest.py` and forces it back to `0` when `use_multiscan == 1`, since multiscan still does not support read-path range conversion.
Reviewed By: xingbowang
Differential Revision: D102493855
fbshipit-source-id: 45bdb86464268e2c4fa2ea735dbb059a5e054f28
Summary:
Replace raw owning pointers with std::unique_ptr for StreamingCompress/Uncompress in Create functions and log::Writer/Reader, eliminating manual delete calls in destructors. Also update the StreamingCompressionTest to use unique_ptr for its local StreamingCompress, StreamingUncompress, and MemoryAllocator objects.
Bonus: improve error message for missing/broken ruby (new build requirement)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14665
Test Plan: ./log_test - all 211 tests pass.
Reviewed By: xingbowang
Differential Revision: D102381893
Pulled By: pdillinger
fbshipit-source-id: 02806b46c8371474a969d6fa3012fff5eb7886c3
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14662
When a file becomes obsolete, `ReleaseObsolete()` calls `ReleaseAndEraseIfLastRef()` to remove it from the table cache. However, if concurrent readers hold references to the cache entry, this is not the last ref, so the entry stays. When those readers later call plain `Release()`, the entry becomes unreferenced but remains in the cache — there is no mechanism to trigger cleanup.
The fix is to call `cache->Erase()` in `ReleaseObsolete()` instead of relying solely on `ReleaseAndEraseIfLastRef()`. `Erase()` marks the entry as Invisible in the cache. The cache's `Release()` implementation already checks `IsInvisible()` and erases Invisible entries when the last reference is released. This guarantees that obsolete file entries are cleaned up regardless of concurrent reader timing.
Also reverts the D102158618 workaround that reordered `EraseUnRefEntries()` before `TEST_VerifyNoObsoleteFilesCached` in `CloseHelper`. With the root cause fixed, the assertion correctly passes in its original position.
Reviewed By: pdillinger
Differential Revision: D102158618
fbshipit-source-id: 5796abee916dfbd99554a36d3bbf579842d2fb8b
Summary:
Seems like crashtests are failing more often now, disabling again to investigate.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14670
Reviewed By: archang19
Differential Revision: D102479287
Pulled By: joshkang97
fbshipit-source-id: 3f147f58b0cf2901ffe3b99e8681ed5a6085e7c8
Summary:
`blob_compression_opts` is dynamically changeable through `SetOptions()`, but
blob direct write cached its per-thread compressor by `CompressionType` only.
After updating `blob_compression_opts`, new blob records on an existing writer
thread could keep using stale compression settings.
This change threads `CompressionOptions` through the direct-write settings and
rebuilds the cached per-thread compressor whenever the published options for
that compression type change. When the options stay the same, the steady-state
cache behavior is unchanged.
It also adds a regression test that toggles the ZSTD checksum flag through
`SetOptions()` and inspects the raw blob records to confirm each direct-write
record used the latest compression options.
## Testing
- `timeout 60s ./db_blob_direct_write_test --gtest_filter='DBBlobDirectWriteTest.DirectWriteCompressionOptionsUpdateRebuildsCachedCompressor' --gtest_repeat=5`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14669
Reviewed By: jainraj91
Differential Revision: D102423376
Pulled By: xingbowang
fbshipit-source-id: 504243d72cf22e62097f7e88ac1cc36c5bf29eee
Summary:
kNoChecksum was not being tested in db_crashtest.py despite being a supported checksum type. Add it to the random selection to ensure crash test coverage of the no-checksum code path.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14667
Test Plan: Ran multiple blackbox and whitebox crash test validation runs with kNoChecksum forced and with random selection including kNoChecksum. All passed crash-recovery verification.
Reviewed By: hx235
Differential Revision: D102404837
Pulled By: pdillinger
fbshipit-source-id: 0120f1269c917dd6e347764f2bdc0095d46ff71a
Summary:
`DBIter::FindNextUserEntryInternal` previously used `iterate_upper_bound_` (user-controlled memory) as the exclusive end-key when flushing an accumulated tombstone run. That pointer is dangerous as it points to user memory space, which can be modified at any moment.
This PR removes that dependency: the end-key is always derived from RocksDB-owned data — either the live `iter_` key that broke the run, or `saved_key_` when `iter_` is exhausted.
The tradeoff here is that there is potentially 1 less tombstone to get covered in this path, which is reasonable.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14664
Test Plan: Updated unit tests, CI passes
Reviewed By: xingbowang
Differential Revision: D102409405
Pulled By: joshkang97
fbshipit-source-id: b2beed73e5fe5b955b2d782fc467eda637c4fcf2
Summary:
CONTEXT: The manifest validation on close feature (verify_manifest_content_on_close) detects corruption but does not increment any statistics counter, making it harder to monitor in production.
WHAT: Add a new ticker MANIFEST_VALIDATION_FAILURE_COUNT that is incremented each time content validation detects manifest corruption during DB::Close(). The counter fires per corruption detection, so it can increment up to 2 times per close (once on initial check, once after rewrite attempt). Updated all existing manifest validation tests to verify the counter value.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14657
Test Plan:
- All 7 manifest validation tests pass with new stat assertions
- 5x repeat with COERCE_CONTEXT_SWITCH=1 shows no flakiness
- Full version_set_test suite (212 tests) passes
Reviewed By: anand1976
Differential Revision: D102404260
Pulled By: dannyhchen
fbshipit-source-id: 21a0aa1ad8de12a935caf5642e41ccf2a47b46d9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14658
Follow up to D101463511. Add a hook and a manager-aware overload of so custom CompressionManagers can provide human-readable names for custom compression types while preserving the existing generic fallback when no compatible manager is available.
Reviewed By: pdillinger
Differential Revision: D102201365
fbshipit-source-id: 0c7456bb9db2e54927a4349d12c035fc8b5ad562
Summary:
There is a separate way to handle incompatibility. Revert this.
- revert the temporary workaround from https://github.com/facebook/rocksdb/issues/14656 that serialized `InternalStats::CompactionStats::counts` with the pre-11.2 compaction-reason count
- restore remote compaction stats serialization to use the full `CompactionReason::kNumOfReasons` array size again
## Context
https://github.com/facebook/rocksdb/issues/14656 temporarily shrank the serialized `counts` array to keep remote compaction metadata readable across the 11.1/11.2 boundary. This follow-up removes that special case because the compatibility issue is being handled separately.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14663
Test Plan: - Not run in this metadata-prep task
Reviewed By: jaykorean
Differential Revision: D102376258
Pulled By: xingbowang
fbshipit-source-id: 55693ea25a97b70de9d53b0d6eebcbd88a325b1c
Summary:
Fixes a correctness bug in read-path range-tombstone synthesis when it races with `IngestExternalFile`. The synthesis path could insert a tombstone into the active memtable at a snapshot's sequence number, while ingestion installed an L0 SST at `LastSequence + 1` — a higher seqno than the synthesized tombstone. This breaks the main assumption of range tombstone reads that all lower levels have lower seqno.
The fix introduces a per-CF `port::RWMutex` (`ColumnFamilyData::ingest_sst_lock_`) plus a per-memtable `ingest_seqno_barrier_`. Ingestion takes the read lock and range tombstone synthesis **tries** to take a write lock.
If iterator lock is successful, then we have a new updated barrier seqno that we can validate the iterator seqno against. An added benefit is we no longer need to gate against empty memtable. This was originally added as an easy fix to prevent memtables from being inserted into while ingestion was happening.
## The bug, by example
`ReadPathRangeTombstoneTest.NewerPointInOlderFileStillVisible` (`db/db_iterator_test.cc:6929`):
1. L0 file `b@1, c@2, d@3`, then L0 file `Delete(b)4, Delete(c)5`.
2. Active memtable: `Put(z)6`. Snapshot taken at seq 6.
3. `IngestExternalFile({c → "vc_live"})` → installed at L0 with seq 7.
4. Iterator at snap 6 walks the deletion run and synthesizes `[b, d) @ seq 6` into the active memtable via `MemTable::AddLogicallyRedundantRangeTombstone`.
5. `Get("c")` at the latest snapshot: memtable returns covering tombstone (seq 6), `Version::Get` short-circuits, **never reads `c@7`** — returns `NotFound` instead of `"vc_live"`.
The invariant `Version::Get` relies on (memtable seqs ≥ any L0 seq for the same key) is broken because synthesis writes at the *snapshot's* seq while ingestion writes at `LastSequence + 1`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14654
Test Plan:
- Updated regression test
- Re-enable crashtests and manually run
| Flavor | Jobs |
|---|---:|
| `fbcode_blackbox_crash_test` | 200 |
| `fbcode_whitebox_crash_test` | 30 |
| `fbcode_asan_blackbox_crash_test` | 30 |
| `fbcode_tsan_blackbox_crash_test` | 30 |
| `fbcode_crash_test_with_atomic_flush` | 30 |
| `fbcode_crash_test_with_wc_txn` | 30 |
| `fbcode_crash_test_with_ts` | 30 |
| **Total** | **380** |
Reviewed By: xingbowang
Differential Revision: D102044512
Pulled By: joshkang97
fbshipit-source-id: 0c69187595edc5a5fa80be24bffdba710a92e56e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14637
Add public API ParseCompressionNameForDisplay() to render TableProperties::compression_name in human-readable form for both legacy and format_version >= 7 SST metadata, including BuiltinV2/compression-manager encodings and filtered no-compression markers for display.
Reviewed By: xingbowang
Differential Revision: D101463511
fbshipit-source-id: 51c44850183cb9757644a323376844adffc7a8c2
Summary:
- Keep remote compaction stats serialization compatible with RocksDB 11.1.
- Serialize `InternalStats::CompactionStats::counts` using the pre-11.2 compaction-reason count so remote compaction metadata remains readable across the version boundary.
- Preserve the existing formatting-only follow-up commit on the branch.
## Notes
- Needs a better mechanism to make OptionTypeInfo::Array ser/des friendly. Will follow up after 11.2 release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14656
Reviewed By: pdillinger
Differential Revision: D102071507
Pulled By: xingbowang
fbshipit-source-id: 6eeb87d86494f528f0d1ce46ce4c3e8e7dd2da53
Summary:
This reverts commit `bad2d5b0af7e160de5ca58869f40941af0f9da95`.
The reverted change switched fbcode Buck dependencies from `//folly/experimental/coro:*` to `//folly/coro:*` in:
- `BUCK`
- `buckifier/buckify_rocksdb.py`
That change broke the internal build, so this PR restores the previous dependency paths.
## Testing
- Not run locally; this task only fetched, rebased, and verified the existing revert branch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14652
Reviewed By: archang19
Differential Revision: D101980062
Pulled By: xingbowang
fbshipit-source-id: 9a6edf14c226caf87e2c63c7f2806d409dc809ef
Summary:
**Summary:**
Assert the DB::Open status before asserting on the DB pointer in FaultInjectionTest::OpenDB(). This avoids aborting the test process when reopen fails and preserves the underlying RocksDB error for ASSERT_OK(OpenDB()). This is to deal with recent failures on a host with nearby tests failing with space issue already.
```
[ RUN ] ExternalSSTFileBasicTest.LargeSizeSstFileWriter
db/external_sst_file_basic_test.cc:3303: Failure
sst_file_writer.Finish()
IO error: No space left on device: While appending to file: /dev/shm/rocksdb_testt/run-external_sst_file_basic_test-shard-3/external_sst_file_basic_test_578578_8746831521190564136/large_key.sst: No space left on device
[ FAILED ] ExternalSSTFileBasicTest.LargeSizeSstFileWriter (8551 ms)
[ RUN ] FaultTest/FaultInjectionTestSplitted.FaultTest/2
fault_injection_test: db/fault_injection_test.cc:269: rocksdb::Status rocksdb::FaultInjectionTest::OpenDB(): Assertion `db_ != nullptr' failed.
Received signal 6 (Aborted)
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14646
Test Plan: Test changes only
Reviewed By: anand1976
Differential Revision: D101742874
Pulled By: hx235
fbshipit-source-id: de3b9349d9d067dd8f538b44ac08d108924faed3
Summary:
**Summary:**
When remote compaction is enabled with skip_stats_update_on_db_open=true, the remote worker's secondary DB skips UpdateAccumulatedStats(), leaving FileMetaData fields (num_entries, num_range_deletions) at 0. This breaks standalone range deletion file filtering in
FilterInputsForCompactionIterator() (https://github.com/facebook/rocksdb/blob/b374ba5968bdd6b16ff58918c561781c8bdab442/db/compaction/compaction.cc#L1085 + https://github.com/facebook/rocksdb/blob/b374ba5968bdd6b16ff58918c561781c8bdab442/db/version_edit.h#L462) , causing the primary and remote worker to disagree on input key count and triggering a "Compaction number of input keys does not match number of keys processed" corruption error.
Repro test (should be landed with the fix in the future): https://github.com/hx235/rocksdb/commit/aa1225c69cd6b6e2d2af735e1d1d9b147edfd6ae
```
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from CompactionServiceTest
[ RUN ] CompactionServiceTest.StandaloneRangeDeletionWithSkipStatsUpdateOnOpen
db/compaction/compaction_service_test.cc:2970: Failure
s
Corruption: Compaction number of input keys does not match number of keys processed. Expected 0 but processed 4. Compaction summary: Base version 15 Base level 5, inputs: [20(1280B)], [18(1270B filtered:true) 19(1270B filtered:true)]
[ FAILED ] CompactionServiceTest.StandaloneRangeDeletionWithSkipStatsUpdateOnOpen (1257 ms)
[----------] 1 test from CompactionServiceTest (1257 ms total)
[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (1257 ms total)
[ PASSED ] 0 tests.
[ FAILED ] 1 test, listed below:
[ FAILED ] CompactionServiceTest.StandaloneRangeDeletionWithSkipStatsUpdateOnOpen
1 FAILED TEST
```
This can happen even when standalone range deletion ingestion is disabled in the current run, because such files may have been ingested in a previous crash test run with different randomized options. So we disable skip_stats_update_on_db_open unconditionally when remote compaction is active until the real fix lands .
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14647
Test Plan:
1. Verify skip_stats_update_on_db_open is forced to 0 with remote compaction
2. Verify skip_stats_update_on_db_open can still be 1 without remote compaction:
Reviewed By: anand1976
Differential Revision: D101744943
Pulled By: hx235
fbshipit-source-id: c1dcd8943433e8a9dfb7385c998be2c339e3b859
Summary:
**Summary:**
Temporarily disables inplace_update_support in db_crashtest.py. See the db_crashtest.py new comments for mow.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14645
Test Plan: - Run db_crashtest.py without passing inplace_update_support option, with short interval, and verify each run shows inplace_update_support is false in the db_stress command line.
Reviewed By: joshkang97
Differential Revision: D101739997
Pulled By: hx235
fbshipit-source-id: 39936b8282e392e688afea5e9bd074b60138f210
Summary:
Restore the iterator contract for blob-backed wide-column entities by eagerly materializing all blob-backed columns before a `DBIter` entry is exposed as valid. Also capture similar error inside compaction filter V4, so that the error is captured a surfaced up like older compaction filter V3.
## Problem
Lazy wide-column blob resolution let `DBIter` report `Valid() == true` before all columns were actually prepared. Callers could read `value()` successfully, then hit a blob-read failure in `columns()`, which flipped the iterator to invalid after the entry had already been observed as valid.
## Solution
- Restore back to old behavior.
- Resolve and materialize all blob-backed wide columns during iterator positioning.
- Keep `columns()` as a pure accessor once `Valid()` is true.
- Add coverage for eager blob resolution on iterator scans.
- Add coverage for blob resolution failures invalidating the iterator before the entry is exposed.
- Capture the same error in compaction filter and set db to bg error status if the error is not retriable.
## Next step
If we want lazy iterator resolution in the future, we will need an API that can surface lazy resolution failures explicitly instead of hiding I/O and status transitions inside value() or columns().
## Testing
- `make -j14 db_wide_blob_direct_write_test`
- `timeout 60s ./db_wide_blob_direct_write_test --gtest_filter='*DirectWriteIteratorValueScanEagerlyResolvesBlobColumns:*DirectWriteIteratorBlobResolutionErrorInvalidatesEntry'`
## Task
T265294130
# Bonus fix
## Non-compaction blob filtering fixes
### 1. Flush-time wide-entity lazy blob resolution now works correctly
Blob direct-write can leave wide-entity blob references in memtables before flush. Flush already runs through `CompactionIterator`, but outside a real compaction it did not have blob read support. As a result, `FilterV4` lazy resolution on flush could fail immediately with
`NotSupported("Blob fetcher not available")` instead of either:
- resolving the blob successfully, or
- surfacing the real blob read error and failing the flush
This stack fixes that by plumbing blob read support into non-compaction table-file creation. `BuildTable()` now gives `CompactionIterator` enough context to construct a `BlobFetcher` for flush/recovery table building, including write-path fallback for direct-write blob files
that are not yet manifest-visible.
With that in place:
- flush-time `FilterV4` can lazily resolve blob-backed wide columns
- if resolution fails, the failure is latched and flush fails after `FilterV4()` returns
- `bg_error` is set instead of silently preserving the entry
### 2. Plain blob fallback outside compaction now uses the resolved user value
For plain blob-backed values, `FilterBlobByKey()` is allowed to return `kUndetermined`, which is supposed to fall back to the normal value-based filter path. That fallback worked in compaction, but the non-compaction `BuildTable()` path still treated plain blob indexes as
unsupported/corrupt because it could not read the blob.
This stack fixes that behavior for flush/recovery table-file creation:
- when `FilterBlobByKey()` returns `kUndetermined`, RocksDB now eagerly resolves the plain blob value
- `FilterV2`/`FilterV3`/`FilterV4` then see the actual user value as `ValueType::kValue`, not the `BlobIndex` encoding
- legacy filters therefore behave the same way outside compaction as they already do inside compaction
Wide-column entities still use the `FilterV4` lazy resolver path. The plain-blob fix is specifically about restoring the documented fallback behavior for blob-backed `kValue` records outside compaction.
## Why this matters
Without these last two commits, flush-time filtering still had two inconsistent behaviors:
- wide-entity lazy resolution could not actually read blob-backed columns in the direct-write case
- plain blob-backed values could not fall back to normal value-based filtering outside compaction
So even after fixing iterator validity and compaction error propagation, non-compaction table-file creation still had remaining contract holes. These commits make flush/recovery behave consistently with compaction.
## Coverage added
The tests added in these commits cover:
- flush-time plain-blob `FilterV3` fallback using the resolved user value
- flush-time plain-blob `FilterV4` fallback using the resolved user value
- flush-time direct-write wide-entity lazy resolver success
- flush-time direct-write wide-entity lazy resolver failure on missing blob file, including flush failure and `bg_error` latching
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14632
Reviewed By: anand1976
Differential Revision: D101425233
Pulled By: xingbowang
fbshipit-source-id: 695b2df5189033d30309b35849815e31fd965664
Summary:
Fix nightly build regressions introduced by recent CI/toolchain changes.
This PR includes two parts:
- CMake/link fixes for the Folly and Folly Lite nightly jobs
- a targeted nightly workflow fix for the clang-21 ASAN/UBSAN + Folly job
Changes:
- link gflags explicitly for USE_FOLLY CMake tool and benchmark targets
- resolve USE_FOLLY_LITE glog via its installed library path instead of bare -lglog
- in the clang-21 nightly job, build Folly/getdeps with gcc/g++ instead of inheriting clang-21 for third-party dependency builds
- disable ccache only for that standalone Folly build step so getdeps/CMake does not inject the broken ccache compiler launcher for ASM
Root cause:
- recent CI/container changes exposed missing explicit link dependencies in the CMake Folly paths
- the clang-21 nightly job exported CC/CXX at job scope, so Folly getdeps inherited clang-21 into libiberty/binutils build logic that expects GCC-style driver behavior such as -print-multi-os-directory
- the same standalone Folly build path also misbehaved when ccache was auto-detected and used as a compiler launcher for assembler-with-cpp inputs
Verification:
- make format-auto
- git diff --check
- local CMake configure sanity check for default config
- upstream nightly reruns used to confirm failure signatures before and after the first-round fixes
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14609
Reviewed By: anand1976
Differential Revision: D100695693
Pulled By: xingbowang
fbshipit-source-id: 703546ad3ddc518ab80c936442709d65fe2d22af
Summary:
The prior fix (https://github.com/facebook/rocksdb/issues/14586) disabled read-path range tombstone synthesis at the `DBIter` level when `table_filter` was set. However, synthesized tombstones persist in the memtable beyond the lifetime of the iterator that created them. A prior unfiltered iterator can synthesize a range tombstone that then silently affects a subsequent filtered iterator's results — the filtered scan sees the tombstone but not the SSTs it was derived from, allowing hidden SST state to corrupt the filtered view. See failed crash test in T264151327.
This PR takes the stronger approach of rejecting iterator creation outright (`InvalidArgument`) when `ReadOptions::table_filter` is used on a column family with `min_tombstones_for_range_conversion > 0`. This eliminates the entire class of interaction bugs between the two features rather than trying to suppress conversion in individual code paths.
## Example
- `min_tombstones_for_range_conversion = 2`
- L1 SST_keep: `Put(a), Put(b), Put(c)` (`c` is the live boundary)
- L0 SST_dels: `Delete(a), Delete(b)` (2 contiguous tombstones, newer)
### Step 1 — Iterator A (no filter)
A walks `Del(a) Del(b)` (2 contig) then `c` (live).
Synthesizes range tombstone `[a, c)` into the active memtable.
### Step 2 — Iterator B (`table_filter` skips SST_dels)
B's filter excludes SST_dels, so `a, b` from SST_keep should appear live.
But the memtable now holds range tombstone `[a, c)` from step 1, which
applies to B regardless of `table_filter`. B sees only `c` —
**a, b are silently hidden**.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14618
Test Plan:
- `TableFilterNotAllowed` test validates the rejection path end-to-end: unfiltered scan synthesizes the tombstone, filtered iterator is rejected with `InvalidArgument`, and the full-DB view remains correct via `VerifyIteration`.
- Crash test sanitization ensures `db_stress` won't hit the incompatible configuration.
- Run crash test sanitization unit test: `python3 tools/db_crashtest_test.py`
- Run iterator tests: `make db_iterator_test && ./db_iterator_test --gtest_filter="*TableFilterNotAllowed*"`
Reviewed By: xingbowang
Differential Revision: D100873156
Pulled By: joshkang97
fbshipit-source-id: dd0d123d94b1bfda2a4a5040681fcc7cf14f760e
Summary:
There is an assumption in the code that within a given prefix all keys can be seen after a seek to that prefix. The problem occurs when there is a Next() followed by a Prev() back into the same prefix. The merging iterator actually uses SeekForPrev during Prev for non-child iterators, and as a result changes the prefix being used. Now that we are back to our original prefix, the view is incomplete.
The fix is to simply disable range tombstone conversion in this legacy prefix mode.
### Example
Use a 1-byte prefix extractor, so the prefix is just the first character.
Think of the merged iterator as combining two children:
- Child A: `b8`
- Child B: `b7, c1`
Global sorted order is:
```text
b7, b8, c1
```
So:
- predecessor of `c1` should be `b8`
- predecessor of `b8` should be `b7`
## Sequence
Start in legacy prefix mode:
- `total_order_seek = false`
- `prefix_same_as_start = false`
Now do:
1. `Seek("b8")`
2. `Next()`
3. `Prev()`
## What should happen
```text
Seek("b8") -> b8
Next() -> c1
Prev() -> b8
```
Because in the global order, `b8` is immediately before `c1`.
## What the buggy iterator did
```text
Seek("b8") -> b8
Next() -> c1
Prev() -> b7
```
So it skipped `b8`.
## Why it happens
When `Prev()` switches direction, the merging iterator tells all non-current
children to `SeekForPrev(current_key)`.
At that moment:
- current key is `c1`
- so the other children get `SeekForPrev("c1")`
But in legacy prefix mode, those child seeks are allowed to use prefix
filtering. So a child that only contains `b*` keys can effectively say:
```text
"I was asked for prefix c; I do not have prefix c."
```
That child drops out instead of returning its real global predecessor `b8`.
Then the current child simply moves back from `c1` to `b7`, and the merged
result becomes `b7`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14625
Test Plan: Update unit tests to disable feature.
Reviewed By: xingbowang
Differential Revision: D101259748
Pulled By: joshkang97
fbshipit-source-id: 23e19c470f931406f96a34a7baebca89e91c7e39
Summary:
- meter blob garbage during `BuildTable()` when flush input can contain direct-write blob references, so overwrite elision and flush-time compaction filters register garbage in the manifest
- plumb `BlobFileGarbage` through flush and recovery manifest edits even when the flush produces no SST output
- add wide-column direct-write tests covering TTL-filtered flushes with both mixed live/expired records and all-expired input
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14623
Reviewed By: anand1976
Differential Revision: D101212841
Pulled By: xingbowang
fbshipit-source-id: e5803ddaf17b5dfd92312e42191b76e311257f3b
Summary:
This fixes the Claude review workflow YAML parse error and adds a CI validation step for GitHub Actions workflow YAML via `make check-
workflow-yaml`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14624
Reviewed By: archang19
Differential Revision: D101225058
Pulled By: xingbowang
fbshipit-source-id: 0f371dcbd70ece2c25220c96e43305d9665554b1
Summary:
Detect out-of-space failures from combined db_stress stdout/stderr in the Python wrapper so both OpenAndCompact stdout failures and verification stderr failures trigger the same diagnostics.
When a match is found, print filesystem usage for /dev/shm and the db roots, then summarize per-directory and per-extension usage to make it clear which files consumed space. Add unit coverage for the failure matcher and suffix accounting.
This help triage any regression in additional file usage in stress test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14619
Reviewed By: archang19
Differential Revision: D100984310
Pulled By: xingbowang
fbshipit-source-id: 23765ff405f0e64382e601b0da173ab6b37dba6d
Summary:
`RangeTreeLockManager::CompareDbtEndpoints()` called `Comparator::Compare()` on range lock endpoint keys that never contain user-defined timestamps. With a timestamp-aware comparator (`timestamp_size() > 0`), this caused assertion failures in debug builds and silent endpoint misordering in release builds.
Replace `Compare()` with the 4-argument `CompareWithoutTimestamp()` using `a_has_ts=false, b_has_ts=false`, which is the correct contract for serialized range lock endpoints (format: `[1-byte suffix][key bytes]`, no timestamp).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14611
Test Plan:
- New test `RangeLockWithTimestampComparator` reopens with `BytewiseComparatorWithU64Ts` and exercises range lock acquisition, conflict detection, and non-overlapping success with short keys.
- Verified RED (assertion failure) before fix, GREEN after.
- Full `range_locking_test` suite passes (16/16).
- Stress tested with `COERCE_CONTEXT_SWITCH=1 --gtest_repeat=5`.
Reviewed By: xingbowang
Differential Revision: D100775201
Pulled By: laurynas-biveinis
fbshipit-source-id: 58e3846e62e9b3cc5bdc69557458b245f90b3967
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14596
When DBOptions::fast_sst_open is enabled, RocksDB retrieves opaque file system
metadata for SST files after flush, compaction, and external file ingestion via
FSRandomAccessFile::GetFileOpenMetadata(). This metadata is persisted in the
MANIFEST using a new forward-compatible NewFileCustomTag (kFileOpenMetadata = 17),
and passed back to the file system via FileOptions::file_metadata on subsequent
file opens. This accelerates DB open time on remote storage systems by allowing
the file system to skip expensive metadata RPCs.
The feature is gated by DBOptions::fast_sst_open (default false). Everything
works seamlessly regardless of the option setting, file metadata support, or
presence/absence of the metadata in the MANIFEST. The MANIFEST change is backward
compatible - older RocksDB versions safely ignore the new tag.
Reviewed By: xingbowang
Differential Revision: D100220973
fbshipit-source-id: f52de9dd853a50653b3297ab4a37a868fe41cc04
Summary:
Temporarily disable the feature until all fixes land and crash tests are stabilized.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14617
Reviewed By: xingbowang
Differential Revision: D100853151
Pulled By: joshkang97
fbshipit-source-id: f2c9018f0b55891f6c0623902c3b961bc7f623aa
Summary:
Blackbox crash tests intentionally terminate `db_stress` with `SIGTERM` on timeout. When `io_uring` is enabled, that shutdown can emit the expected
`PosixRandomAccessFile::MultiRead: io_uring_submit_and_wait returned terminal error: -9.`
message on `stderr`, which currently makes the timeout path look like a real failure.
This change filters only that known post-`SIGTERM` stderr after a timeout when the process output confirms the `SIGTERM` handler ran. Any other stderr is still surfaced and fails the test, while the ignored lines are appended to `stdout` so the signal remains visible in logs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14613
Test Plan:
- Added unit tests covering fully ignored post-`SIGTERM` stderr
- Added unit tests covering mixed stderr where unrelated lines must still fail
- Added unit tests covering guard conditions when the run did not time out or did not print the `SIGTERM` marker
- Not run (not requested)
Reviewed By: archang19
Differential Revision: D100806204
Pulled By: xingbowang
fbshipit-source-id: 01248a371afae91bb43df55f88400baf155373f5
Summary:
- add a `docs/components/` landing page and a stress-test docs index
- document the `db_stress` expected-state trace/replay lifecycle, file invariants, and prefix-recovery contract in `expected_state_trace.md`
- align `db_stress` comments with the restore semantics: missing trace entries are fatal, while extra tail trace entries are tolerated
- keep the new docs tree trackable and point repo instructions at the new component-docs entrypoint
## Testing
- Not run (documentation and comment updates only)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14612
Reviewed By: joshkang97
Differential Revision: D100797173
Pulled By: xingbowang
fbshipit-source-id: 25be8c6239b9fdd84580818efe7520c371f9a46b
Summary:
BlockBuilder was somewhat inconsistent in its treatment of Slices whose size exceeds 4GB, which in a random corruption case could lead to (for example) serializing only the bottom 32 bits of a value size (uncorrupted) but appending the full multi-GB (corrupted) size. Because this is inner loop code, we don't want to pay CPU for extra conversions, data movements, or Status plumbing (already ruled out by Xingbo).
In this change we make BlockBuilder more internally consistent and lift the 32-bit size requirement to callers (of a few functions specifically, for now). To ensure that's satisfied, I've added additional checks near the perimeter of RocksDB to ensure keys and values do not exceed 4GB, plus an extra random corruption or backstop check in BlockBasedTableBuilder. In detail,
BlockBuilder (block_builder.cc/h):
* Add API comments documenting the < 4GB assumption on Add/AddWithLastKey
* Add debug assertions verifying input slice sizes < 4GB
* Simplify AddWithLastKeyImpl to use uint32_t locals, reducing static_cast
* Use only bottom 32 bits when appending derivative slices for consistency
* Update MaybeStripTimestampFromKey to also truncate to 32-bit size
* Add FIXME comments where buffer_/values_buffer_ sizes are truncated
BlockBasedTableBuilder (block_based_table_builder.cc):
* Add value size check (> uint32_t max) as a safety net against random corruption, since we have seen such corruptions in production
WriteBatch (write_batch.cc):
* Tighten existing key size checks to account for kNumInternalBytes (8-byte internal key suffix), using new kMaxWriteBatchKeySize constant
* Add missing size checks to Delete, SingleDelete, and DeleteRange (both Slice and SliceParts variants)
SstFileWriter (sst_file_writer.cc):
* Add key and value size checks in AddImpl and DeleteRangeImpl, which bypass WriteBatch and go directly to the table builder
MergeHelper (merge_helper.cc):
* Add merge result size check (> uint32_t max) in both overloads of TimedFullMergeImpl, returning Corruption if exceeded
* Add PartialMergeMulti result size check in MergeUntil
CompactionIterator (compaction_iterator.cc):
* Add size check on compaction filter output values (kChangeValue and kChangeWideColumnEntity), returning Corruption if > 4GB
meta_blocks.cc:
* Use Slice with uint32_t-truncated sizes in PropertyBlockBuilder::Finish to handle potentially oversized user property collector output
Needed follow-up:
* Check for blocks that are mis-encoded due to overflowing 32 bits for restart point offsets (and similar). See FIXME comments in block_builder.cc for why this is tricky.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14461
Test Plan:
New extreme-size unit tests, along with some testing infrastructure improvements:
Added test::HasBigMem() (in test_util/testharness.h) which returns true when the system has ≥128GB RAM (via sysconf(_SC_PHYS_PAGES) on Linux/macOS) or when ROCKSDB_BIGMEM_TESTS is set. All extreme-size tests use HasBigMem() to skip gracefully on smaller machines rather than being permanently disabled.
Where possible, tests use MemMapping::AllocateLazyZeroed() to supply large keys/values as Slices backed by anonymous mmap (cleaner with new MemMapping::AsSlice()). On Linux, read-only access to these pages maps to the shared kernel zero page, so the source data consumes no physical RAM — only the destination copy (e.g., WriteBatch::rep_) materializes, cutting peak memory roughly in half vs. std::string.
Tests (all enabled, skip via HasBigMem()):
./write_batch_test --gtest_filter='*LargeKeyValueSizeLimit*'
./external_sst_file_basic_test --gtest_filter='*LargeSizeSstFileWriter*'
./merge_test --gtest_filter='*LargeMergeResultRejected*'
./merge_helper_test --gtest_filter='*LargePartialMergeResultRejected*'
./db_compaction_test --gtest_filter='*CompactionFilterLargeValueRejected*'
All 5 pass (verified on a 128GB+ machine). On smaller machines, all 5 bypass cleanly with "insufficient memory for reliable continuous testing".
Reviewed By: xingbowang
Differential Revision: D96521899
Pulled By: pdillinger
fbshipit-source-id: 70f4b5e6a23ab074d60e653fbb7ddc5edbe162ab
Summary:
Reverts the two commits https://github.com/facebook/rocksdb/pull/14578https://github.com/facebook/rocksdb/pull/14575 that changed tracer behavior. The correct behavior is that trace should be written to before WAL.
Although a trace file may not be fully consumed, it is recycles after each crash, so there is no need to worry about applying an uncomited write. The original crash test was a false positive for a different error.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14600
Reviewed By: xingbowang
Differential Revision: D100397989
Pulled By: joshkang97
fbshipit-source-id: 3da6fb80f682ac6f9529c1a76eaf169e38cf2477
Summary:
T263957043 reproduces in whitebox crash testing when DBIter converts a run of point deletes into a range tombstone while reading with an older user-defined timestamp. The scan can observe an older delete for the boundary keys but miss newer live versions of the same keys and interior keys, so the synthesized tombstone is based on partial visibility and later hides valid max-timestamp reads.
## Problem
Read-path range tombstone conversion (`min_tombstones_for_range_conversion`) assumes the scan observes all interior live keys between the start and end of a contiguous delete run. With user-defined timestamps, a read at an older timestamp can see deletes but miss newer Puts for the same keys. The synthesized range tombstone then incorrectly covers those newer versions, causing data loss on subsequent max-timestamp reads.
## Fix
Gate read-path range conversion on full timestamp visibility. The optimization is now only enabled when:
1. There is no `table_filter` (existing guard — SSTs hidden by filter can break the contiguity assumption)
2. There is no `iter_start_ts` (time-travel scans see a subset of versions)
3. Either no read timestamp is set, or the read timestamp is the max timestamp (all `0xff` bytes)
This is done via a new helper `HasFullTimestampVisibility()` checked at `DBIter` construction time.
## Test Changes
- Updated all existing UDT test variants in `ReadPathRangeTombstoneTest` to read at max timestamp so the optimization remains covered where it is valid.
- Added `UDTOlderTimestampDisablesInsertion`: a regression test that writes at ts=1, deletes at ts=2, writes again at ts=3, then reads at ts=2. Verifies no range tombstone is synthesized and a subsequent max-timestamp read still sees all live values.
## Validation
- `make -j192 db_iterator_test`
- `./db_iterator_test --gtest_filter='*ReadPathRangeTombstoneTest*'`
- `./db_iterator_test --gtest_filter='*UDTOlderTimestampDisablesInsertion*' --gtest_repeat=5`
- Reran the original whitebox crash-test seed from T263957043 against the patched tree; crash-recovery verification passed
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14595
Reviewed By: joshkang97
Differential Revision: D100235999
Pulled By: xingbowang
fbshipit-source-id: 7a287f3c7fdc47428fda0ab89eedd5d43f663985
Summary:
TsanAnnotateMappedMemory() always builds a TsanMappedMemoryInfo payload
for TEST_SYNC_POINT_CALLBACK(), but in builds where sync points compile to a
no-op the local variable becomes unused and Clang fails with
-Werror,-Wunused-variable.
Move the explicit (void)info cast so it applies regardless of whether
__SANITIZE_THREAD__ is enabled. This keeps the helper warning-free in:
- non-TSAN builds
- builds where TEST_SYNC_POINT_CALLBACK expands to nothing
- TSAN builds that still want the SyncPoint payload for tests
This was missed by CI because the warning only appears in the
COMPILE_WITH_TSAN=1 + NDEBUG/release configuration. Our CI matrix covers
release builds without TSAN and TSAN builds in debug mode, but not the
TSAN+NDEBUG combination.
No behavior change is intended. The helper still:
- exposes the mapping metadata to SyncPoint-based tests
- calls AnnotateNewMemory() in TSAN builds
- remains a no-op with respect to runtime behavior in non-TSAN builds
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14597
Differential Revision: D100333855
Pulled By: xingbowang
fbshipit-source-id: 3ff0b0c8a51b4959b7a36984e40ac31e49da82eb
Summary:
TSAN was reporting false data-race warnings after the kernel reused a virtual address for a new mapping while TSAN still associated that address range with the old mapping.
This showed up in two RocksDB paths:
- `io_uring` setup on helper threads vs later `io_uring` `MultiRead` access
- `io_uring` setup vs file mmap reads when `use_mmap_reads` is enabled
## Changes
Introduce `TsanAnnotateMappedMemory()`, which calls `AnnotateNewMemory()` as soon as a fresh mapping exists so TSAN drops any stale shadow state for the recycled address range.
Apply the helper to:
- io_uring SQ/CQ/SQE mappings created by `CreateIOUring()`
- `PosixFileSystem` mmap-read mappings
- `PosixFileSystem` raw memory-mapped file buffers
- `PosixMmapFile` writable mmap region growth
Also add deterministic regression tests that force virtual-address reuse with `MAP_FIXED` for both the io_uring and mmap-read cases. The tests pass mapping metadata over a pipe so teardown/remap ordering is deterministic without introducing TSAN-visible synchronization that would mask the problem.
## Testing
- `COMPILE_WITH_TSAN=1 make -j192 env_test`
- `env_test --gtest_filter='EnvPosixTest.SupportedOpsNoAsyncIOOnIOUringInitFailure:EnvPosixTest.IOUringAddressReuseNoTsanFalsePositive:EnvPosixTest.MmapReadAddressReuseNoTsanFalsePositive'`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14594
Reviewed By: archang19
Differential Revision: D100235746
Pulled By: xingbowang
fbshipit-source-id: 8ef8d85c08a7646540b9c11c3741978898a5af3d
Summary:
### Summary
Fix flaky Claude Code Review workflow failures caused by stale workflow_run SHAs.
Problem:
- The auto-review workflow is triggered from pr-jobs via workflow_run.
- In some cases, by the time the review job starts, the PR branch head has already advanced.
- Then Get PR info cannot find an open PR whose head SHA matches the older workflow_run.head_sha, so it fails with:
- Could not find PR for SHA ... after retries
- This creates unnecessary CI noise even though a newer review job will be triggered for the updated commit.
Solution:
- Treat “PR not found for this head SHA” as a stale run, not an error.
- In Get PR info, replace workflow failure with:
- a warning
- skip=true
- skip_reason=stale_sha:<sha>
- Gate all later auto-review steps on steps.pr_info.outputs.skip != 'true'
- Add a small logging step to report the skip reason
Result:
- stale auto-review runs exit successfully instead of failing
- CI becomes less noisy
- the newest commit still gets reviewed by the later workflow run
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14591
Reviewed By: mszeszko-meta
Differential Revision: D100190375
Pulled By: xingbowang
fbshipit-source-id: fa4a0479a10e953e52b3d99c2b483dae01d3e7fb
Summary:
Flush(), Sync(), and Fsync() on WritableFile and FSWritableFile were essentially undocumented. Added comprehensive API contracts covering durability guarantees (process crash vs. power failure), data readability after Flush(), the implication relationship (Sync implies Flush), and thread-safety (referencing IsSyncThreadSafe()). The contracts are written to be implementation-agnostic, avoiding assumptions about OS-level mechanisms, so they apply equally to local, remote, and virtual/wrapper filesystem implementations. Several such implementations have been audited for adherence.
Note that WritableFileWriter currently calls FSWritableFile::Flush() UNNECESSARILY for cases like writing SST files. This should be optimized away in follow-up assuming this interpretation of the contract is agreed upon.
Also removed the dead FileSystem::WriteLifeTimeHint enum, which was never referenced anywhere. The actively used enum is Env::WriteLifeTimeHint.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14590
Test Plan: Documentation-only change to public headers, plus removal of an unused enum with no references. No functional changes. Verified no compilation errors with existing tests.
Reviewed By: xingbowang
Differential Revision: D100183288
Pulled By: pdillinger
fbshipit-source-id: 7f1982c41e09fe39896dbc6cc328316be559ec4a
Summary:
Add Compressor::GetRecommendedParallelThreads() virtual method so that the compression subsystem can influence the parallel compression thread count used when building SST files. The base Compressor returns 0 (no opinion), while built-in compressors (via CompressorBase) return the parallel_threads value from their CompressionOptions. This gives CompressionManager a clean mechanism to override parallel_threads by customizing the CompressionOptions passed to GetCompressor() in its GetCompressorForSST() implementation.
The table builder now reads the thread count from the compressor after creation, rather than only from CompressionOptions directly. Hard structural constraints (partition_filters without decoupled mode, user_defined_index_factory) still force single-threaded compression regardless of the compressor's recommendation.
Also adds a sync point in MaybeStartParallelCompression for test observability, and compression_test coverage for the new functionality.
Bonus: extends CompressionManagerCustomCompression test to cover re-opening a DB at format_version=7 with a non-default compression manager that uses the built-in CompatibilityName ("BuiltinV2"). Verifies that data is readable after re-opening without the original manager, and that neither GetId() nor CompatibilityName() resolves to the custom manager via CreateFromString.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14580
Test Plan:
New unit tests in compression_test:
- GetRecommendedParallelThreads: verifies built-in compressors return the parallel_threads from their CompressionOptions for all supported compression types.
- CompressionManagerOverridesParallelThreads: end-to-end test with a custom CompressionManager that overrides parallel_threads from 1 to 4 in GetCompressorForSST, verified via sync point that parallel compression actually activates with the overridden thread count, plus data readback verification.
Existing parallel compression tests (DBCompressionTestMaybeParallel) continue to pass.
Reviewed By: xingbowang
Differential Revision: D99896343
Pulled By: pdillinger
fbshipit-source-id: 6b3a30856a78641714d33ec7ba2099f33533d3af
Summary:
Add `use_udi_as_primary_index` option to `BlockBasedTableOptions`. When
enabled, the UDI becomes the primary index — all reads (including
internal operations like compaction and VerifyChecksum) automatically
route through the UDI without needing `ReadOptions::table_index_factory`.
Both the standard binary search index and the UDI are always fully
built. The standard index serves as a safety fallback (e.g., for
backup/restore or rollback to a non-UDI configuration). A future
refactor will extract the index abstraction to allow skipping the
standard index build when the UDI is primary (see discussion below).
## Write path
- `UserDefinedIndexBuilderWrapper` always forwards `AddIndexEntry` and
`OnKeyAdded` to both the internal standard builder and the UDI builder
- New `udi_is_primary_index` table property marks primary-mode SSTs
- Validates incompatible options at `DB::Open` and builder creation:
partitioned index, partitioned filters, missing
`user_defined_index_factory`
## Read path
- `UserDefinedIndexReaderWrapper` defaults to UDI when `udi_is_primary_`,
even when `ReadOptions::table_index_factory` is null — this handles
the 15+ internal call sites that don't set `table_index_factory`
- `use_udi_as_primary_index` automatically enforces `fail_if_no_udi_on_open`
to prevent silent data loss if SSTs are opened without UDI support
## Rollback
Since the standard index is always fully populated, rollback from
primary mode is straightforward: set `use_udi_as_primary_index=false`.
No compaction required — SSTs written in primary mode are immediately
readable through the standard index.
## Public API
- `BlockBasedTableOptions::use_udi_as_primary_index` (default: false)
- `UserDefinedIndexBuilder::EstimatedSize()` — pure virtual, O(1) via
running counter in the trie implementation
## Bug fixes (issues https://github.com/facebook/rocksdb/issues/14560, https://github.com/facebook/rocksdb/issues/14561, https://github.com/facebook/rocksdb/issues/14562)
Fixed trie index correctness bugs that caused crash test failures:
- **Always-on seqno encoding**: `must_use_separator_with_seq_` is now
unconditionally true. Non-boundary separators store tag=0 (sentinel),
same-user-key boundaries store the real tag, and the last block stores
its real last-key tag. This fixes the `NonBoundaryTag` bug where
non-boundary separators with `kMaxSequenceNumber` caused the post-seek
correction to incorrectly advance past the correct block.
- **Standard index always built in primary mode**: An empty (stub)
standard index block caused behavioral divergence in
`BlockBasedTableIterator` under concurrent flush/compaction, leading to
`test_batches_snapshots` prefix scan inconsistencies.
## Stress test
- `use_udi_as_primary_index` flag randomized by `db_crashtest.py`
- Both primary and secondary UDI modes exercised in crash tests
- Trie index probability halved (~6%) per reviewer request
- Re-enables trie crash tests (disabled by https://github.com/facebook/rocksdb/issues/14559)
## Tests
- Parameterized `TrieIndexDBTest` on UDI mode (secondary vs primary)
- New factory-level tests: non-boundary separator seek, Prev within
overflow, SeekToLast with overflow, empty trie, overflow exhaustion,
all-scans-exhausted
- New DB-level tests: multi-CF coalescing iterator, GetEntity with
explicit snapshot, reverse iteration across same-user-key blocks,
non-boundary separator seek correctness, rollback from primary
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14547
Reviewed By: anand1976
Differential Revision: D99494181
Pulled By: xingbowang
fbshipit-source-id: ca52a0d5c0e523770c80e1fe2b9b5d50406b67bc
Summary:
Wide-column blob separation: lazy resolution through read, compaction, and write paths
Extend blob direct write to support wide-column entities (PutEntity), and add
lazy blob resolution for wide-column values across all read and compaction paths.
**Write path -- PutEntity blob separation:**
- BlobWriteBatchTransformer::PutEntityCF now extracts large column values
(>= min_blob_size) to blob files and serializes V2 entities with BlobIndex
references, matching the existing Put behavior.
- Add MaybePreprocessWideColumns() static helper to share blob extraction
logic between the WriteBatch transformer and the new PutEntity fast path.
- Add PutEntityFastPath() in DBImpl that preprocesses columns (sort, blob
extract, serialize) before calling WriteImpl, skipping the redundant
WriteBatch transformation pass. Trace batch preserves the original columns.
**Read path -- blob resolution for Get/MultiGet/Iterator:**
- GetContext::SaveValue resolves V2 entity blob columns eagerly: for
value (Get), resolves the default column's blob reference; for columns
(GetEntity), resolves all blob columns and re-serializes as V1.
- DBIter::SetValueAndColumnsFromEntity detects V2 entities, deserializes
with DeserializeV2, and eagerly resolves all blob columns via a new
ReadPathBlobResolver. Resolved values are cached in the resolver and
wide_columns_ Slices point into the cache, avoiding copies.
- Add ReadPathBlobResolver (new file) -- on-demand blob fetcher for the
read path with per-column caching, used by both DBIter and GetContext.
- BlobFetcher gains allow_write_path_fallback to read from in-flight
direct-write blob files not yet visible through Version (pre-flush reads).
- Memtable lookups for Get(key) on V2 entities with a blob default column
now return the blob index with is_blob_index=true, triggering the
existing BDW resolution in MaybeResolveWritePathValue.
- MaybeResolveWritePathValue (renamed from MaybeResolveDirectWriteBlobIndex)
now also resolves V2 entity blob columns for GetEntity/MultiGetEntity,
re-serializing as V1 after resolution.
**Compaction path -- filter, GC, and extraction:**
- CompactionIterator::InvokeFilterIfNeeded handles V2 entities: FilterV3
gets eagerly-resolved column values for backward compatibility; FilterV4
gets a CompactionBlobResolver for lazy on-demand resolution.
- Add CompactionFilter::FilterV4 with WideColumnBlobResolver* parameter
and SupportsFilterV4() opt-in. Default delegates to FilterV3.
- CompactionBlobResolver (new class) implements WideColumnBlobResolver
for the compaction path with stats tracking.
- ExtractLargeColumnValuesIfNeeded extracts inline columns to blob files
during compaction (entities without existing blob columns only).
- GarbageCollectEntityBlobsIfNeeded relocates blob values from old blob
files to new ones during compaction GC, with helpers FetchBlobsNeedingGC,
RelocateBlobValues, and SerializeEntityAfterGC.
- PrepareOutput unified entity deserialization: single DeserializeV2 call
reused by both filter and GC/extraction paths via entity_deserialized_
flag, avoiding redundant parsing.
**Merge path -- V2 entity base value resolution:**
- MergeHelper::MergeUntil, GetContext::MergeWithWideColumnBaseValue, and
DBIter::MergeWithWideColumnBaseValue resolve V2 blob columns before
calling TimedFullMerge, using ResolveEntityForMerge.
**Blob garbage accounting:**
- BlobGarbageMeter tracks blob file in/out flow for V2 entity blob
columns via ForEachBlobFileNumber, used for accurate GC decisions.
- FileMetaData::UpdateBoundaries tracks oldest_blob_file_number for
V2 entities, ensuring blob files referenced by entities are not
prematurely deleted.
**Serialization improvements:**
- WideColumnSerialization::SerializeV2Impl allocates serialized_blob_indices
only for actual blob columns (not all columns) and uses autovector for
name/value sizes.
- Add ForEachBlobFileNumber for lightweight blob file number extraction
without full deserialization.
- Add ResolveEntityForMerge helper for merge-path resolution.
- Add section-size validation in DeserializeV2Impl.
- Add empty blob index and column type validation.
- blob_column_resolver_util.h -- shared helpers (FindBlobColumn, FindInCache,
CacheInlinedBlob) used by both ReadPathBlobResolver and CompactionBlobResolver.
**Testing:**
- db_blob_direct_write_test: end-to-end PutEntity with BDW before/after flush,
verifying Get, GetEntity, MultiGetEntity, and Iterator.
- db_blob_index_test: ~1550 lines covering V2 entity blob resolution through
Get, GetEntity, MultiGet, Iterator, compaction filter (V3 compat and V4 lazy),
merge with blob base, and compaction GC/extraction.
- compaction_iterator_test: ~950 lines testing entity blob GC, extraction,
filter interaction, and combined GC+filter scenarios.
- db_wide_basic_test: ~1200 lines for wide-column lazy blob resolution through
all read paths plus compaction round-trips.
- db_open_with_config_test: ~450 lines for BDW entity config validation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14386
Reviewed By: anand1976
Differential Revision: D99739701
Pulled By: xingbowang
fbshipit-source-id: 6badd89b577f3054802eaaa654738468efb9dbdb
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14581
In `MultiScanIndexIterator::Seek()` Case 3, when re-entering a scan range
after all ranges were exhausted, `block_idx = std::max(cur_scan_start_idx,
cur_idx_)` could produce an out-of-bounds value because `cur_idx_` was left
at `block_handles_.size()` from previous exhaustion. `SeekToBlockIdx()`
unconditionally set `valid_ = true` without checking bounds, causing the
subsequent `value()` call to hit the assertion
`cur_idx_ < block_handles_.size()`.
Added bounds check before `SeekToBlockIdx()` in Case 3 to correctly report
exhaustion instead of crashing.
Reviewed By: joshkang97
Differential Revision: D99604049
fbshipit-source-id: 9d5d91afde7c0984a7b4c2f62604f27f19b07922
Summary:
**Root cause:** DBIter's read-path range tombstone conversion assumes the iterator can observe *every* interior live key between point tombstones. When `ReadOptions.table_filter` is active, whole SSTs can be hidden from the scan. DBIter may then mistake non-contiguous deletes for a contiguous run and synthesize a range tombstone that covers real, still-live keys — silently corrupting the read path.
This is not UDT-specific; UDT just changes the failure surface. The stress test failures that triggered investigation were from `db_stress_tool/no_batched_ops_stress.cc` applying an SQFC-backed `table_filter` for range queries even when `total_order_seek` is forced.
## Fix
`db/db_iter.cc`: When initializing `min_tombstones_for_range_conversion_`, set it to 0 (disabling range conversion) whenever `read_options.table_filter` is non-null. This is the conservative and correct behavior — filtered scans don't provide the full key visibility the optimization requires.
## Test
`db/db_iterator_test.cc` — new test `ReadPathRangeTombstoneTest.TableFilterHiddenInteriorKey`:
- Constructs five flushed SSTs so that the live interior key `"b"` lives in a 2-entry SST that the `table_filter` hides.
- Keeps the active memtable non-empty (key `"zz"`) so range conversion has a valid insertion point — without this the test is a false negative.
- Asserts `inserted_ranges_.size() == 0` (no synthesis occurred).
- Verifies `"b"` is still readable via a normal `Get`.
- Runs both with and without UDT (user-defined timestamps) via `SCOPED_TRACE`.
On unpatched HEAD this test fails with `inserted_ranges_.size() == 1`; with the fix it passes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14586
Reviewed By: joshkang97
Differential Revision: D100023487
Pulled By: xingbowang
fbshipit-source-id: 5ef885c89a6cfd7a97b814e38bdcc48d3a1ab349
Summary:
Crash test T263619547 exposed a flush input accounting bug in
`FragmentedRangeTombstoneList`.
When the constructor sees that the incoming range tombstones are not sorted, it
falls back to rereading all tombstones into `keys`/`values` and sorting them via
`VectorIterator` before fragmentation. However, `num_unfragmented_tombstones_`
was left with the partial count from the aborted first pass. In timestamp
stripping flushes, this stale count could make flush verification report
`Expected X entries in memtables, but read Y` even though all tombstones were
still processed.
Fix the slow path by resetting `num_unfragmented_tombstones_` from
`keys.size()` after the second pass. Add a regression test that feeds unsorted
range tombstones into the fragmenter and verifies the full tombstone count is
preserved.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14579
Test Plan: - New unit test that fails without the fix.
Reviewed By: xingbowang
Differential Revision: D99872374
Pulled By: joshkang97
fbshipit-source-id: 38e2bb73c68dc1c677870b8973c93b933ded9038
Summary:
https://github.com/facebook/rocksdb/pull/14575 addresses an existing issue where trace record before WAL lead to inconsistent expected state.
However, placing the trace record after the WAL is still problematic because. The trace may not have all the records in the WAL, resulting in `Error restoring historical expected values -> Trace ended before replaying all expected write ops` on recovery.
This change minimizes the gap to reduce stress test failures, but it is still not a full solution. It is however still better than previously writing the trace before the WAL because the error is much more obvious when writing the trace after WAL.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14578
Test Plan: Existing CI passes.
Reviewed By: xingbowang
Differential Revision: D99870545
Pulled By: joshkang97
fbshipit-source-id: 024ee3b15e258afd2dd8ff98db7cbe6a01906be7
Summary:
Handle the case where a transaction commit with `commit_bypass_memtable` succeeds in writing the commit marker to WAL, but then `IngestWBWIAsMemtable` fails (e.g., WAL creation failure during `SwitchMemtable` or invalid column family). Previously this failure path was not distinguished from a direct WBWI ingest, so the DB could continue operating with committed data durable in WAL but not published to memtables.
This resulted in a higher seqno in the WAL that what the DB's latest seqno was.
Now `IngestWBWIAsMemtable` takes an `ingest_wbwi_for_commit` parameter. When true and the ingest fails (without a prior memtable update), the DB sets a fatal background error, stopping all writes until the DB is closed and reopened for WAL recovery. A close and reopen is required because the auto-recovery mechanism works by flushing memtables to disk, but here the committed data was never published to memtables in the first place — it only exists as prepare + commit records in the WAL. Only a full WAL replay during `DB::Open` can reconstruct the committed transaction data and insert it into memtables.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14567
Test Plan:
- Added `CommitBypassMemtableTest.SwitchMemtableFailureStopsDBUntilReopen`:
- Injects an IO error during `SwitchMemtable` WAL creation via `TEST_SYNC_POINT_CALLBACK`
- Verifies commit returns `Corruption` status
- Verifies DB enters fatal background error state and rejects further writes
- Verifies that after close and reopen, committed data is recovered from WAL and the DB resumes normal operation
Reviewed By: xingbowang
Differential Revision: D98638705
Pulled By: joshkang97
fbshipit-source-id: f116b1f257b19984b0413f6aeba0ba4a7c0a5b25
Summary:
When BuildTable succeeds during flush, it caches the output SST file in the table cache for subsequent user reads (builder.cc:460). If the flush then fails to install — e.g., LogAndApply MANIFEST I/O error, CF dropped, or shutdown — the cache entry was never evicted. BuildTable only cleans up its own failures (builder.cc:511), not failures in the install path.
The FindObsoleteFiles full-scan backstop would normally catch this by finding the orphan file on disk and evicting the cache entry. However, under crash test metadata read fault injection
(open_metadata_read_fault_one_in), GetChildren fails and the orphan is never found, causing the TEST_VerifyNoObsoleteFilesCached assertion to fire during Close().
This is the same class of bug previously fixed in the compaction path by https://github.com/facebook/rocksdb/issues/14469 (Run failure) and https://github.com/facebook/rocksdb/issues/14549 (Install failure), but in the flush path which had no analogous cleanup.
Fix: call TableCache::ReleaseObsolete in FlushJob::Run() when the flush fails after BuildTable succeeded (meta_.fd.GetFileSize() > 0).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14577
Test Plan:
New unit test DBFlushTest.LeakedTableCacheEntryOnFlushInstallFailure:
- Injects failure after BuildTable via sync point, deactivates filesystem to prevent backstop cleanup
- Without fix: assertion fires ("Leaked table cache entry")
- With fix: passes
```
COMPILE_WITH_ASAN=1 make -j db_flush_test
./db_flush_test --gtest_filter="DBFlushTest.LeakedTableCacheEntry*"
[ PASSED ] 1 test.
```
Existing compaction leak tests still pass:
```
COMPILE_WITH_ASAN=1 make -j db_compaction_test
./db_compaction_test --gtest_filter="*LeakedTableCacheEntry*"
[ PASSED ] 2 tests.
```
Reviewed By: xingbowang
Differential Revision: D99692601
Pulled By: pdillinger
fbshipit-source-id: ff5aa1ad165b3abec915844f97f6af9e59d85774
Summary:
Upgrade the Ubuntu 24 CI Docker image to include clang-21 (from clang-18), and bump all workflow references to the new image tags. Also adds ccache to both images and renames all clang-18 job references to clang-21.
This is mainly for upgrading the clang version later used for generating C API automatically. See PR https://github.com/facebook/rocksdb/issues/14572
### Changes
**`build_tools/ubuntu24_image/Dockerfile`**
- Add clang-21 installation from LLVM snapshot repo (`apt.llvm.org/noble/llvm-toolchain-noble-21`)
- Add ccache
- Add comment pointing Meta employees to internal devvm build guide
**`build_tools/ubuntu22_image/Dockerfile`**
- Add ccache
- Add comment pointing Meta employees to internal devvm build guide
**`.github/workflows/pr-jobs.yml`**
- Rename `build-linux-clang-18-no_test_run` → `build-linux-clang-21-no_test_run`
- Rename `build-linux-clang18-asan-ubsan` → `build-linux-clang21-asan-ubsan`
- Rename `build-linux-clang18-mini-tsan` → `build-linux-clang21-mini-tsan`
- Update all `clang-18`/`clang++-18` references to `clang-21`/`clang++-21`
- Update ccache key prefixes: `clang18-asan-ubsan` → `clang21-asan-ubsan`, `clang18-tsan` → `clang21-tsan`
- Bump `rocksdb_ubuntu:24.0` → `rocksdb_ubuntu:24.1`
**`.github/workflows/nightly.yml`**
- Rename `build-linux-clang-18-asan-ubsan-with-folly` → `build-linux-clang-21-asan-ubsan-with-folly`
- Update clang-18 → clang-21 compiler references
- Bump `rocksdb_ubuntu:24.0` → `rocksdb_ubuntu:24.1`
**`.github/workflows/clang-tidy.yml`**
- Update clang-18 → clang-21
- Bump `rocksdb_ubuntu:24.0` → `rocksdb_ubuntu:24.1`
### New Docker Images
Both images have been built and tested locally. They will be pushed to `ghcr.io/facebook/rocksdb_ubuntu` before this PR is merged.
- `ghcr.io/facebook/rocksdb_ubuntu:22.2` — adds ccache, adds devvm build note
- `ghcr.io/facebook/rocksdb_ubuntu:24.1` — adds clang-21 from LLVM snapshot repo, adds ccache
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14576
Test Plan: Built both images locally and verified CI job names/compiler flags are consistent.
Reviewed By: joshkang97
Differential Revision: D99694293
Pulled By: xingbowang
fbshipit-source-id: c23c27f5cf870fb2c8b4e3d1cba281d0ce63f9d6
Summary:
Fix trace ordering for `preserve_write_order` mode to only record writes that actually succeeded.
Previously, when `preserve_write_order = true`, trace records were emitted **before** confirming the write succeeded — at a point where the write could still fail (e.g., WAL I/O error). This meant the trace could contain records for writes that never actually committed to the DB. When `db_stress` crash testing replays the trace to reconstruct expected state, these phantom records cause the expected state to diverge from reality, leading to false verification failures.
This fix moves the `preserve_write_order` tracing to happen **after** the write is confirmed successful (after memtable insert, right before `SetLastSequence`), in all three write paths: `WriteImpl`, `PipelinedWriteImpl`, and `WriteImplWALOnly`.
default and pipelined write modes. Replays the trace into a fresh DB and confirms only the successful write is present.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14575
Test Plan:
- New unit test: `TracePreserveWriteOrderSkipsFailedWrite` in `db/db_test2.cc`
- Tests both `enable_pipelined_write = false` and `true`
- Uses `FaultInjectionTestEnv` to force a write failure
- Verifies the failed write is absent from both the original DB and the replayed trace
Reviewed By: xingbowang
Differential Revision: D99624672
Pulled By: joshkang97
fbshipit-source-id: 5793727c91cb01a12fbb4b2cd59615802ae3d394
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14569
ReadSet::ReadIndex() moves block values out of pinned_blocks_ via std::move,
but never releases the associated prefetch memory accounting. This causes
ReleaseBlock() and the destructor to skip ReleaseMemory() since they check
pinned_blocks_.GetValue() which returns null after the move. Over time, the
memory budget is exhausted and no further prefetches can be dispatched when
max_prefetch_memory_bytes is set. The bug was introduced in https://github.com/facebook/rocksdb/pull/14401.
The fix releases memory accounting in ReadIndex() when moving values out
(both for Case 1: block already available, and Case 2: after async IO
polling), and zeros block_sizes_ to prevent double-release.
Also adds multiscan_max_prefetch_memory_bytes option to db_stress/crashtest
for stress testing this code path.
Reviewed By: hx235
Differential Revision: D99488961
fbshipit-source-id: 5ddd1f50e2f6ebb357f86e013d781a790e7e558a
Summary:
btrfs accepts fallocate without error but uses copy-on-write, so preallocated extents are not reflected in st_blocks. This caused AllocateTest to fail spuriously on btrfs filesystems. Detect btrfs via statfs and skip only the block-count assertions, keeping the rest of the test (write, flush, close, size checks) intact.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14553
Test Plan: env_test AllocateTest passes on btrfs (skips with message) and would still enforce preallocation checks on ext4.
Reviewed By: joshkang97
Differential Revision: D99307962
Pulled By: pdillinger
fbshipit-source-id: 8f40c0109aa397e9ca7d9ee4496fa0614e971970
Summary:
When MarkAsTrash fails (e.g., rename error due to filesystem conditions), AddFileToDeletionQueue falls back to deleting the file directly via fs_->DeleteFile(). This bypasses DeleteFileImmediately() and its TEST_SYNC_POINT("DeleteScheduler::DeleteFile") callback, causing tests that count deletions via sync points to undercount. This explains flaky failures in DBSSTTest.DeleteSchedulerMultipleDBPaths where bg_delete_file is 4 instead of the expected 8.
Fix by calling DeleteFileImmediately() in the error path instead of fs_->DeleteFile() directly. DeleteFileImmediately already handles the sync point, OnDeleteFile tracking, and FILES_DELETED_IMMEDIATELY stat, so the manual bookkeeping is also removed to avoid duplication.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14556
Test Plan:
- Existing tests: db_sst_test (DeleteSchedulerMultipleDBPaths, DestroyDBWithRateLimitedDelete) and delete_scheduler_test all pass.
- Ran DeleteSchedulerMultipleDBPaths 100x with COERCE_CONTEXT_SWITCH=1.
Reviewed By: joshkang97
Differential Revision: D99308177
Pulled By: pdillinger
fbshipit-source-id: 3918c4800c2420ff1ae667beb321da3f27406420
Summary:
Fix two bugs in read-path range tombstone conversion (`min_tombstones_for_range_conversion`):
1. **SeekToLast stale saved_key_**: `SeekToLast()` did not clear `saved_key_` before calling `PrevInternal()`. A stale key from a prior `Seek()` would be swapped into `range_tomb_end_key_`, corrupting the tombstone tracking bounds and triggering an assertion failure when `range_tomb_first_key_ > end_key`.
2. **Tombstone inserted at wrong sequence**: The current behavior uses the "max" seqno of the tombstones as the insertion seqno. This is not correct because it does not take into account the seqno range deletions seen throughout the iteration. Unfortunately, range deletions are not visible to the db_iter as the are filtered at the merging iterator level. A future refactor may try to expose this, but currently the simplest solution is to just use the iterator seqno.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14573
Test Plan: Unit tests for both bugs that fail without the fix.
Reviewed By: xingbowang
Differential Revision: D99569668
Pulled By: joshkang97
fbshipit-source-id: bb1b154beccced7438831cd5cf2780ad1e11a4cc
Summary:
Add a read-path optimization that converts contiguous point tombstones into range tombstones during forward/reverse iteration. When a configurable threshold of consecutive point deletions (kTypeDeletion, kTypeDeletionWithTimestamp, kTypeSingleDeletion — with no live keys between them) is detected, a range tombstone covering `[first_tombstone_key, next_live_key)` is inserted into the active mutable memtable. This benefits future iterators by enabling efficient skipping via range tombstone fragmentation.
If there is a memtable switch during the read iteration, then the range deletion entry is discarded.
The inserted range tombstones are logically redundant (they don't delete anything that isn't already deleted by point tombstones), skip WAL (they're a derived optimization regenerated by future reads on crash), and use the max tombstone sequence number so they don't interfere with newer writes.
## Key changes
- **New option `min_tombstones_for_range_conversion`** (`AdvancedColumnFamilyOptions`): Threshold of contiguous point tombstones before converting to a range tombstone. Default 0 (disabled). Dynamically changeable via `SetOptions()`.
- **`DBIter` tracking logic** (`db/db_iter.cc`): Tracks contiguous tombstones during `FindNextUserEntryInternal()` (forward) and `PrevInternal()` (reverse). When a live key terminates a run that meets the threshold, `MaybeInsertRangeTombstone()` inserts `[first_tombstone, live_key)` into the active memtable.
- **`FindValueForCurrentKey` `found_visible` output** (`db/db_iter.cc`): Distinguishes "key deleted at this snapshot" from "no visible entries" so reverse tracking doesn't treat post-snapshot keys as tombstones.
- **`IterKey::Swap()`** (`db/dbformat.h`): Efficiently tracks reverse tombstone run end keys without extra allocations.
- **`MemTable::AddLogicallyRedundantRangeTombstone()`** (`db/memtable.cc`): Concurrent-safe range tombstone insertion into the active memtable. Range tombstone skiplist always uses concurrent inserts.
- **`ConstructFragmentedRangeTombstones` race fix** (`db/db_impl/db_impl_write.cc`): Moved after `MarkImmutable()` to prevent lost entries.
- **`MarkImmutable` ordering fix** (`db/memtable_list.cc`): Called before `current_->Add()` to close a race window.
- **Prefix filter awareness** (`db/db_iter.cc`): Tombstone tracking scoped to the seek prefix when prefix filtering is active. See dedicated section below.
- **Transaction awareness** (`db/db_iter.cc`): Tombstones with `seq > snapshot` excluded from tracking. `min_uncommitted` guard uses `insert_seq` (which may be bumped to `earliest_seq`) instead of `range_tomb_max_seq_`. See dedicated section below.
- **Duplicate range check**: Skips insertion if the memtable already covers `[start, end)`.
- **New statistics**: `READ_PATH_RANGE_TOMBSTONES_INSERTED` and `READ_PATH_RANGE_TOMBSTONES_DISCARDED`.
- **Memtable MultiGet batch lookup** (`memtable/inlineskiplist.h`, `db/memtable.cc`): `InlineSkipList::MultiGet()` with cached search path ("finger") for sorted key lookups.
- **New option `memtable_batch_lookup_optimization`** (`AdvancedColumnFamilyOptions`): Enables batch lookup for memtable MultiGet. Default false. Immutable.
## Deciding Range Tombstone Seqno
- The range tombstone is inserted with `insert_seq = max(range_tomb_max_seq_, earliest_seq)` where `range_tomb_max_seq_` is the maximum sequence number across all point tombstones in the contiguous run, and `earliest_seq` is the memtable's earliest sequence number. This preserves the memtable's `earliest_seqno_` invariant.
- If the iterator's snapshot sequence (`sequence_`) predates the memtable's `earliest_seq`, insertion is skipped entirely to avoid unintentionally covering entries between `sequence_` and `earliest_seq`.
## ConstructFragmentedRangeTombstones Race Fix
- `MarkImmutable()` and `ConstructFragmentedRangeTombstones()` are now called before `mutex_.Lock()` in `SwitchMemtable`, keeping this work outside the DB mutex. `MarkImmutable()` blocks concurrent `AddLogicallyRedundantRangeTombstone()` calls via `immutable_mutex_`, ensuring no range tombstones are inserted after the fragmented list is built. `MarkImmutable()` is idempotent, so `MemTableList::Add()` calling it again inside the mutex is harmless.
## Prefix Filter Safety
- When prefix filtering is active, the BBTI bloom filter may reject SST files outside the seek prefix, but the memtable (no bloom filter) returns keys across prefix boundaries. Tombstone tracking is scoped to the seek prefix so that converted range tombstones cannot cover live keys hidden in filtered files.
- `total_order_seek=true` disables prefix filtering — all files are visible, so tombstones safely span prefix boundaries.
- **Behavior change**: Seeking to an out-of-domain key with `total_order_seek=false` now treats it as total-order (prefix_ not set). When `prefix_same_as_start=true`, iterating past an out-of-domain key cleanly invalidates the iterator instead of calling `Transform()` on it (which was UB in release builds with `FixedPrefixTransform`). This is a requirement because an incorrect iterator scan could lead to a range tombstone covering a live key.
## Transaction Support
- Tombstones written by the transaction's own uncommitted writes (sequence > snapshot) are now excluded from contiguous tombstone tracking entirely at the tracking site in `FindNextUserEntryInternal()` and `PrevInternal()`. Previously, tracking relied on a `min_uncommitted` check at insertion time, but this was insufficient — a transaction's own Delete with `seq > snapshot` could extend a run of committed tombstones, and the resulting range tombstone would cover data visible to other snapshots.
- The fix skips any tombstone with `ikey_.sequence > sequence_` during tracking. If a transaction-owned tombstone appears mid-run, it flushes the accumulated committed run first, then resets tracking. This ensures only tombstones visible to the current snapshot are ever converted.
- Both WritePrepared and WriteUnprepared transactions are supported with dedicated test coverage:
- **WritePrepared**: When tombstones are committed before `Prepare()`, their seqnos are below `min_uncommitted` and insertion proceeds safely. When `Prepare()` happens first, tombstone seqnos exceed `min_uncommitted` and insertion is blocked.
- **WriteUnprepared**: Multiple unprepared batches with different seqno ranges are handled correctly. Own transaction Deletes that extend a committed tombstone run block insertion of the entire run. After rollback, data correctness is verified.
## UDT support
- When user-defined timestamps (UDT) are enabled, keys include an 8-byte timestamp suffix. The comparator, Put/Delete APIs, and ReadOptions all require timestamps.
- Forward exhaustion with UDT: `iterate_upper_bound_` is a plain user key without a timestamp suffix. It is padded with min timestamp via `AppendKeyWithMinTimestamp()` so it sorts after all entries with this user key, preserving the exclusive bound semantics.
- Reverse exhaustion with UDT: The end key comes from either the previous live key (which already has a proper timestamp suffix) or the seek target set by `SetSavedKeyToSeekForPrevTarget()` (which appends a timestamp via `SetInternalKey(..., timestamp_ub_)` and `UpdateInternalKey(..., ts)`). In both cases, the end key is already properly timestamped, so no additional padding is needed unlike forward exhaustion.
- Contiguous tombstone detection works correctly with UDT because the underlying `kTypeDeletionWithTimestamp` entries are tracked the same way as `kTypeDeletion`.
## Concurrent Iterators
- Should concurrent iterators happen to read the range at the same time, both will produce the same range and seqno entry. Only one will be accepted by the skip list and the others will be rejected. Future iterators will read the range and not even attempt to insert the range.
- There is nothing preventing similar ranges from being inserted however. Two iterators can produce overlapping ranges, but this protection would be complicated to implement and there is no evidence that it is a likely scenario yet.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14448
Test Plan:
- **Unit tests** (`db/db_iterator_test.cc`): `ReadPathRangeTombstoneTest` parameterized by forward/reverse with cases for basic insertion, non-contiguous (below threshold), memtable switch, exhausted iterator with/without bounds, direction change, mixed Delete/SingleDelete, single-delete-only runs, snapshot predating memtable, block cache tier incomplete, skip when covered by existing range, UDT basic scan, UDT exhaustion, prefix filter cross-prefix scan (`PrefixFilterCrossPrefixScanCoversLiveKey` with default/total_order_seek/prefix_same_as_start variants), stale ikey from forward-then-reverse scan (`StaleIkeyFromForwardThenReverse`), and reseek stale ikey (`ReseekStaleIkey`).
- **Concurrency test** (`db/db_test2.cc`): `DBTestConcurrentRangeTombstoneConversions` parameterized by `(allow_concurrent_memtable_write, min_tombstones_for_range_conversion)` with mixed writers, deleters, range deleters, and concurrent forward/reverse readers.
- **Transaction tests** (`utilities/transactions/write_prepared_transaction_test.cc`, `write_unprepared_transaction_test.cc`): Tests for WritePrepared (insertion allowed when tombstones committed before prepare, blocked when after; seqno bump shadowing prepared writes `RangeTombstoneSeqnoBumpShadowsPreparedWrite`) and WriteUnprepared (multiple batches, extended visibility with CalcMaxVisibleSeq, own deletions with rollback).
- **IterKey::Swap tests** (`db/dbformat_test.cc`): `IterKeySwapTest` parameterized over `(key_len, copy, use_secondary)` × 2 covering all inline/heap/pinned/secondary combinations.
- **InlineSkipList MultiGet tests** (`memtable/inlineskiplist_test.cc`): Basic, exact matches, empty, single key, randomized validation against `std::set::lower_bound`, duplicate keys with callback walk, and concurrent MultiGet with read-after-write consistency.
- **Memtable MultiGet tests** (`db/db_basic_test.cc`): Batch lookup, overwrite, flush, merge, disabled by default, paranoid checks, and snapshot tests.
- **Stress test coverage**: `min_tombstones_for_range_conversion` and `memtable_batch_lookup_optimization` options added to `db_crashtest.py` and `db_stress` flags.
- `make check` passes all tests.
### Benchmark results
Tombstones scattered randomly in clusters via `seek_nexts_to_delete` for realistic workloads.
**DB Setup A (scattered deletes, 100 per seek)**:
```
# Step 1: Create and compact
./db_bench --benchmarks=fillseq,compact --seed=1 --compression_type=none --num=1000000 --db=<DB>
# Step 2: Scatter tombstones (5000 seeks × 100 deletes ≈ 500k tombstones)
./db_bench --benchmarks=seekrandom,flush --seed=1 --compression_type=none --num=2000 \
--seek_nexts=0 --seek_nexts_to_delete=100 --use_existing_db=1 --threads=1 --db=<DB>
```
**DB Setup A2 (scattered deletes, 8 per seek)**:
```
# Step 1: Create and compact
./db_bench --benchmarks=fillseq,compact --seed=1 --compression_type=none --num=1000000 --db=<DB>
# Step 2: Scatter tombstones (5000 seeks × 8 deletes ≈ 40k tombstones)
./db_bench --benchmarks=seekrandom,flush --seed=1 --compression_type=none --num=2000 \
--seek_nexts=0 --seek_nexts_to_delete=8 --use_existing_db=1 --threads=1 --db=<DB>
```
**DB Setup B (no deletes)**:
```
./db_bench --benchmarks=fillseq,compact --seed=1 --compression_type=none --num=1000000 \
[--key_size=100] --db=<DB>
```
**Read workload** (same for all):
```
./db_bench --benchmarks=seekrandom --seek_nexts=100 --threads=8 \
--reverse_iterator={true,false} --seed=1 --use_existing_db=1 \
--compression_type=none --num=1000000 --duration=10 \
--disable_auto_compactions \
[--key_size=100] [--min_tombstones_for_range_conversion=X] --db=<DB_COPY>
```
Each workload averaged over 3 runs.
**Table 1: seekrandom forward, scattered deletes (2000 seeks × 100 deletes/seek)**
| Variant | avg ops/s | % vs main |
|---------|-----------|-----------|
| main | 2,895 | - |
| threshold=0 | 2,869 | -0.9% |
| threshold=8 | 287,334 | +9,824% |
**Table 2: seekrandom reverse, scattered deletes (2000 seeks × 100 deletes/seek)**
| Variant | avg ops/s | % vs main |
|---------|-----------|-----------|
| main | 544 | - |
| threshold=0 | 548 | +0.7% |
| threshold=8 | 206,491 | +37,860% |
**Table 3: seekrandom forward, scattered deletes (2000 seeks × 8 deletes/seek)**
| Variant | avg ops/s | % vs main |
|---------|-----------|-----------|
| main | 194,049 | - |
| threshold=0 | 195,703 | +0.9% |
| threshold=8 | 310,740 | +60.1% |
**Table 4: seekrandom reverse, scattered deletes (2000 seeks × 8 deletes/seek)**
| Variant | avg ops/s | % vs main |
|---------|-----------|-----------|
| main | 63,854 | - |
| threshold=0 | 69,266 | +8.5% |
| threshold=8 | 218,101 | +241.6% |
**Table 5: seekrandom forward, no deletes (regression check)**
| Variant | key=16B avg ops/s | % vs main | key=100B avg ops/s | % vs main |
|---------|-------------------|-----------|---------------------|-----------|
| main | 330,901 | - | 236,048 | - |
| threshold=0 | 328,398 | -0.8% | 238,055 | +0.9% |
| threshold=8 | 332,539 | +0.5% | 233,776 | -1.0% |
**Table 6: seekrandom reverse, no deletes (regression check)**
| Variant | key=16B avg ops/s | % vs main | key=100B avg ops/s | % vs main |
|---------|-------------------|-----------|---------------------|-----------|
| main | 261,445 | - | 192,177 | - |
| threshold=0 | 265,020 | +1.4% | 191,616 | -0.3% |
| threshold=8 | 250,881 | -4.0% | 189,239 | -1.5% |
Reviewed By: xingbowang
Differential Revision: D96203950
Pulled By: joshkang97
fbshipit-source-id: 06ba66ebde3c355f04671d1e681f1b1586e8751d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14557
RocksDB's PerfContext tracks block_decompress_time but has no counter for the
number of block decompressions. The global Statistics ticker
NUMBER_BLOCK_DECOMPRESSED exists but is not accessible through PerfContext,
which is the thread-local, per-operation metric system used by benchmarks.
Add block_decompress_count as a new PerfContext counter (gated at
kEnableCount level) incremented in DecompressBlockData alongside the existing
NUMBER_BLOCK_DECOMPRESSED ticker. This enables benchmarks and applications to
observe how many blocks required decompression per operation, complementing
the existing block_read_count.
Reviewed By: anand1976
Differential Revision: D99233514
fbshipit-source-id: 40a68c1d9321f560cebdb7c30a544a0c62ae64f0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14551
Add two new per-SST TableProperties fields that track how many data blocks
were stored uncompressed:
- `num_data_blocks_compression_rejected`: blocks where compression was
attempted but the compressed output exceeded the ratio limit set by
`CompressionOptions::max_compressed_bytes_per_kb`.
- `num_data_blocks_compression_bypassed`: blocks where compression was
never attempted (e.g., kNoCompression type, no compressor available).
Together with `num_data_blocks`, these give a full decomposition:
num_data_blocks = compressed + rejected + bypassed
Previously this information was only available via global Statistics tickers
(NUMBER_BLOCK_COMPRESSION_REJECTED / BYPASSED) which aggregate across all
SSTs. Now it is persisted per-SST in the properties block and visible via
sst_dump --show_properties.
Reviewed By: anand1976
Differential Revision: D99229339
fbshipit-source-id: 8333f51acff30a759d725e5e94bd80f4fcb18b57
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14568
Port of D98326754 to internal_repo_rocksdb (GitHub-synced repo).
BlobDB blob file compression currently uses hardcoded default
CompressionOptions{} in BlobFileBuilder, ignoring any user-specified
compression options like level, window_bits, or strategy. This means
ZSTD always uses level 3 (doubleFast strategy with two hash tables),
even when a lower level would significantly reduce CPU usage.
This diff adds a new `blob_compression_opts` field to
AdvancedColumnFamilyOptions (and MutableCFOptions) and plumbs it into
BlobFileBuilder::GetCompressor(). The option is registered as mutable
and dynamically changeable via SetOptions().
For example, setting level=1 with ZSTD switches from "doubleFast"
(two hash tables per thread) to "fast" (single hash table), reducing
L3 cache pressure on flush-heavy workloads.
There was an existing TODO in blob_file_builder.cc requesting exactly
this feature — this diff fulfills it.
This change is safe as the default value is always dFast. So this will not impact the current code.
Reviewed By: xingbowang
Differential Revision: D99478562
fbshipit-source-id: e79b847a854f60dea552442b18fd3f1384efac70
Summary:
This PR adds a `BlobFilePartitionStrategy` interface that allows users to plug in custom partition selection logic for blob direct writes. The default behavior (round-robin across partitions) is preserved as the built-in `RoundRobinBlobFilePartitionStrategy`.
### Motivation
The existing blob direct write implementation distributes writes across partitions using a fixed round-robin strategy (atomic counter mod num_partitions). While this works well for uniform workloads, some use cases benefit from custom routing:
- **Key-based routing**: co-locate related keys in the same blob partition for locality-aware reads.
- **CF-based routing**: direct writes for different column families to dedicated partitions.
- **Value-size routing**: send large vs. small values to different partitions.
- **Application-defined affinity**: any domain-specific grouping that the default round-robin cannot express.
### Design
A new pure-virtual interface `BlobFilePartitionStrategy` is added to `include/rocksdb/advanced_options.h`:
```cpp
class BlobFilePartitionStrategy {
public:
virtual ~BlobFilePartitionStrategy() = default;
// Select a partition for the given blob direct write.
// The return value can be any uint32_t; the caller applies
// modulo num_partitions internally.
// Implementations must be thread-safe.
virtual uint32_t SelectPartition(uint32_t num_partitions,
uint32_t column_family_id,
const Slice& key,
const Slice& value) = 0;
};
```
The strategy receives:
- `num_partitions`: the configured partition count (for implementations that want to be partition-count-aware).
- `column_family_id`: useful for CF-based routing.
- `key` / `value`: the key and value being written.
The return value is taken modulo `num_partitions` internally, so implementations can return any `uint32_t` without worrying about bounds.
A new column family option `blob_direct_write_partition_strategy` (type `std::shared_ptr<BlobFilePartitionStrategy>`, default `nullptr`) wires the strategy into `BlobFilePartitionManager`. When `nullptr`, the built-in `RoundRobinBlobFilePartitionStrategy` is used, preserving existing behavior exactly.
### Changes
| File | Change |
|---|---|
| `include/rocksdb/advanced_options.h` | New `BlobFilePartitionStrategy` interface; new `blob_direct_write_partition_strategy` option |
| `db/blob/blob_file_partition_manager.h/.cc` | Accept strategy in constructor; replace atomic counter with strategy call; move `RoundRobinBlobFilePartitionStrategy` here as default |
| `options/cf_options.h/.cc` | Thread strategy through `ImmutableCFOptions` |
| `options/options_helper.cc` | Propagate strategy in `UpdateColumnFamilyOptions` |
| `options/options_settable_test.cc` | Register new option field in options settability test |
| `db/db_impl/db_impl.cc` | Pass strategy when constructing `BlobFilePartitionManager` |
| `db/blob/db_blob_direct_write_test.cc` | New test with `FixedBlobDirectWritePartitionStrategy` |
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14565
Test Plan:
New test `DirectWriteCustomPartitionStrategyRoutesWritesToOneBlobFile` verifies that a `FixedBlobDirectWritePartitionStrategy` that always returns partition 3 correctly routes all writes to a single blob file across 4 configured partitions.
```
make -j128 db_blob_direct_write_test && ./db_blob_direct_write_test --gtest_filter='*CustomPartition*'
make -j128 db_blob_direct_write_test && ./db_blob_direct_write_test
```
Reviewed By: joshkang97
Differential Revision: D99458813
Pulled By: xingbowang
fbshipit-source-id: 54cfbb0f75e24b58a8db61ad8755204d0db6ece7
Summary:
When BatchedOpsStressTest runs on a TransactionDB, db_->Write() creates internal transactions using default_lock_timeout (1 second), which is too short under heavy contention with many threads. This causes sporadic "Timeout waiting to lock key" errors printed to stderr, which the crash test framework treats as fatal failures.
Increase default_lock_timeout to 600000ms (10 minutes) to match the lock_timeout already used for explicit transactions in NewTxn().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14540
Test Plan: Manual `make crash_test_with_wc_txn` as a sanity check
Reviewed By: mszeszko-meta
Differential Revision: D98944474
Pulled By: pdillinger
fbshipit-source-id: 13b9e13fddc6332da010b01f7c41a51748f75624
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14566
**What**: Adds an explicit `file->Close()` call in `WriteStringToFile()` before
the function returns, instead of relying on the `unique_ptr` destructor to close
the file.
**How**: In `file_system.cc`, after the optional `Sync()` and before the
error-cleanup path, calls `file->Close()`. If `Close()` fails, the file is
deleted just like any other write failure. Adds two unit tests in `env_test.cc`:
- `WriteStringToFileClosesFile`: uses `CountedFileSystem` to verify `Close()` is
called and content is written correctly.
- `WriteStringToFileCloseFailureDeletesFile`: uses a custom `CloseFailFS`
wrapper to inject a `Close()` failure and verify the file is deleted on error.
**Why**: The previous implementation never called `Close()` explicitly — it
relied on the `unique_ptr` destructor which may silently swallow write errors
that occur during close (e.g., flushing buffered data). Explicit `Close()`
ensures such errors are detected and propagated to the caller.
Reviewed By: archang19
Differential Revision: D99462173
fbshipit-source-id: 525b5489bd0dbfc3159b007a13f9474f84d3c84e
Summary:
**Summary:**
Disable `use_trie_index` in `db_crashtest.py` to avoid trie UDI stress test failures.
The default param was previously set to randomly enable trie index ~12.5% of the time (`random.choice([0, 0, 0, 0, 0, 0, 0, 1])`). Setting it to `0` until the underlying issues are resolved.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14559
Test Plan: Stress test runs without trie UDI failures.
Reviewed By: hx235
Differential Revision: D99343747
Pulled By: xingbowang
fbshipit-source-id: 6f86b5dd5ddb2869d797e93ddbda521981614bae
Summary:
AnonExpectedState::Open() never initialized the base class
persisted_seqno_ pointer, leaving it as nullptr. Any call to
SetPersistedSeqno() or GetPersistedSeqno() on an AnonExpectedState
(used when expected_values_dir is empty) would dereference a null
pointer. Fix by allocating and assigning it in Open(), matching
FileExpectedState's pattern.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14523
Reviewed By: mszeszko-meta
Differential Revision: D98556119
Pulled By: hx235
fbshipit-source-id: c4c3103272735d36c1d05b96f7d009b6b750bf53
Summary:
Add a git pre-push hook that automatically runs `make format-auto` before every push. If formatting issues are found, the hook fixes them, creates a new commit, and retries the push — all in one shot.
The hook is auto-configured on the first `make` build via `core.hooksPath`, so there is no manual install step needed.
## Changes
**`githooks/pre-push`** — pre-push hook that:
1. Runs `make check-format` to detect formatting issues
2. If issues found, runs `make format-auto` to fix them
3. Creates a new "format" commit with the fixes
4. Retries the push automatically with the format commit included
5. Stashes/restores any uncommitted work so it is not affected
6. Skips in non-interactive environments (CI) unless `ROCKSDB_FORMAT_HOOK` is set
7. Can be skipped with `git push --no-verify`
**`Makefile`**:
- `setup-hooks` target: auto-sets `core.hooksPath=githooks` in the local repo config. Runs as a dependency of `all`, so any `make` build activates the hook with zero manual setup.
- `install-hooks` / `uninstall-hooks` targets: manual copy-based alternative for those who prefer it
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14558
Reviewed By: archang19
Differential Revision: D99324330
Pulled By: xingbowang
fbshipit-source-id: b824e56d572ad423fab7de7e15ead5fa8fd8e847
Summary:
When CompactionJob::Run() succeeds but Install() fails (e.g., LogAndApply MANIFEST I/O error), compact_->status was never updated with the install failure. CleanupCompaction() passed the stale OK status to SubcompactionState::Cleanup(), which skipped ReleaseObsolete -- leaking table cache entries for output files that were cached by VerifyOutputFiles but never installed into any Version.
This is the same class of bug fixed in https://github.com/facebook/rocksdb/issues/14469 (where Run() failed after VerifyOutputFiles), but in the Install() failure path. The FindObsoleteFiles full-scan backstop would normally catch this, but fails under crash test metadata read fault injection
(--open_metadata_read_fault_one_in), causing the
TEST_VerifyNoObsoleteFilesCached assertion to fire during Close().
Fix: propagate Install()'s local status back to compact_->status before CleanupCompaction(), so Cleanup() sees the failure and calls ReleaseObsolete on the output files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14549
Test Plan:
New unit test DBCompactionTest.LeakedTableCacheEntryOnInstallFailure:
- Without fix (ASAN): assertion fires -- "File 12 is not live nor quarantined"
- With fix (ASAN): passes -- ReleaseObsolete properly cleans up the entry
```
COMPILE_WITH_ASAN=1 make -j db_compaction_test
./db_compaction_test --gtest_filter="DBCompactionTest.LeakedTableCacheEntry*"
[ PASSED ] 2 tests.
```
Reviewed By: joshkang97
Differential Revision: D99155908
Pulled By: pdillinger
fbshipit-source-id: ed5374a38d7903866a38a0fe0f5539e12321bc84
Summary:
This file was accidentally included in https://github.com/facebook/rocksdb/issues/14535. It contains local Claude Code permission settings that are user-specific and should not be in the repo.
Also adds it to `.gitignore` to prevent future accidental commits.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14554
Reviewed By: pdillinger
Differential Revision: D99296922
Pulled By: xingbowang
fbshipit-source-id: e229dd85179ae25bb15df1633bba7197dd2fa1f9
Summary:
This PR introduces **blob direct write v1**, a reduced-scope write-path optimization where large values (>= `min_blob_size`) are written directly to blob files during `Put()` and replaced in the memtable with compact `BlobIndex` references. This avoids holding full values in memory until flush time.
### Motivation
In the existing BlobDB architecture, values are written to the WAL and memtable in their full form and separated into blob files only at flush time. This means:
- Large values are held in memory twice (raw in memtable + blob file at flush)
- Blob I/O is serialized through a single flush thread per column family
Blob direct write addresses both: values leave the write path as small `BlobIndex` references, and multiple **partitions** (configurable via `blob_direct_write_partitions`) allow concurrent blob writes with independent locks.
### Design (v1 — single-writer, WAL-disabled, reduced scope)
The v1 design intentionally keeps scope narrow for correctness and reviewability:
- **Single writer thread assumption**: no concurrent writes to the same partition file. One logical writer serializes the batch.
- **WAL-disabled**: direct-write blob files are only registered in MANIFEST at flush time. WAL replay cannot recover unregistered blob references, so WAL is disabled for this v1.
- **Flush-on-write**: each `AddRecord` call flushes to the OS immediately.
- **FIFO generation batching**: each memtable switch creates one generation batch. Direct-write files for that memtable are sealed and registered atomically when the batch is flushed to MANIFEST.
- **Round-robin partitions**: blob writes are distributed across `blob_direct_write_partitions` files using an atomic counter.
### New components
| Component | Description |
|---|---|
| `BlobFilePartitionManager` | Owns N partition files per CF. Manages open/seal/register lifecycle tied to memtable generations. |
| `BlobWriteBatchTransformer` | A `WriteBatch::Handler` that rewrites qualifying `Put` values as `BlobIndex` entries before the batch enters the write group. |
### Write path integration
1. `DBImpl::WriteImpl` calls `BlobWriteBatchTransformer::TransformBatch` before entering the writer group (for default write path), or before joining the batch group (for pipelined/unordered write).
2. Values >= `min_blob_size` are written to a partition file; the key is stored with a `BlobIndex` in the transformed batch. A rollback guard marks blob bytes as initial garbage if the write fails.
3. On `SwitchMemtable`, `RotateCurrentGeneration` moves active partitions into the next immutable batch.
4. `FlushMemTableToOutputFile` / `AtomicFlushMemTablesToOutputFiles` call `PrepareFlushAdditions` to seal partition files and collect `BlobFileAddition` + `BlobFileGarbage` entries registered to MANIFEST alongside the flush.
5. Shutdown paths (`CancelAllBackgroundWork`, `WaitForCompact` with `close_db=true`) force-flush all CFs with active direct-write managers to ensure blob files are registered before close.
### Read path
- **Get/MultiGet**: `MaybeResolveBlobForWritePath` resolves `BlobIndex` references found in memtable or immutable memtable via `BlobFilePartitionManager::ResolveBlobDirectWriteIndex`, which first checks manifest-visible state and falls back to direct blob-file reads via `BlobFileCache`.
- **Iterator**: `DBIter::BlobReader` is extended with a `BlobFilePartitionManager*` to resolve direct-write blob indexes during iteration. The unified `ResolveBlobDirectWriteIndex` path handles both manifest-visible and not-yet-flushed files.
### New options
| Option | Default | Description |
|---|---|---|
| `enable_blob_direct_write` | `false` | Enable write-path blob separation for this CF. Requires `enable_blob_files = true`. Not dynamically changeable. |
| `blob_direct_write_partitions` | `1` | Number of parallel partition files per CF. Not dynamically changeable. |
### Feature incompatibilities (reduced v1 scope)
The following features are *not supported* when `enable_blob_direct_write = true`, and are enforced both in `db_stress_tool` validation and `db_crashtest.py` sanitization:
**Write model constraints:**
- `threads` must be 1 (single writer assumption)
- `allow_concurrent_memtable_write` = 0
- `enable_pipelined_write` = 0 (transformation done before batch group, but pipelined path supported with pre-transform)
- `two_write_queues` = 0
- `unordered_write` = 0 (transformation done before batch group, but unordered path supported with pre-transform)
**WAL and recovery:**
- `disable_wal` = 1 (required — WAL replay of unregistered blob files is out of v1 scope)
- `best_efforts_recovery` = 0
- `reopen` = 0 (no crash-restart with WAL replay)
- All WAL-related stress features disabled: `manual_wal_flush_one_in`, `sync_wal_one_in`, `lock_wal_one_in`, `get_sorted_wal_files_one_in`, `get_current_wal_file_one_in`, `track_and_verify_wals`, `rate_limit_auto_wal_flush`, `recycle_log_file_num`
**Blob GC and dynamic options:**
- `use_blob_db` = 0 (stacked BlobDB not supported)
- `allow_setting_blob_options_dynamically` = 0
- `enable_blob_garbage_collection` = 0
- `blob_compaction_readahead_size` = 0
- `blob_file_starting_level` = 0
**Unsupported value types and APIs:**
- Merge (`use_merge`, `use_full_merge_v1`) — merge values pass through untransformed
- Entity APIs (`use_put_entity_one_in`, `use_get_entity`, `use_multi_get_entity`, `use_attribute_group`)
- `use_timed_put_one_in`
- User-defined timestamps (`user_timestamp_size`, `persist_user_defined_timestamps`, `create_timestamped_snapshot_one_in`)
- Transactions (`use_txn`, `use_optimistic_txn`, `test_multi_ops_txns`, `commit_bypass_memtable_one_in`) — though `WriteCommittedTxn::CommitInternal` falls back from bypass-memtable to normal path when BDW is active
- `IngestWriteBatchWithIndex` returns `NotSupported`
- `inplace_update_support` = 0
**Fault injection:**
- All write/read/metadata fault injection disabled (`sync_fault_injection`, `write_fault_one_in`, `metadata_write_fault_one_in`, `read_fault_one_in`, `metadata_read_fault_one_in`, `open_*_fault_one_in`)
**Infrastructure/snapshot APIs:**
- `remote_compaction_worker_threads` = 0
- `test_secondary` = 0
- `backup_one_in` = 0
- `checkpoint_one_in` = 0
- `get_live_files_apis_one_in` = 0
- `ingest_external_file_one_in` = 0
- `ingest_wbwi_one_in` = 0
### Tests
- `db/blob/db_blob_basic_test.cc`: ~660 lines of new direct-write unit tests covering basic put/get, multi-partition, flush/compaction, recovery, and error injection.
- `db/blob/blob_file_cache_test.cc`: ~96 lines of new tests for direct-write blob file cache behavior.
- `db/write_batch_test.cc`: ~96 lines of tests for WriteBatch with blob index entries.
- `utilities/transactions/transaction_test.cc`: verifies transaction commit path falls back correctly with direct write enabled.
- `db_stress_tool/`: full stress test support with `--enable_blob_direct_write` and `--blob_direct_write_partitions` flags, integrated into `db_crashtest.py` with 10% random selection alongside regular blob params.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14535
Test Plan:
```
make -j128 db_blob_basic_test && ./db_blob_basic_test
make -j128 blob_file_cache_test && ./blob_file_cache_test
make -j128 write_batch_test && ./write_batch_test
make -j128 transaction_test && ./transaction_test
make -j128 check
```
Stress test:
```
python3 tools/db_crashtest.py blackbox --enable_blob_direct_write=1 \
--enable_blob_files=1 --blob_direct_write_partitions=4 \
--disable_wal=1 --threads=1
```
Reviewed By: pdillinger
Differential Revision: D98766843
Pulled By: xingbowang
fbshipit-source-id: 1577653826913a59d05680a87bce5534ac5a5e69
Summary:
Transactions in rockdb supported adding log data using the `put_log_data` api, but this was not exported for other language bindings. Exported this binding allows other languages like rust, go, etc add log data on an transaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14543
Reviewed By: joshkang97
Differential Revision: D99171663
Pulled By: xingbowang
fbshipit-source-id: 24c94d71668a41cc7b2913972427255ab67d0863
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14521
ReadAsync calls io_uring_get_sqe() without checking for nullptr. When the
io_uring submission queue is full (outstanding completions not yet reaped),
io_uring_get_sqe returns NULL and the subsequent io_uring_prep_readv
dereferences it, causing a segfault.
MultiRead already handles this correctly by using io_uring_sq_space_left()
to cap submissions. ReadAsync submits exactly one SQE per call so a simple
null check with error return is sufficient.
On null sqe, clean up the already-allocated Posix_IOHandle and return
IOStatus::Busy so the caller can retry after reaping completions.
The io_uring queue depth is kIoUringDepth (256), and each thread gets its
own io_uring instance via thread-local storage. In practice the SQ rarely
fills because ReadAsync calls io_uring_submit() after each io_uring_get_sqe(),
immediately flushing the SQE to the kernel. The null SQE would only occur
under unusual kernel backpressure where the kernel cannot consume from the
SQ ring fast enough.
IOStatus::Busy was chosen (over IOError) because this is a transient
condition. The caller has two options:
1. Call Poll() to reap outstanding completions from the CQ, then retry
ReadAsync. This mirrors how MultiRead handles queue pressure internally
by capping submissions and reaping between batches.
2. Fall back to synchronous Read(). Existing callers (FilePrefetchBuffer,
IODispatcher) already have synchronous fallback paths for non-OK
ReadAsync status, so IOStatus::Busy naturally triggers that fallback
without additional code changes. Given the rarity of this condition,
the synchronous fallback is pragmatic and avoids adding retry complexity.
Also adds a TEST_SYNC_POINT_CALLBACK on io_uring_get_sqe to enable test
injection, and a new ReadAsyncQueueFull unit test that uses SyncPoint to
force a null SQE and verifies the Busy return, handle cleanup, and no crash.
Reviewed By: xingbowang
Differential Revision: D98533853
fbshipit-source-id: f6d181e5c0d5154b570ff6da39e2f52e2a6aea84
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14514
SupportedOps previously checked only that the ThreadLocalPtr container existed (non-null), which was set during construction based on a one-time probe on the main thread. However, CreateIOUring() can fail on other threads due to kernel resource limits or flag incompatibilities, causing ReadAsync to hit "failed to init io_uring" at runtime.
Now SupportedOps eagerly initializes the thread-local io_uring instance via Get() + CreateIOUring() and only advertises kAsyncIO if it succeeds. Also logs to stderr (with thread id) when io_uring init fails, and adds a TEST_SYNC_POINT_CALLBACK for testing simulated CreateIOUring failures.
Reviewed By: mszeszko-meta, xingbowang
Differential Revision: D98409140
fbshipit-source-id: efa92d9ac920860e95a46710c4a87e36bacbb466
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14522
Add two verification steps to RocksDB CLAUDE.md:
1. **Unit Test flakiness testing**: After writing a test, stress-test for
flakiness with `COERCE_CONTEXT_SWITCH=1 make {test_binary}` followed by
`--gtest_repeat=5`. This catches race conditions and context-switch-dependent
failures early.
2. **Status object verification**: Run `ASSERT_STATUS_CHECKED=1 make check`
to catch missing error handling that can lead to silent data corruption.
These are compile-time checks enforced via a special build mode.
These are existing RocksDB best practices that both the stress test agent
and human developers should follow when writing or fixing code.
Reviewed By: xingbowang
Differential Revision: D98249994
fbshipit-source-id: 25ca6a628a82ecf968d5ed86aaa5c2f4b1060471
Summary:
This PR fixes several correctness and stability issues in the stress test and crash test infrastructure, plus adds regression coverage for trie UDI iterators.
### db_stress: constrain batched prefix scans
`BatchedOpsStressTest::TestPrefixScan` opens 10 iterators with different prefixes and compares results in lockstep. Without `prefix_same_as_start`, unconstrained iterators could return keys beyond their seek prefix, producing more entries than bounded iterators and causing spurious assertion failures. Fix: set `prefix_same_as_start = true` on all iterators so every iterator stops at the seek prefix boundary.
### db_stress: skip invalid cross-prefix iterator checks
When prefix iteration is enabled, `ReadOptions` requires that the seek key and `iterate_lower_bound` share the same prefix. Previously, stress test verification could run against configurations where they don't (e.g. when `lower_bound` spans a prefix boundary), producing false positives. Fix: detect this invalid configuration and mark the check as diverged (skip verification) rather than asserting.
### db_crashtest: disable BlobDB in best-efforts recovery
BlobDB is not compatible with best-efforts recovery mode. Fix: explicitly disable blob file options when `best_efforts_recovery=True` in db_crashtest.py to avoid spurious failures.
### db_crashtest: preserve expected-state dirs across restarts
The expected-state directory was being wiped on certain restart paths, causing verification failures on the next run. Fix: preserve expected-state dirs across db_crashtest restarts. Adds a new `db_crashtest_test.py` test to verify the behavior, runnable via `make db_crashtest_test`.
### Add trie UDI iterator regression coverage
Adds regression tests for trie-based UserDefinedIndex (UDI) iterators covering snapshot-based reads, lower/upper bound iteration, `auto_refresh_iterator_with_snapshot`, and multi-version key handling. Also fixes an MSVC C4244 warning (int→char implicit narrowing) in the test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14512
Reviewed By: hx235
Differential Revision: D98302018
Pulled By: xingbowang
fbshipit-source-id: 84f5878665e5aeb61338ec2b6bfb2d2b077bb9f2
Summary:
CONTEXT: MyRocks and other third-party users currently need to understand deep RocksDB internals to sort unsorted data into SST files. The existing pattern requires: opening a full DB instance, configuring VectorRepFactory with universal compaction, disabling auto-compaction, writing unsorted data, manually triggering CompactRange with kForceOptimized, collecting output via GetColumnFamilyMetaData, and finally ingesting via IngestExternalFile with allow_db_generated_files. This is error-prone and requires ~30 lines of RocksDB configuration knowledge.
WHAT: This adds a new utility class, SortedRunBuilder, that wraps all of the above into a simple Create/Add/Finish API. Callers feed in unsorted key-value pairs (in any order, from any number of threads) and receive sorted SST files with seqno=0 ready for ingestion -- or can iterate sorted output directly.
The name SortedRunBuilder was chosen over alternatives like UnsortedSstFileWriter because this utility targets a broad audience: anyone wanting to use RocksDB as an external sort engine, not just users migrating from SstFileWriter. That said, this explicitly removes the sorted-input requirement that SstFileWriter enforces -- callers no longer need to pre-sort their data before writing SST files.
KEY DESIGN DECISIONS:
- Zero changes to core RocksDB internals. This is a pure utility layer that exclusively calls existing public APIs:
* DB::Open(), DB::Put(), DB::Write(), DB::Flush(), DB::CompactRange(), DB::GetColumnFamilyMetaData(), DB::NewIterator(), DestroyDB()
* VectorRepFactory (sort-on-flush memtable)
* BottommostLevelCompaction::kForceOptimized (zero seqnos)
- No modifications to: compaction logic, memtable implementation, SST file format, ingestion logic, or DB open/close paths
- All new code lives in utilities/sorted_run_builder/ and the public header include/rocksdb/utilities/sorted_run_builder.h
- Build file changes are limited to registering the new source/test files
API SURFACE:
SortedRunBuilderOptions opts; opts.temp_dir = "/tmp/sort_work"; std::unique_ptr<SortedRunBuilder> builder; SortedRunBuilder::Create(opts, &builder);
builder->Add(key, value); // any order, thread-safe
builder->AddBatch(&batch); // WriteBatch for throughput
builder->Finish(); // flush + compact + collect
builder->GetOutputFiles(); // sorted SSTs for ingestion
builder->NewIterator(ro); // iterate sorted output
FILES CHANGED:
New: include/rocksdb/utilities/sorted_run_builder.h (public header) New: utilities/sorted_run_builder/sorted_run_builder.cc (implementation) New: utilities/sorted_run_builder/sorted_run_builder_test.cc (12 tests) New: docs/plans/sorted_run_builder_plan.md (design plan) New: docs/plans/sorted_run_builder_usage_guide.md (usage guide) Modified: src.mk, CMakeLists.txt, Makefile (register new files only)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14499
Test Plan:
$ make clean && make -j$(nproc) sorted_run_builder_test $ ./sorted_run_builder_test [==========] Running 12 tests from 1 test case. [ PASSED ] 12 tests. (688 ms total)
Tests cover: basic sort correctness, empty builder, WriteBatch path, concurrent multi-threaded writes, ingestion into a target DB, large random dataset (10K keys), entry/size counters, error cases (Add after Finish, iterator before Finish, empty temp_dir), cleanup verification, and duplicate key handling.
Reviewed By: xingbowang
Differential Revision: D98935972
Pulled By: dannyhchen
fbshipit-source-id: e3bb7f1ea5f6004aab697e4da667fa21292ca250
Summary:
`teardown-ccache` runs with if: always(), but in folly jobs `setup-folly` precedes `setup-ccache`. When `setup-folly` fails, `setup-ccache` is skipped, so `CCACHE_DIR` is unset and ccache is not on `PATH`. This is exactly what happened [here](https://github.com/facebook/rocksdb/actions/runs/23660261947/job/68928337162). Fix guards `teardown-ccache` to exit gracefully when `CCACHE_DIR` is unset, and tolerate missing ccache binary for stats.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14542
Reviewed By: joshkang97
Differential Revision: D98964757
Pulled By: mszeszko-meta
fbshipit-source-id: 8e83c1b62ef66130ba479142f002897d0a97f6c0
Summary:
The warm_storage_crash_test was failing because UniqueIdVerifier stored its .unique_ids bookkeeping file in the DB directory, which lives on warm storage. After a crash, warm storage's weaker durability guarantees could cause flushed-but-not-synced data to be lost, making the file appear shorter on a second open within the same constructor. This caused CopyFile to return Corruption and trigger assert(false).
Move the file to expected_values_dir (which is always on local filesystem) and always use Env::Default() for file operations, so that POSIX flush semantics are sufficient for read consistency without needing Sync.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14539
Test Plan: Full local run of `make blackbox_crash_test`
Reviewed By: mszeszko-meta
Differential Revision: D98934451
Pulled By: pdillinger
fbshipit-source-id: 5497f23c2382cb5ac2c3d7f238c6c4ca1dd29f5b
Summary:
Adds regression tests and failure-diagnostic tooling for the trie UDI post-seek seqno correction bug (T258590238). The core fix itself already landed in https://github.com/facebook/rocksdb/issues/14466.
## What's in this PR
### Regression tests
- `TrieIndexFactoryTest.ZeroSeqMustNotSkipLeafForSmallerUserKey` — minimal unit test proving the original bug; exercises the seqno-based block advancement with a smaller user key that should NOT trigger advancement
- `TrieIndexDBTest.AutoRefreshSnapshotNextAcrossSameUserKeyBoundaries` — DB-level test for snapshot-refresh iteration across same-user-key block boundaries
- `TrieIndexDBTest.AutoRefreshSnapshotNextAfterCompactionAcrossSameUserKeyBoundaries` — same scenario after compaction reshapes SST layout
- `TrieIndexDBTest.AutoRefreshSnapshotStressLikeSingleCfCoalescingIterator` — stress-like test exercising snapshot-refresh + coalescing-iterator interaction
### Diagnostic logging in db_stress
Extracts the iterator verification failure dump into a `DumpIteratorVerificationFailure()` helper in `NonBatchedOpsStressTest`. On verification failure, logs:
- Expected-state window (pre/post read values, raw state, pending flags)
- Iterator config (UDI, trie, snapshot, multi-CF)
- Replay comparison: creates fresh iterators with standard vs trie index, direct vs coalescing, seek-to-failure-key vs replay-from-mid — making it possible to identify which index/iterator combination diverges
This logging was instrumental in triaging the original bug and will help with future trie UDI stress failures.
## Tests
All existing trie index tests pass. New tests listed above all pass.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14524
Reviewed By: hx235
Differential Revision: D98583342
Pulled By: xingbowang
fbshipit-source-id: 57e099055d6beae9b5f03f4e6605f0af6e65b94a
Summary:
The following performance optimizations are included in this PR:
- Inline NextSetBit/PrevSetBit into header (called on every trie level)
- Unroll Rank1 popcount loop (the single hottest function)
- Advance/Retreat: replace path stack top in-place instead of pop+push
- std::swap for prev_key_scratch_ instead of O(n) string copy
- assign() instead of ToString() to reuse buffer capacity
- Cache IndexValue in wrapper to avoid repeated virtual dispatch
- Mark TrieIndexIterator/TrieIndexBuilder as final for devirtualization
Part of https://github.com/facebook/rocksdb/issues/12396
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14466
Reviewed By: anand1976
Differential Revision: D97979699
Pulled By: xingbowang
fbshipit-source-id: 7ec56ecc8b6d68548a336dd1aaccfb215189eff0
Summary:
Add tracking of uniform index blocks as a table property (`num_uniform_blocks`). When `uniform_cv_threshold` is set, the block builder detects uniformly distributed keys via coefficient of variation of key gaps. This information is now surfaced end-to-end: from block building through index builders to SST table properties.
## Key changes
- **BlockBuilder**: Persist the uniformity result from `ScanForUniformity()` as a member (`is_uniform_`) exposed via `IsUniform()`, rather than a local variable discarded after `Finish()`.
- **Index builders**: All three index builder types (`ShortenedIndexBuilder`, `HashIndexBuilder`, `PartitionedIndexBuilder`) implement `NumUniformIndexBlocks()`. For partitioned indexes, the count accumulates across partition `Finish()` calls.
- **Table property serialization**: Added `TablePropertiesNames::kNumUniformBlocks` (`"rocksdb.num.uniform.blocks"`) with full serialization/deserialization support. Without this, the property was computed in memory but never persisted to SST files.
- **Table property aggregation**: `num_uniform_blocks` included in `Add()` and `GetAggregatablePropertiesAsMap()` for correct cross-SST aggregation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14513
Test Plan:
- **Unit tests**: `block_test` (7504 passed), `table_test` (6917 passed), `table_properties_collector_test` (8 passed).
- **End-to-end verification via db_bench + sst_dump**:
- Positive: `./db_bench --benchmarks=fillseq,compact --num=10000 --uniform_cv_threshold=0.2 --index_shortening_mode=0 --compression_type=none` produces SST with `# uniform blocks: 1`.
- Negative: Same with `--uniform_cv_threshold=-1` (disabled) produces `# uniform blocks: 0`.
Reviewed By: xingbowang
Differential Revision: D98343170
Pulled By: joshkang97
fbshipit-source-id: ed08e83b2dcfb70f074bfb0f7bb5c31d75dc6da9
Summary:
## Problems
**(1) Claude review times out on large PRs**
The `claude-code-base-action` defaults to `timeout_minutes=10`. Large PRs (e.g. https://github.com/facebook/rocksdb/issues/14499) get killed with exit code 124 after 600 seconds mid-review.
**(2) Exported PRs never get reviewed**
PRs exported from Meta's internal pipeline (e.g. https://github.com/facebook/rocksdb/issues/14515) trigger CI within milliseconds of PR creation. The `workflow_run` payload has `pull_requests=[]`, and the SHA-based fallback also misses because GitHub hasn't registered the PR yet — so Claude review never fires.
## Fixes
- Set `timeout_minutes: "60"` on both auto-review and manual-review `Run Claude` steps
- Retry the SHA→PR lookup up to 5 times with a 10s delay, giving GitHub up to ~50s to register the PR. Also bumped `per_page` from 30 to 100.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14526
Reviewed By: pdillinger
Differential Revision: D98636217
Pulled By: xingbowang
fbshipit-source-id: bba19095ab2ddf468c5c19cf5c77d19536f498b0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14520
CreateIOUring() silently returns nullptr on failure, discarding the errno
from io_uring_queue_init. This makes it impossible to diagnose why
io_uring initialization fails on specific threads (e.g. ENOMEM from
memlock limits, EINVAL from unsupported flags, EMFILE from fd exhaustion).
Add a fprintf(stderr, ...) that logs strerror, errno, and pthread thread
ID when io_uring_queue_init fails, so failures are diagnosable from logs
without needing to reproduce.
Reviewed By: xingbowang
Differential Revision: D98526792
fbshipit-source-id: 1eb5042c8b62663c4d24c09f29ef7c55b90032f0
Summary:
Add read-triggered compaction, a new feature that reduces read amplification by compacting SST files that receive high read traffic. When an SST file's read frequency (`num_reads_sampled / file_size`) exceeds a configurable threshold, it is marked for compaction to a lower level.
The feature introduces two new options: a CF option `read_triggered_compaction_threshold` (default 0, disabled) and a DB option `max_periodic_compaction_trigger_seconds` (default 43200s) that controls how often the background thread re-evaluates compaction scores on quiet databases. Both options are dynamically changeable.
Lowering `max_periodic_compaction_trigger_seconds` does add some overhead, but generally is minimal, so running this every couple of minutes in a production environment seems fairly reasonable.
## Key changes
- **New CF option `read_triggered_compaction_threshold`** (`advanced_options.h`): When positive, files with `reads_per_byte > threshold` are marked for compaction. Files at the last non-empty level are skipped (bottommost compaction handles those separately). Marked files are sorted by hotness (reads_per_byte descending).
- **New DB option `max_periodic_compaction_trigger_seconds`** (`options.h`): Replaces the hardcoded 12-hour ceiling in `ComputeTriggerCompactionPeriod()`. Essential for read-triggered compaction on quiet DBs since there are no writes to trigger score re-evaluation.
- **Leveled compaction picker** (`compaction_picker_level.cc`): Adds read-triggered as the lowest-priority compaction reason in `SetupInitialFiles()`, using the existing `PickFileToCompact` helper.
- **Universal compaction picker** (`compaction_picker_universal.cc`): Adds `PickReadTriggeredCompaction` as lowest priority. Refactors shared "find output level + compute overlapping inputs + create Compaction" logic from both `PickDeleteTriggeredCompaction` and `PickReadTriggeredCompaction` into `BuildCompactionToNextLevel`, handling both single-level and multi-level universal cases.
- **Periodic trigger integration** (`db_impl.cc`): `TriggerPeriodicCompaction` now also fires for CFs with `read_triggered_compaction_threshold > 0`, even without time-based compaction configured.
- **Stress test & db_bench support**: Both `db_stress` and `db_bench` support the new options. `db_crashtest.py` randomly enables read-triggered compaction and sets a short periodic trigger interval when enabled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14426
Test Plan:
**Unit tests**:
- `compaction_picker_test` — 7 new tests: `ReadTriggeredCompactionDisabled`, `ReadTriggeredCompactionBelowThreshold`, `ReadTriggeredCompactionAboveThreshold`, `NeedsCompactionReadTriggered`, `ReadTriggeredPicksFile`, `UniversalReadTriggeredCompaction`, `ReadTriggeredSkipsLastLevel`, `UniversalReadTriggeredNoPickWhenNotMarked`
- `db_compaction_test` — `ReadTriggeredCompaction` integration test verifying end-to-end behavior with sync points
- Stress test coverage
**Stress test**:
```
make V=1 -j "CRASH_TEST_EXT_ARGS=--duration=600 --max_key=2500000 --max_compaction_trigger_wakeup_seconds=10
--read_triggered_compaction_threshold=0.0001 --interval=600" blackbox_crash_test
```
- confirmed read triggered compactions from LOGS
**Benchmark** (`db_bench`):
Setup: 5M keys (100B values, 16B keys), leveled compaction, 5 levels, 4MB target file size. DB fully compacted, then 2M overlapping keys written without compaction to create L0/L1 overlap (82 files, ~294MB).
LSM shape change during readrandom with read-triggered compaction:
```
BEFORE: L0=9 files (15MB), L1=4 (16MB), L2=20 (69MB), L3=49 (194MB) — 82 files, 294MB
AFTER: L3=66 files (223MB)
```
| Benchmark | Config | avg ops/s | % change |
|-----------|--------|-----------|----------|
| readrandom (8 threads, 5M reads) | baseline (threshold=0) | 1,086,965 | — |
| readrandom (8 threads, 5M reads) | threshold=0.000001, trigger=5s | 1,453,697 | **+33.7%** |
Reviewed By: xingbowang
Differential Revision: D97838716
Pulled By: joshkang97
fbshipit-source-id: a21fcb270c7fadd4f78d98b9c821982f220dd3f0
Summary:
When `min_blob_size=0`, the existing guard condition:
```cpp
if (value.size() < min_blob_size_) {
return Status::OK(); // skip blob creation
}
```
is **false** for empty values (`0 < 0`), so empty values proceed to blob creation. This writes a 0-byte blob to disk and creates a `BlobContents` object with an empty `Slice`.
When that `BlobContents` is later evicted from the primary blob cache to `CompressedSecondaryCache`, the eviction handler calls `SaveTo(value, 0, 0, out)`, which hits:
```cpp
// typed_cache.h:230
assert(from_offset < slice.size()); // 0 < 0 → CRASH
```
This crash was found by the stress test with `--min_blob_size=0` and `--blob_cache` + secondary cache enabled (T261142690).
## Fix
Add an explicit `value.empty()` check before the blob path:
```cpp
if (value.empty() || value.size() < min_blob_size_) {
return Status::OK();
}
```
Empty values are now always stored inline in the SST, regardless of `min_blob_size`. This is also correct on principle: a `BlobIndex` reference is larger than an empty value, so storing an empty value as a blob is pure overhead with no benefit.
## Root Cause Chain
1. User writes `Put("key", "")` with `min_blob_size=0`
2. `BlobFileBuilder::Add()` — `0 < 0` is false, empty value proceeds to blob creation
3. 0-byte blob written to blob file; `BlobContents` created with 0-size slice
4. `BlobContents` inserted into primary blob cache (LRU)
5. Cache eviction triggers `CacheWithSecondaryAdapter::EvictionHandler()`
6. `CompressedSecondaryCache::InsertInternal()` → `SaveTo(value, 0, 0, out)`
7. `assert(from_offset < slice.size())` → `assert(0 < 0)` → **** assertion failure
## Test
Added `DBBlobBasicTest.EmptyValueNotStoredAsBlob` which:
- Writes an empty value and a non-empty value with `min_blob_size=0`
- Verifies both are readable
- Confirms the empty value is stored **inline** (readable from `kBlockCacheTier` without blob I/O)
- Confirms the non-empty value is stored as a **blob** (returns `IsIncomplete()` from `kBlockCacheTier`)
## Related
- Task: T261142690
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14517
Reviewed By: hx235
Differential Revision: D98499174
Pulled By: xingbowang
fbshipit-source-id: 5713923daec83db6491d00ce58acdf8231fabeba
Summary:
**Summary:**
Add a new boolean flag `include_blob_files` (default: `false`) to `SizeApproximationOptions` and a corresponding `INCLUDE_BLOB_FILES` enum value to `SizeApproximationFlags`. When set to `true`, the returned size includes an approximation of blob file data in the queried key range.
**Algorithm:**
The blob file size contribution is prorated using the SST size ratio:
```
blob_size_in_range ≈ total_blob_size * (sst_size_in_range / total_sst_size)
```
The blob-to-SST ratio (`total_blob_size / total_sst_size`) is computed once before the per-range loop, so iterating levels and blob files only happens once per `GetApproximateSizes` call regardless of how many ranges are queried. The per-range SST size (`ApproximateSize`) is computed once and shared between `include_files` and `include_blob_files`.
**Limitations:**
- Assumes blob data is distributed proportionally to SST data across the key space. May be inaccurate if blob value sizes vary significantly across different key ranges (e.g., one range has large blobs while another has small ones).
- If there are no SST files (all data in memtables), the blob size contribution will be 0 even if blob files exist on disk.
**Changes:**
- `include/rocksdb/options.h`: New `include_blob_files` field in `SizeApproximationOptions`; updated doc comments for `include_memtables`/`include_files`
- `include/rocksdb/db.h`: New `INCLUDE_BLOB_FILES` in `SizeApproximationFlags` enum, updated flags-to-options mapping
- `include/rocksdb/c.h`: New `rocksdb_size_approximation_flags_include_blob_files` C API enum value
- `java/`: Added `INCLUDE_BLOB_FILES` to `SizeApproximationFlag.java` and JNI flag mapping in `rocksjni.cc`
- `db/db_impl/db_impl.cc`: Blob-to-SST ratio computed once before loop, SST range size computed once per range and shared
- `db_stress_tool/db_stress_test_base.cc`: Randomized `include_blob_files` in stress test
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14501
Test Plan:
- New `DBBlobBasicTest.GetApproximateSizesIncludingBlobFiles` — verifies:
- Size with blobs > without (full range)
- Non-overlapping range returns 0
- Partial range returns proportionally less than full range
- `SizeApproximationFlags` API works
- Multi-range query: two sub-ranges sum approximately to the full-range result
- Stress test now exercises the new option randomly
Reviewed By: hx235
Differential Revision: D97984211
Pulled By: xingbowang
fbshipit-source-id: e9127eac3308687fd4f0b17a771fd61fba6a8380
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14518
When RocksDB creates an io_uring instance via `CreateIOUring()`, it calls
`io_uring_queue_init()` under the hood. This function asks the Linux kernel
to set up a submission queue and completion queue for async I/O — the kernel
allocates file descriptors and memory-mapped ring buffers to make this work.
To properly release those kernel resources, you must call
`io_uring_queue_exit()` before freeing the `io_uring` struct. This function
tells the kernel "I'm done with this io_uring" — it unmaps the shared
memory regions and closes the kernel file descriptors. Without it, those
resources leak every time an io_uring instance is destroyed.
`DeleteIOUring()` (the ThreadLocalPtr destructor callback) was only doing
`delete iu` — freeing the C++ struct's heap memory but never telling the
kernel to clean up. This meant every thread exit leaked kernel resources.
The same bug also existed in the `PosixFileSystem` constructor, which
creates a temporary test io_uring to check kernel support and then did
`delete new_io_uring` without cleanup.
Fix:
1. Add `io_uring_queue_exit(iu)` before `delete iu` in `DeleteIOUring()`.
2. Replace bare `delete new_io_uring` in fs_posix.cc with
`DeleteIOUring(new_io_uring)` to reuse the now-correct helper.
Note: the error-recovery path in io_posix.cc (~line 907) already correctly
calls `io_uring_queue_exit()` before `delete`, confirming this was the
intended pattern that was missed in these two spots.
Reviewed By: anand1976
Differential Revision: D98500559
fbshipit-source-id: bd6c10c19d9fd67cf537dd7f100ef9dc49bfe77e
Summary:
Fix build failures, flaky tests, and Windows ccache issues exposed by PR https://github.com/facebook/rocksdb/issues/14478.
### 1. Release build (NDEBUG) compile error
**Job**: `build-linux-release-with-folly`
The SyncPoint cleanup listener in testharness.cc referenced `SyncPoint::GetInstance()` unconditionally, but SyncPoint is only declared in debug builds. Wrapped with `#ifndef NDEBUG`.
### 2. Folly-lite link error
**Job**: `build-linux-cmake-with-folly-lite`
The `USE_FOLLY_LITE` cmake path unconditionally linked `-lglog`, which fails with lld (added in https://github.com/facebook/rocksdb/issues/14478) when glog isn't installed. Use `find_library()` to link only when available.
### 3. Flaky tests — leaked `Env::Default()` thread pool state
Sharded execution runs multiple tests per process. Several tests in `db_compaction_test.cc` modified the global `Env::Default()` BOTTOM thread pool without resetting. With a leaked BOTTOM pool, subsequent tests' last-level compactions get forwarded to the bottom pool, freeing the LOW thread to pick additional compactions unexpectedly.
Fix: add `TearDown()` override to `DBCompactionTest` that captures default thread pool sizes in the constructor and restores them after every test. This is more robust than per-test cleanup because:
- It runs even when a test fails (gtest calls TearDown after assertion failures)
- It catches all current and future leakers without per-test maintenance
### 4. Windows ccache — 0.26% hit rate → should be ~99%
The Windows nightly build takes 57 minutes because ccache has near-zero hit rate. Two issues:
- **Cache key**: The `hendrikmuhs/ccache-action` used a timestamp-based key, so each nightly run created a unique key and never found the previous run's cache (`No cache found.`). Fixed by using a stable key `ccache-windows-<workflow>` with prefix-based restore.
- **Compiler check**: Missing `compiler_check=content` setting, so MSVC path/version changes between runners invalidated all cache entries. Added `compiler_check=content` (same fix as macOS in https://github.com/facebook/rocksdb/issues/14478).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14508
Test Plan:
- Release and debug builds compile cleanly
- 70 consecutive shuffled runs of db_compaction_test (367 tests each) pass — 50 on full cores + 20 on 4 cores
- Format check passes
- Windows ccache fix requires CI run to verify (first run populates cache, second run should see high hit rate)
Reviewed By: anand1976
Differential Revision: D98380757
Pulled By: xingbowang
fbshipit-source-id: d74079b75786ba3299e335b145a6c5fdc81fd5c1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14510
When a filesystem does not return kAsyncIO in SupportedOps(), MultiScan
still sends ReadAsync/Poll/AbortIO requests. Regular iterators check via
CheckFSFeatureSupport in ArenaWrappedDBIter::Init and ForwardIterator,
but MultiScan blindly trusted the caller-provided
MultiScanArgs::use_async_io flag.
Fix: In the MultiScan constructor, after scan_opts_ is initialized,
check CheckFSFeatureSupport and disable use_async_io if the FS doesn't
support it. Also pass the (potentially modified) scan_opts_ to Prepare()
instead of the original scan_opts parameter.
Reviewed By: mszeszko-meta
Differential Revision: D97995735
fbshipit-source-id: 331639d950fd3cb9f491feed996d5820294beb4e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14511
Add `atomic_flush` option to `CreateBackupOptions` that plumbs through
`CheckpointImpl::CreateCustomCheckpoint` to `LiveFilesStorageInfoOptions`.
When `flush_before_backup=true` and `atomic_flush=true`, the backup engine
atomically flushes all column families before creating the backup. This
ensures cross-CF consistency without needing WAL files. Combined with
`BackupEngineOptions::backup_log_files=false`, this allows safely skipping
WAL backup for multi-CF databases, reducing backup size and complexity.
Changes:
- `backup_engine.h`: Add `bool atomic_flush = false` to `CreateBackupOptions`
- `checkpoint_impl.h/.cc`: Add `bool atomic_flush` parameter to
`CreateCustomCheckpoint`, plumb to `LiveFilesStorageInfoOptions::atomic_flush`
- `backup_engine.cc`: Pass `options.atomic_flush` through to
`CreateCustomCheckpoint`
- `checkpoint_test.cc`: Add `BackupWithAtomicFlushSkipsWAL` test verifying
backup with atomic flush + no WAL creates a valid restorable multi-CF backup
Reviewed By: xingbowang
Differential Revision: D98208696
fbshipit-source-id: e818ba8669ac52a206e30e9dd56f1d7573cc0175
Summary:
Improves the Claude Code review CI workflow to produce deeper, more reliable reviews.
**Motivation:** The previous 30-turn limit caused reviews to silently produce "no output" on complex PRs (e.g. https://github.com/facebook/rocksdb/issues/14477). The review prompt was also too shallow — single-pass with no codebase context phase.
**Changes:**
**1. Comprehensive multi-agent review prompt** (`claude_md/ci_review_prompt.md`)
- 9 specialized review agents: design, correctness, cross-component, invariant-adversary, caller-audit, performance, API, serialization, test coverage
- Deep codebase context phase before agents spawn: caller-chain analysis (3-5 levels up), callee side-effect tracing, cross-component data consumer analysis, execution context verification, assumption stress-testing
- Inter-agent debate with round-robin critique assignments
- Final report quality rules: disproven findings removed, no stream-of-consciousness
**2. Incremental findings + recovery flow**
- `Write` tool added so Claude saves findings to `review-findings.md` after each phase
- If the review hits the turn limit, a recovery step launches a Sonnet session to format partial findings into the standard output
- Recovery file existence check prevents crash if recovery step fails
- `getLastAssistantText` fallback truncated to 50KB to avoid enormous PR comments
**3. Prompts extracted to files** (`claude_md/ci_*.md`)
- `ci_review_prompt.md` — full review methodology
- `ci_query_prompt.md` — `/claude-query` system prompt
- `ci_recovery_prompt.md` — recovery formatting prompt
- Secure: checkout is from base branch (main), not PR head
**4. `max_turns` increased from 30 to 300**
- Orchestrator budget for multi-agent workflow; sub-agents get their own turns
- Recovery flow ensures partial results if limit is still hit
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14507
Reviewed By: archang19
Differential Revision: D98170111
Pulled By: xingbowang
fbshipit-source-id: 390626a53e7a7f91c2d3e91ed4403494532425ed
Summary:
Add `SstFileReader::Get` (single-key) and `SstFileReader::MultiGet` (PinnableSlice) overloads to enable zero-copy point lookups directly from SST files. The existing `MultiGet(std::string*)` is refactored to delegate to the new `MultiGet(PinnableSlice*)`, which writes results directly into caller-provided `PinnableSlice` values instead of copying through an intermediate buffer. The single-key `Get` uses `TableReader::Get` with a `GetContext` for efficient single-key lookups without the overhead of MultiGet's sorting and batching machinery.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14500
Test Plan: - New unit tests
Reviewed By: xingbowang
Differential Revision: D97825648
Pulled By: joshkang97
fbshipit-source-id: 17f3edd59bbf4747d17309c44ef12f0d952ea4eb
Summary:
Add `force_atomic_flush` to `FlushOptions` so any flush caller (Checkpoint, `DB::Flush()`, etc.) can force atomic flush of all column families even when `DBOptions::atomic_flush` is disabled.
## Changes
### Public API
- Added `bool force_atomic_flush = false` to `FlushOptions`
- Added `bool atomic_flush = false` to `LiveFilesStorageInfoOptions`
### Flush dispatch
- `FlushAllColumnFamilies()` — uses `immutable_db_options_.atomic_flush || flush_options.force_atomic_flush`
- `DB::Flush()` (single-CF and multi-CF) — respects `force_atomic_flush`
- `FlushForGetLiveFiles()` — constructs `FlushOptions` with `force_atomic_flush` from `LiveFilesStorageInfoOptions::atomic_flush`
### Per-request atomic flush in pipeline
- `FlushRequest.atomic_flush` and `BGFlushArg.atomic_flush_` carry the per-request flag
- `GenerateFlushRequest`, `EnqueuePendingFlush`, `PopFirstFromFlushQueue`, `BackgroundFlush`, `FlushMemTablesToOutputFiles` all use the per-request flag
### Stress test
- Added `--checkpoint_atomic_flush` flag to db_stress, randomly enabled in db_crashtest.py
### Unit tests (6 new)
- End-to-end checkpoint with atomic flush override
- Negative test (default = no atomic flush)
- Single-CF edge case
- Interaction with `DBOptions::atomic_flush=true`
- Mixed non-atomic then atomic flushes in queue
- Mixed atomic then non-atomic flushes in queue
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14477
Reviewed By: joshkang97
Differential Revision: D97639233
Pulled By: xingbowang
fbshipit-source-id: b924c044d6179e68b38c6604eb46c9140516d42a
Summary:
FIFO compaction drops old SST files when total size exceeds `max_table_files_size`
or `max_data_files_size`. The stress test's expected state can't track these drops,
causing false "GetEntity returns NotFound" failures. Confirmed by
`rocksdb.fifo.max.size.compactions COUNT: 7` in an asan_crash_test whitebox run
(`compaction_style=2`, `fifo_compaction_max_data_files_size_mb=100`).
This diff adds a `fifo_compaction_max_table_files_size_mb` flag and sets it to
100GB in db_crashtest.py. Since `max_table_files_size` is always active (default
1GB), it must always be set very high. `fifo_compaction_max_data_files_size_mb`
is randomized between 0 (disabled, defers to `max_table_files_size`) and 100GB
(overrides `max_table_files_size`) to exercise both code paths while preventing
drops in either case. FIFO intra-L0 compaction (`fifo_allow_compaction`) is
still exercised.
Long-term TODO: handle FIFO drops in expected state via `OnCompactionBegin`
listener calling `SetPendingDel()` on affected key ranges, treating drops as
concurrent deletes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14503
Reviewed By: joshkang97
Differential Revision: D97601382
Pulled By: hx235
fbshipit-source-id: 1a3badfa4de47efd73f965807982a966ebe135ef
Summary:
Fix thread-safety issues in memtable stats tracking and `MaybeUpdateNewestUDT` to prepare for PR https://github.com/facebook/rocksdb/issues/14448, which introduces concurrent range tombstone insertion from the read path into the mutable memtable. Currently the non-concurrent write path updates memtable counters (`num_entries_`, `num_deletes_`, `num_range_deletes_`, `data_size_`) using non-atomic load/store pairs. While this is safe today PR https://github.com/facebook/rocksdb/issues/14448 will make range_tombstone memtable concurrent, but the main write path can still be non-concurrent. This fix ensures stats are tracked correctly.
Similarly, `MaybeUpdateNewestUDT` was not thread-safe and was not called in the concurrent write path at all. This PR also fixes the `post_process_info` delete counter which missed `kTypeSingleDeletion` and `kTypeDeletionWithTimestamp`.
## Key changes
- Switch non-concurrent counter updates from `LoadRelaxed`/`StoreRelaxed` pairs to `FetchAddRelaxed` to avoid races with concurrent range tombstone inserts (e.g. `AddLogicallyRedundantRangeTombstone` calling `BatchPostProcess`).
- Make `MaybeUpdateNewestUDT` thread-safe by replacing `Slice newest_udt_` with `RelaxedAtomic<const char*> newest_udt_data_` and using a CAS loop. The pointed-to memory lives in the arena and remains valid for the memtable's lifetime.
- Call `MaybeUpdateNewestUDT` in the concurrent write path (was previously skipped with a TODO).
- Fix `post_process_info->num_deletes++` to also count `kTypeSingleDeletion` and `kTypeDeletionWithTimestamp`, matching the non-concurrent path.
- Change `GetNewestUDT()` return type from `const Slice&` to `Slice` (returning by value) to avoid dangling reference issues with the new atomic pointer storage. Updated across `MemTable`, `ReadOnlyMemTable`, `MemTableListVersion`, and `WBWIMemTable`.
- Remove unused `Slice newest_udt_` member from `WBWIMemTable`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14506
Test Plan:
- Added `ConcurrentWriteMemTableProperties` test in `db_properties_test.cc`: 4 threads writing puts, deletes, and single deletes concurrently, then verifying `rocksdb.num-entries-active-mem-table` and `rocksdb.num-deletes-active-mem-table` properties are correct. Flush with `flush_verify_memtable_count=true` validates integrity.
- Added `ConcurrentGetTableNewestUDT` test in `memtable_list_test.cc`: 4 threads concurrently inserting entries with UDTs via `allow_concurrent=true`, verifying the newest UDT is correctly tracked.
- Extended `DuplicateSeq` and `ConcurrentMergeWrite` tests in `db_memtable_test.cc` to verify stat counters after `BatchPostProcess`.
- Benchmark results show no performance regression (3 runs each, averaged):
**Workload 1: fillrandom** (pure Put workload)
```
./db_bench -benchmarks=fillrandom -seed=1 -compression_type=none -threads=8 -db=<DB>
```
**Workload 2: readrandomwriterandom** (mixed read/write/delete)
```
# Setup: fillrandom,compact to populate DB (single-threaded)
./db_bench -benchmarks=readrandomwriterandom -seed=1 -compression_type=none \
-use_existing_db=1 -readwritepercent=50 -deletepercent=20 -threads=8 -db=<DB>
```
**Workload 3: fillrandom with range tombstones** (puts interleaved with range deletes)
```
./db_bench -benchmarks=fillrandom -seed=1 -compression_type=none \
-writes_per_range_tombstone=100 -range_tombstone_width=10 \
-max_num_range_tombstones=1000 -threads=8 -db=<DB>
```
```
| Benchmark | avg ops/s (main) | avg ops/s (feature) | % change |
|------------------------------------|-----------------|--------------------:|----------|
| fillrandom | 531,986 | 532,099 | +0.02% |
| readrandomwriterandom (50r/30w/20d)| 786,400 | 815,143 | +3.65% |
| fillrandom (range tombstones) | 541,656 | 531,008 | -1.97% |
```
Reviewed By: xingbowang
Differential Revision: D97974456
Pulled By: joshkang97
fbshipit-source-id: ea95f23953fe37771a599aed61ece1a504b12c2e
Summary:
Reduce `make check` time from ~9.4 minutes to ~3.7 minutes (2.6x) on a 192-core machine, speed up CI by 1.8x across all jobs, and fix ccache hit rates (macOS cmake: 0.51% → 99.80%).
## Changes
### 1. Auto-detect lld linker (`build_detect_platform`)
- Probe for `lld` at configure time; fall back to default `ld.bfd` if unavailable
- Linux-only guard (`TARGET_OS = Linux`)
- **Measured: 12x faster linking** (7.4s → 0.6s per test binary)
- Add `-L/usr/local/lib` for lld library resolution on CI
- Install lld in CI pre-steps action
- Opt out with `ROCKSDB_NO_FAST_LINKER=1`
### 2. Sharded test execution (`Makefile`)
**Before:** `make check` enumerated every individual gtest case via `--gtest_list_tests`, then created a separate shell script for each case with `--gtest_filter=TestName`. For example, `block_based_table_reader_test` has 9,788 parameterized test cases — each spawned its own process, loading a ~50MB binary, initializing gtest, registering all ~10K tests, running exactly ONE, then tearing down. The per-process overhead was ~1.5s, so 9,788 × 1.5s ≈ 15,000s wasted on overhead alone. Across all 302 test binaries this produced 39,467 individual processes.
**After:** Use gtest's built-in sharding (`GTEST_TOTAL_SHARDS`/`GTEST_SHARD_INDEX`) to group ~10 test cases per process. Each shard loads the binary once and runs multiple tests sequentially — eliminating the per-process overhead. The 9,788 cases in `block_based_table_reader_test` become ~980 shards. Total process count drops from 39,467 to ~4,036. Same tests, same coverage, just fewer process spawns.
The shard count adapts to machine size: `min(ceil(test_count / GTEST_SHARD_SIZE), NCORES * 8)`. This ensures large machines (192 cores) get many small shards for parallelism, while small CI runners (4 cores) get fewer larger shards to avoid excessive overhead. Both `GTEST_SHARD_SIZE` (default 10) and `NCORES` can be overridden.
CI sharding uses round-robin distribution across 3 shards to balance heavy tests (db_test, db_compaction_test, etc.) instead of contiguous alphabetical ranges.
Note: gtest continues running all tests after a failure (`GTEST_THROW_ON_FAILURE=0`), so grouping tests does not mask failures — all failures within a shard are reported.
**Measured: 3.6x CPU reduction** (76,121s → 21,033s)
### 3. Fix SyncPoint leaks for test isolation (5 files)
Sharded execution exposed 43 test files that set SyncPoint callbacks but never clean them up. Stale callbacks with captured local variables cause segfaults or data corruption when subsequent tests in the same process trigger them.
**Systematic fix:** Global gtest `TestEventListener` in `testharness.cc` that calls `SyncPoint::DisableProcessing()` + `ClearAllCallBacks()` + `ClearTrace()` + `LoadDependency({})` after every test case. This cleans up all SyncPoint state: callbacks, dependency maps, cleared points, and the point filter. Registered via static initialization — no changes needed to individual test `main()` functions.
**SyncPoint infrastructure fixes:**
- `DisableProcessing()` now calls `cv_.notify_all()` to wake threads blocked in `Process()`. Previously, threads waiting for predecessor sync points that would never fire (because the test ended) would hang forever.
- `Process()` now rechecks `enabled_` after waking from `cv_.wait()`, so threads exit promptly when processing is disabled instead of looping forever.
**Specific fixes (defense-in-depth):**
- `SstFileReaderTest::VerifyNumEntriesCorruption` — leaked `PropertyBlockBuilder::AddTableProperty:Start` callback that corrupted SST entry counts
- `WritePreparedTransactionTest::CommitAndSnapshotDuringCompaction` — leaked `CompactionIterator:AfterInit` callback causing segfault via dangling pointers
- `TransactionTestBase` destructor — cleanup for 26 SyncPoint uses across transaction tests
- `RetriableLogTest` destructor — cleanup for log reader SyncPoint callbacks
### 4. Fix ccache for reliable cross-run caching (`setup-ccache`, `CMakeLists.txt`)
- **`CCACHE_COMPILERCHECK=content`**: Default `mtime` compared compiler binary modification time, which differs on every fresh CI runner → 0% hit rate. `content` hashes compiler output instead — stable across runner instances
- **`CCACHE_SLOPPINESS`**: CMake generates varying `-MF` paths and timestamps. Without sloppiness settings, ccache treated these as different compilations
- **`CMAKE_C/CXX_COMPILER_LAUNCHER`** instead of `RULE_LAUNCH_COMPILE`: Avoids double-wrapping ccache when also injected via PATH. Removes `RULE_LAUNCH_LINK` since ccache cannot cache link operations
- **macOS cmake hit rate: 0.51% → 99.80%**
### 5. Align GTEST_THROW_ON_FAILURE (`Makefile`)
- Set `GTEST_THROW_ON_FAILURE=0` in Makefile default to match CI's `pre-steps/action.yml`
- `=1` caused `std::terminate` in multi-threaded stress tests (e.g., `point_lock_manager_stress_test`, `rate_limiter_test`) when assertions fail — the gtest exception propagates across thread boundaries, triggering undefined behavior and heap-use-after-free under ASAN
### 6. Portability fixes
- `[[maybe_unused]]` instead of `__attribute__((unused))` for MSVC compatibility
- Remove blanket `-fPIC` from `COMMON_FLAGS` — only apply via `PLATFORM_SHARED_CFLAGS` for shared builds, preserving optimal codegen for release/static builds
- `make -s list_all_tests` to suppress make noise in test enumeration
## Results
### Local (192-core devvm, clean build)
| Metric | Before | After | Improvement |
|--------|--------|-------|-------------|
| Build time (clean) | 2m43s | ~2m | lld linker |
| Test wall clock | 400s | 200s | 2.0x |
| Test CPU total | 76,121s | 21,033s | 3.6x |
| Test failures | 0 | 0 | — |
| **Total make check** | **~9.4min** | **3m41s** | **2.6x** |
### CI (GitHub Actions)
| Metric | Before | After | Improvement |
|--------|--------|-------|-------------|
| Aggregate CI time | 15,220s | 8,425s | 1.8x |
| build-linux-arm | 763s | 221s | 3.5x |
| build-linux-release | 616s | 244s | 2.5x |
| build-windows-vs2022 | 2,868s | 1,205s | 2.4x |
| build-linux-java-static | 858s | 335s | 2.6x |
### ccache hit rates (all jobs)
| Job | Before | After |
|-----|--------|-------|
| macOS cmake (0-3) | 0.51% | **99.80%** |
| build-linux | 99.68% | **99.84%** |
| build-linux-clang18-asan-ubsan | 98.58% | **99.84%** |
| build-linux-clang18-mini-tsan | 98.74% | **99.84%** |
| build-windows-vs2022 | 99.22% | **99.83%** |
| All 27 ccache jobs | — | **>94%** |
## Remaining bottleneck
The critical path is now dominated by genuinely slow integration tests:
- `external_sst_file_test` shards: 97–113s each
- `db_test` shards: 109–124s each
- `db_bloom_filter_test`: 103s (non-parallel)
Further improvement would require test optimization or a `make check-fast` target.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14478
Test Plan:
- `make check -j192 SKIP_FORMAT_BUCK_CHECKS=1` — all 4,045 shards pass (ASAN+UBSAN), zero failures
- Verified SyncPoint fixes: each crash/corruption reproduces before fix, passes after
- Verified SyncPoint::DisableProcessing wakes blocked threads (previously caused indefinite hangs)
- Verified gtest does NOT stop after failure with `GTEST_THROW_ON_FAILURE=0` — all tests in a shard run and all failures are reported
- CI: all 34 checks pass across two consecutive runs
- ccache: validated hit rates >94% on second run after cache seeding
Reviewed By: hx235
Differential Revision: D97623305
Pulled By: xingbowang
fbshipit-source-id: b299e6d43c8713a9ef2b5659e8bbd74958afe155
Summary:
The ExternalTableReader `Get` API has been modified to use PinnableSlice instead of std::string, this will allow implementations to utilize zero-copy Gets (e.g. reading from mmap or a cache). This is not a compatible change, but the API is marked as experimental, so should be allowed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14497
Test Plan: - New unit tests
Reviewed By: xingbowang
Differential Revision: D97782011
Pulled By: joshkang97
fbshipit-source-id: 8a9e8c5bc5ff5e8dee6c0f2ee745521f09042cef
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14502
The asan_crash_test_with_ts randomly selected use_trie_index=1 alongside
user_timestamp_size=8. TrieIndexFactory requires plain BytewiseComparator,
but user-defined timestamps wrap it as BytewiseComparator.u64ts, causing
Status::NotSupported errors during flush/compaction.
Fix:
1. Add use_trie_index=0 to ts_params in db_crashtest.py to prevent the
incompatible combination (matches existing pattern for other features).
2. Add early validation in db_stress_tool.cc to reject this combination
with a clear error message (defense in depth).
Reviewed By: xingbowang
Differential Revision: D97592648
fbshipit-source-id: 2259f82d95ea8410c4278b52f60983a4b5e84ec2
Summary:
Add GitHub Actions workflows for AI-powered code review on PRs using Claude (Anthropic). Follows the clang-tidy security pattern with two separate workflows for privilege separation.
## Trigger Modes
**1. Auto** — Runs after `pr-jobs` workflow completes successfully via `workflow_run`. Safe for fork PRs (runs from default branch, never executes PR code).
**2. Manual** — Maintainers comment `/claude-review [focus area]` or `/claude-query <question>` on any PR. Restricted to 15 authorized team members.
**3. workflow_dispatch** — For manual testing.
## Security Model (Two-Workflow Separation)
Same pattern as clang-tidy:
**`claude-review.yml`** (analysis):
- Runs Claude with `ANTHROPIC_API_KEY`
- Has ONLY `contents: read` — no PR write, no issue write
- Saves review markdown + metadata as artifact
**`claude-review-comment.yml`** (posting):
- Triggers on `workflow_run` completion
- Downloads artifact and posts/updates PR comment
- Has `pull-requests: write` but never runs AI
This separation prevents a crafted PR from tricking Claude into exfiltrating write tokens.
## Review Methodology
Review prompt in `claude_md/code_review.md` (shared with local Claude Code reviews). Five perspectives:
- Call-chain analysis (3-5 levels up/down)
- Correctness & edge cases
- Cross-component & adversarial (10 execution contexts)
- Performance
- API compatibility & test coverage
## Shared Scripts
- `.github/scripts/post-pr-comment.js` — Create-or-update PR comment with marker-based dedup. Now used by both clang-tidy and Claude review.
- `.github/scripts/parse-claude-review.js` — Parses `claude-code-base-action` execution log into markdown.
## Files Changed
| File | Description |
|------|-------------|
| `.github/workflows/claude-review.yml` | Analysis workflow (476 lines) |
| `.github/workflows/claude-review-comment.yml` | Comment posting workflow (146 lines) |
| `.github/scripts/post-pr-comment.js` | Shared PR comment utility (57 lines) |
| `.github/scripts/parse-claude-review.js` | Execution log parser (78 lines) |
| `.github/workflows/clang-tidy-comment.yml` | Updated to use shared script |
| `claude_md/code_review.md` | Review methodology (104 lines) |
## Setup Required
Add `ANTHROPIC_API_KEY` secret to the repo settings.
## Testing
Tested end-to-end on `xingbowang/rocksdb` fork — both auto and manual triggers, artifact upload/download, comment posting, and duplicate detection all verified working.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14480
Reviewed By: omkarhgawde
Differential Revision: D97832666
Pulled By: xingbowang
fbshipit-source-id: f80c7d8683ac980614dc4ca66c1e545deb3be504
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14475
GetLiveFiles was previously blocked on secondary instances with
Status::NotSupported, even though the operation is safe to perform.
The only reason GetLiveFiles was originally blocked is that it defaults
to flushing the memtable (flush_memtable=true), which is a write
operation. However, the actual implementation in DBImpl::GetLiveFiles
(db_filesnapshot.cc) does two things:
1. Optionally flush the memtable — which we skip by passing
flush_memtable=false. The secondary already overrides
FlushForGetLiveFiles() as a no-op, so even if true were passed
it would not actually flush.
2. Read live file state from versions_ under the mutex — this is
purely read-only. It iterates the ColumnFamilySet to collect live
table and blob file numbers, builds relative file paths for SST
files, blob files, CURRENT, MANIFEST, and OPTIONS, and reads the
manifest file size.
The secondary maintains its own VersionSet which it keeps up to date
via MANIFEST replay in TryCatchUpWithPrimary(). So all of this state
is valid and accurate — it reflects exactly which files the secondary
considers live at its current replay point.
This is the same approach used by DBImplReadOnly and CompactedDBImpl,
which both delegate to DBImpl::GetLiveFiles with flush_memtable=false.
Reviewed By: xingbowang
Differential Revision: D97563143
fbshipit-source-id: 8d5b52e26a478ef190eba598819de6527817bcfc
Summary:
Fix leaked table cache entries that cause `TEST_VerifyNoObsoleteFilesCached` assertion failure during `DB::Close()` in ASAN crash test builds (T258745630):
```
File 126519 is not live nor quarantined
Assertion `cached_file_is_live_or_quar' failed.
```
When a compaction fails *after* `VerifyOutputFiles` succeeds (e.g., at `VerifyCompactionRecordCounts`), the overall `compact_->status` is set to error but each subcompaction's individual `status` remains OK. `SubcompactionState::Cleanup` only checked the individual subcompaction status, so it skipped calling `ReleaseObsolete` on the output files' table cache entries — leaking them.
Normally, `Close()`'s backstop (`FindObsoleteFiles(force=true)` + `PurgeObsoleteFiles`) would catch this by finding the orphan file on disk, evicting the cache entry, and deleting the file. However, `FindObsoleteFiles` calls `FaultInjectionTestFS::GetChildren` which can fail under metadata read fault injection (`--open_metadata_read_fault_one_in=8`). The error is silently ignored (`s.PermitUncheckedError()` in db_impl_files.cc:199), so the orphan file is never found and the leaked cache entry is never evicted.
The `Cleanup` bug is latent and predates recent changes. PR https://github.com/facebook/rocksdb/issues/14433 added `verify_output_flags` randomization to the crash test, and PR https://github.com/facebook/rocksdb/issues/14456 fixed false-positive corruptions that https://github.com/facebook/rocksdb/issues/14433 caused. Before https://github.com/facebook/rocksdb/issues/14456, `VerifyOutputFiles` would produce false corruption errors that *accidentally prevented the leak* by setting the subcompaction status to non-OK.
### How it triggers
**Step 1 — VerifyOutputFiles adds cache entries for compaction output files:**
```
CompactionJob::Run()
RunSubcompactions() // all subcompactions succeed
// output file 12 written to disk
// each sub_compact.status = OK
SyncOutputDirectories() // status = OK
VerifyOutputFiles()
for each output_file:
table_cache()->NewIterator(output_file.meta)
FindTable()
cache->Insert(file_number=12) // <<< ENTRY ADDED TO TABLE CACHE
return iterator holding handle
delete iter // releases handle, entry stays in LRU
// status = OK
```
**Step 2 — A post-verification step fails, overall status set but NOT subcompaction status:**
```
CompactionJob::Run()
VerifyCompactionRecordCounts() // returns Status::Corruption(...)
FinalizeCompactionRun(status=Corruption)
compact_->status = Corruption // <<< OVERALL status = error
// each sub_compact.status is still OK!
```
**Step 3 — Install skips InstallCompactionResults (file 12 never enters a Version):**
```
CompactionJob::Install()
status = compact_->status // Corruption
if (status.ok()) // FALSE
InstallCompactionResults() // <<< SKIPPED — file 12 not in any version
```
**Step 4 — CleanupCompaction skips ReleaseObsolete (THE BUG):**
```
CompactionJob::CleanupCompaction()
for each sub_compact:
sub_compact.Cleanup(table_cache)
if (!status.ok()) // checks sub_compact.status = OK
// <<< FALSE — individual status is OK!
ReleaseObsolete(...) // <<< NEVER CALLED — cache entry leaked!
```
**Step 5 — Metadata read fault prevents backstop from finding the orphan:**
`FindObsoleteFiles(force=true)` calls `GetChildren()` to scan the DB directory.
`FaultInjectionTestFS::GetChildren` injects a metadata read error
(`--open_metadata_read_fault_one_in=8`). The error is silently ignored
(`s.PermitUncheckedError()`), so the directory listing is empty and the orphan
file is never found. This happens both in the post-compaction cleanup
(`BackgroundCallCompaction`) and in `Close()`'s backstop:
```
FindObsoleteFiles(force=true)
fs->GetChildren(path, ...) // returns IOError (fault injected)
s.PermitUncheckedError(); // error silently ignored
// files vector is empty — orphan file 12 not found
PurgeObsoleteFiles() // nothing to do — no Evict called
```
**Step 6 — Assertion fires during Close():**
```
CloseHelper()
FindObsoleteFiles(force=true) // GetChildren fails again → orphan missed
PurgeObsoleteFiles() // nothing to evict
TEST_VerifyNoObsoleteFilesCached()
for each cache entry:
file_number = 12
live_and_quar_files.find(12) == end() // NOT in any version!
>>> assert(cached_file_is_live_or_quar) FAILS <<<
```
**Crash test call stack:**
```
frame https://github.com/facebook/rocksdb/issues/9: __assert_fail_base("cached_file_is_live_or_quar", "db_impl_debug.cc", 389)
frame https://github.com/facebook/rocksdb/issues/11: DBImpl::TEST_VerifyNoObsoleteFilesCached()::lambda // finds leaked entry
frame https://github.com/facebook/rocksdb/issues/18: LRUCacheShard::ApplyToSomeEntries(...) // iterating cache shard
frame https://github.com/facebook/rocksdb/issues/19: ShardedCache::ApplyToAllEntries(...) // iterating all shards
frame https://github.com/facebook/rocksdb/issues/20: DBImpl::TEST_VerifyNoObsoleteFilesCached() // the verification
frame https://github.com/facebook/rocksdb/issues/21: DBImpl::CloseHelper() // during Close
frame https://github.com/facebook/rocksdb/issues/22: DBImpl::CloseImpl()
frame https://github.com/facebook/rocksdb/issues/23: DBImpl::Close()
frame https://github.com/facebook/rocksdb/issues/24: StressTest::Reopen() // crash test reopen
frame https://github.com/facebook/rocksdb/issues/25: StressTest::OperateDb() // worker thread
```
### Fix
Pass the overall `compact_->status` to `SubcompactionState::Cleanup` and call
`ReleaseObsolete` when *either* the subcompaction status or the overall status
is non-OK:
```cpp
// Before:
if (!status.ok()) { ReleaseObsolete(...); }
// After:
if (!status.ok() || !overall_status.ok()) { ReleaseObsolete(...); }
```
## Key changes
- **`SubcompactionState::Cleanup`**: Now takes an `overall_status` parameter and calls `ReleaseObsolete` when *either* the subcompaction status or the overall compaction status is non-OK.
- **`CompactionJob::CleanupCompaction`**: Passes `compact_->status` (the overall status) to each subcompaction's `Cleanup`.
- **Sync point**: Added `CompactionJob::Run():AfterVerifyOutputFiles` for error injection in tests.
- **Unit test**: `DBCompactionTest.LeakedTableCacheEntryOnCompactionFailure` uses `FaultInjectionTestFS` to reproduce the crash test scenario — injects error after `VerifyOutputFiles` and deactivates the filesystem so `GetChildren` fails in `FindObsoleteFiles`, preventing the backstop from evicting the leaked cache entry.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14469
Test Plan:
- `DBCompactionTest.LeakedTableCacheEntryOnCompactionFailure`:
- **Without fix (ASAN)**: assertion fires during `Close()` — `File 12 is not live nor quarantined`
- **With fix (ASAN)**: passes — `ReleaseObsolete` properly cleans up the cache entry
- **With fix (non-ASAN)**: passes
```
$ COMPILE_WITH_ASAN=1 make -j db_compaction_test
$ ./db_compaction_test --gtest_filter="DBCompactionTest.LeakedTableCacheEntryOnCompactionFailure"
[ PASSED ] 1 test.
```
Reviewed By: xingbowang
Differential Revision: D97190944
Pulled By: joshkang97
fbshipit-source-id: fdfd481cc1e192803cfb7d64052ccb9162c21b94
Summary:
**Context/Summary:**
Remove assertions on flush_running/flush_scheduled from the BackgroundJobPressure listener test. These checks are inherently racy because Flush() can return before the background thread fires the pressure callback.
The race: when BackgroundCallFlush re-acquires the DB mutex after cleanup (line 3661), COERCE_CONTEXT_SWITCH=1 or others may call bg_cv_->SignalAll() and sleep before actually acquiring the lock. This spurious signal wakes WaitForFlushMemTables(), which sees the flush is already installed and returns — so Flush() "completes" seen as foreground in test. The test then starts the next Put()+Flush(), scheduling new flush work. When the previous background thread finally wakes and captures the pressure snapshot, it sees the newly scheduled flush (flush_scheduled > 0). If the race happens on the last flush, its callback may not have fired at all by the time the test reads GetSnapshots(), so snapshots.back() can also show stale data.
The remaining assertions (compaction counts, speedup, write stall proximity) are not affected: compaction is blocked by sleeping_task in Phase 1-2, and Phase 3 uses TEST_WaitForCompact() whose wait condition (bg_*_scheduled_ counts) is only satisfied after the mutex is re-acquired and callbacks have fired.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14476
Test Plan:
COERCE_CONTEXT_SWITCH=1 make -j56 listener_test
Before: ~22/50 failures at line 1715 (flush_running != 0)
After: 50/50 pass
Reviewed By: xingbowang
Differential Revision: D97572545
Pulled By: hx235
fbshipit-source-id: e8a57a7a00e41d17bf47ef8e04cea977225a5909
Summary:
**Summary:**
Add a new EventListener callback `OnBackgroundJobPressureChanged` that fires after every flush or compaction background job completes. The callback delivers a `BackgroundJobPressure` snapshot containing:
- Compaction scheduling counters (scheduled/running, combined and per-priority LOW/BOTTOM breakdown)
- Flush scheduling counters (scheduled/running)
- Write stall proximity percentage (0=healthy, 100=at stall threshold, can exceed 100 when stalling)
- Whether compaction speedup is active
`CaptureBackgroundJobPressure()` reads scheduling counters and computes write stall proximity from L0 sorted run count and pending compaction bytes (same inputs as `RecalculateWriteStallConditions()`). TODO: add memory-related write stall triggers later.
Introduces `num_running_bottom_compactions_` counter to track BOTTOM- priority compactions separately from LOW, enabling per-pool breakdown in the pressure snapshot.
The callback fires on the background thread after counter decrements and `MaybeScheduleFlushOrCompaction()`, so the snapshot reflects post- completion state. Uses the same mutex unlock/lock pattern as `NotifyOnFlushCompleted`. A `bg_pressure_callback_in_progress_` counter ensures destructor safety since the callback fires after `bg_flush_scheduled_`/`bg_compaction_scheduled_` are decremented.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14474
Test Plan:
- listener_test BackgroundJobPressure: 3-phase test verifying no pressure, pressure build-up (speedup, scheduling, proximity), and pressure relief after compaction completes
- db_compaction_test CompactRangeBottomPri: verifies num_running_bottom_compactions_ via sync point
- db_stress_tool exercises the callback with RandomSleep()
Reviewed By: xingbowang
Differential Revision: D97423623
Pulled By: hx235
fbshipit-source-id: 07003c8de226ec29d32b8a88e2d86e5de85cd2cc
Summary:
Skip WAL recovery when opening a secondary DB instance in OpenAndCompact() for remote compaction. WAL replay is unnecessary in this flow since only LSM state from MANIFEST is needed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14462
Test Plan:
- make -j db_secondary_test && ./db_secondary_test — 35/35 passed
- make -j compaction_service_test && ./compaction_service_test — 43/43 passed (includes new SkipWALRecoveryInOpenAndCompact test)
- make -j options_settable_test && ./options_settable_test --gtest_filter="*DBOptionsAllFieldsSettable*" — 1/1 passed
- Removed temporary hack in stress test that disables WAL
Reviewed By: hx235
Differential Revision: D96788211
Pulled By: jaykorean
fbshipit-source-id: f91a2f861f2450ebc83423ed4c6f5b70da7d9e8b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14473
Add separate PerfContext byte counters for data, index, filter, compression-dictionary, and metadata block reads while preserving block_read_byte as the aggregate. Wire the new counters through the block fetch and multi-read data block paths, expose them via the C perfcontext API, and extend table tests to verify byte attribution and that the classified counters sum back to block_read_byte.
Reviewed By: xingbowang
Differential Revision: D97333746
fbshipit-source-id: 3411844d7fa9c76c9ff28af477b3a72a5d6e5d9b
Summary:
Add a new mutable DB option `verify_manifest_content_on_close` (default: false).
When enabled, on DB close the MANIFEST file is read back and all records are
validated (CRC checksums via log::Reader and logical content via
VersionEdit::DecodeFrom). If corruption is detected, a fresh MANIFEST is written
from in-memory state using the existing LogAndApply recovery path.
This complements the existing size validation in VersionSet::Close() with content
validation, reusing the same manifest reading pattern as VersionSet::Recover().
Implementation plan:
## Part 1: New DB Option — verify_manifest_content_on_close
- A new mutable bool DB option (default: false) that can be dynamically toggled
via SetDBOptions() at runtime, following the pattern of other mutable manifest
options like max_manifest_file_size.
- Propagation: SetDBOptions() -> DBImpl::mutable_db_options_ ->
versions_->UpdatedMutableDbOptions() -> VersionSet::verify_manifest_content_on_close_
## Part 2: Core Implementation — Content Validation in VersionSet::Close()
- Inserted after existing size check, before closed_ = true
- Opens manifest as SequentialFileReader, creates log::Reader with checksum=true
- Loops ReadRecord with WALRecoveryMode::kAbsoluteConsistency, decodes each
record as VersionEdit
- On corruption: fires OnIOError listeners, logs error, calls LogAndApply with
empty edit to trigger manifest rewrite from in-memory state
- If manifest can't be opened for reading: logs warning, doesn't fail close
## Part 3: Unit Tests (in version_set_test.cc)
- ManifestContentValidationOnClose_Clean: enable option, normal close, verify
no manifest rotation
- ManifestContentValidationOnClose_CorruptRecord: enable option, corrupt manifest
via SyncPoint, verify rotation occurs and DB reopens cleanly
- ManifestContentValidationOnClose_Disabled: default off, verify content
validation does not run
- ManifestContentValidationOnClose_SizeCheckFails: truncate manifest so size
check fails first, verify recovery via size-check path
## What Happens If a Corruption is Detected
If corruption was detected, four things happen:
1. **Notify listeners** — Fires `OnIOError` on all registered event listeners
(from db_options_->listeners) so monitoring/alerting systems can observe
the corruption event. Uses `FileOperationType::kVerify` to categorize it.
2. **Permit unchecked errors** — `PermitUncheckedError()` silences RocksDB's
debug-mode assertion that every `IOStatus` must be inspected. These statuses
are informational-only here; the real recovery is via `LogAndApply`.
3. **Log the error** — Writes a `ROCKS_LOG_ERROR` message with the filename
for operational visibility (grep-able in production logs).
4. **Rewrite the manifest via `LogAndApply`** — This is the actual recovery.
`LogAndApply` is called with an empty `VersionEdit` (no changes). Internally,
`LogAndApply` detects that the current `descriptor_log_` is null (it was
reset at line 5551, or by the previous `LogAndApply` in the size-check
path) and creates a brand-new MANIFEST file. It serializes the entire
current in-memory LSM state — all column families, all levels, all file
metadata, sequence numbers, etc. — into this new file. It then atomically
updates the `CURRENT` file pointer to reference the new MANIFEST.
This works because the in-memory state was built from the original manifest
during `DB::Open()` and has been kept fully up to date through all
subsequent operations (flushes, compactions, etc.) during the DB's lifetime.
The on-disk manifest is essentially a journal of changes; `LogAndApply`
with an empty edit produces a fresh, compacted snapshot of that state.
## Flow Diagram of Manifest Content Validation
VersionSet::Close()
│
├─ Close descriptor_log_ and check size
│ └─ Size mismatch? → LogAndApply (rewrite manifest)
│
├─ Content validation (if s.ok() && option enabled)
│ ├─ Open manifest for sequential reading
│ │ └─ Can't open? → WARN log, continue
│ │
│ ├─ For each record:
│ │ ├─ ReadRecord (CRC32 check, kAbsoluteConsistency)
│ │ └─ DecodeFrom (VersionEdit logical check)
│ │
│ └─ Corruption detected?
│ ├─ Notify OnIOError listeners
│ ├─ LOG_ERROR
│ └─ LogAndApply (rewrite manifest from in-memory state)
│
└─ closed_ = true; return s;
## How This Relates to the Existing Size Check
The existing size check (lines 5556-5582) and the new content validation are
complementary:
| Check | What it catches | How it checks |
|----------------|-----------------------------------------|----------------------------|
| Size check | Truncation, partial writes, extra bytes | Compare expected vs actual file size |
| Content check | Bit-rot, silent corruption, bad records | CRC32 + VersionEdit decode |
The size check catches gross corruption (file too short or too long). The
content check catches subtle corruption where the file is the right size but
individual bytes have been flipped (e.g., storage media bit-rot, buggy
filesystem, incomplete block write).
Both recovery paths use the same mechanism: `LogAndApply` with an empty
`VersionEdit` to rewrite the manifest from in-memory state.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14451
Reviewed By: xingbowang
Differential Revision: D96004906
Pulled By: dannyhchen
fbshipit-source-id: 0b0ecdada3a74e97d2cadbba2091b8b577f1d684
Summary:
InjectedErrorLog is a lock-free circular ring buffer designed to be safe to call from signal handlers (which cannot use locks). It has an intentional benign data race between Record() (called by worker threads) and PrintAll() (called by the main thread from a signal/termination handler). The code documents this trade-off in comments, but was missing TSAN suppression annotations.
Simply adding TSAN_SUPPRESSION (__attribute__((no_sanitize("thread")))) is insufficient because TSAN still intercepts libc functions like vsnprintf/snprintf -- accesses through these interceptors are still tracked even when the calling function is annotated.
The fix:
1. Add TSAN_SUPPRESSION to both Record() and PrintAll() to suppress direct field reads/writes in the function body.
2. Restructure both functions to use local stack buffers for vsnprintf/snprintf operations instead of operating directly on shared entry data. This avoids passing shared memory through TSAN-intercepted libc functions.
Also adds fault_injection_fs_test with a ConcurrentRecordAndPrintAll test that exercises the concurrent Record() + PrintAll() pattern and verifies no TSAN race is reported.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14467
Test Plan:
- fault_injection_fs_test passes under TSAN (buck2 test fbcode//mode/dbg-tsan)
- Reverted fix, re-ran: Fatal (6 TSAN warnings) -- round-trip confirmed
- fault_injection_fs_test passes under debug mode (no regression)
Reviewed By: mszeszko-meta
Differential Revision: D96948483
Pulled By: xingbowang
fbshipit-source-id: efdd5eafa12a5a82f973e40aa327901cc5f95033
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14464
SetBackupInfoFromBackupMeta passes file_ptr->filename (which may contain
directory components like "private/1/000008.log") directly to
ParseFileName. ParseFileName expects a bare filename, so it fails and
file_type stays at the default kTempFile for all files in file_details.
Fix by extracting the basename before calling ParseFileName.
Reviewed By: mszeszko-meta
Differential Revision: D96793040
fbshipit-source-id: 7bb6b633eb07bb7ebda06edbc039a96b9b77b410
Summary:
The callback loop in InlineSkipList::MultiGet updated finger.prev_[0] as it walked forward through entries (e.g., merge operands). When the MultiGet batch contained duplicate user keys, the next lookup for the same key would find finger.prev_[0] pointing to an entry that sorts AFTER the lookup key in internal key order (because the lookup key has a high sequence number which sorts first), violating the FindSpliceForLevel precondition: before == head_ || KeyIsAfterNode(key, before).
Fix: stop updating finger.prev_[0] in the callback loop. Only finger.next_[0] needs advancing to track the walk-forward position. The prev_[0] from FindGreaterOrEqualWithFinger is always a valid lower bound for any subsequent key, whether it uses kMaxSequenceNumber or a snapshot sequence number.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14465
Reviewed By: xingbowang
Differential Revision: D96882549
Pulled By: anand1976
fbshipit-source-id: a733fa9d4f23f8b55a027c257a114c4cf35abe2b
Summary:
write_prepared_transaction_test_seqno keeps showing up in git changes
test binaries need to end with _test so they are ignored by .gitignore
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14453
Reviewed By: xingbowang
Differential Revision: D96202671
Pulled By: joshkang97
fbshipit-source-id: 91642226bedde37ca72c6af03f300990dca8ff3a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14458
In `CheckpointImpl::ExportColumnFamily()`, the assignment
`*metadata = result_metadata` is inside the `for` loop over
`db_metadata.levels`. If the column family has no levels, the loop body
never executes, so the `new ExportImportFilesMetaData()` is leaked and
the caller receives a nullptr despite a success status.
Fix: Move `*metadata = result_metadata` outside the loop.
Reviewed By: anand1976
Differential Revision: D95303457
fbshipit-source-id: 6a9be47bcca257803969eb3daac7b91e95143ebf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14443
In `PosixDirectory::FsyncWithDirOptions()`, when handling btrfs file rename
syncing, if `open()` fails and `fd` is -1, the code unconditionally calls
`close(fd)`. Calling `close(-1)` is undefined behavior per POSIX (returns
EBADF on Linux), and overwrites the original meaningful open error with a
misleading "While closing file after fsync" error message.
Fix: Guard the `close()` call with `fd >= 0`.
Reviewed By: xingbowang
Differential Revision: D95303407
fbshipit-source-id: 9b64b45a09e6ba41d87d164a1c094b4d22f7c186
Summary:
Fix a bug where `VerifyOutputFiles()` produces false positive "Key-value checksum of compaction output doesn't match what was computed when written" errors when `verify_output_flags` includes `kVerifyIteration` but `paranoid_file_checks` is false.
The root cause is a hash enable flag mismatch between writing and verification:
- During compaction writing (`OpenCompactionOutputFile`), the `OutputValidator` hash computation was gated solely by `paranoid_file_checks_`. When false, `enable_hash=false` and the hash stays at 0.
- During verification (`VerifyOutputFiles`), a new `OutputValidator` is always created with `enable_hash=true`, computing a non-zero hash.
- `CompareValidator()` then compares 0 vs non-zero, producing a false positive corruption.
This was exposed by the crash test randomization of `verify_output_flags` added in https://github.com/facebook/rocksdb/issues/14433. Before that change, `verify_output_flags` was always 0, so `kVerifyIteration` was only exercised via `paranoid_file_checks=true` (which correctly enabled the hash during writing).
The fix ensures hash computation is enabled during writing whenever either `paranoid_file_checks_` is true OR `verify_output_flags` includes `kVerifyIteration`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14456
Test Plan:
- Added `DBCompactionTest.VerifyIterationWithoutParanoidFileChecks`
- Added `DBCompactionTest.VerifyAllOutputFlagsWithoutParanoidFileChecks`
- Round-trip verified: tests FAIL without fix, PASS with fix
Reviewed By: jaykorean
Differential Revision: D96371769
Pulled By: xingbowang
fbshipit-source-id: 2f7406327496c7b541e4fa2668894df89eb813e8
Summary:
One of the follow ups from https://github.com/facebook/rocksdb/pull/14103. Users will have the option to verify file checksums for all compaction output files before they are installed. This feature helps prevent corrupted SST files from being added to the LSM tree.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14433
Test Plan: Unit test added
Reviewed By: archang19
Differential Revision: D95648000
Pulled By: jaykorean
fbshipit-source-id: 512f3c1a7449b96a660865531f3537624c89a9cc
Summary:
Add a circular ring buffer (InjectedErrorLog) to FaultInjectionTestFS that records the last 1000 injected errors. Each entry captures the timestamp, thread ID, FS API name with all arguments (file path, offset, buffer size, first 8 bytes of data in hex), and the injected error status. For example:
Append("/path/035354.sst", size=4096, head=[1a 2b 3c ...]) -> IOError (Retryable): injected write error
RenameFile("/path/tmp.sst", "/path/035355.sst") -> IOError: injected metadata write error
Read(offset=16384, size=4096) -> IOError (Retryable): injected read error
The ring buffer is printed to a file automatically:
- On any fatal signal (SIGABRT, SIGSEGV, SIGTERM, SIGINT, SIGHUP, SIGFPE, SIGBUS, SIGILL, SIGQUIT, SIGXCPU, SIGXFSZ, SIGSYS) via a registered crash callback
- At the end of db_stress main(), for diagnostic visibility even when the test completes normally
This addresses a key debugging gap: when write fault injection causes secondary failures (e.g., the builder error propagation issue in T257612259), the injected errors were previously completely silent with no logging trail. The ring buffer provides the missing diagnostic context to correlate fault injection with downstream failures.
Changes:
- port/stack_trace.h/.cc: Add RegisterCrashCallback() API; extend InstallStackTraceHandler() to catch all catchable termination signals
- utilities/fault_injection_fs.h: Add InjectedErrorLog class with printf-style Record(), HexHead() for data bytes, and signal-safe PrintAll()
- utilities/fault_injection_fs.cc: Record full API arguments and error status at all 31 fault injection call sites
- db_stress_tool/db_stress_tool.cc: Register crash callback and print ring buffer at end of main()
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14431
Reviewed By: hx235
Differential Revision: D95435430
Pulled By: xingbowang
fbshipit-source-id: 6c18e1b072044575d6c8c3f198070127b0f80608
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14447
`rocksdb_create_column_family()` and
`rocksdb_transactiondb_create_column_family()` allocate a
`rocksdb_column_family_handle_t` but always return it even when
`CreateColumnFamily()` fails. This leaks the handle and returns an object
with an indeterminate `rep` pointer. Other similar functions like
`rocksdb_create_column_family_with_import()` correctly delete the handle
and return nullptr on error.
Fix: Initialize `handle->rep = nullptr`, check `SaveError()` return value,
and on error delete the handle and return nullptr.
Reviewed By: anand1976
Differential Revision: D95303444
fbshipit-source-id: 5fde7a3ed588794d78429d5cb3d9f621f0fb6388
Summary:
When RocksDB operates with tiered or remote storage (e.g., Warm Storage, HDFS, S3), reading recently compacted data incurs high-latency remote reads because compaction output files are not present in the block cache. The existing `prepopulate_block_cache = kFlushOnly` avoids this for flush output but leaves compaction output cold until first access.
Add a new `PrepopulateBlockCache::kFlushAndCompaction` enum value that warms all block types (data, index, filter, compression dict) into the block cache during both flush and compaction. Flush-warmed blocks use `LOW` priority (unchanged from kFlushOnly behavior), while compaction-warmed blocks use `BOTTOM` priority — compaction data is less temporally local than freshly flushed data, so it should be the first to be evicted when the cache is full. This gives the remote-read avoidance benefit without risking cache thrashing.
The enum uses `kFlushAndCompaction` rather than separate `kCompactionOnly` + `kFlushAndCompaction` values because there is no practical use case for warming compaction output without also warming flush output. Flush output is by definition the hottest data (just written by the user), so if a workload benefits from warming the colder compaction output, it would always benefit from warming flush output too.
The implementation reuses the existing `InsertBlockInCacheHelper` / `WarmInCache` infrastructure in `BlockBasedTableBuilder`. The only internal change is adding a `warm_cache_priority` field to `Rep` alongside the existing `warm_cache` bool, and plumbing it through to the `WarmInCache` call instead of the previously hardcoded `Cache::Priority::LOW`.
### Key changes
- New `PrepopulateBlockCache::kFlushAndCompaction` enum value in table.h
- `Rep::warm_cache_priority` field in BlockBasedTableBuilder for per-reason priority control
- Serialization support ("kFlushAndCompaction" in string map)
- db_bench support (--prepopulate_block_cache=2)
- Crash test coverage (random choice includes new value)
**NOTE:** Unlike flush output (which is inherently hot — just written by the user), it is hard to distinguish hot from cold blocks in compaction output. Warming all compaction output therefore risks polluting the block cache and evicting genuinely hot entries. The kFlushAndCompaction mode is recommended only for use cases where most or all of the database is expected to reside in cache (e.g., the working set fits in cache). For workloads where only a fraction of the data is hot, kFlushOnly remains the safer choice.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14445
Test Plan:
- New `WarmCacheWithDataBlocksDuringCompaction` test: verifies data blocks from compaction output are present in the block cache and served without misses
- Extended `DynamicOptions` test: verifies dynamic switching through kDisable -> kFlushAndCompaction -> kFlushOnly -> kDisable via SetOptions
- Existing `WarmCacheWithDataBlocksDuringFlush` and parameterized `WarmCacheWithBlocksDuringFlush` tests continue to pass (kFlushOnly behavior unchanged)
- db_block_cache_test: 81/81 passed
- options_test: 74/74 passed
- table_test: 6910/6910 passed
Reviewed By: xingbowang
Differential Revision: D95997952
Pulled By: mszeszko-meta
fbshipit-source-id: 4ad568264992532053947df298e63e343821ddb5
Summary:
The trie UDI sanitization in db_crashtest.py disables use_txn because TransactionDB ROLLBACK writes DELETE entries that violate UDI's Put-only restriction. However, it was not clearing use_optimistic_txn or test_multi_ops_txns, which both require use_txn to be true.
Since use_trie_index is randomly enabled with 1/8 probability, the optimistic_txn and multiops_txn crash tests would intermittently fail with:
- "You cannot set use_optimistic_txn true while use_txn is false"
- "-use_txn must be true if -test_multi_ops_txns"
Fix by also setting use_optimistic_txn=0 and test_multi_ops_txns=0 in the trie UDI sanitization block alongside the existing use_txn=0.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14446
Test Plan: Watch crash test CI results
Reviewed By: xingbowang
Differential Revision: D95986370
Pulled By: pdillinger
fbshipit-source-id: 0751dcb4e99425ba9eb9aa34c2a99efa8eb3a194
Summary:
The function handles more than just lock timeouts — it also covers TryAgain from optimistic transactions. Rename it and update the comment to reflect its broader scope.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14441
Reviewed By: hx235
Differential Revision: D95877799
Pulled By: xingbowang
fbshipit-source-id: 0651ffd39ac9d3a7979750be6dfe5b293eab7a64
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14442
In `BackgroundEmptyTrash()`, the bucket counter uses post-decrement
(`iter->second--`), which assigns the pre-decrement value to
`pending_files_in_bucket`. When a bucket transitions from 1 to 0 files,
`pending_files_in_bucket` receives the value 1 (not 0), so the check
`pending_files_in_bucket == 0` fails. This means `WaitForEmptyTrashBucket()`
is never woken up when a specific bucket empties, unless all global pending
files are also zero.
Fix: Use pre-decrement (`--iter->second`) so the decremented value correctly
triggers the condition variable signal.
Reviewed By: xingbowang
Differential Revision: D95303385
fbshipit-source-id: 3c5f0978ff33600acaf406b3f839cf13d9983055
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14427
In `MultiGetBlob()`, the `adjustments` autovector is populated only for blob
requests that pass validation. Requests that fail validation are skipped via
`continue`, so `adjustments` has fewer entries than `blob_reqs`. When consuming
results, `adjustments[i]` uses the `blob_reqs` index instead of the
`read_reqs`/`adjustments` index, causing an out-of-bounds read when any request
fails validation.
Fix: Replace `adjustments[i]` with `adjustments[j - 1]`, since `j` tracks the
position in `read_reqs`/`adjustments` and has already been post-incremented.
Reviewed By: hx235
Differential Revision: D95303356
fbshipit-source-id: a264ae6481f74ce33f64e40624441d666135bcd0
Summary:
Add per-file sampling of "collapsible" entry reads (single deletions, merges, and kNotFound results) that may later be used to help inform read-triggered compactions. This is a better metric than `num_reads_sampled` as it is more targeted towards reads that could be avoided via compaction.
The existing behavior of `num_reads_sampled` is that reads only gets sampled on iterator creation for a file. It is problematic because next/prev() calls are not sampled, nor are additional seeks().
This PR moves sampling to per-seek/next granularity within `LevelIterator` and adds a new `num_collapsible_entry_reads_sampled` counter that tracks how often a file serves entries that could be eliminated by compaction.
Note only L1+ files have iterator seeks/nexts/prevs sampled. Introducing this at L0 would require wrapping table reader iterators, introducing a performance cost.
## Key changes
- **New counter `num_collapsible_entry_reads_sampled`** in `FileSampledStats` tracks sampled reads that encounter deletions, single deletions, merges, or kNotFound results in both Get and Iterator paths.
- **Moved sampling from file-open to per-operation** in `LevelIterator`: sampling now happens in `SampleRead()` called from `Seek()`, `SeekForPrev()`, `SeekToFirst()`, `SeekToLast()`, `Next()`, `NextAndGetResult()`, and `Prev()`. The `should_sample` parameter was removed from `LevelIterator`'s constructor.
- **Differentiated sampling rate for Next() vs Seek()**: `should_sample_file_read_next()` uses a 64x lower sampling rate (`kFileReadSampleRate * 64`) since Next() is cheaper than Seek() and called more frequently.
- **Collapsible tracking in Get path**: `Version::Get()` now increments the collapsible counter when `GetContext::State()` is `kNotFound`, `kMerge`, or `kDeleted`.
- **Collapsible tracking in MultiGet path**: `MultiGetFromSST` also increments the collapsible counter for the same states.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14434
Test Plan:
- Added new DB tests for both num_reads_sampled and num_collapsible_entry_reads_sampled
### Benchmark results (readrandom, readseq)
Setup: 1M keys, 16-byte keys, 100-byte values, no compression, fillrandom+compact
| Benchmark | Params | ops/s (main) | ops/s (feature) | % change |
|------------|--------------------|-------------|--------------------------|----------|
| readrandom | seed=1, threads=1 | 387,194 | 389,449 | +0.6% |
| readseq | seed=1, threads=1 | 5,598,371 | 5,572,975 | -0.5% |
No meaningful performance regression observed — differences are within run-to-run noise.
Reviewed By: xingbowang
Differential Revision: D95613793
Pulled By: joshkang97
fbshipit-source-id: 9dd09c9b7527b148424bde5686f4157c7a9e1214
Summary:
### Context/Summary:
**_See below for an example of the bug:_**
L1 has 5 SST files at indices [0, 1, 2, 3, 4] where files at indices 1/2/3
share user key boundaries (e.g., file[1].largest and file[2].smallest have
the same user key).
Round-robin picks file[1] at start_index=1.
Before fix:
1. ExpandInputsToCleanCut expands {file[1]} → {file[1], file[2], file[3]}
2. Loop starts at i=2 (start_index+1), adds file[2] → duplicate!
3. start_level_inputs_ = {file[1], file[2], file[3], file[2]}
4. GetRange uses back()=file[2], returns [file[1].smallest, file[2].largest]
5. ExpandInputsToCleanCut converges on {file[1], file[2]}, missing file[3]
6. AssertCleanCut crashes (debug) or data corruption (release)
After fix:
1. ExpandInputsToCleanCut expands {file[1]} → {file[1], file[2], file[3]}
2. Find last file = file[3] at index 3, loop starts at i=4
3. start_level_inputs_ = {file[1], file[2], file[3], file[4]} (no duplicates)
4. GetRange correctly returns [file[1].smallest, file[4].largest]
_**More details:**_
When round-robin compaction picks a file, PickFileToCompact calls ExpandInputsToCleanCut which may expand start_level_inputs_ from 1 file to N files (when adjacent files share user key boundaries). Then SetupOtherFilesWithRoundRobinExpansion loops from start_index + 1 to add more files, not knowing the expansion already happened. This re-adds files already in start_level_inputs_, creating duplicates.
The duplicates corrupt GetRange, which trusts inputs.back()->largest for non-L0 levels. With a duplicate earlier file at back(), GetRange returns a truncated range. ExpandInputsToCleanCut then converges on an incomplete file set, violating the clean-cut invariant.
In debug builds this crashes at AssertCleanCut. In release builds the compaction proceeds with a non-clean-cut input, causing data corruption (newer data in a later level while older data remains in an earlier level).
The fix finds the position of the last file already in start_level_inputs_ and starts the loop after it, avoiding duplicates. When start_level_inputs_ has only 1 file (no expansion happened), the behavior is unchanged.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14436
Test Plan:
New test RoundRobinCleanCutWithSharedBoundary:
- Without fix: crashes at AssertCleanCut in SetupOtherFilesWithRoundRobinExpansion
```
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from DBCompactionTest
[ RUN ] DBCompactionTest.RoundRobinCleanCutWithSharedBoundary
db_compaction_test: db/compaction/compaction_picker.cc:81: void rocksdb::AssertCleanCut(const rocksdb::InternalKeyComparator*, rocksdb::VersionStorageInfo*, rocksdb::CompactionInputFiles*, int, rocksdb::Logger*): Assertion `false' failed.
Received signal 6 (Aborted)
Invoking GDB for stack trace...
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/usr/local/fbcode/platform010/lib/libthread_db.so.1".
0x00007fd9ba8f0e13 in __GI___wait4 (pid=1004850, stat_loc=0x7fd9b8bfb2ac, options=0, usage=0x0) at ../sysdeps/unix/sysv/linux/wait4.c:30
30 ../sysdeps/unix/sysv/linux/wait4.c: No such file or directory.
https://github.com/facebook/rocksdb/issues/4 __pthread_kill_internal (signo=6, threadid=<optimized out>) at pthread_kill.c:45
45 pthread_kill.c: No such file or directory.
https://github.com/facebook/rocksdb/issues/5 __GI___pthread_kill (threadid=<optimized out>, signo=6) at pthread_kill.c:62
62 in pthread_kill.c
https://github.com/facebook/rocksdb/issues/6 0x00007fd9ba8444ad in __GI_raise (sig=6) at ../sysdeps/posix/raise.c:26
26 ../sysdeps/posix/raise.c: No such file or directory.
https://github.com/facebook/rocksdb/issues/7 0x00007fd9ba82c433 in __GI_abort () at abort.c:79
79 abort.c: No such file or directory.
https://github.com/facebook/rocksdb/issues/8 0x00007fd9ba83bc28 in __assert_fail_base (fmt=0x7fd9ba9e11d8 "%s%s%s:%u: %s%sAssertion `%s' failed.\n%n", assertion=0x7fd9bcc04819 "false", file=0x7fd9bcc04700 "db/compaction/compaction_picker.cc", line=81, function=<optimized out>) at assert.c:92
92 assert.c: No such file or directory.
https://github.com/facebook/rocksdb/issues/9 0x00007fd9ba83bc93 in __GI___assert_fail (assertion=0x7fd9bcc04819 "false", file=0x7fd9bcc04700 "db/compaction/compaction_picker.cc", line=81, function=0x7fd9bcc04780 "void rocksdb::AssertCleanCut(const rocksdb::InternalKeyComparator*, rocksdb::VersionStorageInfo*, rocksdb::CompactionInputFiles*, int, rocksdb::Logger*)") at assert.c:101
101 in assert.c
https://github.com/facebook/rocksdb/issues/10 0x00007fd9bc2f8181 in rocksdb::AssertCleanCut (icmp=0x7fd9b9a7a840, vstorage=0x7fd9b9b6d040, inputs=0x7fd9b8bfc690, level=1, logger=0x7fd9b9aa2790) at db/compaction/compaction_picker.cc:81
81 assert(false);
https://github.com/facebook/rocksdb/issues/11 0x00007fd9bc2f8fa6 in rocksdb::CompactionPicker::ExpandInputsToCleanCut (this=0x7fd9b9b32900, vstorage=0x7fd9b9b6d040, inputs=0x7fd9b8bfc690, next_smallest=0x0) at db/compaction/compaction_picker.cc:310
310 AssertCleanCut(icmp_, vstorage, inputs, level, ioptions_.logger);
https://github.com/facebook/rocksdb/issues/12 0x00007fd9bc30cbfb in rocksdb::(anonymous namespace)::LevelCompactionBuilder::SetupOtherFilesWithRoundRobinExpansion (this=0x7fd9b8bfc910) at db/compaction/compaction_picker_level.cc:423
```
- With fix: compaction completes, data correctness verified
Reviewed By: joshkang97
Differential Revision: D95718659
Pulled By: hx235
fbshipit-source-id: 6ef125455ef5ae8a07c323835ff25588dbbb3634
Summary:
The whitebox_crash_test crashes with Assertion `checking_set_.count(cfd) == 0` failed in `FlushScheduler::ScheduleWork()` during a read-only DB open (e.g., StressTest::TestBackupRestore).
## Root cause:
Commit 3aa706c2b ("Enforce WriteBufferManager during WAL recovery") added logic to schedule flushes when `WriteBufferManager::ShouldFlush()` is true during WAL recovery. However, the drain logic in `MaybeWriteLevel0TableForRecovery()` was gated by !read_only, so in read-only mode the flush
scheduler queue was never cleared. On subsequent WAL records, `ScheduleWork()` is called again for the same CFD still in the queue, triggering the duplicate assertion.
## Fix:
Add a `read_only` parameter to `InsertLogRecordToMemtable()` and skip WBM flush scheduling entirely in read-only mode. Flushes cannot be performed during read-only recovery, so scheduling them is pointless and causes the assertion failure when the scheduler is never drained.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14440
Test Plan:
- flush_job_test — all 15 tests pass
- db_basic_test --gtest_filter="*ReadOnly*:*Recovery*:*WAL*" — all 9 tests pass
- db_write_buffer_manager_test — all 18 tests pass
- `python3 tools/db_crashtest.py whitebox --simple --duration=600 --interval=30`
Reviewed By: xingbowang
Differential Revision: D95829493
Pulled By: jaykorean
fbshipit-source-id: bdfef9ce66d507a1169381064344afd612a4b318
Summary:
Add automatic per-block interpolation search selection (`kAuto` mode) for index blocks. During SST construction, each index block's key distribution is analyzed using the coefficient of variation (CV) of gaps between restart-point keys. Blocks with uniformly distributed keys are flagged via a new bit in the data block footer, and at read time, `kAuto` resolves to interpolation search for uniform blocks and binary search otherwise.
## Key changes
- **New `BlockSearchType::kAuto` enum value**: Resolves per-block at read time to either `kInterpolation` or `kBinary` based on the block's uniformity flag. Falls back to `kBinary` on older versions that don't recognize it.
- **Write-path uniformity analysis**: `BlockBuilder::ScanForUniformity()` uses Welford's online algorithm to incrementally compute the CV of key gaps at restart points. The result is stored in a new bit (bit 30) of the data block footer's packed restart count.
- **New table option `uniform_cv_threshold`** (default: -1 `disabled`): Controls how strict the uniformity check is. Set to negative to disable. Exposed in C++, Java (JNI), and `db_bench`.
- **Code reorganization**: Block entry decode helpers (`DecodeEntry`, `DecodeKey`, `DecodeKeyV4`, `ReadBe64FromKey`) moved from `block.cc` to a new shared header `block_util.h` so they can be reused by `BlockBuilder` on the write path.
- **New histogram `BLOCK_KEY_DISTRIBUTION_CV`**: Records the CV (scaled by 10000) of each index block's key distribution for observability.
- **Java bindings**: `IndexSearchType.kAuto`, `uniformCvThreshold` getter/setter, JNI portal constructor signature updated, and `HistogramType.BLOCK_KEY_DISTRIBUTION_CV` added.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14383
Test Plan:
- `IndexBlockTest.IndexValueEncodingTest` parameterized to include `kAuto` search type alongside `kBinary` and `kInterpolation`, verifying correct seek/iteration behavior across all combinations of key distributions, restart intervals, and key lengths.
- Uniformity detection validated: blocks with uniform key distribution correctly set `is_uniform = true`, blocks with clustered/non-uniform keys set `is_uniform = false`.
- Stress test coverage
- Updated check_format_compatible to also include a "uniform" dataset. By default using uniform_cv_threshold=-1 does not result in an incompatibility issues. When manually changing the threshold (e.g. `uniform_cv_threshold=1000`), I see `bad block contents`, which is expected
## Benchmark
readrandom with `fillrandom,compact -seed=1 --statistics`:
| Benchmark | Branch | Params | avg ops/s | % change vs main | CV P50 |
|-----------|--------|--------|-----------|------------------|--------|
| readrandom | main | `binary_search, shortening=1` | 335,791 | baseline | N/A |
| readrandom | feature | `binary_search, shortening=1` (default) | 335,749 | -0.0% | 1,500 |
| readrandom | feature | `auto_search, shortening=1` (kAuto) | 366,832 | **+9.2%** | 1,500 |
| readrandom | feature | `interpolation_search, shortening=1` | 366,598 | **+9.2%** | 1,500 |
| readrandom | feature | `auto_search, shortening=2` (kAuto) | 344,631 | **+2.6%** | 1,030,000 |
| readrandom | feature | `interpolation_search, shortening=2` | 201,178 | **-40.1%** | 1,030,000 |
As seen with shortening=2, a non-uniform distribution produces a high CV, which does not use interpolation search.
## Write benchmark
There is a write overhead which scans each restart entry for a block upon Finish. In practice this is very low because currently it is only applied to index blocks.
See cpu profile (https://fburl.com/strobelight/io5hwj9h) here of `-benchmarks=fillseq,compact -compression_type=none -disable_wal=1`. Only 0.08% attributed to `ScanForUniformity`.
Reviewed By: pdillinger
Differential Revision: D94738890
Pulled By: joshkang97
fbshipit-source-id: 9661ac593c5fef89d49f3a8a027f1338a0c96766
Summary:
Address 2 compatibility issue of UDI:
A:
Trie UDI (UserDefinedIndex) does not support SeekToFirst/SeekToLast. The crash test already disabled prefix scanning (prefixpercent=0) when use_trie_index=1, but iteration (iterpercent) was still enabled.
During iteration, LevelIterator::SkipEmptyFileForward() internally calls file_iter_.SeekToFirst() when Next() crosses SST file boundaries within a level. This propagates to UserDefinedIndexIteratorWrapper::SeekToFirst() which returns NotSupported, causing "Iterator diverged from control iterator" / "VerifyIterator failed" errors across many crash test variants.
B:
BlobDB is incompatible with trie UDI (user-defined index). When BlobDB is enabled (`enable_blob_files=1`), the flush job stores large values in separate blob files and writes `kTypeBlobIndex` entries in the SST instead of `kTypeValue`. The UDI builder (`UserDefinedIndexBuilderWrapper::OnKeyAdded()`) rejects any entry whose type is not `kTypeValue`, causing the flush to fail with `"user_defined_index_factory only supported with Puts"`.
Once the flush fails, the DB enters background error state, and all subsequent `Write()` calls fail with the same error -- printed as the repeated `"multiput error: ..."` messages from
`BatchedOpsStressTest::TestPut`. Eventually, `ProcessStatus` encounters this error and triggers `assert(false)`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14432
Reviewed By: pdillinger
Differential Revision: D95457462
Pulled By: xingbowang
fbshipit-source-id: bf2bc47bd1ed75926765b2e3ff068f99f89a7793
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14424
## Context
`MemPurge` releases db_mutex_ to do its merge work and re-acquires it before adding the output memtable to the immutable list. During this window, concurrent writers can fill the active memtable and trigger a switch, adding a new immutable memtable with a higher ID. `MemPurge` then assigns its output memtable the stale ID from mems_.back() (the newest memtable in the original flush batch), which is now lower than the front of the immutable list, violating the ordering assertion in `MemTableListVersion::AddMemTable `.
## Fix
After re-acquiring db_mutex_, use `std::max(mems_.back()->GetID(), imm()->GetLatestMemTableID())` so the output memtable's ID is never lower than any existing immutable memtable. Two `TEST_SYNC_POINT`s are added to the `MemPurge` success path to enable deterministic repro.
Reviewed By: pdillinger
Differential Revision: D94756638
fbshipit-source-id: 2e9e15b4285dc6b996c8744795228180dbd73ed3
Summary:
Add a new immutable DB option `enforce_write_buffer_manager_during_recovery` to control whether WriteBufferManager buffer_size is enforced during WAL recovery. When multiple RocksDB instances share a WriteBufferManager, a recovering instance could exceed the global memory limit by replaying large amounts of WAL data into memtables. This can cause OOM, especially when other instances are actively using the shared memory budget. When this option is enabled (and a WriteBufferManager is configured), RocksDB will check `WriteBufferManager::ShouldFlush()` after each batch insertion during WAL recovery and schedule flushes when needed to keep memory bounded.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14305
Test Plan:
Two unit tests added - `WriteBufferManagerLimitDuringWALRecoverySingleDB` and `WriteBufferManagerLimitDuringWALRecoveryMultipleDBs`
```
./db_write_buffer_manager_test --gtest_filter="*WriteBufferManagerLimitDuringWALRecovery*"
```
Basic stress test added with options toggled
## To follow up
Multiple DB scenario in stress test
Reviewed By: hx235
Differential Revision: D92533792
Pulled By: jaykorean
fbshipit-source-id: 35ca70e2300a8bfd6b81a6646a65ef284fb90a9a
Summary:
Propagate builder error when flush produces empty output
When write fault injection causes the table builder to enter an error
state during flush, all subsequent Add() calls return early (ok() is
false), leaving the builder empty (IsEmpty() == true). Previously,
BuildTable() would call builder->Abandon() but not propagate the
builder's error status to 's', leaving it OK. This caused the downstream
key count validation in flush_job.cc to fire a misleading Corruption
error ('Number of keys in flush output SST files does not match...'),
which the stress test harness couldn't identify as a retryable injected
fault error, leading to SafeTerminate().
This started failing recently because ('Separate keys and
values in data blocks', ) introduced a new SST
block format (separate_key_value_in_data_block) that stores keys and
values in separate sections within data blocks. This format requires
additional write operations during Flush() inside the table builder,
increasing the probability that write fault injection
(--write_fault_one_in=128) hits a data block write and puts the builder
into an error state before any entries are committed. The bug in
BuildTable() existed before, but was rarely triggered because the old
interleaved block format had fewer write points susceptible to fault
injection during the critical Add() path.
Fix: After builder->Abandon(), propagate the builder's error status to
's' when the builder is empty due to an internal error. This ensures the
actual IOError from write fault injection is reported, which the stress
test can properly handle via IsErrorInjectedAndRetryable().
The analysis was based on stack trace. However, it would be great
if we could get direct evidence from fault injection.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14418
Test Plan: Unit test
Reviewed By: hx235
Differential Revision: D95121070
Pulled By: xingbowang
fbshipit-source-id: cfb513bd744ac34ac90cda11c1cbe49a9d0a7c6c
Summary:
This fixes a longstanding bug in which `ldb dump` swallows iterator errors. This can affect check_format_compatible.sh test results; if lucky, it will misleadingly look like a data mismatch instead of an outright failure. If unlucky, it could cause a test false negative.
However the compatibility test uses old versions of ldb, so the best way to improve the test (for the foreseeable future) is to replace `ldb dump` with `ldb scan`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14422
Test Plan: manual
Reviewed By: joshkang97
Differential Revision: D95332577
Pulled By: pdillinger
fbshipit-source-id: bef1b427dd8aaa2cabbd23b7ad9f3cad1f67a349
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14401
Unify the MultiScan and regular iterator codepaths in BlockBasedTableIterator by introducing a MultiScanIndexIterator that implements InternalIteratorBase<IndexValue>. During Prepare(), the original index iterator is swapped out for a MultiScanIndexIterator that wraps the prefetched block handles and scan range metadata. This allows SeekImpl() and FindBlockForward() to use the same code flow for both regular and MultiScan operations, eliminating the need for separate MultiScan-specific methods (SeekMultiScan, FindBlockForwardInMultiScan, MultiScanSeekTargetFromBlock, MultiScanUnexpectedSeekTarget, MultiScanLoadDataBlock, MarkPreparedRangeExhausted).
Key changes:
- New MultiScanIndexIterator class that manages scan range tracking, block handle iteration, forward-only seek enforcement, and wasted block counting
- InitDataBlock() loads blocks from ReadSet when MultiScan is active
- FindBlockForward() detects scan range boundaries via IsScanRangeExhausted() after index_iter_->Next()
- Disabled reseek optimization for MultiScan so MultiScanIndexIterator::Seek() is always called to update scan range tracking state
- Removed MultiScanState struct and all MultiScan-specific methods from BlockBasedTableIterator
- No changes to CheckDataBlockWithinUpperBound or CheckOutOfBound — they work as-is through iterate_upper_bound
- multi_scan_status_ intentionally not checked in Valid() hot path to avoid performance regression; when status is non-OK, block_iter_points_to_real_block_ is already false
- Fixed pre-existing bug in ReadSet::SyncRead() that used the base decompressor without compression dictionary, causing ZSTD data corruption when blocks with dictionary compression needed synchronous fallback reads
Reviewed By: xingbowang
Differential Revision: D93300655
fbshipit-source-id: 231059208e0cc512bc2ec43ff7055fcb2a2dc72d
Summary:
When the same user key spans adjacent data blocks with different sequence
numbers, the trie index cannot distinguish them because it only stores
user keys. This adds a side-table that stores per-leaf sequence numbers
and overflow block metadata, enabling correct post-seek correction using
seqno comparison.
Design:
- Trie always stores user-key-only separators (unchanged)
- Duplicate separators from same-key boundaries are de-duplicated
- Side-table stores leaf_seqnos[], leaf_block_counts[], and overflow
arrays (offsets, sizes, seqnos) serialized after the trie data
- Post-seek correction: after trie Seek lands on a leaf, compare
target_seq vs leaf_seqno to decide whether to advance through
overflow blocks or to the next trie leaf
- Zero overhead when no same-user-key boundaries exist (no side-table
serialized, no seqno checks at seek time)
Public API changes (user_defined_index.h):
- AddIndexEntry: single pure virtual with IndexEntryContext parameter
(replaces old 4-arg version). Context carries last_key_seq and
first_key_seq for block boundaries.
- SeekAndGetResult: single pure virtual with SeekContext parameter
(replaces old 2-arg version). Context carries target_seq.
- Fix trailing semicolons on NewBuilder/NewReader default overrides.
Wrapper layer (user_defined_index_wrapper.h):
- Extract sequence numbers from parsed internal keys and pass via
context structs to UDI builder/iterator.
- Fix CurrentIndexSizeEstimate() to delegate to internal builder
(was returning 0).
- Fix ApproximateMemoryUsage() to include UDI reader memory.
- Fix typos: lof_err_key -> log_err_key, COuld -> Could,
"Bad index name" -> "Bad index name: ".
Trie builder (trie_index_factory.cc):
- Buffer separator entries during building; at Finish(), detect whether
any same-user-key boundary was seen (sticky flag, same strategy as
ShortenedIndexBuilder::must_use_separator_with_seq_).
- When seqno encoding is needed, re-encode all separators with seqno
side-table metadata in the trie.
- Default null comparator to BytewiseComparator() to prevent crash.
Trie iterator (trie_index_factory.cc):
- Post-seek correction: compare target_seq vs leaf_seqno to advance
through overflow runs.
- Overflow run state tracking: overflow_run_index_, overflow_run_size_,
overflow_base_idx_ for O(1) access to overflow block handles.
- NextAndGetResult advances within overflow runs before moving to next
trie leaf.
- Return kOutOfBound instead of kUnknown when Seek/Next finds no blocks.
LOUDS trie (louds_trie.h, louds_trie.cc):
- Seqno side-table: builder AddKeyWithSeqno/AddOverflowBlock methods,
BFS reordering of seqno/block_count arrays, serialization/deserialization
of per-leaf seqnos, block counts, overflow handles/seqnos, and
overflow_base_ prefix sum for O(1) access.
- AppendKeySlot() helper with debug bounds assert on all key-append sites.
Dead code removal:
- LoudsTrieBuilder::NumKeys() (never called externally)
- sparse_leaf_count_ (serialized but never read by the reader)
- DenseChildNodeNum(pos), DenseLeafIndex(pos), DenseNodeNum(pos)
(superseded by FromRank variants that avoid redundant Rank1 calls)
Deserialization hardening (all in InitFromData, cold path only):
- Move num_keys_ validation before any dependent arithmetic.
- Validate dense bitvector sizes match dense_node_count_ (d_labels
must be node_count*256, d_has_child must equal d_labels.NumOnes(),
d_is_prefix_key must equal node_count).
- Validate sparse bitvector sizes match s_labels_size_ (s_has_child
and s_louds must have the same number of bits as the label array).
- Validate child position table values within s_labels_size_ bounds.
- Validate chain suffix offset+length within suffix data blob.
- Validate chain end child indices < num_internal or == UINT32_MAX.
- EliasFano: add count_ upper-bound check (<=2^30) to prevent
count_*low_bits_ integer overflow.
- Bitvector: validate select hint values < num_rank_samples_ to
prevent OOB in FindNthOneBit/FindNthZeroBit.
- EliasFano: custom move ctor/assignment to re-seat low_words_
after owned_low_data_ move (fixes dangling pointer for SSO strings).
https://github.com/facebook/rocksdb/issues/14406
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14412
Reviewed By: anand1976
Differential Revision: D95160933
Pulled By: xingbowang
fbshipit-source-id: f2c3681c1059c03d540ce1cb9cc14cc79cd9730c
Summary:
It contains 3 commits.
Commit 1: Fix /dev/shm exhaustion during make check
Fix /dev/shm exhaustion during make check
1. Add space-heavy tests to NON_PARALLEL_TEST: perf_context_test (1GB
write_buffer_size), obsolete_files_test (~1GB), backup_engine_test
(1GB), prefetch_test (1GB), and db_io_failure_test (256MB).
2. Add per-test cleanup on success: each parallel test script now
removes its test directory after passing. Failed test directories
are preserved for debugging.
3. Add age-based stale directory cleanup: at the start of make check,
remove /dev/shm/rocksdb.* directories older than 3 hours from
previous failed runs, using age-based filtering to avoid disturbing
concurrent runs.
Commit 2: Fix SIGSEGV in prefetch_test due to stale SyncPoint callbacks
Both FilePrefetchBufferTest and FSBufferPrefetchTest fixtures set up
SyncPoint callbacks capturing local variables by reference and enable
processing, but neither fixture's TearDown() clears them. When a
subsequent test runs, the stale callbacks fire with dangling
references, causing memory corruption and SIGSEGV.
Fixed by adding DisableProcessing() and ClearAllCallBacks() to both
fixtures' TearDown() methods.
Commit 3: Fix OpenFilesAsyncTest using excessive disk space
Fix OpenFilesAsyncTest using excessive disk space in /dev/shm
OpenFilesAsyncTest::SetupData() set write_buffer_size to SIZE_MAX to
prevent automatic flushes. This caused each of its ~20 parallel
instances to consume 6-43 GB in /dev/shm (validated: a single
Shutdown/3 instance used 43 GB), totaling ~233 GB and filling the
disk. This was the primary cause of "No space left on device" errors
in make check. The sst flushed is small, but it causes issue in WAL
space reservation, which bloated the disk usage.
The SIZE_MAX is unnecessary — the test writes only 4 tiny key-value
pairs and does explicit Flush() after each. The default 64 MB
write_buffer_size will never auto-flush for such small writes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14425
Test Plan: Unit Test
Reviewed By: joshkang97
Differential Revision: D95253518
Pulled By: xingbowang
fbshipit-source-id: 03907df59b3d89d90413a0e33996ec205944d48b
Summary:
wal_in_db_path_ was declared without an in-class initializer in DBImpl. While DB::Open, OpenAsSecondary, and OpenAsFollower all explicitly set this field, OpenForReadOnly and CompactedDBImpl::Open did not.
This was harmless until recently because CloseHelper() only calls PurgeObsoleteFiles when opened_successfully_ is true, and read-only DBs previously did not set opened_successfully_. After https://github.com/facebook/rocksdb/issues/14322 added opened_successfully_ = true to the read-only path, CloseHelper now executes PurgeObsoleteFiles -> DeleteObsoleteFileImpl, which reads wal_in_db_path_ to decide whether WAL files should be deleted in the foreground. Reading an uninitialized bool is undefined behavior, caught by UBSan as "invalid-bool-load" in the secondary_cache_crash_test.
Fixed by adding an in-class initializer (= false) to wal_in_db_path_. The default of false means WAL deletions use foreground deletion, which is safe for read-only DBs that don't write WALs. The existing DB::Open path continues to set the correct value explicitly.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14419
Test Plan:
- Added regression test ReadOnlyDBWalInDbPathInitialized in db_test2
- Ran test 5 times without failure
- Ran all OpenForReadOnly tests (3 tests pass)
Task:
T257988006
Reviewed By: joshkang97
Differential Revision: D95122268
Pulled By: xingbowang
fbshipit-source-id: 7f65fe1f09e7e0b42ba68f44f246615abc0757d4
Summary:
The trie-based User Defined Index (UDI) has a bug in its iterator implementation that causes "Not implemented: SeekToFirst not supported" errors during prefix scanning. This causes assertion failures in BatchedOpsStressTest::TestPrefixScan and may cause silent wrong results in other prefix scan tests.
Until the trie index is fixed, disable prefix scanning when use_trie_index is enabled by setting prefixpercent=0 and redistributing the percentage to reads.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14421
Test Plan:
- Reproduced: db_stress with --use_trie_index=1 --prefixpercent=5 crashes with assertion failure. Instrumentation revealed iterator status "Not implemented: SeekToFirst not supported" on prefix 0x35.
- Verified: db_stress with --use_trie_index=1 --prefixpercent=0 completes cleanly.
Reviewed By: anand1976
Differential Revision: D95135250
Pulled By: xingbowang
fbshipit-source-id: 81540b2426aa1a855a58e6ca9e68a035d53aa2d8
Summary:
The kv ratio compaction option sanitization is too strict. Sometimes, it blocks engine start up. Relax the validation and convert it to runtime check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14397
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D94709471
Pulled By: xingbowang
fbshipit-source-id: cc076c397b3acfa426112063224771a196684798
Summary:
When using optimistic transactions (--use_optimistic_txn=1), the transaction commit can fail with Status::TryAgain() if the memtable history is insufficient for conflict detection (controlled by max_write_buffer_size_to_maintain). ExecuteTransaction retries up to 10 times, and if all retries fail, it returns TryAgain.
Previously, this TryAgain status was not handled in TestPut, TestDelete, and TestSingleDelete, causing the stress test to call SafeTerminate() and crash with "put or merge error" / "delete error".
This is an expected condition with optimistic transactions and should be handled gracefully by rolling back the pending expected value.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14410
Test Plan:
- make -j db_stress && run crash_test_with_optimistic_txn
- Stress test with --use_optimistic_txn=1 --use_txn=1 with small max_write_buffer_size_to_maintain to verify no crash on TryAgain.
Reviewed By: hx235
Differential Revision: D95077227
Pulled By: xingbowang
fbshipit-source-id: 15c343cc1c5f888fb79e95a87cfe46145f98b3a1
Summary:
The trie-based UDI (LoudsTrie/Bitvector) stores zero-copy reinterpret_cast pointers into the serialized UDI block data. When mmap_read is enabled, the block data resides in memory-mapped file pages. If the SST file's mmap mapping is later invalidated (e.g. through table cache eviction or DB reopen), the trie's internal pointers become dangling, causing SIGSEGV.
For now, disable mmap_read when use_trie_index is enabled:
- db_crashtest.py: force mmap_read=0 when use_trie_index=1
- db_stress_tool.cc: reject the combination with an error message
The proper fix, involve loading index then fix the pointer address.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14409
Reviewed By: pdillinger
Differential Revision: D94971965
Pulled By: xingbowang
fbshipit-source-id: 6abbd4bfbfd26e7ba8725a93a6c68beb26f6393e
Summary:
In StressTest::OperateDb(), read_opts.table_index_factory is set to udi_factory_.get() (a raw pointer) once at function entry. When Reopen() is called during the stress test loop, Open() recreates udi_factory_ via std::make_shared<TrieIndexFactory>(), destroying the old factory. But read_opts.table_index_factory still holds the dangling pointer.
When MultiGet subsequently calls UserDefinedIndexReaderWrapper::NewIterator() and accesses read_options.table_index_factory->Name(), it dereferences a dangling pointer, causing a segmentation fault.
Fix: After each reopen, update read_opts.table_index_factory to point to the newly created udi_factory_ instance.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14411
Test Plan: - Ran db_stress with --use_trie_index=1 --reopen=20 --use_multiget=1 4 times without failure (previously crashed with SIGSEGV)
Reviewed By: pdillinger
Differential Revision: D95064144
Pulled By: xingbowang
fbshipit-source-id: 7139baacda659d8a3cb84fdcc4822a428542bf0c
Summary:
Add `open_files_async` option for faster DB startup. When enabled, SST file opening and validation is deferred to a background thread after `DB::Open` returns, reducing startup latency for databases with many SST files. WAL recovery remains synchronous.
To support this, `FindTable` is extended with a pinning mechanism that stores the cache handle directly on `FileMetaData` via a new `PinnedTableReader` class, and sets the table reader atomically so subsequent reads skip cache lookups. `FileDescriptor::table_reader` is replaced with `PinnedTableReader pinned_reader` which wraps a `std::atomic<TableReader*>` with acquire/release ordering to safely handle concurrent access between the background opener and read threads.
Should validations fail, the background opener sets a `kAsyncFileOpen` background error. Future read requests will look up the table reader again via the cache, and if any validations fail there it will get propagated to the user (existing behavior when `max_open_files > 0`).
This feature is most useful when `max_open_files=-1`, because otherwise file opening is already capped at 16 files and DB open should be fast.
## Restrictions
- This feature also is incompatible with fifo compaction because fifo compaction requires reading table properties under DB mutex. When table reader is unpinned, this may cause a DB hang.
- This feature is also incompatible with `skip_stats_update_on_db_open=false` because it will result in even longer DB open
## Key changes
- New `open_files_async` DB option with C, Java, and `db_bench` bindings
- `BGWorkAsyncFileOpen` background worker that opens all SST files post-`DB::Open`, with shutdown awareness via `shutting_down_` flag
- New `PinnedTableReader` class in `version_edit.h` — thread-safe wrapper holding `std::atomic<TableReader*>` and `Cache::Handle*` with proper acquire/release ordering. Replaces the old `FileDescriptor::table_reader` raw pointer and `FileMetaData::table_reader_handle`
- Extract `LoadTableHandlersHelper` into `db/version_util.cc` — shared between `VersionBuilder::LoadTableHandlers` (for version edits during recovery) and `BGWorkAsyncFileOpen` (for base storage post-open)
- `FindTable` extended with `pin_table_handle` and `out_table_reader` params — when pinning is enabled, the table reader is stored on `FileMetaData` so Get/MultiGet/Iterator skip redundant cache lookups. `FindTable` now performs the pinned-reader fast-path check internally instead of requiring callers to check `fd.table_reader` beforehand
- Note: pinning is explicit (not default) because some callers create temporary `FileMetaData`s that would need to properly clean up table handles
- `CompactedDBImpl` updated to use `FindTable` + pinning instead of raw `fd.table_reader` access for Get/MultiGet
- New `kAsyncFileOpen` background error reason in `listener.h` and `error_handler.cc`
- Add a check in ~DBImpl to ensure async file open task has not been forgotten to be scheduled in (future) subclasses of DBImpl. Certain subclasses that never use it will need to explicitly mark it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14322
Test Plan:
- `OpenFilesAsyncTest` parameterized over `num_flushes` (1, 20), `ReadType` (Get, MultiGet, Iterator), `max_open_files` (-1, 10), and `read_only` (true, false)
- **ConcurrentFileAccess**: concurrent reads and compactions race with async opener
- **AfterRead**: reads happen before async opener, verifying lazy open and that the opener sees already-pinned readers
- **BeforeRead**: async opener completes first, verifying reads use pre-loaded table readers
- **Shutdown**: DB closes before async opener starts, verifying clean cancellation with 0 file opens
- **Error**: corrupted SST files, verifying `kAsyncFileOpen` background error is set and reads return corruption
- **DropColumnFamily**: CF dropped before async opener runs, verifying the opener gracefully skips dropped CFs
- Added to crash test
### Benchmark
To simulate a high-latency remote filesystem, I set up a virtual filesystem with dm-delay using 10ms reads, 0 ms writes.
```
# Generate a DB with many L0 files
TEST_TMPDIR=/data/users/jkangs/dm-delay-test/mnt ./db_bench -benchmarks=fillseq -disable_auto_compactions=true -write_buffer_size=1000 -num=1000000
```
```
./db_bench -use_existing_db=true -db=/data/users/jkangs/dm-delay-test/mnt/dbbench -benchmarks=readrandom -reads=1 -report_open_timing=true -open_files_async=true -use_direct_reads -file_opening_threads=1 -skip_stats_update_on_db_open
OpenDb: 25.1419 milliseconds
```
```
./db_bench -use_existing_db=true -db=/data/users/jkangs/dm-delay-test/mnt/dbbench -benchmarks=readrandom -reads=1 -report_open_timing=true -open_files_async=false -use_direct_reads -file_opening_threads=1 -skip_stats_update_on_db_open
OpenDb: 23109.4 milliseconds
```
### No read regressions
On main branch
```
./db_bench -use_existing_db=true -db=/dev/shm/dbbench -benchmarks=readrandom -seed=1 -threads=8 -duration=30
readrandom : 4.827 micros/op 1657100 ops/sec 30.005 seconds 49720992 operations; 183.3 MB/s (6198999 of 6198999 found)
```
On this branch
```
./db_bench -use_existing_db=true -db=/dev/shm/dbbench -benchmarks=readrandom -seed=1 -threads=8 -duration=30
readrandom : 4.863 micros/op 1644808 ops/sec 30.007 seconds 49354992 operations; 182.0 MB/s (6099999 of 6099999 found)
./db_bench -use_existing_db=true -db=/dev/shm/dbbench -benchmarks=readrandom -seed=1 -threads=8 -duration=30 -open_files_async=true
readrandom : 4.803 micros/op 1665392 ops/sec 30.004 seconds 49968992 operations; 184.2 MB/s (6222999 of 6222999 found)
```
Reviewed By: pdillinger, xingbowang
Differential Revision: D93538033
Pulled By: joshkang97
fbshipit-source-id: 32ac70c112cd733b7c1e1c1e2e7ce6422318a5ae
Summary:
Fix stress test to check UDI usage for unsupported iterator ops
UserDefinedIndexIterator only supports Seek(target) + Next() - it requires
a target key for all seeks. The following operations are not supported:
- SeekToFirst() - no target key
- SeekToLast() - no target key
- SeekForPrev() - not in UDI interface
- Prev() - not in UDI interface
This fix checks for UDI usage at both levels:
1. ReadOptions level: ro.table_index_factory != nullptr
2. CF/table level: udi_factory_ != nullptr (BlockBasedTableOptions)
This is future-proof for any UDI implementation, not just trie index.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14407
Reviewed By: joshkang97
Differential Revision: D94925001
Pulled By: xingbowang
fbshipit-source-id: bf2ceacc7acebbc2347be445e2f208820a97b817
Summary:
As a follow-up to https://github.com/facebook/rocksdb/issues/13736, allow a "quiet" DB to react much sooner to time-based compaction triggers. For details see DBImpl::ComputeTriggerCompactionPeriod() implementation.
Also based on review feedback, fixing a bug where only column families setting periodic compaction would be triggered, rather than any time-based compaction.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14396
Test Plan: extended+added unit tests to cover much of the logic
Reviewed By: xingbowang
Differential Revision: D94626166
Pulled By: pdillinger
fbshipit-source-id: de9ca19e46bdba5d9715474efbc61805354d730d
Summary:
When the UDI wrapper's OnKeyAdded() encountered a non-Put key type (e.g. Delete or Merge), it set its internal status_ to non-OK and stopped forwarding OnKeyAdded() to the wrapped internal index builder. However, AddIndexEntry() was always forwarded unconditionally. This asymmetry left the internal ShortenedIndexBuilder's current_block_first_internal_key_ empty, triggering an assertion failure in GetFirstInternalKey() during the buffered-block replay in MaybeEnterUnbuffered().
The crash required three conditions to co-occur:
1. UDI enabled (use_trie_index=1)
2. Compression dictionary enabled (triggering kBuffered mode)
3. Non-Put entries in the data (Delete, Merge, etc.)
Fix: move the internal_index_builder_->OnKeyAdded() call before the status_ guard so the internal builder always receives every key, matching the unconditional forwarding in AddIndexEntry().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14404
Reviewed By: archang19
Differential Revision: D94791493
Pulled By: xingbowang
fbshipit-source-id: 882e409f61ae9aca084e9511794ca32ba4ff090f
Summary:
Implement a Fast Succinct Trie (FST) index based on LOUDS encoding as a User Defined Index plugin for RocksDB's block-based tables. First step toward trie-based indexing per https://github.com/facebook/rocksdb/issues/12396.
The trie uses hybrid LOUDS-Dense (upper levels, 256-bit bitmaps) + LOUDS-Sparse (lower levels, label arrays) encoding inspired by the [SuRF paper](https://www.pdl.cmu.edu/PDL-FTP/Storage/surf_sigmod18.pdf) (Zhang et al., SIGMOD 2018). The boundary between dense and sparse levels is automatically chosen to minimize total space.
Key Components
- **Bitvector** with O(1) rank and O(log n) select using a rank LUT sampled every 256 bits with popcount intrinsics. Uses uint32_t rank LUT entries (halving memory vs uint64_t). Includes word-level `AppendWord()` for efficient dense bitmap construction and `AppendMultiple()` optimized at word granularity for bulk bit fills.
- **Streaming trie builder** using flat per-level arrays with deferred internal marking, handle migration for prefix keys, and lazy node creation. Infers trie structure directly from sorted keys via LCP analysis in a single pass (no intermediate tree).
- **LoudsTrie** for immutable querying with BFS-ordered handle reordering built into single-pass level-by-level serialization. Move-only semantics with correct pointer re-seating after `std::string` move.
- **LoudsTrieIterator** with rank-based traversal, key reconstruction from trie path, and stack-based backtracking for `Next()`. Uses packed 8-byte `LevelPos` (is_dense flag in bit 63) and `autovector<LevelPos, 24>` to avoid heap allocation. Key reconstruction uses a raw char buffer allocated once to `MaxDepth()+1` bytes.
- **TrieIndexFactory/Builder/Reader/Iterator** implementing the `UserDefinedIndexFactory` interface.
- **Zero-copy block handle loading** using two fixed-width uint64_t arrays (offsets + sizes) with 8-byte alignment, enabling O(1) initialization via direct pointer assignment.
Seek Hot Path Optimizations
- **Fanout-1 sparse fast path**: Most sparse nodes in tries built from zero-padded numeric keys have exactly one child. Detected via `start_pos + 1 == end_pos` and inlined as a single byte comparison, avoiding the full `SparseSeekLabel` call.
- **Linear scan for small sparse nodes**: `SparseSeekLabel` uses sequential scan for nodes with ≤16 labels instead of binary search. Faster for common 10-child digit nodes where branch misprediction cost outweighs linear scan cost.
- **Rank reuse**: `DenseLeafIndexFromRankAndHasChildRank` and `SparseLeafIndexFromHasChildRank` overloads accept pre-computed `has_child_rank` from the Seek descent, avoiding redundant `Rank1` calls.
General Performance Optimizations
- **Select-free sparse traversal**: Precomputed child position lookup tables (`s_child_start_pos_`/`s_child_end_pos_`) eliminate `Select1` calls during Seek. Sparse traversal tracks `(start_pos, end_pos)` directly, using only `Rank1` (O(1)) + array lookup (O(1)) for child descent.
- **Cached label_rank pattern**: Eliminates redundant `Rank1` calls in hot paths (Seek, Next, Advance all cache and reuse the label_rank computed for has_child checking).
- **Leaf index fast path**: When no prefix keys exist (common case), `SparseLeafIndex` and `DenseLeafIndexFromRank` skip the prefix-key `Rank1` calls entirely, reducing from 3 to 1 `Rank1` call.
- **Popcount-based Select64** via 6-step binary search within 64-bit words.
- MSVC portability using RocksDB's `BitsSetToOne`/`CountTrailingZeroBits`.
Benchmark Results
Trie Seek at 32K keys (16-byte keys, 5M lookups, median of 5 runs):
| Configuration | ns/op |
|---|---:|
| Trie (optimized) | **118** |
| Binary search (native) | 134 |
Trie Seek is **~12% faster** than native binary search index at 32K keys per block.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14310
Reviewed By: anand1976
Differential Revision: D93921511
Pulled By: xingbowang
fbshipit-source-id: ba18604bc6bd4f575311a1ae3047a541274869b6
Summary:
**Summary:**
This is to sync an internal change of passing `Statistics*` to the GetContext constructor and collecting more stats as well as fix a bug this change created.
SstFileReader::MultiGet was passing nullptr for both `SystemClock*` and `Statistics*` to the GetContext constructor. After `Statistics*` was passed to the GetContext constructor (the internal change), this caused a segfault when a merge operation was triggered with statistics enabled, because the merge helper's StopWatchNano attempted to dereference the null clock pointer. Fix by passing `r->ioptions.clock` from the reader's options.
Additionally, add `assert(clock_)` guards to `StopWatchNano::Start()` and `ElapsedNanos()` to catch null clock bugs in debug builds. Can't do so in release build because it's on hot path.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14393
Test Plan:
- `./sst_file_reader_test --gtest_filter='SstFileReaderTableMultiGetTest.Basic'` exercises merge with statistics enabled, previously segfaulted without the fix with the clock, now passes.
- `./sst_file_reader_test` - all tests pass.
Reviewed By: xingbowang
Differential Revision: D94599343
Pulled By: hx235
fbshipit-source-id: 0a748bb00ee27bb202d01d410b52657101c05de0
Summary:
The nightly build had a flag for portable build on folly, but rocksdb does not. This causes link error. Fix this by removing portable build flag in folly build in nightly run. Nightly run will always build without cache using native flag. Only PR jobs uses cache.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14391
Test Plan: nightly run
Reviewed By: joshkang97
Differential Revision: D94520182
Pulled By: xingbowang
fbshipit-source-id: 7c7d3b089744c7e2e8ea073158f1b5b80db420d4
Summary:
-march=native in cmake builds causes SIGILL after cache hits
Issue: PORTABLE=1 env var only works for Makefile builds. cmake ignores it and injects -march=native via its own CPU detection. Since ccache hashes the flag string literally, a cache compiled on an AVX-512 runner and restored on a non-AVX-512 runner produces a SIGILL crash.
Fix: Added -DPORTABLE=ON to build-linux-cmake-with-benchmark-no-thread-status and build-linux-cmake-with-folly-coroutines cmake commands.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14388
Test Plan: Github CI
Reviewed By: joshkang97
Differential Revision: D94367557
Pulled By: xingbowang
fbshipit-source-id: 5aa39cc36fc004d0ee636cd3fa197880d6cb773a
Summary:
**Context/Summary:**
It's very easy to make mistake in removing deprecated option such as deleting the corresponding entry in type info that breaks backward compatibility or missing tests to test backward compatibility. Example: https://github.com/facebook/rocksdb/pull/14350#discussion_r2834554287. Added a claude md file for that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14360
Test Plan: In claude code, prompted for deprecated option removal with this claude md file in a repo separated from my previous deprecation efforts as much as possible and received correct change at one try.
Reviewed By: pdillinger
Differential Revision: D93920127
Pulled By: hx235
fbshipit-source-id: 1977ce7404b0188f84258552b4843e70394a5aea
Summary:
PR https://github.com/facebook/rocksdb/issues/14266 ("Remove compression support") removed compression-related functionality from blob_dump_tool and ldb commands. While this was valid for the legacy stacked BlobDB (which no longer supports compression), the integrated blob storage (`enable_blob_files`) uses the same blob file format and fully supports compression via `blob_compression_type`.
This partial revert restores the ability to view uncompressed blob data from compressed blob files, which is essential for debugging and analysis of integrated blob storage.
Restored functionality:
- `blob_dump --show_uncompressed_blob` option
- `ldb dump --dump_uncompressed_blobs` option
- `ldb dump_live_files --dump_uncompressed_blobs` option
- Related ldb_test.py test coverage
The decompression implementation in utilities/blob_db/blob_dump_tool.cc has been updated to use the modern
`GetBuiltinV2CompressionManager()->GetDecompressorOptimizeFor()` API.
Suggested follow-up:
* Tests for blob_dump (or integrate it into sst_dump/ldb?)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14382
Test Plan:
Restored ldb_test.py
Some manual testing
Reviewed By: mszeszko-meta
Differential Revision: D94257588
Pulled By: pdillinger
fbshipit-source-id: c6d8a556a51ec9422df208ae161806ccc1f20b36
Summary:
Accelerate CI: ccache integration, replace scan-build, upgrade runners
1. ccache integration via reusable composite action (setup-ccache)
- Created .github/actions/setup-ccache/action.yml
- Applied to 13+ Linux compilation jobs in pr-jobs.yml
- PORTABLE=1 by default to avoid illegal instruction errors on
heterogeneous runners; disabled for jobs linking pre-built Folly
- On cache hit, reducing compilation from 20+ min to ~2-5 min per job
- ccache placed before Folly build so Folly compilation also
benefits from cache on Folly cache misses
2. Replace scan-build with clang-tidy (30+ min -> seconds)
- Removed build-linux-clang18-clang-analyze job from pr-jobs.yml
- Expanded .clang-tidy to enable all clang-analyzer-* checks,
matching scan-build's full coverage
- Existing clang-tidy-comment.yml workflow now handles both
clang-tidy and clang-analyzer checks on changed files only
3. Upgrade Folly job runners for faster builds and tests
- build-linux-make-with-folly: 16-core -> 32-core, -j32 -> -j64
- build-linux-cmake-with-folly-coroutines: 16-core -> 32-core,
-j20 -> -j64 (both make and ctest)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14368
Reviewed By: joshkang97
Differential Revision: D94166095
Pulled By: xingbowang
fbshipit-source-id: 3dfbd71aace3ba9cb2e790873bdbedba20215657
Summary:
Add proper synchronization in mempurge unit test to fix flaky test
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14377
Test Plan: Unit test
Reviewed By: nmk70
Differential Revision: D94153290
Pulled By: xingbowang
fbshipit-source-id: 2a2cdcc6f8b5500bb5e68d2d223255cd1a5660b3
Summary:
Sanitize crash test arg for interpolation search when UDTs are used. Currently its not supported and returns invalid arg on DB open.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14374
Test Plan: CI
Reviewed By: archang19
Differential Revision: D94123220
Pulled By: joshkang97
fbshipit-source-id: 89b63f14a721bc692465620298161af23eec2446
Summary:
Summary
- Fix a race condition in three WritePreparedTransactionSeqnoTest tests (SeqnoGoesBackwardsDuringErrorRecovery, SeqnoDiscrepancyDuringErrorRecovery, ConcurrentWritesDuringErrorRecovery) that could cause permanent hangs.
- The tests inject a filesystem error during flush via a WriteManifest sync point callback, then wait for background error recovery to complete. The bug was in the ordering of operations after recovery starts: SetFilesystemActive(true) was called before ClearCallBack, allowing a window where recovery's ResumeImpl could trigger the callback and re-disable the filesystem. This left the filesystem permanently disabled, causing all recovery retries to fail and exit without firing the RecoverSuccess sync point, leaving the test thread blocked forever.
- The fix swaps the order so ClearCallBack is called before SetFilesystemActive(true), ensuring the filesystem cannot be re-disabled by a late callback firing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14361
Test Plan:
- Stress tested with gtest_parallel (500 iterations, 32 workers, 60s timeout) with no hangs observed.
- Previously reproduced the hang at ~7% rate under stress with 15s timeout before the fix.
Reviewed By: pdillinger
Differential Revision: D93929251
Pulled By: anand1976
fbshipit-source-id: 9cb6844ed20146c754091575156b08d5551b3034
Summary:
Introduce a new V2 serialization format for wide column entities that supports storing individual column values in blob files. The V2 format adds a column type section that marks each column as either inline or blob-index, enabling per-column blob storage for large values.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14314
Reviewed By: pdillinger
Differential Revision: D92832066
Pulled By: xingbowang
fbshipit-source-id: 13c24347e1f481a059d67eef987d2d2b184b4a51
Summary:
RocksDB has been using clang-tidy for a long time inside Meta. However, it is not efficient for external contributor, as the result from clang-tidy has to be ferried back through internal contributor. This PR added support to run clang-tidy on external github CI. It added .clang-tidy file based on internal version. It run clang-tidy in a separate pr job and a workflow step would post the pr job result to the PR itself. See example below.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14347
Test Plan: Github CI
Reviewed By: archang19
Differential Revision: D93862467
Pulled By: xingbowang
fbshipit-source-id: bb4330241036894deb619470efd73a7041a8b62f
Summary:
**Context/Summary:**
Remove deprecated, unused APIs and options:
- ReadOptions::managed: This option was not used anymore. The functionality it controlled has been removed long ago.
- ColumnFamilyOptions::snap_refresh_nanos: Deprecated and unused option.
Corresponding C API (rocksdb_readoptions_set_managed) and Java API (ReadOptions.managed/setManaged) are also removed. All related checks an db_impl and db_impl_secondary iterators are cleaned up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14350
Test Plan: make check
Reviewed By: pdillinger
Differential Revision: D93812438
Pulled By: hx235
fbshipit-source-id: e4a9d21c65f83294b6d0878286ba14024f049bac
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14355
SupportedOps advertised kAsyncIO based only on the IsIOUringEnabled() weak symbol check, without verifying that the constructor's io_uring probe actually succeeded. Add a thread_local_async_read_io_urings_ null check so kAsyncIO is only reported when the probe passed. Also update the constructor to probe with the same IORING_SETUP_SINGLE_ISSUER | IORING_SETUP_DEFER_TASKRUN flags that ReadAsync and MultiRead use at runtime.
Reviewed By: anand1976
Differential Revision: D93780065
fbshipit-source-id: 6f51f544b267cb39d09b49949a9485f55eeae12e
Summary:
**Context/Summary:**
Remove `SstFileWriter::Add()` (deprecated in favor of `Put()`) and the `skip_filters` parameter from `SstFileWriter` constructors (deprecated in favor of setting `BlockBasedTableOptions::filter_policy` to `nullptr`).
Both APIs have zero active callers. The `skip_filters` field is also removed from `TableBuilderOptions` (write-side only; the read-side `TableReaderOptions::skip_filters` is unchanged).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14352
Test Plan: make check
Reviewed By: xingbowang
Differential Revision: D93812389
Pulled By: hx235
fbshipit-source-id: 236b36a6e664758ab5ad90e606bc195d0a6de70f
Summary:
a couple recent failures in this test. Waiting for purge and disabling sync points before Close should resolve the issues.
Also fixing EventListenerTest.BlobDBOnFlushCompleted because it showed up as flaky in CI for this PR
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14349
Test Plan: watch CI
Reviewed By: mszeszko-meta
Differential Revision: D93619322
Pulled By: pdillinger
fbshipit-source-id: bb9fc7d3c0ecaaeaffe4305e1ad403cbcd597484
Summary:
In my last version bump, I forgot that the next release would be a major release. We can fix that now ahead of release cut.
I'm also updating folly now because I have experience resolving folly issues. Folly commit e04860553 changed libevent to build as static-only so required a change in our build.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14357
Test Plan: CI
Reviewed By: xingbowang
Differential Revision: D93797666
Pulled By: pdillinger
fbshipit-source-id: 22179da900f9dc6c5544163071079a4701c7c663
Summary:
**Summary/Context:**
Remove the `InRange()` virtual method from `SliceTransform` and all its overrides. This method was marked DEPRECATED, never called by RocksDB, and existed only for backward compatibility.
Also removes the `in_range` callback parameter from `rocksdb_slicetransform_create()` in the C API, which is a breaking change appropriate for a major version release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14353
Test Plan: Make check
Reviewed By: xingbowang
Differential Revision: D93795070
Pulled By: hx235
fbshipit-source-id: 5eba23f1d038b19c494997a55e5d8ca379fbedcb
Summary:
There was an edge case missed in the implementation of interpolation search for target keys that had a length smaller than the shared prefix.
E.g. first_key = "aaaaaa", last_key = "aaaaaz", target_key = "aaz". In the existing setup, we will seek to position 0, but in reality is should be seeked to the end.
#### The fix
The solution here was to also do a bounds check on the first search iteration. We utilize memcmp on the target key with the shared_prefix to determine if the target key is outside the bounds. An edge case here is if the target key itself a prefix of the shared prefix (e.g. target = "aaaa"), in this case memcmp return return 0, but the target key is actually smaller.
### Minor optimizations
- cache left,right values so we don't need to re-compute it when left/right boundaries don't change
- In ReadBe64FromKey, utilize memcpy + swap for fast path
- since we have already computed a shared_prefix, every other comparison only needs to compare the non-shared suffixes
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14343
Test Plan:
Added new unit tests to test this case
### Benchmarks
No significant regressions due to additional memcmp.
#### Configuration
- **CPU:** 192 * AMD EPYC-Genoa Processor
- **RocksDB Version:** 10.12.0
- **Compression:** Snappy
- **Entries:** 1,000,000
- **Value Size:** 100 bytes
- **Index Search Type:** interpolation_search
- **Index Shortening Mode:** 1
#### Results
| Benchmark | Params | ops/s (main) | ops/s (feature) | % change |
|-----------|--------|-------------|-----------------|----------|
| readrandom | 16B keys, no prefix | 367,264 | 369,163 | +0.52% |
| readrandom | 100B keys, prefix_size=50 | 376,066 | 371,193 | -1.29% |
Reviewed By: pdillinger
Differential Revision: D93535267
Pulled By: joshkang97
fbshipit-source-id: beda182efce1e914ff587e697b927347cfa42656
Summary:
Add clang-tidy-comment workflow. This workflow allows pr clang tidy pr job to post the clang-tidy finding directly on the PR page.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14348
Test Plan: Will be tested with next clang-tidy PR
Reviewed By: joshkang97
Differential Revision: D93670150
Pulled By: xingbowang
fbshipit-source-id: 8245f9d5bde8cf800d88034c4339de9f387c5692
Summary:
Extending https://github.com/facebook/rocksdb/issues/14323 by testing scenarios for compaction after downgrade. Detail: we shouldn't need to test loading options with compaction, as options file inclusion is mostly a sanity check for "can you open the DB with options file?"
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14344
Test Plan: manual run of SHORT_TEST=1 J=140 tools/check
Reviewed By: xingbowang
Differential Revision: D93553897
Pulled By: pdillinger
fbshipit-source-id: ec08ae2a3d49971e24a215e38df9506fe1133096
Summary:
and remove deprecated DB::MaxMemCompactionLevel(). In the process of pushing through a relatively clean refactoring of uses of the old functions, some other minor public APIs are also migrated from raw DB pointers to unique_ptr.
Claude did pretty much all the work, but requiring dozens of prompts to actually push through relatively clean phase out of raw DB pointers from what needed to be touched, and leaving that code in better shape. (Hundreds of `DB*` still remain all over the place even outside C and Java bindings.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14335
Test Plan: existing tests; no functional changes intended
Reviewed By: xingbowang, mszeszko-meta
Differential Revision: D93523820
Pulled By: pdillinger
fbshipit-source-id: e4ca22ad81cd2cfe91122d7507d7ca34fe03d043
Summary:
After PR https://github.com/facebook/rocksdb/issues/14315 dropped support for block-based table format_version < 2, several code paths became obsolete. This change removes them.
Investigation findings:
1. Table properties are now a hard requirement for block-based SST files:
- format_version >= 2 guarantees a properties block exists
- Removed defensive conditionals like `if (rep_->table_properties)`
- Missing properties block now returns Status::Corruption instead of just logging an error. This is important because some properties affect the semantic interpretation of the file.
2. Index type property (kIndexType) is now required:
- kIndexType was introduced in Feb 2014 (commit 74939a9e1), ~11 months BEFORE format_version was introduced in Jan 2015
- BlockBasedTablePropertiesCollector::Finish() has always written kIndexType unconditionally for all block-based tables
- Therefore all format_version >= 2 files have this property
- Now returns Status::Corruption if missing instead of silently defaulting to kBinarySearch
3. Removed SetOldTableOptions() from sst_file_dumper:
- This fallback handled files without a properties block
- Dead code since format_version >= 2 guarantees properties exist
4. Removed kPropertiesBlockOldName ("rocksdb.stats") fallback:
- The properties block was renamed from "rocksdb.stats" to "rocksdb.properties" in RocksDB 2.7 (April 2014)
- format_version 2 was introduced in RocksDB 3.10 (Oct 2015)
- All table formats (block-based, plain, cuckoo) were created after the rename, so they all use "rocksdb.properties"
- The backward compatibility fallback in FindOptionalMetaBlock() was dead code for all supported table formats
5. Removed obsolete assertion about format_version 0 checksum in BlockBasedTableBuilder::WriteFooter()
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14327
Test Plan: some tests updated for updated requirements. Mostly, CI including format compatible test
Reviewed By: mszeszko-meta
Differential Revision: D93124820
Pulled By: pdillinger
fbshipit-source-id: eb12cbdca0e69f34a08051d5160c282384128a4a
Summary:
I'm implementing this intending it to be used for https://github.com/facebook/rocksdb/issues/14287
Refactor the data block footer encoding/decoding to use a struct-based Encode/Decode API (DataBlockFooter), reserving the top 4 bits of the footer for metadata:
- Bit 31: Hash index present (kDataBlockBinaryAndHash) - existing use
- Bits 28-30: Reserved for future features
Comments have some detail for why it is safe to assume no practical existing SST files would use these newly reserved bits. And for forward compatibility, existing versions detect (non-zero) use of these new bits as impossibly large num_restarts and report "bad block contents". Not perfect, but not bad.
Key changes:
- Replace PackIndexTypeAndNumRestarts/UnPackIndexTypeAndNumRestarts with DataBlockFooter::EncodeTo/DecodeFrom methods
- DecodeFrom returns a detailed error when reserved bits are set, enabling graceful failure on newer format versions
- Reduce kMaxNumRestarts from 2^31-1 to 2^28-1 (268M), which is adequate for the maximum possible restarts in a 4GiB block
- Add GetCorruptionStatus() to Block for detailed error messages (Note that we are sensitive to the size of Block objects, so have to avoid adding unnecessary new members.)
- Remove obsolete kMaxBlockSizeSupportedByHashIndex size checks
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14332
Test Plan:
- Existing unit tests and format compatibility test
- Add test for reserved bit detection (ReservedBitInDataBlockFooter)
Reviewed By: joshkang97
Differential Revision: D93293152
Pulled By: pdillinger
fbshipit-source-id: b65a83e96bb09a98fb9b8b2dd9f754653ca7ed4d
Summary:
This PR fixes a bug in the interaction between WritePrepared/WriteUnprepared TransactionDB (with two_write_queues=true) and background error recovery. This bug caused crash tests to fail with a "sequence number going backwards" error during DB open.
Root Cause
------------
When two_write_queues=true, sequence numbers are allocated via FetchAddLastAllocatedSequence() before a write completes, but are only published via SetLastSequence() after the write succeeds. If a background error occurs (e.g., a MANIFEST write failure during flush), the error recovery path in DBImpl::ResumeImpl creates new memtables and WAL files. The new WAL's starting sequence number is based on LastSequence() (the published value), which can be lower than already-allocated sequence numbers that were written to the old WAL. On subsequent recovery, RocksDB detects that sequence numbers in the new WAL are lower than those in the old WAL and reports a "sequence number going backwards" corruption error, causing the DB to fail to open.
Fix
---
The fix adds a call to a new VersionSet::SyncLastSequenceWithAllocated() method at the beginning of DBImpl::ResumeImpl, before any new memtables or WALs are created. This method advances last_sequence_ to match last_allocated_sequence_ if the latter is higher, ensuring the new WAL starts with a sequence number that is at least as high as any previously allocated one.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14313
Test Plan:
---------
Add new unit tests in write_prepared_transaction_test_seqno
Reviewed By: pdillinger
Differential Revision: D92746944
Pulled By: anand1976
fbshipit-source-id: 34385fc13fd74435dd1c3283637eb118f45d887e
Summary:
Fixes blog author display issues on rocksdb.org/blog by:
* Adding missing authors to authors.yml: pdillinger, alanpaxton, akankshamahajan15, anand1976, poojam23
* Standardizing on GitHub usernames: renamed sdong → siying
* Fixing typo in 2016-02-25-rocksdb-ama.markdown: yhchiang → yhciang
* A short note in CLAUDE.md
Authors were not showing on the blog because they were referenced in post frontmatter but not defined in the _data/authors.yml lookup file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14342
Test Plan: push & see ;)
Reviewed By: mszeszko-meta
Differential Revision: D93523972
Pulled By: pdillinger
fbshipit-source-id: 757c33e80f3c1d99ff4134a37321f40634d6e294
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14321
Add file_checksum and file_checksum_func_name fields to FileOptions so that downstream FileSystem implementations can access per-file checksum metadata when SST files are opened. The fields are populated from FileMetaData at all call sites where SST files are opened via NewRandomAccessFile: TableCache::GetTableReader, Version::GetTableProperties, and CompactionJob::ReadTablePropertiesDirectly. Also fixes the fallback path in TableCache::GetTableReader to use the local fopts (with temperature and checksum) instead of the original file_options.
Added a kNoFileChecksumFuncName which is distinct from kUnknownFileChecksumFuncName:
- kUnknownFileChecksumFuncName ("Unknown"): We have FileMetaData for this file, and the metadata says no checksum was computed (no factory was configured when the file was written). This is a property of the file itself.
- kNoFileChecksumFuncName ("Unavailable"): We don't even have FileMetaData — we're opening this file in a context where there's no checksum metadata to propagate at all (e.g., SstFileDumper, SstFileReader, checksum generation). It's a property of the call site, not the file.
So the assertion file_checksum.empty() is correct for both, but for different reasons — one says "the file has no checksum," the other says "we have no idea about this file's checksum."
Reviewed By: pdillinger
Differential Revision: D92728944
fbshipit-source-id: 8fd34ea22ca87090b26d0a55c921f354f97f1ffc
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14337
## Context:
D91624185 changed `FilePrefetchBuffer::PollIfNeeded` from `void` to returning `Status`, correctly propagating `Poll` errors instead of silently swallowing them. A side effect is that when io_uring fails to initialize at runtime (e.g., sandcastle seccomp restrictions), `ReadAsync` returns `NotSupported` which now propagates through `PrefetchRemBuffers` and `HandleOverlappingAsyncData`, causing `PrefetchInternal` to return early before executing the synchronous `Read` that the caller actually depends on. This leaves iterators invalid with no recovery path — both in `db_stress` crash tests and in production. The filesystem advertises async IO support (`CheckFSFeatureSupport` passes), so the failure only surfaces at runtime when io_uring initialization fails. The prior behavior silently degraded to sync reads because `PollIfNeeded` swallowed the error.
## Changes
Add a sync fallback in `FilePrefetchBuffer::ReadAsync` — the single chokepoint for all async reads. When `reader->ReadAsync()` returns `NotSupported`, fall back to `reader->Read()` synchronously, populate the buffer inline, and return OK. Since `async_read_in_progress_` stays false, `PollIfNeeded` becomes a no-op (nothing to poll, data is already there). All callers — `PrefetchRemBuffers`, `HandleOverlappingAsyncData`, `PrefetchAsync` — work transparently without any per-site changes.
Reviewed By: archang19
Differential Revision: D93432284
fbshipit-source-id: daef185fc3535e347d182e75dd443ae921eeb495
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14340
`ReadBe64FromKey` pads its result with `val <<= (8 - len) * 8` to right-align partial reads. When the seek target's user key is shorter than shared_prefix_len, len is 0 and this becomes a shift by 64, which is undefined behavior for uint64_t. On x86 this happens to produce 0 (the correct result), but UBSan rightfully flags it. Guard the shift with `len > 0 && len < 8`.
Reviewed By: joshkang97
Differential Revision: D93435715
fbshipit-source-id: bab128e9a65ea18d401670268cbac77d45e11340
Summary:
FIFO crash tests fail on DB open when `fifo_compaction_max_data_files_size_mb` is randomly set to 100 or 500 MB, because `max_table_files_size` defaults to 1GB and the validation requires `max_data_files_size` >= `max_table_files_size` when non-zero. Cap `max_table_files_size` to `max_data_files_size` in db_stress when the latter is set. `max_table_files_size` is ignored at runtime when `max_data_files_size` is non-zero, so this only satisfies the validation constraint.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14341
Reviewed By: pdillinger
Differential Revision: D93503113
Pulled By: mszeszko-meta
fbshipit-source-id: 5c3e7c9b568661244c71c548cb0fe5e55472c0ca
Summary:
Add a new kv ratio based compaction picking algorithm in fifo compaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14326
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D93257941
Pulled By: xingbowang
fbshipit-source-id: fd2d0e1356c7b54682a1197475a1bd26cb45c9d4
Summary:
Interpolation search is an alternative algorithm to binary search, which performs better on uniformly distributed keys. Instead of binary search always computing the mid point of the left and right boundaries, interpolation search "interpolates" the mid point based on the distance to the target. Fortunately, we can re-use existing block format to support interpolation search.
For a given block, we compute the shared_prefix length of the first and last key. Interpolation search is usually done with numerical target values, so for a variable binary length key, we calculate the "value" as the first 8 non-shared bytes. This also means interpolation search would only really be effective for bytewise comparator (guarded via options validations).
#### Fallback to binary search
- if the the val(left_key) == val(right_key) then we fallback to classic binary search (to avoid divide by 0)
- interpolation search is significantly more computationally expensive than binary search, so when the search distance is small, we also fallback to binary search.
- if interpolation search does not make significant progress (i.e. reduces search space by more than half each iteration), we can assume data is non-uniform and fallback.
Interpolation search also performs best when there is minimal shortening, especially shortening of the last block, as it can heavily skew the distribution of the actual keys.
Note that each search algorithm is guaranteed to make progress because at each iteration the search space is guaranteed to be reduce by at least 1.
For now this change only applies to index block seeks, as data block seeks and other blocks do not have as many entries and would not require significant number of search rounds, but it could be easily extended to include that support.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14247
Test Plan:
Updated unit tests and crash test with new search option
### Benchmark
The default benchmark sets up keys in generally uniform distribution, so it was a good way to test performance improvements.
Setup: `./db_bench -benchmarks=fillseq,compact -index_shortening_mode=1`
#### Before this change
```
./db_bench -use_existing_db=true -benchmarks=readrandom -seed=1
readrandom : 2.899 micros/op 344973 ops/sec 2.899 seconds 1000000 operations; 38.2 MB/s (1000000 of 1000000 found)
```
#### After this change
Notice how key comparison counts are the same between the two.
```
./db_bench -use_existing_db=true -benchmarks=readrandom -seed=1 -index_search_type=binary_search
readrandom : 2.881 micros/op 347128 ops/sec 2.881 seconds 1000000 operations; 38.4 MB/s (1000000 of 1000000 found)
```
```
./db_bench -use_existing_db=true -benchmarks=readrandom -seed=1 -index_search_type=interpolation_search
readrandom : 2.609 micros/op 383209 ops/sec 2.610 seconds 1000000 operations; 42.4 MB/s (1000000 of 1000000 found)
```
With a non-uniform distribution, `i.e. index_shortening_mode=2`
```
./db_bench -use_existing_db=true -benchmarks=readrandom -seed=1 -index_search_type=binary_search
readrandom : 2.958 micros/op 338075 ops/sec 2.958 seconds 1000000 operations; 37.4 MB/s (1000000 of 1000000 found)
```
```
./db_bench -use_existing_db=true -benchmarks=readrandom -seed=1 -index_search_type=interpolation_search
readrandom : 5.502 micros/op 181750 ops/sec 5.502 seconds 1000000 operations; 20.1 MB/s (1000000 of 1000000 found)
```
Reviewed By: pdillinger
Differential Revision: D91063163
Pulled By: joshkang97
fbshipit-source-id: 151d6aa76f8713740b714de6e406aff40d28ccbc
Summary:
In follow-up to https://github.com/facebook/rocksdb/issues/14315
Remove obsolete code replaced by new Compressor/Decompressor interface:
* OLD_CompressData and OLD_UncompressData
* Individual compression/decompression functions (Snappy_*, Zlib_*, BZip2_*, LZ4_*, LZ4HC_*, XPRESS_*, ZSTD_Compress, ZSTD_Uncompress)
* CompressionInfo and UncompressionInfo classes
* UncompressionDict class
* compression::PutDecompressedSizeInfo and GetDecompressedSizeInfo
The only small refactoring in this change that is not pure code removal or movement is in blob_file_builder_test.cc.
Move some function implementations etc. from compression.h to compression.cc:
* CompressionTypeToString, CompressionTypeFromString, CompressionOptionsToString
* ZSTD_TrainDictionary (both overloads), ZSTD_FinalizeDictionary
* DecompressorDict::Populate
* Most compression library includes
Also cleaned up other includes of compression.h, which caused some other files to need new includes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14325
Test Plan: existing tests
Reviewed By: hx235
Differential Revision: D93120580
Pulled By: pdillinger
fbshipit-source-id: ab5c50db7379c0387a8c0e379642c9ea2799eae5
Summary:
**Context/Summary:**
Internal adoption has demonstrated stability and measurable improvements of this feature without much cost so we can turn it on by default. Eventually we'd like to remove this configuration and make this an expected behavior.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14329
Test Plan: Existing unit test
Reviewed By: mszeszko-meta
Differential Revision: D93210059
Pulled By: hx235
fbshipit-source-id: 04f77954e6624c8e60a2db030eb19eb341dd0fcf
Summary:
See https://github.com/facebook/rocksdb/issues/14240 which brought this to my attention. Here I've added range deletions and compactions to the format compatible test, and fixed or worked-around compatibility issues (likely longstanding).
The first fix was in Version::MaybeInitializeFileMetaData for an assertion failure simply from adding range deletions from some 5.x version.
The second fix is a broader work-around for older SST files with unreliable num_entries/num_range_deletions/num_deletions statistics in their table properties. We depend on them only for some paranoid checks for compaction, so in my assessment the best way to deal with those files is to exclude the paranoid checks when dealing with the files with unrelaible data. (Details in code comments.) The important part is that compacting old files is exceptionally rare, so we aren't really interefering with the paranoid checks doing thier job on an ongoing basis.
This depends on https://github.com/facebook/rocksdb/issues/14315 (just landed) because there is a remaining undiagnosed problem with some very early releases, but I'm not fixing that because its support is being dropped.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14323
Test Plan: test extended (ran locally excluding some releases)
Reviewed By: xingbowang
Differential Revision: D93032653
Pulled By: pdillinger
fbshipit-source-id: f90b32f30ba4764692e68d23705f42c778e0dc1d
Summary:
**Context/Summary**:
compaction_job_test does low level assertions on what keys were saved and to resume in https://github.com/facebook/rocksdb/commit/1e5fa69c99ac8765783f5ce8a3a065b08f5b08a7 before the integration of the feature is done in a separate PR https://github.com/facebook/rocksdb/commit/f7e4009de1d16421a254dd7e799dd91c522d832c. Such low-level test makes it difficult to assert data correctness, is hard to understand by being tied to implementation details.
Therefore they are now replaced with db-level tests with data correctness check, which is what we ultimately care out of those details. I also expand the test to cover wide column and TimedPut() which associates a key with write time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14191
Test Plan:
- Only test change; I also manually traced every test to ensure correct resumption point; also removing
```
if (c_iter->IsCurrentKeyAlreadyScanned()) {
return false;
}
```
correctly leads to the expected error with resumption at merge, single delete and deletion at bottom
Reviewed By: jaykorean
Differential Revision: D89492846
Pulled By: hx235
fbshipit-source-id: 6c6ab3cbd643ca1b15d049a062da2c76165ef9db
Summary:
... and remove some old code and tech debt in the process.
This is arguably a great milestone and precendent in RocksDB history as for the first time we are explicitly dropping support for the ability to read source-of-truth data in old formats. (We previously dropped support for reading some old bloom filters, but those are performance optimizers not source-of-truth. https://github.com/facebook/rocksdb/issues/10184) However, DBs written with default settings since release 4.6.0, which is very nearly 10 years ago, can still be read. And by using compaction with intermediate versions, there's an upgrade path going back to (AFAIK) early releases of LevelDB (from which RocksDB was forked).
Some detail:
* The magic number for LevelDB SST files (0xdb4775248b80fb57, most recently called kLegacyBlockBasedTableMagicNumber) now only exists in the code to provide a good error message and to test that good error message.
* There is some notable refactoring and renaming around format_version handling. This is a bit of a messy area of code because the footer code being shared between different table formats (block-based, plain, cuckoo) means format_version in the footer is in ways tied to all of them, but in other ways is just tied to block-based table where we have been making updates. Hopefully code comments keep this clear.
* Now that there are old format_versions we can't read (and can't write authoritatively in tests), I've needed to split out kMinSupportedFormatVersion into a constant for reads and for writes, currently the same at format_version=2. Comments describe how to update these in the future.
* The idea of versioning the compression format is basically going away, though we're keeping BuiltinV2 in places just because it's already there. There's lots of room in the BuiltinV2 schema to expand to new built-in compression types, or new ways of handling existing compression algorithms. CompressionManager with CompatibilityName gives users the power to customize compression without the need for versions tied to format_version.
Immediate follow-up:
* Clean up compression loose ends like OLD_Compress, OLD_Uncompress
Suggested follow-up:
* Update plain table builder to migrate to new footer version so that we can drop support for legacy footer. We have to be careful that the (likely untested) forward compatibility path I put in place a while back works (or fix it and wait a while) before dropping support for plain table with legacy footer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14315
Test Plan:
* Some tests updated / added
* A couple tests are obsolete: removed
* Also updated format compatible test, which now doesn't need to dig as far back into history building RocksDB.
Reviewed By: hx235
Differential Revision: D92577766
Pulled By: pdillinger
fbshipit-source-id: a23be846189d901ce087af4ca9a99cef18445cb7
Summary:
This could is triggering `-Wstring-conversion`, which presents as:
```
warning: implicit conversion turns string literal into bool: A to B
```
This is often a bug and what was intended. The most frequent cause is the code was:
```
void foo(bool) { ... }
void foo(std::string) { ... }
foo("this gets interpreted as a bool");
```
It is also possible the issue is innocuous as part of an assert:
```
assert(false && "this string is true, so the assertion is false");
EXPECT_FALSE("this string is true, so the expect fails");
```
in these cases the use is to "cute", so we modify the code to make it more obvious.
```
assert(false && "the compiler recognizes and doesn't complain about this pattern");
FAIL() << "much more obvious";
```
Differential Revision: D92886041
fbshipit-source-id: 6adfaa102f12e293491cc579ec92b48834d1d0a8
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14319
The test asserted h1->is_finished and h2->is_finished immediately after AbortIO({H0}), before calling Poll. This is invalid because AbortIO only guarantees that handles in its abort set are finalized. Non-aborted handles' CQEs may or may not be consumed during AbortIO depending on io_uring completion ordering. If H0's two CQEs (original read + cancel) arrive before H1/H2's CQEs, AbortIO breaks out of its wait loop without processing them. Move the H1/H2 is_finished assertions to after Poll, which correctly handles either case. Also remove the racy req_count checks for non-aborted handles since Poll does not increment req_count.
Reviewed By: jaykorean
Differential Revision: D92848827
fbshipit-source-id: 0c09b44ceada99877e8311cff799fa94f1056545
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14312
This could is triggering `-Wstring-conversion`, which presents as:
```
warning: implicit conversion turns string literal into bool: A to B
```
This is often a bug and what was intended. The most frequent cause is the code was:
```
void foo(bool) { ... }
void foo(std::string) { ... }
foo("this gets interpreted as a bool");
```
It is also possible the issue is innocuous as part of an assert:
```
assert(!"this string is true, so the assertion is false");
EXPECT_FALSE("this string is true, so the expect fails");
```
in these cases the use is to "cute", so we modify the code to make it more obvious.
```
assert(false && "the compiler recognizes and doesn't complain about this pattern");
FAIL() << "much more obvious";
```
Reviewed By: dmm-fb
Differential Revision: D92528316
fbshipit-source-id: 93fbb624e8731c4cdb559746b44c1aa71d786304
Summary:
To make CI consistent with internal meta clang version.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14311
Test Plan:
CI shows correct version
```
Successfully installed clang-format-21.1.2
clang-format version 21.1.2
```
Reviewed By: xingbowang
Differential Revision: D92535441
Pulled By: joshkang97
fbshipit-source-id: ea21ea97b13a35b286f0c2ce18b3f01ffbf49afd
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14301
When AbortIO was called with a subset of outstanding async read handles,
it would consume io_uring completions for handles NOT in the abort set
but fail to finalize them. This caused subsequent Poll calls on those
handles to hang forever waiting for completions that had already been
consumed.
The fix adds an `is_being_aborted` flag to Posix_IOHandle that is set
when submitting the cancel request. When processing completions in
AbortIO, handles with this flag wait for req_count==2 (original + cancel),
while handles without the flag are finalized immediately at req_count==1.
Also refactored the completion finalization logic into a shared
FinalizeAsyncRead() helper function used by both Poll and AbortIO.
Reviewed By: mszeszko-meta, archang19
Differential Revision: D92230883
fbshipit-source-id: e6d11e009a4930e5608459771990f6cf7d46d827
Summary:
This change adds the ability to open and operate on databases as TransactionDB in the ldb command-line tool.
New Command-Line Options
- --use_txn - Opens the database as a TransactionDB instead of a regular DB
- --txn_write_policy=<0|1|2> - Sets the transaction write policy:
- 0 = WRITE_COMMITTED (default)
- 1 = WRITE_PREPARED
- 2 = WRITE_UNPREPARED
Use Case
This is needed to inspect or modify databases that were created with WritePrepared or WriteUnprepared transactions, which require opening via TransactionDB::Open() rather than the regular DB::Open().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14304
Test Plan:
Tests (tools/ldb_test.py): Adds testTxnPutGet() covering:
- Basic put/get/delete with TransactionDB
- All three write policies
- Validation that --use_txn and --ttl are mutually exclusive
Reviewed By: pdillinger
Differential Revision: D92323195
Pulled By: anand1976
fbshipit-source-id: 0a62b8ea4e2985feed977fad72595d6fff75db09
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14306
GetLiveFilesStorageInfo crashes when called on a read-only RocksDB because
it calls FlushWAL(), which accesses logs_.back() on an empty deque.
Root cause: DBImplReadOnly overrides SyncWAL() to return NotSupported, but
does NOT override FlushWAL(). Read-only DBs have an empty logs_ deque
because they don't create WAL writers during recovery - there's nothing to
write, so no WAL infrastructure is initialized.
The reason SyncWAL was originally marked NotSupported is that these WAL
operations (SyncWAL syncs buffer to disk, FlushWAL flushes to OS buffer)
require an active WAL writer at logs_.back().writer. Since read-only DBs:
1. Cannot perform writes
2. Don't create WAL files for writing
3. Have an empty logs_ deque
...there's no WAL writer to sync or flush. The operations are semantically
meaningless, not just "forbidden write operations."
The fix adds a FlushWAL override matching the SyncWAL pattern. The caller
in db_filesnapshot.cc:403-405 already handles IsNotSupported() gracefully:
if (s.IsNotSupported()) { s = Status::OK(); }
Reviewed By: pdillinger
Differential Revision: D92419557
fbshipit-source-id: 7079071209b3c7be41a2c98c9b691e68bc031595
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14296
This could is triggering `-Wstring-conversion`, which presents as:
```
warning: implicit conversion turns string literal into bool: A to B
```
This is often a bug and what was intended. The most frequent cause is the code was:
```
void foo(bool) { ... }
void foo(std::string) { ... }
foo("this gets interpreted as a bool");
```
It is also possible the issue is innocuous as part of an assert:
```
assert(!"this string is true, so the assertion is false");
EXPECT_FALSE("this string is true, so the expect fails");
```
in these cases the use is to "cute", so we modify the code to make it more obvious.
```
assert(false && "the compiler recognizes and doesn't complain about this pattern");
FAIL() << "much more obvious";
```
Reviewed By: anand1976
Differential Revision: D92013593
fbshipit-source-id: 0b4e00339bef3f76fc5b9ad35e2383c5e4f828f9
Summary:
I don't think this option was ever useful. There was no compressed secondary cache compatibility issue that needed to accommodate compression format version 1. It was needlessly imported from legacy SST file formats. Version 1 is simply an inefficient format because it requires guessing the uncompressed size on decompression.
And as far as I know, we don't have any plans to make compressed secondary cache entries persistable across RocksDB versions. I.e. if persisting, we would simply tag the persistence layer with the version (perhaps major and minor) and throw out the cache whenever that changes. Then we don't have to deal with explicit schema versioning in persistenct caches. This is a workable approach because unlike SSTs, caches are not source-of-truth that need to survive version rollback.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14302
Test Plan: existing tests
Reviewed By: anand1976
Differential Revision: D92315003
Pulled By: pdillinger
fbshipit-source-id: 0b82cfdbd92bcd2b8fbddd6586824f53c88069c4
The internal and external repositories are out of sync. This Pull Request attempts to brings them back in sync by patching the GitHub repository. Please carefully review this patch. You must disable ShipIt for your project in order to merge this pull request. DO NOT IMPORT this pull request. Instead, merge it directly on GitHub using the MERGE BUTTON. Re-enable ShipIt after merging.
fbshipit-source-id: 08b287a3f343f6ac5872c2a059d91d1bed9ff0a8
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14286
Add the db_c leanup target which can be used by CI test scripts to delete the db on failure. The db_crashtest.py doesn't automatically delete on error.
Reviewed By: jaykorean
Differential Revision: D91912877
fbshipit-source-id: d36ec0896fba64faaafe055d8673e437e85d0c3a
Summary:
* Add CLAUDE.md This CLAUDE.md is generated through the analysis of 9,012 commits from the RocksDB repository since 2016. It aggregate the commits into 8 major components based on the component it is changing. It selected top 100 most complex PR based on line of changes to collect the code review feedbacks. For each PR, we collected:
• PR title and description to understand the change context
• Files changed to identify affected components and measure complexity
• Inline code review comments from reviewers
• General review summaries and approval/change request feedback
The feedback was then categorized by RocksDB component and analyzed for recurring themes, patterns, and best practices.
This CLAUDE.md file could be used for guiding code generation and code review.
* Optimize tooling for claude code Add make check-progress and format-auto targets for automation
Add machine-parseable progress reporting for `make check` to support automated monitoring tools like Claude Code:
- Add `build_tools/check_progress.sh` script that outputs JSON progress
- Add `make check-progress` target to poll build/test progress
- Detects phases: compiling -> linking -> generating -> testing
- Reports failed tests with exit codes, signals, and log output
- Limits to 10 failures with last 50 lines of output each
Also add non-interactive formatting support:
- Add `-y` flag to format-diff.sh for auto-apply without prompts
- Add `make format-auto` target for CI/automation use
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14293
Test Plan: Local build
Reviewed By: pdillinger
Differential Revision: D92085763
Pulled By: xingbowang
fbshipit-source-id: ba122a4ff51087aec5c06bab804edfee34e13880
Summary:
Change adds `log_data_` function callback for when iterating over a `WriteBatch`. Previously only the `Put`, `Delete`, `Merge` operations were called into when iterating over an `WriteBatch` (and their `*_cf` equivalent through a different `WriteBatch::Handler` implementation).
To maintain backwards compatibility, previously exported function definitions remain the same, but new functions are exported for different languages to use the `LogData` callback on an iteration.
### Background
Hi - we use the [`rust-rocksdb`](https://github.com/rust-rocksdb/rust-rocksdb) bindings to work with `rocksdb` at Stripe. We are starting to make small contributions https://github.com/facebook/rocksdb/pull/14183 & https://github.com/facebook/rocksdb/pull/14136 and this adds on top of it. I saw that the `PutLogData` method is already exported for a `WriteBatch`, but there's no way to consume that. This change allows us to consume that information (with a follow up change on the [`rust-rocksdb`](https://github.com/rust-rocksdb/rust-rocksdb) repo.).
Thanks for your time looking into this. Previously we had trouble with meta's internal linters - I am happy to make appropriate change if something like that pops up again.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14245
Reviewed By: archang19
Differential Revision: D92069503
Pulled By: jaykorean
fbshipit-source-id: a4a3c885462f641c8df9e3401a0e4c1d38871c6f
Summary:
This diff introduces the IODispatcher into the BlockBasedTableIterator. This replaces much of the prefetch logic with the logic found in IODispatcher.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14255
Test Plan:
I ran the following benchmark, %change is within noise tolerance. There shouldn't be any large performance improvement with this change, more so there should also not be any performance degradation.
MultiScan Benchmark: Current Branch vs Main
Configuration:
- Threads: 4
- Ranges per scan: 10
- Stride: 5000
- Seek nexts: 100
- Cache: Cold (dropped before each run)
- Runs: 3
Results:
│ Mode │ Main (ops/sec) │ Current (ops/sec) │ Change │
│ Sync │ 8,901 │ 9,032 │ +1.5% │
│ Async │ 11,297 │ 11,947 │ +5.8% │
I further run db_stress test
```
make -j32 -f crash_test.mk J=32 blackbox_crash_test
```
Against my local machine for 60 minutes, on local flash, with async-io for multiscans always on.
Reviewed By: anand1976
Differential Revision: D91705195
Pulled By: krhancoc
fbshipit-source-id: acf2f944e8b715e99384c8cee79f8d241eadf5b8
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14282
Previously, `PollIfNeeded` returned `void` and silently ignored errors from
`fs_->Poll()` by calling `PermitUncheckedError()`. This could lead to silent
data corruption or unexpected behavior when Poll operations fail.
This diff changes `PollIfNeeded` to return `Status` and properly propagate
Poll errors to callers. When Poll fails:
1. The IO handle is cleaned up via `DestroyAndClearIOHandle`
2. The error status is returned to the caller
3. Callers (`HandleOverlappingAsyncData` and `PrefetchInternal`) now check
and propagate this error
Also adds a `TEST_SYNC_POINT_CALLBACK` to allow tests to inject Poll errors.
Reviewed By: anand1976
Differential Revision: D91624185
fbshipit-source-id: 8dd0ee6588ed1ce4bf080bcf857b778c5140ccf5
Summary:
This adds a new public API to allow applications to abort all running compactions and prevent new ones from starting. Unlike DisableManualCompaction() which only pauses manual compactions and waits for them to finish naturally, AbortAllCompactions() actively signals running compactions (both automatic and manual) to terminate early and waits for them to complete before returning.
The abort signal is checked periodically during compaction (every 100 keys), so ongoing compactions abort quickly. Any output files from aborted compactions are automatically cleaned up to prevent partial results from being installed.
This is useful for scenarios where applications need to quickly stop all compaction activity, such as during graceful shutdown or when performing maintenance operations.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14227
Test Plan:
- Unit tests in db_compaction_abort_test.cc cover various abort scenarios including: abort before/during compaction, abort with multiple subcompactions, nested abort/resume calls, abort with CompactFiles API, abort across multiple column families, and timing guarantees
- Updated compaction_job_test.cc to include the new parameter
Reviewed By: anand1976
Differential Revision: D91480994
Pulled By: xingbowang
fbshipit-source-id: 36837971d8a540cd34d3ec28a78bc94b582625b0
Summary:
When building a Release on Windows RTTI is not available, so asserts that use dynamic_cast need to be disabled
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14280
Reviewed By: nmk70
Differential Revision: D91807791
Pulled By: mszeszko-meta
fbshipit-source-id: e29c19c757bcd076a1f09ed40b306bb50ba9e882
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14278
Remove dead code from `BlobDBImpl`:
- `debug_level_ `member and associated unreachable debug logging
- `CopyBlobFiles()` - private method that was never called
- `FileDeleteOk_SnapshotCheckLocked()` - declared but never implemented
- `RemoveTimerQ()` - declared but never implemented
Remove unused includes:
- rocksdb/wal_filter.h from blob_db_impl.h
- rocksdb/utilities/transaction.h from blob_db_impl.cc
- table/meta_blocks.h from blob_db_impl.cc
- util/random.h from blob_db_impl.cc
Remove from BlobFile:
- `GetColumnFamilyId()` - declared/implemented but never called
Reviewed By: xingbowang
Differential Revision: D91089144
fbshipit-source-id: d9bce24122b3bb790644fe4e51ce4403c77a1abf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14277
Remove `GetBlobDBOptions()` and `SyncBlobFiles()` from the public `BlobDB` interface. These methods were only used internally or in tests and are not needed by any production code. `GetBlobDBOptions()` is now replaced by storing bdb_options_ as a member in the test class. `SyncBlobFiles()` is moved to private in `BlobDBImpl` since it's only called internally. Also remove unused `kDeleteCheckPeriodMillisecs` constant.
Reviewed By: xingbowang
Differential Revision: D91089111
fbshipit-source-id: 9c92b6d9563cf241c69d8880b418e8bcb7acb6c5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14276
No production user of Legacy overrides the default value of `bytes_per_sync`. Replace the option with a constant `kBytesPerSync` to further reduce legacy blob db customizability / configuration surface.
Reviewed By: xingbowang
Differential Revision: D91089096
fbshipit-source-id: 162df65646a4f3a3fab3586cf6ff223e1917d86e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14275
No one ever sets `blob_dir` to a non-default value. Replace the configurable `blob_dir` option with a constant `kBlobDirName`. This simplifies the code and further reduces the configurability surface.
Reviewed By: xingbowang
Differential Revision: D91089039
fbshipit-source-id: 7d82e86415cc4bc89a7fe1399c29d4cc3058d1de
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14273
The `path_relative` option in `BlobDBOptions` was never used in practice - all
production deployments use the default value of `true` (relative path). The
absolute path mode (`path_relative = false`) was essentially unsupported:
- `GetLiveFiles()` returned `NotSupported` for absolute paths
- `GetLiveFilesMetaData()` had an assertion that would crash for absolute paths
This change removes the option and simplifies the code to always use relative
paths for the blob directory.
Changes:
- Remove `path_relative` field from `BlobDBOptions`
- Simplify `blob_dir_` construction in `BlobDBImpl` constructor
- Simplify path construction in `DestroyBlobDB()`
- Remove `NotSupported` check in `GetLiveFiles()`
- Remove assertion in `GetLiveFilesMetaData()`
- Remove logging of `path_relative` in `Dump()`
- Remove redundant `path_relative = true` in tests
Reviewed By: xingbowang
Differential Revision: D91089016
fbshipit-source-id: 947b129e405a315b94ac73bc48b23103ba12d73b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14272
All production deployments have the hardcoded (non-configurable) default for`garbage_collection_cutoff = 0.25`. This change removes the configurable option and replaces it with a fixed constant `kGarbageCollectionCutoff = 0.25`, simplifying the configuration surface.
Changes:
- Remove `garbage_collection_cutoff` from `BlobDBOptions`
- Add `kGarbageCollectionCutoff` constant (0.25) in blob_db_impl.cc
- Remove `--blob_db_gc_cutoff` flag from db_bench tool and db_stress
- Update tests to work with the fixed cutoff value
Reviewed By: xingbowang
Differential Revision: D91088998
fbshipit-source-id: 820fc7f1ad4c3fe8a15f22a92cd53fb96c56c6e1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14270
Legacy BlobDB's inline values feature (storing small values directly in the LSM tree via `min_blob_size` threshold) is unused in production - all
deployments use `min_blob_size = 0`. This removes the functionality entirely.
Changes:
- Remove `min_blob_size` from `BlobDBOptions`
- Remove `IsInlined()` check from compaction filter (dead code path)
- Remove inline-related statistics (`BLOB_DB_WRITE_INLINED*`)
- Remove `InlineSmallValues` test
- Update stale comments referencing inlined data
Reviewed By: xingbowang
Differential Revision: D91088985
fbshipit-source-id: ec67848ece1a7dc071ca8e8a17faebb435394733
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14268
FIFO eviction is unused in production, making this dead code that adds complexity.
Core changes:
- Remove `is_fifo` from `BlobDBOptions`
- Remove `CheckSizeAndEvictBlobFiles` and related methods
- Remove `fifo_eviction_seq` and `evict_expiration_up_to` from `BlobCompactionContext`
CLI tool cleanup:
- Remove `--blob_db_is_fifo` flag from db_bench
- Remove stale FIFO eviction comments
Tests:
- Remove FIFO-related tests (`FIFOEviction_*`, `FilterForFIFOEviction`)
Note: TTL-based expiration (`EvictExpiredFiles`) is preserved as it handles blob file cleanup based on TTL, which is separate from FIFO eviction.
Reviewed By: xingbowang
Differential Revision: D91088968
fbshipit-source-id: 123df98d1132095cef15473b76011de030c5df34
Summary:
Work around a warning/linter false positive related to the use of string::insert. The code in question is legal C++, but GCC 12's libstdc++ implementation of string::insert internally uses memcpy, which can trigger undefined behavior warnings when the source and destination overlap.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14265
Reviewed By: pdillinger
Differential Revision: D91594561
Pulled By: mszeszko-meta
fbshipit-source-id: faa1487aba11a6581bf9ac8eb89442b6e4120427
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14257
Removes the 'unused' `PutUntil` API and updates `Put/PutWithTTL` to inline the previous implementation. Test helpers are updated to use `PutWithTTL` with computed TTL values instead.
Reviewed By: xingbowang
Differential Revision: D90900841
fbshipit-source-id: c6ab89fe32773f426b0bedc706bf5a2683ec31cf
Summary:
When running make check on aarch64, hash_test reports an integer underflows:
util/math.h:44:46: runtime error: signed integer overflow:
-2147483648 - 1 cannot be represented in type 'int'
util/math.h:44:46: runtime error: signed integer overflow:
-9223372036854775808 - 1 cannot be represented in type 'long long'
util/math.h:44:46: runtime error: signed integer overflow:
-9223372036854775808 - 1 cannot be represented in type 'long'
The issue is when BottomNBits(int32 value, 31) does not use BMI2, it executes the following:
return static_cast<T>(v & ((T{1} << nbits) - 1));
For int32_t, (1 << 31) is the minimum value, and -1 is an integer underflow. The fix is to cast T to an unsigned type and use that for the bit manipulation.
I used Compiler Explorer to verify that this still compiles to the BZHI instruction mentioned in the comment with -march=x86-64-v3: https://godbolt.org/z/8bcTE8xbf
To reproduce these errors on x86-64, disable the BMI code path:
```
USE_CLANG=1 PORTABLE=x86-64-v2 LDFLAGS=-fsanitize=undefined CXXFLAGS=-fsanitize=undefined make -j20 hash_test
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14231
Reviewed By: mszeszko-meta
Differential Revision: D91353147
Pulled By: pdillinger
fbshipit-source-id: 64cc191ccb9ecba20c260fab759e8881e30d2352
Summary:
Update HISTORY, version number, format compatible test, and folly version
folly build now depends on libaio
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14259
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D91356493
Pulled By: pdillinger
fbshipit-source-id: 9d85960c647758d5cb33e3910e714e2f7785fd06
Summary:
Seeing many errors like this
```
Iterator diverged from control iterator which has value ...
iterator is not valid with status: IO error: Req failed: Unknown error -14
VerifyIterator failed. Control CF default
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14263
Test Plan: CI
Reviewed By: archang19
Differential Revision: D91478886
Pulled By: pdillinger
fbshipit-source-id: 94b955b6ecdb7a3cab39dac8e7b0d1047d49a0bb
Summary:
... in addition to those derived from samples. This could be useful when trade-offs favor an offline trained dictionary that's good for the whole work load, which can involve heavy-weight training, vs. on-the-fly training on samples for each file, which has limitations.
This involves some breaking changes to some deeper parts of the new compression API. I'm not concerned about performance because this doesn't touch the per-block parts of the API, just the per-file parts.
Bonus: change to
CompressionManagerWrapper::FindCompatibleCompressionManager to implement what is likely the preferred behavior.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14253
Test Plan: unit test included
Reviewed By: hx235
Differential Revision: D91082208
Pulled By: pdillinger
fbshipit-source-id: 1442db65e15c9435437204c19787c96f7a40a207
Summary:
Clearing DB dir for crash test is currently a hodgepodge of
1. Caller of db_crashtest.py maybe tries to clear the dir
2. db_crashtest.py tries to clear the dir in get_dbname() (but ignoring failure)
3. db_crashtest.py passes --destroy_db_initially to some db_stress calls as needed
4. db_crashtest.py tries to clear the dir between some db_stress calls
5. db_crashtest.py tries to clear the dir after everything is done and successful (no artifacts to investigate or save) (but ignoring failure)
6. Try to add more uniqueness to the directory from https://github.com/facebook/rocksdb/issues/14249
This change reverts or replaces 2, 4, 5, and 6 by doubling-down on (expanding) 3 and a small variant of it:
* crash_test.mk passes --destroy_db_initially=1 so that the first run of db_stress clears the db dir.
* After each db_stress invocation, db_crashtest.py resets destroy_db_initially=0 so that the next invocation reuses the same DB, except in cases where there is an incompatibility that requires a fresh DB (from cases 3 and 4 above).
* On success, uses new `db_stress --destroy_db_and_exit` option to clean up the DB dir without needing a custom cleanup_cmd (now ignored)
Note that although case 1 is likely obsolete, it is out of control of an open source PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14254
Test Plan: some manual runs
Reviewed By: xingbowang
Differential Revision: D91164731
Pulled By: pdillinger
fbshipit-source-id: 0a66c8c0e130c9eeacc55af411a18a09bc9debdf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14248
### Overview
This diff introduces the addition of multi-scan statistics to RocksDB, enhancing the database's ability to monitor and analyze performance during multi-scan operations.
### Key Changes
#### Implemented Multi-Scan Statistics
The following statistics were implemented to provide deeper insights into multi-scan operations:
- **MULTISCAN_PREPARE_MICROS**: Measures the time (in microseconds) spent preparing for multi-scan operations.
- **MULTISCAN_BLOCKS_PER_PREPARE**: Tracks the number of blocks processed per multi-scan prepare operation.
- **Wasted Prefetch Blocks Count**: Counts the number of prefetched blocks that were not used (i.e., wasted) if the iterator is abandoned before accessing them.
- **MULTISCAN_TOTAL_BLOCKS_SCANNED**: Tracks the total number of blocks scanned during all multi-scan operations.
- **MULTISCAN_TOTAL_KEYS_SCANNED**: Measures the total number of keys scanned across all multi-scan operations.
- **MULTISCAN_TOTAL_MICROS**: Captures the total time (in microseconds) spent in multi-scan operations.
- **MULTISCAN_PREFETCHED_BLOCKS**: Counts the number of blocks that were prefetched during multi-scan operations.
- **MULTISCAN_USED_PREFETCH_BLOCKS**: Tracks the number of prefetched blocks that were actually used during multi-scan operations.
### Impact
This diff provides more fine-grained statistics for multi-scan operations, allowing developers and users to better understand and optimize the performance of their RocksDB instances.
Reviewed By: krhancoc
Differential Revision: D91053297
fbshipit-source-id: 7158741b9f026c0b5ce8ba1264dbd137e7fe985d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14252
Fixed a bug in PosixFileSystem::AbortIO that could cause an infinite hang
when aborting multiple concurrent async IO handles.
The bug occurred in the completion processing loop: when an io_uring
completion arrived for a handle other than the one currently being waited
for (io_handles[i]), the code would increment that handle's req_count but
only mark it as finished if it also matched io_handles[i]. This meant
completions for other handles were consumed but those handles were never
marked as finished.
Later, when iterating to those handles, the code would enter
io_uring_wait_cqe expecting more completions, but they had already been
consumed - causing an infinite hang.
The fix aligns AbortIO's completion handling with what Poll() already does:
mark handles as finished whenever their completions arrive, regardless of
which handle we're currently waiting for in the outer loop. Only the break
statement remains conditional on matching io_handles[i].
Reviewed By: anand1976
Differential Revision: D91070044
fbshipit-source-id: 47faf5f0df3e26a2aa83444bbac623f43f560933
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14251
The AbortIO API documentation incorrectly stated that the callback
should NOT be called for aborted io_handles. However, the actual
implementation in fs_posix.cc does invoke the callback with
IOStatus::Aborted() status after cancelling requests:
```
// fs_posix.cc:1252-1260
if (posix_handle->req_count == 2 &&
static_cast<Posix_IOHandle*>(io_handles[i]) == posix_handle) {
posix_handle->is_finished = true;
FSReadRequest req;
req.status = IOStatus::Aborted();
posix_handle->cb(req, posix_handle->cb_arg);
break;
}
```
This change corrects the documentation to match the actual behavior
in RocksDB.
Reviewed By: anand1976
Differential Revision: D91073466
fbshipit-source-id: 47ae14a09e9386cc68049ca272d6b712f5a9bed7
Summary:
Since it's been > 6 months and we have production uses, migrate to fv=7 by default. One unit test needed an update for the change to table properties with fv=7.
On making this change, PresetCompressionDictTest tests detected extra memory usage by decompressing LZ4 with dictionary compression. This turned out to be a bug in `std::find` usage that led to using the ZSTD-optimized decompressor (with digested dictionary usage) in cases where it is not needed. I've fixed the bug and improved the unit tests that found the bug.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14239
Test Plan: existing tests, including format compatible CI job (updated, and run locally with SHORT_TEST=1)
Reviewed By: hx235
Differential Revision: D90728697
Pulled By: pdillinger
fbshipit-source-id: 8f1a0e9ca59a88c18eaa4cdfdea00309175ce30a
Summary:
Some of the stress tests script run tests multiple times with TEST_TMPDIR set. When TEST_TMPDIR is set, the db directory is a fixed string. This caused the same DB directory was reused across db_crashtest.py script run. Typically, the DB folder is cleaned up after db_crashtest.py complete. But sometimes, the clean up command could fail. This caused the DB folder to be reused across different db_crashtest.py runs. Meantime, each db_crashtest.py run would randomize some of the parameters. This caused different parameters to be used with same DB directory, violating some of the assumption such as use_put_entity_one_in parameter to be not changed between runs. This change added a suffix to DB directory, so that each db_crashtest.py script run would generate a unique DB directory, which prevents the clean up failure issue causing test flaky.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14249
Test Plan:
Stress test local run
```
TEST_TMPDIR=/tmp/aaa /usr/local/bin/python3 -u tools/db_crashtest.py --stress_cmd=./db_stress --cleanup_cmd='' --simple blackbox --duration 15 --interval 10
>>> Running db_stress with pid=113810: ./db_stress ... --db=/tmp/aaa/rocksdb_crashtest_blackbox_6967584463401575611 ...
```
Reviewed By: hx235
Differential Revision: D91069655
Pulled By: xingbowang
fbshipit-source-id: 327fc3cd0d8e3ef4b49e182e21bcd91a10647710
Summary:
Problem
The TEST_WaitForCompact in
TimestampCompatibleCompactionTest.UdtTombstoneCollapsingTest would sometimes
run forever, indicating an infinite compaction loop.
Issue https://github.com/facebook/rocksdb/issues/14223
Root Cause
In ComputeBottommostFilesMarkedForCompaction(), files were marked for
bottommost compaction based only on the condition largest_seqno <
oldest_snapshot_seqnum. However, for User-Defined Timestamps (UDT) columns,
compaction can only zero sequence numbers when the file's maximum timestamp is
below full_history_ts_low.
When timestamps were above this threshold:
1. File gets marked for compaction (seqno condition met)
2. Compaction runs but cannot zero seqno (timestamp condition not met)
3. Output file immediately gets re-marked for compaction
4. Infinite loop
Solution
Added timestamp range tracking to FileMetaData and updated the marking logic to
check timestamps before marking files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14228
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D90586045
Pulled By: xingbowang
fbshipit-source-id: addfa4f988db8c87fb513a1bf58ee54623a6c210
Summary:
Surface async read errors instead of asserting on them. This makes it easier to debug stress test failures. Async reads can fail for legitimate reasons, such as fs errors.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14244
Reviewed By: hx235
Differential Revision: D90878515
Pulled By: anand1976
fbshipit-source-id: 6335d4b06ddf250b26842ce94e3f5263356b2695
Summary:
These will be useful for qualifying non-tiered workloads for tiered storage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14243
Test Plan:
unit test included
I'm not concerned about performance because this fits pretty nicely into some existing code and only adds overhead when (expensive) IOs are done.
Reviewed By: jaykorean
Differential Revision: D90870348
Pulled By: pdillinger
fbshipit-source-id: 984411123bcd54c249a949da813ff04fedacc6a4
Summary:
To move away from OLD_CompressData / OLD_UncompressData. Also improved some error/warning messages.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14241
Test Plan: manual tests showing similar performance, runs with ASAN/UBSAN to check for issues
Reviewed By: hx235
Differential Revision: D90793708
Pulled By: pdillinger
fbshipit-source-id: e0655f7bed8d85e5ea110167dca73c6664f7465b
Summary:
Trying to get rid of uses of OLD_CompressData / OLD_UncompressData. Some performance optimizations and corrections for better accounting also.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14230
Test Plan:
* exanded unit test to be more complete / rigorous
* manual before-and-after db_bench runs with the option, seeing table properties as expected
Reviewed By: hx235
Differential Revision: D90545476
Pulled By: pdillinger
fbshipit-source-id: 2f7c577574bcc4b2acafa002761ec1cad7fdb093
Summary:
as part of the effort to get rid of OLD_CompressData and OLD_UncompressData and the old implementations in compression.h.
It's unfortunate the the existing blob file schema doesn't allow storing blobs uncompressed when the compressed version is larger, so we have to work around that.
Note that use of GrowableBuffer in place of std::string is intended to avoid the potential performance overhead of zeroing out memory before overwriting it.
Also includes some cleanup of includes
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14234
Test Plan:
some unit test updates as needed. Crash test covers integrated blob support.
I'm not too concerned about performance, as until a future schema change, this code is committing the grave performance error of storing compressed data larger than uncompressed.
Reviewed By: mszeszko-meta, hx235
Differential Revision: D90544049
Pulled By: pdillinger
fbshipit-source-id: 2f2ed16de63990b797cc06c8dad36b5869dac302
Summary:
Deadlock or timeout is possible in TestPut, when TestMultiGet was executed at the same time, because it executes MaybeAddKeyToTxnForRYW, which writes to the same key space but does not acquire stress test level mutex. Therefore, RocksDB could return deadlock error.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14235
Test Plan: Stress test
Reviewed By: hx235
Differential Revision: D90621772
Pulled By: xingbowang
fbshipit-source-id: eb808193ded06b69a8161320f88d5ba4e20b4901
Summary:
Folly download dependencies directly from external source. Sometimes, this could fail due to external website instability. To solve this, we added github cache to cache the dependencies. We also added a python script to try different sources during download to reduce the chance of failure.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14226
Test Plan: github CI
Reviewed By: krhancoc, archang19
Differential Revision: D90343051
Pulled By: xingbowang
fbshipit-source-id: 3faad6aaa6c1bfd361b9e405c298856cd64bf457
Summary:
Currently to set options for multiple CFs, the caller must repeatedly call SetOptions() for each CF. This in turn serializes the entire options file each time. This PR exposes a new API that allows SetOptions to be called on multiple CFs at once, thus only paying the OPTIONS file serialization once.
Also added a new unit test for SetOptions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14201
Reviewed By: pdillinger
Differential Revision: D89735181
Pulled By: joshkang97
fbshipit-source-id: 9b7a721b7e8769b653243b1581678ffd05d038e8
Summary:
This diff introduces the IO Dispatcher, which will be used to simplify the code path for MultiScan, while further providing a centralized place to enact policy on how MultiScan is done (i.e., limit memory usage and pinned buffers for example). Right now this diff only encapsulates the functionality done during the Prepare of MultiScan.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14135
Reviewed By: anand1976
Differential Revision: D87837261
Pulled By: krhancoc
fbshipit-source-id: 2698910ade02bc3d182413ae07ce69fe7abb7ec5
Summary:
As compaction scheduling is not deterministic, the existing check is too strict sometimes, causing test to be flaky.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14220
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D90143556
Pulled By: xingbowang
fbshipit-source-id: 6780423c63324a4b20fc8b8ccac2051a094c9f4a
Summary:
My PR https://github.com/facebook/rocksdb/issues/14195 regressed a case in which db_crashtest.py calling db_stress with --destroy_db_initially=1 could lead to dbname directory being nonexistant for subsequent calls to gen_cmd -> finalize_and_sanitize -> is_direct_io_supported which would fail in creating a temporary file. Fix this (and clean up existing related code) using os.makedirs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14219
Test Plan: I don't have a good reproducer for the error but some manual testing indicates this change is at least safe
Reviewed By: virajthakur
Differential Revision: D90138248
Pulled By: pdillinger
fbshipit-source-id: 0ed6524cd50f8632346a8583f26bf1f4941817ce
Summary:
* Some existing commentary and motivation around my atomic wrappers in atomic.h was based on a misreading of documentation. seq_cst *is* a safe substitute for acq_rel in all cases. I still like having a distinct type for RelaxedAtomic (as folly does) and a wrapper also for other cases to avoid readability traps like implicit conversion and implicit memory order. This PR is only comment changes and renaming.
* Create a blog post about bit fields API to help with lock-free (and low-lock) programming.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14213
Test Plan: esiting tests
Reviewed By: xingbowang
Differential Revision: D89971581
Pulled By: pdillinger
fbshipit-source-id: 9bd1181d692258d668189c2da8bd0e5d98fd6230
Summary:
This change builds on https://github.com/facebook/rocksdb/issues/14027 and https://github.com/facebook/rocksdb/issues/13965 to complete migration
of the HyperClockCache implementation to using the hygienic BitFields API.
No semantic change in the implementation details is intended, just
greatly improving readability and safety of the code while maintaining
the same performance.
In more detail,
* Refactor the main metadata atomic for each slot in an HCC table into
SlotMeta using BitFields.
* Extended BitFields APIs with some additional features, and renamed
BlahTransform classes to BlahTransformer to resolve potential naming
conflicts with member functions to create them.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14154
Test Plan:
for correctness, mostly existing tests. but also added tests
for new BitFields features. I especially ran local TSAN whitebox crash
test extensively which caught a couple of refactoring errors.
For performance, I verified with release builds of cache_bench, using
default options, that there was no noticeable/consistent difference
after all these HCC migrations vs. backing them out. That test was with
GCC 11 and -O2, which is a reasonable baseline for expected compiler
optimizations.
Reviewed By: xingbowang
Differential Revision: D87960540
Pulled By: pdillinger
fbshipit-source-id: e0257b7fea8a5c7709daef18911959201ce4e0f3
Summary:
This bug caused seqno to be incorrectly zeroed when UDT is enabled. This is one of the contributing factor that caused tombstones to be accumulated at bottommost level, causing high space amp.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14207
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D89826564
Pulled By: xingbowang
fbshipit-source-id: 62ab1e37c36ae1ed95f26213c97a591a17e962a6
Summary:
## Context
1. OpenAndCompact required CompactionServiceOptionsOverride
2. Currently there are no C APIs to create CompactionServiceOptionsOverride
## Changes
1. Create C API for compactionServiceOptionsOverride
2. Create helper function to create compactionServiceOptionsOverride from Options. This was added in because The C API lacks getter methods for non-serializable options (comparator, table_factory, etc.). Without this, users would need to maintain separate references to all these options just to pass them to the override. If the user need to create a new comparator or table factory then C API for compactionServiceOptionsOverride already as the setters for the same.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14183
Reviewed By: hx235
Differential Revision: D89690005
Pulled By: jaykorean
fbshipit-source-id: efe8211feec9d144b32be0f5e66c8cf8bde8dac0
Summary:
Let db_crashtest.py work with TEST_TMPDIR on remote filesystem, by infering whether it's remote from the env_uri argument. Note that some other paths passed to db_stress are local paths and we can't reuse TEST_TMPDIR for those cases when it's remote.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14195
Test Plan: public and private CI
Reviewed By: archang19
Differential Revision: D89590246
Pulled By: pdillinger
fbshipit-source-id: db6eb9c16d4e76617183780747353c798cc9bef6
Summary:
Causing failures and not yet supported. Also putting a note in db.h about the combination being unsupported.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14189
Test Plan: started up blackbox_crash_test_with_ts many times and checked command line to be confident it's excluded.
Reviewed By: hx235
Differential Revision: D89297971
Pulled By: pdillinger
fbshipit-source-id: c5134351d9ecb37879c7e3319c17dd9228d7f12a
Summary:
`PosixRandomAccessFile::MultiRead` was introduced in Dec 2019 in https://github.com/facebook/rocksdb/pull/5881. Subsequently, 2 years after, we introduced the `PosixRandomAccessFile::ReadAsync` API in https://github.com/facebook/rocksdb/pull/9578, which was reusing the same `PosixFileSystem` IO ring as `MultiRead` API, consequently writing to the very same ring's submission queue (without waiting!). This 'shared ring' design is problematic, since sequentially interleaving `ReadAsync` and `MultiRead` API calls on the very same thread might result in reading 'unknown' events in `MultiRead` leading to `Bad cqe data` errors (and therefore falsely perceived as a corruption) - which, for some services (running on local flash), in itself is a hard blocker for adopting RocksDB async prefetching ('async IO') that heavily relies on the `ReadAsync` API. This change aims to solve this problem by maintaining separate thread local IO rings for `async reads` and `multi reads` assuring correct execution. In addition, we're adding more robust error handling in form of retries for kernel interrupts and draining the queue when process is experiencing terse memory condition. Separately, we're enhancing the performance aspect by explicitly marking the rings to be written to / read from by a single thread (`IORING_SETUP_SINGLE_ISSUER` [if available]) and defer the task just before the application intends to process completions (`IORING_SETUP_DEFER_TASKRUN` [if available]). See https://man7.org/linux/man-pages/man2/io_uring_setup.2.html for reference.
## Benchmark
**TLDR**
There's no evident advantage of using `io_uring_submit` (relative to proposed `io_uring_submit_and_wait`) across batches of size 10, 250 and 1000 simulating significantly-less, close-to and 4x-above `kIoUringDepth` batch size. `io_uring_submit` might be more appealing if (at least) one of the IOs is slow (which was NOT the case during the benchmark). More notably, with this PR switching from `io_uring_submit_and_wait` -> `io_uring_submit` can be done with a single line change due to implemented guardrails (we can followup with adding optional config for true ring semantics [if needed]).
**Compilation**
```
DEBUG_LEVEL=0 make db_bench
```
**Create DB**
```
./db_bench \
--db=/db/testdb_2.5m_k100_v6144_16kB_LZ4 \
--benchmarks=fillseq \
--num=2500000 \
--key_size=100 \
--value_size=6144 \
--compression_type=LZ4 \
--block_size=16384 \
--seed=1723056275
```
**LSM**
* L0: 2 files, L1: 5, L2: 49, L3: 79
* Each file is roughly ~35M in size
### MultiReadRandom (with caching disabled)
Each run was preceded by OS page cache cleanup with `echo 1 | sudo tee /proc/sys/vm/drop_caches`.
```
./db_bench \
--use_existing_db=true \
--db=/db/testdb_2.5m_k100_v6144_16kB_LZ4 \
--compression_type=LZ4 \
--benchmarks=multireadrandom \
--num= **<N>** \
--batch_size= **<B>** \
--io_uring_enabled=true \
--async_io=false \
--optimize_multiget_for_io=false \
--threads=4 \
--cache_size=0 \
--use_direct_reads=true \
--use_direct_io_for_flush_and_compaction=true \
--cache_index_and_filter_blocks=false \
--pin_l0_filter_and_index_blocks_in_cache=false \
--pin_top_level_index_and_filter=false \
--prepopulate_block_cache=0 \
--row_cache_size=0 \
--use_blob_cache=false \
--use_compressed_secondary_cache=false
```
| B=10; N=100,000 | B = 250; N=80,000 | B = 1,000; N=20,000
-- | -- | -- | --
baseline | 31.5 (± 0.4) us/op | 17.5 (± 0.5) us/op | 13.5 (± 0.4) us/op
io_uring_submit_and_wait | 31.5 (± 0.6) us/op | 17.7 (± 0.4) us/op | 13.6 (± 0.4) us/op
io_uring_submit | 31.5 (± 0.6) us/op | 17.5 (± 0.5) us/op | 13.4 (± 0.45) us/op
### Specs
| Property | Value
-- | --
RocksDB | version 10.9.0
Date | Tue Dec 9 15:57:03 2025
CPU | 56 * Intel Sapphire Rapids (T10 SPR)
Kernel version | 6.9.0-0_fbk12_0_g28f2d09ad102
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14158
Reviewed By: anand1976
Differential Revision: D88172809
Pulled By: mszeszko-meta
fbshipit-source-id: 5198de3d2f18f76fee661a2ec5f447e79ba06fbd
Summary:
**Context/Summary:**
Truncated range deletion in input files can be output by CompactionIterator with type kMaxValid instead of kTypeRangeDeletion, to satisfy ordering requirement between the truncated range deletion start key and a file's point keys. There was a plan to skip such key in https://github.com/facebook/rocksdb/pull/14122 but blockers remain to fulfill the plan.
Resumable compaction is not able to handle resumption from range deletion well at this point and should consider kMaxValid type same as kTypeRangeDeletion for resumption. Previously, it didn't and mistakenly allow resumption from a delete range. That led to an assertion failure, complaining about lacking information to update file boundaries in the presence of range deletion needed during cutting an output file, after the compaction resumes from that delete range and happens to cut the output file shortly after without any point keys in between.
```
frame https://github.com/facebook/rocksdb/issues/9: 0x00007f4f4743bc93 libc.so.6`__GI___assert_fail(assertion="meta.smallest.size() > 0", file="db/compaction/compaction_outputs.cc", line=530, function="rocksdb::Status rocksdb::CompactionOutputs::AddRangeDels(rocksdb::CompactionRangeDelAggregator&, const rocksdb::Slice*, const rocksdb::Slice*, rocksdb::CompactionIterationStats&, bool, const rocksdb::InternalKeyComparator&, rocksdb::SequenceNumber, std::pair<long unsigned int, long unsigned int>, const rocksdb::Slice&, const string&)") at assert.c:101:3
frame https://github.com/facebook/rocksdb/issues/10: 0x00007f4f4808c68c librocksdb.so.10.9`rocksdb::CompactionOutputs::AddRangeDels(this=0x00007f4f0c27e1a0, range_del_agg=0x00007f4f0c21ecc0, comp_start_user_key=0x0000000000000000, comp_end_user_key=0x0000000000000000, range_del_out_stats=0x00007f4f0dffa140, bottommost_level=false, icmp=0x00007f4ef4c93040, earliest_snapshot=13108729, keep_seqno_range=<unavailable>, next_table_min_key=0x00007f4ef4c8f540, full_history_ts_low="") at compaction_outputs.cc:530:7
frame https://github.com/facebook/rocksdb/issues/11: 0x00007f4f480480dd librocksdb.so.10.9`rocksdb::CompactionJob::FinishCompactionOutputFile(this=0x00007f4f0dffb890, input_status=<unavailable>, prev_table_last_internal_key=0x00007f4f0dffa650, next_table_min_key=0x00007f4ef4c8f540, comp_start_user_key=0x0000000000000000, comp_end_user_key=0x0000000000000000, c_iter=0x00007f4ef4c8f400, sub_compact=0x00007f4f0c27e000, outputs=0x00007f4f0c27e1a0) at compaction_job.cc:1917:31
```
This PR simply prevents MaxValid from being a resumption point like regular range deletion - see commit 842d66eb18ea67e965d6acb1fce12c18eeb778d2
Besides that, the PR also improves the testing, variable naming, logging in resumable compaction codes that were needed to debug this assertion failure - see commit https://github.com/facebook/rocksdb/pull/14184/commits/aecd4e7f971f6dd4df672d9e5f1409fe4747c561. These improvements are covered by existing tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14184
Test Plan:
- The stress initially surfaced the error. Using the exact same LSM shapes and files that were used in stress test but in a unit test, I'm able to get a deterministic repro and confirmed the fix resolves the error. This is the repro test https://github.com/hx235/rocksdb/commit/1075936e693c68c960761855900c53f5b894f57a
```
./compaction_service_test --gtest_filter=ResumableCompactionServiceTest.CompactSpecificFilesFromExistingDBWithCancelAndResume
# Before fix
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from ResumableCompactionServiceTest
[ RUN ] ResumableCompactionServiceTest.CompactSpecificFilesFromExistingDBWithCancelAndResume
compaction_service_test: db/compaction/compaction_outputs.cc:530: rocksdb::Status rocksdb::CompactionOutputs::AddRangeDels(rocksdb::CompactionRangeDelAggregator&, const rocksdb::Slice*, const rocksdb::Slice*, rocksdb::CompactionIterationStats&, bool, const rocksdb::InternalKeyComparator&, rocksdb::SequenceNumber, std::pair<long unsigned int, long unsigned int>, const rocksdb::Slice&, const string&): Assertion `meta.smallest.size() > 0' failed.
Received signal 6 (Aborted)
Invoking GDB for stack trace...
[New LWP 2621610]
[New LWP 2621611]
[New LWP 2621612]
[New LWP 2621613]
[New LWP 2621614]
[New LWP 2621630]
[New LWP 2621631]
# After fix
Note: Google Test filter = ResumableCompactionServiceTest.CompactSpecificFilesFromExistingDBWithCancelAndResume
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from ResumableCompactionServiceTest
[ RUN ] ResumableCompactionServiceTest.CompactSpecificFilesFromExistingDBWithCancelAndResume
[ OK ] ResumableCompactionServiceTest.CompactSpecificFilesFromExistingDBWithCancelAndResume (4722 ms)
[----------] 1 test from ResumableCompactionServiceTest (4722 ms total)
[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (4722 ms total)
[ PASSED ] 1 test.
```
- Follow-up: I tried a couple time to coerce the truncated range delete from scratch in the unit test but failed doing so. Considering kMaxValid may not be outputted by compaction iterator anymore after https://github.com/facebook/rocksdb/pull/14122/files gets landed again (and obsolete the bug) ADN the simple nature of this fix 842d66eb18ea67e965d6acb1fce12c18eeb778d2 AND the worst case of such fix going wrong is just less resumption, I decided to leave writing a unit test to coerce truncated ranged deletion from scratch a follow-up. Maybe I will draw inspiration from https://github.com/facebook/rocksdb/pull/14122/files.
Reviewed By: jaykorean
Differential Revision: D88912663
Pulled By: hx235
fbshipit-source-id: 80a01135684c8fea659650faaa00c2dc452c482a
Summary:
**Context/Summary:**
Stress test flag printed by db_crashtest.py like `./db_stres ....-secondary_cache_uri=compressed_secondary_cache://capacity=8388608;enable_custom_split_merge=true --otherflags=xxxx` is not copy-paste-run friendly. Directly running this command will cause parsing hiccups due to special characters like // or ;. This PR made the db_crashtest.py print a single-quoted value so at least the copy-paste-run works for unix-like shell (the most common case).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14180
Test Plan:
`python3 tools/db_crashtest.py --simple blackbox ...` display the following
Before fix, no single-quoted
```
Use random seed for iteration 9698536012932546857
Running db_stress with pid=1280640:./db_stress --secondary_cache_uri=compressed_secondary_cache://capacity=8388608;enable_custom_split_merge=true ...
// Directly copy, paste and run the ./db_stress command will encounter
Error: Read(-readpercent=0)+Prefix(-prefixpercent=0)+Write(-writepercent=45)+Delete(-delpercent=0)+DeleteRange(-delrangepercent=30)+Iterate(-iterpercent=40)+CustomOps(-customopspercent=0) percents != 100!
bash: --set_options_one_in=0: command not found
```
After fix, has single-quoted
```
se random seed for iteration 6017815530972723112
Running db_stress with pid=1234632: ./db_stress --secondary_cache_uri='compressed_secondary_cache://capacity=8388608;enable_custom_split_merge=true' ....
// Directly copy, paste and run the ./db_stress command is fine
```
Reviewed By: archang19
Differential Revision: D88688584
Pulled By: hx235
fbshipit-source-id: 88b8b2de7c2c5619b6e19900f4144dcd8e032f7b
Summary:
r? cbi42
Exposes RocksDB's remote compaction functionality through the C API, enabling C/FFI clients (Go, Rust, Python, etc.) to offload compaction work to remote workers.
## API Components
### Compaction Service
Create service with schedule, wait, cancel, and on_installation callbacks
Ownership transfers to options object (auto-destroyed, no manual cleanup)
### Job Info (13 getters)
DB/CF metadata and compaction details (priority, reason, levels, flags)
### Schedule Response
Create with job ID and status (validated with errptr)
Status: success, failure, aborted, use_local
### OpenAndCompact (for remote workers)
Execute compaction on worker node with environment/comparator overrides
Cancellation support via atomic flags
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14136
Reviewed By: hx235
Differential Revision: D88316558
Pulled By: jaykorean
fbshipit-source-id: 60a0fee69ff1e650dd785d96ec656649263214f8
Summary:
Crash tests have been failing of late with this assertion failure - db_stress: `./table/block_based/block_based_table_iterator.h:656: void rocksdb::BlockBasedTableIterator::PrepareReadAsyncCallBack(rocksdb::FSReadRequest &, void *): Assertion `async_state->status.IsAborted()' failed.` Instead of asserting, surface the failure status so we can troubleshoot.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14171
Reviewed By: xingbowang
Differential Revision: D88396654
Pulled By: anand1976
fbshipit-source-id: 8d59d7ace0c522c17b7af17c50e16af876911bad
Summary:
To help find potential issues not showing up in ARM unit tests. I'm running it with and without TransactionDB (write-committed) for better coverage. The job expands the size of /dev/shm for adequate space on maximum performance storage, and adds swap space to reduce risk of OOM in case we fill that up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14172
Test Plan: earlier drafts of this PR added the job to PR jobs, and the last before putting in "nightly" can be seen here: https://github.com/facebook/rocksdb/actions/runs/19945493840/job/57193797390?pr=14172
Reviewed By: archang19
Differential Revision: D88429479
Pulled By: pdillinger
fbshipit-source-id: bd4d9cda9256950c3c6c126c299a44dbbbc30c7e
Summary:
Fix missing const for arg of OptionChangeMigration
We switched from std::string to std::string & for API OptionChangeMigration, which caused const qualifier to be lost at call site, which causes compilation failure.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14173
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D88431457
Pulled By: xingbowang
fbshipit-source-id: a705f3b80cc5ff56dab73aa6a31c940798d8df45
Summary:
Revert "Fix a bug where compaction with range deletion can persist kTypeMaxValid in file metadata (https://github.com/facebook/rocksdb/issues/14122)"
Add a new unit test to capture the situation found by stress test
This reverts commit 8c7c8b8dab.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14170
Test Plan: Unit Test
Reviewed By: anand1976
Differential Revision: D88395956
Pulled By: xingbowang
fbshipit-source-id: 226649dc79a86010ad326ffb2eae35109dc96bc4
Summary:
Continuing work from https://github.com/facebook/rocksdb/issues/13965. Here I'm migrating the "next with shift" kind of bit field and for that I've added an API for atomic additive transformations that can be combined into a single atomic update for multiple fields. (I implemented more features than needed, just in case they are needed someday and to demonstrate what is possible.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14027
Test Plan: BitFields unit test updated/added, existing HCC tests
Reviewed By: xingbowang
Differential Revision: D83895094
Pulled By: pdillinger
fbshipit-source-id: e4487f34f5607b20f94b85a645ca654e6401e35d
Summary:
I want to reduce the time from when we call `StopBackup` to `CreateNewBackup` returning `BackupStopped`. We already check for the `stop_backup_` inside `CopyOrCreateFile` and `ReadFileAndComputeChecksum`, but we should add a check at the top of these methods to abort immediately. This could help save some latency from the file system metadata operations, like creating the sequential file and writable file.
We also want to update the API documentation for `StopBackup` which currently does not indicate that once it is called, all subsequent requests to create backups will fail.
In a follow up PR, we should also add coverage of `StopBackup` to the crash tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14129
Test Plan:
We were missing unit test coverage for `StopBackup`. I added test cases which cancel backups at different points in time.
Once this change is rolled out to production, we can monitor the DB close latencies, which depend on first cancelling ongoing backups
Reviewed By: pdillinger
Differential Revision: D87356536
Pulled By: archang19
fbshipit-source-id: 687094a41f096f6a156be65b2cce0b5054fb26f2
Summary:
Support ccache in make file
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14123
Test Plan: local build
Reviewed By: cbi42
Differential Revision: D87332892
Pulled By: xingbowang
fbshipit-source-id: 2088bd19bdab1bd7070734c886200be80f1a65af
Summary:
... from https://github.com/facebook/rocksdb/issues/14140. The assertion in the default implementation of CompressorWrapper::MaybeCloneSpecialized() could fail because this wrapper wasn't overriding it when it should. (See the NOTE on that implementation.)
Because this release already has a breaking modification to the Compressor API (adding Clone()), I took this opportunity to add 'const' to MaybeCloneSpecialized(). Also marked some compression classes as 'final' that could be marked as such.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14150
Test Plan: unit test expanded to cover this case (verified failing before). Audited the rest of our CompressorWrappers.
Reviewed By: archang19
Differential Revision: D87793987
Pulled By: pdillinger
fbshipit-source-id: 61c4469b84e4a47451a9942df09277faeeccfe63
Summary:
This change enables a custom CompressionManager / Compressor to adopt custom handling for data and index blocks. In particular, index blocks for format_version >= 4 use a distinct variant of the block format. Thus, a potentially format-aware compression algorithm such as OpenZL should be told which kind of block we are compressing. (And previously I avoided passing block type in CompressBlock for efficient handling of things like dictionaries but also avoiding checks on every CompressBlock call.)
Most of the change is in BlockBasedTableBuilder to call MaybeCloneSpecialized for both kDataBlock and for kIndexBlock. But I also needed some small tweaks/additions to the public API also:
* Require a Clone() function from Compressors, to support proper implementations of MaybeCloneSpecialized() in wrapper Compressors.
* Assert that the default implementation of CompressorWrapper::MaybeCloneSpecialized() is only used in allowable cases.
* Convenience function Compressor::CloneMaybeSpecialized()
This also fixes a serious bug/oversight in ManagedPtr for (ManagedWorkingArea) that somehow wasn't showing up before. It probably doesn't need a release note because CompressionManager stuff is still considered experimental.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14140
Test Plan: Greatly expanded DBCompressionTest.CompressionManagerWrapper to make sure the distinction between data blocks and index blocks is properly communicated to a custom CompressionManager/Compressor. The test includes processing the expected structure of data and index blocks, to serve as a tested example for structure-aware compressors.
Reviewed By: hx235
Differential Revision: D87600019
Pulled By: pdillinger
fbshipit-source-id: 252ef78910073a0e45f2c81dd45ac87ff8a41fc6
Summary:
Range deletion start keys are considered during compaction for cutting output files. Due to some ordering requirement (see comment above InsertNextValidRangeTombstoneAtLevel()) between truncated range deletion start key and a file's point keys, there was logic in https://github.com/facebook/rocksdb/blob/f6c9c3bf1cf05096e8ff8c03ded60c1e199edbb7/db/range_del_aggregator.cc#L39 that changes the value type to be kTypeMaxValid. However, kTypeMaxValid is not supposed to be persisted per https://github.com/facebook/rocksdb/blob/f6c9c3bf1cf05096e8ff8c03ded60c1e199edbb7/db/dbformat.h#L75-L76. This can cause forward compatibility issues reported in https://github.com/facebook/rocksdb/issues/14101. This PR fixes this issue by removing the logic that sets kTypeMaxValid and always skip truncated range deletion start key in CompactionMergingIterator.
For existing SST files, we want to avoid using this kTypeMaxValid, so this PR also introduces a new placeholder value type. This allows us to re-strengthen the relevant value type checks (IsExtendedValueType()) that was loosen for kTypeMaxValid.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14122
Test Plan:
- a unit test that persists kTypeMaxValid before this fix
- crash test with frequent range deletion: `python3 ./tools/db_crashtest.py blackbox --delrangepercent=11 --readpercent=35`
- Generate SST files with 0x1A as value type (kTypeMaxValid before this change) in file metadata. Run ldb with the strengthened check in IsExtendedValueType() to dump the MANIFEST. It failed to parse MANIFEST as expected before this PR and succeeds after this PR.
```
Error in processing file /tmp/rocksdbtest-543376/db_range_del_test_2549357_6547198162080866792/MANIFEST-000005 Corruption: VersionEdit: new-file4 entry The file /tmp/rocksdbtest-543376/db_range_del_test_2549357_6547198162080866792/MANIFEST-000005 may be corrupted.
```
Reviewed By: pdillinger
Differential Revision: D87016541
Pulled By: cbi42
fbshipit-source-id: 9957a095db2cd9947463b403f352bd9a1fd70a76
Summary:
Fixing internal validator failure
```
Every project specific source file must contain a doc block with an appropriate copyright header. Unrelated files must be listed as exceptions in the Copyright Headers Exceptions page in the repo dashboard.
A copyright header clearly indicates that the code is owned by Meta. Every open source file must start with a comment containing "Meta Platforms, Inc. and affiliates"
https://github.com/facebook/rocksdb/blob/main/buckifier/targets_cfg.py:
The first 16 lines of 'buckifier/targets_cfg.py' do not contain the patterns:
(Meta Platforms, Inc. and affiliates)|(Facebook, Inc(\.|,)? and its affiliates)|([0-9]{4}-present(\.|,)? Facebook)|([0-9]{4}(\.|,)? Facebook)
```
While fixing the text to pass the linter, I took the opportunity to modify `format-diff.sh` script to add the copyright header automatically if missing in new files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14143
Test Plan:
```
$> make format
```
**new python file**
```
build_tools/format-diff.sh
Checking format of uncommitted changes...
Checking for copyright headers in new files...
Added copyright header to build_tools/test.py
Copyright headers were added to new files.
Nothing needs to be reformatted!
```
**new header file**
```
build_tools/format-diff.sh
Checking format of uncommitted changes...
Checking for copyright headers in new files...
Added copyright header to db/db_impl/db_impl_jewoongh.h
Copyright headers were added to new files.
Nothing needs to be reformatted!
```
Reviewed By: hx235
Differential Revision: D87653124
Pulled By: jaykorean
fbshipit-source-id: 164322cfcd2c162bb3b41bb8f3bafefa3f20b695
Summary:
**Context/Summary:**
.. because verify_output_flags contains information of usage of paranoid_file_check that is currently not yet compatible with resumable remote compaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14139
Test Plan: Existing tests
Reviewed By: jaykorean
Differential Revision: D87582635
Pulled By: hx235
fbshipit-source-id: ef21223da53a0696fa3ca9b1617c2c1ee2e19878
Summary:
**Context/Summary:**
Due to double's 53-bit mantissa limitation, large uint64_t values lose precision when converted to double. Value equals to or smaller than UINT64_MAX (but greater than 2^64 - 1024) round up to 2^64 since rounding up results in less error than rounding down, which exceeds UINT64_MAX. `std::numeric_limits<uint64_t>::max() / op1 < op2` won't catch those cases. Casting such out-of-range doubles back to uint64_t causes undefined behavior. T
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14132
UndefinedBehaviorSanitizer: undefined-behavior options/cf_options.cc:1087:32 in
```
before the fix but not after.
Test Plan:
```
COMPILE_WITH_ASAN=1 COMPILE_WITH_UBSAN=1 CC=clang-18 CXX=clang++-18 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j55 db_stress
python3 tools/db_crashtest.py --simple blackbox --compact_range_one_in=5 --target_file_size_base=9223372036854775807 // Half of std::numeric_limits<uint64_t>::max()
```
It fails with
```
stderr:
options/cf_options.cc:1087:32: runtime error: 1.84467e+19 is outside the range of representable values of type 'unsigned long'
Reviewed By: pdillinger
Differential Revision: D87434936
Pulled By: hx235
fbshipit-source-id: 65563edf9faf732410bdba8b9e4b7fd61b958169
Summary:
I have been using sst_dump --command=recompress for some ad hoc automation for compression engineering and these new options help with that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14133
Test Plan: manual
Reviewed By: hx235
Differential Revision: D87453635
Pulled By: pdillinger
fbshipit-source-id: 2ae54e13a9221ec27c6637fea16623465a9163ae
Summary:
Saw a mysterious failure of assertion
`assert(rep_->props.num_data_blocks == 0)` in
DBCompressionTest/CompressionFailuresTest.CompressionFailures/45. This seems to be caused by a parallel compression failure arriving after the emit thread has started Finish() but before the Flush() at the start of Finish(). We can fix this by relaxing the assertion to allow for the !ok() case. Testing revealed more ok() assertions that needed to be relaxed/moved.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14130
Test Plan: Added a sync point to inject a failure status in the right place and added to unit test to be sure the case is essentially covered. It would arguably be a more realistic test to force a particular thread interleaving but I believe simple is good here.
Reviewed By: hx235
Differential Revision: D87377709
Pulled By: pdillinger
fbshipit-source-id: 4bd465673b084afcc235688503d1c2f464eed32d
Summary:
**Context/Summary:**
This PR adds multi-cf support to option migration. The original implementation sets options, opens db, compacts files and reopens the db in almost all the three branches below. Such design makes expanding to multi-cf difficult as it needs to change all these places within each of the branch causing code redundancy.
```
Status OptionChangeMigration(std::string dbname, const Options& old_opts,
const Options& new_opts) {
if (old_opts.compaction_style == CompactionStyle::kCompactionStyleFIFO) {
// LSM generated by FIFO compaction can be opened by any compaction.
return Status::OK();
} else if (new_opts.compaction_style ==
CompactionStyle::kCompactionStyleUniversal) {
return MigrateToUniversal(dbname, old_opts, new_opts);
} else if (new_opts.compaction_style ==
CompactionStyle::kCompactionStyleLevel) {
return MigrateToLevelBase(dbname, old_opts, new_opts);
} else if (new_opts.compaction_style ==
CompactionStyle::kCompactionStyleFIFO) {
return CompactToLevel(old_opts, dbname, 0, 0 /* l0_file_size */, true);
} else {
return Status::NotSupported(
"Do not how to migrate to this compaction style");
}
}
```
Therefore this PR
- Refactor the option migration implementation by moving the common parts into the high-level `OptionChangeMigration()` through `PrepareNoCompactionCFDescriptors()` and `OpenDBWithCFs()` so `MigrateAllCFs()` can focus on compaction only.
- Treat the original OptionChangeMigration() API as a special case of the multi-cf version option migration
- Add multiple-cf support
A few notes:
- CompactToLevel() originally modifies the compaction-related options conditionally before doing compaction. This is moved into earlier steps through `ApplySpecialSingleLevelSettings()` in `PrepareNoCompactionCFDescriptors()`
- MigrateToUniversal() originally opens the db twice with essentially the same option. This PR reduces that to one open
- Option migration does not always use the old option to compact the db and reopen the db after migration, see ` return CompactToLevel(new_opts, dbname, new_opts.num_levels - 1,/*l0_file_size=*/0, false);`. `PrepareNoCompactionCFDescriptors()` is where we handle those decisions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14059
Test Plan:
- Existing UTs
- New UTs
Reviewed By: cbi42
Differential Revision: D84852970
Pulled By: hx235
fbshipit-source-id: 936b456cf9fb4c3ccb687e5d1387f2d67a1448be
Summary:
This diff introduces the async prepare of all iterators within a MultiScan. The current state has each iterator be prepared as its needed, and with this diff, we prepare all iterators during the prepare phase of the Level Iterator, this will allow more time for each IO to be dispatched and serviced, increasing the odds that a block is ready as the scan seeks to it.
Benchmark is prefilled using
```
KEYSIZE=64
VALUESIZE=512
NUMKEYS=5000000
SCAN_SIZE=100
DISTANCE=25000
NUM_SCANS=15
THREADS=1
./db_bench --db=$DB \
--benchmarks="fillseq" \
--write_buffer_size=5242880 \
--max_write_buffer_number=4 \
--target_file_size_base=5242880 \
--disable_wal=1 --key_size=$KEYSIZE \
--value_size=$VALUESIZE --num=$NUMKEYS --threads=32
}
```
And benchmark ran is
```
run() {
echo 1 | sudo tee /proc/sys/vm/drop_caches
./db_bench --db=$DB --use_existing_db=1 \
--benchmarks=multiscan \
--disable_auto_compactions=1 --seek_nexts=$SCAN_SIZE \
--multiscan-use-async-io=1 \
--multiscan-size=$NUM_SCANS --multiscan-stride=$DISTANCE \
--key_size=$KEYSIZE --value_size=$VALUESIZE \
--num=$NUMKEYS --threads=$THREADS --duration=60 --statistics
}
```
The benchmark uses large stride sides to ensure that two scans would touch separate files. We reduce the size of the block cache to increase likelyhood of reads (and simulate larger data sets)
**Branch:**
```
Integrated BlobDB: blob cache disabled
RocksDB: version 10.8.0
Date: Tue Nov 11 13:26:29 2025
CPU: 166 * AMD EPYC-Milan Processor
CPUCache: 512 KB
Keys: 64 bytes each (+ 0 bytes user-defined timestamp)
Values: 512 bytes each (256 bytes after compression)
Entries: 5000000
Prefix: 0 bytes
Keys per prefix: 0
RawSize: 2746.6 MB (estimated)
FileSize: 1525.9 MB (estimated)
Write rate: 0 bytes/second
Read rate: 0 ops/second
Compression: Snappy
Compression sampling rate: 0
Memtablerep: SkipListFactory
Perf Level: 1
------------------------------------------------
multiscan_stride = 25000
multiscan_size = 15
seek_nexts = 100
DB path: [/data/rocksdb/mydb]
multiscan : 837.941 micros/op 1193 ops/sec 60.001 seconds 71605 operations; (multscans:71605)
```
**Baseline:**
```
Set seed to 1762898809121995 because --seed was 0
Initializing RocksDB Options from the specified file
Initializing RocksDB Options from command-line flags
Integrated BlobDB: blob cache disabled
RocksDB: version 10.9.0
Date: Tue Nov 11 14:06:49 2025
CPU: 166 * AMD EPYC-Milan Processor
CPUCache: 512 KB
Keys: 64 bytes each (+ 0 bytes user-defined timestamp)
Values: 512 bytes each (256 bytes after compression)
Entries: 5000000
Prefix: 0 bytes
Keys per prefix: 0
RawSize: 2746.6 MB (estimated)
FileSize: 1525.9 MB (estimated)
Write rate: 0 bytes/second
Read rate: 0 ops/second
Compression: Snappy
Compression sampling rate: 0
Memtablerep: SkipListFactory
Perf Level: 1
------------------------------------------------
multiscan_stride = 25000
multiscan_size = 15
seek_nexts = 100
DB path: [/data/rocksdb/mydb]
multiscan : 1129.916 micros/op 885 ops/sec 60.001 seconds 53102 operations; (multscans:53102)
```
Repeated for confirmation.
This introduces a ~20% improvement in latency and op/s.
Note: Benchmarks are single threaded as, when increasing thread count, we start seeing large amounts of overhead being induced by block cache contention, finally resulting in both baseline and branch becoming equal.
Further on network attached storage with high latency, the level iterator, preparing all iterators so a 20% improvement even at high thread counts.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14100
Reviewed By: anand1976
Differential Revision: D86913584
Pulled By: krhancoc
fbshipit-source-id: da9d0c890e25e392a33389ce6b80f9bfb84d3f85
Summary:
Oversight in https://github.com/facebook/rocksdb/issues/13964. More detail:
* Applies to cache_bench and db_bench (db_stress already using it)
* Make sure those along with db_stress treat "hyper_clock_cache" as "auto_hyper_clock_cache" because this is now the blessed implementation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14120
Test Plan: manual runs of the tools
Reviewed By: krhancoc
Differential Revision: D86913202
Pulled By: pdillinger
fbshipit-source-id: 07b425d3522103417f4b034735376b9d759af5fb
Summary:
Right now, in Java's Get() calls, the way Get() is treated is inefficient. Status.NotFound is turned into an exception in the JNI layer, and is caught in the same function to turn into not found return. This causes significant overhead in the scenario where most of the queries ending up with not found. For example, in Spark's deduplication query, this exception creation overhead is higher than Get() itself. With the proposed change, if return status is NotFound, we directly return, rather than going through the exception path
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14095
Test Plan: Existing tests should cover all Get() cases, and they are passing.
Reviewed By: jaykorean
Differential Revision: D86797594
Pulled By: cbi42
fbshipit-source-id: 1202d24e46a2358976bb7c8ff38a2fd4783d0f99
Summary:
There are instances where an application might be interested in knowing the distribution in SST files for a key range in a particular level.
This implementation creates an overloaded GetColumnFamilyMetaData api where (startKey, EndKey) can be passed along with level information to filter the necessary sst files along with the keyranges for each sst file
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14009
Reviewed By: anand1976
Differential Revision: D83389707
fbshipit-source-id: 6df1dc1f9233efe9000b03cc1831b3c618cbcef3
Summary:
Support trivial move in CompactFiles API, which is not supported previously.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14112
Test Plan: Unit test
Reviewed By: cbi42
Differential Revision: D86546150
Pulled By: xingbowang
fbshipit-source-id: 08a3ae9a055f3d3d41711403b1695f44977e6ea8
Summary:
**Summary:**
Merge the BuiltinFilterBitsBuilder into FilterBitsBuilder. This enables using
CalculateSpace() for accurate filter size estimation instead of hardcoded
bits-per-key which could result in incorrect estimations for different filter types.
The previous hardcoded estimate of 15 bits per key was in the filter block builders UpdateFilterSizeEstimate().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14111
Test Plan: - Existing filter tests pass (bloom_test, full_filter_block_test, filter_bench, db_bloom_filter_test)
Reviewed By: pdillinger
Differential Revision: D86473287
Pulled By: nmk70
fbshipit-source-id: cd4a47351e67444e944d5b1b375b3b13274dd6e3
Summary:
For all compactions, RocksDB performs a lightweight sanity check on output SST files before installation (in `CompactionJob::VerifyOutputFiles()`). However, this lightweight check may not catch corruption that is small enough to allow the SST files to still be opened.
There is an existing feature, `paranoid_file_check`, which opens the SST file, iterates through all keys, and checks the hash of each key. While this provides the ultimate level of data integrity checking, it comes at a high computational cost.
In this PR, we introduce a new mutable CF option, `verify_output_flags`. The `verify_output_flags` is a bitmask enum that allows users to select various verification types, including block checksum verification, full key iteration, and file checksum verification (to be added in subsequent PRs). Note that the existing `paranoid_file_check` option is equivalent to a full key iteration check. Block-level checksum verification is much lighter than the full key iteration check.
Please note that the previously deprecated `verify_checksums_in_compaction` option (removed in version 5.3.0) was for verifying the checksum of **input SST files**. RocksDB continues to perform this verification for both local and remote compactions, and this behavior remains unchanged. In contrast, this PR focuses on verifying the **output SST files**.
## To follow up
- File-level Checksum verification for output SST files
- Deprecate `paranoid_file_checks` option in favor of the new option
- Add to stress test / db_bench
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14103
Test Plan:
New Unit Test added. The corruption is both detected by `paranoid_file_check` and various types of verification set by this new option, `verify_output_flags`
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.CorruptedOutput*"
```
Reviewed By: pdillinger
Differential Revision: D86357924
Pulled By: jaykorean
fbshipit-source-id: a9e04798f249c7e977231e179622a0830d6675fe
Summary:
MultiScanUnexpectedSeekTarget() currently uses user key comparison to decide on the next data block for multiscan. This can cause a multiscan to move backward in the following scenario:
data block 1: ..., k@7, k@6
data block 2: k@5, ...
DB iter scan through k@7, k@6 and k@5 and decides to seek to k@0 due to option [`max_sequential_skip_in_iterations`](https://github.com/facebook/rocksdb/blob/d56da8c112b4e6968fd79ce2bf15e6435df40656/include/rocksdb/advanced_options.h#L621-L629). Multiscan was on data block 2, but moves to data block 1 after the seek.
This can cause assertion failure in debug mode and seg fault in prod since older data blocks are unpinned and freed as we advanced a multiscan. This PR fixes the issue by forcing a multiscan to never go backward.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14106
Test Plan: - added a new unit test that reproduces the scenario: `./db_iterator_test --gtest_filter="*ReseekAcrossBlocksSameUserKey*"`
Reviewed By: xingbowang
Differential Revision: D86428845
Pulled By: cbi42
fbshipit-source-id: ab623f93e73298a60857fb2ff268366f289092a0
Summary:
This test is now taking > 6 hours, timing out, and has low signal, so creating a weekly job for it, with an explicit timeout of 12 hours.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14110
Test Plan: watch CI
Reviewed By: virajthakur
Differential Revision: D86428262
Pulled By: pdillinger
fbshipit-source-id: 44103518064ca378f3fd2ff8d21967ede698c8ea
Summary:
Adds auto-tuning of manifest file size to avoid the need to scale `max_manifest_file_size` in proportion to things like number of SST files to properly balance (a) manifest file write amp and new file creation, vs. (b) manifest file space amp and replay time, including non-incremental space usage in backups. (Manifest file write amp comes from re-writing a "live" record when the manifest file is re-created, or "compacted"; space amp is usage beyond what would be used by a compacted manifest file.) In more detail,
* Add new option `max_manifest_space_amp_pct` with default value of 500, which defaults to 0.2 write amp and up to roughly 5.0 space amp, except `max_manifest_file_size` is treated as the "minimum" size before re-creating ("compacting") the manifest file.
* `max_manifest_file_size` in a way means the same thing, with the same default of 1GB, but in a way has taken on a new role. What is the same is that we do not re-create the manifest file before reaching this size (except for DB re-open), and so users are very unlikely to see a change in default behavior (auto-tuning only kicking in if auto-tuning would exceed 1GB for effective max size for the current manifest file). The new role is as a file size lower bound before auto-tuning kicks in, to minimize churn in files considered "negligibly small." We recommend a new setting of around 1MB or even smaller like 64KB, and expect something like this to become the default soon.
* These two options along with `manifest_preallocation_size` are now mutable with SetDBOptions. The effect is nearly immediate, affecting the next write to the current manifest file.
Also in this PR:
* Refactoring of VersionSet to allow it to get (more) settings from MutableDBOptions. This touches a number of files in not very interesting ways, but notably we have to be careful about thread-safe access to MutableDBOptions fields, and even fields within VersionSet. I have decided to save copies of relevant fields from MutableDBOptions to simplify testing, etc. by not saving a reference to MutableDBOptions but getting notified of updates.
* Updated some logging in VersionSet to provide some basic data about final and compacted manifest sizes (effects of auto-tuning), making sure to avoid I/O while holding DB mutex.
* Added db_etc3_test.cc which is intended as a successor to db_test and db_test2, but having "test.cc" in its name for easier exclusion of test files when using `git grep`. Intended follow-up: rename db_test2 to db_etc2_test
* Moved+updated `ManifestRollOver` test to the new file to be closer to other manifest file rollover testing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14076
Test Plan:
As for correctness, new unit test AutoTuneManifestSize is pretty thorough. Some other unit tests updated appropriately. Manual tests in the performance section were also audited for expected behavior based on the new logging in the DB LOG. Example LOG data with -max_manifest_file_size=2048 -max_manifest_space_amp_pct=500:
```
2025/10/24-11:12:48.979472 2150678 [/version_set.cc:5927] Created manifest 5, compacted+appended from 52 to 116
2025/10/24-11:12:49.626441 2150682 [/version_set.cc:5927] Created manifest 24, compacted+appended from 2169 to 1801
2025/10/24-11:12:52.194592 2150682 [/version_set.cc:5927] Created manifest 91, compacted+appended from 10913 to 8707
2025/10/24-11:13:02.969944 2150682 [/version_set.cc:5927] Created manifest 362, compacted+appended from 52259 to 13321
2025/10/24-11:13:18.815120 2150681 [/version_set.cc:5927] Created manifest 765, compacted+appended from 80064 to 13304
2025/10/24-11:13:35.590905 2150681 [/version_set.cc:5927] Created manifest 1167, compacted+appended from 79863 to 13304
```
As you can see, it only took a few iterations of ramp-up to settle on the auto-tuned max manifest size for tracking ~122 live SST files, around 80KB and compacting down to about 13KB. (13KB * (500 + 100) / 100 = 78KB). With the default large setting for max_manifest_file_size, we end up with a 232KB manifest, which is more than 90% wasted space. (A long-running DB would be much worse.)
As for performance, we don't expect a difference, even with TransactionDB because actual writing of the manifest is done without holding the DB mutex. I was not able to see a performance regression using db_bench with FIFO compaction and >1000 ~10MB SST files, including settings of -max_manifest_file_size=2048 -max_manifest_space_amp_pct={500,10,0}. No "hiccups" visible with -histogram either.
I also tried seeding a 1 second delay in writing new manifest files (other than the first). This had no significant effect at -max_manifest_space_amp_pct=500 but at 100 started causing write stalls in my test. In many ways this is kind of a worst case scenario and out-of-proportion test, but gives me more confidence that a higher number like 500 is probably the best balance in general.
Reviewed By: xingbowang
Differential Revision: D85445178
Pulled By: pdillinger
fbshipit-source-id: 1e6e07e89c586762dd65c65bb7cb2b8b719513f9
Summary:
**Summary:**
This change introduces tail size estimation during SST construction to improve compaction file cutting accuracy to prevent oversized files. The BlockBasedTableBuilder now estimates the SST tail size (index and filter blocks) and uses this estimate, in addition to the data size, to determine when to cut files during compaction.
**Problem:**
Currently, file cutting logic only considers data size when determining where to cut a file, failing to reserve space for index and filter blocks that are added when the file is finalized. This often leads to SST files that exceed target file size limits.
**Behavior Change:**
Implement size estimation methods for index and filter builders, and integrate these estimates into BlockBasedTableBuilder via a new EstimatedTailSize() method. This method aggregates estimates from all tail components and is used for file cutting decisions during compaction.
**Performance Considerations:**
To minimize CPU overhead, size estimates are updated when data blocks are finalized rather than on every key add. For index builders, estimates are updated when index entries are added (one per data block). For filter builders, the OnDataBlockFinalized() hook triggers estimate updates when data blocks are cut/finalized.
This approach provides:
* Minimal impact to compaction hot path (key additions)
* Near real-time estimates for file cutting decisions
* Meaningful estimate changes only when data blocks are finalized
**Usage:**
* Set true mutable cf option `compaction_use_tail_size_estimation`
to use tail size estimation for compaction file cutting decisions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14051
Test Plan:
* Assert tail size estimate is an overestimate in BlockBasedTableBuilder::Finish
* Add new test to verify compaction output file is below target file size
**Next steps:**
* Enable tail size estimation for compaction file cutting by default (and other improvements)
Reviewed By: pdillinger, cbi42
Differential Revision: D84852285
Pulled By: nmk70
fbshipit-source-id: c43cf5dbd2cb2f623a0622591ef24eee30ce0c87
Summary:
* Fix nightly build-linux-cmake-with-folly-lite-no-test for real this time
with correct include directory. (CMakeLists.txt)
* Add test runs to that build (and rename)
* Improve folly build caching with a folly.mk file with most of the relevant
parts of Makefile that contribute to the checkout_folly and
build_folly builds. This reduces the risk of false passing of CI job with
cache folly build. This caching is still only for folly debug builds, (which
is probably OK with just a single nightly build relying on release folly
build, which also serves as a rough canary against false passing
because of caching).
* Use `make VERBOSE=1` after cmake calls for detailed output
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14099
Test Plan:
temporary CI change to put the relevant parts in pr-jobs,
then back to homes including in nightly
Reviewed By: mszeszko-meta
Differential Revision: D86243363
Pulled By: pdillinger
fbshipit-source-id: f7975fa190ef45195c6d0b74417f7886e551516a
Summary:
... caused by public headers depending on build parameters (macro definitions). This change also adds a check under 'make check-headers' (already in CI) looking for potential future violations.
I've audited the uses of '#if' in public headers and either
* Eliminated them
* Systematically excluded them because they are intentional or similar (details in comments in check-public-header.sh
* Manually excluded them as being ODR-SAFE
In the case of ROCKSDB_USING_THREAD_STATUS, there was no good reason for this to appear in public headers so I've replaced it with a static bool ThreadStatus::kEnabled. I considered getting rid of the ability to disable this code but some relatively recent PRs have been submitted for fixing that case. I've added a release note and updated one of the CI jobs to use this build configuration. (I didn't want to combine with some jobs like no_compression and status_checked because the interaction might limit what is checked.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14096
Test Plan: manual 'make check-headers' + manual cmake as in new CI config + CI
Reviewed By: jaykorean
Differential Revision: D86241864
Pulled By: pdillinger
fbshipit-source-id: d16addc9e3480706b174a006720a4def0740bf2e
Summary:
Following up on https://github.com/facebook/rocksdb/pull/14071, updating folly to
https://github.com/facebook/folly/commit/8a9fc1e80a18cafadbec85e33d5042ce13a7c634 or beyond was failing an F14Table assertion for a very subtle reason: ODR violation between the folly build and RocksDB build because folly build was release mode and RocksDB build was debug mode. What was happening was that folly change introduced a dependence on kDebug (whether build is debug) in a hashing implementation in a .h file, and the inconsistency between the inlined implementation during RocksDB build and the linked-to implementation from the folly build was leading to inconsistencies in the data structure.
The primary fix is to ensure we build folly in debug mode for debug mode RocksDB builds. Also,
* Needed to use the `patchelf` tool in `build_folly` to ensure the glog dependency shared library can always find its own gflags dependency. I explored many options for working around this, and this is what would work without reworking folly's own build.
* Updated folly to latest commit.
* Thrown in an ad hoc folly patch to use ftp.gnu.org mirrors (the canonical is super slow)
* Moved the placement of GETDEPS_USE_WGET=1 to apply to local builds also, to avoid the issue of a large download almost reaching completion and then stalling indefinitely.
* Fix failing nightly build-linux-cmake-with-folly-lite-no-test with fmt includes in cmake build (as was done with make build)
* Add a release mode folly+RocksDB to nightly CI, including both cmake and make. This also serves as a non-cached folly build to detect potential problems with PR jobs working from cached folly build.
* Move build-linux-cmake-with-folly to nightly because it's mostly covered by build-linux-cmake-with-folly-coroutines
Intended follow-up:
* folly-lite build with tests
* Make the folly build caching more friendly+accurate by hashing the relevant Makefile parts and tagging whether debug or release. Not in this PR because then you wouldn't be able to see what changed in the folly build steps themselves.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14094
Test Plan: manual + CI
Reviewed By: mszeszko-meta
Differential Revision: D85864871
Pulled By: pdillinger
fbshipit-source-id: 50009b33422d5781074fcbbdf18089be9e36800d
Summary:
Resolving this folly upgrade required fixing the FOLLY_LITE build with header include from the 'fmt' library.
I was close to timing out on fixing USE_FOLLY_LITE and removing it altogether - it could be considered obsolete and/or not worth the maintenance cost.
Follow-up: make the folly build caching more friendly by hashing the relevant makefile parts. Not in this PR because then you wouldn't be able to see what changed in the folly build steps themselves.
UPDATE/NOTE: I wasn't able to fully update to latest due to a failure seen in F14, using the next folly commit or later. The source of the bug is likely outside of F14 but investigation is in progress.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14071
Test Plan: CI
Reviewed By: jaykorean
Differential Revision: D85268833
Pulled By: pdillinger
fbshipit-source-id: 1d0a2d61f095524a20e6ec796ef46c02d0696f4e
Summary:
Change PosixWritableFile's Truncate to the new end offset. This ensures that future appends are written with no holes or overwrites. RocksDB doesn't guarantee this in the FileSystem contract, and its left up to the specific implementation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14088
Reviewed By: cbi42
Differential Revision: D85786398
Pulled By: anand1976
fbshipit-source-id: 3520d9d6336362f5128a17bbf396297d821a5da3
Summary:
Comprehensive performance optimizations for the RocksDB C API that eliminate unnecessary memory allocations and copies.
## Key Changes
### 1. PinnableSlice for Get Operations (50% reduction in copies)
- Changed all `rocksdb_get*` functions to use `PinnableSlice` internally instead of `std::string`
- **Before:** RocksDB → std::string → malloc'd buffer (2 copies)
- **After:** RocksDB → malloc'd buffer (1 copy)
- Affects: Get, Transaction Get, TransactionDB Get, WriteBatch Get variants
### 2. Array-Based MultiGet with PinnableSlice (30% allocation reduction)
- Switched MultiGet operations to use optimized array-based RocksDB API with `PinnableSlice`
- Eliminates vector overhead and string allocations
- Affects: MultiGet, Transaction MultiGet, TransactionDB MultiGet variants
### New Zero-Copy APIs
Added high-performance zero-copy functions for applications that can use them:
- `rocksdb_iter_key_slice()` / `value_slice()` / `timestamp_slice()` - Return slices by value (eliminates output param overhead)
- `rocksdb_batched_multi_get_cf_slice()` - Batched get with slice array input
- `rocksdb_slice_t` - ABI-compatible slice type
Note that this pr builds on top of https://github.com/facebook/rocksdb/pull/13911
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14036
Reviewed By: pdillinger
Differential Revision: D85604919
Pulled By: jaykorean
fbshipit-source-id: 7f04b935eea79af1d45b3125a79b90e4706666f6
Summary:
Stress test can fail with assertion inside MultiScan in some reseek scenario. E.g., data block 1 ends with k@9, data block 2 starts with k@8, when a DB iter seeks to k@0 (see option `max_sequential_skip_in_iterations`), MultiScan will land in data block 1 due to https://github.com/facebook/rocksdb/blob/fd0b4e0cf08315f6a644d54d585fe70ca958d4ba/table/block_based/block_based_table_iterator.cc#L1258-L1263.
We can't just use internal key as separator since index block might not use it. I plan to follow up with a fix that never moves `cur_data_block_idx` backward within a MultiScan.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14087
Test Plan: CI and internal crash tests
Reviewed By: anand1976
Differential Revision: D85701668
Pulled By: cbi42
fbshipit-source-id: d3f1aaff40a12be4e3d1b4b7160bf2547f43b849
Summary:
All remote compaction test failures had `mmap_read=1` in common. Unfortunately, the failure hasn't been very reproducible. Try disabling `mmap_read` to see if that shed some light.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14083
Test Plan: CI
Reviewed By: hx235
Differential Revision: D85622229
Pulled By: jaykorean
fbshipit-source-id: bbe9e08efc369813f0fec388c910446089e43650
Summary:
As titled, this fixes some internal crash test failures when UDT is enabled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14085
Test Plan: monitor crash tests.
Reviewed By: anand1976
Differential Revision: D85617949
Pulled By: cbi42
fbshipit-source-id: da6fb21c0ca5803ea24e8daf7de8558321babcf4
Summary:
Due to some internal requirements, what's being used for`$SSH` and `$SCP` has changed and it broke the regression test. (e.g. tarball streaming to remote host no longer works)
Minor behavior changes to the script to make the internal workflow work.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14079
Test Plan:
```
./tools/regression_test.sh
```
Meta Internal automation
Reviewed By: pdillinger
Differential Revision: D85502798
Pulled By: jaykorean
fbshipit-source-id: d294c2ee47661fbe368ccc318062e891f3ac7c81
Summary:
The TTL-based WAL archive cleanup logic could incorrectly delete an archived WAL if the system clock moved backwards between the last write to that WAL and `WALManager::PurgeObsoleteWALFiles()`. This happened due to unsigned underflow in subtraction of two wall clock based timestamps: `now_seconds - file_m_time`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14016
Test Plan: unit test repro
Reviewed By: pdillinger
Differential Revision: D83879806
Pulled By: hx235
fbshipit-source-id: 643e7f623c6b5c31711565854314cfd6cbbcf3a7
Summary:
Fixed a missing CV signal when `FindObsoleteFiles()` decides there is nothing to purge and then decrements `pending_purge_obsolete_files_` to zero. This bug could cause `DB::GetSortedWalFiles()` to hang, at least.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14069
Test Plan: unit test repro
Reviewed By: hx235
Differential Revision: D85453534
Pulled By: cbi42
fbshipit-source-id: cf5cfe7f5087459ca1f1f28ce81ea6afc84178f0
Summary:
* Address feedback from https://github.com/facebook/rocksdb/issues/14040
* Add additional test for MultiScan
* Fix a bug when del range and data are in same file for multi-scan
* Rewrite the cases need to be handled in SeekMultiScan
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14055
Test Plan: Unit test
Reviewed By: cbi42, anand1976
Differential Revision: D84851788
Pulled By: xingbowang
fbshipit-source-id: 0f69632733afb99685f6341badbf239681010c38
Summary:
Linter complains like this
```
void foo(Arg parameter_name) {}
void bar() {
Arg a;
foo(/*some_other_name=*/ a); // Wrong! Comment/parameter name mismatch
foo(/*parameter_name=*/ a); // This is OK; the names match.
}
```
```
Argument name in comment (`read_only`) does not match parameter name (`unchanging`).
```
This used to be warning, but now treated as an error :(
Fixing a few other linter warnings before they become errors in the future.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14074
Test Plan: CI
Reviewed By: archang19
Differential Revision: D85370353
Pulled By: jaykorean
fbshipit-source-id: 20e96aad740d516a29c0424282674e655f99c0a2
Summary:
When a standalone range deletion file is ingested in L0, currently it is compacted with any overlapping L0 files. This is not desirable when we ingest new data on top of the range deletion file. This PR fixes the compaction picking logic to only consider L0 files older than the standalone range deletion file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14061
Test Plan: added a new unit test and updated an existing one.
Reviewed By: xingbowang
Differential Revision: D84930780
Pulled By: cbi42
fbshipit-source-id: 65f4403ccb40ba964b9e65b09e2f7f7efebe81df
Summary:
**Context/Summary:**
- Add resumable compaction to stress test with adaptive progress cancellation
- Add fault injection to remote compaction
- Fix a real minor bug in a couple testing framework bugs with remote compaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14041
Test Plan: - Rehearsal stress test, finding bugs for https://github.com/facebook/rocksdb/pull/13984 effectively and did not create new failures.
Reviewed By: jaykorean
Differential Revision: D84524194
Pulled By: hx235
fbshipit-source-id: 42b4264e428c6739631ed9aa5eb02723367510bc
Summary:
With cache hit and compiler option optimization, the compilation time build time is reduced from 40 min to 2 min. Overall build time is reduced from 60 min to less 20 minutes on cache hit on majority of the source file. On 100% cache miss, it would be around 40 minutes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14064
Test Plan: Github CI
Reviewed By: mszeszko-meta
Differential Revision: D85023882
Pulled By: xingbowang
fbshipit-source-id: 98551880c98f14d36133ff43e6af8c3be94ab465
Summary:
Fixing a nullptr access in multiscan, under following situation.
```
Block Based Table: blk1:[k1,k2], blk2:[k3, k8], blk3:[k9]
Scan ranges: [k1, k4), [k5,k6), [k7, k10)
Prepared block ranges: [0,2], [2,2], [1,3]
```
1. Seek key k1 on the first range, read key k1, k2.
2. Seek key k4 on the 2nd range, blocks 0,1 would be unpinned.
3. Seek key k9, block 1 would be accessed, but it is unpinned, which trigger assert failure in debug mode and nullptr access on release build.
This fix changes how blocks are unpinned. It is now only unpinning the block, when the cur_data_block_idx has passed it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14062
Test Plan:
Unit Test
rand_seed 304010984 on UserDefinedIndexStressTest
Reviewed By: cbi42
Differential Revision: D84976410
Pulled By: xingbowang
fbshipit-source-id: 6b99bf85fc9d4108c5267ae77be77ccfe08923cd
Summary:
**Problem:** RocksDB was making unnecessary prefetch system calls on file systems that don't support prefetch operations, potentially leading to wasted CPU cycles.
**Fix:** Add kFSPrefetch to FSSupportedOps enum to allow file systems to indicate prefetch support capability. File systems can now opt out of prefetch calls by not setting this field.
**Backwards compatibility:** File systems that don't override SupportedOps() continue to receive prefetch calls exactly as before. Only file systems that explicitly opt out by not setting kFSPrefetch will avoid the calls.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13917
Test Plan:
- Added a new test in block_based_table_reader.
- Run existing tests: ```make prefetch_test && ./prefetch_test```
Reviewed By: anand1976
Differential Revision: D81607145
Pulled By: nmk70
fbshipit-source-id: 3bbefa05919034e8776ea4e4540cdc695cdc6d3f
Summary:
Currently we return `File is too large for PlainTableReader!` when the file size exceeds our pre-defined constant. There was a request to have the file size information logged when this error is returned.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14056
Reviewed By: nmk70
Differential Revision: D84834869
Pulled By: archang19
fbshipit-source-id: 8f332b6a31d51f320c7e2db06ad49f50798ff70e
Summary:
* Reduce build time of folly from 45m~1hr down to 25m. This is achieved by caching folly build artifact from previous build.
* Reduce windows build time of folly from 1hr 15m down to 50m. This is done by increase windows build machine size.
* Fix build on macos on other macos target.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14057
Test Plan: github CI
Reviewed By: archang19, nmk70
Differential Revision: D84848041
Pulled By: xingbowang
fbshipit-source-id: 00306750737070e7e446ee436d607ed6ecae79ae
Summary:
We simulate remote compaction in our stress test by running a separate set of worker threads to run compactions. In reality, these remote compactions run on a different host or (at least in a different process) where we cannot share the TableFactory and BlockCache with the main DB process.
To make this simulated remote compaction closer to reality, create a new TableFactory for each remote compaction in stress test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14050
Test Plan:
```
python3 -u tools/db_crashtest.py --cleanup_cmd='' --simple blackbox --remote_compaction_worker_threads=8 --interval=10
```
Reviewed By: hx235
Differential Revision: D84775656
Pulled By: jaykorean
fbshipit-source-id: d6203fcbe0eca3539e008a19fd47b742553537ed
Summary:
We are adding more and more tests, so we need to increase the number of shards in macos build to reduce overall CI time.
macos-15-xlarge image is ARM, which has 5 vCPU cores, but is still 50% faster than the intel x86 12 vCPU.
Test time reduced from 1h 37m to 14m.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14048
Reviewed By: archang19
Differential Revision: D84741917
Pulled By: xingbowang
fbshipit-source-id: 9ba9bd696d3b2152f11dec2fb4280572b98233d5
Summary:
Currently in BlockBasedTableIterator's Prepare(), the index lookup for a MultiScan range is expected to return atleast 1 data block (unless UDI is in use). This is because there's an implicit assumption that only ranges intersecting with the keys in the file will be prepared. This assumption, however, doesn't hold if there are range deletions and the smallest and/or largest keys in the file extend beyond the keys in the file. The LevelIterator prunes the MultiScan ranges based on the smallest/largest key, so its possible for a range to only overlap the range deletion portion of the file and not overlap any of the data blocks. Furthermore, the BlockBasedTableIterator is now much more forgiving of Seek to targets outside of prepared ranges after https://github.com/facebook/rocksdb/issues/14040 .
Keeping the above in mind, this PR removes the check in BlockBasedTableIterator for non-empty index result. It adds assertions in LevelIterator to verify that ranges are being properly pruned. Another side effect is we can no longer rely solely on a scan range having 0 data blocks (i.e cur_scan_start_idx >= cur_scan_end_idx) to decide if the iterator is out of bound. We can only do so for all but the last range prepared range.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14046
Test Plan:
1. Add unit test in db_iterator_test
2. Run crash test
Reviewed By: xingbowang
Differential Revision: D84623871
Pulled By: anand1976
fbshipit-source-id: 2418e629f92b1c46c555ddea3761140f700819e4
Summary:
The current seek key validation is too strict. This change relaxes it at block iterator level, and add additional check at DB iterator level. The new contract is that when MultiScan is used, after prepared is called, each following seek must seek the start key of the prepared scan range in order. Otherwise, the iterator is set with error status.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14040
Test Plan: Unit test
Reviewed By: anand1976
Differential Revision: D84292297
Pulled By: xingbowang
fbshipit-source-id: 7b31f727e67e7c0bfc53c2f9a6552e0c3d324869
Summary:
Multi scan crash/stress tests are failing when skip_stats_update_on_db_open is true, because LevelIterator::Prepare relies on these stats in FileMetaData to make decisions. Disable it in crash tests until the proper fix is ready.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14039
Reviewed By: archang19
Differential Revision: D84280059
Pulled By: anand1976
fbshipit-source-id: f9f58b94c24d1f455432b05f3bf97f25c7233e3c
Summary:
**Context/Summary:**
There is no way to tag or rate-limit write IO occurs during FlushWAL() with priority. Under `Options::manual_wal_flush=true`, it is the major source of write IO during user writes so we decide to add that support. A new option struct `FlushWALOptions` is introduced to avoid making the API ugly for future new fields.
Also, we can't use the WriteOptions (https://github.com/facebook/rocksdb/blob/main/include/rocksdb/options.h#L2293-L2302 i) since is associated with that particular Put/Merge/.. associated with that option but FlushWAL() can happen after that write. There is no way to carry that write option over in RocksDB. I also avoided using the WriteOptions since it's mostly for live write.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14037
Test Plan: New UTs `TEST_P(DBRateLimiterOnManualWALFlushTest, ManualWALFlush)`
Reviewed By: archang19
Differential Revision: D84193522
Pulled By: hx235
fbshipit-source-id: 18feb5235672010d19a101ce52c8abdcc4a789f2
Summary:
- Include Status in RemoteCompactionResultMap in SharedState so that we can directly check the status of the remote compaction in `DbStressCompactionService::Wait()`
- If result is empty, populate the result with the status that was returned from `GetRemoteCompactionResult()` so that the status can be bubbled up to the primary (main db thread)
- Get rid of Timeout in `Wait()`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14022
Test Plan:
With fall-back
```
python3 -u tools/db_crashtest.py blackbox --remote_compaction_worker_threads=8 --remote_compaction_failure_fall_back_to_local=1
```
Without fall-back
```
python3 -u tools/db_crashtest.py blackbox --remote_compaction_worker_threads=8 --remote_compaction_failure_fall_back_to_local=0
```
Reviewed By: hx235
Differential Revision: D83789172
Pulled By: jaykorean
fbshipit-source-id: 08f710c4ece5fcc1d4b95b3f9c353831882851b7
Summary:
Fix the binutils truncated download issue by switching to wget in the folly build scripts for downloading dependencies.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14030
Test Plan: make build_folly
Reviewed By: jaykorean
Differential Revision: D84033126
Pulled By: anand1976
fbshipit-source-id: bc6706d7e57c97d6edff149a965aa12c7959825f
Summary:
MultiScan currently doesn't handle delete range properly. In this specific case, a file with only delete range will have an empty index resulting in BlockBasedTableIterator wrongly thinking that a scan doesn't intersect the file due to empty result.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14026
Test Plan: Run crash test
Reviewed By: xingbowang
Differential Revision: D83881266
Pulled By: anand1976
fbshipit-source-id: dc1faa494ea23f36391b700dd1ee0430a1f20ac5
Summary:
When there is an ingested SST file that only contains delete range operations, MultiScan may return error "Scan does not intersect with file". This is due to file selection during Prepare uses the file smallest and largest key without considering whether there is any key in the file. This is only a temporary fix.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14028
Test Plan: Unit test
Reviewed By: anand1976
Differential Revision: D83986964
Pulled By: xingbowang
fbshipit-source-id: e0961ca854e2062c2457be4324817ba073ae785d
Summary:
Implicit reseek in the middle of an iteration is not supported with MultiScan. Avoid this for now in crash tests by setting max_sequential_skip_in_iterations to an absurdly high value.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14015
Reviewed By: xingbowang
Differential Revision: D83761612
Pulled By: anand1976
fbshipit-source-id: 16f4e856374b79170c0a79c11c275cbb0fc83a70
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14024
Fix some typo found along the codebase
Reviewed By: pdillinger
Differential Revision: D83789182
fbshipit-source-id: feb24d7d47a6faaf735fcfd50dd3ecce4a6c8cd5
Summary:
When inplace_update_support and memtable_veirfy_per_key_checksum_on_seek are enabled at the same time, it would cause data race in memtable.
inplace_update_support allows key/value pair in place update in memtable.
memtable_veirfy_per_key_checksum_on_seek performs key checksum verification during seek. It is possible that one thread is updating the key/value pair in place, while another thread is reading the key/value pair for checksum verification during seek.
Therefore, there these 2 configurations could not be enabled at the same time
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14023
Test Plan: local stress test run stops reporting race condition
Reviewed By: anand1976
Differential Revision: D83812322
Pulled By: xingbowang
fbshipit-source-id: 6cb9f0f3faa8deba97305bfe87266f2fe78e0501
Summary:
In RocksDB 10.6 with https://github.com/facebook/rocksdb/issues/13805, due to inaccurate testing of an async system, it went undetected at the time that LZ4 compression was using more CPU despite making a change to reuse stream objects which dramatically improved LZ4HC compression efficiency.
This change switches to using a basic LZ4 compress API which appears to be faster than all of these:
* Legacy behavior of creating LZ4_stream_t for each compression
* 10.6-10.7 behavior of re-using streams between compressions for the same file (with stream-as-WorkingArea)
* using LZ4's extState APIs without streams (with extState-as-WorkingArea) (data not shown in below results)
Also in this PR: more improvements to sst_dump --recompress, which is arguably the best SST construction benchmark right now since db_bench seems to be so noisy due to backgroun flush+compaction, even with no compaction (FIFO). Streamlined some output and added a SST read time test, mostly for decompression performance.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14017
Test Plan:
Performance test using sst_dump --recompress with newer sst_dump back-ported to 10.5:
```
./sst_dump --command=recompress --compression_types=kLZ4Compression
test5.sst --compression_level_from=-6 --compression_level_to=-1
```
and with default compression level.
10.5:
```
Cx level: -6 Cx size: 61608137 Write usec: 880404
Cx level: -5 Cx size: 60793749 Write usec: 840903
Cx level: -4 Cx size: 58134030 Write usec: 836365
Cx level: -3 Cx size: 55193773 Write usec: 857113
Cx level: -2 Cx size: 54013891 Write usec: 855642
Cx level: -1 Cx size: 50400393 Write usec: 865194
Cx level: 32767 Cx size: 50400393 Write usec: 886310
```
Before this change (showing the regression, more time, from 10.6:
```
Cx level: -6 Cx size: 61608137 Write usec: 933448
Cx level: -5 Cx size: 60793749 Write usec: 893826
Cx level: -4 Cx size: 58134030 Write usec: 891138
Cx level: -3 Cx size: 55193773 Write usec: 898461
Cx level: -2 Cx size: 54013891 Write usec: 897485
Cx level: -1 Cx size: 50400393 Write usec: 936970
Cx level: 32767 Cx size: 50400393 Write usec: 958764
```
After this change (faster than both the above):
```
Cx level: -6 Cx size: 63641883 Write usec: 874190
Cx level: -5 Cx size: 58860032 Write usec: 834662
Cx level: -4 Cx size: 57150188 Write usec: 832707
Cx level: -3 Cx size: 58791894 Write usec: 850305
Cx level: -2 Cx size: 53145885 Write usec: 839574
Cx level: -1 Cx size: 49809139 Write usec: 845639
Cx level: 32767 Cx size: 49809139 Write usec: 875199
```
Similar tests with dictionary compression show essentially no difference (need to use stream APIs and reuse doesn't seem to matter). LZ4HC also unaffected (still improved vs. 10.5)
Reviewed By: hx235
Differential Revision: D83722880
Pulled By: pdillinger
fbshipit-source-id: 30149dd187686d5dd98321e6aa7d74bd7653a905
Summary:
Pad block based table based on super block alignment
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13909
Test Plan:
Unit Test
No impact on perf observed due to change in the inner loop of flush.
upstream/main branch 202.15 MB/s
```
for i in `seq 1 10`; do ./db_bench --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 >> /tmp/x1 2>&1; grep fillseq /tmp/x1 | grep -Po "\d+\.\d+ MB/s" | grep -Po "\d+\.\d+" | awk '{sum+=$1} END {print sum/NR}'
```
After the change without super block alignment 203.44 MB/s
```
for i in `seq 1 10`; do ./db_bench --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 >> /tmp/x1 2>&1
```
After the change with super block alignment 204.47 MB/s
```
for i in `seq 1 10`; do ./db_bench --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 --super_block_alignment_size=131072 --super_block_alignment_max_padding_size=4096 >> /tmp/x1 2>&1;
```
Reviewed By: pdillinger
Differential Revision: D83068913
Pulled By: xingbowang
fbshipit-source-id: eecd65088ab3e9dbc7902aab8c2580f1bc8575df
Summary:
### Context/Summary:
Flow of resuming: DB::OpenAndCompact() -> Compaction progress file -> SubcompactionProgress -> CompactionJob
Flow of persistence: CompactionJob -> SubcompactionProgress -> Compaction progress file -> DB that is called with OpenAndCompact()
This PR focuses on SubcompactionProgress -> CompactionJob and CompactionJob -> SubcompactionProgress -> Compaction progress file. For now only single subcompaction is supported as OpenAndCompact() does not partition compaction anyway.
The actual triggering of progress persistence and resuming (i.e, integration) is through DB::OpenAndCompact() in the upcoming PR.
**Resume Flow**
1. input_iter->Seek(next_internal_key_to_compact) // Position iterator
2. ReadTableProperties() // Validate existing outputs
3. RestoreCompactionOutputs() in CompactionOutputs // Rebuild output file metadata
4. Restore critical statistics about processed input and output records count for verification later
5. AdvanceFileNumbers() // Prevent file number conflicts
6. Continue normal compaction from positioned iterator or fallback to not resuming compaction in limited case or fail the compaction entirely
**Persistence Strategy**
1. When: At each SST file completion (FinishCompactionOutputFile()). This is the simplest but most expensive frequency. See below for benchmarking and potential follow-up items
2. What: Serialize, write and sync the in-memory SubcompactionProgress to a dedicated manifest-like file
3. For simplicity: Only persist at "clean" boundaries (no overlapping user keys, no range deletions, no timestamp for now)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13983
Test Plan:
- New unit test in CompactionJob level to cover basic compaction progress resumption
- Existing UTs and stress/crash test to test no correctness regression to existing compaction code
- Run benchmark to ensure no performance regression to existing compaction code
```
./db_bench --benchmarks=fillseq[-X10] --db=$db --disable_auto_compactions=true --num=100000 --value_size=25000 --compression_type=none --target_file_size_base=268435456 --write_buffer_size=268435456
```
Pre-PR:
fillseq [AVG 10 runs] : 45127 (± 799) ops/sec; 1076.6 (± 19.1) MB/sec
fillseq [MEDIAN 10 runs] : 45375 ops/sec; 1082.5 MB/sec
Post-PR (regressed 0.057%, ignorable)
fillseq [AVG 10 runs] : 45101 (± 920) ops/sec; 1076.0 (± 22.0) MB/sec
fillseq [MEDIAN 10 runs] : 45385 ops/sec; 1082.8 MB/sec
Reviewed By: jaykorean
Differential Revision: D82889188
Pulled By: hx235
fbshipit-source-id: 8553fd478f134969d331af2c5a125b94bd747268
Summary:
This method will be used to improve the compaction logic by accounting for the tail size, in addition to the data size, when determining when to cut a file.
Problem: Currently the file cutting logic only considers data size when determining where to cut a file, failing to reserve space for index and filter blocks that are added when the file is finalized.
Key changes:
- Add EstimateCurrentIndexSize() to IndexBuilder interface
- Implement in ShortenedIndexBuilder with buffer that accounts for the next index entry. The buffer addresses under-estimation where the current index size doesn't account for the next index entry associated with the data block currently being built. The 2x multiplier bounds the estimate in the right direction and handles outlier cases with large keys.
- Add num_index_entries_ member to track added index entries (== data blocks emitted). This is thread-safe since it's updated/read in the serialized emit step.
Next steps:
- Partitioned index size estimation implementation
- Update compaction file cutting logic to consider index size estimation
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14010
Test Plan: Added a new test class with unit tests for new builder size estimation across all IndexBuilder implementations.
Reviewed By: pdillinger
Differential Revision: D83501741
Pulled By: nmk70
fbshipit-source-id: d58fc2a9e92e12a162f6244d4abd707a9c9e1885
Summary:
This PR fixes a bug in how MultiScan handled a scan range limit falling in the key range between files. The bug was in LevelIterator, where Prepare() relied on FindFile to determine the lower bound file for the range limit. FindFile returns the smallest file index with `range.limit < file.largest_key`. However, that doesn't guarantee that the range overlaps the file, as the `range.limit` could be smaller than `file.smallest_key`.
This also fixes a bug in BlockBasedTableIterator of Valid() returning true even if status() returned error. This was exposed by the previous bug.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14011
Test Plan: Add unit tests in db_iterator_test and table_test
Reviewed By: cbi42
Differential Revision: D83496439
Pulled By: anand1976
fbshipit-source-id: a9d2d138d69d0c816d9f4160a984b273d00d683f
Summary:
Pretty self-explanatory from the changes, including re-arranging the "COOL" entries for easier tracking of which values are used.
I'm not touching the TICKER_ENUM_MAX issue because IIRC we've gotten in trouble in the past for changing any Java ticker values.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14012
Test Plan: CI, sufficient prompts to get AI to discover the known issues relayed by hx235, to help ensure we found any other outstanding issues.
Reviewed By: hx235
Differential Revision: D83497503
Pulled By: pdillinger
fbshipit-source-id: ec0bd7e28188e0430fb03fc5bd79c2ed7b28f3ad
Summary:
Pass the comparator to UDI interface for both reader and builder.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14001
Test Plan: Unit test
Reviewed By: anand1976
Differential Revision: D83339943
Pulled By: xingbowang
fbshipit-source-id: 7f6541776b0995260e28224329f0cca37f13b3d4
Summary:
currently BlockBasedTableIterator::Prepare() fails the iterator with non-ok status if an out-of-range scan option is detected. This is due to the interaction between LevelIterator and BlockBasedTableIterator, see added comment above BlockBasedTableIterator::Prepare(). This can fail stress test for L0 files since it doesn't use LevelIterator and scan options are not pruned. This PR fixes this by adding an internal option to MultiScanArgs that enables this check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13995
Test Plan:
- new unit test
- stress test that fails before this pr: `python3 -u ./tools/db_crashtest.py whitebox --iterpercent=60 --prefix_size=-1 --prefixpercent=0 --readpercent=0 --test_batches_snapshots=0 --use_multiscan=1 --read_fault_one_in=0 --kill_random_test=88888 --interval=60 --multiscan_use_async_io=0 --mmap_read=0 --level0_file_num_compaction_trigger=20`
Reviewed By: anand1976
Differential Revision: D83166088
Pulled By: cbi42
fbshipit-source-id: 241a7d43c8c00d9a98eea0cabb03d2174d51aae5
Summary:
There can be concurrent reads/writes to fields in `IODebugContext`. One example we have seen is for the `cost_info` field which is of type `std::any`. In fact, in RocksDB's async MultiRead implementation, the same `IODebugContext` is re-used across separate async read requests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13993
Test Plan: Update code which reads/writes to `cost_data` to first acquire shared/exclusive lock on the `mutex` field. There should not be any race conditions when async MultiRead is used.
Reviewed By: pdillinger
Differential Revision: D83091423
Pulled By: archang19
fbshipit-source-id: 4db86d33cf162ed39114b1cd115fcd8964c8ff9b
Summary:
Remove the restriction of only using BytewiseComparator(). In a follow on PR, the UDI interface will be updated to take the Comparator as a parameter.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13999
Test Plan: Add a unit test in table_test.cc
Reviewed By: cbi42
Differential Revision: D83179747
Pulled By: anand1976
fbshipit-source-id: 60222533c71022aa0701ac61c39268d36ca86338
Summary:
In https://github.com/facebook/rocksdb/issues/13964 I changed an expensive DEBUG check in ~AutoHyperClockTable to only run in ASAN builds. It's still expensive so I'm modifying it to scan only about one page beyond what we expect to have written to the anonymous mmap, rather than scanning the whole thing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13998
Test Plan: manually checked that lru_cache_test running time went from 5.0s to 4.0s after the change. Verified that existing unit test ClockCacheTest.Limits uses the full anonymous mmap to be sure it is sized as expected, by temporarily breaking AutoHyperClockTable::Grow() to allow slightly exceeding the anonymous mmap size.
Reviewed By: cbi42
Differential Revision: D83178493
Pulled By: pdillinger
fbshipit-source-id: a2bf093e98bf68b540c073800be7e193021f2692
Summary:
This combination causes MultiScan iteration to fail due to internal reseek by the iterator.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13992
Reviewed By: cbi42
Differential Revision: D83094631
Pulled By: anand1976
fbshipit-source-id: 96410747d88de391e6d65857d39063d4fb113d65
Summary:
Fix the bug in Improve random seed override support in stress test.
The Bug:
`parser.parse_known_args()` is used to parse command line argument. When it is called without any argument, it uses sys.argv as input parameter. In sys.argv, the first argument is the command itself, so parser.parse_known_args skip the first argument. Meantime, the return value `remain_argv` of `parser.parse_known_args()` does not contain the command itself. When `remain_arg` replaces `sys.argv`, the first argument is treated as the command itself, which is skipped by `parser.parse_known_args()`. In the internal stress test tool, the first argument is `--stress_cmd`, therefore, it is skipped. Instead, the default value `./stress_db` is used. This is why `./stress_db` showed up in the error message. This is also why it works in local, as stress_db is located in the local folder.
The Fix:
When `parser.parse_known_args()` is called first time, the remain_argv is saved as a global variable. It is used in the second call of the `parser.parse_known_args(remain_argv)`. When argument is passed to `parser.parse_known_args` directly, the first argument will not be skipped.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13991
Test Plan:
The the value of first argument `--stress_cmd` is parsed correctly, and shown up in the error message.
```
/usr/local/bin/python3 -u tools/db_crashtest.py --stress_cmd=/data/sandcastle/boxes/trunk-hg-full-fbsource/buck-out/v2/gen/fbcode/d7db8b24dd42e2db/internal_repo_rocksdb/repo/__db_stress__/db_stress --cleanup_cmd='' --simple blackbox --print_stderr_separately
Start with random seed 11107847853133580500
Running blackbox-crash-test with
interval_between_crash=120
total-duration=6000
Use random seed for iteration 8577470137673434540
Traceback (most recent call last):
File "/home/xbw/workspace/ws1/rocksdb/tools/db_crashtest.py", line 1650, in <module>
main()
File "/home/xbw/workspace/ws1/rocksdb/tools/db_crashtest.py", line 1639, in main
blackbox_crash_main(args, unknown_args)
File "/home/xbw/workspace/ws1/rocksdb/tools/db_crashtest.py", line 1358, in blackbox_crash_main
hit_timeout, retcode, outs, errs = execute_cmd(cmd, cmd_params["interval"])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/xbw/workspace/ws1/rocksdb/tools/db_crashtest.py", line 1294, in execute_cmd
child = subprocess.Popen(cmd, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/fbcode/platform010/lib/python3.12/subprocess.py", line 1028, in __init__
self._execute_child(args, executable, preexec_fn, close_fds,
File "/usr/local/fbcode/platform010/lib/python3.12/subprocess.py", line 1957, in _execute_child
raise child_exception_type(errno_num, err_msg, err_filename)
FileNotFoundError: [Errno 2] No such file or directory: '/data/sandcastle/boxes/trunk-hg-full-fbsource/buck-out/v2/gen/fbcode/d7db8b24dd42e2db/internal_repo_rocksdb/repo/__db_stress__/db_stress'
```
Reviewed By: hx235
Differential Revision: D83068960
Pulled By: xingbowang
fbshipit-source-id: 28334d38a444c6f8525444e15f460ec6b257ef38
Summary:
Return a failure status for multi scan if Prepare fails, or if the scan options are unsupported, instead of falling back on a regular scan. This PR also fixes a bug in LevelIterator that caused max_prefetch_size to be ignored.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13974
Test Plan: Add new test in db_iterator_test and table_test
Reviewed By: xingbowang
Differential Revision: D82843944
Pulled By: anand1976
fbshipit-source-id: f12756c40ebd38d8d4e4425e97438b6e766a4663
Summary:
**Context/Summary**
This reverts commit 73432a3f36. This is due to it mysteriously fails our internal CI running with this change to db_crashtest.py. The root-cause is unknown but the error only reproed with this commit frequently but not the one before it. The error message appears to be the command parsing leading to the db_stress binary can't be found
```
Traceback (most recent call last):
File "/data/sandcastle/boxes/trunk-hg-full-fbsource/fbcode/internal_repo_rocksdb/repo/tools/db_crashtest.py", line 1638, in <module>
main()
File "/data/sandcastle/boxes/trunk-hg-full-fbsource/fbcode/internal_repo_rocksdb/repo/tools/db_crashtest.py", line 1627, in main
blackbox_crash_main(args, unknown_args)
File "/data/sandcastle/boxes/trunk-hg-full-fbsource/fbcode/internal_repo_rocksdb/repo/tools/db_crashtest.py", line 1347, in blackbox_crash_main
hit_timeout, retcode, outs, errs = execute_cmd(cmd, cmd_params["interval"])
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/data/sandcastle/boxes/trunk-hg-full-fbsource/fbcode/internal_repo_rocksdb/repo/tools/db_crashtest.py", line 1283, in execute_cmd
child = subprocess.Popen(cmd, stderr=subprocess.PIPE, stdout=subprocess.PIPE)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/fbcode/platform010/lib/python3.12/subprocess.py", line 1028, in __init__
self._execute_child(args, executable, preexec_fn, close_fds,
File "/usr/local/fbcode/platform010/lib/python3.12/subprocess.py", line 1957, in _execute_child
raise child_exception_type(errno_num, err_msg, err_filename)
FileNotFoundError: [Errno 2] No such file or directory: './db_stress'
```
**Test plan**
- Rehearsal crash test
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13989
Reviewed By: xingbowang
Differential Revision: D83010751
Pulled By: hx235
fbshipit-source-id: d8cfc70564074065b6bb8a3986d6c1011064dd5e
Summary:
This is causing some internal failure, we decide to revert this for now until we have a proper fix.
This reverts commit 961880b458.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13987
Reviewed By: anand1976
Differential Revision: D82990294
Pulled By: cbi42
fbshipit-source-id: 5f5b4d18d0afe47599738d27e11e3eb2d08d88a0
Summary:
**Context**
Resuming compaction is designed to periodically record the progress of an ongoing compaction and can resume from that saved progress after interruptions such as cancellation, database shutdown, or crashes.
This PR introduces the data structures needed to store subcompaction progress in memory, along with serialization and deserialization support to persist and parse this progress to/from "a manifest-like compaction progress file" (the actual creation of such file is in upcoming PRs).
Flow of resuming: DB::OpenAndCompact() -> Compaction progress file -> SubcompactionProgress -> CompactionJob
Flow of persistence: CompactionJob -> SubcompactionProgress -> Compaction progress file -> DB that is called with OpenAndCompact()
**Summary**
Progress represented by `SubcompactionProgress` will be tracked at the scope of a subcompaction, which is the smallest independent unit of compaction work.
The frequency of recording this progress is once every N compaction output files (to be detailed in future PRs).
When recording, all fields, except for the output files metadata in `SubcompactionProgress`, will directly overwrite the corresponding fields from the last saved progress (See `SubcompactionProgress` and `SubcompactionProgressBuilder` for more).
As a bonus, this PR refactors the file metadata encoding and decoding utilities into two static helper functions, EncodeToNewFile4() and DecodeNewFile4From(), to support subcompaction progress usage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13928
Test Plan:
- Added various `SubcompactionProgressTest` unit tests in version_edit_test.cc to verify basic serialization/deserialization and forward compatibility handling
- Existing UTs and stress/crash test
**Follow up:**
- Move output entry number and file verification to after each file creation so we can remove kNumProcessedOutputRecords persistence support and make resuming compaction work with `paranoid_file_checks=true` (by default false). Output verification will be done before persistence of progress. As long as this follow-up is done before the landing of the integration PR to create the progress file, we can change the manifest-like compaction progress file format freely.
Reviewed By: jaykorean
Differential Revision: D81986583
Pulled By: hx235
fbshipit-source-id: b42766da7d9c2e2f596c892d050c753238d1039f
Summary:
for MultiScan and UDI we start to use bound check from index iterator, so removing this assert here.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13988
Test Plan: existing test
Reviewed By: hx235
Differential Revision: D82993180
Pulled By: cbi42
fbshipit-source-id: 442b2e83cb3aef96fc1a825bf733af9ce59c21c1
Summary:
It is useful to be able to specify output temperatures in the CompactFiles API. For example it may be useful to store small L0 files produced by flushes locally, while larger intra-L0 compactions can store the compacted L0 file remotely.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13955
Test Plan: New unit tests
Reviewed By: jaykorean
Differential Revision: D82492503
Pulled By: joshkang97
fbshipit-source-id: e1225fe572a15d7c5c30a265762b048a4a9e7f0b
Summary:
- updated release note
- updated version to 10.8 in version.h
- added 10.7 to check_format_compatible.sh
- did not updated folly commit hash due to some build failure.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13980
Reviewed By: xingbowang
Differential Revision: D82882035
Pulled By: cbi42
fbshipit-source-id: b5e0e78570fdd492d592ee77bd3901e4b39c25fb
Summary:
the test did not consider the ingestion_option settings that can result in different error message. This PR fixes the relevant check and ensure we have enough randomness in this test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13979
Test Plan: `gtest-parallel --repeat=20 --workers=20 ./external_sst_file_test --gtest_filter="*VaryingOptions/IngestDBGeneratedFileTest2.NonZeroSeqno/*"`
Reviewed By: hx235
Differential Revision: D82873439
Pulled By: cbi42
fbshipit-source-id: b0d74bf26a502ca3db59b4a0ea9717bf7d027400
Summary:
Start the process of migrating the HCC implementation over to my new system of "bit field atomics" to clean up the code. Here I took on the simplest of the three "bit field atomic" formats in HCC, but ended up moving some things around to end up with less plumbing of definitions and values overall.
In the process, updated BitFields to use the CRTP pattern to simplify some things (see updated example, etc.)
https://en.wikipedia.org/wiki/Curiously_recurring_template_pattern
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13965
Test Plan: existing tests. ClockCacheTest.ClockEvictionEffortCapTest caught a regression during my development, and the crash test has a history of finding subtle HCC bugs.
Reviewed By: xingbowang
Differential Revision: D82669582
Pulled By: pdillinger
fbshipit-source-id: b73dd47361cbe9fbd334413dd4ce01b3c667159e
Summary:
longtime wanted e.g. for easy tab-completion, now implemented
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13978
Test Plan: pretty good unit test updates, manual testing
Reviewed By: cbi42
Differential Revision: D82857671
Pulled By: pdillinger
fbshipit-source-id: d2b63b7d15e61ebf22c58a6ecd3003311e2d03cb
Summary:
* There was a bug where the compression manager would actually not be used for recompress because the options passed to SstFileDumper were not respected. That is now fixed by respecting the Options.
* Refactored SstFileDumper not to take explicit options that could naturally be embedded in Options.
* Report compressed and uncompressed data block sizes (and ratio) instead of total file size (without a useful ratio). Needed to add a new table property to support that.
* Allow --block_size instead of --set_block_size to be consistent with other tools
* Allow --compression_level as shorthand for both _from and _to options, for simplicity and consistency with other tools
* Support --compression_parallel_threads option
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13977
Test Plan:
* sst_dump manual testing
* TableProperties unit tests updated
* Made it much easier to detect when a functional change requires an update to ParseTablePropertiesString() (rather than causing cryptic downstream failures)
Reviewed By: cbi42
Differential Revision: D82841412
Pulled By: pdillinger
fbshipit-source-id: 8d3421be4d2a3e25b7590cd59d204a3779c2a928
Summary:
Currently in MultiScan we only unpins a block after we scan through it. This PR adds unpinning during Seek to release all blocks pinned by the previous scan range. This is useful when users do not scan through the entire scan range. I plan to follow up with support for aborting async IOs from the previous scan.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13972
Test Plan: new test MultiScanUnpinPreviousBlocks validates unpinning behavior
Reviewed By: xingbowang
Differential Revision: D82779504
Pulled By: cbi42
fbshipit-source-id: 17ba7d1e5a6d8ff09ceea57b79c18febfba75584
Summary:
This change adds FFI support for exporting column family checkpoints, basic access to the export/import files metadata, and creating column families by import.
I've been able to successfully use this to [add checkpoint export and import support to `rust-rocksdb`](https://github.com/pcholakov/rust-rocksdb/pull/2), a forked version of which has been successfully used in production for some time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13874
Reviewed By: hx235
Differential Revision: D82343565
Pulled By: jaykorean
fbshipit-source-id: fb4182bdfd5cce10743c021a1ac636fd6ac48df3
Summary:
If there's a static initialization of Options() this could now instantiate an AutoHyperClockTable before kPageSize is initialized. Break the dependency because it's a very minor optimization.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13973
Test Plan: internal CI (not able to reproduce locally)
Reviewed By: hx235
Differential Revision: D82789849
Pulled By: pdillinger
fbshipit-source-id: 3f32b5779a4f56d2071be5aadacda2bf0f4b895d
Summary:
Add a new CF immutable option `paranoid_memory_check_key_checksum_on_seek` that allows additional data integrity validations during seek on SkipList Memtable. When this option is enabled and memtable_protection_bytes_per_key is non zero, skiplist-based memtable will validate the checksum of each key visited during seek operation. The option is opt-in due to performance overhead. This is an enhancement on top of paranoid_memory_checks option.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13902
Test Plan:
* new unit test added for paranoid_memory_check_key_checksum_on_seek=true.
* existing unit test for paranoid_memory_check_key_checksum_on_seek=false.
* enable in stress test.
Performance Benchmark: we check for performance regression in read path where data is in memtable only. For each benchmark, the script was run at the same time for main and this PR:
### Memtable-only randomread ops/sec:
* Value size = 100 Bytes
```
for B in 0 1 2 4 8; do (for I in $(seq 1 50);do ./db_bench --benchmarks=fillseq,readrandom --write_buffer_size=268435456 --writes=250000 --value_size=100 --num=250000 --reads=500000 --seed=1723056275 --paranoid_memory_check_key_checksum_on_seek=true --memtable_protection_bytes_per_key=$B 2>&1 | grep "readrandom"; done;) | awk '{ t += $5; c++; print } END { print 1.0 * t / c }'; done;
```
1. Main: 928999
2. PR with paranoid_memory_check_key_checksum_on_seek=false: 930993 (+0.2%)
3. PR with paranoid_memory_check_key_checksum_on_seek=true:
3.1 memtable_protection_bytes_per_key=1: 464577 (-50%)
3.2 memtable_protection_bytes_per_key=2: 470319 (-49%)
3.3 memtable_protection_bytes_per_key=4: 468457 (-50%)
3.4 memtable_protection_bytes_per_key=8: 465061 (-50%)
* Value size = 1000 Bytes
```
for B in 0 1 2 4 8; do (for I in $(seq 1 50);do ./db_bench --benchmarks=fillseq,readrandom --write_buffer_size=268435456 --writes=250000 --value_size=1000 --num=250000 --reads=500000 --seed=1723056275 --paranoid_memory_check_key_checksum_on_seek=true --memtable_protection_bytes_per_key=$B 2>&1 | grep "readrandom"; done;) | awk '{ t += $5; c++; print } END { print 1.0 * t / c }'; done;
```
1. Main: 601321
2. PR with paranoid_memory_check_key_checksum_on_seek=false: 607885 (+1.1%)
3. PR with paranoid_memory_check_key_checksum_on_seek=true:
3.1 memtable_protection_bytes_per_key=1: 185742 (-69%)
3.2 memtable_protection_bytes_per_key=2: 177167 (-71%)
3.3 memtable_protection_bytes_per_key=4: 185908 (-69%)
3.4 memtable_protection_bytes_per_key=8: 183639 (-69%)
Reviewed By: pdillinger
Differential Revision: D81199245
Pulled By: xingbowang
fbshipit-source-id: e3c29552ab92f2c5f360361366a293fa26934913
Summary:
Force caller of MultiScanArgs to pass comparator. Pass comparator from CF handle to MultiScanArgs in NewMultiScan.
Expand MultiScanArgs unit test with different comparator.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13970
Test Plan: unit test
Reviewed By: cbi42
Differential Revision: D82739270
Pulled By: xingbowang
fbshipit-source-id: e709f4a333ad547c0ba6d24d8fb2b22e50e8a12f
Summary:
**Context/Summary:**
`Status::state` can be nullptr when created with no specific error message. std::strstr on nullptr caused some segfault in our stress test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13968
Test Plan: Monitor stress test
Reviewed By: jaykorean
Differential Revision: D82695541
Pulled By: hx235
fbshipit-source-id: cf08f70163a9ee6c911cdc3a3d79acd3429f0d15
Summary:
After seeing more people hit issues with thrashing small LRUCache shards and AutoHCC running fully in production for a while on a very large service, here I make these updates:
* In the public API, mark the case of `estimated_entry_charge = 0` (which is how you select AutoHCC) as production-ready and generally preferred. That means devoting a lot less space to how to tune FixedHCC (`estimated_entry_charge > 0`) because it is not generally recommended anymore even though in theory it is the fastest (conditional on a fragile configuration).
* In the public API, add more detail about potential problems with LRUCache and explicitly endorse HCC.
* When a default block cache is created, use AutoHCC instead of LRUCache. It's still a 32MB cache but that's just one cache shard for AutoHCC so the risk of issues with small cache shards is dramatically reduced. And a single AutoHCC shard is still essentially wait-free.
* Improve the handling of the hypothetical scenario of a failed anonymous mmap. This is hardly a concern for 64-bit Linux and likely most other OSes. It would in theory be possible to fall back on LRUCache in that case but the code structure makes that annoying/challenging. Instead we crash with an appropriate message.
* Cleaned up some includes
* Fixed some previously unreported leaks (better assertions on HCC perhaps, some subtle behavior changes)
* Added a new mode to cache_bench (detailed below)
* Avoid a particularly costly sanity check in `~AutoHyperClockTable()` even in debug builds so that unit testing, etc., isn't bogged down, except keep it in ASAN build.
Planned follow-up:
* Update HCC implementation to use my new "bit field atomics" API introduced in https://github.com/facebook/rocksdb/issues/13910 to make it easier to read and maintain
Possible follow-up:
* Re-engineer table cache to use AutoHCC also, instead of LRUCache and a single mutex to ensure no duplication across threads. (a) Pad table cache key to 128 bits for AutoHCC. (b) Stripe/shard the no-duplication mutex. (HCC's consistency model is too weak for concurrent threads to use its API to agree on a winner, even if entries could be inserted in an "open in progress" state.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13964
Test Plan:
existing tests. ClockCacheTest.ClockEvictionEffortCapTest caught a regression during my development, and the crash test has a history of finding subtle HCC bugs.
## Performance
Although we've validated AutoHCC performance under high load, etc., before we haven't really considered whether there will be unacceptable overheads for small DBs and CFs, e.g. in unit tests. For this, I have added a new mode to cache_bench: with the -stress_cache_instances=n parameter, it will create and destroy n empty cache instances several times. In the debug build, this found that a particular check in `~AutoHyperClockTable()` was extremely costly for short-lived caches (fixed). Beyond that, we can answer the question of whether it is feasible for a single process to host 1000 DBs each with 1000 CFs with default block cache instances, after moving LRUCache -> AutoHCC, for example:
```
/usr/bin/time ./cache_bench -stress_
cache_instances=1000000 -cache_type=auto_hyper_clock_cache -cache_size=33554432
```
Release build:
Average 9.8 us per 32MB LRUCache creation, 2.9 us per destruction, 24.6GB max RSS (~25KB each)
->
Average 4.3 us per 32MB AutoHCC creation, 4.9 us per destruction, 4.8GB max RSS (~5KB each)
Debug build:
Average 10.9 us per 32MB LRUCache creation, 3.5 us per destruction, 28.7GB max RSS (~29KB each)
->
Average 4.5 us per 32MB AutoHCC creation, 4.9 us per destruction, 4.7GB max RSS (~5KB each)
Despite the anonymous mmaps, it's apparently more efficient for default/small/empty structures. This is likely due to the dramatically low number of cache shards at this size. If we switch to `-stress_cache_instances=10000 -cache_size=1073741824`:
Release build:
Average 10.6 us per 1GB LRUCache, 2.8 us per destruction, 2.3 GB max RSS (~230KB each)
->
Average 130 us per 1GB AutoHCC creation, 153 us per destruction, 1.5 GB max RSS (~150KB each)
Debug build:
Average 11.2 us per 1GB LRUCache, 3.6 us per destruction, 2.4 GB max RSS (~240KB each)
->
Average 130 us per 1GB AutoHCC creation, 150 us per destruction, 1.6 GB max RSS (~160KB each)
Here it's clear that we are paying a price in time for setting up all those mmaps for the good number of cache shards and potential table growth, even though the RSS is well under control. However, I am not concerned about this at all, as it's unlikely to slow down anything notably such as unit tests. Before and after full testsuite runs confirm:
3327.73user 5188.71system 3:38.88elapsed -> 3312.07user 5704.77system 3:41.61elapsed
There is increased kernel time but acceptable. With ASAN+UBSAN:
11618.70user 15671.30system 5:54.68elapsed -> 12595.81user 16159.67system 6:32.77elapsed
Acceptable given that our ASAN+UBSAN builds are not the slowest in CI
Reviewed By: hx235
Differential Revision: D82661067
Pulled By: pdillinger
fbshipit-source-id: ab25c766ca70f2b8664849c2a838b9e1b4e72d3b
Summary:
when ingesting DB generated file with non-zero sequence number, we need smallest seqno of each file for file meta data. To avoid full table scan, we record this information in table property and use it during file ingestion.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13942
Test Plan: new unit test and updated existing unit test.
Reviewed By: hx235
Differential Revision: D82331802
Pulled By: cbi42
fbshipit-source-id: 3009a6801ca7092cd0fde33692db1a13567068a9
Summary:
This PR fixes a bug in BlockBasedTableIterator::Prepare in conjunction with a user defined index (UDI). If the UDI determines a scan range to be empty and thus returns the kOutOfBound iteration result during Seek, the iteration result is not propagated up and Prepare() assumes end of file and aborts the remaining scans. This results in incorrect behavior and unpredictable multi scan results.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13960
Test Plan: Add unit test to table_test.cc
Reviewed By: xingbowang
Differential Revision: D82590892
Pulled By: anand1976
fbshipit-source-id: 8cfaaae2bb1a9509ddf8ec967cb8a8801748413d
Summary:
* Fix compaction/flush CPU usage stats to include CPU usage by parallel compression workers. (Validated with manual db_bench testing.)
* Disable the parallel compression framework when compression is disabled. See new code comment for details, because in theory it could be useful to hide SST write latency, but manual testing with db_bench and -rate_limiter_bytes_per_sec or -simulate_hdd options shows no useful increase in throughput, just more CPU usage.
* Fix some minor clean-up items in the implementation
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13959
Test Plan: Also ran some tests like in https://github.com/facebook/rocksdb/issues/13910 to ensure the new CPU usage tracking did not regress performance, all good.
Reviewed By: xingbowang
Differential Revision: D82556686
Pulled By: pdillinger
fbshipit-source-id: 77c522159a7e6ab0ab6f7fb1d662070a46661557
Summary:
The stress test runs concurrent transactions through many threads at the same time on a shared key space. It is possible that a dead lock or a timeout is detected from the transactiondb layer. When this happens, simply return from the function and continue the test, instead of fail the test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13950
Test Plan: Stress test pass locally with the same random seed from stress test 14723229280871643749.
Reviewed By: hx235
Differential Revision: D82373959
Pulled By: xingbowang
fbshipit-source-id: 5d72e89998171c5844fb22f13d8f061f81014c7d
Summary:
... reporting false positive double-lock on some of the new parallel compression code. Switching from std::condition_variable to condition_variable_any simply changes the FP from double-lock to lock inversion. In addition, leaking ParallelCompressionRep instances to avoid memory location reuse fails to fix the FP reports. Thus, I've decided to disable the watchdog with GCC+TSAN.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13958
Test Plan: local crash test runs could reproduce, now don't reproduce. CLANG TSAN doesn't seem to be reporting the same supposed issues
Reviewed By: xingbowang
Differential Revision: D82555968
Pulled By: pdillinger
fbshipit-source-id: 537fbc3a787f917915a6faf0bdedd1449a7f378a
Summary:
Complete redo of parallel compression in block_based_table_builder.cc to greatly reduce cross-thread hand-off and blocking. A ring buffer of blocks-in-progress is used to essentially bound working memory while enabling high throughput. Unlike before, all threads can participate in compression work, for a kind of work-stealing algorithm that reduces the need for threads to block. This builds on improvements in https://github.com/facebook/rocksdb/pull/13850
Previously, there was either
* parallel_threads==1, the *emit thread* (caller from flush/compaction) doing all the work
* parallel_threads > 1, the emit thread generates uncompressed blocks, `parallel_threads` worker threads compress blocks, and a writer thread writes to the SST file. Total of `parallel_threads + 2` threads participating. (Other bookkeeping in emit and write steps omitted from description for simplicity.)
Now we have either
* parallel_threads==1 (same), the emit thread doing all the work
* parallel_threads > 1, the emit thread generates uncompressed blocks and can take up compression work when the ring buffer is full; `parallel_threads` worker threads have as their top priority to write compressed blocks to the SST file but also take up compression work in priority order of next-to-write. Total of `parallel_threads + 1` threads participating. In some cases, this could result in less throughput than before, but arguably the previous implementation was using more threads than explicitly allowed.
## Future/alternate considerations
Although we could likely have used some framework for micro-work sharing across threads, that could be difficult with the asymmetry of work loads and thread affinity. Specifically, (a) it would be quite challenging to allow emit work in other threads, because it happens in the caller of BlockBasedTableBuilder, (b) async programming is unlikely to pay off until we have an async interface for writing SST files, and (c) this implementation will nevertheless serve as a benchmark for what we lose or gain in such a framework vs. a hand-tuned system.
This implementation still creates and destroys threads for each SST file created. We hope in the future to have more governance and/or pooling of worker threads across various flushes and compactions, but that is not available currently and would require significant design and implementation work.
## More details
* This implementation makes use of semaphores for idling and re-waking threads. `std::counting_semaphore` and `binary_semaphore` offer the best performance (see benchmark results below) but some implementations are known to have correctness bugs. Also, my attempt at upgrading CI for C++20 support (required for these) in https://github.com/facebook/rocksdb/pull/13904 is actually incomplete. Therefore, using these structures is opt-in with `-DROCKSDB_USE_STD_SEMAPHORES` at compile time, and a naive semaphore implementation based on mutex and condvar is used by default. A folly alternative (folly::fibers::Semaphore) was dropped in during development and found to be less efficient than the naive implementation. One CI job is upgraded to test with the new opt-in.
* One of the biggest concerns about correctness/reliability for this implementation is the possibility of hitting a deadlock, in part because that is not well checked in the DB crash test (a challenging problem!). Note also that with the parallel compression improvements in this release, I am calling the feature production-ready, so there is an extra level of confidence needed in the reliability of the feature. Thus, for DEBUG builds including crash test, I have added a watchdog thread to each parallel SST construction that heuristically checks for the most likely kinds of deadlock that could happen, including for the case of buggy semaphore implementations. It periodically verifies that some thread is outside of its "idle" state, and if the watchdog wakes up repeatedly to see all live threads stuck in their idle state (even if wake-up was attempted) then it declares a deadlock. This feature was manually verified for several seeded deadlock bugs. (More details in code comments.)
* For CPU efficiency, this implementation greatly simplifies the logic to estimate the outstanding or "inflight" size not yet written to the SST file. I expect this size to generally be insignificant relative to the full SST file size so is not worth careful engineering. And based on Meta's current needs, landing under-size for an SST file is better than over-size. See comments on `estimated_inflight_size` for details.
* Some other existing atomics in block_based_table_builder.cc modified to use safe atomic wrappers.
* Status handling in BlockBasedTableBuilder was streamlined to get rid of essentially redundant `status`+`io_status` fields and associated code. Made small optimizations to reduce unnecessary IOStatus copies (with StatusOk()) and mark status conditional branches as LIKELY or UNLIKELY.
* Prefer inline field initialization to initialization in constructor.
* Minimize references to the `parallel_threads` configuration parameter for better separation of concerns / sanitization / etc. For example, use non-nullity of `pc_rep` to indicate that parallel compression is enabled (and active).
* Some other refactoring to aid the new implementation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13910
Test Plan:
## Correctness
Already integrated into unit tests and crash test. CI updated for opt-in semaphore implementation. Basic semaphore unit tests added/updated.
As for the tremendous simplification of logic relating to hitting target SST file size, as expected, the new behavior could under-shoot the single-threaded behavior by a small number of blocks, which will typically affect the file size by ~1/1000th or less. I think that's a good trade-off for cutting out unnecessarily complex code with non-trivial CPU cost (FileSizeEstimator).
```
./db_bench -db=/dev/shm/dbbench_filesize_after8 -benchmarks=fillseq,compact -num=10000000 -compression_type=zstd -compression_level=8 -compression_parallel_threads=8
```
Before, PT=8 & PT=1, and After PT=1 the same or very similar
```
-rw-r--r-- 1 peterd users 67474097 Sep 12 15:32 000052.sst
-rw-r--r-- 1 peterd users 67474214 Sep 12 15:32 000053.sst
-rw-r--r-- 1 peterd users 67473834 Sep 12 15:32 000054.sst
-rw-r--r-- 1 peterd users 67473437 Sep 12 15:32 000055.sst
-rw-r--r-- 1 peterd users 67473835 Sep 12 15:32 000056.sst
-rw-r--r-- 1 peterd users 67473204 Sep 12 15:33 000057.sst
-rw-r--r-- 1 peterd users 67473294 Sep 12 15:33 000058.sst
-rw-r--r-- 1 peterd users 67473839 Sep 12 15:33 000059.sst
```
After, PT=8 (worst case here ~0.05% smaller)
```
-rw-r--r-- 1 peterd users 67463189 Sep 12 14:55 000052.sst
-rw-r--r-- 1 peterd users 67465233 Sep 12 14:55 000053.sst
-rw-r--r-- 1 peterd users 67466822 Sep 12 14:55 000054.sst
-rw-r--r-- 1 peterd users 67466221 Sep 12 14:55 000055.sst
-rw-r--r-- 1 peterd users 67441675 Sep 12 14:55 000056.sst
-rw-r--r-- 1 peterd users 67467855 Sep 12 14:55 000057.sst
-rw-r--r-- 1 peterd users 67455132 Sep 12 14:55 000058.sst
-rw-r--r-- 1 peterd users 67458334 Sep 12 14:55 000059.sst
```
## Performance, modest load
We are primarily interested in balancing throughput in building SST files and CPU usage in doing so. (For example, we could maximize throughput by having worker threads only spin waiting for work, but that would likely be extra CPU usage we want to avoid to allow other productive CPU work to be scheduled.) No read path code has been touched.
A benchmark script running "before" and "after" configurations at the same time to minimize random machine load effects:
```
$ SUFFIX=`tty | sed 's|/|_|g'`; for CT in none lz4 zstd; do for PT in 1 2 3 4 6 8; do echo -n "$CT pt=$PT -> "; (for I in `seq 1 10`; do BIN=/tmp/dbbench${SUFFIX}.bin; rm -f $BIN; cp db_bench $BIN; /usr/bin/time $BIN -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 -compression_type=$CT -compression_parallel_threads=$PT 2>&1; done) | awk '/micros.op/ {n++; sum += $5;} /system / { cpu += $1 + $2; } END { print "ops/s: " int(sum/n) " cpu*s: " cpu; }'; done; done
```
Before this change:
```
none pt=1 -> ops/s: 1999603 cpu*s: 72.08
none pt=2 -> ops/s: 1871094 cpu*s: 148.3
none pt=3 -> ops/s: 1882907 cpu*s: 147.7
lz4 pt=1 -> ops/s: 1987858 cpu*s: 94.74
lz4 pt=2 -> ops/s: 1590192 cpu*s: 182.65
lz4 pt=3 -> ops/s: 1896294 cpu*s: 174.7
lz4 pt=4 -> ops/s: 1949174 cpu*s: 172.26
lz4 pt=6 -> ops/s: 1912517 cpu*s: 175.91
lz4 pt=8 -> ops/s: 1930585 cpu*s: 176.71
zstd pt=1 -> ops/s: 1239379 cpu*s: 129.85
zstd pt=2 -> ops/s: 1171742 cpu*s: 226.12
zstd pt=3 -> ops/s: 1832574 cpu*s: 214.21
zstd pt=4 -> ops/s: 1887124 cpu*s: 212.51
zstd pt=6 -> ops/s: 1920936 cpu*s: 211.7
zstd pt=8 -> ops/s: 1885544 cpu*s: 214.87
```
After this change:
```
none pt=1 -> ops/s: 1964361 cpu*s: 72.66
none pt=2 -> ops/s: 1914033 cpu*s: 104.95
none pt=3 -> ops/s: 1978567 cpu*s: 100.24
lz4 pt=1 -> ops/s: 2041703 cpu*s: 92.88
lz4 pt=2 -> ops/s: 1903210 cpu*s: 121.64
lz4 pt=3 -> ops/s: 1973906 cpu*s: 122.22
lz4 pt=4 -> ops/s: 1952605 cpu*s: 123.05
lz4 pt=6 -> ops/s: 1957524 cpu*s: 124.31
lz4 pt=8 -> ops/s: 1986274 cpu*s: 129.06
zstd pt=1 -> ops/s: 1233748 cpu*s: 130.43
zstd pt=2 -> ops/s: 1675226 cpu*s: 158.41
zstd pt=3 -> ops/s: 1929878 cpu*s: 159.77
zstd pt=4 -> ops/s: 1916403 cpu*s: 160.99
zstd pt=6 -> ops/s: 1942526 cpu*s: 166.21
zstd pt=8 -> ops/s: 1966704 cpu*s: 171.56
```
For parallel_threads=1, results are very similar, as expected.
For parallel_threads>1, throughput is usually improved a bit, but cpu consumption is dramatically reduced. For zstd, maximum throughput is essentially achieved with pt=3 rather than the previous roughly pt=4 to 6. And the old used about 30% more CPU.
We can also compare with more expensive compression by raising the compression level.
```
SUFFIX=`tty | sed 's|/|_|g'`; CT=zstd; for CL in 4 6 8; do for PT in 1 4 8; do echo -n "$CT@$CL pt=$PT -> "; (for I in `seq 1 10`; do BIN=/tmp/dbbench${SUFFIX}.bin; rm -f $BIN; cp db_bench $BIN; /usr/bin/time $BIN -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 -compression_type=$CT -compression_parallel_threads=$PT -compression_level=$CL 2>&1; done) | awk '/micros.op/ {n++; sum += $5;} /system / { cpu += $1 + $2; } END { print "ops/s: " int(sum/n) " cpu*s: " cpu; }'; done; done
```
Before:
```
zstd@4 pt=1 -> ops/s: 883630 cpu*s: 161.12
zstd@4 pt=4 -> ops/s: 1878206 cpu*s: 243.25
zstd@4 pt=8 -> ops/s: 1885002 cpu*s: 245.89
zstd@6 pt=1 -> ops/s: 710767 cpu*s: 189.44
zstd@6 pt=4 -> ops/s: 1706377 cpu*s: 277.29
zstd@6 pt=8 -> ops/s: 1866736 cpu*s: 275.07
zstd@8 pt=1 -> ops/s: 529047 cpu*s: 237.87
zstd@8 pt=4 -> ops/s: 1401379 cpu*s: 330.61
zstd@8 pt=8 -> ops/s: 1895601 cpu*s: 321.59
```
After:
```
zstd@4 pt=1 -> ops/s: 889905 cpu*s: 161.03
zstd@4 pt=4 -> ops/s: 1942240 cpu*s: 193.18
zstd@4 pt=8 -> ops/s: 1922367 cpu*s: 205.21
zstd@6 pt=1 -> ops/s: 713870 cpu*s: 188.91
zstd@6 pt=4 -> ops/s: 1832314 cpu*s: 219.66
zstd@6 pt=8 -> ops/s: 1949631 cpu*s: 229.34
zstd@8 pt=1 -> ops/s: 530324 cpu*s: 238.02
zstd@8 pt=4 -> ops/s: 1479767 cpu*s: 271.65
zstd@8 pt=8 -> ops/s: 1949631 cpu*s: 275.6
```
And we can also look at the cumulative effect of this change and https://github.com/facebook/rocksdb/pull/13850 that will combine for the parallel compression improvements in the upcoming 10.7 release:
Before both:
```
lz4 pt=1 -> ops/s: 1954445 cpu*s: 95.14
lz4 pt=3 -> ops/s: 1687043 cpu*s: 186.62
lz4 pt=5 -> ops/s: 1708196 cpu*s: 188.33
zstd pt=1 -> ops/s: 1220649 cpu*s: 131.2
zstd pt=3 -> ops/s: 1658100 cpu*s: 227.08
zstd pt=5 -> ops/s: 1685074 cpu*s: 226.08
```
After:
```
lz4 pt=1 -> ops/s: 2048214 cpu*s: 93.24
lz4 pt=3 -> ops/s: 1922049 cpu*s: 122.9
lz4 pt=5 -> ops/s: 1980165 cpu*s: 122.49
zstd pt=1 -> ops/s: 1245165 cpu*s: 128.84
zstd pt=3 -> ops/s: 1956961 cpu*s: 158.73
zstd pt=5 -> ops/s: 1970458 cpu*s: 161.02
```
In summary, before with zstd default level, you could see only
* about 38% increase in throughput for about 73% increase in CPU usage
Now you can get
* about 58% increase in throughput for about 25% increase in CPU usage
## Performance, high load
To validate this for usage on remote compaction workers, we also need to test whether it falls over at high load or anything concerning like that. For this I did a lot of testing with concurrent db_bench and zstd compression_level=8 and parallel_thread (PT) in {1,8} trying to observe "bad" behaviors such as stalls due to preempted threads and such. On a 166 core machine where a "job" is a db_bench process running a fillseq benchmark similar to above in parallel with others, I could summarize the results like this:
10 jobs PT=8 vs. PT=1 -> 12% more CPU usage, 75% reduction in wall time, 1.9 jobs/sec (vs. 0.5)
50 jobs PT=8 vs. PT=1 -> 89% more CPU usage, 27% reduction in wall time, 3.1 jobs/sec (vs. 2.3)
100 jobs PT=8 vs. PT=1 -> 24% more CPU usage, 5% reduction in wall time, 3.25 jobs/sec (vs. 3.1)
150 jobs PT=8 vs. PT=1 -> 4% more CPU usage, 2% increase in wall time, 3.3 jobs/sec (vs. 3.4)
500 jobs PT=8 vs. PT=1 -> 1% more CPU usage, insignificant difference in wall time, 3.3 jobs/sec
Even when there are 4000 threads potentially competing for 166 cores, the throughput (3.3 jobs / sec) is still very close to maximum (3.4). Enabling parallel compression didn't result in notably less throughput (based on wall clock time for all jobs to complete) in any case tested above, and much higher throughput for many cases. If parallel compression causes us to tip from comfortably under-saturating to over-saturating the cores (as in the 50 jobs case), the overall CPU usage can be much higher, presumably due to lower CPU cache hit rates and maybe clock throttling, but parallel compression still has the throughput advantage in those cases.
In other words, what would we stand to gain from being able to intelligently share worker threads between compaction jobs? It doesn't seem that much.
Reviewed By: xingbowang
Differential Revision: D81365623
Pulled By: pdillinger
fbshipit-source-id: 5db5151a959b5d25b84dbe185bc208bd188f2d1c
Summary:
we saw some crash test failure at https://github.com/facebook/rocksdb/blob/f46242cef631351a5c8f4a7b0fb0935ec7fa61c8/table/block_based/block_based_table_iterator.cc#L964-L965. This is likely due to timestamp not being considered properly in some places in MultiScan code paths. This PR fixes the issue.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13938
Test Plan: crash test with timestamp and multiscan: `python3 -u ./tools/db_crashtest.py whitebox --enable_ts --iterpercent=60 --prefix_size=-1 --prefixpercent=0 --readpercent=0 --test_batches_snapshots=0 --use_multiscan=1 --read_fault_one_in=0 --kill_random_test=88888 --interval=60`
Reviewed By: anand1976
Differential Revision: D82175263
Pulled By: cbi42
fbshipit-source-id: 5d40ede1aec15f8faeaa7fd041b939e68611ff73
Summary:
This PR enables Stress Test to fall back to local compaction when a remote compaction fails, allowing the compaction to be retried on the main thread.
If the local compaction succeeds, the stress test will continue without failing. The main thread will log that the remote compaction failed and was retried locally, while detailed failure logs from the remote compaction attempt will still be printed by the worker thread for further investigation.
This approach allows us to keep collecting useful logs for diagnosing remote compaction failures in Stress Test, while ensuring the test continues to run with remote compaction enabled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13945
Test Plan:
```
python3 -u tools/db_crashtest.py --cleanup_cmd='' --simple blackbox --remote_compaction_worker_threads=8 --interval=10
```
# Internal Only
https://www.internalfb.com/sandcastle/workflow/1315051091202224133https://www.internalfb.com/sandcastle/workflow/3382203320165521367https://www.internalfb.com/sandcastle/workflow/2616591383512372892https://www.internalfb.com/sandcastle/workflow/4607182418810099066
Reviewed By: hx235
Differential Revision: D82279337
Pulled By: jaykorean
fbshipit-source-id: 6f663ec2eeb642fd4ad885a90efb344432a32f89
Summary:
We should add error logging to be able to pinpoint why RocksDB is returning status `NotSupported` for `ReadAsync`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13936
Test Plan: Look at logs (and client logs of error status)
Reviewed By: anand1976
Differential Revision: D82141529
Pulled By: archang19
fbshipit-source-id: c71b70967457be35ef5168321d449f96b2b9441d
Summary:
Fix uninitialized value complaint in valgrind due to gtest print padded struct.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13934
Test Plan: CI. Verified that valgrind no longer complains about it.
Reviewed By: pdillinger
Differential Revision: D82124983
Pulled By: xingbowang
fbshipit-source-id: 99eb7bab99726c45affe0a231777e5951844d73b
Summary:
... and associated statistics, etc. Someone needs it, so here it is.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13927
Test Plan: Updated / extended / added some unit tests
Reviewed By: cbi42
Differential Revision: D81981469
Pulled By: pdillinger
fbshipit-source-id: 52558c08741890b781310906acbc18d9eb479363
Summary:
There are some internal use cases that do not map cleanly onto the existing `IOActivity` enums. This PR creates new custom IOActivity types that internal users can use as they see fit.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13924
Test Plan: Wrote a simple unit test
Reviewed By: pdillinger
Differential Revision: D82029992
Pulled By: archang19
fbshipit-source-id: a3e23c360baa96cd2e9adf570e71c6e43947bfc8
Summary:
PointLockManager manages point lock per key. The old implementation partition the per key lock into 16 stripes. Each stripe handles the point lock for a subset of keys. Each stripe have only one conditional variable. This conditional variable is used by all the transactions that are waiting for its turn to acquire a lock of a key that belongs to this stripe.
In production, we notified that when there are multiple transactions trying to write to the same key, all of them will wait on the same conditional variables. When the previous lock holder released the key, all of the transactions are woken up, but only one of them could proceed, and the rest goes back to sleep. This wasted a lot of CPU cycles. In addition, when there are other keys being locked/unlocked on the same lock stripe, the problem becomes even worse.
In order to solve this issue, we implemented a new PerKeyPointLockManager that keeps a transaction waiter queue at per key level. When a transaction could not acquire a lock immediately, it joins the waiter queue of the key and waits on a dedicated conditional variable. When previous lock holder released the lock, it wakes up the next set of transactions that are eligible to acquire the lock from the waiting queue. The queue respect FIFO order, except it prioritizes lock upgrade/downgrade operation.
However, this waiter queue change increases the deadlock detection cost, because the transaction waiting in the queue also needs to be considered during deadlock detection. To resolve this issue, a new deadlock_timeout_us (microseconds) configuration is introduced in transaction option. Essentially, when a transaction is waiting on a lock, it will join the wait queue and wait for the duration configured by deadlock_timeout_us without perform deadlock detection. If the transaction didn't get the lock after the deadlock_timeout_us timeout is reached, it will then perform deadlock detection and wait until lock_timeout is reached. This optimization takes the heuristic where majority of the transaction would be able to get the lock without perform deadlock detection.
The deadlock_timeout_us configuration needs to be tuned for different workload, if the likelihood of deadlock is very low, the deadlock_timeout_us could be configured close to a big higher than the average transaction execution time, so that majority of the transaction would be able to acquire the lock without performing deadlock detection. If the likelihood of deadlock is high, deadlock_timeout_us could be configured with lower value, so that deadlock would get detected faster.
The new PerKeyPointLockManager is disabled by default. It can be enabled by TransactionDBOptions.use_per_key_point_lock_mgr. The deadlock_timeout_us is only effective when PerKeyPointLockManager is used. When deadlock_timeout_us is set to 0, transaction will perform deadlock detection immediately before wait.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13731
Test Plan:
Unit test.
Stress unit test that validates deadlock detection and exclusive, shared lock guarantee.
A new point_lock_bench binary is created to help perform performance test.
Reviewed By: pdillinger
Differential Revision: D77353607
Pulled By: xingbowang
fbshipit-source-id: 21cf93354f9a367a78c8666596ed14013ac7240b
Summary:
A follow-up to https://github.com/facebook/rocksdb/issues/13904 which was incomplete in updating CI jobs to support C++20 because the C++20 usage was only in tests. Here we add subtle C++20 usage in the public API ("using enum" feature in db.h) to force the issue.
A lot of the work for this PR was in updating the Ubuntu22 docker image, for earlier compiler/runtime versions supporting C++20, and generating a new Ubuntu24 docker image, for later compiler/runtime versions. The Ubuntu22 image needed to be updated because there are incompatibilities with clang-13 + c++20 + libstdc++ for gcc 11, seen on these examples
```
#include <chrono>
int main(int argc, char *argv[]) {
std::chrono::microseconds d = {}; return 0;
}
```
and
```
#include <coroutine>
int main() { return 0; }
```
The second was causing recurring failures in build-linux-clang-13-asan-ubsan-with-folly, now fixed.
So we have to install clang's libc++ to compile with clang-13. I haven't been able to get this to work with some of the libraries like benchmark, glog, and/or gflags, but I'm able to compile core RocksDB with clang-13. On this docker image, an extra compiler parameter is needed to compile with gcc and glog because it's built from source perhaps not perfectly, because the ubuntu package transitively conflicts with libc++.
The Ubuntu24 image seems to be low-drama and generally work for testing out newer compiler versions. The mingw build uses Ubuntu24 because the mingw package on Ubuntu22 uses a gcc version that is too old.
And the mass of other code changes are trying to work around new warnings, mostly from clang-analyze, which I upgraded to clang-18 in CI.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13915
Test Plan: CI, including temporarily including the nightly jobs in the PR jobs in earlier revisions to test and stabilize
Reviewed By: archang19
Differential Revision: D81933067
Pulled By: pdillinger
fbshipit-source-id: 7e33823006a79d5f3cf5bc1d625f0a3c08a7d74c
Summary:
After running stress test over a week, we've identified more failures to fix. While we work on the fix, disable the remote compaction temporarily to reduce noise and avoid these failures hiding other failures.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13925
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D81934248
Pulled By: jaykorean
fbshipit-source-id: 9ac11926429eebe1aebf7b520a548dc5987b7d76
Summary:
This diff adds logging in various places in the external file ingestion code where we check for non-OK status codes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13905
Test Plan: Debugging external file ingestion should be easier with additional logging.
Differential Revision: D81814033
Pulled By: archang19
fbshipit-source-id: 77f8b342cbad892acedc4603c02865c38886f2f4
Summary:
If user_defined_index_factory in BlockBasedTableOptions is configured and we try to open an SST file without the corresponding UDI (either during DB open or file ingestion), ignore a failure to load the UDI by default. If fail_if_no_udi_on_open in BlockBasedTableOptions is true, then treat it as a fatal error.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13921
Test Plan: Update unit tests
Reviewed By: xingbowang
Differential Revision: D81826054
Pulled By: anand1976
fbshipit-source-id: f4fe0b13ccb02b9448622af487680131e349c52b
Summary:
Add a new option `MultiScanArgs::max_prefetch_size` that limits the memory usage of per file pinning of prefetched blocks. Note that this only accounts for compressed block size. This is intended to be a stopgap until we implement some kind of global prefetch manager that limits the global multiscan memory usage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13920
Test Plan: new unit test `./block_based_table_reader_test --gtest_filter="*MultiScanPrefetchSizeLimit/*"`
Reviewed By: xingbowang
Differential Revision: D81630629
Pulled By: cbi42
fbshipit-source-id: 9f66678915242fe1220620531a4b9fd22747cdea
Summary:
# Summary
Until we get WAL + Remote Compaction in Stress Test working, temporarily disable this
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13919
Test Plan: Meta Internal CI run
Reviewed By: anand1976
Differential Revision: D81605621
Pulled By: jaykorean
fbshipit-source-id: 6e1f9a0a7a0f27e7465512689b51364b63ef3e2b
Summary:
Re-enabling Remote Compaction Stress Test with some changes to stress test feature combo sanitization changes
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13913
Test Plan:
Ran Meta Internal Tests for a few days
# Follow up
- Skip recovering from WAL in remote worker and re-enable WAL
- Investigate and fix races with Integrated BlobDB
Reviewed By: hx235
Differential Revision: D81509225
Pulled By: jaykorean
fbshipit-source-id: 949762c48ece0a25e3d0281e3510f1e7d3fe3667
Summary:
**Context/Summary:**
A small change as titled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13891
Test Plan: - Existing UT and rehearsal stress test
Reviewed By: jaykorean
Differential Revision: D80588011
Pulled By: hx235
fbshipit-source-id: 6987e08a4855782305ad742eef6c0196da0d67ca
Summary:
I am wanting to use std::counting_semaphore for something and the timing seems good to require C++20 support. The internets suggest:
* GCC >= 10 is adequate, >= 11 preferred
* Clang >= 10 is needed
* Visual Studio >= 2019 is adquate
And popular linux distributions look like this:
* CentOS Stream 9 -> GCC 11.2 (CentOS 8 is EOL)
* Ubuntu 22.04 LTS -> GCC 11.x (Ubuntu 20 just ended standard support)
* Debian 12 (oldstable) -> GCC 12.2
* (Debian 11 has ended security updates, uses GCC 10.2)
This required generating a new docker image based on Ubuntu 22 for CI using gcc. The existing Ubuntu 20 image works for covering appropriate clang versions (though we should maybe add a much later version as well, in the next increment of our Ubuntu 22 image; however the minimum available clang build from apt.llvm.org for Ubuntu 22 is clang 13).
Update to SetDumpFilter is to quiet a mysterious gcc-13 warning-as-error.
Removed --compile-no-warning-as-error from a cmake command line because cmake in the new docker image is too old for this option.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13904
Test Plan: CI, one minor unit test added to verify std::counting_semaphor works
Reviewed By: xingbowang
Differential Revision: D81266435
Pulled By: pdillinger
fbshipit-source-id: 26040eeccca7004416e29a6ff4f6ea93f2052684
Summary:
**Context/Summary:**
`ProcessKeyValueCompaction()` has grown too long to resonate or add any logic to resume from some key and save progress for resumable compaction. This PR breaks this function into smaller functions. Almost all of them are cosmetic changes, except for one thing pointed out in below PR conversation.
Specially, this PR did the following:
- Added `SubcompactionInternalIterators`, `SubcompactionKeyBoundaries` and `BlobFileResources` to manage the lifetime of the local variables of the original functions to be used across smaller functions
- Moved AutoThreadOperationStageUpdater, some IO stats measurement to a different place that makes more sense
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13879
Test Plan: Existing UT
Reviewed By: jaykorean
Differential Revision: D80216092
Pulled By: hx235
fbshipit-source-id: 515615906e5e5fd5ec191bcdd4126f17d282cac2
Summary:
The implementation of parallel compression has historically scaled rather poorly, or perhaps modestly with heavy compression, topping out around 3x throughput vs. serial and incurring big overheads in CPU consumption relative to the throughput.
This change addresses one source of that extra CPU consumption: stashing all the keys of a block for later processing into building index and filter blocks. Historically with parallel compression, the index and filter block updates were handled in the last stage of processing along with writing each data block to the file writer. This was because the index blocks needed to know the BlockHandle of the new data block, which could only be known after every preceeding data block was compressed, to know the starting location for the BlockHandle. And because index and filter partitions were historically coupled (see decouple_partitioned_filters), filter updates had to happen at the same time.
Here we get rid of stashing the keys for later processing and the extra CPU associated with it, by
* Creating a two stage process of adding to index blocks ("prepare" and "finish" each entry; one entry per data block). The two stages must be executable in parallel for separate index entries. NOTE: not yet supported by UserDefinedIndex
* Requiring decouple_partitioned_filters=true for parallel compression, because we now add to filters in the first stage of processing when each key is readily available and we cannot couple that with finalizing index entries in the last stage of processing.
It might seem like adding to filters is something that is expensive (hashing etc.) and should be kept out of the bottle-neck first stage of processing (which includes walking the compaction iterator) but it's probably similar cost to simply stashing the keys away for later processing. (We might be able to reduce a bottle-neck by stashing hashes, but we're not to a point where that is worth the effort.)
And it makes sense to make two more simple public API updates in conjunction with this:
* Set decouple_partitioned_filters=true by default. No signs of problems in production.
* Mark parallel compression as production-ready. It's being thoroughly tested in the crash test, successfully, and in limited production uses.
Follow-up:
* Improve the threading/sychronization model of parallel compression for the next major efficiency improvement
* Consider supporting the parallel-compatible index building APIs with UserDefinedIndex, unless it's considered too dangerous to expect users to safely handle the multi-threading.
* (In a subsequent release) remove all the code associated with coupling filter and index partitions and mark the option as ignored.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13850
Test Plan:
for correctness, existing tests
## Performance Data
The "before" data here includes revert of https://github.com/facebook/rocksdb/issues/13828 for combined performance measurement of this change and that one.
```
SUFFIX=`tty | sed 's|/|_|g'`; for CT in lz4 zstd lz4; do for PT in 1 2 3 4 6 8; do echo "$CT pt=$PT"; (for I in `seq 1 1`; do BIN=/dev/shm/dbbench${SUFFIX}.bin; rm -f $BIN; cp db_bench $BIN; /usr/bin/time $BIN -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=30000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 -compression_type=$CT -compression_parallel_threads=$PT 2>&1 | tail -n 3 | head -n 2; done); done; done
```
To get a sense of the overall performance relative to number of parallel threads, we vary that with popular fast compression and popular heavier weight compression (some noise in this data, don't interpret each data point too strongly)
lz4 pt=1
2107431 -> 2112941 ops/sec (+0.3% - improvement)
(26.51 + 0.75) = 27.26 CPU sec -> (26.63 + 0.79) = 27.42 CPU sec (+0.6% - regression)
lz4 pt=2
1606660 -> 1580333 ops/sec (-1.6% - regression)
(47.10 + 8.37) = 55.47 CPU sec -> (45.05 + 9.23) = 54.28 CPU sec (-2.2% - improvement)
lz4 pt=3
1701353 -> 1889283 ops/sec (+11.1% - improvement)
(47.23 + 8.29) = 55.52 CPU sec -> (43.89 + 8.33) = 52.22 CPU sec (-6.0% - improvement)
lz4 pt=4
1651504 -> 1817890 ops/sec (+10.1% - improvement)
(48.07 + 8.31) = 56.38 CPU sec -> (44.77 + 8.45) = 53.22 CPU sec (-5.6% - improvement)
lz4 pt=6
1716099 -> 1888523 ops/sec (+10.1% - improvement)
(47.50 + 8.45) = 55.95 CPU sec -> (44.25 + 8.73) = 52.98 CPU sec (-5.3% - improvement)
lz4 pt=8
1696840 -> 1797256 ops/sec (+5.9% - improvement)
(48.09 + 8.61) = 56.70 CPU sec -> (45.90 + 8.68) = 54.58 CPU sec (-3.8% - improvement)
Clearly parallel threads do not help with fast compression like LZ4, but it's not as bad as it was before.
zstd pt=1
1214258 -> 1202863 ops/sec (-0.9% - regression)
(38.26 + 0.66) = 38.92 CPU sec -> (39.37 + 0.69) = 40.06 CPU sec (+2.9% - regression)
zstd pt=2
1194673 -> 1152746 ops/sec (-3.5% - regression)
(61.01 + 9.85) = 70.86 CPU sec -> (58.28 + 9.99) = 68.27 CPU sec (-3.7% - improvement)
zstd pt=3
1653661 -> 1825618 ops/sec (+10.4% - improvement)
(60.07 + 8.45) = 68.52 CPU sec -> (56.03 + 8.43) = 64.46 CPU sec (-5.9% - improvement)
zstd pt=4
1691723 -> 1890976 ops/sec (+11.8% - improvement)
(59.72 + 8.46) = 68.18 CPU sec -> (55.96 + 8.27) = 64.23 CPU sec (-5.7% - improvement)
zstd pt=6
1684982 -> 1900002 ops/sec (+12.8% - improvement)
(58.89 + 8.26) = 67.15 CPU sec -> (55.98 + 8.48) = 64.46 CPU sec (-4.0% - improvement)
zstd pt=8
1648282 -> 1892531 ops/sec (+14.8% - improvement)
(59.43 + 8.63) = 68.06 CPU sec -> (56.49 + 8.32) = 64.81 CPU sec (-4.8% - improvement)
The throughput is now able to increase by *more than half* with lots of parallelism, rather than only *about a third*.
Scalability is a bit better with higher compression level, and we still see a benefit from this change. (We've also enabled partitioned indexes and filters here, which sees essentially the same benefits):
zstd pt=1 compression_level=7
595720 -> 597359 ops/sec (+0.3% - improvement)
(63.45 + 0.73) = 64.18 CPU sec -> (63.25 + 0.71) = 63.96 CPU sec (-0.3% - improvement)
zstd pt=4 compression_level=7
1527116 -> 1501779 ops/sec (-1.7% - regression)
(85.00 + 8.14) = 93.14 CPU sec -> (81.85 + 9.02) = 90.87 CPU sec (-2.5% - improvement)
zstd pt=6 compression_level=7
1678239 -> 1956070 ops/sec (+16.5% - improvement)
(83.77 + 8.11) = 91.88 CPU sec -> (79.87 + 7.78) = 87.65 CPU sec (-4.6% - improvement)
zstd pt=8 compression_level=7
1696132 -> 1953041 ops/sec (+15.1% - improvement)
(83.97 + 8.14) = 92.11 CPU sec -> (80.61 + 7.78) = 88.39 CPU sec (-4.1% - improvement)
With more tests, not really seeing any consistent differences with no parallelism (despite some micro-optimizations thrown in)
Reviewed By: hx235
Differential Revision: D79853111
Pulled By: pdillinger
fbshipit-source-id: 7a34fd7811217fb74fa6d3efaea7ffcce72beec7
Summary:
**Context/Summary:**
RocksDB stress test verifies IOActivity is set correctly through reusing the pass-in Read/Write options through assertion. This is too strict for API that does not take or do not need to take Read/WriteOptions yet hence assertion failure.
```
stderr:
db_stress: ... db_stress_tool/db_stress_env_wrapper.h:24: void rocksdb::(anonymous namespace)::CheckIOActivity(const IOOptions &): Assertion `io_activity == Env::IOActivity::kUnknown || io_activity == options.io_activity' failed.
Received signal 6 (Aborted)
```
An example is ManagedSnapshot snapshot_guard(db_); in TestMultiScan().
This PR ignores such check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13898
Test Plan: The same command repro-ed this assertion failure passes after this fix
Reviewed By: archang19
Differential Revision: D80983214
Pulled By: hx235
fbshipit-source-id: d8b660f8c8771198bc7fa0e805c3e86d2584f03e
Summary:
**Context/Summary:**
Clear statistics reference from options_ to intentionally shorten the statistics object lifetime to be same as the db object (which is the common case in practice) and detect if RocksDB access the statistics beyond its lifetime.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13899
Test Plan: - [Ongoing] Stress test rehearsal
Reviewed By: pdillinger
Differential Revision: D80985435
Pulled By: hx235
fbshipit-source-id: ab238231cd81f47fa451aea12a0c85fa11d9ac81
Summary:
`IngestExternalFileOptions::allow_db_generated_files` requires SST files to have zero sequence number. This PR opens it up for any DB generated SST files. Currently we don't do global sequence number assignment when `allow_db_generated_files` is true, so we require that files do not overlap with any key in the CF. One behavior difference is that now we allow ingesting overlapping files when `allow_db_generated_files` is true. Users need to ensure that files are ordered such that later files have more recent updates.
Intended follow ups:
- Record smallest seqno in table property, so that we don't need to scan the file for it.
- Cover allow_db_generated_files in crash test. We may create a new DB and ingest all files from a CF for verification.
- Add APIs that uses allow_db_generated_files. For example, an API for ingesting SST files from a source CF, so that we take care of ingestion file ordering for user. If we are already getting metadata from the source CF, we may be use it as a hint for level placement instead of dividing input files into batches again (`ExternalSstFileIngestionJob::DivideInputFilesIntoBatches`).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13878
Test Plan: two new unit tests.
Reviewed By: hx235, xingbowang
Differential Revision: D80233727
Pulled By: cbi42
fbshipit-source-id: 74209386d8426c434bff2d9a734f06db537eb50c
Summary:
RocksDB currently aborts whenever `io_uring_wait_cqe` returns an error code. It also does not log what error code was returned.
While experimenting with `IO_URING`, my application crashed because of this.
I asked the Linux Kernel user group the best way to handle unsuccessful `io_uring_wait_cqe`.
It was recommended to retry on `EINTR`, `EAGAIN`, and `ETIME`. `ETIME` only happens when waiting with a timeout, so I am not handling it.
I also write to `stderr` so that we have some debugging information if we abort.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13890
Test Plan: Unfortunately this is hard to cover through unit/stress tests. We have to see what sort of errors get encountered in production.
Reviewed By: anand1976
Differential Revision: D80639955
Pulled By: archang19
fbshipit-source-id: e3a230bd37552ec0f36be34e6a4e53cfd2a254f1
Summary:
When fill_cache is ReadOptions is false, multi scan Prepare crashes with the following assertion failure. In this case, CreateAndPibBlockInCache needs to directly create a block with full ownership.
https://github.com/facebook/rocksdb/issues/9 0x00007f2fc003bc93 in __GI___assert_fail (assertion=0x7f2fc2147361 "pinned_data_blocks_guard[block_idx].GetValue()", file=0x7f2fc2146e08 "table/block_based/block_based_table_iterator.cc", line=1178, function=0x7f2fc2147262 "virtual void rocksdb::BlockBasedTableIterator::Prepare(const rocksdb::MultiScanArgs *)") at assert.c:101
101 in assert.c
https://github.com/facebook/rocksdb/issues/10 0x00007f2fc1d73088 in rocksdb::BlockBasedTableIterator::Prepare(rocksdb::MultiScanArgs const*) () from /data/users/anand76/rocksdb_anand76/librocksdb.so.10.6
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13889
Test Plan: Parameterize the DBMultiScanIteratorTest tests with fill_cache
Reviewed By: cbi42
Differential Revision: D80552069
Pulled By: anand1976
fbshipit-source-id: 1a0b64af1e14c63d826add1f994a832ebff12757
Summary:
I ran multiple runs of crash test jobs internally, so far I've seen one iterator mismatch and one assertion failure. I've added relevant logging improvements to help debugging them. use_multiscan will be stable within a crash test run to make it easier to triage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13888
Test Plan: `python3 tools/db_crashtest.py whitebox --prefix_size=-1 --test_batches_snapshots=0 --use_multiscan=1 --read_fault_one_in=0 --kill_random_test=88888`
Reviewed By: anand1976
Differential Revision: D80627399
Pulled By: cbi42
fbshipit-source-id: 2fa3f77e730f5bc7d1d200dc122cf84e3558c588
Summary:
The assert occasionally throws off the stress test runs. We already have sufficient logging in place to collect the signal about secondary cache capacity exceeding primary cache reservation for further investigation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13885
Reviewed By: anand1976
Differential Revision: D80355513
Pulled By: mszeszko-meta
fbshipit-source-id: b36926f0493a3aca19818a1980ef79277db9fe7e
Summary:
Add the --list_meta_blocks option to sst_dump. This PR also refactors some of the test code in sst_dump_test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13838
Reviewed By: cbi42
Differential Revision: D80320812
Pulled By: anand1976
fbshipit-source-id: 921b6560fbd756f5f8b364893700d240d3b7ad00
Summary:
Two instances of change that are not just cosmetic:
* InlineSkipList<>::Node::CASNext() was implicitly using memory_order_seq_cst to access `next_` while it's intended to be accessed with acquire/release. This is probably not a correctness issue for compare_exchange_strong but potentially a previously missed optimization.
* Similar for `max_height_` in Insert which is otherwise accessed with relaxed memory order.
* One non-relaxed access to `is_range_del_table_empty_` in a function only used in assertions. Access to this atomic is otherwise relaxed (and should be - comment added)
Didn't do all of memtable.h because some of them are more complicated changes and I should probably add FetchMin and FetchMax functions to simplify and take advantage of C++27 functions where available (intended follow-up).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13844
Test Plan: existing tests
Reviewed By: xingbowang
Differential Revision: D79742552
Pulled By: pdillinger
fbshipit-source-id: d97ce72ba9af6c105694b7d40622db9e994720cd
Summary:
This is an important feature for avoiding (reducing) unfair block cache treatment for a lot of blocks. It should also unlock some parallel optimizations (https://github.com/facebook/rocksdb/issues/13850) and code simplification.
Consider for follow-up:
* Feature to avoid majorly under0sized data blocks and filter and index partition blocks
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13881
Test Plan: existing tests, been looking good in production
Reviewed By: hx235
Differential Revision: D80288192
Pulled By: pdillinger
fbshipit-source-id: 5e274ffffb044713278d2a286db6bceaab2dadec
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13882
The `expect_valid_internal_key` parameter was always passed as true, with false only used in one unit test. This change removes the parameter and always fail compaction when encountering corrupted internal keys, which is the expected production behavior.
Reviewed By: mszeszko-meta
Differential Revision: D80287672
fbshipit-source-id: e30a282ac30d7fded677504cec11173de8d15167
Summary:
Allow a user defined index to be configured from a string
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13880
Test Plan: Add a unit test in table_test.cc
Reviewed By: bikash-c
Differential Revision: D80237701
Pulled By: anand1976
fbshipit-source-id: 8b3d0bcdfbb4bb76803916ea1b1f940a4d985dfd
Summary:
The original intention of the User Defined Index interface was to use the user key. However, the implementation mixed user and internal key usage. This PR makes it consistent. It also clarifies the UDI contract.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13865
Test Plan: Update tests in table_test.cc
Reviewed By: pdillinger
Differential Revision: D80050344
Pulled By: anand1976
fbshipit-source-id: ace47737d21684ec19709640a09e198cee2d98bd
Summary:
... as we see some issues that rehearsal stress test didn't surface.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13869
Reviewed By: cbi42
Differential Revision: D80103341
Pulled By: hx235
fbshipit-source-id: 8b2c1d76d4c3099727ba3a69de44de67afd64369
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13846
This diff addresses few issues that was identified during testing of the user defined index.
1. During the finishing of the index blocks, we run into an infinite loop because the user defined index wrapper returns
early on incomplete status. This happens because the wrapper blindly returns the status if it not OK. But, the status
could legitimately be `Incomplete()` for some indices like Partitioned Index (serving as the internal index for the UDI
wrapper). Fix is to exclude `Incomplete()` check from the status check early in the UDI wrapper's finish.
2. Once we fixed (1), we noticed that the meta blocks for the UDI-based index writer were not written out to the final
SST file. This is because the UDI's meta blocks are created after the internal index's meta blocks and the block-based
index builder didn't account for this. The fix is to finish the UDI wrapper first which will create the necessary meta blocks
and then finish the internal index. If the internal index is incomplete, the block-based index builder should still continue
to write out the meta blocks.
3. OnKeyAdded when delegating to the user-defined index should only pass the user key. The UDI builder doesn't
understand RocksDB's internal key format and while that poses interesting challenges when the UDI is used for non
last level SST files, our plan is to restrict the usage of the UDI to last level files only (for now).
Reviewed By: pdillinger
Differential Revision: D79781453
fbshipit-source-id: 2239c8fc016da55df5c24be6aacc8f6357cab029
Summary:
fix the following error showing up in continuous tests:
```
Makefile:186: Warning: Compiling in debug mode. Don't use the resulting binary in production
port/mmap.cc:46:15: error: first argument in call to 'memcpy' is a pointer to non-trivially copyable type 'rocksdb::MemMapping' [-Werror,-Wnontrivial-memcall]
46 | std::memcpy(this, &other, sizeof(*this));
| ^
port/mmap.cc:46:15: note: explicitly cast the pointer to silence this warning
46 | std::memcpy(this, &other, sizeof(*this));
| ^
| (void*)
1 error generated.
make: *** [Makefile:2580: port/mmap.o] Error 1
make: *** Waiting for unfinished jobs....
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13864
Test Plan: `make USE_CLANG=1 j=150 check` with https://github.com/facebook/rocksdb/blob/13f054febb26100184eeefaac11877d735d45ac2/build_tools/build_detect_platform#L61-L70 commented out.
Reviewed By: mszeszko-meta
Differential Revision: D80033441
Pulled By: cbi42
fbshipit-source-id: b2330eea71fe28243236b75128ec6f3f1e971873
Summary:
while debugging stress test failure, I noticed that sst_dump and ldb do not work if custom db_stress compression manager is used. This PR adds support for it.
```
./sst_dump --command=raw --show_properties --file=/tmp/rocksdb_crashtest_whitebox4ny5mass/000589.sst
options.env is 0x7f2b1f4b9000
Process /tmp/rocksdb_crashtest_whitebox4ny5mass/000589.sst
Sst file format: block-based
/tmp/rocksdb_crashtest_whitebox4ny5mass/000589.sst: Not implemented: Could not load CompressionManager: DbStressCustom1
/tmp/rocksdb_crashtest_whitebox4ny5mass/000589.sst is not a valid SST file
./ldb idump --db=/tmp/rocksdb_crashtest_whiteboxy_emah11 --ignore_unknown_options --hex >> /tmp/i_dump
Failed: Not implemented: Could not load CompressionManager: DbStressCustom1
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13827
Test Plan: manually tested that ldb and sst_dump work with DbStressCustomCompressionManager after this PR
Reviewed By: pdillinger
Differential Revision: D79461175
Pulled By: cbi42
fbshipit-source-id: c8c092b10b4fde3a295b00751057749e8f0cf095
Summary:
To better support future options, and changes, we need to convert the std::vector<ScanOptions> to something more malleable.
This diff introduces the MultiScanOptions structure and pipes it through the various points in the code in the Prepare path.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13837
Test Plan:
Ensure all associated tests pass
```
make check all
```
Reviewed By: cbi42
Differential Revision: D79655229
Pulled By: krhancoc
fbshipit-source-id: 3a90fb7420e9655021de85ed0158b866f8bfba05
Summary:
**Context/Summary:**
This update, which should have been part of a previous refactoring [PR](https://github.com/facebook/rocksdb/commit/d2ac955881e856fc69d5b15427d742fc635aaead), involves simple renaming for clarity and ensures output table properties are only set when compaction succeeds. Output properties are not meaningful if compaction fails, so this change prevents their population in such cases. Additionally, subsequent statistics updates already do not rely on output file table properties, maintaining correctness regardless of compaction success.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13851
Test Plan: Existing unit tests
Reviewed By: jaykorean
Differential Revision: D79862244
Pulled By: hx235
fbshipit-source-id: 1db16b8dc7b820fab3ec1d5c8a4b757466590e2c
Summary:
**Context/Summary:**
The `CompactionJob::Run()` method has grown too large and complex, making it difficult to implement moderate changes or reason about the code flow (e.g., determining where to save compaction progress for resuming). This PR refactors the method into smaller, more focused functions to improve readability and maintainability.
The refactoring consists mostly of cosmetic changes that extract logical sections into separate methods, with two notable functional improvements:
1. **Relocated output processing logic**: Moved code under `RemoveEmptyOutputs()` and `HasNewBlobFiles()` to where it's actually needed, rather than piggy-backing on the subcompaction state loop. While this introduces 2 additional loops over subcompactions, the performance impact should be negligible given the improved code clarity.
2. **Repositioned statistics updates**: Moved `UpdateCompactionJobInputStats()` and `UpdateCompactionJobOutputStats()` from the record verification section to the end `FinalizeCompactionRun()` methods. This change is safe since record verification is a read-only operation that doesn't modify any statistics.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13849
Test Plan: Existing unit tests
Reviewed By: jaykorean
Differential Revision: D79824429
Pulled By: hx235
fbshipit-source-id: 6b73136f32ecc6842a04a77502b7dbb0bbf507f7
Summary:
We temporarily disabled WAL when Remote Compaction is enabled in Stress Test (https://github.com/facebook/rocksdb/pull/13843). There are few others to incompatible features when WAL is disabled. Due to the sanitization order, WAL was disabled at the end of the sanitization and these incompatible features weren't set properly. Stress Test failed with an error like the following.
e.g. `reopen` stress test is not compatible with `disable_wal` - `Error: Db cannot reopen safely with disable_wal set!`
This PR changes the order of sanitization so that `disable_wal` is set earlier when `remote_compaction_worker_threads > 0`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13845
Test Plan:
```
python3 -u tools/db_crashtest.py blackbox --remote_compaction_worker_threads=8 --interval=5 --duration=6000 --continuous_verification_interval=10 --disable_wal=1 --use_txn=1 --txn_write_policy=2 --enable_pipelined_write=0 --checkpoint_one_in=0 --use_timed_put_one_in=0
```
Reviewed By: cbi42
Differential Revision: D79758670
Pulled By: jaykorean
fbshipit-source-id: aa6f4a74cc86c23f442928c301187b06e8137f53
Summary:
https://github.com/facebook/rocksdb/issues/13676 unfortunately treated some IOErrors as corruption, which is not appropriate when remote storage is involved. To help enforce this, our crash test injects errors that are expected to be propagated back to the user rather than causing some other failure.
Saw crash test failures like this:
```
TestMultiGetEntity (AttributeGroup) error: Corruption: Failed to get file size: Not implemented: GetFileSize Not Supported for file ...
```
So fixing this handling by not injecting a false Corruption failure and allowing smooth fallback from FSRandomAccessFile::GetFileSize to FileSystem::GetFileSize
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13842
Test Plan: unit test added
Reviewed By: xingbowang
Differential Revision: D79728861
Pulled By: pdillinger
fbshipit-source-id: 33f7dfc85d86d88cb4ab24a8defd26618c95c954
Summary:
To reduce the noise, disable the incompatible ones for now when `remote_compaction_worker_threads > 0`. We will investigate each, fix as needed and re-enable them as follow up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13843
Test Plan:
```
python3 -u tools/db_crashtest.py blackbox --remote_compaction_worker_threads=8 --interval=5 --duration=6000 --continuous_verification_interval=10 --disable_wal=1 --use_txn=1 --enable_pipelined_write=0 --checkpoint_one_in=0 --use_timed_put_one_in=0
```
Reviewed By: cbi42
Differential Revision: D79735166
Pulled By: jaykorean
fbshipit-source-id: ae3be38a21073fd3282d6e8cd7d71f0363df3590
Summary:
**Summary:**
This test verifies that compaction respects the min_file_size parameter when triggered by deletions, preventing the compaction of files with deletions smaller than the threshold. The test logic includes two scenarios:
1. Verify that a large L0 file with deletions exceeding the minimum file size threshold triggers deletion-triggered compaction (DTC) and compacts to L1.
2. Verify that a small L0 file with deletions, but below the minimum file size threshold, does not trigger DTC and remains at L0.
Added the DeletionTriggeredCompactionWithMinFileSizeTestListener, which verifies that files selected for compaction based on deletion triggers meet the minimum file size threshold. The listener validates in OnCompactionBegin that all input files have sizes greater than or equal to the configured min_file_size parameter.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13825
Test Plan:
Tested this feature on our devserver using the following commands:
```
DEBUG_LEVEL=2 make -j64 db_compaction_test && KEEP_DB=1 ./db_compaction_test --gtest_filter="*DBCompactionTest.CompactionWith*"
```
Test output confirms the expected behavior:
```
2025/07/31-11:24:49.473181 1431671 [/compaction/compaction_job.cc:2291] [default] [JOB 6] Compacting 2@0 files to L1, score 0.04
2025/07/31-11:24:49.473240 1431671 [/compaction/compaction_job.cc:2297] [default]: Compaction start summary: Base version 6 Base level 0, inputs: [15(52KB) 9(103KB)]
2025/07/31-11:24:49.473304 1431671 EVENT_LOG_v1 {"time_micros": 1753986289473273, "job": 6, "event": "compaction_started", "cf_name": "default", "compaction_reason": "FilesMarkedForCompaction", "files_L0": [15, 9], "score": 0.04, "input_data_size": 159848, "oldest_snapshot_seqno": -1}
```
**Tasks:**
T228156639
Reviewed By: cbi42
Differential Revision: D79395851
Pulled By: nmk70
fbshipit-source-id: 4c2a80a95521b40543981dd81b347f3984cd2a8b
Summary:
Remote Compaction in the stress test previously failed with the following error, so we temporarily disabled it in PR https://github.com/facebook/rocksdb/issues/13815 :
```
reference std::vector<rocksdb::ThreadState *>::operator[](size_type) [_Tp = rocksdb::ThreadState *, _Alloc = std::allocator<rocksdb::ThreadState *>]: Assertion '__n < this->size()' failed.
```
The error was from accessing `remote_compaction_worker_threads[i]` when `i < remote_compaction_worker_threads.size()` which leads to an undefined behavior. This PR fixes the issue by properly setting the worker thread pointers in `remote_compaction_worker_threads`.
Note: We are still encountering errors when both BlobDB and Remote Compaction are enabled. It appears to be a race condition. For now, BlobDB is temporarily disabled if remote compaction is enabled. We will fix the race condition and re-enable BlobDB as a follow-up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13835
Test Plan:
```
python3 -u tools/db_crashtest.py blackbox --remote_compaction_worker_threads=16 --interval=2 --duration=180
```
Reviewed By: hx235
Differential Revision: D79684447
Pulled By: jaykorean
fbshipit-source-id: 65f5809f651865c3df76c2cf3b9e7b8d654bb90a
Summary:
this option has the same functionality as DBOptions::allow_ingest_behind but allows the feature at per CF level. `DBOptions::allow_ingest_behind` is deprecated after this PR and users should use `cf_allow_ingest_behind` instead.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13810
Test Plan: updated some existing tests to use the new option.
Reviewed By: xingbowang
Differential Revision: D79191969
Pulled By: cbi42
fbshipit-source-id: 0da45f6be472ace6754ad15df93d45ac86313837
Summary:
**Context/Summary:**
The `RoundRobinSubcompactionsAgainstResources` test, specifically the `SubcompactionsUsingResources` case, is now disabled. This decision was made because the test's reliability depends on the absence of any concurrent compactions other than the round-robin compaction. Addressing this issue while maintaining the test's focus on resource reservation requires a deeper investigation, which is currently beyond my available bandwidth. Given the increased frequency of test failures, it has been temporarily disabled to prevent further disruptions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13839
Test Plan: - Should be no test failure from RoundRobinSubcompactionsAgainstResources.SubcompactionsUsingResources anymore.
Reviewed By: cbi42
Differential Revision: D79686366
Pulled By: hx235
fbshipit-source-id: 3a226cfd2b67cabc6c585ea567e2b0c25aa5f345
Summary:
#Summary
Quick follow-up from https://github.com/facebook/rocksdb/pull/13816: `CompactFiles()` and `CompactRange()` in CompactionPickers do not run compaction as their names might suggest. What they actually do is create the Compaction object that will be passed to `CompactionJob` to run the compaction.
Renaming these two functions to better represent their purposes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13831
Test Plan: No functional change. Existing CI should be sufficient.
Reviewed By: hx235
Differential Revision: D79660196
Pulled By: jaykorean
fbshipit-source-id: ca831dbef5120e7115b52fd07b0059ca16c8f1e8
Summary:
... by ensuring that files in dropped column family are not returned to the caller upon successful, offline MANIFEST iteration.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13832
Test Plan: `DBTest2, GetFileChecksumsFromCurrentManifest_CRC32`
Reviewed By: pdillinger
Differential Revision: D79607298
Pulled By: mszeszko-meta
fbshipit-source-id: e7948e086ba6e6fb953a3959fdcc81300613d73e
Summary:
Introduce `CompactionJob::VerifyOutputRecordCount()` and make it align with `VerifyInputRecordCount()`.
Functionality-wise, it should be the same except when `db_options_.compaction_verify_record_count` is false. RocksDB will only print WARN message upon verification failure and not return `Status::Corruption()`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13830
Test Plan:
Existing tests cover both
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.VerifyInputRecordCount*"
```
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.CorruptedOutput*"
```
Reviewed By: hx235
Differential Revision: D79584795
Pulled By: jaykorean
fbshipit-source-id: 5851328999005601b28504085b688b80880bca7c
Summary:
In anticipation of an enhancement related to parallel compression
* Rename confusing state variables `seperator_is_key_plus_seq_` -> `must_use_separator_with_seq_`
* Eliminate copy-paste code in `PartitionedIndexBuilder::AddIndexEntry`
* Optimize/simplify `PartitionedIndexBuilder::flush_policy_` by allowing a single policy to be re-targetted to different block builders. Added some additional internal APIs to make this work, and it only works because the FlushBlockBySizePolicy is otherwise stateless (after creation).
* Improve some comments, including another proposed optimization especially for the common case of no live snapshots affecting a large compaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13828
Test Plan:
existing tests are pretty exhaustive, especially with crash test
Planning to validate performance in combination with next change. (This change is saving some extra allocate/deallocate with partitioned index.)
Reviewed By: cbi42
Differential Revision: D79570576
Pulled By: pdillinger
fbshipit-source-id: f7a16f0e6e6ad2023a3d1a2ebaa3cc22aac717af
Summary:
This diff introduces the IntervalSet data structure, which will be used to help create sets of non overlapping sets of intervals for MultiScan scan options. Specifically, we add specializations for Slices to assist in this.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13787
Test Plan: Added test to catch various cases within adding intervals.
Reviewed By: anand1976
Differential Revision: D78624970
Pulled By: krhancoc
fbshipit-source-id: 9a3e4a28738ab8428788467540fc05ab5c1a1b67
Summary:
One of the parameters for constructing a Compaction object is `earliest_snapshot`, which is required for Standalone Range Deletion Optimization (introduced in [https://github.com/facebook/rocksdb/pull/13078](https://github.com/facebook/rocksdb/pull/13078)). Remote Compaction has been using the `CompactionPicker::CompactFiles()` API to create the Compaction object, but this API never sets the `earliest_snapshot` parameter. To address this, update `CompactionPicker::CompactFiles()` to optionally accept `earliest_snapshot` and pass it during the call in `DBImplSecondary::CompactWithoutInstallation()`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13816
Test Plan:
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.StandaloneDeleteRangeTombstoneOptimization*"
```
\+ Tested in Meta's internal offload infra.
Reviewed By: hx235
Differential Revision: D79284769
Pulled By: jaykorean
fbshipit-source-id: 164834ef6972d5e0ddfc2970bb9234ef166d6e52
Summary:
Fix a bug in MultiScan where BlockBasedTableIterator should not return out-of-bound when the all blocks of the last scan are exhausted. This prevented LevelIterator from entering the next file so iterator is returning less keys than expected.
Also fixed stress testing to specify iterate_upper_bound correctly.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13822
Test Plan:
- the following fails quickly before this PR and finishes after this PR
```python3 tools/db_crashtest.py whitebox --iterpercent=60 --prefix_size=-1 --prefixpercent=0 --readpercent=0 --test_batches_snapshots=0 --use_multiscan=1 --seed=1 --fill_cache=1 --read_fault_one_in=0 --column_families=1 --allow_unprepared_value=0 --kill_random_test=88888```
- new unit test that fails before this PR
Reviewed By: krhancoc
Differential Revision: D79308957
Pulled By: cbi42
fbshipit-source-id: c9eafd1c8750b959b0185d7c63199b503493cbd2
Summary:
The main motivation for this change is to more flexibly and efficiently support compressing data without extra copies when we do not want to support saving compressed data that is LARGER than the uncompressed. We believe pretty strongly that for the various workloads served by RocksDB, it is well worth a single byte compression marker so that we have the flexibility to save compressed or uncompressed data when compression is attempted. Why? Compression algorithms can add tens of bytes in fixed overheads and percents of bytes in relative overheads. It is also an advantage for the reader when they can bypass decompression, including at least a buffer copy in most cases, after reading just one byte.
The block-based table format in RocksDB follows this model with a single-byte compression marker, and at least after https://github.com/facebook/rocksdb/pull/13797 so does CompressedSecondaryCache. (Notably, the blob file format DOES NOT. This is left to follow-up work.)
In particular, Compressor::CompressBlock now takes in a fixed size buffer for output rather than a `std::string*`. CompressBlock itself rejects the compression if the output would not fit in the provided buffer. This also works well with `max_compressed_bytes_per_kb` option to reject compression even sooner if its ratio is insufficient (implemented in this change). In the future we might use this functionality to reduce a buffer copy (in many cases) into the WritableFileWriter buffer of the block based table builder.
This is a large change because we needed to (or were compelled to)
* Update all the existing callers of CompressBlock, sometimes with substantial changes. This includes introducing GrowableBuffer to reuse between calls rather than std::string, which (at least in C++17) requires zeroing out data when allocating/growing a buffer.
* Re-implement built-in Compressors (V2; V1 is obsolete) to efficiently implement the new version of the API, no longer wrapping the `OLD_CompressData()` function. The new compressors appropriately leverage the CompressBlock virtual call required for the customization interface and no rely on `switch` on compression type for each block. The implementations are largely adaptations of the old implementations, except
* LZ4 and LZ4HC are notably upgraded to take advantage of WorkingArea (see performance tests). And for simplicity in the new implementation, we are dropping support for some super old versions of the library.
* Getting snappy to work with limited-size output buffer required using the Sink/Source interfaces, which appear to be well supported for a long time and efficient (see performance tests).
* Replace awkward old CompressionManager::GetDecompressorForCompressor with Compressor::GetOptimizedDecompressor (which is optional to implement)
* Small behavior change where we treat lack of support for compression closer to not configuring compression, such as incompatibility with block_align. This is motivated by giving CompressionManager the freedom of determining when compression can be excluded for an entire file despite the configured "compression" type, and thus only surfacing actual incompatibilities not hypothetical ones that might be irrelevant to the CompressionManager (or build configuration). Unit tests in `table_test` and `compact_files_test` required update.
* Some lingering clean up of CompressedSecondaryCache and a re-optimization made possible by compressing into an existing buffer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13805
Test Plan:
for correctness, existing tests
## Performance Test
As I generally only modified compression paths, I'm using a db_bench write benchmark, with before & after configurations running at the same time. vc=1 means verify_compression=1
```
USE_CLANG=1 DEBUG_LEVEL=0 LIB_MODE=static make -j100 db_bench
SUFFIX=`tty | sed 's|/|_|g'`; for CT in zlib bzip2 none snappy zstd lz4 lz4hc none snappy zstd lz4 bzip2; do for VC in 0 1; do echo "$CT vc=$VC"; (for I in `seq 1 20`; do BIN=/dev/shm/dbbench${SUFFIX}.bin; rm -f $BIN; cp db_bench $BIN; $BIN -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 -compression_type=$CT -verify_compression=$VC 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done; done
```
zlib vc=0 524198 -> 524904 (+0.1%)
zlib vc=1 430521 -> 430699 (+0.0%)
bzip2 vc=0 61841 -> 60835 (-1.6%)
bzip2 vc=1 49232 -> 48734 (-1.0%)
none vc=0 1802375 -> 1906227 (+5.8%)
none vc=1 1837181 -> 1950308 (+6.2%)
snappy vc=0 1783266 -> 1901461 (+6.6%)
snappy vc=1 1799703 -> 1879660 (+4.4%)
zstd vc=0 1216779 -> 1230507 (+1.1%)
zstd vc=1 996370 -> 1015415 (+1.9%)
lz4 vc=0 1801473 -> 1943095 (+7.9%)
lz4 vc=1 1799155 -> 1935242 (+7.6%)
lz4hc vc=0 349719 -> 1126909 (+222.2%)
lz4hc vc=1 348099 -> 1108933 (+218.6%)
(Repeating the most important ones)
none vc=0 1816878 -> 1952221 (+7.4%)
none vc=1 1813736 -> 1904622 (+5.0%)
snappy vc=0 1794816 -> 1875062 (+4.5%)
snappy vc=1 1789363 -> 1873771 (+4.7%)
zstd vc=0 1202592 -> 1225164 (+1.9%)
zstd vc=1 994322 -> 1016688 (+2.2%)
lz4 vc=0 1786959 -> 1971518 (+10.3%)
lz4 vc=1 1829483 -> 1935871 (+5.8%)
I confirmed manually that the new WorkingArea for LZ4HC makes the huge difference on that one, but not as much difference for LZ4, presumably because LZ4HC uses much larger buffers/structures/whatever for better compression ratios.
Reviewed By: hx235
Differential Revision: D79111736
Pulled By: pdillinger
fbshipit-source-id: 1ce1b14af9f15365f1b6da49906b5073a8cecc14
Summary:
Unit Test for a repro for the fix that was reported by https://github.com/facebook/rocksdb/pull/13743
There's potential dataloss when Remote Compaction entries are all removed due to various reasons (CompactionFilter, DeleteRange covering all keys of the SST file, etc)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13812
Test Plan:
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.EmptyResult*"
```
Failed before merging https://github.com/facebook/rocksdb/pull/13743, now passing
Reviewed By: cbi42
Differential Revision: D79192829
Pulled By: jaykorean
fbshipit-source-id: e200300c4a7993de21c63cd92bda65b692921b89
Summary:
We were seeing some internal builds apparently failing the `-d /mnt/gvfs/third-party` check. Although third-party2 is likely a better check (see dependencies_platform010.sh), that would create a big headache with check_format_compatible.sh which has to work across codebase versions.
* Report a WARNING when we detect on a Meta machine but the `-d /mnt/gvfs/third-party` check fails
* Let USE_CLANG influence default compiler choice so that things might still work in that case (e.g. `USE_CLANG=1 make -j24 check`)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13820
Test Plan: manual, CI
Reviewed By: jaykorean
Differential Revision: D79277197
Pulled By: pdillinger
fbshipit-source-id: 19b2d45ed794f64bbf838f4414568d77ae9ca6f1
Summary:
Preserve tombstone when allow_ingest_behind` is enabled so that they can be applied to ingested files. This can be useful when users use ingest_behind to buffer updates where Deletion needs to be preserved. This fixes https://github.com/facebook/rocksdb/issues/13571.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13807
Test Plan: updated a unit test to verify that tombstones are not dropped during compaction.
Reviewed By: hx235
Differential Revision: D79016109
Pulled By: cbi42
fbshipit-source-id: c4d31ef32c88468ababcc1ea5af5db6de42a3b0d
Summary:
As title. We will re-enable it once fixed
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13815
Test Plan: N/A - Disabling the test.
Reviewed By: archang19
Differential Revision: D79172697
Pulled By: jaykorean
fbshipit-source-id: 936de3743816049cda811bde48b3b2207ed256ee
Summary:
**Issue**:
When running remote compaction, if all entries in the input files are expired, RocksDB incorrectly deletes an active file from the primary DB, leading to data loss and corruption.
**Root Cause**:
The current logic mistakenly mixed up the input and output file paths during the cleanup phase when no keys survive the compaction (all expired). This results in deleting the input files (which belong to the primary DB) instead of the output files (which belong to the SecondaryDB).
**Fix**:
Use `GetTableFileName` (virtual function) instead of `TableFileName`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13743
Reviewed By: hx235
Differential Revision: D79108650
Pulled By: jaykorean
fbshipit-source-id: 1c9ba971a0e9a62c15ebc014436cb8fc961af95c
Summary:
Building db_bench with clang and DEBUG_LEVEL=0 was failing with unused variable. This was not caught by CI so I have added this to the build-linux-clang-13-no_test_run job.
Also, while I was touching CI:
* Fold build-linux-release-rtti into build-linux-release by reducing the number of combinations tested between static/dynamic lib and rtti/not. I don't expect these to interact meaningfully with an extremely mature compiler.
* Combine build-linux-clang10-asan and build-linux-clang10-ubsan because clang is extremely reliable running both together
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13813
Test Plan: manual builds, CI
Reviewed By: krhancoc
Differential Revision: D79112643
Pulled By: pdillinger
fbshipit-source-id: 4ffc672718c05fa4597d637aacbc5a179ad8a0cf
Summary:
• Guard on __cpp_lib_atomic_shared_ptr to use std::atomic<std::shared_ptr<T>>::load()/store()
• Fallback to std::atomic_load_explicit()/store_explicit() under C++17
When attempting to build with CXX 20 using clang in a Linux environment, the build fails due to deprecation of atomic_load_explicit.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13744
Reviewed By: xingbowang
Differential Revision: D78997919
Pulled By: cbi42
fbshipit-source-id: f829c282cba878f072d4b0ad44192a87f73b8a90
Summary:
Simulate Remote Compaction in Stress Test by running a separate set of threads that runs remote compaction.
Queue and ResultMap for the remote compactions are stored in memory as part of the `SharedState`. They are shared across main worker threads and remote compaction worker threads.
`enable_remote_compaction` is replaced by `remote_compaction_worker_threads`.
If `remote_compaction_worker_threads` is set to 0, remote compaction is not enabled in Stress Test.
**To Follow up**
This PR covers happy path only. Failure injection in the remote worker thread will be added as a follow up.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13800
Test Plan:
```
./db_stress --remote_compaction_worker_threads=4 --flush_one_in=1000 --writepercent=40 --readpercent=40 --iterpercent=10 --prefixpercent=0 --delpercent=10 --destroy_db_initially=0 --clear_column_family_one_in=0 --reopen=0
```
```
python3 -u tools/db_crashtest.py blackbox --remote_compaction_worker_threads=8
```
Reviewed By: hx235
Differential Revision: D78862084
Pulled By: jaykorean
fbshipit-source-id: b262058c92d7fecc5e014cef5df9cca4a209921b
Summary:
To be compatible with some upcoming compression change/refactoring where we supply a fixed size buffer to CompressBlock, we need to support CompressedSecondaryCache storing uncompressed values when the compression ratio is not suitable. It seems crazy that CompressedSecondaryCache currently stores compressed values that are *larger* than the uncompressed value, and even explicitly exercises that case (almost exclusively) in the existing unit tests. But it's true.
This change fixes that with some other nearby refactoring/improvement:
* Update the in-memory representation of these cache entries to support uncompressed entries even when compression is enabled. AFAIK this also allows us to safely get rid of "don't support custom split/merge for the tiered case".
* Use more efficient in-memory representation for non-split entries
* For CompressionType and CacheTier, which are defined as single-byte data types, use a single byte instead of varint32. (I don't know if varint32 was an attempt at future-proofing for a memory-only schema or what.) Now using lossless_cast will raise a compiler error if either of these types is made too large for a single byte.
* Don't wrap entries in a CacheAllocationPtr object; it's not necessary. We can rely on the same allocator being provided at delete time.
* Restructure serialization/deserialization logic, hopefully simpler or easier to read/understand.
* Use a RelaxedAtomic for disable_cache_ to avoid race.
Suggested follow-up on CompressedSecondaryCache:
* Refine the exact strategy for rejecting compressions
* Still have a lot of buffer copies; try to reduce
* Revisit the split-merge logic and try to make it more efficient overall, more unified with non-split case
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13797
Test Plan:
Unit tests updated to use actually compressible strings in many places and more testing around non-compressible string.
## Performance Test
There was some pre-existing issue causing decompression failures in compressed secondary cache with cache_bench that is somehow fixed in this change. This decompression failures were present before the new compression API, but since then cause assertion failures rather than being quietly ignored. For the "before" test here, they are back to quietly ignored. And the cache_bench changes here were back-ported to the "before" configuration.
### No compressed secondary (setting expectations)
```
./cache_bench --cache_type=auto_hyper_clock_cache -cache_size=8000000000 -populate_cache
```
Max key : 3906250
Before:
Complete in 12.784 s; Rough parallel ops/sec = 2503123
Thread ops/sec = 160329; Lookup hit ratio: 0.686771
After:
Complete in 12.745 s; Rough parallel ops/sec = 2510717 (in the noise)
Thread ops/sec = 159498; Lookup hit ratio: 0.68686
### Compressed secondary, no split/merge
Same max key and approximate total memory size
```
/usr/bin/time ./cache_bench --cache_type=auto_hyper_clock_cache -cache_size=4000000000 -populate_cache -resident_ratio=0.125 -compressible_to_ratio=0.4 --secondary_cache_uri=compressed_secondary_cache://capacity=4000000000
```
Before:
Complete in 18.690 s; Rough parallel ops/sec = 1712144
Thread ops/sec = 108683; Lookup hit ratio: 0.776683
Latency: P50: 4205.19 P75: 15281.76 P99: 43810.98 P99.9: 71487.41 P99.99: 165453.32
max RSS (according to /usr/bin/time): 9341856
After:
Complete in 17.878 s; Rough parallel ops/sec = 1789951 (+4.5%)
Thread ops/sec = 114957; Lookup hit ratio: 0.792998 (+0.016)
Latency: P50: 4012.70 P75: 14477.63 P99: 40039.70 P99.9: 62521.04 P99.99: 167049.18
max RSS (according to /usr/bin/time): 9235688
The improved hit ratio is probably from fixing the failed decompressions (somehow). And my modifications could have improved CPU efficiency, or it could be the small penalty the benchmark naturally imposes on most misses (generate another value and insert it).
### Compressed secondary, with split/merge
```
/usr/bin/time ./cache_bench --cache_type=auto_hyper_clock_cache -cache_size=4000000000 -populate_cache -resident_ratio=0.125 -compressible_to_ratio=0.4 --secondary_cache_uri='compressed_secondary_cache://capacity=4000000000;enable_custom_split_merge=true'
```
Before:
Complete in 20.062 s; Rough parallel ops/sec = 1595075
Thread ops/sec = 101759; Lookup hit ratio: 0.787129
Latency: P50: 5338.53 P75: 16073.46 P99: 46752.65 P99.9: 73459.11 P99.99: 201318.75
max RSS (according to /usr/bin/time): 9049852
After:
Complete in 18.564 s; Rough parallel ops/sec = 1723771 (+8.1%)
Thread ops/sec = 110724; Lookup hit ratio: 0.813414 (+0.026)
Latency: P50: 5234.75 P75: 14590.43 P99: 41401.03 P99.9: 65606.50 P99.99: 157248.04
max RSS (according to /usr/bin/time): 8917592
Looks like an improvement
Reviewed By: anand1976
Differential Revision: D78842120
Pulled By: pdillinger
fbshipit-source-id: 5f754b160c37ebee789279178ebb5e862071bdb2
Summary:
Create a new API FileSystem::SyncFile for file sync, so that we could use file sync directly in places where we need to sync file content to file system without any modification. This is mostly used combined with link file. In some file system link file does not guarantee the file content is synced to file system.
https://github.com/facebook/rocksdb/issues/13741
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13762
Test Plan:
Unit test
T229418750
Reviewed By: pdillinger
Differential Revision: D78121137
Pulled By: xingbowang
fbshipit-source-id: 0ea8a5a3b486e0b61636700400613fed6bbd3faa
Summary:
The yield is actually of not much use because waitFor should already be doing that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13796
Reviewed By: pdillinger
Differential Revision: D78823656
Pulled By: jainpr
fbshipit-source-id: 040eaf596938ce8db535bc810ad77a9e50b2d551
Summary:
This diff fixes up a miss in which the property_bag was not pushed down to the BlockBasedIterator.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13795
Reviewed By: anand1976
Differential Revision: D78762294
Pulled By: krhancoc
fbshipit-source-id: 8970b0a87e35d07d5a0dd16f360ec96859f66550
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13794
LLVM has a warning `-Wdeprecated-redundant-constexpr-static-def` which raises the warning:
> warning: out-of-line definition of constexpr static data member is redundant in C++17 and is deprecated
Since we are now on C++20, we can remove the out-of-line definition of constexpr static data members. This diff does so.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D78635037
fbshipit-source-id: a90c68469947705c65f36588b2d575237689dbe8
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13793
LLVM has a warning `-Wdeprecated-redundant-constexpr-static-def` which raises the warning:
> warning: out-of-line definition of constexpr static data member is redundant in C++17 and is deprecated
Since we are now on C++20, we can remove the out-of-line definition of constexpr static data members. This diff does so.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: meyering
Differential Revision: D78635005
fbshipit-source-id: bd7cbfff0580b9579e78237ec4371615d3609536
Summary:
This patch reverted "NewRandomRWFile" back to "ReopenWritableFile" in external sst file ingestion job when file is linked instead of copied. The reason is that some of the file systems do not support "NewRandomRWFile". A long term fix is being worked in progress.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13791
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D78697825
Pulled By: xingbowang
fbshipit-source-id: d3651223ab1f2369aac34b772bba8049c6c2c628
Summary:
This diff introduces the ScanOption Pruning, previously the intent was to do prefetching for each sub-iterator of the level iterator, however since BlockBasedIterator does not prefetch asynchronously, this optimization does not make sense just yet.
For now we will prune the ScanOptions to the overlapping ranges and make sure they are properly piped to the underlying layers (during Prepare, and Seek).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13780
Reviewed By: cbi42
Differential Revision: D78436869
Pulled By: krhancoc
fbshipit-source-id: 681fe7f7f88b04b5c2d60cb3a5de01e03f6f8431
Summary:
So that we can use --command=recompress with a custom CompressionManager. (It's not required for reading files using a custom CompressionManager because those can already use ObjectLibrary for dependency injection.)
Suggested follow-up:
* These tests should not be using C arrays, snprintf, manual delete, etc. except for thin compatibility with argc/argv.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13783
Test Plan: unit test added, some manual testing
Reviewed By: archang19
Differential Revision: D78574434
Pulled By: pdillinger
fbshipit-source-id: 609e6c6439090e6b7e9b63fbd4c2d3f04b104fcf
Summary:
initial support for Prepare() to optimize the performance of MultiScan when using block-based tables. In Prepare(), we do the following:
1. Load all data blocks that will be read in multiscan to block cache
2. Pin the data blocks during the scan
3. if I/O is needed, coalesce I/Os when they are adjacent.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13778
Test Plan:
Added a new unit test.
Benchmark:
1. Set up the DB, I use FIFO here so that files will be in L0 and iterator will use BlockBasedTableIterator directly instead of LevelIterator, where Prepare() call is not implemented yet.
```
./db_bench --benchmarks="fillseq,compact" --disable_wal=1 --threads=1 --num_levels=1 --compaction_style=2 --fifo_compaction_max_table_files_size_mb=1000 --write_buffer_size=268435456
```
2. Multi-scan: based on https://github.com/facebook/rocksdb/issues/13765
```
./db_bench --db="/tmp/rocksdbtest-543376/dbbench" --use_existing_db=1 --benchmarks=multiscan --disable_auto_compactions=1 --seek_nexts=100 --threads=32 --duration=10 --statistics=1
multiscan_stride = 100
multiscan_size = 10
seek_nexts = 100
Main:
multiscan : 449.386 micros/op 70562 ops/sec 10.359 seconds 730968 operations; (multscans:22999)
multiscan : 453.606 micros/op 69433 ops/sec 10.369 seconds 719968 operations; (multscans:22999)
rocksdb.non.last.level.read.bytes COUNT : 47763519421
rocksdb.non.last.level.read.count COUNT : 21573878
Branch:
multiscan : 332.670 micros/op 94698 ops/sec 10.285 seconds 973968 operations; (multscans:29999)
rocksdb.non.last.level.read.bytes COUNT : 111791308336
rocksdb.non.last.level.read.count COUNT : 1062942
With direct-IO:
./db_bench --db="/tmp/rocksdbtest-543376/dbbench" --use_existing_db=1 --benchmarks=multiscan --disable_auto_compactions=1 --seek_nexts=100 --threads=32 --duration=10 --statistics=1 --use_direct_reads=1
Main:
multiscan : 586.045 micros/op 53825 ops/sec 10.366 seconds 557968 operations; (multscans:14999)
rocksdb.non.last.level.read.bytes COUNT : 69107458693
rocksdb.non.last.level.read.count COUNT : 6724651
Branch:
multiscan : 386.679 micros/op 81282 ops/sec 10.359 seconds 841968 operations; (multscans:25999)
rocksdb.non.last.level.read.bytes COUNT : 96605800558
rocksdb.non.last.level.read.count COUNT : 918973
```
Throughput is 36% higher with non-direct IO and 50% higher with direct IO. The improvement is likely from doing less number of I/Os due to I/O coalescing during Prepare(), as shown in `rocksdb.non.last.level.read.count`. The total bytes read is more with this PR for the same reason.
3. Regular iterator:
```
./db_bench --use_existing_db=1 --db="/tmp/rocksdbtest-543376/dbbench" --benchmarks=seekrandom --disable_auto_compactions=1 --seek_nexts=10 --threads=32 --duration=10
Main:
seekrandom : 13.014 micros/op 2456735 ops/sec 10.014 seconds 24602968 operations; 2717.8 MB/s (773999 of 773999 found)
Branch:
seekrandom : 13.048 micros/op 2450554 ops/sec 10.013 seconds 24537968 operations; 2710.9 MB/s (772999 of 772999 found)
```
The result fluctuates but without noticeable regression.
Reviewed By: anand1976
Differential Revision: D78440807
Pulled By: cbi42
fbshipit-source-id: 80ac6fd222696fa65ac0b4b5441748be5ee0b979
Summary:
When max_table_files_size was accidentally configured with 0 value, engine could crash on divide by 0 operation. Although RocksDB do configuration validation during bootstrap, it typically does not do this for runtime dynamic parameter validation. Therefore, there is a chance where max_table_files_size could be set to 0. This PR only focuses on fixing a code path where max_table_files_size ack as divisor.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13767
Test Plan: Unit test.
Reviewed By: cbi42
Differential Revision: D78420516
Pulled By: xingbowang
fbshipit-source-id: 6fdcc85b28a2c6319066665262b981e513719703
Summary:
This diff introduces the ability to override behavior of exits, allow for users to catch exits in a try catch for example as opposed to fully exiting the process.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13772
Reviewed By: hx235
Differential Revision: D78244499
Pulled By: krhancoc
fbshipit-source-id: b403327ed5b494a22b6beeaad4083945a1def0c7
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13728
The Prepare interface allows the user defined index iterator to prefetch index entries, as well as take custom scan termination criteria specified in the property_bag into account.
Reviewed By: pdillinger
Differential Revision: D76165546
fbshipit-source-id: 83d628598924aa7a60dff7ed62a16ae575b2c8ec
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13727
Add UserDefinedIndexReader and UserDefinedIndexIterator. The BlockBasedTable reads the user defined index meta block during open, verifies the checksum, pins in cache or heap depending on configuration, and allocates a UserDefinedIndexReader object with the contents. Similar to the builder, an IndexReader wrapper is allocated. The wrapper forwards the calls to the native reader and/or user defined index reader as appropriate.
A new option, table_index_name, in ReadOptions specifies the index to use when creating a new Iterator.
Reviewed By: pdillinger
Differential Revision: D76165694
fbshipit-source-id: c30bde4c5ce91ea3dc9ad302e73fe4963c1ed457
Summary:
In SstFileWriter::Finish, the call to DeleteFile to delete the output file in case of an error may fail. The current behavior is to ignore the error. In stress tests, there may be expected failures due to error injection. Not acting on the return status will cause the ASSERT_STATUS_CHECKED test to fail, so silence it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13776
Reviewed By: mszeszko-meta
Differential Revision: D78307124
Pulled By: anand1976
fbshipit-source-id: d27d9397c15cac5cb33b27094c9123a3fde7fa24
Summary:
* A number of comments clarifying contracts, etc.
* Make ReleaseWorkingArea public instead of protected because there are some limited cases where a wrapper implementation might want to call it directly
* Check non-empty dictionary precondition on MaybeCloneForDict
* Expand testing of wrapped WorkingAreas
* Random documentation improvement in block_builder.cc
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13775
Test Plan: existing and expanded tests and assertions
Reviewed By: hx235
Differential Revision: D78304550
Pulled By: pdillinger
fbshipit-source-id: e5f064e8405a5a49be123ee13145cb3626bbbfbf
Summary:
The Stress test was broken due to a change in switching from ReopenWritableFile to FSRandomRWFile for sync linked file in external Sst ingestion job. The Stress test is using FaultInjectionFs, which tracks the opening of ReopenWritableFile properly, but does not track FSRandomRWFile properly. This change fixes the tracking of FSRandomRWFile in FaultInjectionFs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13771
Test Plan: unit test, stress test
Reviewed By: mszeszko-meta
Differential Revision: D78282719
Pulled By: xingbowang
fbshipit-source-id: f8f2ed8a5b28a76836f75effbdfa2c3bb172dc51
Summary:
in log_reader.cc.
* `for (;;)` (with no matching break inside) should be more structurally recognizable to compilers as unreachable after compared to `while (true)` which compilers can treat as conditional for warning/error purposes because `true` might have come from a macro, etc.
* Comment the `break` statements to indicate they are for the `switch` (not the `for`)
* No code or annotation is apparently needed for the unreachable end of the non-void function, so just a comment
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13764
Test Plan: CI
Reviewed By: archang19
Differential Revision: D78135493
Pulled By: pdillinger
fbshipit-source-id: e313435a846a6e15346acf40404f755be98ab09a
Summary:
This diff reverts D77424529
Unland reason: This diff broke our Windows 2022 build for Open Source CI (T230460952).
Depends on D77424529
Reviewed By: pdillinger
Differential Revision: D78107313
fbshipit-source-id: 6177448e1015c239abcebb0e68470dfd841b6fa0
Summary:
I was going to use this in some code I was working on but ended up not needing it. But it's useful nonetheless and I'm using it in a few places to replace reinterpret_cast.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13648
Test Plan: existing tests, manually see compilation fail when pointed-to types are not same size integral types
Reviewed By: cbi42
Differential Revision: D75576195
Pulled By: pdillinger
fbshipit-source-id: e10c7a4959340f6f2b536de8088072a90e871fcf
Summary:
LLVM has a warning `-Wdeprecated-redundant-constexpr-static-def` which raises the warning:
> warning: out-of-line definition of constexpr static data member is redundant in C++17 and is deprecated
Since we are now on C++20, we can remove the out-of-line definition of constexpr static data members. This diff does so.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Differential Revision: D77423205
fbshipit-source-id: 4ee4a390431a5d25e7733311f3fa40395dfd4bc0
Summary:
LLVM has a warning `-Wunreachable-code-return` which identifies return statements that cannot be reached.
In innocuous situations such statements are often present:
* to satisfy a compiler warning that existed before `[[noreturn]]` was introduced. Now that we have `[[noreturn]]`, this use is not necessary.
* to specify a return type. But there are clearer ways to do this.
* in place of the more legible `__builtin_unreachable()` (which will soon become `std::unreachable()`). In this case, we should use the more legible alternative.
* because the programmer was afraid of the function unexpectedly returning. But we check for this condition with `-Wreturn-type`.
In dangerous situations such statements can obscure the intended execution of the program or even hide an erroneous early return.
In this diff, we remove one or more unreachable returns.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Differential Revision: D77424529
fbshipit-source-id: fe41b5a640264d0a299d5ad330c645f94b147323
Summary:
Add file size validation in ReadFooterFromFile function.
Deprecate skip_checking_sst_file_sizes_on_db_open option.
This change is used to address this issue
https://github.com/facebook/rocksdb/issues/13619
It supports file size validation in ReadFooterFromFile. In favor of this
change, CheckConsistency function and
skip_checking_sst_file_sizes_on_db_open flag are deprecated.
The CheckConsistency function checks each file size matches what was
recorded in manifest during DB open. Meantime, ReadFooterFromFile was
called for each file in LoadTables function. Since ReadFooterFromFile
always validates file size, the CheckConsistency is redundant.
In addtion, CheckConsistency is executed in a single thread. This could
slow down DB open when a network file system is used. Therefore, the
flag skip_checking_sst_file_sizes_on_db_open was added to skip this
check. After this change, ReadFooterFromFile was executed in parallel
through multiple threads. Therefore, the concern of DB open slowness is
eliminated, and the flag could be deprecated.
When paranoid check flag is set to true, corrupted file will fail to open the DB.
When paranoid check flag is set to false, DB will still be able to open, the
healthy ones can be accessed, while the corrupted ones not.
There is 2 slight concerns of this change.
*If max_open_files is set with smaller value, engine will not open all
the files during DB open. This means if there is a corruption on file
size, it will not be detected during DB open, but rather at a later
time. Since the default is -1, which means open all the files, and it is
rarely overridden and a lot of new features rely on it to be -1, the
risk is very low.
*If FIFO compaction is used, engine could fail to open DB unnecessarily
on the corrupted files that would never be used again. However, this is
a very rare case as well. The error could still be ignored by setting
paranoid_checks operationally. The risk is very low.
To remain backward compatibility. The public facing flag was kept and
marked as no-op internally. Another change is required to fully remove
the flag.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13676
Test Plan:
make check
A new unit test was added to validate file size check API works as
expected.
Reviewed By: pdillinger
Differential Revision: D76168033
Pulled By: xingbowang
fbshipit-source-id: 8ceacf39bcfe02ff7aa289868c341366ee9f3a8e
Summary:
`SstFileManager` is supposed to be thread-safe for all of its public methods, but `SetStatisticsPtr` leads to a race condition because the access to `stat_` is not synchronized. We don't use `stat_` internally so we can get rid of it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13757
Test Plan: Existing unit tests.
Reviewed By: mszeszko-meta
Differential Revision: D77962592
Pulled By: archang19
fbshipit-source-id: e8e56194dda034935ddef44e479243770a73d065
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13726
Add UserDefinedIndexFactory and UserDefinedIndexBuilder interfaces to allow users to plugin custom index implementation into block based table. The factory is specified in BlockBasedTableOptions. If non-null, BlockBasedTableBuilder allocates a wrapper index builder encapsulating the native index and the custom index. The custom index is exposed to BlockBasedTableBuilder as a meta_block of type kUserDefinedIndex. This block type is not compressed.
The IndexBuilder OnKeyAdded interface is enhanced to accept the value in addition to the key. Only full values are supported, and parallel compression is not supported since we cannot obtain the value when calling OnKeyAdded.
Reviewed By: pdillinger
Differential Revision: D76165614
fbshipit-source-id: dfad9cbd6d0359987b7f4abe64cae58c472836f9
Summary:
While a transaction is waiting on a lock, we can use GetWaitingTxns() to determine the transactionID of the blocking transaction and the contended key. However, this gets cleared when the lock times out, so if a client has widespread timeout errors, you need to catch a transaction 'in the act' before they actually hit the timeout in order to understand the contention pattern. This diff adds a new TransactionOptions variable enable_get_waiting_txn_after_timeout, which persists the lock contention information after timeout so it can be accessed by the client after they have received the timeout error.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13754
Test Plan:
- updated TransactionTest.WaitingTxn to test the changed behavior
- ran production shadow tests on traffic with frequent timeouts
Reviewed By: cbi42
Differential Revision: D77703598
Pulled By: akabcenell
fbshipit-source-id: b4448ca1b6a3694d51bfe1ce801b09eb376ff3e9
Summary:
See `Makefile` for actual changes:
* ZLIB remains the same
* BZIP2 remains the same
* SNAPPY is a minor update
* LZ4 is a significant update with multithreaded/multicore compression https://github.com/lz4/lz4/releases/tag/v1.10.0
* ZSTD is a significant update RocksDB is called out as benefiting in particular from the performance improvements herein https://github.com/facebook/zstd/releases/tag/v1.5.7
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13609
Reviewed By: archang19
Differential Revision: D77877295
Pulled By: mszeszko-meta
fbshipit-source-id: bf9a257e8f68dec3d02743b339aa2df65df4ab2c
Summary:
Was reading sec_cache_res_ratio_ outside of mutex and using the result for computation that needs to be synchronized
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13747
Test Plan: existing tests. Has been showing up in crash test, and there's no interesting concurrency here that would warrant a regression test based on sync points.
Reviewed By: cbi42
Differential Revision: D77607660
Pulled By: pdillinger
fbshipit-source-id: 12a71936b3558c7528d229a11c7d2e43982ad06b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13752
... to github repo. This include changes from D77323287, D77473923 and the release note change in patch release: D77611483.
Reviewed By: archang19
Differential Revision: D77670619
fbshipit-source-id: 37d877f3317c71de190128fa4da6b18f6dfcf3c5
Summary:
address an existing limitation on compaction triggering mechanism that relies on events like flush/compaction/SetOptions. This is important for periodic compactions where files can become eligible without any of these events. The periodic task now runs every 12 hours and check CFs that enables `periodic_compaction_second` (TBD if we want to expand to all CFs) for eligible compactions.
Some of the periodic tasks probably don't need to run immediately after Register(). I'm keeping the existing behavior for now for patch release and to makes tests happy.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13736
Test Plan:
- new unit test that fails before this change.
- ran crash test for hours with the periodic task running every 5 seconds: `python3 ./tools/db_crashtest.py blackbox --test_batches_snapshot=0 --periodic_compaction_seconds=10`
Reviewed By: pdillinger
Differential Revision: D77460715
Pulled By: cbi42
fbshipit-source-id: 00f61502753185e76830c9ed44c5ccc4f4f16bfa
Summary:
CostAwareCompressor simply ignores the preferred compression type as compression manager setting takes precedence over the compression type setting. Thus, I am removing the assert statement as it itself is unnecessary for this case.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13739
Test Plan:
Run nightly build test
```bash
make V=1 J=4 -j4 check
```
Reviewed By: hx235
Differential Revision: D77470932
Pulled By: shubhajeet
fbshipit-source-id: ebb69367d2ffb9bd72432fd04b0cd12ce2d6240a
Summary:
improve assertions, one apparently a previous typo in https://github.com/facebook/rocksdb/issues/13606 and one a suspected possible area of logic error
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13737
Test Plan: watch crash test
Reviewed By: anand1976
Differential Revision: D77453102
Pulled By: pdillinger
fbshipit-source-id: d4196910a9e8d59ef814130a52ff4ebf188a976d
Summary:
I am seeing crashes during backups. The stack trace points back to `WritableFileWriter` creation inside `BackupEngineImpl::CopyOrCreateFile`. I believe the issue is that we are calling `writable_file_->GetRequiredBufferAlignment()` with a `null` `writable_file`.
https://github.com/facebook/rocksdb/blob/v10.2.1/utilities/backup/backup_engine.cc#L2396-L2397https://github.com/facebook/rocksdb/blob/v10.2.1/file/writable_file_writer.h#L210
Here's how I think the flow is:
```cpp
io_s = dst_env->GetFileSystem()->NewWritableFile(dst, dst_file_options,
&dst_file, nullptr);
// say there was some issue and dst_file is nullptr
// evaluates to false
if (io_s.ok() && !src.empty()) {
// we don't go down this branch
auto src_file_options = FileOptions(src_env_options);
src_file_options.temperature = *src_temperature;
io_s = src_env->GetFileSystem()->NewSequentialFile(src, src_file_options,
&src_file, nullptr);
}
// say this evaluates to true
if (io_s.IsPathNotFound() && *src_temperature != Temperature::kUnknown) {
// Retry without temperature hint in case the FileSystem is strict with
// non-kUnknown temperature option
io_s = src_env->GetFileSystem()->NewSequentialFile(
src, FileOptions(src_env_options), &src_file, nullptr);
}
// this is now from the NewSequentialFile call, not NewWritableFile
if (!io_s.ok()) {
return io_s;
}
// dst_file is still nullptr
```
If the first `NewWritableFile` fails and `IsPathNotFound
Tests: existing unit tests
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13734
Reviewed By: pdillinger
Differential Revision: D77390694
Pulled By: archang19
fbshipit-source-id: 865a3a646079ae2349a3b6f25e53ae85df8e4985
Summary:
`data_` and `size_` fields are duplicated in `Block` class, as `contents_` field already have a `data` member variable, which contains `data` and `size` already. This reduces memory consumption by 16 bytes per block.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13733
Test Plan: Unit test
Reviewed By: pdillinger
Differential Revision: D77389791
Pulled By: xingbowang
fbshipit-source-id: 50a56bc5fae494ed5bc39bdfde7303ca06ce87c6
Summary:
Corrected misspelling of "Compression". Changed "Compresssion" to "Compression".
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13735
Test Plan:
All the test case for compression is still working properly.
```bash
./compression_test
```
Reviewed By: hx235
Differential Revision: D77390273
Pulled By: shubhajeet
fbshipit-source-id: f5310e393e23f5d6c8310154cb929db4b6c60a77
Summary:
Respect the scan upper bound/limit, if specified, in `MultiScan`. This applies to block based table and other native RocksDB SSTs. In order to properly support it, the `MultiScan` object caches the `ReadOptions` passed by the user and sets the `iterate_upper_bound` as appropriate. We optimize for the case of either all scans specifying the upper bound, or none of them. In case of mixed scans, we reallocate the DB iterator anytime `ReadOptions` has to be updated.
Tests:
New unit tests in `db_iterator_test`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13723
Reviewed By: cbi42
Differential Revision: D77385049
Pulled By: anand1976
fbshipit-source-id: 9c02d125770cbedbe6e8c10767ba537e7f7540e1
Summary:
Large txn commit optimization requires all updates are added to a transaction's WriteBatchWithIndex. However, some usage of transactions may add updates directly to the WBWI's underlying write batch. In these cases, we should not attempt to ingest the WBWI since it will drop these updates. This PR adds sanity checking for this.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13722
Test Plan:
- added checks in unit test and stress test
- manually check LOG files for the new unit test
Reviewed By: hx235
Differential Revision: D77247688
Pulled By: cbi42
fbshipit-source-id: 3d1c0c6e64d6d7dfd5578bc4d77abe44cac1e419
Summary:
The nightly build was failing because we were using the AutoSkipCompressionManager with kNoCompression. The test cases should not be running with NoCompression.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13716
Test Plan:
Run the test code being run on the nightly build.
```bash
make V=1 J=4 -j4 check
```
Reviewed By: hx235
Differential Revision: D77042874
Pulled By: shubhajeet
fbshipit-source-id: 821643b30ca53b1855fc24e3bc0a319e4fec2876
Summary:
Port changes made directly in fbcode in order to facilitate the 10.4 release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13714
Test Plan: Existing tests
Reviewed By: mszeszko-meta
Differential Revision: D77038668
Pulled By: anand1976
fbshipit-source-id: 6b9b16d62bccf75923b525c1c24597a59920a948
Summary:
* Make new format_version=7 a supported setting.
* Fix a bug in compressed_secondary_cache.cc that is newly exercised by custom compression types and showing up in crash test with tiered secondary cache
* Small change to handling of disabled compression in fv=7: use empty compression manager compatibility name.
* Get rid of GetDefaultBuiltinCompressionManager() in public API because it could cause unexpected+unsafe schema change on a user's CompressionManager if built upon the default built-in manager and we add a new built-in schema. Now must be referenced by explicit compression schema version in the public API. (That notion was already exposed in compressed secondary cache API, for better or worse.)
* Improve some error messages for compression misconfiguration
* Improve testing with ObjectLibrary and CompressionManagers
* Improve testing of compression_name table property in BlockBasedTableTest.BlockBasedTableProperties2
* Improve some comments
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13713
Test Plan: existing and updated tests. Notably, the crash test has already been running with (unpublished) format_version=7
Reviewed By: mszeszko-meta, hx235
Differential Revision: D77035482
Pulled By: pdillinger
fbshipit-source-id: 95278de8734a79706a22361bff2184b1edb230ca
Summary:
Auto skip compression manager code is currently running only in context of test / db bench. Disable failing test to unblock monthly minor release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13715
Test Plan: Disable test.
Reviewed By: hx235
Differential Revision: D77039218
Pulled By: mszeszko-meta
fbshipit-source-id: f9eeec8d5ca4efeaf1f490c5f091b3aff7861a4a
Summary:
Some pieces of follow-up to https://github.com/facebook/rocksdb/issues/13659.
_Recommend hiding whitespace for review_
* Add support for instantiating CompressionManagers through CreateFromString/ObjectLibrary.
* Pull CompressorCustomAlg and DecompressorCustomAlg out of db_test2, refactor/improvement them a bit, and put them in testutil.h for sharing with db_stress. Switched it from being built on snappy to being built on lz4 so that it can properly test dictionary compression.
* Add a custom compression manager for db_stress that uses these, and add to crash test. This depends on the ObjectLibrary stuff because some invocations of db_stress will not be configured with the custom compression manager but will need to access it to read some existing SST files.
* Remove some pieces where the concern of setting compression=kZSTD for compatibility purposes had leaked into configuring some tests and compression managers. After https://github.com/facebook/rocksdb/issues/13659 this compatibility concern is contained in the SST building code.
* Fix BuiltinDecompressorV2SnappyOnly hiding the (ignored) compression dictionary. SST read logic expects the serialized dictionary to be returned by the decompressor even if it's effectively ignored. Updated DBBlockCacheTest.CacheCompressionDict to cover this case.
For follow-up:
* Combine custom compression and mixed compression types in a file (not clean/easy without duplicating or majorly refactoring the mixed/random compressor)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13710
Test Plan: unit tests updated
Reviewed By: hx235
Differential Revision: D76928974
Pulled By: pdillinger
fbshipit-source-id: 772cf9cb048d737699b0e2887c624fb64a68aa8c
Summary:
add the `min_file_size` parameter to CompactOnDeletionCollector. A file must be at least this size for it to qualify for DTC. This is useful when a user wants to specific a min file size requirement that is larger than the size constraint imposed by the sliding window's `deletion_trigger` requirement.
Added some comment explaining that the file_size provided to table property collector only includes data blocks and may not be up-to-date. This PR also updates DTC to consider SingleDelete and DeletionWithTimestamp as tombstones.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13707
Test Plan:
- new unit test for when min_file_size is specified.
- existing unit test for when min_file_size is not specified.
Reviewed By: hx235, pdillinger
Differential Revision: D76837231
Pulled By: cbi42
fbshipit-source-id: 0782144e75aef9961bf03da2a2c4b3c613ce5db3
Summary:
Some usage of vector memtable is bottlenecked in the memtable insertion path when using multiple writers. This PR adds support for concurrent writes for the vector memtable. The updates from each concurrent writer are buffered in a thread local vector. When a writer is done, MemTable::BatchPostProcess() is called to flush the thread local updates to the main vector. TSAN test and function comment suggest that ApproximateMemoryUsage() needs to be thread-safe, so its implementation is updated to provide thread-safe access.
Together with unordered_write, benchmark shows much improved insertion throughput.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13675
Test Plan:
- new unit test
- enabled some coverage of vector memtable in stress test
- Performance benchmark: benchmarked memtable insertion performance with by running fillrandom 20 times
- Compare branch and main performance with one thread and write batch size 100:
- main: 4896888.950 ops/sec
- branch: 4923366.350 ops/sec
- Benchmark this branch by configuring different threads, allow_concurrent_memtable_write, and unordered_write. Performance ratio is computed as current ops/sec divided by ops/sec at 1 thread with the same options.
allow_concurrent | unordered_write | Threads | ops/sec | Performance Ratio
-- | -- | -- | -- | --
0 | 0 | 1 | 4923367 | 1.0
0 | 0 | 2 | 5215640 | 1.1
0 | 0 | 4 | 5588510 | 1.1
0 | 0 | 8 | 6077525 | 1.2
1 | 0 | 1 | 4919060 | 1.0
1 | 0 | 2 | 5821922 | 1.2
1 | 0 | 4 | 7850395 | 1.6
1 | 0 | 8 | 10516600 | 2.1
1 | 1 | 1 | 5050004 | 1.0
1 | 1 | 2 | 8489834 | 1.7
1 | 1 | 4 | 14439513 | 2.9
1 | 1 | 8 | 21538098 | 4.3
```
mkdir -p /tmp/bench_$1
export TEST_TMPDIR=/tmp/bench_$1
memtablerep_value=${6:-vector}
(for I in $(seq 1 $2)
do
/data/users/changyubi/vscode-root/rocksdb/$1 --benchmarks=fillrandom --seed=1722808058 --write_buffer_size=67108864 --min_write_buffer_number_to_merge=1000 --max_write_buffer_number=1000 --enable_pipelined_write=0 --memtablerep=$memtablerep_value --disable_auto_compactions=1 --disable_wal=1 --avoid_flush_during_shutdown=1 --allow_concurrent_memtable_write=${5:-0} --unordered_write=$4 --batch_size=1 --threads=$3 2>&1 | grep "fillrandom"
done;) | awk '{ t += $5; c++; print } END { printf ("%9.3f\n", 1.0 * t / c) }';
```
Reviewed By: pdillinger
Differential Revision: D76641755
Pulled By: cbi42
fbshipit-source-id: c107ba42749855ad4fd1f52491eb93900757542e
Summary:
* Improve debugability with better error messages (including the returned status, not just log messages)
* Tolerate user providing file checksums recognized by the factory but not the same function as currently, generally provided by the factory. This makes it practical to transition from one type of checksum to another without major hiccups in ingestion workflows.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13708
Test Plan: updated unit test, manually inspect LOG file from the unit test
Reviewed By: cbi42
Differential Revision: D76837804
Pulled By: pdillinger
fbshipit-source-id: 45b744829b3a125e9d0ee6874bd37ce534c2e13c
Summary:
**Summary:**
We need to move the Predictor to WorkingArea so that it is local to each thread and thus is thread safe.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13706
Test Plan: It should pass the test case written in ./compression_test.
Reviewed By: pdillinger
Differential Revision: D76836846
Pulled By: shubhajeet
fbshipit-source-id: 0d0170baf65f4bb95ba107fec77151e66b8a4449
Summary:
**Summary**:
Clang tidy was throwing error that the gtest assertion EXPECT_EQ was being carried out variables that was not initialized.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13703
Test Plan:
Ran the clang-tidy operations to make sure the same error does not appear.
```bash
CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 CLANG_ANALYZER="/usr/bin/clang++-10" CLANG_SCAN_BUILD=scan-build-10 USE_CLANG=1 make V=1 -j32 analyze
```
Reviewed By: hx235, pdillinger
Differential Revision: D76777988
Pulled By: shubhajeet
fbshipit-source-id: b9bfe26a2264d4c21224ab53a0b0307596d7f49d
Summary:
The Object registry requires object to be allocated as std::unique_ptr. Hence we provide a new API for external table plugins to allocate and return a unique_ptr ExternalTableFactory wrapper.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13694
Reviewed By: jaykorean
Differential Revision: D76767974
Pulled By: anand1976
fbshipit-source-id: ac59c523a11679ca7c9f0b280325c7873c6b4c07
Summary:
This change builds on https://github.com/facebook/rocksdb/issues/13540 and https://github.com/facebook/rocksdb/issues/13626 in allowing a CompressionManager / Compressor / Decompressor to use a custom compression algorithm, with a distinct CompressionType. For background, review the API comments on CompressionManager and its CompatibilityName() function.
Highlights:
* Reserve and name 127 new CompressionTypes that can be used for custom compression algorithms / schemas. In many or most cases I expect the enumerators such as `kCustomCompression8F` to be used in user code rather than casting between integers and CompressionTypes, as I expect the supported custom compression algorithms to be identifiable / enumerable at compile time.
* When using these custom compression types, a CompressionManager must use a CompatibilityName() other than the built-in one AND new format_version=7 (see below).
* When building new SST files, track the full set of CompressionTypes actually used (usually just one aside from kNoCompression), using our efficient bitset SmallEnumSet, which supports fast iteration over the bits set to 1. Ideally, to support mixed or non-mixed compression algorithms in a file as efficiently as possible, we would know the set of CompressionTypes as SST file open time.
* New schema for `TableProperties::compression_name` in format_version=7 to represent the CompressionManager's CompatibilityName(), the set of CompressionTypes used, and potentially more in the future, while keeping the data relatively human-readable.
* It would be possible to do this without a new format_version, but then the only way to ensure incompatible versions fail is with an unsupported CompressionType tag, not with a compression_name property. Therefore, (a) I prefer not to put something misleading in the `compression_name` property (a built-in compression name) when there is nuance because of a CompressionManager, and (b) I prefer better, more consistent error messages that refer to either format_version or the CompressionManager's CompatibilityName(), rather than an unrecognized custom CompressionType value (which could have come from various CompressionManagers).
* The current configured CompressionManager is passed in to TableReaders so that it (or one it knows about) can be used if it matches the CompatibilityName() used for compression in the SST file. Until the connection with ObjectRegistry is implemented, the only way to read files generated with a particular CompressionManager using custom compression algorithms is to configure it (or a known relative; see FindCompatibleCompressionManager()) in the ColumnFamilyOptions.
* Optimized snappy compression with BuiltinDecompressorV2SnappyOnly, to offset some small added overheads with the new tracking. This is essentially an early part of the planned refactoring that will get rid of the old internal compression APIs.
* Another small optimization in eliminating an unnecessary key copy in flush (builder.cc).
* Fix some handling of named CompressionManagers in CompressionManager::CreateFromString() (problem seen in https://github.com/facebook/rocksdb/issues/13647)
Smaller things:
* Adds Name() and GetId() functions to Compressor for debugging/logging purposes. (Compressor and Decompressor are not expected to be Customizable because they are only instantiated by a CompressionManager.)
* When using an explicit compression_manager, the GetId() of the CompressionManager and the Compressor used to build the file are stored as bonus entries in the compression_options table property. This table property is not parsed anywhere, so it is currently for human reading, but still could be parsed with the new underscore-prefixed bonus entries. IMHO, this is preferable to additional table properties, which would increase memory fragmentation in the TableProperties objects and likely take slightly more CPU on SST open and slightly more storage.
* ReleaseWorkingArea() function from protected to public to make wrappers work, because of a quirk in C++ (vs. Java) in which you cannot access protected members of another instance of the same class (sigh)
* Added `CompressionManager:: SupportsCompressionType()` for early options sanity checking.
Follow-up before release:
* Make format_version=7 official / supported
* Stress test coverage
Sooner than later:
* Update tests for RoundRobinManager and SimpleMixedCompressionManager to take advantage of e.g. set of compression types in compression_name property
* ObjectRegistry stuff
* Refactor away old internal compression APIs
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13659
Test Plan:
Basic unit test added.
## Performance
### SST write performance
```
SUFFIX=`tty | sed 's|/|_|g'`; for ARGS in "-compression_type=none" "-compression_type=snappy" "-compression_type=zstd" "-compression_type=snappy -verify_compression=1" "-compression_type=zstd -verify_compression=1" "-compression_type=zstd -compression_max_dict_bytes=8180"; do echo $ARGS; (for I in `seq 1 20`; do BIN=/dev/shm/dbbench${SUFFIX}.bin; rm -f $BIN; cp db_bench $BIN; $BIN -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -format_version=7 $ARGS 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```
Ops/sec, Before -> After, both fv=6:
-compression_type=none
1894386 -> 1858403 (-2.0%)
-compression_type=snappy
1859131 -> 1807469 (-2.8%)
-compression_type=zstd
1191428 -> 1214374 (+1.9%)
-compression_type=snappy -verify_compression=1
1861819 -> 1858342 (+0.2%)
-compression_type=zstd -verify_compression=1
979435 -> 995870 (+1.6%)
-compression_type=zstd -compression_max_dict_bytes=8180
905349 -> 940563 (+3.9%)
Ops/sec, Before fv=6 -> After fv=7:
-compression_type=none
1879365 -> 1836159 (-2.3%)
-compression_type=snappy
1865460 -> 1830916 (-1.9%)
-compression_type=zstd
1191428 -> 1210260 (+1.6%)
-compression_type=snappy -verify_compression=1
1866756 -> 1818989 (-2.6%)
-compression_type=zstd -verify_compression=1
982640 -> 997129 (+1.5%)
-compression_type=zstd -compression_max_dict_bytes=8180
912608 -> 937248 (+2.7%)
### SST read performance
Create DBs
```
for COMP in none snappy zstd; do echo $ARGS; ./db_bench -db=/dev/shm/dbbench-7-$COMP --benchmarks=fillseq,flush -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -compression_type=$COMP -format_version=7; done
```
And test
```
for COMP in none
snappy zstd none; do echo $COMP; (for I in `seq 1 8`; do ./db_bench -readonly -db=/dev/shm/dbbench
-7-$COMP --benchmarks=readrandom -num=10000000 -duration=20 -threads=8 2>&1 | grep micros/op; done
) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```
Ops/sec, Before -> After (both fv=6)
none
1491732 -> 1500209 (+0.6%)
snappy
1157216 -> 1169202 (+1.0%)
zstd
695414 -> 703719 (+1.2%)
none (again)
1491787 -> 1528789 (+2.4%)
Ops/sec, Before fv=6 -> After fv=7:
none
1492278 -> 1508668 (+1.1%)
snappy
1140769 -> 1152613 (+1.0%)
zstd
696437 -> 696511 (+0.0%)
none (again)
1500585 -> 1512037 (+0.7%)
Overall, I think we can take the read CPU improvement in exchange for the hit (in some cases) on background write CPU
Reviewed By: hx235
Differential Revision: D76520739
Pulled By: pdillinger
fbshipit-source-id: e73bd72502ff85c8779cba313f26f7d1fd50be3a
Summary:
Add an atomic bool to CompactionOptions to cancel an ongoing CompactFiles() operation, in the same fashion we do for CompactRange().
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13687
Test Plan: ./db_test2 --gtest_filter=DBTest2.TestCancelCompactFiles
Reviewed By: jaykorean
Differential Revision: D76538529
Pulled By: virajthakur
fbshipit-source-id: 77db5b4fb4cbd5280584834df28e51a72b084dab
Summary:
**Summary:**
This pull request configures RocksDB to optionally utilize this customized compressor (RandomCompressor) in the db stress test. It randomly selects the compression algorithm among the blocks.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13691
Test Plan: Testing was performed by verifying the stdout output from both RandomCompressor.
Reviewed By: hx235
Differential Revision: D76624220
Pulled By: shubhajeet
fbshipit-source-id: d9c458eeee930b25e8a87a77dc29f0647836310e
Summary:
**Context:**
RocksDB's current compression approach rejects blocks if the compressed size exceeds a predefined threshold. To optimize performance, we aim to develop an algorithm that dynamically stops and resumes block compression attempts based on past rejection data.
**Summary:**
The goal of this milestone is to design, implement, and evaluate an algorithm that intelligently skips and resumes block compression attempts in RocksDB. The algorithm tracks whether randomly selected blocks was rejected, compressed or bypassed and using data of window size to determine the current rejection rate. The calculate rejection rate is used to decide whether to pause and resume compression attempts. We measure the effectiveness of skipping and resuming compression using DB bench and identify any concerning regressions in correctness and performance.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13674
Test Plan:
1. Test case to see if it can automatically start compression on compression friendly workload and see if it can automatically stop compression on non-compression friendly workload (auto_skip_compresor_test.cc)
3. Regression analysis to prove that no significant performance attempt
```bash
SUFFIX=`tty | sed 's|/|_|g'`; for ARGS in "-compression_parallel_threads=1 -compression_type=zstd -compression_manager=none" "-compression_parallel_threads=4 -compression_type=zstd -compression_manager=none" "-compression_parallel_threads=1 -compression_type=zstd -compression_manager=autoskip" "-compression_parallel_threads=4 -compression_type=zstd -compression_manager=autoskip" ; do echo $ARGS; (for I in `seq 1 20`; do ./db_bench -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 $ARGS 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```
Measurement experiment | throughput (% change from main branch) |
|---------------|--------------------------------|
compression manager = none (main branch) | 1106890.35 ops/s
compression manager = none (auto skip) | 1097574.55 ops/s (-0.84%)
compression manager = auto skip (auto skip branch) | 1133432.9 ops/s (+2.4%)
Reviewed By: hx235
Differential Revision: D76220795
Pulled By: shubhajeet
fbshipit-source-id: 0f46ab34da1b451f8907306afba221503e6e22a5
Summary:
Since `request_id` is a raw pointer to a string, copying `IODebugContext` becomes a little bit more complicated. We need to ensure that `request_id` gets its memory freed, but by we don't have ownership of the memory by default. The `request_id` inside `IODebugContext` is meant to point to a string allocated outside of the RocksDB read request. To get around this issue without refactoring `request_id`'s type entirely, we can store a private member variable and have `request_id` point to it, so the memory deallocation happens automatically for us.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13690
Test Plan:
I updated the `RequestIdPlumbingTest` unit test from https://github.com/facebook/rocksdb/issues/13616
```
./db_test --gtest_filter=DBTest.RequestIdPlumbingTest
```
Reviewed By: anand1976
Differential Revision: D76613051
Pulled By: archang19
fbshipit-source-id: 053a5b9c4cde20606ec7854ada29904bdf11d40c
Summary:
This field will be used internally to feed Warm Storage cost information back through the Sally IO stack. This is needed for cost accounting / reporting.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13666
Test Plan: I made the additional changes needed to set/record the new cost info field, and confirmed that this information could be fed through.
Reviewed By: anand1976
Differential Revision: D76070434
Pulled By: archang19
fbshipit-source-id: 2fab975f14fd8f7c20b5d0d85c31686ccf682068
Summary:
**Context:**
RocksDB currently selects files for long-running compaction outputs to the bottommost level, preventing these selected files files from being selected, but does not execute the compaction immediately like other compactions. Instead, this compaction is forwarded to another Env::Priority::bottom thread pool, where it waits (potentially for a long time) until its thread is ready to execute. This extended L0 lock time in universal compaction caused our users write stall and read performance regression.
**Summary:**
This PR is to eliminate L0 lock time during bottom priority compaction waiting to execute by the following
- Create and forward an intended compaction only consists of last input file (or sorted run if non-L0) instead of all the input files. This eliminate the locking for non-bottommost level input files while waiting for bottom priority thread is up to run.
- Re-pick compaction that outputs to max output level when bottom priority thread is up to run
- Refactor universal compaction picking logic to make it cleaner and easier to force picking compaction with max output level when bottom priority thread is up to run
- Guard feature behind a temporary option as requested
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13633
Test Plan:
- New unit test to cover the case that's not covered by existing tests - bottom priority thread re-picks compaction ends up picking nothing due to LSM shape changes
- Adapted existing unit tests to verify various bottom priority compaction behavior with this new option
- Stress test `python3 tools/db_crashtest.py --simple blackbox --compaction_style=1 --target_file_size_base=1000 --write_buffer_size=1000 --compact_range_one_in=10000 --compact_files_one_in=10000 `
Reviewed By: cbi42
Differential Revision: D76005505
Pulled By: hx235
fbshipit-source-id: 9688f22d4a84f619452820f12f15b765c17301fd
Summary:
**Context/Summary:**
Since LDB manifest dump including printing the LSM shape does not open the db and manifest itself does not have info about Options.num_levels, LDB tool (the only caller of `DumpManifestHandler` has to set a "hopefully-large-enough" level number (i.e,64) to print info of every level for the LSM shape in the manifest. This can mislead whoever that's reading the manifest to believe there are actually 64 levels configured with the CF. This PR clarifies that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13681
Test Plan:
Manual test
`./ldb manifest_dump --hex --verbose --json --path=<some manifest file path>`
```
--------------- Column family "9" (ID 9) --------------
log number: 115873
comparator: leveldb.BytewiseComparator
--- level 0 --- version# 19 ---
--- level 1 --- version# 19 --- compact_cursor: '000000000000000900000000000000DF78787878787878' seq:3418519, type:2 ---
--- level 2 --- version# 19 --- compact_cursor: '000000000000000900000000000000D8' seq:3446619, type:2 ---
--- level 3 --- version# 19 --- compact_cursor: '000000000000000900000000000000DF78787878787878' seq:3418519, type:2 ---
--- level 4 --- version# 19 --- compact_cursor: '000000000000000900000000000000DF78787878787878' seq:3418519, type:2 ---
--- level 5 --- version# 19 --- compact_cursor: '0000000000000009000000000000012B0000000000000065' seq:3447830, type:2 ---
--- level 6 --- version# 19 ---
115931:376281[0 .. 0]['0000000000000000' seq:0, type:1 .. '00000000000003E7000000000000012B00000000000002B1' seq:0, type:1]
--- level 7 --- version# 19 ---
--- level 8 --- version# 19 ---
--- level 9 --- version# 19 ---
--- level 10 --- version# 19 ---
--- level 11 --- version# 19 ---
....
--- level 61 --- version# 19 ---
--- level 62 --- version# 19 ---
--- level 63 --- version# 19 ---
By default, manifest file dump prints LSM trees as if 64 levels were configured, which is not necessarily true for the column family (CF) this manifest is associated with. Please consult other DB files, such as the OPTIONS file, to confirm.
```
Reviewed By: jaykorean
Differential Revision: D76391064
Pulled By: hx235
fbshipit-source-id: 3e1c58e0eeb39a5fa020040201b07b181f8977a6
Summary:
In the function WritableFileWriter::WriteBufferedWithChecksum, since the alignment parameter passed to RequestToken defaults to 4096, when data_size is less than 4096, subtracting a larger value from data_size (which is of type unsigned long) will cause an underflow. This results in an infinite loop. Since WriteBuffered does not require alignment, it is sufficient to pass alignment == 0.
issue:https://github.com/facebook/rocksdb/issues/13640
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13641
Reviewed By: jaykorean
Differential Revision: D76341973
Pulled By: hx235
fbshipit-source-id: 8912f2b6598bb5a48b6b813c53146d9ecfd31d30
Summary:
**Context/Summary:**
`RoundRobinSubcompactionsAgainstResources.SubcompactionsUsingResources` has been flaky and difficult to de-flake. One of the reasons is the complicated usage of sync points and unnecessarily strict verification.
- The sync points don't seem necessary to verify the number of extra reserved threads for sub-compactions so are removed.
- The full reservation after compaction to verify extra reserved threads were release is indirect and hard to get right. So it's replaced with simpler sync-point callback check.
- Since we already have tests (see https://github.com/facebook/rocksdb/blob/7d80ea45442e84c25669db61cb7376ba0cd10ba5/env/env_test.cc#L841 and )for testing pure functionality of reserve/release does reserve/release the threads, verifying the relevant code paths are called should be enough to verify extra reserved threads were released after compaction
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13672
Test Plan: Monitor future flakiness.
Reviewed By: cbi42
Differential Revision: D76108242
Pulled By: hx235
fbshipit-source-id: 30113f16455688f113f296bda0098a66a7a198a3
Summary:
**Summary**
This pull request fixes the issue of having a single file simple_mixed_compressor.h containing both implementation and declaration. To improve code organization and follow best practices, I have separated the implementation into a new file simple_mixed_compressor.cc and updated the original file to only contain the necessary declarations.
**Testing**
Testing was performed by verifying the stdout output from both RoundRobinCompressor and BuiltInCompressorV2.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13665
Reviewed By: pdillinger
Differential Revision: D76060831
Pulled By: shubhajeet
fbshipit-source-id: c034868be51ea7b89c1a8dd12082b0159f49f588
Summary:
Iterator seek returns "SeekAndValidate() not implemented" error if the disallow_memtable_writes CF option is set along with paranoid_memory_checks. The fix is to sanitize the paranoid_memory_checks option to false, which should be safe since the memtable is guaranteed to be empty.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13663
Test Plan: Update unit test in db_basic_test.cc
Reviewed By: pdillinger
Differential Revision: D75973515
Pulled By: anand1976
fbshipit-source-id: 3f381f19dcda72e3b78ee375f755fb4809c6b99c
Summary:
Some tests were failing due to apparent missing include of iomanip. I suspect this was from a gtest upgrade, because in open source, the include iomanip comes from gtest.h. To ensure we maintain compatibility with older gtest as well as the newer one, I pulled the include iomanip out of the in-repo gtest.h. Note that other places in gtest code only instantiate floating-point related templates with `float` and `double` types.
Also, to avoid `make format` being insanely slow on gtest.h, I've excluded third-party from the formatting check.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13661
Test Plan: make check, internal CI, manually ensure formatting check works outside of third-party/
Reviewed By: jaykorean
Differential Revision: D75963897
Pulled By: pdillinger
fbshipit-source-id: ed5737dd456e74068185f1ac5d57046d7509df7a
Summary:
**Summary**
This pull request introduces a mixed compressor, RoundRobinManager and RoundRobinCompressor, which selects algorithms in a loop. This implementation replaces the current hacky approach to round-robin compression in BuiltInCompressorV2. Additionally, it configures RocksDB to optionally utilize this customized compressor in the db stress test.
**Testing**
Testing was performed by verifying the stdout output from both RoundRobinCompressor and BuiltInCompressorV2.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13647
Reviewed By: pdillinger
Differential Revision: D75921997
Pulled By: shubhajeet
fbshipit-source-id: 8f42ac46f08ba982b2cd70241bd7dc13ff5a1225
Summary:
... to support SmallEnumSet over CompressionType with allowed custom compression types using most of the available byte. This is accomplished using an std::array<uint64_t> in place of just uint64_t. Also adds an std::bitset-like count() operation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13657
Test Plan: unit tests included
Reviewed By: hx235
Differential Revision: D75827601
Pulled By: pdillinger
fbshipit-source-id: 519ae97ac671fd9885d6485976abbd969d1392d3
Summary:
**Summary**
This pull request aims to populate num_input_files and total_input_bytes in the CompactionJobStats object, which is accessible through EventListener::OnCompactionBegin(DB*, const CompactionJobInfo&). This change will enable RocksDB users to access accurate compaction input information.
**Context/Goals**
Provide accurate compaction input statistics to RocksDB users
Populate num_input_files and total_input_bytes in CompactionJobStats
Ensure correct population of these fields before EventListener::OnCompactionBegin() is called
**Test Plan**
Added test code to capture num_input_file and total_num_bytes when EventHandler is triggered
Asserted that these values are populated correctly
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13637
Reviewed By: cbi42
Differential Revision: D75690774
Pulled By: shubhajeet
fbshipit-source-id: 8236546f8ce7743f46048b302b376b7ef6429887
Summary:
Somehow this was previously not being tested in our Windows CI jobs so was accidentally broken in https://github.com/facebook/rocksdb/pull/13540 This fix will need to be backported to 10.3.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13649
Test Plan: CI
Reviewed By: hx235
Differential Revision: D75655418
Pulled By: pdillinger
fbshipit-source-id: a56bb213270904a1b7a13b905c2cc1919116df1c
Summary:
Also revamping test
GeneralTableTest::ApproximateOffsetOfCompressed so that it's not sensitive to adding new metadata to SST files
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13646
Test Plan: manually inspect new table property, which is not parsed anywhere, just for information to human reader
Reviewed By: hx235
Differential Revision: D75561241
Pulled By: pdillinger
fbshipit-source-id: c076c01a8b540bc4cb771964d48fa919c4c48ae4
Summary:
to make it easier to use 0 for disabled. And deprecate the use of txn db option `txn_commit_bypass_memtable_threshold`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13636
Test Plan: updated unit tests.
Reviewed By: jowlyzhang
Differential Revision: D75262136
Pulled By: cbi42
fbshipit-source-id: 9040e5a9c918c1d0906a2db4600cc012d2436b22
Summary:
Larger key/values can cause memtable write to take longer time. Add new option `TransactionOptions::large_txn_commit_optimize_byte_threshold` that enables the optimization by transaction write batch size.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13634
Test Plan:
- new unit test
- added option to stress test and ran stress test for some time: `python3 ./tools/db_crashtest.py --txn blackbox --txn_write_policy=0 --commit_bypass_memtable_one_in=50 --test_batches_snapshots=0`
Reviewed By: jowlyzhang
Differential Revision: D75248126
Pulled By: cbi42
fbshipit-source-id: 9522db93457729ba60e4176f7d47f7c2c7778567
Summary:
This exposes CompressionManager and related classes to the public API and adds `ColumnFamilyOptions::compression_manager` for tying a custom compression strategy to a column family. At the moment, this does not support custom/pluggable compression algorithms, just custom strategies around the built-in algorithms, e.g. which compression to use when and where.
A large part of the change is moving code from internal compression.h to a new public header advanced_compression.h, with some minor changes:
* `Decompressor::ExtractUncompressedSize()` is out-of-lined
* CompressionManager inherits Customizable and some related changes to members of CompressionManager are made. (Core functionality of CompressionManager is unchanged.)
This depends on a smart pointer I'm calling `ManagedPtr` which I'm adding to data_structure.h.
Additionally, advanced_compression.h gets CompressorWrapper and CompressionManagerWrapper as building blocks for overriding aspects of compression strategy while leveraging existing compression algorithms / schemas.
Some pieces needed to support the `compression_manager` option and rudimentary Customizable implementation are included. More work will be needed to make this general and well-behaved (see e.g. https://github.com/facebook/rocksdb/issues/8641; I still hit inscrutible problems every time I touch Customizable).
I'll add a release note for the experimental feature once pluggable compression algorithms and more of the Customizable things are working.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13626
Test Plan:
Added a unit test demonstrating how a custom compressor can "bypass" or "reject" compressions.
Expected next follow-up (probably someone else): use a custom CompressionManager/Compressor to replace the internal hack for testing mixed compressions.
Reviewed By: hx235
Differential Revision: D75028850
Pulled By: pdillinger
fbshipit-source-id: 8565bb8ba4b5fa923b1e29e76b4f7bb4faa42381
Summary:
Some specific old versions around RocksDB 2.5 would compress the metaindex and properties blocks. This hasn't been done since, probably because it interferes with the properties block indicating how to set up for decompression (so the reader can read those blocks before doing any decompression).
To fix backward compatibility, we establish a decompressor early if format_version indicates the file could come from a sufficiently old version.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13628
Test Plan: local and CI runs of tools/check_format_compatible.sh. (I don't believe we need special code to set up a unit test for this case.)
Reviewed By: jowlyzhang
Differential Revision: D75107623
Pulled By: pdillinger
fbshipit-source-id: 97132b8c5e0602e8e27254a11386d866b23cb4f5
Summary:
This PR introduces a new CF option, `memtable_avg_op_scan_flush_trigger`, to support triggering a memtable flush when an iterator skips too many invisible keys from the active memtable. This is a follow up to https://github.com/facebook/rocksdb/pull/13523#discussion_r2038261975, which introduced the option `memtable_op_scan_flush_trigger` for a single expensive iterator step. This PR focus on an expensive stretch of iterator steps, between Seeks and until iterator destruction. To avoid triggering a memtable flush for a stretch that is too small, this option only takes effect when the total number of entries skipped from the active memtable in a stretch of iterator steps exceeds `memtable_op_scan_flush_trigger`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13593
Test Plan:
* New unit tests covering the new option
* Add the option to the crash test.
Reviewed By: hx235
Differential Revision: D74434263
Pulled By: cbi42
fbshipit-source-id: 64f1101efb79c7498e2038eff630713ead8f6f41
Summary:
Instead of using FileSystem::GetFileSize() for each CompactionOutputFile, use the file size that is being tracked internally as part of the output file's metadata. FileSize is now part of `CompactionServiceOutputFile` and serialized in the `CompactionServiceResult`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13620
Test Plan:
Tested with logging Meta's internal offload Infra
```
./compaction_job_test
```
Reviewed By: jowlyzhang
Differential Revision: D75006961
Pulled By: jaykorean
fbshipit-source-id: 008f9dc22bd672746ac180380ada4188713a6b85
Summary:
Usual release steps
* Release notes from 10.3 branch
* Update version.h
* Add 10.3.fb to check_format_compatible.sh
* Update folly commit hash. Added a few hacks to fix build errors.
Bonus:
* Add a check_format_compatible.sh sanity check to the per-PR GitHub actions jobs. It should be quick enough and catch typos in release diffs as we've seen in the past.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13622
Test Plan: CI
Reviewed By: jowlyzhang
Differential Revision: D74943843
Pulled By: pdillinger
fbshipit-source-id: 4ff1db9a635e111f8830cadff2d3ee51cf2de512
Summary:
showing up in the crash test after https://github.com/facebook/rocksdb/issues/13540
* For an assertion `dict_samples.sample_data.size() <= opts_.max_dict_bytes` we needed to ensure that `zstd_max_train_bytes` only takes effect with kZSTD compression.
* For an assertion with `r->table_options.verify_compression == (verify_decomp != nullptr)` we needed to ensure that `data_block_verify_decompressor` is set even when dictionary compression is attempted but not used.
* Noticed along the way: finish an optimization in `CompressAndVerifyBlock` that was incomplete.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13621
Test Plan:
Both failures were reproducible with hard-coding of some crash test params, and now not getting a failure.
```
--compression_type=zstd --compression_max_dict_bytes=16384 --compression_zstd_max_train_bytes=65536 --compression_max_dict_buffer_bytes=131071 --compression_use_zstd_dict_trainer=1
```
Write performance test like in https://github.com/facebook/rocksdb/issues/13540 shows essentially no change, maybe slightly faster (+0.4%) with verify_compression.
Reviewed By: virajthakur
Differential Revision: D74939103
Pulled By: pdillinger
fbshipit-source-id: 8bac8891bc08e1356eff52cc524e5bb409b0f86f
Summary:
[internal use] Allow the application to pass a request_id per read request to RocksDB and pass it down to the FileSystem (via IODebugContext)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13616
Test Plan:
./db_test --gtest_filter=DBTest.RequestIdPlumbingTest
Validates that RocksDB Api calls with request_id set result in request_id being passed to the filesystem through IODebugContext
Reviewed By: pdillinger
Differential Revision: D74912824
Pulled By: virajthakur
fbshipit-source-id: 4f15fef3ff7b5d700563f993f9b211c991020fb6
Summary:
Remove the dependency on `allow_db_generated_files` option in `IngestExternalFile` to be set for ingesting external tables. The files are created by SstFileWriter, and we should be able to ingest them. We could make it work by having the external table implementation provide the version and global sequence number related properties, but its safer to have RocksDB generate the table properties block and store it as is in the file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13608
Test Plan: Add unit test to test basic ingestion and ingestion with atomic_replace_range
Reviewed By: pdillinger
Differential Revision: D74830707
Pulled By: anand1976
fbshipit-source-id: 4a9bea4a4f38f7c24c584262095c5c98cd771ddc
Summary:
Add stats to monitor the large transaction optimization. A stat is added for how many times wbwi ingestion is used. A histogram is added to track transaction size. We could also just track write batch size for all writes but I don't want to add the overhead to all writes yet.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13611
Test Plan:
ran `python3 ./tools/db_crashtest.py --txn blackbox --txn_write_policy=0 --commit_bypass_memtable_one_in=50 --test_batches_snapshots=0 --stats_dump_period_sec=2 --dump_malloc_stats=0 --statistics=1` and manually check LOG files
```
rocksdb.number.wbwi.ingest COUNT : 57
...
rocksdb.num.op.per.transaction P50 : 1.000000 P95 : 1.000000 P99 : 1.000000 P100 : 1.000000 COUNT : 2265 SUM : 2265
```
Reviewed By: jowlyzhang
Differential Revision: D74829087
Pulled By: cbi42
fbshipit-source-id: 5a9c3ab2d4cb6071cedfc47201ce2cf65a77d3c6
Summary:
Similar to https://github.com/facebook/rocksdb/pull/13555, add more info, ColumnFamily Id and name, to `CompactionServiceJobInfo`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13615
Test Plan:
Updated Unit Test
```
./compaction_service_test
```
Reviewed By: archang19
Differential Revision: D74845661
Pulled By: jaykorean
fbshipit-source-id: e2fc61006092b9febec1c6637b92cb00fb6cb73e
Summary:
Adds new classes etc. in internal compression.h that are intended to become public APIs for supporting custom/pluggable compression. Some steps remain to allow for pluggable compression and to remove a lot of legacy code (e.g. now called `OLD_CompressData` and `OLD_UncompressData`), but this change refactors the key integration points of SST building and reading and compressed secondary cache over to the new APIs.
Compared with the proposed https://github.com/facebook/rocksdb/issues/7650, this fixes a number of issues including
* Making a clean divide between public and internal APIs (currently just indicated with comments)
* Enough generality that built-in compressions generally fit into the framework rather than needing special treatment
* Avoid exposing obnoxious idioms like `compress_format_version` to the user.
* Enough generality that a compressor mixing algorithms/strategies from other compressors is pretty well supported without an extra schema layer
* Explicit thread-safety contracts (carefully considered)
* Contract details around schema compatibility and extension with code changes (more detail in next PR)
* Customizable "working areas" (e.g. for ZSTD "context")
* Decompression into an arbitrary memory location (rather than involving the decompressor in memory allocation; should facilitate reducing number of objects in block cache)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13540
Test Plan:
This is currently an internal refactor. More testing will come when the new API is migrated to the public API. A test in db_block_cache_test is updated to meaningfully cover a case (cache warming compression dictionary block) that was previously only covered in the crash test.
SST write performance test, like https://github.com/facebook/rocksdb/issues/13583. Compile with CLANG, run before & after simultaneously:
```
SUFFIX=`tty | sed 's|/|_|g'`; for ARGS in "-compression_parallel_threads=1 -compression_type=none" "-compression_parallel_threads=1 -compression_type=snappy" "-compression_parallel_threads=1 -compression_type=zstd" "-compression_parallel_threads=1 -compression_type=zstd -verify_compression=1" "-compression_parallel_threads=1 -compression_type=zstd -compression_max_dict_bytes=8180" "-compression_parallel_threads=4 -compression_type=snappy"; do echo $ARGS; (for I in `seq 1 20`; do ./db_bench -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 $ARGS 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```
Before (this PR and with https://github.com/facebook/rocksdb/issues/13583 reverted):
-compression_parallel_threads=1 -compression_type=none
1908372
-compression_parallel_threads=1 -compression_type=snappy
1926093
-compression_parallel_threads=1 -compression_type=zstd
1208259
-compression_parallel_threads=1 -compression_type=zstd -verify_compression=1
997583
-compression_parallel_threads=1 -compression_type=zstd -compression_max_dict_bytes=8180
934246
-compression_parallel_threads=4 -compression_type=snappy
1644849
After:
-compression_parallel_threads=1 -compression_type=none
1956054 (+2.5%)
-compression_parallel_threads=1 -compression_type=snappy
1911433 (-0.8%)
-compression_parallel_threads=1 -compression_type=zstd
1205668 (-0.3%)
-compression_parallel_threads=1 -compression_type=zstd -verify_compression=1
999263 (+0.2%)
-compression_parallel_threads=1 -compression_type=zstd -compression_max_dict_bytes=8180
934322 (+0.0%)
-compression_parallel_threads=4 -compression_type=snappy
1642519 (-0.2%)
Pretty neutral change(s) overall.
SST read performance test (related to https://github.com/facebook/rocksdb/issues/13583). Set up:
```
for COMP in none snappy zstd; do echo $ARGS; ./db_bench -db=/dev/shm/dbbench-$COMP --benchmarks=fillseq,flush -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 -compression_type=$COMP; done
```
Test (compile with CLANG, run before & after simultaneously):
```
for COMP in none snappy zstd; do echo $COMP; (for I in `seq 1 5`; do ./db_bench -readonly -db=/dev/shm/dbbench-$COMP --benchmarks=readrandom -num=10000000 -duration=20 -threads=8 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```
Before (this PR and with https://github.com/facebook/rocksdb/issues/13583 reverted):
none
1495646
snappy
1172443
zstd
706036
zstd (after constructing with -compression_max_dict_bytes=8180)
656182
After:
none
1494981 (-0.0%)
snappy
1171846 (-0.1%)
zstd
696363 (-1.4%)
zstd (after constructing with -compression_max_dict_bytes=8180)
667585 (+1.7%)
Pretty neutral.
Reviewed By: hx235
Differential Revision: D74626863
Pulled By: pdillinger
fbshipit-source-id: dc8ff3178da9b4eaa7c16aa1bb910c872afaf14a
Summary:
Addresses https://github.com/facebook/rocksdb/issues/13587.
This PR exposes the optimized implementation of batched reads through a `Transaction` object to Java clients.
The latency improvement of transactional multiget on production workload achieved by switching the implementation is roughly:
```
quantile=0.2: 21%
quantile=0.5: 28%
quantile=0.8: 46%
quantile=1.0: 239%
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13589
Reviewed By: jaykorean
Differential Revision: D74660169
Pulled By: cbi42
fbshipit-source-id: d01780173e0500c96e5e431ff6645008cbf6e8b5
Summary:
Expose pinned WriteBatchWithIndex::GetFromBatchAndDB through C bindings so that one can read data from the `WriteBatchWithIndex` and db w/o copying the data.
This fixes https://github.com/facebook/rocksdb/issues/12969.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12970
Reviewed By: cbi42
Differential Revision: D74586418
Pulled By: jaykorean
fbshipit-source-id: a5a4d2e8ce3ddf4c2371fdfdb4e9c3309966a05d
Summary:
As titled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13605
Test Plan: This is removing a test
Reviewed By: mszeszko-meta
Differential Revision: D74660230
Pulled By: jowlyzhang
fbshipit-source-id: 9c1d46b56d2f9ee43eba645563d4f954645d1ace
Summary:
We saw some crash test failure for secondary db. It happens during crash recovery verification. This PR logs the manifest number when such failure happens. This PR also includes a small fix in `TryCatchUpWithPrimary()` that could incorrectly check WAL not found case.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13594
Test Plan: monitor further secondary DB crash test failure.
Reviewed By: archang19
Differential Revision: D74488769
Pulled By: cbi42
fbshipit-source-id: 226e55b2f99a739e93abda3ee91c05b80f59bf6a
Summary:
This PR fixes a bug where the file checksum for an external table file was not being calculated by SstFileWriter. The checksum is calculated in WritableFileWriter, so we need to pass that the the external table builder rather than the FSWritableFile pointer directly. However, WritableFileWriter is private to RocksDB, so wrap it in an FSWritableFile and pass it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13591
Test Plan: Add a new test in table_test.cc
Reviewed By: jaykorean
Differential Revision: D74410563
Pulled By: anand1976
fbshipit-source-id: c7fa8142e20da8836589dee5fa50919951cf4046
Summary:
Prior to this PR, for FIFO kChangeTemperature compaction was done by iterating and reading thru the input sst and generate the output sst. This was wasteful since for FIFO we could apply the "trivial" move by copying the input sst to the out sst without need decompress/compress and reading thru the input sst content at all. This PR added "allow_trivial_copy_when_change_temperature" to the CompactionOptionsFIFO.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13562
Reviewed By: cbi42
Differential Revision: D73295404
Pulled By: mikechuangmeta
fbshipit-source-id: 02241c7389797730ecd4a3b636837cb5f912b424
Summary:
Allow specifying ReadOptions for WBWI iterator when creating it through the C bindings. This allows to specify upper and lower bounds for the created iterator.
This fixes https://github.com/facebook/rocksdb/issues/12963.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12968
Reviewed By: pdillinger
Differential Revision: D74188049
Pulled By: jaykorean
fbshipit-source-id: 970d9910472dfedaa29a800c6d52bec14c656f3c
Summary:
clarify in comments and fix one implementation under NewIterator where option `memtable_op_scan_flush_trigger` does not work correctly with tailing iterator yet. This is because tailing iterator can rebuild iterator internally which reads from a newer memtable, and DBIter's reference to active memtable needs to be refreshed. This PR clarifies that `memtable_op_scan_flush_trigger` will have no effect on tailing iterator. We can add the support in the future if needed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13586
Test Plan: existing tests.
Reviewed By: jaykorean
Differential Revision: D74108099
Pulled By: cbi42
fbshipit-source-id: 7c6608485d57755abc44f3be0b3c5d82a7bc5ca9
Summary:
While working on some compression refactoring, I noticed that `NotifyCollectTableCollectorsOnBlockAdd()` was being called from multiple threads (with `parallel_threads` > 1), meaning we were violating the promise that TablePropertiesCollectors need not be thread safe (and typically will not be, for efficiency).
Fixing this is a bit awkward or intrusive. Even though it seems weird to expose `block_compressed_bytes_fast` and `block_compressed_bytes_fast` in the public `BlockAdd()` function, and NOT the actual compressed block size used, there are some Meta-internal uses that would at least require negotiation / coordination to deprecate and remove. So it's probably easiest to just keep the awkward functionality and do the necessary modifications to call from a single thread.
The simplest solution that preserves the functionality with `parallel_threads` > 1 (provide the sampling data, expected ordering between `BlockAdd()` and `AddUserKey()`, no races) is to do the compression sampling in the thread building uncompressed blocks. Specifically, moving `NotifyCollectTableCollectorsOnBlockAdd()` and the compression sampling from `CompressAndVerifyBlock()`, which is called in parallel, to table builder `Flush()`, which is only called serially (per file). Even though this adds some compression to that single thread when sampling is enabled, that should be tolerable without complicating the code or regressing performance. Some related or nearby optimizations are included to ensure this.
* Got rid of a lot of unnecessary indirection and unnecessary fields in BlockRep, which should be a step in improving parallel compression performance (still bad IMHO).
* Restructured some `if`s etc. to streamline some logic
This satisfies my original refactoring need to moving the sampling code higher up the stack from `CompressBlock()`, to set up some other upcoming refactorings. The other caller of `CompressBlock()` (legacy BlobDB) doesn't need it, and in fact is better off calling `CompressData()` directly because it does not appear to be dealing with the various "no compression" outcomes introduced by `CompressBlock()`.
Eventual follow-up:
* Performance data below shows how the overhead of parallel compression can make it slower, with available CPUs, compared to serial compression. This infrastructure should be re-designed/re-engineered to reduce thread creation, context switches, etc. Also, more of the processing such as checksumming could be parallelized. (Things dependent on the block location in the file, such as ChecksumModifierForContext and cache warming, cannot be parallelized.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13583
ThreadSanitizer: data race /data/users/peterd/rocksdb/./db_stress_tool/db_stress_table_properties_collector.h:36:5 in rocksdb::DbStressTablePropertiesCollector::BlockAdd(unsigned long, unsigned long, unsigned long)
```
Performance:
```
SUFFIX=`tty | sed 's|/|_|g'`; for ARGS in "-compression_parallel_threads=1 -compression_type=none" "-compression_parallel_threads=1 -compression_type=snappy" "-compression_parallel_threads=4 -compression_type=snappy"; do echo $ARGS; (for I in `seq 1 100`; do ./db_bench -db=/dev/shm/dbbench$SUFFIX --benchmarks=fillseq -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -disable_wal -write_buffer_size=12000000 $ARGS 2>&1 | grep micros/op; done) | awk '{n++; sum += $5;} END { print int(sum / n); }'; done
```
Average ops/s of 100 runs, running before & after at the same time, using clang DEBUG_LEVEL=0:
-compression_parallel_threads=1 -compression_type=none
Before: 1976319
After: 1983840 (+0.3%)
-compression_parallel_threads=1 -compression_type=snappy
Before: 1945576
After: 1953473 (+0.4%)
-compression_parallel_threads=4 -compression_type=snappy
Before: 1573190
After: 1611881 (+2.4%)
-compression_parallel_threads=4 -sample_for_compression=100 (pretty high sample rate)
Before: 1577167
After: 1589704 (+0.8%)
-compression_parallel_threads=4 -sample_for_compression=10 (crazy high sample rate)
Before: 1581276
After: 1393453 (-11.9%)
As seen, you need a very very high compression sample rate to see a regression. I would expect a setting like 1000 to be more typical.
Test Plan:
Along with existing unit tests + CI, expanded crash test to make its TablePropertiesCollector non-trivial, to exercise the bug (and other potential bugs), which was confirmed with local run of whitebox_crash_test with TSAN:
```
Reviewed By: hx235
Differential Revision: D73944593
Pulled By: pdillinger
fbshipit-source-id: f1dcba4ebdc01e735251037395003945c9b34e62
Summary:
I added `TransactionDBOptions::txn_commit_bypass_memtable_threshold` previously but per DB option is not dynamically changeable. Adding it as a per transaction option to make it easier to use. The option naming is updated to make it easier for customer to understand `large_txn_commit_optimize_threshold`. The transaction DB option `TransactionDBOptions::txn_commit_bypass_memtable_threshold` is marked as deprecated.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13582
Test Plan:
- new unit test
- updated stress test to use this new transaction option
Reviewed By: jowlyzhang
Differential Revision: D73960981
Pulled By: cbi42
fbshipit-source-id: 406f6e0f5f4eb6b336976f9a93b0bc08e61a9662
Summary:
AcquireLocked() returns transaction ids that currently hold the lock for deadlock detection purpose. We should not include the id of the transaction that is trying to acquire the lock, since this would lead to a false-positive deadlock detection where the deadlock is a self-loop. Note that since `wait_ids` is never cleared, there is another bug where if AcquireLocked() fails with kLockLimit, we could do deadlock detection based on `wait_ids` from a previous lock acquire attempt. This PR fixes both bugs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13575
Test Plan: added a unit test repro that shows deadlock status can be incorrectly returned.
Reviewed By: jaykorean
Differential Revision: D73617887
Pulled By: cbi42
fbshipit-source-id: a6388b3ec53db13e2c502d60199378ea95885841
Summary:
**Context/Summary:**
Similar to https://github.com/facebook/rocksdb/commit/0a43d8a261b9c633c0a4e369b1ef33aa5ee32810, this is to verify flush output file contains the exact number of keys (represented by its `TableProperties::num_entries`) as added to table builder for block-based and plain table format. The implementation reuses a temporary compaction stats to record output record and existing input record (with some refactoring)
**Bonus:**
following https://github.com/facebook/rocksdb/commit/0a43d8a261b9c633c0a4e369b1ef33aa5ee32810#r154313564, limit compaction output record count check within block based table and plain table format as well as removing extra test setting; fix some typo
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13556
Test Plan: New test
Reviewed By: jaykorean
Differential Revision: D73229644
Pulled By: hx235
fbshipit-source-id: 2a7796450048b3bcb2d5c38f2b5fc6b53e4aae37
Summary:
we want to limit the maximum disk space used by RocksDB in one of our Go services, as it runs on a highly disk-constrained network switch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13404
Reviewed By: cbi42
Differential Revision: D73517940
Pulled By: jaykorean
fbshipit-source-id: ae91fc7a4992399e20f06cc67dad8130cf19049e
Summary:
This test was failing sporadically for me, like
```
db/db_properties_test.cc:247: Failure
Expected: (static_cast<double>(dbl_a - dbl_b) / (dbl_a + dbl_b)) <
(bias), actual: 0.113964 vs 0.1
```
I tried waiting for compaction in the test, but that made it fail consistently. Based on inspection of the test and the related test AggregatedTablePropertiesAtLevel already using `disable_auto_compactions = true`, I'm applying that to this test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13568
Test Plan: Parallel runs of the unit test, before and after
Reviewed By: jaykorean
Differential Revision: D73463685
Pulled By: pdillinger
fbshipit-source-id: 84df7cc9bdcd1caa108a7be254ffbebbe9a77de7
Summary:
* Mostly, remove `sample_for_compression` from CompressionInfo because it's not used by the core function it serves, `CompressData()`. Confusing (and inefficient), especially in db_bench where it appears to use `FLAGS_sample_for_compression` in places where it is actually ignored.
* Various clarifying comments, clean-ups, and tiny optimizations
* Prepare some structures like `CompressionDict` for more usage
* Some TODOs and FIXMEs about some things I've noticed are amiss, confusing, or excessive
* A notable optimization opportunity that might become a "pay as you go" improvement for the potential indirection costs of customizable compression: use C++23's resize_and_overwrite() in compress functions to avoid zeroing the string buffer contents before populating it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13539
Test Plan: existing tests / CI
Reviewed By: hx235
Differential Revision: D73451273
Pulled By: pdillinger
fbshipit-source-id: 0373627466d695043d21146ce34d52f189ae9432
Summary:
Updated version, HISTORY and compatibility script for 10.3 release (no folly hash update in this release).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13566
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D73391839
Pulled By: jaykorean
fbshipit-source-id: 075bb1f9f25caf96c4fcca7f4a315666acd5a288
Summary:
Adding an arbitrary options map so that any additional overridable options can be added without RocksDB change. Unknown options will be ignored
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13552
Test Plan:
Unit Test added
```
./db_secondary_test -- --gtest_filter="*OptionsOverrideTest*"
```
Reviewed By: hx235
Differential Revision: D73203789
Pulled By: jaykorean
fbshipit-source-id: 176bd9849d2bc60e78657c119e10a1a2a0988cd1
Summary:
This PR adds a DB::GetNewestUserDefinedTimestamp API to get the newest timestamp of the column family. This is only for when the column family enables user defined timestamp.
It checks the mutable memtable, the immutable memtable and the SST files, and returns the first newest user defined timestamp found. When user defined timestamp is not persisted in SST files, there is metadata in MANIFEST tracking upperbound of flushed timestamps, so the newest timestamp in SST files can be found. If user defined timestamps are
persisted in SST files, currently no timestamp metadata info is persisted. A NotSupported status will be returned if SST files need to be checked in that case.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13547
Test Plan: Added tests
Reviewed By: cbi42
Differential Revision: D73123575
Pulled By: jowlyzhang
fbshipit-source-id: 460ac4f9c96926d3c8fcf7944edab8dc0feae1dd
Summary:
add support for ingesting a WriteBatchWithIndex into the DB with the new API `IngestWriteBatchWithIndex()`. This ingestion works similarly as `TransactionOptions::commit_bypass_memtable` where the WBWI will be ingested as an immutable memtable. Since this skips memtable writes, it improves the write performance when writing a large write batch into the DB. Currently this API only supports `disableWAL=true`. Support for WAL write will be in a follow up if needed.
For a WBWI to be ingestable, we needed to call `SetTrackPerCFStat()` at WBWI creation. This PR removes this step for simpler usage and per CF stats will always be tracked in WBWI. `WBWIIteratorImpl::TestOutOfBound()` is optimized to offset the performance impact.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13550
Test Plan:
- new unit test
- stress test option ingest_wbwi_one_in and ran a few runs of `python3 ./tools/db_crashtest.py blackbox --enable_pipelined_write=0 --use_timed_put_one_in=0 --use_put_entity_one_in=0 --ingest_wbwi_one_in=10 --test_batches_snapshots=0 --enable_blob_files=0 --preserve_unverified_changes=1 --avoid_flush_during_recovery=1 --disable_wal=1 --inplace_update_support=0 --interval=40`
Reviewed By: jowlyzhang
Differential Revision: D73152223
Pulled By: cbi42
fbshipit-source-id: 339f8ed26ac5a798238870df3ba857ba1add759b
Summary:
A reopened writable file's size is not correctly tracked in the `WritableFile`'s internal state. This PR adds a querying to the file system to get the initial file size in the reopen case and use it to populate posix `WritableFile`'s internal state.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13534
Reviewed By: anand1976
Differential Revision: D72756628
Pulled By: jowlyzhang
fbshipit-source-id: 6f02b5c5da069fe49055d7b75bec9e7e47d5cd71
Summary:
Add a test to cover an internal user's expected behavior of using atomic_replace_range feature to atomically ingest a version key and a data file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13549
Test Plan: This is a test
Reviewed By: cbi42
Differential Revision: D73142626
Pulled By: jowlyzhang
fbshipit-source-id: a5bdc24b762cbe91dd4d94242b9e1539c9feaf61
Summary:
add support for atomic_flush when using WBWI ingestion [feature](https://github.com/facebook/rocksdb/blob/29c6610617ddc1b486f12b99c16e7c9851e80430/include/rocksdb/utilities/transaction_db.h#L387). Transaction DB usually uses WAL so atomic_flush is not as helpful. This is to prepare for a follow up PR that enables ingesting WBWI without using transaction DB.
This PR also removes a redundant parameter `prep_log` for the WBWI ingestion feature.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13545
Test Plan:
- unti test added
- stress test will be added as we add support to ingest WBWI without using transaction DB.
Reviewed By: jowlyzhang
Differential Revision: D73062342
Pulled By: cbi42
fbshipit-source-id: e05da55dfabb8241a042214b9d50b1b49d42613e
Summary:
**Context/Summary:**
This PR adds new stats to measure compaction readahead size for rocksdb managed prefetching (not FS prefetching). It can be used to verify compaction read-ahead is doing what's configured. This PR also excludes compaction readahead stats from user scan readahead stats measured in existing stats so there is a cleaner separating between these two.
Bonus: this PR also included some typo fixing about "io activities"
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13520
Test Plan: Modified existing test to verify stats
Reviewed By: archang19
Differential Revision: D72892850
Pulled By: hx235
fbshipit-source-id: 1a73182061baa044c9c9193a2b0fd967ffe75c4a
Summary:
* Clarify in API comments which `log_` options in DBOptions relate to WALs, info log, and/or manifest files.
* Rename a bunch of "log" things to "wal" for clarity, especially in DBImpl. (More to go, especially some more challenging cases like `DBImpl::logs_`, but a step in the right direction IMHO)
* Simplify DBImpl ctor by moving constant initializers to field definitions.
* Use RelaxedAtomic for (renamed) `wals_total_size_`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13490
Test Plan: existing tests
Reviewed By: cbi42
Differential Revision: D71939382
Pulled By: pdillinger
fbshipit-source-id: 852f4737eca83e6ad653010cc197ad1b6e6bae13
Summary:
Introduce a mutable CF option `memtable_op_scan_flush_trigger`. When a DB iterator scans this number of hidden entries (tombstones, overwritten puts) from the active memtable in a Seek() or Next() operation, it marks the memtable to be eligible for flush. Subsequent write operations will schedule the marked memtable for flush.
The main change is small and is in db_iter.cc. Some refactoring is done to consolidate and simplify creation of `ArenaWrappedDBIter` and `DBIter`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13523
Test Plan:
- new unit tests added.
- added `memtable_op_scan_flush_trigger` in crash test
- benchmark:
The following benchmark was done with a previous version of the PR where the option was `memtable_tombstone_scan_limit` and it concerns tombstone only. The results should still be applicable for the case when there's no overwritten puts.
Tests that when memtable has many tombstones, the option helps to improve scan performance:
```
TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=seekrandomwhilewriting --expand_range_tombstones=true --writes_per_range_tombstone=1 --max_num_range_tombstones=10000000 --perf_level=2 --range_tombstone_width=100 --memtable_tombstone_scan_limit=
memtable_tombstone_scan_limit = 10000
seekrandomwhilewriting : 18.527 micros/op 53973 ops/sec 18.527 seconds 1000000 operations; (7348 of 1000000 found)
next_on_memtable_count = 122305248
grep "flush_started" /dev/shm/dbbench/LOG | wc
8 200 2417
memtable_tombstone_scan_limit=200
seekrandomwhilewriting : 4.918 micros/op 203315 ops/sec 4.918 seconds 1000000 operations; (4510 of 1000000 found)
next_on_memtable_count = 1853167
grep "flush_started" /dev/shm/dbbench/LOG | wc
184 4600 54121
When memtable_tombstone_scan_limit=200, more flush is trigged to drop tombstones sooner and improve scan performance.
```
Tests that the new option does not introduce noticeable regression:
```
TEST_TMPDIR=/dev/shm ./db_bench --benchmarks=seekrandomwhilewriting[-X5] --expand_range_tombstones=true --writes_per_range_tombstone=1 --max_num_range_tombstones=10000000 --perf_level=2 --range_tombstone_width=100 --seed=123
Main:
seekrandomwhilewriting [AVG 5 runs] : 46049 (± 4512) ops/sec
PR:
seekrandomwhilewriting [AVG 5 runs] : 46100 (± 4470) ops/sec
The results are noisy with this PR performing better and worse in different runs, with no noticeable regression.
```
Reviewed By: pdillinger
Differential Revision: D72596434
Pulled By: cbi42
fbshipit-source-id: 2d51a0221dc20dac844aeba2ad3999d075a4cf91
Summary:
When ReadOptions.allow_unprepared_value is true, a `Iterator::PrepareValue()` call is needed to prepare the value after an entry is pinpointed, to only load the blob when it's actually needed. And it uses the `saved_key_.GetUserKey()` to prepare value.
https://github.com/facebook/rocksdb/blob/6d802639f7dc35bf765dbe1ed6b3942e4d76375d/db/db_iter.cc#L319
In the reverse iteration case, when the `FindValueForCurrentKeyUsingSeek()` path is used, `saved_key_` is only updated when `ReadOptions.iter_start_ts` is specified. This PR fixes it by updating `saved_key_` for the other case too.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13531
Test Plan: The FIXME test that reproduce the bug is updated
Reviewed By: pdillinger
Differential Revision: D72681397
Pulled By: jowlyzhang
fbshipit-source-id: 6c239da53c9beed1560d30013474f2ba542b245c
Summary:
Fix a reported data race, accessing `manifest_reader_` without locking `mutex_` could race with another `DBImpl::Secondary::TryCatchUpWithPrimary` thread that is updating to a new manifest in `ReactiveVersionSet::MaybeSwitchManifest`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13529
Test Plan: Existing tests
Reviewed By: hx235
Differential Revision: D72655645
Pulled By: jowlyzhang
fbshipit-source-id: 08599862346bb39a6872c3adfd7f0097fc633849
Summary:
As titled. This option has been marked deprecated since introduction of a better option `max_write_buffer_size_to_maintain` and acts as its fallback since RocksDB 6.5.0 The internal user we know these options were created for migrated to `max_write_buffer_size_to_maintain` for a long time too.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13491
Test Plan: existing tests
Reviewed By: cbi42
Differential Revision: D71984601
Pulled By: jowlyzhang
fbshipit-source-id: c264d4809e311f60fdbad817ebfade256db549b6
Summary:
**Context/Summary:**
Rebased on https://github.com/facebook/rocksdb/pull/13522/files, this is to use the refactored function to calculate tail size from table property "tail_start_offset"
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13524
Test Plan: CI
Reviewed By: anand1976
Differential Revision: D72576262
Pulled By: hx235
fbshipit-source-id: 78c126bc64024c2341d183d6871e06d55fd27501
Summary:
**Context/Summary:**
This is to fix a bug that tail size of remote compaction output SST file is not persisted to manifest in primary instance. This prevent us from using direct tail prefetch optimization each time opening this SST file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13522
Test Plan: Modify existing UT that failed before the fix
Reviewed By: anand1976
Differential Revision: D72479612
Pulled By: hx235
fbshipit-source-id: 1ba8aa66fac71b9196589f60076229c29a103706
Summary:
Public APIs like `DB::GetFullHistoryTsLow` and `DB::IncreaseFullHistoryTsLow` have such safeguarding, allowing them to only be invoked when user defined timestamp is enabled. This PR adds safeguarding into related internal APIs in `ColumnFamilyData` to properly handle the case when the UDT feature are toggled.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13521
Test Plan: ./db_with_timestamp_basic_test --gtest_filter="*EnableDisableUDT*"
Reviewed By: cbi42
Differential Revision: D72475234
Pulled By: jowlyzhang
fbshipit-source-id: 194c07287e3100da95450b04c76552c9d4a86c2d
Summary:
A multi scan API for users to pass a set of scan ranges and have the table readers determine the optimal strategy for performing the scans. This might include coalescing of IOs across scans, for example. The requested scans should be in increasing key order. The scan start keys and other info is passed to NewMultiScanIterator, which in turn uses the newly added Prepare() interface in Iterator to update the iterator. The Prepare() takes a vector of ScanOptions, which contain the start keys and optional upper bounds, as well as user defined parameters in the property_bag taht are passed through as is to external table readers.
The initial implementation plumbs this through to the ExternalTableReader. This PR also fixes an issue of premature destruction of the external table iterator after the first scan of the multi-scan. The `LevelIterator` treats an invalid iterator as a potential end of file and destroys the table iterator in order to move to the next file. To prevent that, this PR defines the `NextAndGetResult` interface that the external table iterator must implement. The result returned by `NextAndGetResult` differentiates between iterator invalidation due to out of bound vs end of file.
Eventually, I envision the `MultiScanIterator` to be built on top of a producer-consumer queue like container, with RocksDB (producer) enqueueing keys and values into the container and the application (consumer) dequeueing them. Unlike a traditional producer consumer queue, there is no concurrency here. The results will be buffered in the container, and when the buffer is empty a new batch will be read from the child iterators. This will allow the virtual function call overhead to be amortized over many entries.
TODO (in future PRs):
1. Update the internal implementation of Prepare to trim the ScanOptions range based on the intersection with the table key range, taking into consideration unbounded scans and opaque user defined bounds.
2. Long term, take advantage of Prepare in BlockBasedTableIterator, atleast for the upper bound case.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13473
Reviewed By: pdillinger
Differential Revision: D71447559
Pulled By: anand1976
fbshipit-source-id: 31668abb0c529aa1ac1738ae46c36cbddf9148f1
Summary:
**Context/Summary:**
- This is an attempt to fix our [build-window-vs2022 failure](https://github.com/facebook/rocksdb/actions/runs/14215681026/job/39831770554?fbclid=IwZXh0bgNhZW0CMTAAAR2BQLjp8kC1u1yyvN1_S5qwmrHEZOfzxJdcbj2vq7mvwwq83n1cbkmiBCA_aem_ygYxQA5EUmxh2y4EjMlTfg) below. snappy-1.1.8's cmake_minimum_required being less than 3.5 seems to trigger the complaint. Hopefully downloading the 1.2.2 which is the [first version starting to use higher cmake_minimum_required version](https://github.com/google/snappy/releases/tag/1.2.2) solves the failure.
```
Directory: D:\a\rocksdb\rocksdb\thirdparty\snappy-1.1.8
Mode LastWriteTime Length Name
---- ------------- ------ ----
d---- 4/2/2025 9:02 AM build
CMake Error at CMakeLists.txt:29 (cmake_minimum_required):
Compatibility with CMake < 3.5 has been removed from CMake.
Update the VERSION argument <min> value. Or, use the <min>...<max> syntax
to tell CMake that the project requires at least <min> but has been updated
to work with policies introduced by <max> or earlier.
Or, add -DCMAKE_POLICY_VERSION_MINIMUM=3.5 to try configuring anyway.
```
- The downloaded snappy do not include the content under nested repos Google Test and Google Benchmark. But snappy cmake by default will attempt to build them. Since we don't change snappy, we don't need building such development suit. This PR also disabled snappy cmake's attempt to build them.
- By running above changes, the same build [complained](https://github.com/facebook/rocksdb/actions/runs/14228883966/job/39874927730?pr=13514) about java cmakelists requiring too low cmake_minimum_required as well. So this PR also upgraded its cmake_minimum_required to be 3.11 aligning with its warning message
```
if(${CMAKE_VERSION} VERSION_LESS "3.11.4")
message("Please consider switching to CMake 3.11.4 or newer")
endif()
```
**Test plan**
Monitor build-window-vs2022 for this PR
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13514
Reviewed By: pdillinger
Differential Revision: D72333581
Pulled By: hx235
fbshipit-source-id: 1a9096738d39c8b1d270fe17fbd78c1ea4c4c45e
Summary:
Essentially fix https://github.com/facebook/rocksdb/issues/13429 by
* Avoiding publishing to readers a partial write batch written to memtable. Also clarify in DB::Write that WriteBatch is applied atomically, and improve some logging.
* When we know we have written a bad write batch to WAL due to memtable insert failure, make a good effort to roll it back to make the DB recoverable. (Not compatible with all options.)
Fixes https://github.com/facebook/rocksdb/issues/13429
Follow-up items:
* More rigorously test and fix the code paths and option combinations where these features could be useful.
* Allow default CF with disallow_memtable_writes (with caveat that violation stops writes on your open DB)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13489
Test Plan: Updated existing test, manually verified the DB went into a "stopped" state at least in this example.
Reviewed By: jaykorean
Differential Revision: D71917670
Pulled By: pdillinger
fbshipit-source-id: c9b9dfc102817fc4c160a6c7170c04011c228aaf
Summary:
We are seeing some occasional failures with WRITE_(UN)PREPARED crash test runs, and it's alarming when these are grouped in with WRITE_COMMITTED, which AFAIK is the only one considered mature and mission-critical at this point.
* Mark WRITE_(UN)PREPARED as EXPERIMENTAL in the public APIs
* Separate out the `_with_txn` crash test jobs by write policy, now `_with_wc_txn`, `_with_wp_txn` and `_with_wup_txn` so that the major functional and maturity differences are better grouped.
* Add `_with_multiops_wup_txn` which was apparently missing
* Clean up db_crashtest.py for better consistency
* Get rid of awkard "write_policy" parameter that could conflict with authoritative "txn_write_policy" parameter.
* Similarly, move some multiops logic from different parameter sets to finalize_and_sanitize logic.
Immediate internal follow-up:
* Migrate from `_with_txn` which are now deprecated aliases of `_with_wc_txn` to more jobs with the new variants. And likely also add new multiops job.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13499
Test Plan: manual runs of modified jobs, at least long enough to spot check things like txn_write_policy
Reviewed By: hx235
Differential Revision: D72015307
Pulled By: pdillinger
fbshipit-source-id: 06b99b2d1f15ac76fe7b8e22c93a51aaa2a42ecf
Summary:
**Context/Summary:**
MaxMemCompactionLevel() developed 10 years ago simply returns the level a memtable flushed to, which has historically been L0 and have no plan to change to something different for future. It is also not used in test or internally.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13503
Test Plan: CI + fake release
Reviewed By: cbi42
Differential Revision: D72066092
Pulled By: hx235
fbshipit-source-id: 5ff5b16a6664ef3efabd3a6fbd8a2d0529b62460
Summary:
based on the option comment, `ignore_range_deletions` was added due to the overhead of range deletions in read path when a DB does not use DeleteRange(). The current implementation should not have a noticeable performance difference in this case.
`experimental::PromoteL0()` can be replaced by doing a manual compaction with proper CompactRangeOptions.
There are some internal use of these option and API so we will remove them later after the usages are updated.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13500
Test Plan:
comment change only.
Performance: benchmark the performance difference with `ignore_range_deletions` and without (borrowed flag `universal_incremental` for this purpose), ran at the same time on the same machine.
- random point get:
- ignore_range_deletions=false: 343078 ops/sec
- ignore_range_deletions=true: 340219 ops/sec (0.8% slower)
```
(for I in $(seq 1 1); do TEST_TMPDIR=/dev/shm/t1 /data/users/changyubi/vscode-root/rocksdb/db_bench --benchmarks=fillseq,waitforcompaction,readrandom --write_buffer_size=67108864 --writes=1000000 --num=2000000 --reads=1000000 --seed=1723056275 --universal_incremental=false 2>&1 | grep "readrandom"; done;) | awk '{ t += $5; c++; print } END { print 1.0 * t / c }';
```
- sequential scan:
- ignore_range_deletions=false: 5378104 ops/sec
- ignore_range_deletions=true: 5393809 ops/sec (0.3% faster)
```
(for I in $(seq 1 10); do TEST_TMPDIR=/dev/shm/t1 /data/users/changyubi/vscode-root/rocksdb/db_bench --benchmarks=fillseq,waitforcompaction,readseq[-X10] --write_buffer_size=67108864 --writes=1000000 --num=2000000 --universal_incremental=true --seed=1723056275 2>1 | grep "\[AVG 10 runs\]"; done;) | awk '{ t += $6; c++; print; } END { printf "%.0f\n", 1.0 * t / c }';
```
The difference in ops/sec for the two benchmarks is likely noise.
Reviewed By: hx235
Differential Revision: D72069223
Pulled By: cbi42
fbshipit-source-id: ad82a051aa4682790d2178cd4fb2d1467397fbb5
Summary:
Acquiring a lock here can take a long time and cause a user mode scheduler to hold up, as it relies on explicit yielding. Hence, forcing a check here but ignoring any abort requests. Would rely on upstream to take action on aborts.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13498
Reviewed By: pdillinger
Differential Revision: D71987173
Pulled By: jainpr
fbshipit-source-id: 4aec40bdf0bc657e29f72c306c576b3117f97a25
Summary:
Based on passing address of uninit variable in ReadOnlyMemTable::Get() in memtable.h. The contract and other implementations suggest it is a pure out parameter that is always overwritten, so we initialize it in the function before checking its value in a loop
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13486
Test Plan: watch build-linux-valgrind in CI
Reviewed By: cbi42
Differential Revision: D71819843
Pulled By: pdillinger
fbshipit-source-id: 1e06f3ee6998099791af27de5b2872eb476ceb7c
Summary:
The new API in https://github.com/facebook/rocksdb/issues/13453 is awkward and precarious because of using RangePtr, which encodes optional keys using raw pointers to Slice. We could use `std::optional<Slice>` instead but that is unsatisfyingly a larger object with an inefficient size (typically 17 bytes).
Here I introduce a custom optional Slice type, `OptSlice`, that is the same size as a Slice, and use it in a number of places to clean up code and make some public APIs easier to work with. This includes
* `atomic_replace_range` (not yet released, OK to change)
* `GetAllKeyVersions()` which gets a behavior change because of its unusual handling of empty keys.
* `DeleteFilesInRanges()`
* TODO in follow-up: `CompactRange()`
Most of the diff is associated updates and refactorings. Also
* Move some relevant things out of db.h to keep it as tidy as possible.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13481
Test Plan: tests updated
Reviewed By: hx235
Differential Revision: D71747774
Pulled By: pdillinger
fbshipit-source-id: b4c8519608d119b8bceca9bb0fd778608f62a141
Summary:
As titled. This API returns the table properties of files per level. It can be handy for use cases that needed file's leveling info while retrieving TableProperties. We will use this API to later aggregate per level data write time info.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13469
Test Plan: Added unit tests
Reviewed By: pdillinger
Differential Revision: D71353096
Pulled By: jowlyzhang
fbshipit-source-id: dc1fbb2c97e4365fc8d7241f9a59c65fbf4fb766
Summary:
This PR adds a new field `CompactionJobStats.num_input_files_trivially_moved` representing the number of files this compaction trivially moved. It should either equal to the total number of input files, or being 0.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13479
Test Plan: Added tests
Reviewed By: hx235
Differential Revision: D71638796
Pulled By: jowlyzhang
fbshipit-source-id: 794c085408a0dc95f11874ca60fca3e6b5b92cba
Summary:
Adding a new option (argument) for file ingestion `atomic_replace_range` which is intended to support a couple forms of "atomic replacement of a key range":
* (Experimental implementation here) With snapshot_consistency=false, the feature acts like an atomic DeleteFilesInRange prior to the ingestion, though requires no existing files to partially overlap the range. (Consider using SstPartitioner.) This is especially useful for "always compacted" workloads, perhaps along with CF option `disallow_memtable_writes` and ingestion option `fail_if_not_bottommost_level`. If both bounds are nullptr, the whole CF is replaced.
* (To implement in follow-up) With snapshot_consistency=true (and perhaps in some fallback cases from above such as partial overlap), a "giant tombstone file" as in https://github.com/facebook/rocksdb/issues/13078 is generated and ingested at the beginning of the list.
Because I see this as a more elaborate DeleteRange, I would naturally expect the upper bound/limit key to be exclusive, but it has been challenging getting that to work. The inclusive/exclusive handling is currently a documented bug for the experimental feature to sort out in follow-up work. (I would love to take advantage of proposed SliceBound, but that would be ambitious to adapt to DeleteRange. Even getting the "replace whole CF" variant of the functionality might be difficult to get worthing with DeleteRange underneath. Nevertheless, I feel it's best to consolidate these two forms of "atomic replacement" under variants of the same API.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13453
Test Plan:
Unit tests added / updated.
db_stress integration left as follow-up work (experimental feature, will be challenging)
Reviewed By: anand1976
Differential Revision: D71584295
Pulled By: pdillinger
fbshipit-source-id: 307abff426e4b7d0a340008918ebcddc896ef747
Summary:
This PR is a followup to https://github.com/facebook/rocksdb/pull/13461. We're introducing an experimental option / killswitch to control SST write lifetime hint calculation based on the selected compaction style. By default (and mostly for backwards compatibility reasons), we'll calculate the SST hints only for level compactions. With this change users have an option to configure SST lifetime hint policy in their environments to enable the calculations in the universal compaction mode as well. It's important to underline that as currently implemented, SST write lifetime hints are calculated in a static way and solely based on the level, which might not be suitable for non-uniform workloads with dynamic / high-variance lifespan of data within the same level. In those cases (or when the performance is not satisfactory), it's recommended to disable the hints by setting the set to empty. Please see the comment in `options.h` for more.
**NOTE:** We deliberately decided to introduce a new option to ensure no impact to external users running their RocksDB instances on local flash with the default `PosixWritableFile` file implementation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13472
Reviewed By: pdillinger, anand1976
Differential Revision: D71445488
Pulled By: mszeszko-meta
fbshipit-source-id: 57dc5e56662fa0b0fd686e183c0ec7090ff12d66
Summary:
## Issue
Thanks to PRs https://github.com/facebook/rocksdb/issues/13455 and https://github.com/facebook/rocksdb/issues/13464 , we were able to find another issue with compaction stats.
When there are multiple sub-compactions and they are processed remotely, some compaction stats are not collected correctly.
Here's an example of how `num_input_records` can be double-counted during a compaction with multiple sub-compactions executed remotely. Please note that this problem is not limited to `num_input_records`.
Input File: 1 SST file with 100 keys.
- Key 1~50 are in one sub compaction
- Key 51~100 in another sub compaction
`UpdateOutputLevelCompactionStats()` currently retrieves the total number of entries from the input files and sets `num_input_records` in the internal_stats to 100. In `CompactionJob::Run()`, this method is called once after all sub-compactions have finished. However, during remote compaction, `UpdateOutputLevelCompactionStats()` is called for each offloaded sub-compaction on the remote side and then aggregated on the primary host. The internal_stats for the first sub-compaction will have 100 `num_input_records`, and the second sub-compaction will have another 100 `num_input_records`. We end up having 200 `num_input_records` in the aggregated internal_stats.
There was another issue that `num_input_record` was not properly excluding `num_input_range_del` in `UpdateCompactionJobStats()`. `job_stats_->num_input_record` originally has correct value set by compaction iterator, but then later overwritten in `UpdateCompactionJobStats()`. `UpdateCompactionJobStats()` was called during `CompactionJob::Install()`, so not caught by `VerifyInputRecordCount()`.
## Refactor and other changes before the fixes
* Renamed `UpdateOutputLevelCompactionStats()` to `BuildStatsFromInputTableProperties()` to make the function more descriptive. `BuildStatsFromInputTableProperties()` builds input stats by scanning through entries from TableProperties in the Input Files and it's at the top compaction level, not at the sub-compaction level. (It also updates a couple of non-input stats, `bytes_read_blob` and `num_dropped_records`, but will be refactored in a later PR.)
* `UpdateCompactionJobStats()` was moved from `CompactionJob::Install()` to `CompactionJob::Run()` and separated into `UpdateCompactionJobInputStats()` and `UpdateCompactionJobOutputStats()`.
## Fixes
* Remote Compaction no longer updates the subcompaction-job-level input stats from InputTableProperties to avoid double-counted stats in case of multiple sub-compactions. Subcompaction-job-level input stats are aggregated to the compaction-job-level input stats in the primary host after all sub-compactions are finished.
* Remote Compaction now only calls `UpdateCompactionJobOutputStats()` to update the job-level output stats by copying from internal stats.
* `UpdateCompactionJobInputStats()` now takes `num_input_range_del` and properly subtracts it from the input record count. `VerifyInputRecordCount()` expected `job_stats.num_input_records` to be equal to `internal_stats_.output_level_stats.num_input_records - num_input_range_del`. However, when updating the job-level stats, we were taking the entire `internal_stats_.output_level_stats.num_input_records` after verification.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13470
Test Plan:
Local Compaction
```
./db_compaction_test -- --gtest_filter="*DBCompactionTest.VerifyRecordCount*"
```
Remote Compaction
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.VerifyInputRecordCount*"
```
Reviewed By: pdillinger
Differential Revision: D71566149
Pulled By: jaykorean
fbshipit-source-id: c8aafcde701dec8901fd5e5a9ec186e26b896c19
Summary:
Continuing cbi42 's work in 602cc0f9a4be89020fb870dba2816f11dd515d16.
In this PR, we are adding record count verification for each compaction by comparing number of entries summed from Table Properties with the number of output records from the compaction stats.
If the count does not match, `Status::Corruption(msg)` is returned with detailed message including the actual number (from table property) and the expected number (from compaction stats)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13455
Test Plan:
New UT added
```
./db_compaction_test -- --gtest_filter="*Verify*"
```
The check had to be disabled for some of the existing tests using MockTable/MockTableFactory, because TableProperties aren't populated properly for the MockTables.
Reviewed By: hx235
Differential Revision: D71235790
Pulled By: jaykorean
fbshipit-source-id: 3a86a878d13e79d948409d6a9843d1c992d2c98e
Summary:
## Background
Compaction statistics are collected at various levels across different classes and structs.
* `InternalStats::CompactionStats`: Per-level Compaction Stats within a job (can be at subcompaction level which later get aggregated to the compaction level)
* `InternalStats::CompactionStatsFull`: Contains two per-level compaction stats - `output_level_stats` for primary output level stats and `proximal_level_stats` for proximal level stats. Proximal level statistics are only relevant when using Tiered Storage with the per-key placement feature enabled.
* `InternalStats::CompactionOutputsStats`: Simplified version of `InternalStats::CompactionStats`. Only has a subset of fields from `InternalStats::CompactionStats`
* `CompactionJobStats`: Job-level Compaction Stats. (can be at subcompaction level which later get aggregated to the compaction level)
Please note that some fields in Job-level stats are not in Per-level stats and they don't map 1-to-1 today.
## Issues
* In non-remote compactions, proximal level compaction statistics were not being aggregated into job-level statistics. Job level statistics were missing stats for proximal level for tiered storage compactions with per-key-replacement feature enabled.
* During remote compactions, proximal level compaction statistics were pre-aggregated into job-level statistics on the remote side. However, per-level compaction statistics were not part of the serialized compaction result, so that primary host lost that information and weren't able to populate `per_key_placement_comp_stats_` and `internal_stats_.proximal_level_stats` properly during the installation.
* `TieredCompactionTest` was only checking if (expected stats > 0 && actual stats > 0) instead actual value comparison
## Fixes
* Renamed `compaction_stats_` to `internal_stats_` for `InternalStats::CompactionStatsFull` in `CompactionJob` for better readability
* Removed the usage of `InternalStats::CompactionOutputsStats` and consolidated them to `InternalStats::CompactionStats`.
* Remote Compactions now include the internal stats in the serialized `CompactionServiceResult`. `output_level_stats` and `proximal_level_stats` get later propagated in sub_compact output stats accordingly.
* `CompactionJob::UpdateCompactionJobStats()` now takes `CompactionStatsFull` and aggregates the `proximal_level_stats` as well
* `TieredCompactionTest` is now doing the actual value comparisons for input/output file counts and record counts. Follow up is needed to do the same for the bytes read / written.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13464
Test Plan:
Unit Tests updated to verify stats
```
./compaction_service_test
```
```
./tiered_compaction_test
```
Reviewed By: pdillinger
Differential Revision: D71220393
Pulled By: jaykorean
fbshipit-source-id: ad70bffd9614ced683f90c7570a17def9b5c8f3f
Summary:
This PR adds a check for an invariant of sequence number during recovery, that it should not be set backward. This is inspired by a recent SEV that is caused by a software bug. It is a relatively cheap and straightforward check that RocksDB can do to avoid silently opening the DB in a corrupted state.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13465
Test Plan:
Existing tests should cover the case when the invariant is met
The corrupted state is manually tested using aforementioned bug.
Reviewed By: hx235
Differential Revision: D71226513
Pulled By: jowlyzhang
fbshipit-source-id: cd8056fa6653d44ceeb9ba9b4693ab0660a53b4e
Summary:
**Context/Summary:**
For users who are interested in knowing how efficient their compaction in reducing L0 files or how bad their long-running compaction in "locking" L0 files, they now have a reference point "L0 files in the CF pre compaction" for their input compaction files.
- Compared to the existing stats or exposing in some other way, exposing this info in CompactionJobInfo allows users to compare it with other compaction data (e.g, compaction input num, compaction reason) of within **one** compaction (of per-compaction granularity).
- If this number is high while their "short-running" compaction has little L0 files input, then those compaction may have a room for improvement. Similar for those long-running compaction. This PR is to add a new field `CompactionJobInfo::num_l0_files_pre_compaction` for that.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13462
Test Plan: - Piggyback on an existing test
Reviewed By: jaykorean
Differential Revision: D71124938
Pulled By: hx235
fbshipit-source-id: aa47c9c86c62d9425771b320f5636e50671fd289
Summary:
The original implementation of NVMe write lifetime hints (https://github.com/facebook/rocksdb/pull/3095) assumed a flexible interface which decouples file creation from the explicit act of setting write lifetime hint (see `PosixWritableFile` for more context). However, there are existing file systems implementations (ex. Warm Storage) that require all the options (including file write lifetime hints) to be specified once at the time of the actual `FSWritableFile` object instantiation. We're extending the `FileOptions` with `Env::WriteLifeTimeHint` and patch existing callsites accordingly to enable one-shot metadata setup for those more constraint implementations.
NOTE: Today `CalculateSSTWriteHint` only sets write lifetime hint for Level compactions. We'll fill that gap in following PRs and add calculation for Universal Compactions which would unblock Zippy's use case.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13461
Reviewed By: anand1976
Differential Revision: D71144645
Pulled By: mszeszko-meta
fbshipit-source-id: 6c09b62a360d48bd6e4fb08a1265bce2a49f3f4a
Summary:
In hopes of eventually removing some ugly and awkard code for compress_format_version < 2, users can no longer write files in that format and its read support is marked deprecated. For continuing to test that read support, there is a back door to writing the files in unit tests.
If format_version < 2 is specified, it is quietly sanitized to 2. (This is similar to other BlockBasedTableOptions.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13463
Test Plan: unit tests updated.
Reviewed By: hx235
Differential Revision: D71152916
Pulled By: pdillinger
fbshipit-source-id: 95be55e86f93f09fd898223578b9381385c3ccd8
Summary:
With generalized age-based tiering (work-in-progress), the "warm tier" data will no longer necessarily be placed in the second-to-last level (also known as the "penultimate level").
Also, the cold tier may no longer necessarily be at the last level, so we need to rename options like `preclude_last_level_seconds` to `preclude_cold_tier_seconds`, but renaming options is trickier because it can be a breaking change for consuming applications. We will do this later as a follow up.
**Minor fix included**: Fixed one `use-after-move` in CompactionPicker
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13460
Test Plan: CI
Reviewed By: pdillinger
Differential Revision: D71059486
Pulled By: jaykorean
fbshipit-source-id: fd360cdf719e015bf9f9e3f6f1663438226566a4
Summary:
This PR adds support for PerKeyPlacement in Remote Compaction.
The `seqno_to_time_mapping` is already available from the table properties of the input files. `preserve_internal_time_seconds` and `preclude_last_level_data_seconds` are directly read from the OPTIONS file upon db open in the remote worker. The necessary changes include:
- Add `is_penultimate_level_output` and `file_temperature` to the `CompactionServiceOutputFile`
- When building the output for the remote compaction, get the outputs for penultimate level and last level separately, serialize them with the two additional information added in this PR.
- When deserializing the result from the primary, SubcompactionState's `GetOutputs()` now takes `is_penultimate_level`. This allows us to determine which level to place the output file.
- Include stats from `compaction_stats.penultimate_level_stats` in the remote compaction result
# To Follow up
- Stats to be fixed. Stats are not being populated correctly for PerKeyPlacement even for non-remote compactions.
- Clean up / Reconcile the "penultimate" naming by replacing with "proximal"
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13459
Test Plan:
Updated the unit test
```
./compaction_service_test
```
Reviewed By: pdillinger
Differential Revision: D71007211
Pulled By: jaykorean
fbshipit-source-id: f926e56df17239875d849d46b8b940f8cd5f1825
Summary:
[Experiment]
This PR is a followup to https://github.com/facebook/rocksdb/pull/13408. Thick bandaid of ignoring all injected read errors in context of periodic iterator auto refreshes in db stress proved to be effective. We confirmed our theory that errors are not a really a consequence / defect related to this new feature but rather due to subtle ways in which downstream code paths handle their respective IO failures. In this change we're replacing a thick 'ignore all IO read errors' bandaid in `no_batched_ops_stress` with a much smaller, targeted patches in obsolete files purge / delete codepaths, table block cache reader, table cache lookup to make sure we don't miss signal and ensure there's a single mechanism for ignoring error injection in db stress tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13447
Reviewed By: hx235
Differential Revision: D70794787
Pulled By: mszeszko-meta
fbshipit-source-id: c5fcd4780d82357c407f53bf0bb22fc38f7bd277
Summary:
Add debug logging when the Wait() does not return `kSuccess` so that we can compare the version state that was printed by the logging added in https://github.com/facebook/rocksdb/issues/13427 upon InputFileCheck failure.
# Test Plan
CI + Tested with Temporary Change in Meta Internal Infra
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13452
Reviewed By: hx235
Differential Revision: D70898963
Pulled By: jaykorean
fbshipit-source-id: d591b82f2df173b5e01f6552230844ce95155256
Summary:
`nullptr` is preferable to `0` or `NULL`. Let's use it everywhere so we can enable `-Wzero-as-null-pointer-constant`.
- If you approve of this diff, please use the "Accept & Ship" button :-)
Reviewed By: dtolnay
Differential Revision: D70818166
fbshipit-source-id: 4658fb004676fe2686249fdd8ecb322dec8aa63d
Summary:
Primarily, fix an issue from https://github.com/facebook/rocksdb/issues/13316 with opening secondary DB with preserve/preclude option (crash test disable in https://github.com/facebook/rocksdb/issues/13439). The issue comes down to mixed-up interpretations of "read_only" which should now be resolved. I've introduced the stronger notion of "unchanging" which means the VersionSet never sees any changes to the LSM tree, and the weaker notion of "read_only" which means LSM tree changes are not written through this VersionSet/etc. but can pick up externally written changes. In particular, ManifestTailer should use read_only=true (along with unchanging=false) for proper handling of preserve/preclude options.
A new assertion in VersionSet::CreateColumnFamily to help ensure sane usage of the two boolean flags is incompatible with the known wart of allowing CreateColumnFamily on a read-only DB. So to keep that assertion, I have fixed that issue by disallowing it. And this in turn required downstream clean-up in ldb, where I cleaned up some call sites as well.
Also, rename SanitizeOptions for ColumnFamilyOptions to SanitizeCfOptions, for ease of search etc.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13441
Test Plan:
* Added preserve option to a test in db_secondary_test, which reproduced the failure seen in the crash test.
* Revert https://github.com/facebook/rocksdb/issues/13439 to re-enable crash test functionality
* Update some tests to deal with disallowing CF creation on read-only DB
* Add some testing around read-only DBs and CreateColumnFamily(ies)
* Resurrect a nearby test for read-only DB to be sure it doesn't write to the DB dir. New EnforcedReadOnlyReopen should probably be used in more places but didn't want to attempt a big migration here and now. (Suggested follow-up.)
Reviewed By: jowlyzhang
Differential Revision: D70808033
Pulled By: pdillinger
fbshipit-source-id: 486b4e9f9c9045150a0ebb9cb302753d03932a3f
Summary:
In case the primary host has a new option added which isn't available in the remote worker yet, the remote compaction currently fails. In most cases, these new options are not relevant to the remote compaction and the worker should be able to move on by ignoring it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13443
Test Plan: Verified internally in Meta Infra.
Reviewed By: anand1976
Differential Revision: D70744359
Pulled By: jaykorean
fbshipit-source-id: eb6a388c2358a7f8089f2e35a378b7017b9e03f3
Summary:
If compaction job needs to be aborted inside `Schedule()` or `Wait()` today (e.g. Primary host is shutting down), the only two options are the following
- Handle it as failure by returning `CompactionServiceJobStatus::kFailure`
- Return `CompactionServiceJobStatus::kUseLocal` and let the compaction move on locally and eventually succeed or fail depending on the timing
In this PR, we are introducing a new status, `CompactionServiceJobStatus::kAborted`, so that the implementation of `Schedule()` and `Wait()` can return it. Just like how `CompactionServiceJobStatus::kFailure` is handled, compaction will not move on and fail, but the status will be returned as `Status::Aborted()` instead of `Status::Incomplete()`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13438
Test Plan:
Unit Test added
```
./compaction_service_test --gtest_filter="*CompactionServiceTest.AbortedWhileWait*"
```
Reviewed By: anand1976, hx235
Differential Revision: D70655355
Pulled By: jaykorean
fbshipit-source-id: 22614ce9c7455cda649b15465625edc93978fe11
Summary:
I have a place I want to use this helper method inside the Sally codebase. I have this functionality in my Sally diff right now, but I think it is generic enough to warrant putting alongside `Env::PriorityToString`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13440
Test Plan: Just the compiler and CI checks are sufficient IMO.
Reviewed By: hx235
Differential Revision: D70664597
Pulled By: archang19
fbshipit-source-id: 341de6c6e311a3f421ad093c2c216e5caa5034dd
Summary:
This PR adds the ability to use an ExternalTableBuilder through the SstFileWriter to create external tables. This is a counterpart to https://github.com/facebook/rocksdb/issues/13401 , which adds the ExternalTableReader. The support for external tables is confined to ingestion only DBs, with external table files ingested into the bottommost level only. https://github.com/facebook/rocksdb/issues/13431 enforces ingestion only DBs by adding a disallow_memtable_writes column family option.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13434
Test Plan: New unit tests in table_test.cc
Reviewed By: pdillinger
Differential Revision: D70532054
Pulled By: anand1976
fbshipit-source-id: a837487eadfabed9627a0eceb403bfc5fc2c427c
Summary:
Add an unordered_map of name/value pairs in ReadOptions::property_bag, similar to IOOptions::property_bag. It allows users to pass through some custom options to an external table.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13436
Reviewed By: jaykorean
Differential Revision: D70649609
Pulled By: anand1976
fbshipit-source-id: 9b14806a9f3599b861827bd4ae6e948861edc51a
Summary:
PR https://github.com/facebook/rocksdb/issues/13316 broke some crash test cases in DBImplSecondary, from combining test_secondary=1 and preserve_internal_time_seconds>0. Disabling that while investigating the fix.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13439
Test Plan: manual blackbox_crash_test runs with forced test_secondary=1
Reviewed By: anand1976
Differential Revision: D70656373
Pulled By: pdillinger
fbshipit-source-id: fa2139e90bbe64ec8ebb062877d9337894ea3b43
Summary:
... to better support "ingestion only" column families such as those using an external file reader as in https://github.com/facebook/rocksdb/issues/13401.
It would be possible to implement this by getting rid of the memtable for that CF, but it quickly because clear that such an approach would need to update a lot of places to deal with such a possibility. And we already have logic to optimize reads when a memtable is empty. We put a vector memtable in place to minimize overheads of an empty memtable.
There are three layers of defense against writes to the memtable:
* WriteBatch ops to a disallowed CF will fail immediately, without waiting for Write(). For this check to work, we need a ColumnFamilyHandle and because of that, we don't support disallow_memtable_writes on the default column family.
* MemtableInserter will reject writes to disallowed CFs. This is needed to protect re-open with disallow when there are existing writes in a WAL.
* The placeholder memtable is marked immutable. This will cause an assertion failure on attempt to write, such as in case of bug or regression.
Suggested follow-up:
* Remove the limitation on using the option with the default column family, perhaps by solving https://github.com/facebook/rocksdb/issues/13429 more generally or perhaps with some specific check before the first memtable write of the batch (but potential CPU overhead for such a check - there's likely optimization opportunities around ColumnFamilyMemTables).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13431
Test Plan:
unit tests added
Performance: A db_bench call designed to realistically focus on the CPU cost of writes:
```
./db_bench -db=/dev/shm/dbbench1 --benchmarks=fillrandom -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=1000 -fifo_compaction_allow_compaction=0 -num_column_families=20 -disable_wal -write_buffer_size=1234000
```
Running before & after tests at the same time on the same machine, 40 iterations each, average ops/s, DEBUG_LEVEL=0, remove slowest run of each:
Before: 772466
After: 773785 (0.2% faster)
Likely within the noise, as if there was any change, we would expect a slight regression.
Reviewed By: anand1976
Differential Revision: D70495936
Pulled By: pdillinger
fbshipit-source-id: 306f7e737f87c1fbb52c5805f3cadb6e8ced9b40
Summary:
This is an unexpectedly complex follow-up to https://github.com/facebook/rocksdb/issues/13269.
This change solves (and detects regressed) inconsistencies between whether a CF's SuperVersion is configured with a preserve/preclude option and whether it gets a usable SeqnoToTimeMapping. Operating with preserve/preclude and no usable mapping is degraded functionality we need to avoid. And no mapping is useful for actually disabling the feature (except with respect to existing SST files, but that's less of a concern for now).
The challenge is that how we maintain the DB's SeqnoToTimeMapping can depend on all the column families, and we don't want to iterate over all column families *for each column family* (e.g. on initially creating each). The existing code was a bit relaxed:
* On initially creating or re-configuring a CF, we might install an empty mapping, but soon thereafter (after releasing and re-acquiring the DB mutex) re-install another SuperVersion with a useful mapping.
The solution here is to refactor the logic so that there's a distinct but related workflow for (a) ensuring a quality set of mappings when we might only be considering a single CF (`EnsureSeqnoToTimeMapping()`), and (b) massaging that set of mappings to account for all CFs (`RegisterRecordSeqnoTimeWorker`) which doesn't need to re-install new SuperVersions because each CF already has good mappings and will get updated SuperVersions when the periodic task adds new mappings. This should eliminate the extra SuperVersion installs associated with preserve/preclude on CF creation or re-configure, making it the same as any other CF.
Some more details:
* Some refactorings such as removing new_seqno_to_time_mapping from SuperVersionContext. (Now use parameter instead of being stateful.)
* Propagate `read_only` aspect of DB to more places so that we can pro-actively disable preserve/preclude on read-only DBs, so that we don't run afoul of the assertion expecting SeqnoToTime entries.
* Introduce a utility struct `MinAndMaxPreserveSeconds` for aggregating preserve/preclude settings in a useful way, sometimes on one CF and sometimes across multiple CFs. Much cleaner! (IMHO)
* Introduce a function `InstallSuperVersionForConfigChange` that is a superset of `InstallSuperVersionAndScheduleWork` for when a CF is new or might have had a change to its mutable options.
* Eliminate redundant re-install SuperVersions of created "missing" CFs in DBImpl::Open.
Intended follow-up:
* Ensure each flush has an "upper bound" SeqnoToTime entry, which would resolve a FIXME in tiered_compaction_test, but causes enough test churn to deserve its own PR + investigation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13316
Test Plan:
This change is primarily validated by a new assertion in SuperVersion::Init to ensure consistency between (a) presence of any SeqnoToTime mappings in the SuperVersion and (b) preserve/preclude option being currently set.
One unit test update was needed because we now ensure at least one SeqnoToTime entry is created on any DB::Open with preserve/preclude, so that there is a lower bound time on all the future data writes. This required a small hack in associating the time with Seqno 1 instead of 0, which is reserved for "unspecified old."
Reviewed By: cbi42
Differential Revision: D70540638
Pulled By: pdillinger
fbshipit-source-id: bb419fdbeb5a1f115fc429c211f9b8efaf2f56d7
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13435
We've noticed the default CRC32c function gets executed when running on aarch64 cpus within our servers
Issue is that ROCKSDB_AUXV_GETAUXVAL_PRESENT evaluates to false
This fix enables the flag internally and reverts the previous fix, landed with D70423483
Reviewed By: pdillinger
Differential Revision: D70584250
fbshipit-source-id: 28e41316187c474fdfaf854f301ad14b6721fcad
Summary:
when reading with ReadOptions::read_tier = kPersistedTier and with a snapshot, MultiGet allows the case where some CF is read before a flush and some CF is read after the flush. This is not desirable, especially when atomic_flush is enabled and users use MultiGet to do some consistency checks on the data in SST files. This PR updates the code path for SuperVersion acquisition to get a consistent view across when kPersistedTier is used.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13433
Test Plan: a new unit test that could be flaky without this change.
Reviewed By: jaykorean
Differential Revision: D70509688
Pulled By: cbi42
fbshipit-source-id: 80de96f94407af9bb2062b6a185c61f65827c092
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13432
We've noticed the default CRC32c function gets executed when running on aarch64 cpus within our servers
Issue is that ROCKSDB_AUXV_GETAUXVAL_PRESENT evaluates to false
This fix allows the usage of hardware-accelerated crc32 within our fleet
Reviewed By: jaykorean
Differential Revision: D70423483
fbshipit-source-id: 601da3fbf156e3e40695eb76ee5d37f67f83d427
Summary:
This adds a test that attempts DeleteRange() with PlainTable (not supported) and shows that it not only puts the DB in failed write mode, it (a) breaks WriteBatch atomicity for readers, because they can see just part of a failed WriteBatch, and (b) makes the DB not recoverable (without manual intervention) if using WAL.
Note: WriteBatch atomicity is not clearly documented but indicated at the top of write_batch.h and the wiki page for Transactions, even without Transactions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13428
Test Plan: this is the test
Reviewed By: anand1976
Differential Revision: D70332226
Pulled By: pdillinger
fbshipit-source-id: 67bc4de68833a80578e48baa9d3a4f23f1600f3c
Summary:
The existing format compatibility test had limited coverage of compression options, particularly newer algorithms with and without dictionary compression. There are some subtleties that need to remain consistent, such as index blocks potentially being compressed but *not* using the file's dictionary if they are. This involves detecting (with a rough approximation) builds with the appropriate capabilities.
The other motivation for this change is testing some potentially useful reader-side functionality that has been in place for a long time but has not been exercised until now: mixing compressions in a single SST file. The block-based SST schema puts a compression marker on each block; arguably this is for distinguishing blocks compressed using the algorithm stored in compression_name table property from blocks left uncompressed, e.g. because they did not reach the threshold of useful compression ratio, but the marker can also distinguish compression algorithms / decompressors.
As we work toward customizable compression, it seems worth unlocking the capability to leverage the existing schema and SST reader-side support for mixing compression algorithms among the blocks of a file. Yes, a custom compression could implement its own dynamic algorithm chooser with its own tag on the compressed data (e.g. first byte), but that is slightly less storage efficient and doesn't support "vanilla" RocksDB builds reading files using a mix of built-in algorithms. As a hypothetical example, we might want to switch to lz4 on a machine that is under heavy CPU load and back to zstd when load is more normal. I dug up some data indicating ~30 seconds per output file in compaction, suggesting that file-level responsiveness might be too slow. This agility is perhaps more useful with disaggregated storage, where there is more flexibility in DB storage footprint and potentially more payoff in optimizing the *average* footprint.
In support of this direction, I have added a backdoor capability for debug builds of `ldb` to generate files with a mix of compression algorithms and incorporated this into the format compatibility test. All of the existing "forward compatible" versions (currently back to 8.6) are able to read the files generated with "mixed" compression. (NOTE: there's no easy way to patch a bunch of old versions to have them support generating mixed compression files, but going forward we can auto-detect builds with this "mixed" capability.) A subtle aspect of this support that is that for proper handling of decompression contexts and digested dictionaries, we need to set the `compression_name` table property to `zstd` if any blocks are zstd compressed. I'm expecting to add better info to SST files in follow-up, but this approach here gives us forward compatibility back to 8.6.
However, in the spirit of opening things up with what makes sense under the existing schema, we only support one compression dictionary per file. It will be used by any/all algorithms that support dictionary compression. This is not outrageous because it seems standard that a dictionary is *or can be* arbitrary data representative of what will be compressed. This means we would need a schema change to add dictionary compression support to an existing built-in compression algorithm (because otherwise old versions and new versions would disagree on whether the data dictionary is needed with that algorithm; this could take the form of a new built-in compression type, e.g. `kSnappyCompressionWithDict`; only snappy, bzip2, and windows-only xpress compression lack dictionary support currently).
Looking ahead to supporting custom compression, exposing a sizeable set of CompressionTypes to the user for custom handling essentially guarantees a path for the user to put *versioning* on their compression even if they neglect that initially, and without resorting to managing a bunch of distinct named entities. (I'm envisioning perhaps 64 or 127 CompressionTypes open to customization, enough for ~weekly new releases with more than a year of horizon on recycling.)
More details:
* Reduce the running time (CI cost) of the default format compatibility test by randomly sampling versions that aren't the oldest in a category. AFAIK, pretty much all regressions can be caught with the even more stripped-down SHORT_TEST.
* Configurable make parallelism with J environment variable
* Generate data files in a way that makes them much more eligible for index compression, e.g. bigger keys with less entropy
* Generate enough data files
* Remove 2.7.fb.branch from list because it shows an assertion violation when involving compression.
* Randomly choose a contiguous subset of the compression algorithms X {dictionary, no dictionary} configuration space when generating files, with a number of files > number of algorithms. This covers all the algorithms and both dictionary/no dictionary for each release (but not in all combinations).
* Have `ldb` fail if the specified compression type is not supported by the build.
Other future work needed:
* Blob files in format compatibility test, and support for mixed compression. NOTE: the blob file schema should naturally support mixing compression algorithms but the reader code does not because of an assertion that the block CompressionType (if not no compression) matches the whole file CompressionType. We might introduce a "various" CompressionType for this whole file marker in blob files.
* Do more to ensure certain features and code paths e.g. in the scripts are actually used in the compatibility test, so that they aren't accidentally neutralized.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13414
Test Plan: Manual runs with some temporary instrumentation, also a recent revision of this change included a GitHub Actions run of the updated format compatible test: https://github.com/facebook/rocksdb/actions/runs/13463551149/job/37624205915?pr=13414
Reviewed By: hx235
Differential Revision: D70012056
Pulled By: pdillinger
fbshipit-source-id: 9ea5db76ba01a95338ed1a86b0edd71a469c4061
Summary:
added merge support for WBWIMemTable. Most of the preparation work is done in https://github.com/facebook/rocksdb/issues/13387 and https://github.com/facebook/rocksdb/issues/13400. The main code change to support merge is in wbwi_memtable.cc to support reading the Merge value type. The rest of the changes are mostly comment change and tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13410
Test Plan:
- new unit test
- ran `python3 ./tools/db_crashtest.py --txn blackbox --txn_write_policy=0 --commit_bypass_memtable_one_in=100 --test_batches_snapshots=0 --use_merge=1` for several runs.
Reviewed By: jowlyzhang
Differential Revision: D69885868
Pulled By: cbi42
fbshipit-source-id: b127d95a3027dc35910f6e5d65f3409ba27e2b6b
Summary:
... to ensure proper cache charging. However, this is a somewhat hazardous combination if there are many CFs and could be the target of future work.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13398
Test Plan: this is the test
Reviewed By: hx235
Differential Revision: D69619977
Pulled By: pdillinger
fbshipit-source-id: 9841768584e4688d8fdd0258f3ba9608b67408e5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13411
We should intialize statuses with OK rather than IOError to correctly handle cases
like NotFound due to bloom filter. In case of IOError status would be updated
appropriately by the reader
Reviewed By: anand1976
Differential Revision: D69886976
fbshipit-source-id: 92b130168f23633224ff4153bfe46a7d86482b90
Summary:
This is a preparation for supporting merge in `WBWIMemTable`. This PR updates the sequence number assignment method so that it allows efficient and simple assignment when there are multiple entries with the same user key. This can happen when the WBWI contains Merge operations. This assignment relies on tracking the number of updates issued for each key in each WBWI entry (`WriteBatchIndexEntry::update_count`). Some refactoring is done in WBWI to remove `last_entry_offset` as part of the WBWI state which I find it harder to use correctly.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13400
Test Plan: updated unit tests to check that update count is tracked correctly and WBWIMemTable is assigning sequence number as expected.
Reviewed By: pdillinger
Differential Revision: D69666462
Pulled By: cbi42
fbshipit-source-id: 9b18291825017a67c4da3318e8a556aa2971326b
Summary:
This PR introduces an interface to plug in an external table file reader into RocksDB. The external table reader may support custom file formats that might work better for a specific use case compared to RocksDB native formats. This initial version allows the external table file to be loaded and queried using an `SstFileReader`. In the near future, we will allow it to be used with a limited RocksDB instance that allows bulkload but not live writes.
The model of a DB using an external table reader is a read only database allowing bulkload and atomic replace in the bottommost level only. Live writes, if supported in the future, are expected to use block based table files in higher levels. Tombstones, merge operands, and non-zero sequence numbers are expected to be present only in non-bottommost levels. External table files are assumed to have only Puts, and all keys implicitly have sequence number 0.
TODO (in future PRs) -
1. Add support for external file ingestion, with safety mechanisms to prevent accidental writes
2. Add support for atomic column family replace
3. Allow custom table file extensions
4. Add a TableBuilder interface for use with `SstFileWriter`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13401
Reviewed By: pdillinger
Differential Revision: D69689351
Pulled By: anand1976
fbshipit-source-id: c5d5b92d56fd4d0fc43a77c4ceb0463d4f479bda
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13403
Add MultiGet support in SstReader. Today we only have iteration support and this change
also adds MultiGet support to SstFileReader if some application wants to use it.
Reviewed By: anand1976
Differential Revision: D69514499
fbshipit-source-id: 20e85a4bd13a3a9f45dacb223c1a4541fb87f561
Summary:
Noticed that the `do_merge` parameter is not properly set while working on memtable code.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13396
Test Plan: updated unit test for the read-only db case.
Reviewed By: jaykorean
Differential Revision: D69505015
Pulled By: cbi42
fbshipit-source-id: d4c64ca7bba31fe26aa41a29cbc55835d9f1f116
Summary:
**Context/Summary:**
https://github.com/facebook/rocksdb/pull/13263 and https://github.com/facebook/rocksdb/pull/13360 disabled `track_and_verify_wals` with some injection under TXN temporarily but recent stress tests has found more issues this feature surfaced even with the previous disabling. Disabling the feature **completely** now for stabilizing CI while debugging.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13405
Test Plan: Monitor CI
Reviewed By: cbi42
Differential Revision: D69759276
Pulled By: hx235
fbshipit-source-id: 501a3561acb9daa834f874095f9a66ae6ae5aa42
Summary:
**Context/Summary:**
It's [documented (https://github.com/facebook/rocksdb/blob/affcad0cc997958e93bc560202ed107c80d00395/db/job_context.h#L230) that `// For non-empty JobContext Clean() has to be called at least once before before destruction`. This is violated in a UT accidentally so causing the assertion failure `assert(logs_to_free.size() == 0);` in` ~JobContext`. This PR is to fix it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13406
Test Plan: Monitor for future UT assertion failure in `TEST_F(DBWALTest, FullPurgePreservesRecycledLog) `
Reviewed By: cbi42
Differential Revision: D69759725
Pulled By: hx235
fbshipit-source-id: dd1617b370a2c69daba657287dcf258542f92ef5
Summary:
**Context/Summary:**
https://github.com/facebook/rocksdb/commit/02b4197544f758bdf84d80fe9319238611848c48 recently added the ability to detect WAL hole presents in the predecessor WAL. It forgot to update the corrupted wal number to point to the predecessor WAL in that corruption case. This PR fixed it.
As a bonus, this PR also (1) fixed the `FragmentBufferedReader()` constructor API to expose less parameters as they are never explicitly passed in in the codebase (2) a INFO log wording (3) a parameter naming typo (4) the reporter naming
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13359
Test Plan:
1. Manual printing to ensure the corrupted wal number is set to the right number
2. Existing UTs
Reviewed By: jowlyzhang
Differential Revision: D69068089
Pulled By: hx235
fbshipit-source-id: f7f8a887cded2d3a26cf9982f5d1d1ab6a78e9e1
Summary:
**Context/Summary:**
Secondary DB relies on open file descriptor of the shared SST file in primary DB to continue being able to read the file even if that file is deleted in the primary DB. However, this won't work if the file is truncated instead of deleted, which triggers an "truncated block read" corruption in stress test on secondary db reads. Truncation can happen if RocksDB implementation of SSTFileManager and `bytes_max_delete_chunk>0` are used. This PR is to disable such testing combination in stress test and clarify the related API.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13395
Test Plan:
- Manually repro-ed with below UT. I'm in favor of not including this UT in the codebase as it should be self-evident from the API comment now about the incompatiblity. Secondary DB is in a direction of being replaced by Follower so we should minimize edge-case tests for code with no functional change for a to-be-replaced functionality.
```
TEST_F(DBSecondaryTest, IncompatibleWithPrimarySSTTruncation) {
Options options;
options.env = env_;
options.disable_auto_compactions = true;
options.sst_file_manager.reset(NewSstFileManager(
env_, nullptr /*fs*/, "" /*trash_dir*/, 2024000 /*rate_bytes_per_sec*/,
true /*delete_existing_trash*/, nullptr /*status*/,
0.25 /*max_trash_db_ratio*/, 1129 /*bytes_max_delete_chunk*/));
Reopen(options);
ASSERT_OK(Put("key1", "old_value"));
ASSERT_OK(Put("key2", "old_value"));
ASSERT_OK(Flush());
ASSERT_OK(Put("key1", "new_value"));
ASSERT_OK(Put("key3", "new_value"));
ASSERT_OK(Flush());
Options options1;
options1.env = env_;
options1.max_open_files = -1;
Reopen(options);
OpenSecondary(options1);
ASSERT_OK(db_secondary_->TryCatchUpWithPrimary());
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->SetCallBack(
"DeleteScheduler::DeleteTrashFile:Fsync", [&](void*) {
std::string value;
Status s = db_secondary_->Get(ReadOptions(), "key2", &value);
assert(s.IsCorruption());
assert(s.ToString().find("truncated block read") !=
std::string::npos);
});
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->EnableProcessing();
ASSERT_OK(db_->CompactRange(CompactRangeOptions(), nullptr, nullptr));
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->DisableProcessing();
ROCKSDB_NAMESPACE::SyncPoint::GetInstance()->ClearAllCallBacks();
}
```
- Monitor future stress test
Reviewed By: jowlyzhang
Differential Revision: D69499694
Pulled By: hx235
fbshipit-source-id: 57525b9841897f42aecb758a4d3dd3589367dcd9
Summary:
# Problem
Once opened, iterator will preserve its' respective RocksDB snapshot for read consistency. Unless explicitly `Refresh'ed`, the iterator will hold on to the `Init`-time assigned `SuperVersion` throughout its lifetime. As time goes by, this might result in artificially long holdup of the obsolete memtables (_potentially_ referenced by that superversion alone) consequently limiting the supply of the reclaimable memory on the DB instance. This behavior proved to be especially problematic in case of _logical_ backups (outside of RocksDB `BackupEngine`).
# Solution
Building on top of the `Refresh(const Snapshot* snapshot)` API introduced in https://github.com/facebook/rocksdb/pull/10594, we're adding a new `ReadOptions` opt-in knob that (when enabled) will instruct the iterator to automatically refresh itself to the latest superversion - all that while retaining the originally assigned, explicit snapshot (supplied in `read_options.snapshot` at the time of iterator creation) for consistency. To ensure minimal performance overhead we're leveraging relaxed atomic for superversion freshness lookups.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13354
Test Plan:
**Correctness:** New test to demonstrate the auto refresh behavior in contrast to legacy iterator: `./db_iterator_test --gtest_filter=*AutoRefreshIterator*`.
**Stress testing:** We're adding command line parameter controlling the feature and hooking it up to as many iterator use cases in `db_stress` as we reasonably can with random feature on/off configuration in db_crashtest.py.
# Benchmarking
The goal of this benchmark is to validate that throughput did not regress substantially. Benchmark was run on optimized build, 3-5 times for each respective category or till convergence. In addition, we configured aggressive threshold of 1 second for new `Superversion` creation. Experiments have been run 'in parallel' (at the same time) on separate db instances within a single host to evenly spread the potential adverse impact of noisy neighbor activities. Host specs [1].
**TLDR;** Baseline & new solution are practically indistinguishable from performance standpoint. Difference (positive or negative) in throughput relative to the baseline, if any, is no more than 1-2%.
**Snapshot initialization approach:**
This feature is only effective on iterators with well-defined `snapshot` passed via `ReadOptions` config. We modified the existing `db_bench` program to reflect that constraint. However, it quickly turned out that the actual `Snapshot*` initialization is quite expensive. Especially in case of 'tiny scans' (100 rows) contributing as much as 25-35 microseconds, which is ~20-30% of the average per/op latency unintentionally masking _potentially_ adverse performance impact of this change. As a result, we ended up creating a single, explicit 'global' `Snapshot*` for all the future scans _before_ running multiple experiments en masse. This is also a valuable data point for us to keep in mind in case of any future discussions about taking implicit snapshots - now we know what the lower bound cost could be.
## "DB in memory" benchmark
**DB Setup**
1. Allow a single memtable to grow large enough (~572MB) to fit in all the rows. Upon shutdown all the rows will be flushed to the WAL file (inspected `000004.log` file is 541MB in size).
```
./db_bench -db=/tmp/testdb_in_mem -benchmarks="fillseq" -key_size=32 -value_size=512 -num=1000000 -write_buffer_size=600000000 max_write_buffer_number=2 -compression_type=none
```
2. As a part of recovery in subsequent DB open, WAL will be processed to one or more SST files during the recovery. We're selecting a large block cache (`cache_size` parameter in `db_bench` script) suitable for holding the entire DB to test the “hot path” CPU overhead.
```
./db_bench -use_existing_db=true -db=/tmp/testdb_in_mem -statistics=false -cache_index_and_filter_blocks=true -benchmarks=seekrandom -preserve_internal_time_seconds=1 max_write_buffer_number=2 -explicit_snapshot=1 -use_direct_reads=1 -async_io=1 -num=? -seek_nexts=? -cache_size=? -write_buffer_size=? -auto_refresh_iterator_with_snapshot={0|1}
```
| seek_nexts=100; num=2,000,000 | seek_nexts = 20,000; num=50000 | seek_nexts = 400,000; num=2000
-- | -- | -- | --
baseline | 36362 (± 300) ops/sec, 928.8 (± 23) MB/s, 99.11% block cache hit | 52.5 (± 0.5) ops/sec, 1402.05 (± 11.85) MB/s, 99.99% block cache hit | 156.2 (± 6.3) ms / op, 1330.45 (± 54) MB/s, 99.95% block cache hit
auto refresh | 35775.5 (± 537) ops/sec, 926.65 (± 13.75) MB/s, 99.11% block cache hit | 53.5 (± 0.5) ops/sec, 1367.9 (± 9.5) MB/s, 99.99% block cache hit | 162 (± 4.14) ms / op, 1281.35 (± 32.75) MB/s, 99.95% block cache hit
_-cache_size=5000000000 -write_buffer_size=3200000000 -max_write_buffer_number=2_
| seek_nexts=3,500,000; num=100
-- | --
baseline | 1447.5 (± 34.5) ms / op, 1255.1 (± 30) MB/s, 98.98% block cache hit
auto refresh | 1473.5 (± 26.5) ms / op, 1232.6 (± 22.2) MB/s, 98.98% block cache hit
_-cache_size=17680000000 -write_buffer_size=14500000000 -max_write_buffer_number=2_
| seek_nexts=17,500,000; num=10
-- | --
baseline | 9.11 (± 0.185) s/op, 997 (± 20) MB/s
auto refresh | 9.22 (± 0.1) s/op, 984 (± 11.4) MB/s
[1]
### Specs
| Property | Value
-- | --
RocksDB | version 10.0.0
Date | Mon Feb 3 23:21:03 2025
CPU | 32 * Intel Xeon Processor (Skylake)
CPUCache | 16384 KB
Keys | 16 bytes each (+ 0 bytes user-defined timestamp)
Values | 100 bytes each (50 bytes after compression)
Prefix | 0 bytes
RawSize | 5.5 MB (estimated)
FileSize | 3.1 MB (estimated)
Compression | Snappy
Compression sampling rate | 0
Memtablerep | SkipListFactory
Perf Level | 1
Reviewed By: pdillinger
Differential Revision: D69122091
Pulled By: mszeszko-meta
fbshipit-source-id: 147ef7c4fe9507b6fb77f6de03415bf3bec337a8
Summary:
Options File Number to be read by remote worker is part of the `CompactionServiceInput`. We've been setting this in `ProcessKeyValueCompactionWithCompactionService()` while the db_mutex is not held. This needs to be accessed while the mutex is held. The value can change as part of `SetOptions() -> RenameTempFileToOptionsFile()` as in following.
https://github.com/facebook/rocksdb/blob/e6972196bca115e841a6b88d361ba945b49e1e5d/db/db_impl/db_impl.cc#L5595-L5596
Keep this value in memory during `CompactionJob::Prepare()` which is called while the mutex is held, so that we can easily access this later without mutex when building the CompactionInput for the remote compaction.
Thanks to the crash test. This was surfaced after https://github.com/facebook/rocksdb/issues/13378 merged.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13394
Test Plan:
Unit Test
```
./compaction_service_test
```
Crash Test
```
COERCE_CONTEXT_SWITCH=1 COMPILE_WITH_TSAN=1 CC=clang-13 CXX=clang++-13 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j100 dbg
```
```
python3 -u tools/db_crashtest.py blackbox --enable_remote_compaction=1
```
Reviewed By: jowlyzhang
Differential Revision: D69496313
Pulled By: jaykorean
fbshipit-source-id: 7e38e3cb75d5a7708beb4883e1a138e2b09ff837
Summary:
as a preparation to support merge in [WBWIMemtable](https://github.com/facebook/rocksdb/blob/d48af213860054a7696e7ea2764f266c88a3263e/memtable/wbwi_memtable.h#L31), this PR updates how we [order updates to the same key](https://github.com/facebook/rocksdb/blob/d48af213860054a7696e7ea2764f266c88a3263e/utilities/write_batch_with_index/write_batch_with_index_internal.cc#L694-L697) in WriteBatchWithIndex. Specifically, the order is now reversed such that more recent update is ordered first. This will make iterating from WriteBatchWithIndex much easier since the key ordering in WBWI now matches internal key order where keys with larger sequence number are ordered first. The ordering is now explicitly documented above the declaration for `WriteBatchWithIndex` class.
Places that use `WBWIIteratorImpl` and assume key ordering are updated. The rest is test and comments update.
This will affect users who use WBWIIterator directly, the output of GetFromBatch, GetFromBatchAndDB or NewIteratorWithBase are not affected. Users are only affected if they may issue multiple updates to the same key. If WriteBatchWithIndex is created with `overwrite_key=true`, one the the updates needs to be Merge.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13387
Test Plan: we have some good coverage of WBWI, I updated some existing tests and added a test for `WBWIIteratorImpl`.
Reviewed By: pdillinger
Differential Revision: D69421268
Pulled By: cbi42
fbshipit-source-id: d97eec4ee74aeac3937c9758041c7713f07f9676
Summary:
Motivated by code review issue in https://github.com/facebook/rocksdb/issues/13316, we don't want to release the DB mutex in SetOptions between updating the cfd latest options and installing the new Version and SuperVersion. SetOptions uses LogAndApply to install a new Version but this currently incurs an unnecessary manifest write. (This is not a big performance concern because SetOptions dumps a new OPTIONS file, which is much larger than the redundant manifest update.) Since we don't want IO while holding the DB mutex, we need to get rid of the manifest write, and that's what this change does. We introduce a kind of dummy VersionEdit that allows the existing code paths of LogAndApply to install a new Version (with the updated mutable options), recompute resulting compaction scores etc., but without the manifest write.
Part of the validation for this is new assertions in SetOptions verifying the consistency of the various copies of MutableCFOptions. (I'm not convinced we need it in SuperVersion in addition to Version, but that's not for here and now.) These checks depend on defaulted `operator==` so depend on C++20.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13384
Test Plan:
New unit test in addition to new assertions. SetOptions already tested heavily in crash test. Used
`ROCKSDB_CXX_STANDARD=c++20 make -j100 check` to ensure the new assertions are verified
Reviewed By: cbi42
Differential Revision: D69408829
Pulled By: pdillinger
fbshipit-source-id: 4cf026010c6bb381e0ea27567cce2708d4678e7d
Summary:
I found a failed crash test with this error message:
```
Verification failed: Failed to flush primary's WAL before secondary verification
```
`manual_wal_flush_one_in` does not make sense / is not applicable when we are disabling the WAL.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13382
Test Plan: Monitor future crash test runs
Reviewed By: jowlyzhang, anand1976
Differential Revision: D69314053
Pulled By: archang19
fbshipit-source-id: b69d2e1e2869943c0df8cdc4f0623906f4ec7a7a
Summary:
There was a stress test that failed at the assertion check for `IsDataBlockInBuffer`.
`IsDataBlockInBuffer` is too strict of a condition if we are trying to read past the end of the file.
This seems to be a bug from the original 2019 commit https://github.com/siying/rocksdb/commit/3737d06adc01a59e7eb29710a2a4ec64adfaa528: https://github.com/siying/rocksdb/blob/4eb51130917c260f5637731cd77baaa45dfdc5ec/file/file_prefetch_buffer.cc#L130
If the caller tries requesting more bytes than are available, then we still return `n` bytes, even if the buffer really only contains `m < n` bytes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13376
Test Plan: I added a unit test which caused the original `IsDataBlockInBuffer ` assertion to fail. I also updated the unit test to check for the result size, which triggered the bug (without this fix) where we return a size of `n` even if less than `n` bytes exist.
Reviewed By: anand1976
Differential Revision: D69269608
Pulled By: archang19
fbshipit-source-id: 1dc0d5930e2b73089850f6e996afbd6192cd5ac8
Summary:
First step to add (simulated) Remote Compaction in Stress Test. More PRs to come. Just first PR to add the FLAG to enable it. `DbStressCompactionService` will return `kUseLocal` for all compactions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13378
Test Plan:
```
python3 -u tools/db_crashtest.py whitebox --enable_remote_compaction=1
```
```
python3 -u tools/db_crashtest.py blackbox --enable_remote_compaction=1
```
Reviewed By: hx235
Differential Revision: D69269568
Pulled By: jaykorean
fbshipit-source-id: 5119bb6afd4d52f66923fb095150d3132226f7ba
Summary:
**This PR adds a new statistic to track the total number of sorted runs for running compactions.**
Context: I am currently working on a separate project, where I am trying to tune the read request sizes made by `FilePrefetchBuffer` to the storage backend. In this particular case, `FilePrefetchBuffer` will issue larger reads and have to buffer larger read responses. This means we expect to see higher memory utilization. At least for the initial rollout, we only want to enable this optimization for compaction reads.
**I want some way to get a sense of what the memory usage _impact_ will be if the prefetch read request size is increased from (for instance) 8MB to 64MB.**
**If I know the number of files that compactions are actively reading from (i.e. the number of sorted runs / "input iterators"), I can determine how much the memory usage will increase if I bump up the readahead size inside `FilePrefetchBuffer`.** For instance, if there are 16 sorted runs at any given point in time and I bump up the readahead size by 64MB, I can project an increase of 16 * 64 MB.
In most cases, the number of sorted runs processed per compaction is the number of L0 files plus the number of non-L0 levels. However, we need to be aware of exceptions like trivial compactions, deletion compactions, and subcompactions. This is a major reason why this PR chooses to implement the stats counting inside `CompactionMergingIterator`, since by the time we get down to that part of the stack, we know the "true" values for the number of input iterators / sorted runs.
Alternatives considered:
- https://github.com/facebook/rocksdb/issues/13299 gives you a histogram for the number of sorted runs ("input iterators") for a _single compaction_. While this statistic is interested and in the direction of what we want, we are going to be assessing the memory impact across _all_ compactions that are currently running. Thus, this statistic does not give us all the information we need.
- https://github.com/facebook/rocksdb/issues/13302 gives you the total prefetch buffer memory usage, but it doesn't tell you what happens when the readahead size is increased. Furthermore, the code change is error prone and very "invasive" -- look at how many places in the code had to be updated. This would be useful in the future for general memory accounting purposes, but it does not serve our immediate needs.
- https://github.com/facebook/rocksdb/issues/13320 aimed to track the same metric, but did this inside `DbImpl:: BackgroundCallCompaction`. It turns out that this does not handle the case where a compaction is divided into multiple subcompactions (in which case, there would be _more_ sorted runs being processed at the same time than you would otherwise predict.) The current PR handles subcompactions automatically, and I think it is cleaner overall.
Note: When I attempted to put this statistic as part of the `cf_stats_value_` array, even after updating the array to use `std::atomic<uint64_t>`, I still was able to get assertions to _fail_ inside the crash tests. These assertions checked that the unsigned integer would not underflow below zero during compaction. I experimented for many hours but could not figure out a solution, even though it would seem like things "should" work with `fetch_add` and `fetch_sub`. One possibility is that the values in `cf_stats_value_` are being cleared to 0, but I added a `fprintf` to that portion of the code and didn't see it getting printed out before my assertions failed. Regardless, I think that this statistic is different enough from the CF-specific and the other DB-wide stats that the best solution is to just have it defined as a separate `std::atomic<uint64_t>`. I also do not want to spend more hours trying to debug why the crash test assertions break, when the solution in the current version of the PR can get the assertions to consistently pass.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13325
Test Plan:
- I updated one unit test to confirm that `num_running_compaction_sorted_runs` starts and ends at 0. This checks that all the additions and subtractions cancel out. I also made sure the statistic got incremented at least once.
- When I added `fprintf` manually, I confirmed that my statistics updating code was being exercised numerous times inside `db_compaction_test`. I printed out the results before and after the increments/decrements, and the numbers looked good.
- We will monitor the generated statistics after this PR is merged.
- There are assertion checks after each increment and before each decrement. If there are bugs, the crash test will almost certainly find them, since they quickly found issues with my initial implementation for this PR which tried using the `cf_stats_value_` array (modified to use `std::atomic`).
Reviewed By: anand1976, hx235
Differential Revision: D68527895
Pulled By: archang19
fbshipit-source-id: 135cf210e0ff1550ea28ae4384d429ae620b1784
Summary:
This test is flaky likely due to synchronization of the file ingestion thread and the live write thread with test sync points are not working as expected sometimes. Very occasionally, the live write thread can enter the write queue after file ingestion job already dequeued. Or it entered and waited for a very short period of time and quickly returned in the fast path: https://github.com/facebook/rocksdb/blob/833a2266a394fe5f140d2a22f406c82bb605c726/db/write_thread.cc#L83-L86
To fix the flakiness, I moved the test sync points to make sure the write thread is already linked into the write queue before the file ingestion writer get dequeued, so it definitely would need to wait some time in order to do its write.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13374
Test Plan:
I'm able to reproduce the flakiness with this command before the fix with every two or three runs:
./gtest-parallel external_sst_file_basic_test --gtest_filter=ExternalSSTFileBasicTest.Basic --repeat=10000 --workers=100
After the fix, I have tried the command for 10 runs, and there is no failure detected.
Reviewed By: cbi42
Differential Revision: D69258712
Pulled By: jowlyzhang
fbshipit-source-id: adcbad4dd53ccddab5c137d3f9d740b9f9623207
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13370
We have a class called `DefaultSecondaryIndex` in `TransactionTest.SecondaryIndexPutDelete` that contains generally useful functionality. The patch generalizes it a bit to make the column name configurable, renames it to `SimpleSecondaryIndex`, and moves it to the public API so applications can use it.
Reviewed By: jowlyzhang
Differential Revision: D69147890
fbshipit-source-id: 0d2d1cc5adcde01f3978a450ec841c9e990d2170
Summary:
There was a failed TSAN crash test run that involved BlobDB and secondary instances. ltamasi said that BlobDB is not compatible with secondary instances, so I have updated the crash test script accordingly.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13371
Test Plan:
I confirmed there were no blob-related parameters after running
```
python3 tools/db_crashtest.py --simple blackbox --test_secondary=1
```
Reviewed By: jowlyzhang
Differential Revision: D69193105
Pulled By: archang19
fbshipit-source-id: b545d7765928a385a792fc070c1d432d1c002b3d
Summary:
As titled. unreleased_history directory now only contain release notes for the next 10.0 release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13373
Reviewed By: ltamasi
Differential Revision: D69196468
Pulled By: jowlyzhang
fbshipit-source-id: 849193c7901c5938d3d7c938e3b6c805532d7de4
Summary:
We want to disable WAL for RoWS stress tests (anand1976 made a config change to explicitly do this), but it turns out that is not compatible with `reopen` > 0.
I found this error in the logs:
```
Error: Db cannot reopen safely with disable_wal set!
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13372
Test Plan: We should not get this error message in the RoWS stress tests.
Reviewed By: jowlyzhang
Differential Revision: D69193849
Pulled By: archang19
fbshipit-source-id: 933252926a906183c9abdef0b47f641073c5de37
Summary:
`DynamicLevelCompressionPerLevel` test started _somewhat occasionally_ failing post refactoring in https://github.com/facebook/rocksdb/pull/13322. In order for `DeleteFilesInRange`-replacement to behave according to our expectations (that is delete exactly that very single file given its' key range), we must first ensure that input `keys` are NOT randomly shuffled, but rather preserved in their natural, sequential order. That change was originally a part of the PR, but got somehow deleted due to human error and since tests passed locally and in CI, spilled unnoticed. We're removing random keys reshuffling (as intended originally) and, in addition, asserting that all such constructed files are 1) non-overlapping and 2) contain full range of keys BEFORE we actually get to test the on table deletion callbacks.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13349
Test Plan: Confirmed that key range overlap is an issue by volume testing: `./db_test --gtest_filter=*DynamicLevelCompressionPerLevel --gtest_repeat=1000 --gtest_break_on_failure` (2-3 times is enough). Could not longer repro after the fix.
Reviewed By: jaykorean
Differential Revision: D68857018
Pulled By: mszeszko-meta
fbshipit-source-id: 873b1ba44f32d40192da4265aeeb39702c22a1d0
Summary:
**Context/Summary:**
archang19 found the place in code where no injected error status is returned on effectively injected error (empty result or corrupted bytes). I can't find a good argument for doing so. In these cases where such empty result and corrupted result is not expected, the file system should return error (< 0). Our fault injection framework should align with that to simulate fault returned by file system.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13369
Test Plan: Monitor stress test
Reviewed By: archang19
Differential Revision: D69136015
Pulled By: hx235
fbshipit-source-id: 6ee7a7bd5e0aa19837e4dfd73817d4a9d5af76f9
Summary:
The crash tests are failing during secondary database verification due to a "truncated block read" error.
https://github.com/facebook/rocksdb/issues/13366 attempted to resolve the issue by checking for injected errors. However, that did not work.
It turns out that sometimes faults are injected yet the return status is still "OK."
See https://github.com/facebook/rocksdb/blob/main/utilities/fault_injection_fs.cc#L1407-L1414 for an example:
```cpp
} else if (Random::GetTLSInstance()->OneIn(8)) {
assert(result);
// For a small chance, set the failure to status but turn the
// result to be empty, which is supposed to be caught for a check.
*result = Slice();
msg << "empty result";
ctx->message = msg.str();
ret_fault_injected = true;
```
My hypothesis is that this particular fault injection is the root cause of the "truncated block read" error.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13368
Test Plan: Hopefully the recurring crash tests start passing consistently for secondary db verification
Reviewed By: hx235
Differential Revision: D69132024
Pulled By: archang19
fbshipit-source-id: 941406165a2fd306f10048614457261cda99d762
Summary:
Leading up to some compression code refactoring, we have a bit of an ifdef nightmare in compression.h relating to zstd support. With the major release RocksDB 10.0.0 coming up, it is a good time to clean up much of this tech debt by requiring zstd >= 1.4.0 (April 2019) if building RocksDB with ZSTD support. For example, Ubuntu 20, the first LTS version to properly support C++17 in its built-in gcc, comes with zstd version 1.4.4. This should not be a significant limitation.
* Almost all of the `ZSTD_VERSION_NUMBER` checks are simplified to just `ZSTD`, though
* `ROCKSDB_ZSTD_DDICT` still needs to be separate because of dependency on `ZSTD_STATIC_LINKING_ONLY` (added to fbcode_config_platform010.sh by the way)
* Similar for ZDICT_finalizeDictionary, which is only generally available in >= 1.4.5
* Eliminate deprecated `kZSTDNotFinalCompression`
* Reduce some cases of unnecessary copying definitions across `#if` branches (e.g. `ZSTDUncompressCachedData`)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13362
Test Plan:
minor unit test updates. `make check` on several build variants with/without zstd and with/without `ZSTD_STATIC_LINKING_ONLY`
Also deflaked DBTest.DynamicLevelCompressionPerLevel which was flaky before this change but failed once in CI
Reviewed By: cbi42
Differential Revision: D69129453
Pulled By: pdillinger
fbshipit-source-id: ef0cbf9f0fea4e7684fa0999320aa170cfbec233
Summary:
**Context/Summary:**
https://github.com/facebook/rocksdb/pull/13263 temporally disable `track_and_verify_wals=1` with write fault injection in all cases to mitigate a WAL hole not fully debugged. Fully debugging shows the WAL hole only happens under pessimistic TXN when two-phase-commit (2pc) was used.
The bug essentially is about 2pc won't be able to discard the corrupted WAL as it would in non-2pc case as part of the WAL write error recovery. So the corrupted WAL will still present in the next DB open and caught by `track_and_verify_wals=1`.
This fix is going to take a while. So for now, let's reduce the scope of disabling the testing.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13360
Test Plan: Monitor stress test for WAL recovery error/corruption
Reviewed By: jaykorean
Differential Revision: D68973022
Pulled By: hx235
fbshipit-source-id: ea8db6fa11ba25ace896da7cdb1dc1cd757742f6
Summary:
https://github.com/facebook/rocksdb/issues/13281 added secondary database verification to the crash tests.
I am seeing failures in the crash test that trace back to these two code sections:
1. https://github.com/facebook/rocksdb/blob/main/db_stress_tool/no_batched_ops_stress.cc#L2969-L2975
```cpp
VerificationAbort(
shared,
msg_prefix + "Non-OK status" + read_u64ts.str() + s.ToString(), cf,
key, "", Slice(expected_value_data, expected_value_data_size));
```
2. https://github.com/facebook/rocksdb/blob/main/table/block_fetcher.cc#L327-L331
```cpp
io_status_ = IOStatus::Corruption(
"truncated block read from " + file_->file_name() + " offset " +
std::to_string(handle_.offset()) + ", expected " +
std::to_string(block_size_with_trailer_) + " bytes, got " +
std::to_string(slice_.size()));
```
The error messages look like
```
Secondary get verificationNon-OK statusCorruption: truncated block read from /dev/shm/rocksdb_test/rocksdb_crashtest_blackbox/011887.sst offset 11780096, expected 16274 bytes, got 0
```
As you can see, the issue is not that the values of the secondary DB differ from what we expect. Rather, the `get` request itself is returning a non-OK status. I looked at the test configurations for the failed test runs, and I saw that both of them enabled fault injections (e.g. `read_fault_one_in`).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13366
Test Plan:
Before merging: `python3 tools/db_crashtest.py --simple blackbox --test_secondary=1`
After merging: monitor for crash test failures
Reviewed By: jaykorean
Differential Revision: D69059138
Pulled By: archang19
fbshipit-source-id: a9c07d80381f52bdff220b0db3302748ebccd96c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13361
After https://github.com/facebook/rocksdb/pull/13346 and https://github.com/facebook/rocksdb/pull/13348, K-nearest-neighbors queries no longer have to be exposed via an iterator API. The patch makes the interface for KNN search more natural by replacing `KNNIterator` in `FaissIVFIndex` with a new method `FindKNearestNeighbors`. This simplifies both the use and the implementation of `FaissIVFIndex`.
Reviewed By: jowlyzhang
Differential Revision: D68973541
fbshipit-source-id: cd6fec44c202e7cfa7219af482d1ca800e2d672d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13353
The patch changes `SecondaryIndexIterator` to a standalone concrete class that mimics most of `Iterator`'s interface but no longer derives from `Iterator`. This eliminates the need to implement `Iterator` methods which are not applicable in the context of secondary indices (namely `SeekToFirst`, `SeekToLast`, and `SeekForPrev`). The class is also moved to the public interface; with this move, the earlier factory method doesn't really add much value anymore and is thus removed.
Reviewed By: jowlyzhang
Differential Revision: D68923662
fbshipit-source-id: 9e1af250bb392535537d6c867f36d23dae5b01b9
Summary:
This bug was spotted by cbi42 and should be the root cause for the crash test data races 🤞 .
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13351
Test Plan: Monitor recurring crash tests.
Reviewed By: hx235
Differential Revision: D68909000
Pulled By: archang19
fbshipit-source-id: e0bdfda9f92eacd2513fc8894f8cde35da88da68
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13348
This eliminates the need to shoehorn all index queries into a single method signature. With this change, `SecondaryIndex` implementations can expose the queries they support via the most natural interface. For `FaissIVFIndex`, this means that KNN search need not be modeled using an iterator anymore; however, for now, the class still has a (non-virtual) `NewIterator` method that takes a read options structure `FaissIVFIndexReadOptions`.
Reviewed By: jowlyzhang
Differential Revision: D68852927
fbshipit-source-id: b4f63bfea9cd73a6c99a547de2a0676e1e8dee0d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13346
As the first step of revising the secondary index query API, the patch moves `FaissIVFIndex` to the public header. This will enable querying the index without having a `NewIterator` virtual in the `SecondaryIndex` interface (which will be removed in the next step of this cleanup).
Reviewed By: jowlyzhang
Differential Revision: D68846678
fbshipit-source-id: 37617d7da87a5c31b1ec7d82ef9694f8519d78d6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13326
This diff introduces ToolHooks, a class which allows for users to interpose their own set of logic for various functionality with db_bench_tool (i.e., various OpenDB implementations).
Reviewed By: anand1976
Differential Revision: D67868126
fbshipit-source-id: df433b0c8a064a86735b92a8ef5f38527dbc9112
Summary:
Fixing the GetMergeOperands() in ReadOnlyDB and SecondaryDB as reported in https://github.com/facebook/rocksdb/issues/13243. Refactor in https://github.com/facebook/rocksdb/issues/11799 introduced this regression.
Follow ups to come
- Large Result Optimization (done in https://github.com/facebook/rocksdb/issues/10458 ) for ReadOnlyDB and SecondaryDB
- Stress Test / Crash Test coverage
- Consider removing some duplicate logic between ReadOnlyDB's GetImpl() and SecondaryDB's `GetImpl()`. The only difference is between acquiring/referencing Superversion.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13340
Test Plan:
`DBMergeOperandTest` and `DBSecondaryTest` updated
```
./db_merge_operand_test --gtest_filter="*GetMergeOperandsBasic*"
```
```
./db_secondary_test -- --gtest_filter="*GetMergeOperands*"
```
Reviewed By: ltamasi
Differential Revision: D68791652
Pulled By: jaykorean
fbshipit-source-id: 760925e257ab10993c207094718dc0659822ae64
Summary:
This is a continuation of https://github.com/facebook/rocksdb/pull/13338, which aims to address crash test failures caused by https://github.com/facebook/rocksdb/pull/13281.
This PR attempts to address the TSAN failures.
I searched for wherever we call `column_families_.clear()` and made sure that we also clear the secondary column families as well. I made a helper method since it is easy to forget to clear both sets of column families.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13343
Test Plan: Monitor recurring crash test results.
Reviewed By: cbi42
Differential Revision: D68790580
Pulled By: archang19
fbshipit-source-id: 96ed758a21545dd20181b8db71b81dd660546e18
Summary:
https://github.com/facebook/rocksdb/issues/13281 added support for verifying secondaries in the crash tests. We are trying to check that the values returned by the secondary in `Get` requests fall within an expected range of values. We do reads from the shared expected state before and after we read from the secondary.
There are some rare verification failures where `VerifyValueRange` fails with `Unexpected value found outside of the value base range`.
I have some ideas on what the root cause could be. The secondary can read the WAL, MANIFEST, and SST files, but in some scenarios some of these pieces may not be present.
I noticed that the failures had `manual_wal_flush_one_in=1000`, which means that `options.manual_wal_flush` is set to `true`. With this setting, RocksDB has its own internal buffers that need to be manually flushed for the WAL to be persisted.
Although the test failures I looked at did not disable the WAL, I realized that, when the WAL is disabled, we should flush the primary's memtables, since the secondary needs to be able to find SST files to fully catch up.
Injected faults further complicate matters, so I have a check to skip secondary verification whenever the WAL or memtable flushes fail due to fault injection.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13338
Test Plan:
Locally:
```
python3 tools/db_crashtest.py --simple blackbox --test_secondary=1
python3 tools/db_crashtest.py --simple whitebox --test_secondary=1
python3 tools/db_crashtest.py --simple blackbox --test_secondary=1 --disable_wal=1
python3 tools/db_crashtest.py --simple blackbox --test_secondary=1 --disable_wal=0 --manual_wal_flush_one_in=1000
```
I will monitor the recurring crash tests after this gets merged.
Reviewed By: anand1976
Differential Revision: D68741287
Pulled By: archang19
fbshipit-source-id: 86f474c41a68b7b06f2ed80a851c6cb52a47ebe7
Summary:
https://github.com/facebook/rocksdb/issues/13281 added support to the crash tests for secondary DB verification.
I looked at our recurring crash tests to see what impact https://github.com/facebook/rocksdb/issues/13281 had. The actual secondary verification looks okay to me (no `assert` failures), but I noticed memory leaks were detected.
The problematic areas were tracked down to the call to `DB::OpenAsSecondary` from `rocksdb::StressTest::Open`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13337
Test Plan:
Monitor recurring crash tests. It is likely hard to reproduce the ASAN failures locally if they are rare enough.
```
make -j100 db_stress COMPILE_WITH_ASAN=1
python3 tools/db_crashtest.py --simple blackbox --test_secondary=1
```
Reviewed By: cbi42
Differential Revision: D68721624
Pulled By: archang19
fbshipit-source-id: 9c3044884c505c43c1819a3e98ce99b2d171f3ca
Summary:
Cleanup post https://github.com/facebook/rocksdb/pull/13284.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13322
Test Plan:
1. We did not find any evidence of breakage in internal pre-release integration pipeline runs after renaming the deprecated API in `9.10`.
2. _To the extent possible_, we manually validated partner use cases of file deletion and confirmed deprecated API is no longer in use.
Reviewed By: jaykorean
Differential Revision: D68476852
Pulled By: mszeszko-meta
fbshipit-source-id: fbe1f873e16ae7c60d7706a3c44ecc695ab86a4b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13329
The patch adds two convenience methods `ConvertFloatsToSlice` and `ConvertSliceToFloats` that can be used to convert embeddings from a contiguous range of floats to a RocksDB `Slice` or vice versa. The methods are added to the public API so they can be utilized by applications as well.
Reviewed By: jowlyzhang
Differential Revision: D68581494
fbshipit-source-id: 2207fa3e668a6546b7de6d8ab78be2ba9f2ffd8c
Summary:
TLDR: This PR enables secondary DB verification inside the "simple" crash tests (`NonBatchedOpsStressTest`). Essentially, we want to be able to verify that the secondary is a valid "prefix" of the primary. This PR allows us to do this by piggybacking on the existing verification of the primary through `Get()` requests.
I originally proposed replaying the trace file to recreate the `ExpectedState` as of a specific sequence number. This could be used to run verifications against the secondary database. I did some experimenting in https://github.com/facebook/rocksdb/issues/13266 and got a "mostly working" implementation of this approach. I could sometimes get through entire key space verifications but eventually one of the keys would fail verification. I have not figured out the root cause yet, but I assume that something caused the sequence number to trace record alignment to break.
The approach in this PR is considerably simpler. We can just check that the secondary database's value is in the correct "range," which we already have functionality for checking that. Compared to the approach in https://github.com/facebook/rocksdb/issues/13266, this approach is _much, much simpler_ since we do not have to go through the whole headache of replaying the trace and creating an entire new `ExpectedState`. (Look at https://github.com/facebook/rocksdb/issues/13266 to see how much of a mess that creates.) I think this approach is better than my original approach in almost most aspects: it's faster, uses less space, and has less room for implementation errors.
Other nice aspects of this approach:
1. We don't need to block the primary. (Another approach you could imagine would be to block writes to the primary, have the secondary catch up, do the whole verification, and then re-enable writes to the primary.)
2. We don't need to block the secondary or do any special coordination (locks, sync points, etc). (If we insist on one "golden" expected value to be read from the secondary, then we need to make sure that another thread does not call `TryCatchUpWithPrimary` while we are trying to perform a `Get()`)
3. More "realistic" usage of the secondary. For instance, writes to the primary and secondary would continue on in production while we try to read from the secondary.
The main drawback of course is that we verify against a range of expected values, rather than one particular expected value. However, I think this is acceptable and "good enough" especially with all of other the aforementioned benefits.
Historical context: There is some very old code that attempted to verify secondaries, but is not enabled. This code has not been touched or executed in an extremely long time, and the crash tests started failing when I tried enabling it, most likely because the code is not compatible with certain other crash test options. This code is for the "continuous verification" and involves long iterator scans over the secondary database. Some of the code involved the cross CF consistency test type. I don't think the old checks are what we really want for our purposes of verifying the secondary functionality. Since I don't think we will get much value out of this old "continuous verification" code, I integrated my secondary verification with the "regular" database verification. This also makes the rollout simpler on my end, since I can control whether my secondary verifications are enabled through one `test_secondary` configuration. To make sure the old code does not execute for our recurring crash test runs, I had to enforce that `continuous_verification_interval` is 0 whenever `test_secondary` is set.
Monitoring: I will want to monitor the Sandcastle "simple" runs for failures where `test_secondary` is set. All of my error messages are prefixed with "Secondary" so it should be easy to tell if this PR causes any crash test issues.
Future work:
1. Extend this to followers. I think the same verification method should work, so most of the code from this PR should be reusable
2. Add additional checks to make sure the sequence number of the follower/secondary is actually increasing. For instance, if the primary's sequence number has advanced, and in that period the secondary has not (even after calling `TryCatchUpWithPrimary`), then we know there is a problem
3. Potentially checking things other than `Get()` for the secondary (i.e. iterators). I think the focus here should be testing replication-specific logic, and since we will already have separate unit tests, we do not need to repeat all of tests against both the primary and the secondary.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13281
Test Plan:
The primary crash test commands I ran were:
```
python3 tools/db_crashtest.py --simple blackbox --test_secondary=1
python3 tools/db_crashtest.py --simple whitebox --test_secondary=1
```
As a sanity check, I added an `assert(false)` right after my secondary verification code to make sure that my code was actually being run.
Reviewed By: anand1976
Differential Revision: D67953821
Pulled By: archang19
fbshipit-source-id: 0bd853580ea53566be41639f5499eb9b5e0e9376
Summary:
The patch adds a unit test that reproduces an issue we have been seeing in our stress tests that affects reverse iteration when BlobDB and user-defined timestamps are both enabled. If in addition to the above, lazy loading of blobs (`allow_unprepared_value`) is enabled and `max_sequential_skip_in_iterations` is exceeded during the reverse scan, calling `PrepareValue` can result in an error status (`Corruption: Key mismatch when reading blob`). We plan to fix the issue in a follow-up patch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13332
Reviewed By: jowlyzhang
Differential Revision: D68642615
fbshipit-source-id: a09b24e2dda6b5fa97ae576708ab278f540251bf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13327
The patch adds a public API method `NewSecondaryIndexIterator` that can be leveraged by users providing their own `SecondaryIndex` implementations.
Reviewed By: jaykorean
Differential Revision: D68569198
fbshipit-source-id: 07f77837c3ce7ab8ea2d9bac172df3d64ce4f745
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13324
There are actually some use cases which would benefit from the ability to use the primary key when forming the secondary key prefix or value. One such use case, which is demonstrated using a unit test, is building a secondary index on non-initial part(s) of the primary key. The patch adds back this ability, which was was removed in https://github.com/facebook/rocksdb/pull/13207, with a twist: the earlier `GetSecondaryKeyPrefix` is essentially split into two parts, with `GetSecondaryKeyPrefix` now being responsible only for computing whatever the secondary index is built on (let's call this "index function result") and a new `FinalizeSecondaryKeyPrefix` method having the responsibility of dealing with serialization concerns like adding a length indicator for disambiguation. This also means a slight change for the `SecondaryIndexIterator` class: it now treats its `Seek` argument as an "index function result" and thus only calls the new `FinalizeSecondaryKeyPrefix` on it (but not `GetSecondaryKeyPrefix`).
Reviewed By: jaykorean
Differential Revision: D68514201
fbshipit-source-id: d3750d049b0aee37e6c20edc19f5e4a0d3fce91e
Summary:
Today, backup verification is serial, which could pose a challenge in rare, high urgency recovery scenarios where we want to timely assess whether candidate backup is not corrupted and eligible for the restore. The _timely_ part will become increasingly more important in case of disaggregated storage.
### Semantics
Given the very simple thread pool implementation in `backup_engine` today, we do not really have a control over initialized threads and consequently do not have an option to unschedule / cancel in-progress tasks. As a result, `VerifyBackup` won't bail out on a very first mismatch (as it was the case for serial implementation) and instead will iterate over all the files logging success / degree_of_failure for each. We _could_, in theory, not `.wait()` on remaining `std::future<WorkItem>`s (upon previously detected failure) and therefore decrease the observed API latency, but that _could_ cause more confusion down the road as verification threads would still be occupied with inflight/scheduled work and would not be reclaimed by the pool for a while. It's a tradeoff where we choose a solution with clear and intuitive semantics.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13292
Test Plan:
Kudos to pdillinger who pointed out that we should already have appropriate fuzzing for max_background_operations and verify_checksum=true parameters in scope of ::VerifyBackup calls in existing backup restore stress test collateral.
[1]
https://github.com/facebook/rocksdb/blob/main/db_stress_tool/db_stress_test_base.cc#L1296
Reviewed By: pdillinger
Differential Revision: D68046714
Pulled By: mszeszko-meta
fbshipit-source-id: 980253174aa9dfd3064866a51c53345277e3a032
Summary:
... to makes it easier to use the new transaction feature `commit_bypass_memtable`. Instead of needing to specify the option when creating a transaction, this option allows users to specify a threshold on the number of updates in a transaction to determine when to skip memtables writes for a transaction.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13304
Test Plan: a new unit test for the new option
Reviewed By: pdillinger
Differential Revision: D68288579
Pulled By: cbi42
fbshipit-source-id: d3076629891d8b1d427878d20f0ac40dc0dadd35
Summary:
With this change we are adding native library support for incremental restores. When designing the solution we decided to follow 'tiered' approach where users can pick one of the three predefined, and for now, mutually exclusive restore modes (`kKeepLatestDbSessionIdFiles`, `kVerifyChecksum` and `kPurgeAllFiles` [default]) - trading write IO / CPU for the degree of certainty that the existing destination db files match selected backup files contents. New mode option is exposed via existing `RestoreOptions` configuration, which by this time has been already well-baked into our APIs. Restore engine will consume this configuration and infer which of the existing destination db files are 'in policy' to be retained during restore.
### Motivation
This work is motivated by internal customer who is running write-heavy, 1M+ QPS service and is using RocksDB restore functionality to scale up their fleet. Given already high QPS on their end, additional write IO from restores as-is today is contributing to prolonged spikes which lead the service to hit BLOB storage write quotas, which finally results in slowing down the pace of their scaling. See [T206217267](https://www.internalfb.com/intern/tasks/?t=206217267) for more.
### Impact
Enable faster service scaling by reducing write IO footprint on BLOB storage (coming from restore) to the absolute minimum.
### Key technical nuances
1. According to prior investigations, the risk of collisions on [file #, db session id, file size] metadata triplets is low enough to the point that we can confidently use it to uniquely describe the file and its' *perceived* contents, which is the rationale behind the `kKeepLatestDbSessionIdFiles` mode. To find more about the risks / tradeoffs for using this mode, please check the related comment in `backup_engine.cc`. This mode is only supported for SSTs where we persist the `db_session_id` information in the metadata footer.
2. `kVerifyChecksum` mode requires a full blob / SST file scan (assuming backup file has its' `checksum_hex` metadata set appropriately, if not additional file scan for backup file). While it saves us on write IOs (if checksums match), it's still fairly complex and _potentially_ CPU intensive operation.
3. We're extending the `WorkItemType` enum introduced in https://github.com/facebook/rocksdb/pull/13228 to accommodate a new simple request to `ComputeChecksum`, which will enable us to run 2) in parallel. This will become increasingly more important as we're moving towards disaggregated storage and holding up the sequence of checksum evaluations on a single lagging remote file scan would not be acceptable.
4. Note that it's necessary to compute the checksum on the restored file if corresponding backup file and existing destination db file checksums didn't match.
### Test plan ✅
1. Manual testing using debugger: ✅
2. Automated tests:
* `./backup_engine_test --gtest_filter=*IncrementalRestore*` covering the following scenarios: ✅
* Full clean restore
* Integration with `exclude files` feature (with proper writes counting)
* User workflow simulation: happy path with mix of added new files and deleted original backup files,
* Existing db files corruptions and the difference in handling between `kVerifyChecksum` and `kKeepLatestDbSessionIdFiles` modes.
* `./backup_engine_test --gtest_filter=*ExcludedFiles*` ✅
* Integrate existing test collateral with newly introduced restore modes
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13239
Reviewed By: pdillinger
Differential Revision: D67513875
Pulled By: mszeszko-meta
fbshipit-source-id: 273642accd7c97ea52e42f9dc1cc1479f86cf30e
Summary:
Offer new DB::Open and variants that use `std::unique_ptr<DB>*` output parameters and deprecate the old versions that use `DB**` output parameters.
This shouldn't have weird downstream effects because these are just static functions. (And a constructor for StackableDB)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13311
Test Plan: existing tests
Reviewed By: anand1976
Differential Revision: D68340779
Pulled By: pdillinger
fbshipit-source-id: 30f4448398b479b5abecfc2406447f200a5fe073
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13312
The patch moves the `AsSlice` and `AsString` methods to a new `SecondaryIndexHelper` class to facilitate reuse and eliminate some code duplication.
Reviewed By: jaykorean
Differential Revision: D68342378
fbshipit-source-id: 9cb55bfd64a7db810898739dde01b128e15c81f4
Summary:
FlushReason enum in C++ has members up to 15, but in Java, the mirroring FlushReason only supports reason codes up to 12. This causes exceptions when adding a flush listener.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13246
Reviewed By: pdillinger
Differential Revision: D68241620
Pulled By: jaykorean
fbshipit-source-id: 1e2856dad28dff0cbb1772f5a8ea03cc1e224088
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13305
The patch adds a public factory method `NewFaissIVFIndex` that can be used to create a FAISS inverted file based secondary index object. (Note that at the moment, FAISS secondary indices require using the Meta-internal BUCK build; this will be addressed in a follow-up patch.) As a small code organization improvement, the patch also moves `SecondaryIndexReadOptions` to its own header file.
Reviewed By: jaykorean
Differential Revision: D68284544
fbshipit-source-id: b46351c110589ec05606710452016deaa5028626
Summary:
As follow-up to https://github.com/facebook/rocksdb/issues/13239, this change is primarily motivated by simplifying the calling conventions of LogAndApply. Since it must be called while holding the DB mutex, it can read safely read cfd->GetLatestMutableCFOptions(), until it releases the mutex within ProcessManifestWrites. Before it releases the mutex, it makes a copy of the mutable options in a new, unpublished Version object, which can be used when not holding the DB mutex. This eliminates the need for callers of LogAndApply to copy mutable options for its sake, or even specify mutable options at all. And it eliminates the need for *another* copy to be saved in ManifestWriter.
Other functions that don't need the mutable options parameter:
* ColumnFamilyData::CreateNewMemtable()
* CompactionJob::Install() / InstallCompactionResults()
* MemTableList::*InstallMemtable*()
* Version::PrepareAppend()
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13301
Test Plan: existing tests, CI with sanitizers
Reviewed By: mszeszko-meta
Differential Revision: D68234865
Pulled By: pdillinger
fbshipit-source-id: 6ce95f9cc479834e09ffc8ce93cbae7b664329e5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13300
The patch adds a new unit test for `FaissIVFIndex` that compares its results with a regular in-memory FAISS index. Specifically, it trains two identical IVF indices using the same training vectors, passes the ownership of one to `FaissIVFIndex`, adds the same set of database vectors to both, and then queries them using the same query vectors (with a variety of values for number of neighbors and number of probes).
Reviewed By: jaykorean
Differential Revision: D68233815
fbshipit-source-id: 7577a65c03c7b811707a4dbcd81e69ed85202a51
Summary:
To start, I wanted to remove the unnecessary new_options parameter of `InstallSuperVersionAndScheduleWork()`. Passing it something other than the latest mutable options would be inconsistent/outdated. There was even a comment "Use latest MutableCFOptions" on a place that was using the saved options in effect for the compaction.
On investigation, this fixes an undiagnosed but longstanding serious bug in SetOptions() where the new settings can be reverted if a flush or compaction started before the SetOptions() finishes after. Fix confirmed with new unit test in db_test.cc.
I also got tired of seeing the cumbersome usage of pointer rather than const reference for related options accesses, so there's kind of a large (but trivial) refactoring tied in here as well. (Sorry for combining them; wasn't planning a major bug fix)
Intended follow-up: Clarify/simplify the crazy calling conventions of LogAndApply, and remove some unnecessary copying of MutableCFOptions (see new FIXMEs)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13294
Test Plan: test for bug fix, confirmed fails on main and at least as far back as version 8.10. Plus existing tests and CI
Reviewed By: mszeszko-meta
Differential Revision: D68141563
Pulled By: pdillinger
fbshipit-source-id: f6c3290145afa06cc2fe8b485a5de17560a5deea
Summary:
Currently, when the primary instance shuts down, remote compaction continues to run and `CompactionService::Wait()` does not get aborted. This slows down `DB::Close()` as it waits for the completion of `CompactionService::Wait()`. Moreover, since shutdown has already begun, the compaction is unnecessary and will be wasted.
This PR introduces `CancelAwaitingJobs()` to the CompactionService interface. This allows users to implement cancellation of running remote compactions from the primary instance. When `CancelAllBackgroundWork()` is called on the primary instance, `CancelAwaitingJobs()` will be invoked, enabling a more efficient shutdown process.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13286
Test Plan:
Unit Test added
```
./compaction_service_test --gtest_filter="*CancelCompactionOnPrimarySide*"
```
Reviewed By: anand1976, cbi42
Differential Revision: D68035191
Pulled By: jaykorean
fbshipit-source-id: 47da641f7cbed1267f0a1f16924f57efde46216d
Summary:
The patch implements support for `Delete` and `SingleDelete` with secondary indices, leveraging the earlier pieces built for `Put` / `PutEntity`. As expected, deleting an entry using these APIs also deletes any associated secondary index entries.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13291
Reviewed By: jaykorean
Differential Revision: D68041422
fbshipit-source-id: c8afc9ff69dea834f89ae855a72c1d76e7db0e35
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13289
The patch adds support for `Put` / `PutUntracked` to the secondary indexing logic. Similarly to `PutEntity` (see https://github.com/facebook/rocksdb/pull/13180), calling these APIs automatically add or remove secondary index entries as needed in an atomic and transparent fashion.
Reviewed By: jaykorean
Differential Revision: D68035089
fbshipit-source-id: db37bce62151ae1909b46b1020592c8348156653
Summary:
We added a removal warning for public `DB::DeleteFile` API ~4 years ago in https://github.com/facebook/rocksdb/pull/7337. This API seems to sit at wrong layer of abstraction, where instead of exposing a clear interface to delete specific range of keys, callers rely on their own discovery / interpretation of where their data / log possibly resides 'as-of-now'. For example, in case of data, the physical location of the keys might very well change after user obtained their mapping from key(s) to specific SST file. This will lead to `InvalidArgument` response, which if repeated, would put a user in a race condition spinning wheel - the behavior that's inefficient, fairly indeterministic and therefore one that should be strongly discouraged. We're employing a graceful approach to prefixing the public API with `DEPRECATED_` first for better discoverability and ease of self service for product teams should they still use that legacy API. If everything goes smoothly, we intend to remove all the deprecated API references in the next release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13284
Reviewed By: pdillinger
Differential Revision: D67981502
Pulled By: mszeszko-meta
fbshipit-source-id: adc7fe5cf4e2180bcfd21878b8f78f3fb6ead355
Summary:
The warm storage crash test sometimes fails due to the cleanup command failing if the db_stress exited successfully and we already cleaned up. This results in false alarms. Don't treat a cleanup command failure as crash test failure.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13287
Reviewed By: archang19
Differential Revision: D68023398
Pulled By: anand1976
fbshipit-source-id: f95fff030a5ea8eb7d2dfb248d08d7876e2de2b2
Summary:
As advertised and recommended by original authors comment, we're removing the now-outdated special handling logic for bloom filters perf regression (timing ~release 7.0.X). I decided to keep the `CompatibilityName` as-is since 1) it's publicly exposed API and 2) it's generally useful to have a dedicated name used for identifying whether a filter on disk is readable by the FilterPolicy.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13277
Test Plan:
'Dead code' / tech debt. As a smoke test, I manually run a similar benchmark to the one in https://github.com/facebook/rocksdb/pull/9736, with ./db_bench built pre and post change.
**Generate DB:**
```hcl
./db_bench -db=/dev/shm/rocksdb.9.11 -bloom_bits=10 -cache_index_and_filter_blocks=1 -benchmarks=fillrandom -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0
```
**Before removing the 'if' block:**
```hcl
./db_bench -db=/dev/shm/rocksdb.9.11 -use_existing_db -readonly -bloom_bits=10 -benchmarks=readrandom -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -duration=10 2>&1 | grep micros/op
readrandom : 17.216 micros/op 58085 ops/sec 10.002 seconds 580999 operations; 4.1 MB/s (367256 of 580999 found)
```
**After removing the 'if' block:**
```hcl
./db_bench -db=/dev/shm/rocksdb.9.11 -use_existing_db -readonly -bloom_bits=10 -benchmarks=readrandom -num=10000000 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -duration=10 2>&1 | grep micros/op
readrandom : 16.776 micros/op 59607 ops/sec 10.015 seconds 596999 operations; 4.2 MB/s (377846 of 596999 found)
```
Reviewed By: jaykorean, pdillinger
Differential Revision: D67908020
Pulled By: mszeszko-meta
fbshipit-source-id: b904b8eaf9d106f0b47e4ff175242795ac1c5e73
Summary:
In https://github.com/facebook/rocksdb/pull/13177, I discussed an unsigned integer overflow issue that affects compaction reads inside `FilePrefetchBuffer` when we attempt to enable the file system buffer reuse optimization. In that PR, I disabled the optimization whenever `for_compaction` was `true` to eliminate the source of the bug.
**This PR safely re-enables the optimization when `for_compaction` is `true`.** We need to properly set the overlap buffer through `PrefetchInternal` rather than simply calling `Prefetch`. `Prefetch` assumes `num_buffers_` is 1 (i.e. async IO is disabled), so historically it did not have any overlap buffer logic. What ends up happening (with the old bug) is that, when we try to reuse the file system provided buffer, inside the `Prefetch` method, we read the remaining missing data. However, since we do not do any `RefitTail` method when `use_fs_buffer` is true, normally we would rely on copying the partial relevant data into an overlap buffer. That overlap buffer logic was missing, so the final main buffer ends up storing data from an offset that is greater than the requested offset, and we effectively end up "throwing away" part of the requested data.
**This PR also unifies the prefetching logic for compaction and non-compaction reads:**
- The same readahead size is used. Previously, we read only `std::max(n, readahead_size_)` bytes for compaction reads, rather than `n + readahead_size_` bytes
- The stats for `PREFETCH_HITS` and `PREFETCH_BYTES_USEFUL` are tracked for both. Previously, they were only tracked for non-compaction reads.
These two small changes should help reduce some of the cognitive load required to understand the codebase. The test suite also became easier to maintain. We could not come up with good reasons why the logic for the readahead size and stats should be different for compaction reads.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13187
Test Plan:
I removed the temporary test case from https://github.com/facebook/rocksdb/issues/13200 and incorporated the same test cases into my updated parameterized test case, which tests the valid combinations between `use_async_prefetch` and `for_compaction`.
I went further and added a randomized test case that will simply try to hit `assert`ion failures and catch any missing areas in the logic.
I also added a test case for compaction reads _without_ the file system buffer reuse optimization. I am thinking that it may be valuable to make a future PR that unifies a lot of these prefetch tests and parametrizes as much of them as possible. This way we can avoid writing duplicate tests and just look over different parameters for async IO, direct IO, file system buffer reuse, and `for_compaction`.
Reviewed By: anand1976
Differential Revision: D66903373
Pulled By: archang19
fbshipit-source-id: 351b56abea2f0ec146b83e3d8065ccc69d40405d
Summary:
This option has been officially deprecated in 5.4.0. We're removing all the references to `random_access_max_buffer_size`, related rules and all the clients wrappers. As a part of this refactoring, we're also getting rid of the `options-1-false` (and consequently its' `multiple-conds-all-false` corresponding rule), as condition would not make much sense anymore without the bounding RA max buffer size limit. Motivated by ongoing tech debt reduction effort.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13278
Test Plan: Validated that internal users do not rely on this long-gone option in their workflows.
Reviewed By: jaykorean
Differential Revision: D67909674
Pulled By: mszeszko-meta
fbshipit-source-id: 8f4b59a4a92b0b32b8b91b71ac318aafc17f1da2
Summary:
The crash test with COERCE_CONTEXT_SWITCH=1 is showing a failure:
```
db_stress: db/seqno_to_time_mapping.cc:480: bool rocksdb::SeqnoToTimeMapping::Append(rocksdb::SequenceNumber, uint64_t): Assertion `false' failed.
```
with `DBImpl::SetOptions()` in the call stack. This assertion and those around it are mostly there for catching systematic problems with recording the mappings, as small imprecisions here and there are not a problem in production. Nevertheless, we need to fix this to maintain the assertions for catching possible future systematic problems.
Because the seqno and time are acquired before holding the DB mutex, there could be a race where T1 acquires latest seqno, T1 acquires latest seqno, T2 acquires unix time, T1 acquires unix time, and entries are not just saved out-of-order, but would represent an inconsistent (time traveling) mapping if they were saved.
We can fix this by getting the seqno and unix times while under the mutex. (Hopefully this is not caused by non-monotonic clock adjustments.)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13279
Test Plan: local run blackbox_crash_test with COERCE_CONTEXT_SWITCH=1. This is not really a production concern, and the conditions are not really reproducible in a unit test after the fix.
Reviewed By: cbi42
Differential Revision: D67923314
Pulled By: pdillinger
fbshipit-source-id: 6bfb6b05d6d449154fbaeb9196eedcfa21fe5ae1
Summary:
Reflect RocksDB DailyOffpeakTimeUTC option in Java API. As is standard for options, there are a number of different places where this option needs to be added: it is an option, a DB option, and it is mutable (can be changed while running).
The new option is a string value. This requires an extension to the internal MutableDBOptions parse code, which received the entire options string from C++ and parses it on the Java side.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13148
Reviewed By: cbi42
Differential Revision: D67870402
Pulled By: jaykorean
fbshipit-source-id: 975af69773206da936d230cbadb5f69a002d92a3
Summary:
The patch is the read-side counterpart of https://github.com/facebook/rocksdb/pull/13197 . It adds support for K-nearest-neighbor vector similarity searches to `FaissIVFIndex`. There are two main pieces to this:
1) `KNNIterator` is an `Iterator` implementation that is returned by `FaissIVFIndex` upon a call to `NewIterator`. `KNNIterator` treats its `Seek` target as a vector embedding and passes it to FAISS along with the number of neighbors requested `k` as well as the number of probes to use (i.e. the number of inverted lists to check). Applications can then use `Next` (and `Prev`) to iterate over the the vectors in the result set. `KNNIterator` exposes the primary keys associated with the result vectors (see below how this is done), while `value` and `columns` are empty. The iterator also supports a property `rocksdb.faiss.ivf.index.distance` that can be used to retrieve the distance/similarity metric for the current result vector.
2) `IteratorAdapter` takes a RocksDB secondary index iterator (see https://github.com/facebook/rocksdb/pull/13257) and adapts it to the interface required by FAISS (`faiss::InvertedListsIterator`), enabling FAISS to read the inverted lists stored in RocksDB. Since FAISS only supports numerical vector ids of type `faiss::idx_t`, `IteratorAdapter` uses `KNNIterator` to assign ephemeral (per-query) ids to the inverted list items read during iteration, which are later mapped back to the original primary keys by `KNNIterator`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13258
Reviewed By: jaykorean
Differential Revision: D67684898
fbshipit-source-id: 5b5c4c438deb86b35d5d45262ce290caee083bca
Summary:
To resolve a crash test failure in
`FlushJob::GetPrecludeLastLevelMinSeqno()`
To fix this properly, I will work on ensuring that (a) FlushJob is created with a consistent view on mutable options and seqno_to_time_mapping (from a single SuperVersion) and (b) SuperVersions always have a non-null seqno_to_time_mapping when a relevant option is set.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13269
Test Plan: watch crash test
Reviewed By: ltamasi
Differential Revision: D67843008
Pulled By: pdillinger
fbshipit-source-id: cedbac4b2255398eefade46240c5481b57a98b1e
Summary:
The primary goal of this change was to support full dynamic mutability of options `preclude_last_level_data_seconds` and `preserve_internal_time_seconds`, which was challenging because of subtle design holes referenced from https://github.com/facebook/rocksdb/issues/13124.
The fix is, in a sense, "doubling down" on the idea of write-time-based tiering, by simplifying the output level decision with a single sequence number threshold. This approach has some advantages:
* Allows option mutability in presence of long snapshots (or UDT)
* Simpler to believe correct because there's no special treatment for range tombstones, and output level assignment does not affect sequence number assignment to the entries (which takes some care to avoid circular dependency; see CompactionIterator stuff below).
* Avoids extra key comparisons, in `WithinPenultimateLevelOutputRange()`, in relevant compactions (more CPU efficient, though untested).
There are two big pieces/changes to enable this simplification to a single `penultimate_after_seqno_` threshold:
* Allow range tombstones to be sent to either output level, based on sequence number.
* Use sequence numbers instead of range checks to avoid data in the last level from moving to penultimate level outside of the permissable range on that level (due to compaction selecting wider range in the later input level, which is the normal output level). With this change, data can only move "back up the LSM" when entire sorted runs are selected for comapction.
Possible disadvantages:
* Extra CPU to iterate over range tombstones in relevant compactions *twice* instead of once. However, work loads with lots of range tombstones relative to other entries should be rare.
* Data might not migrate back up the LSM tree on option changes as aggressively or consistently. This should a a rare concern, however, especially for universal compaction where selecting full sorted runs is normal compaction.
* This approach is arguably "further away from" a design that allows for other kinds of output level placement decisions, such as range-based input data hotness. However, properly handling range tombstones with such policies will likely require flexible placement into outputs, as this change introduces.
Additional details:
* For good code abstraction, separate CompactionIterator from the concern of where to place compaction outputs. CompactionIterator is supposed to provide a stream of entries, including the "best" sequence number we can assign to those entries. If it's safe and proper to zero out a sequence number, the placement of entries to outputs should deal with that safely rather than having complex inter-dependency between sequence number assignment and placement. To achieve this, we migrate all the compaction output placement logic that was in CompactionIterator to CompactionJob and similar. This unfortunately renders some unit tests (PerKeyPlacementCompIteratorTest) depending on the bad abstraction as obsolete, but tiered_compaction_test has pretty good coverage overall, catching many issues during this development.
Intended follow-up:
* See FIXME items in tiered_compaction_test
* More testing / validation / support for tiering + UDT
* Consider generalizing this work to split results at other levels as appropriate based on stats (auto-tuning essentially). Allowing only the last level to be cold is limiting.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13256
Test Plan: tests were added in previous changes (https://github.com/facebook/rocksdb/issues/13244#13124), and updated here to reflect correct operation (with some known problems for leveled compaction)
Reviewed By: cbi42
Differential Revision: D67683210
Pulled By: pdillinger
fbshipit-source-id: ca3f2bbc2fcc6891516a2a4220f1b0da09af5ade
Summary:
The RocksDB backup engine code currently derives the IO buffer size based on the following criteria:
1. If specified, use the rate limiter burst size
2. Otherwise, use the default size (5 MiB)
We want to be able to explicitly choose the IO size based on the storage backend. We want the new criteria to be:
1. If specified, use the size in `BackupEngineOptions`
2. If specified, use the rate limiter burst size
3. Otherwise, use the default size (5 MiB)
This PR adds a new option called `io_buffer_size` to `BackupEngineOptions` and updates the logic used to set the buffer size.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13236
Test Plan:
I added a separate unit test and verified that we can either use the `io_buffer_size`, rate limiter burst size, or the default size.
I decided to use a `TEST_SYNC_POINT_CALLBACK`. I considered the alternative of updating the `Read` implementation of `DummySequentialFile` / `CheckIOOptsSequentialFile` to check the value of `n`. However, that would have considerably complicated the whole test code, and we also do not need to be checking for this in every single test case. I think the `TEST_SYNC_POINT_CALLBACK` turned out to be quite elegant.
Reviewed By: sushilpa
Differential Revision: D67765000
Pulled By: archang19
fbshipit-source-id: 2122fab7379335de44ba4423af47aa0563635688
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13257
The patch adds a new API `NewIterator` to `SecondaryIndex`, which should return an iterator that can be used by applications to query the index. This method takes a `ReadOptions` structure, which can be used by applications to provide (implementation-specific) query parameters to the index, and an underlying iterator, which should be an iterator over the index's secondary column family, and is expected to be leveraged by the returned iterator to read the actual secondary index entries. (Providing the underlying iterator this way enables querying the index as of a specific point in time for example.)
Querying the index can be performed by calling the returned iterator's `Seek` API with a search target, and then using `Next` (and potentially `Prev`) to iterate through the matching index entries. `SeekToFirst`, `SeekToLast`, and `SeekForPrev` are not expected to be supported by the iterator. The iterator should expose primary keys, that is, the secondary key prefix should be stripped from the index entries.
The exact semantics of the returned iterator depend on the index and are implementation-specific. For simple indices, the search target might be a primary column value, and the iterator might return all primary keys that have the given column value. (This behavior can be achieved using the new class `SecondaryIndexIterator`.) However, other semantics are also possible: for vector indices, the search target might be a vector, and the iterator might return similar vectors from the index. (This will be implemented for `FaissIVFIndex` in a subsequent patch.)
Reviewed By: jaykorean
Differential Revision: D67684777
fbshipit-source-id: 59bc33919405a3e9e316a1fa4790c1708788eb85
Summary:
**Context/Summary:**
After https://github.com/facebook/rocksdb/pull/13226, our crash test appears to find a WAL hole caused by mishandling of an injected error during writing the buffer in writable file writer into the underlying log file. It will take some time for me to fully root-cause and fix it. Before then, let's disable this combination.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13263
Test Plan: Monitor crash test
Reviewed By: ltamasi
Differential Revision: D67755485
Pulled By: hx235
fbshipit-source-id: 5f7bb422f7722c2696872232b1fed8ffa5c0f4c3
Summary:
we saw this [assertion](https://github.com/facebook/rocksdb/blob/02b4197544f758bdf84d80fe9319238611848c48/db/error_handler.cc#L576) failing in crash test. The LOG shows that there's a call to SetOptions() concurrent to ResumeImpl(). It's possible that while waiting for error recovery flush (with mutex released), SetOptions() failed to write to MANIFEST and added a file to be quarantined. This triggered the assertion failure when ResumeImpl() calls ClearBGError().
This PR fixes the issue by setting background error when SetOptions() fails to write to MANIFEST.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13251
Test Plan: monitor future crash test failures.
Reviewed By: hx235
Differential Revision: D67660106
Pulled By: cbi42
fbshipit-source-id: 1b52bb23005c4b544f8f9bceefd3b9dcbaf0edfa
Summary:
The patch tweaks the new `SecondaryIndex` interface a bit by removing the `primary_key` parameter of `GetSecondaryKeyPrefix` and `GetSecondaryValue`. This parameter is currently unused by existing implementations and it actually does not make sense to have the secondary index prefix depend on the primary key since it would lead to potential chicken-and-egg problems at query time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13207
Reviewed By: jaykorean
Differential Revision: D67184936
fbshipit-source-id: 5707a35225a0160132e5e87e9fe6c36bee5eada1
Summary:
**Context/Summary:**
This PR provides a new Options `track_and_verify_wals` to detect and handle WAL hole where new WAL data presents while some old WAL data is missing as well as db opened with no WAL. It's for https://github.com/facebook/rocksdb/issues/12488.
It's intended to be a future replacement to `track_and_verify_wals_in_manifest` for its simplicity, better handling of WAL hole in `WALRecoveryMode::kPointInTimeRecovery` and potentials to cover more scenarios for `WALRecoveryMode::kTolerateCorruptedTailRecords/kAbsoluteConsistency`(in future PRs).
The verification is done in `LogReader::MaybeVerifyPredecessorWALInfo()` and tracking is done in `log::Writer::MaybeAddPredecessorWALInfo()`. This PR also groups common utilities in `log::Writer` into functions `MaybeHandleSeenFileWriterError()`, `MaybeSwitchToNewBlock()` to avoid adding redundant code
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13226
Test Plan:
- New UT
- Integrate into existing UT
- Intense rehearsal stress/crash test
- db bench
- The only potential performance implication it has is to the write path since now we keep track of the last seqno recorded in the WAL in `log::Writer`. Below benchmark show no regression.
```
./db_bench --benchmarks=fillrandom[-X3] --num=2500000 --db=/dev/shm/db_bench_new --disable_auto_compactions=1 --threads=1 --enable_pipelined_write=0 --disable_wal=0 --track_and_verify_wals=1
Pre
fillrandom [AVG 3 runs] : 310517 (± 5641) ops/sec; 34.4 (± 0.6) MB/sec
fillrandom [MEDIAN 3 runs] : 308848 ops/sec; 34.2 MB/sec
Post
fillrandom [AVG 3 runs] : 311469 (± 4096) ops/sec; 34.5 (± 0.5) MB/sec
fillrandom [MEDIAN 3 runs] : 311961 ops/sec; 34.5 MB/sec
```
Reviewed By: pdillinger
Differential Revision: D67550260
Pulled By: hx235
fbshipit-source-id: 623e29bbe293ef03a45c20c348f84c8cb5bdaf91
Summary:
**Context/Summary:**
This is to solve https://github.com/facebook/rocksdb/issues/12152. We persist the largest flushed seqno before crash just like how we persist the ExpectedState. And we verify the db lates seqno after recovery is no smaller than this flushed seqno.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12787
Test Plan:
- Manually observe that the persisted sequence after flush completion is used to verify db's latest sequence
- python3 tools/db_crashtest.py --simple blackbox --interval=30
- CI
Reviewed By: archang19
Differential Revision: D58860150
Pulled By: hx235
fbshipit-source-id: 99cb4403964d0737908855f92af7327867079e3e
Summary:
* Expand RangeTombstoneSnapshotMigrateFromLast in tiered_compaction_test (originally from https://github.com/facebook/rocksdb/issues/13124) to reproduce a failure in universal compaciton (as well as leveled), when a specific part of the test is uncommented.
* Small refactoring to eliminate unnecessary fields in SubcompactionState. Adding a bool parameter to SubcompactionState::AddToOutput here will make more sense in the next PR (which I'm trying to keep
from getting too big).
* Improve debuggability and performance of some other tests
* Remove accidentally committed test "BlahPrecludeLastLevel" which was a temporary copy of CompactionServiceTest.PrecludeLastLevel
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13244
Test Plan: existing tests, updated/expanded tests
Reviewed By: cbi42
Differential Revision: D67605076
Pulled By: pdillinger
fbshipit-source-id: 9be83c2173f77545b5fe17ff9dc67db497c7afc9
Summary:
Followup to https://github.com/facebook/rocksdb/pull/13228. This fix is not a critical one in a sense that `else`-branch is only supposed to act as a guard just in case when new work item type is being introduced, scheduled but not handled. However, we're in control of the work item types and currently we only support a single one (which has appropriate handling logic to it).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13238
Reviewed By: pdillinger
Differential Revision: D67512001
Pulled By: mszeszko-meta
fbshipit-source-id: 71e74b3dac388882dd3757871f500c334667fbd1
Summary:
This test assertion was added in https://github.com/facebook/rocksdb/issues/13219. It checks the concurrent write thread's wait time is not longer than the file ingestion thread's write blocking time since the former entered the write thread after the blocking already started in the test. This test runs into flakiness like this:
```db/external_sst_file_basic_test.cc:300: Failure
Expected: (perf_context.file_ingestion_blocking_live_writes_nanos) > (write_thread_perf_context->write_thread_wait_nanos), actual: 166210 vs 279681
```
In reality the write thread is yielding starting with a 1 micro period and then every 100 micros: https://github.com/facebook/rocksdb/blob/54b614de5bd3e26d332b85557d44bde86b2a2e87/db/write_thread.cc#L68-L70
So this 113 micros errors is within this margin
This fix the test with just removing this assertion. The other assertion `ASSERT_GT(write_thread_perf_context->write_thread_wait_nanos, 0)` should be sufficient for the test's purpose.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13241
Reviewed By: hx235
Differential Revision: D67526804
Pulled By: jowlyzhang
fbshipit-source-id: 23ee9771247e4c13444054a1e86ad9293902cb56
Summary:
* Simplify some testing callbacks for tiered_compaction_test ahead of some significant functional updates.
* Refactor CompactionJob::Prepare() for sharing with CompactionServiceCompactionJob. This is a minor functional change in computing preserve/preclude sequence numbers for remote compaction, but it is a start toward support for tiered storage with remote compaction. A test is added that is only partly working but does check that outputs are being split (just not to the correct levels).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13230
Test Plan: mostly test changes and additions. Arguably makes tiered storage + remote compaction MORE broken as a step toward supporting it.
Reviewed By: jaykorean
Differential Revision: D67493682
Pulled By: pdillinger
fbshipit-source-id: fd6db74e08ef0e4fc7fdd599ff8555aab0c8ddc4
Summary:
`DBErrorHandlingFSTest.AtomicFlushNoSpaceError` is flaky due to seg fault during error recovery:
```
...
frame https://github.com/facebook/rocksdb/issues/5: 0x00007f0b3ea0a9d6 librocksdb.so.9.10`rocksdb::VersionSet::GetObsoleteFiles(std::vector<rocksdb::ObsoleteFileInfo, std::allocator<rocksdb::ObsoleteFileInfo>>*, std::vector<rocksdb::ObsoleteBlobFileInfo, std::allocator<rocksdb::ObsoleteBlobFileInfo>>*, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char>>>>*, unsigned long) [inlined] std::vector<rocksdb::ObsoleteFileInfo, std::allocator<rocksdb::ObsoleteFileInfo>>::begin(this=<unavailable>) at stl_vector.h:812:16
frame https://github.com/facebook/rocksdb/issues/6: 0x00007f0b3ea0a9d6 librocksdb.so.9.10`rocksdb::VersionSet::GetObsoleteFiles(this=0x0000000000000000, files=size=0, blob_files=size=0, manifest_filenames=size=0, min_pending_output=18446744073709551615) at version_set.cc:7258:18
frame https://github.com/facebook/rocksdb/issues/7: 0x00007f0b3e8ccbc0 librocksdb.so.9.10`rocksdb::DBImpl::FindObsoleteFiles(this=<unavailable>, job_context=<unavailable>, force=<unavailable>, no_full_scan=<unavailable>) at db_impl_files.cc:162:30
frame https://github.com/facebook/rocksdb/issues/8: 0x00007f0b3e85e698 librocksdb.so.9.10`rocksdb::DBImpl::ResumeImpl(this=<unavailable>, context=<unavailable>) at db_impl.cc:434:20
frame https://github.com/facebook/rocksdb/issues/9: 0x00007f0b3e921516 librocksdb.so.9.10`rocksdb::ErrorHandler::RecoverFromBGError(this=<unavailable>, is_manual=<unavailable>) at error_handler.cc:632:46
```
I suspect this is due to DB being destructed and reopened during recovery. Specifically, the [ClearBGError() call](https://github.com/facebook/rocksdb/blob/c72e79a262bf696faf5f8becabf92374fc14b464/db/db_impl/db_impl.cc#L425) can release and reacquire mutex, and DB can be closed during this time. So it's not safe to access DB state after ClearBGError(). There was a similar story in https://github.com/facebook/rocksdb/issues/9496. [Moving the obsolete files logic after ClearBGError()](https://github.com/facebook/rocksdb/pull/11955) probably makes the seg fault more easily triggered.
This PR updates `ClearBGError()` to guarantee that db close cannot finish until the method is returned and the mutex is released. So that we can safely access DB state after calling it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13234
Test Plan: I could not trigger the seg fault locally, will just monitor future test failures.
Reviewed By: jowlyzhang
Differential Revision: D67476836
Pulled By: cbi42
fbshipit-source-id: dfb3e9ccd4eb3d43fc596ec10e4052861eeec002
Summary:
This change refactors existing `CopyOrCreateWorkItem` async task definition to a more generic one (`WorkItem`) with an assigned `type` indicative of intended action. This would allow us to reuse existing, battle-tested async tasks initialization code to handle wider range of incoming use cases in B/R space.
### Motivation
Historically, the two main use cases for `BackupEngineImpl`'s async work items were either creating a file in backup workflow or copying files in restore workflow. However, as we're now exploring opportunities in incremental restore (and potentially speeding up backup verification), we need the work item abstraction to be capable of processing different workflow types concurrently (computing checksum comes to mind).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13228
Test Plan: Since this is purely cosmetic change where behavior remains intact, existing test collateral will suffice.
Reviewed By: pdillinger
Differential Revision: D67441210
Pulled By: mszeszko-meta
fbshipit-source-id: 78803e8cf3cf40b9d81831fac3a99193e1a30ef0
Summary:
As titled. And also added some documentation for an approach to name perf context metrics that can help identify the starting `PerfLevel` that enables collecting it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13219
Test Plan: Unit test
Reviewed By: hx235
Differential Revision: D67362022
Pulled By: jowlyzhang
fbshipit-source-id: 7ed1bb475b5497961612d4e331600609da42074b
Summary:
To set up for splitting range deletes between penultimate and last level with per-key-placement compaction. This will solve some issues in combining RangeDelete+snapshot+mutable preclude_last, and probably also RangeDelete+UDT+preclude_last
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13231
Test Plan: existing tests
Reviewed By: cbi42
Differential Revision: D67481038
Pulled By: pdillinger
fbshipit-source-id: 597f0c991e4d7eae73b36b36aad493c2d2a15f24
Summary:
Originally I was trying to update `build-linux-clang10-mini-tsan` to actually use `clang10` (as the name implied). https://github.com/facebook/rocksdb/issues/13220 was supposed to also update this configuration, but I did not see that we had a definition for `build-linux-clang10-mini-tsan` in both `config.yml` and `pr-jobs.yml`. I was wondering why I could not see my changes reflected in the CI checks after merging. After I updated `pr-jobs.yml` for this PR, I found that the CI check started failing https://github.com/facebook/rocksdb/actions/runs/12417441052/job/34668411263?pr=13232. I don't think it makes sense for me to tackle looking into all the TSAN warnings being reported in `clang10` (at least in this PR), so for now I have updated the name of the PR job to accurately reflect the command that is being run.
This PR also gets rid of the entire `.circleci` folder, which I think is the more significant change.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13232
Test Plan: Existing CI check is unchanged
Reviewed By: pdillinger
Differential Revision: D67462454
Pulled By: archang19
fbshipit-source-id: f1aabfe4c8793616d6cbaae36fdf007319bf7ab2
Summary:
... which is the default for CentOS 9 and Ubuntu 24, the latter of which is now available in GitHub Actions. Relevant CI job updated.
Re-formatted all cc|c|h files except in third-party/, using
```
clang-format -i `git ls-files | grep -E '[.](cc|c|h)$' | grep -v third-party/`
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13233
Test Plan: CI
Reviewed By: jaykorean, archang19
Differential Revision: D67461638
Pulled By: pdillinger
fbshipit-source-id: 0c9ac21a3f5eea6f5ade68bb6af7b6ba16c8b301
Summary:
I found this mismatch between the CI job title and the actual command ran incidentally while trying to work on https://github.com/facebook/rocksdb/issues/13213.
`build-linux-clang10-mini-tsan` was added in https://github.com/facebook/rocksdb/issues/7122 with `clang-10`.
In https://github.com/facebook/rocksdb/issues/10496 it was changed to use `clang-13` but the name was not also updated. I do not know what the author's intent was, but given that `build-linux-clang10-mini-tsan` is right next to`build-linux-clang10-ubsan` and `build-linux-clang10-asan`, I think it is more likely we originally intended to use `clang-10`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13220
Test Plan: I think we need to wait for the next set of CI checks after this PR is merged, since I don't see my changes incorporated into this PR's `build-linux-clang10-mini-tsan` check.
Reviewed By: hx235
Differential Revision: D67407034
Pulled By: archang19
fbshipit-source-id: 9c22b6c6c330a367920eb3d4a387f37b760d722c
Summary:
This is a follow up to https://github.com/facebook/rocksdb/issues/13189. As mentioned in the description in the previous PR, to guard against similar bugs in the future, we should update our test implementations to reflect the real-world assumptions that we can make about `fs_scratch` when we issue reads with the filesystem buffer reuse optimization. The current test implementations reinforce the misconception that `fs_scratch` points to the same place as `result.data()` (i.e. to the start of the valid data buffer for the read result). `fs_scratch` can point to any arbitrary data structure, but for our purposes, I think we achieve what we want if we just have it point to a `Slice` which wraps the underlying result buffer inside one of its class variables.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13195
Test Plan: Existing unit tests test the same functionality but in an improved way with this change.
Reviewed By: hx235
Differential Revision: D66896380
Pulled By: archang19
fbshipit-source-id: 377e67ec70427716f2b7b7388d99b78003c01eb0
Summary:
In https://github.com/facebook/rocksdb/pull/13118#discussion_r1842848359, we decided to make a separate follow-up PR that refactors `FilePrefetchBuffer` to determine `use_fs_buffer` once at construction time.
The change would have involved passing in the `RandomAccessFileReader*` directly to the constructor, and using that to determine `use_fs_buffer`. This would avoid repeatedly calling `UseFSBuffer(RandomAccessFileReader* reader)` during the actual prefetch requests.
I started working on this refactoring change but ran into issues with these 2 files, which used `GetOrCreatePrefetchBuffer`
- https://github.com/facebook/rocksdb/blob/main/db/compaction/compaction_iterator.cc
- https://github.com/facebook/rocksdb/blob/main/db/merge_helper.cc
As I explained in the added code comments, sometimes the `RandomAccessFileReader*` is not available when we construct the `FilePrefetchBuffer`, so although it is not the most elegant, I think right now it makes sense to pass in the `reader` into the `Prefetch` / `PrefetchAsync` / `TryReadFromCache` calls. Maybe there is a workaround but I don't think the refactor would be worth it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13159
Test Plan: N/A (comments)
Reviewed By: anand1976
Differential Revision: D66473731
Pulled By: archang19
fbshipit-source-id: ce3473694c2cd82513da1a76ad5995afa5bc9cfa
Summary:
I saw these compiler warnings while preparing for the 9.10 release:
```cpp
'~CompactOnDeletionCollectorFactory' overrides a destructor but is not marked 'override' [-Werror,-Wsuggest-destructor-override]
'~CompactForTieringCollectorFactory' overrides a destructor but is not marked 'override' [-Werror,-Wsuggest-destructor-override]
```
This code is from a while ago so I assume that this CI check has been failing for quite some time. We should still clean this up to avoid confusion in the future.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13212
Test Plan: Existing CI checks should pass, and we should not see this CI check failure the next time we try to make a release/patch.
Reviewed By: jaykorean
Differential Revision: D67287794
Pulled By: archang19
fbshipit-source-id: a11230a919c0b7ef21a7219bf05f567d3d44b2d1
Summary:
I had an extra comma after `9.9.fb` when I updated `tools/check_format_compatible.sh` in https://github.com/facebook/rocksdb/issues/13210. This caused the nightly builds to start failing https://github.com/facebook/rocksdb/actions/workflows/nightly.yml on the `build-format-compatible` step. The error message is
```
2024-12-14T11:55:23.3413129Z == Building 9.9.fb, debug
2024-12-14T11:55:23.3427208Z fatal: ambiguous argument '_tmp_origin/9.9.fb,': unknown revision or path not in the working tree.
```
Notice the extra comma after `9.9.fb`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13211
Test Plan: The nightly builds should start passing again.
Reviewed By: jowlyzhang
Differential Revision: D67286484
Pulled By: archang19
fbshipit-source-id: 57a754c88af004ee879d9c9f82819b3c410a66a9
Summary:
**Context/Summary:**
`DBImpl::RecoverLogFiles()` has ~500 lines of code with nested loops and various return/continue/break statements. This becomes too difficult to understand and make change for the upcoming wal hole detection.
This PR broke it into multiple smaller functions and left a couple FIXME where the EXISTING ugly code is too complicated to clean up right now. Most of them are copy-and-paste excepts for `ProcessLogRecord()` that needs some thoughts into how to translate existing behaviors of `break`, `continue`, `return non-ok status`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13184
Test Plan: Pass existing test
Reviewed By: jowlyzhang
Differential Revision: D66799568
Pulled By: hx235
fbshipit-source-id: d15617a47ee2d1c02652f1fd8336e82a2c5434b1
Summary:
I followed the release instructions and referenced https://github.com/facebook/rocksdb/pull/13146
1. HISOTRY update
2. version.h
3. Format compatability test
4. Folly Git hash
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13210
Test Plan: CI
Reviewed By: pdillinger
Differential Revision: D67210980
Pulled By: archang19
fbshipit-source-id: cfbc02c643aeae19453c8c36d03d93478ea81c4e
Summary:
expand the test coverage to the more comprehensive no_batched_ops_stress. Small refactoring in db_crashtest.py.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13203
Test Plan: ran a couple stress test jobs internally: https://fburl.com/sandcastle/nohosh7i
Reviewed By: jowlyzhang
Differential Revision: D67057497
Pulled By: cbi42
fbshipit-source-id: eccc033f3ae3dbd20729cd8f1f8f8d8b7c2cd057
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13197
The patch adds initial support for backing FAISS's inverted file based indices with data stored in RocksDB. It introduces a `SecondaryIndex` implementation called `FaissIVFIndex` which takes ownership of a `faiss::IndexIVF` object. During indexing, `FaissIVFIndex` treats the original value of the specified primary column as an embedding vector, and passes it to the provided FAISS index object to perform quantization. It replaces the original embedding vector with the result of the coarse quantizer (i.e. the inverted list id), and puts the result of the fine quantizer (if any) into the secondary index value. Note that this patch is only one half of the equation; it provides a way of storing FAISS inverted lists in RocksDB but there is currently no retrieval/search support (this will be a follow-up change). Also, the integration currently works only with our internal Buck build. I plan to add support for `cmake` / `make` based builds similarly to how we handle Folly.
Reviewed By: jowlyzhang
Differential Revision: D66907065
fbshipit-source-id: 63fdf29895d5feeffc230254a7ddfb0aac050967
Summary:
This is a follow up to https://github.com/facebook/rocksdb/issues/13177, which was supposed to disable the file system buffer optimization for compaction reads. However, it did not work as expected because I did not pass through `use_fs_buffer` to the `Read` method, which also calls `UseFSBuffer`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13200
Test Plan:
I added simple tests to verify we do not hit the overflow issue when we are doing compaction prefetches.
```
./prefetch_test --gtest_filter="*FSBufferPrefetchForCompaction*"
```
Of course I will be looking through the warm storage crash test logs as well once the change is merged.
Reviewed By: anand1976
Differential Revision: D66996079
Pulled By: archang19
fbshipit-source-id: b4d9254f1354ccfc53a307174de5f2388b7e5474
Summary:
https://github.com/facebook/rocksdb/issues/13182 successfully fixed the heap `use-after-free` issue.
However, there was one additional error I found while looking through the warm storage crash test logs. There are repeated (though infrequent) unsigned pointer arithmetic overflow errors that look like this:
```cpp
file_prefetch_buffer.cc:860:46: runtime error: addition of unsigned offset to 0x7f282001880f overflowed to 0x7f2820017667
```
It took me a while to figure it out, but I was finally able to reproduce the issue locally. It turns out the issue is when we call `TryReadFromCache` with `for_compaction` set to `true`. The default value for `for_compaction` is `false`, and this was not covered in the unit tests written for https://github.com/facebook/rocksdb/issues/13118.
When I run the same unit tests with `for_compaction` set to `true`, I am able to break this assertion that I added at the end of `TryReadFromCacheUntracked`:
```cpp
assert(buf->offset_ <= offset);
```
If `buf->offset_` is greater than `offset`, then that explains the overflow we get in the following lines:
```cpp
uint64_t offset_in_buffer = offset - buf->offset_;
*result = Slice(buf->buffer_.BufferStart() + offset_in_buffer, n);
```
I will have another PR out that fixes the issue and enables the optimization when `for_compaction` is set to `true`. I will need to add some overlap buffer logic, similar to what I have inside `PrefetchInternal`. For now, since I have confirmed that there is indeed a bug, we should disable the optimization where needed. It will take me some time to implement the fix and write new test cases.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13177
Test Plan: I kept the existing unit tests which test the file system buffer reuse code when `for_compaction` is `false`. I expect that the warm storage crash test logs will no longer show the integer overflow issue once we merge this PR.
Reviewed By: anand1976
Differential Revision: D66721857
Pulled By: archang19
fbshipit-source-id: 22d523646f969a7a0ccbbea73f63c32601f1179a
Summary:
This is a followup to https://github.com/facebook/rocksdb/issues/13190. We're patching the targets generating script to construct `BUCK` file instead of deprecated `TARGETS` file + adding safety checks to ensure that `BUCK` file does not go missing (either as a direct renaming / removal OR as a modification to buckfier's script(s)).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13196
Test Plan:
1. Manually verify 'Compare buckify output' step produces expected results (vs previously soft-failed one [here](https://github.com/facebook/rocksdb/actions/runs/12202756083/job/34044173548?pr=13178)).
2. Manually test following scenarios (for both of which we expect the buckifier script to fail):
-> Simulate removing `BUCK` file via commit
-> Simulate buckifier script removing the `BUCK` file
Reviewed By: pdillinger
Differential Revision: D66903948
Pulled By: mszeszko-meta
fbshipit-source-id: 0f83fd2f87b600981f640ccdbc3a4640974a63d4
Summary:
After https://github.com/facebook/rocksdb/pull/13118 was merged, I did some investigation to see whether the file system buffer reuse code was actually being used.
The good news is that I was able to see from the CPU profiling results that my code is getting invoked through the warm storage stress tests.
The bad news is that most of the time, the optimization is not being used, so we end up going through the regular old `RandomAccessFileReader::Read` path.
Here is the entire function call chain up to `FilePrefetchBuffer::Read`
1. rocksdb::DB::MultiGet
2. rocksdb::DBImpl::MultiGet
3. rocksdb::DBImpl::MultiGetCommon
4. rocksdb::DBImpl::MultiGetImpl
5. rocksdb::Version::MultiGet
6. rocksdb::Version::MultiGetFromSST
7. rocksdb::TableCache::MultiGet
8. rocksdb::TableCache::FindTable
9. rocksdb::TableCache::GetTableReader
10. rocksdb::BlockBasedTableFactory::NewTableReader
11. rocksdb::BlockBasedTable::Open
12. rocksdb::BlockBasedTable::PrefetchTail
13. rocksdb::FilePrefetchBuffer::Prefetch
14. rocksdb::FilePrefetchBuffer::Read
At this point, we split into `rocksdb::RandomAccessFileReader::Read` and
`rocksdb::FilePrefetchBuffer::FSBufferDirectRead`. `FSBufferDirectRead` gets called <3% of the time.
I think the root cause is that the `FileSystem* fs` parameter is not getting passed into the `FilePrefetchBuffer` constructor. When `fs` is `nullptr`, `UseFSBuffer()` will always return `false` and we do not end up calling `FSBufferDirectRead`.
Luckily, it does not seem like there are too many places I need to change. `BlockBasedTable` resets its `prefetch_buffer` in 3 separate places. When it disables the prefetch buffer (2/3 of the instances), we don't care about whether the `fs` parameter is there. This PR is addressing the third instance, where it is not trying to disable the buffer.
Note that there is another method, `PrefetchBufferCollection::GetOrCreatePrefetchBuffer` that creates new `FilePrefetchBuffer`s without the `fs` parameter. This method gets called by `compaction_iterator` and `merge_helper`. I think we can address this in a subsequent PR:
1. Each of these changes effectively "unlocks" the buffer reuse feature. Separating the changes would be helpful when I look at the profiling results again, since I can isolate what impact this PR had on the percentage of time that `rocksdb::FilePrefetchBuffer::FSBufferDirectRead` was invoked.
2. I still need to look into what exactly I would need to changes I need to make to `PrefetchBufferCollection`
3. This code seems to be for blob prefetching in particular, and I don't think it has the biggest ROI anyways.
```cpp
const Status s = blob_fetcher_->FetchBlob(
user_key(), blob_index, prefetch_buffer, &blob_value_, &bytes_read);
```
4. I am not sure if the current benchmark I am using for warm storage exercises this blob prefetching code, so I may need to find another way to assess the performance impact.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13157
Test Plan: The existing unit test coverage guards against obvious bugs. I ran another set of performance tests to confirm there were no regressions in CPU utilization.
Reviewed By: anand1976
Differential Revision: D66464704
Pulled By: archang19
fbshipit-source-id: 260145cfcc05ac46cf2dd77a53a85e8808031dea
Summary:
This change introduces a new, lightweight _experimental_ API that reconstructs the [file # -> file checksum -> file checksum function] 1-1-1 mapping directly from the `MANIFEST` file considered `CURRENT` in scope of specific DB instance at the time. The goal is to provide a cheap alternative to `DB::GetLiveFilesMetaData` that doesn't require opening the database, reconstructing version sets and/or accessing files that are _potentially_ in disaggregated storage.
### Housekeeping:
1. Moved the `GetCurrentManifestPath` out of `version_set` to a new `manifest_ops` file(s) dedicated to manifest related operations.
2. Introduced new `Env::IOActivity::kReadManifest` to better reflect the IO intent in offline file checksum retrieving function.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13178
Test Plan:
Added a unit test comparing the outcome of newly introduced API against the established `GetLiveFilesMetaData`:
```hcl
./db_test2 --gtest_filter="*GetFileChecksumsFromCurrentManifest_CRC32*"
```
Reviewed By: pdillinger
Differential Revision: D66711910
Pulled By: mszeszko-meta
fbshipit-source-id: 57091c550a14ac2e832bf7eea136dab5450e71bc
Summary:
https://github.com/facebook/rocksdb/pull/13182 seems to have resolved the `heap-use-after-free` / `heap-buffer-overflow` issues, but not for the reasons we had in mind.
I believe I have figured out the root cause after doing more thinking / reading into the warm storage code.
**`fs_scratch` cannot be assumed to point to the start of the data buffer. It must be treated as a pointer to any arbitrary object / data structure. As such, we must rely only on result.data().**
I think that part of the reason for the bug was that the comment for `fs_scratch` was
> fs_scratch is a data buffer allocated and provided by underlying FileSystem
which is _extremely misleading_.
To avoid confusion in the future, I have updated the comments related to `FsReadRequest` with some of my learnings and included `WARNING`s in all caps to hopefully steer future engineers aware from the same issue.
In another PR, I will update some of our mock file system test classes that support `FSSupportedOps::kFSBuffer`. The test class implementation also contributed to my confusion, since `fs_scratch` did point to the start of the valid data in those implementations. This cannot and should not be assumed to be true in general, and we should try to guard against potential future bugs by updating those mock implementations.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13189
Test Plan: These are just comments.
Reviewed By: anand1976, hx235
Differential Revision: D66849436
Pulled By: archang19
fbshipit-source-id: c264007647af9cc2a4dfd58dbe7287af86fa2261
The internal and external repositories are out of sync. This Pull Request attempts to brings them back in sync by patching the GitHub repository. Please carefully review this patch. You must disable ShipIt for your project in order to merge this pull request. DO NOT IMPORT this pull request. Instead, merge it directly on GitHub using the MERGE BUTTON. Re-enable ShipIt after merging.
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13180
The patch adds initial support for secondary indices using write-committed transactions. Currently, only the `PutEntity` API is supported; other APIs like `Put` and `Delete` will be added separately. Applications can set up secondary indices using the new configuration option `TransactionDBOptions::secondary_indices`. When secondary indices are enabled, calling `PutEntity` via a (n explicit or implicit) transaction performs the following steps:
1) It retrieves the current value (if any) of the primary key using `GetEntityForUpdate`.
2) If there is an existing primary key-value, it removes any existing secondary index entries using `SingleDelete`. (Note: as a later optimization, we can avoid removing and recreating secondary index entries when neither the secondary key nor the value changes during an update.)
3) It invokes `UpdatePrimaryColumnValue` for all applicable `SecondaryIndex` objects, that is, those for which the primary column family matches the column family from the `PutEntity` call and for which the primary column appears in the new wide-column structure.
4) It writes the new primary key-value. Note that the values of the indexing columns might have been changed in step 3 above.
5) It builds the secondary key-value for each applicable secondary index using `GetSecondaryKeyPrefix` and `GetSecondaryValue`, and writes it to the appropriate secondary column family.
All the above operations are performed as part of the same transaction. The logic uses `SavePoint`s to roll back any earlier operations related to a primary key if a subsequent step fails.
Implementation-wise, the code uses a mixin template `SecondaryIndexMixin` that can inherit from any kind of transaction and use the write APIs and concurrency control mechanisms of the base class to implement the index maintenance logic. The mixin will enable us to later extend secondary indices to optimistic or write-prepared/write-unprepared pessimistic transactions as well.
Reviewed By: jowlyzhang
Differential Revision: D66672931
fbshipit-source-id: cdf6ef9c40dec46d928156bad0a3cc546aa8b887
Summary:
`StartV2()` and `WaitForCompleteV2()` were deprecated and replaced by`Schedule()` and `Wait()` in 9.1.0. This PR removes them from the codebase completely.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13188
Test Plan: CI
Reviewed By: archang19
Differential Revision: D66843687
Pulled By: jaykorean
fbshipit-source-id: f13d05845bf5ac4ae736c105035ca1a4d5a96047
Summary:
add a new transaction option `TransactionOptions::commit_bypass_memtable` that will ingest the transaction into a DB as an immutable memtables, skipping memtable writes during transaction commit. This helps to reduce the blocking time of committing a large transaction, which is mostly spent on memtable writes. The ingestion is done by creating WBWIMemTable using transaction's underlying WBWI, and ingest it as the latest immutable memtable. The feature will be experimental.
Major changes are:
1. write path change to ingest the transaction, mostly in WriteImpl() and IngestWBWI() in db_impl_write.cc.
2. WBWI changes to track some per CF stats like entry count and overwritten single deletion count, and track which keys have overwritten single deletions (see 3.). Per CF stat is used to precompute the number of entries in each WBWIMemTable.
3. WBWIMemTable Iterator changes to emit overwritten single deletions. The motivation is explained in the comment above class WBWIMemTable definition. The rest of the changes in WBWIMemTable are moving the iterator definition around.
Some intended follow ups:
1. support for merge operations
2. stats/logging around this option
3. tests improvement, including stress test support for the more comprehensive no_batched_op_stress.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13144
Test Plan:
* added new unit tests
* enabled in multi_ops_txns_stress test
* Benchmark: applying the change in 8222c0cafc4c6eb3a0d05807f7014b44998acb7a, I tested txn size of 10k and check perf context for write_memtable_time, write_wal_time and key_lock_wait_time(repurposed for transaction unlock time). Though the benchmark result number can be flaky, this shows memtable write time improved a lot (more than 100 times). The benchmark also shows that the remaining commit latency is from transaction unlock.
```
./db_bench --benchmarks=fillrandom --seed=1727376962 --threads=1 --disable_auto_compactions=1 --max_write_buffer_number=100 --min_write_buffer_number_to_merge=100 --writes=100000 --batch_size=10000 --transaction_db=1 --perf_level=4 --enable_pipelined_write=false --commit_bypass_memtable=1
commit_bypass_memtable = false
fillrandom : 3.982 micros/op 251119 ops/sec 0.398 seconds 100000 operations; 27.8 MB/s PERF_CONTEXT:
write_memtable_time = 116950422
write_wal_time = 8535565
txn unlock time = 32979883
commit_bypass_memtable = true
fillrandom : 2.627 micros/op 380559 ops/sec 0.263 seconds 100000 operations; 42.1 MB/s PERF_CONTEXT:
write_memtable_time = 740784
write_wal_time = 11993119
txn unlock time = 21735685
```
Reviewed By: jowlyzhang
Differential Revision: D66307632
Pulled By: cbi42
fbshipit-source-id: 6619af58c4c537aed1f76c4a7e869fb3f5098999
Summary:
[Venice](https://venicedb.org/) is a derived data platform using RocksDB as its storage engine. It is LinkedIn's ML feature store, powering thousands of recommender use cases, including the Feed, Video recommendations, and People You May Know.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13179
Reviewed By: hx235
Differential Revision: D66724729
Pulled By: cbi42
fbshipit-source-id: a027d6664f2924473884a3d5d129748ea1e5fe37
Summary:
This PR is an attempt to address https://github.com/facebook/rocksdb/issues/13118. The warm storage crash tests show use-after-free errors. They do not occur in every single crash test run, but with enough attempts, they are repeatable.
Theory 1:
I am wondering if the `fs_buffer` is being prematurely freed before we take ownership of it. In `SetBuffer`, I was passing in `FSAllocationPtr&& new_buf` rather than `FSAllocationPtr new_buf`. When I pass the parameter as `FSAllocationPtr&& new_buf`, only after the `buf_ = std::move(new_buf);` line is run is ownership transferred from the original `FSAllocationPtr`. But before that I had a line `bufstart_ = reinterpret_cast<char*>(buf_.get());`. So I am hypothesizing that it is possible, under certain race conditions, that between the first `buf_.get()` and the `buf_ = std::move(new_buf);`, the `fs_buffer` was altered, leaving `bufstart_` pointing to some freed memory area.
Theory 2 (from anand1976):
Perhaps we need to set the `bufstart_` based on the `Slice` rather than the `FSAllocationPtr`. This would be more consistent with what we do here https://github.com/facebook/rocksdb/blob/main/table/block_fetcher.cc#L275.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13182
Test Plan: The existing unit tests and CI ensures I am not making anything worse, but I will want to wait and see if the daily crash tests runs still have the same `heap-use-after-free` errors with this change. Alternatively, if we fail the `assert` I just added, then I can make a follow-up PR to return `false` from `TryReadFromCache` whenever we get handed back a `nullptr`.
Reviewed By: anand1976
Differential Revision: D66771852
Pulled By: archang19
fbshipit-source-id: 5b585d86d657ec050a04e892d3b1cf4383f377f9
Summary:
During `FinishCompactionOutputFile()` if there's an IOError, we may end up having the output in memory, but table properties are not populated, because `outputs.UpdateTableProperties();` is called only when `s.ok()` is true.
However, during remote compaction result serialization, we always try to access the `table_properties` which may be null. This was causing a segfault.
We can skip building the output files in the result completely if the status is not ok.
# Unit Test
New test added
```
./compaction_service_test --gtest_filter="*CompactionOutputFileIOError*"
```
Before the fix
```
Received signal 11 (Segmentation fault)
Invoking GDB for stack trace...
https://github.com/facebook/rocksdb/issues/4 0x00000000004708ed in rocksdb::TableProperties::TableProperties (this=0x7fae070fb4e8) at ./include/rocksdb/table_properties.h:212
212 struct TableProperties {
https://github.com/facebook/rocksdb/issues/5 0x00007fae0b195b9e in rocksdb::CompactionServiceOutputFile::CompactionServiceOutputFile (this=0x7fae070fb400, name=..., smallest=0, largest=0, _smallest_internal_key=..., _largest_internal_key=..., _oldest_ancester_time=1733335023, _file_creation_time=1733335026, _epoch_number=1, _file_checksum=..., _file_checksum_func_name=..., _paranoid_hash=0, _marked_for_compaction=false, _unique_id=..., _table_properties=...) at ./db/compaction/compaction_job.h:450
450 table_properties(_table_properties) {}
```
After the fix
```
[ RUN ] CompactionServiceTest.CompactionOutputFileIOError
[ OK ] CompactionServiceTest.CompactionOutputFileIOError (4499 ms)
[----------] 1 test from CompactionServiceTest (4499 ms total)
[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (4499 ms total)
[ PASSED ] 1 test.
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13183
Reviewed By: anand1976
Differential Revision: D66770876
Pulled By: jaykorean
fbshipit-source-id: 63df7c2786ce0353f38a93e493ae4e7b591f4ed9
Summary:
* Test tiered storage FIFO setting `file_temperature_age_thresholds` in crash test, with dynamic mutability.
* Re-organize db_crashtest.py slightly to better handle tiered storage parameters and their interaction with compaction_style and num_levels. I have put most of this logic in the python script so that `db_stress` command lines reflect settings in effect as best as possible.
* Tweak crash test settings for preclude_last_level_data_seconds. This seems to have amplified the possibility of hitting "Corruption: Unsafe to store Seq later" even with universal compaction, which I am working on a fix for. We should also be able to enable tiered+leveled when this is fixed. (TODO / follow-up items)
* Code formatting / small simplifications in db_crashtest.py
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13176
Test Plan:
no production code changes
Kicked off about 24 CI jobs (temporary internal link https://fburl.com/sandcastle/s61rzusr)
Reviewed By: cbi42
Differential Revision: D66674123
Pulled By: pdillinger
fbshipit-source-id: 33dd7f9d291ec4a9516665b4adb998fd9a2b9266
Summary: I missed in the previous diff that this is generated. Let's fix that codegen script
Reviewed By: dtolnay
Differential Revision: D66725403
fbshipit-source-id: ec9fa773c8309040da98677a128c4cb0309542a8
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13165
This diff migrates TARGETS file to BUCK files that are synced for an open source project.
Reviewed By: dtolnay
Differential Revision: D66561335
fbshipit-source-id: 9c91a19ef59a81adc31b763a63134aeef1eb00ed
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13175
The patch is the first step in adding support for secondary indices via the transaction layer. It introduces a new `SecondaryIndex` interface, which enables creating secondary indices over a set of (plain or wide-column) primary key-values to facilitate queries by (column) value instead of key. This interface will be automagically invoked by the transaction logic to add and remove secondary index entries as needed when the application issues write operations for the primary data. Classes deriving from `SecondaryIndex` can implement the methods `GetPrimaryColumn{Family,Name}` and `GetSecondaryColumnFamily` to respectively define the primary column family and wide column to index and the column family to use for the secondary index entries. The format of the secondary index entries can be defined by implementing `GetSecondaryKeyPrefix` and `GetSecondaryValue`. In addition, `UpdatePrimaryColumnValue` can be used to optionally update the value of the indexing column in the primary key-value before it is added to the transaction.
Reviewed By: jowlyzhang
Differential Revision: D66672758
fbshipit-source-id: 0b7441ffff626c13956220e6efc98215303ef57e
Summary:
In buffered IO mode, without checksum calculation for buffered data enabled, try to align writes to the file system on a power of two. This can improve performance, especially on a distributed file system like Warm Storage that does erasure coding and benefits from full stripe writes. We do this by filling up the writable buffer, with a partial append if necessary, before flushing. When checksum calculation for buffered data is enabled, we don't do this since its preferable to not split the data, especially if the caller provides the checksum. We don't guarantee alignment if the caller manually flushes before finishing the file.
Tests:
Add unit tests in file_reader_writer_test and external_sst_file_basic_test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13158
Reviewed By: pdillinger
Differential Revision: D66669367
Pulled By: anand1976
fbshipit-source-id: 6df1b4538bda696e2170515420ee4c3766c83bb8
Summary:
This PR adds the definition for the public APIs for surfacing data write time info. It only contains minimum implementation. The implementations will be in follow ups. I need to sync with customers if these public APIs meet their requirements and are easy to use. And make modifications accordingly before proceeding with implementations.
- `struct DataCollectionUnixWriteTimeInfo` is a struct for the unix write time info for a collection of data
- `DB::GetPropertiesOfTablesForLevels` returns table properties collection per level
- `GetDataCollectionUnixWriteTimeInfoForFile` returns the data write time info for a file.
- `GetDataCollectionUnixWriteTimeInfoForLevels` returns the data write time info for levels.
- The user property names for recording write time stats in the user collected properties are defined.
Follow ups:
Implement collecting the write time related user table properties
Use the data write time info recorded in the table properties to implement these APIs
Test Plan:
No functional change, also follow ups should have tests covering the minimum implementation added in this PR.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13138
No functional change, also follow ups should have tests covering the minimum implementation added in this PR.
Reviewed By: pdillinger
Differential Revision: D65952586
Pulled By: jowlyzhang
fbshipit-source-id: b1ebf61a35005e9ca6b4ecc28c864beb6fb4bc59
Summary:
The compaction will incorrectly drop a key under the following conditions:
1. Open an empty database.
2. Use the `IngestExternalFile` API to ingest an SST file (the global sequence number will be 0).
3. Create a snapshot (the snapshot sequence number will be 0).
4. Trigger compaction; the key in the above SST file will be dropped.
The drop condition is found here: https://github.com/facebook/rocksdb/blob/f20d12adc85ece3e75fb238872959c702c0e5535/db/compaction/compaction_iterator.cc#L875-L878
The condition does not explicitly check if a previous key exists.
Fix: Add a check of `last_sequence != kMaxSequenceNumber` to verify if there is a previous key
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13155
Reviewed By: jowlyzhang
Differential Revision: D66473015
Pulled By: cbi42
fbshipit-source-id: 93a3ec5c103f95e9bb97e3944ba6e752a5394421
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13168
The patch moves `WideColumnSerialization::Find` to `WideColumnsHelper` to facilitate reuse in non-serialization-related contexts. It also generalizes the method to take a range of iterators, and templatizes it on the iterator type to enable using it with both `const` and non-`const` iterators. Finally, it adds an assertion to ensure the method is called with a properly sorted range, which is a precondition for binary search.
Reviewed By: jaykorean
Differential Revision: D66602558
fbshipit-source-id: 841a885af31e183edeb7e3314167c55f8ed53ff1
Summary:
Adding ability to kill mysql queries traversing long lists of tombstones. Outside of mysql where RocksDbThreadYieldAndCheckAbort is not implemented all of this should still be optimized out by the compiler.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13164
Reviewed By: cbi42
Differential Revision: D66556004
Pulled By: george-reynya
fbshipit-source-id: 727875569209cd6d2f29c07f89ecfa641d5ee36f
Summary:
This change aims at increasing general memory safety in scope of selected `/db` files (`db_impl/db_impl.cc`, `dbformat.cc`, `log_reader.cc` and `transaction_log_impl.cc`).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13154
Test Plan:
Verify logging structure & formatting parity by manually running the `/db` related tests exercising respective code paths pre and post change.
Note: As per request, we'll address the `internal_stats.cc` in the followup PR.
Reviewed By: pdillinger
Differential Revision: D66392729
Pulled By: mszeszko-meta
fbshipit-source-id: 107fd11221554721d9c1669a24031be3049afd01
Summary:
`OptionTypeInfo::ParseStruct()` was not honoring `config_options.ignore_unknown_options` when unknown properties are found in the serialized string. This caused a compatibility issue in Remote Compaction. When the worker was updated with RocksDB 9.9, the remote worker started including a new table property, `newest_key_time` (added in PR https://github.com/facebook/rocksdb/issues/13083), in the compaction output files. However, parsing that table property in the serialized compaction result from the primary (running with `9.8`) was returning a non-ok status, even though `config_options.ignore_unknown_options` was `true`.
In this fix, we will ignore unused properties if `config_options.ignore_unknown_options` is set to true.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13152
Test Plan: Unit Test Added
Reviewed By: archang19
Differential Revision: D66374541
Pulled By: jaykorean
fbshipit-source-id: 78fd8309909279390438c247c4d390bbee4fa914
Summary:
This PR adds support for reusing the file system provided buffer to avoid an extra `memcpy` into RockDB's buffer. This optimization has already been implemented for point lookups, as well as compaction and scan reads _when prefetching is disabled_.
This PR extends this optimization to work with synchronous prefetching (`num_buffers == 1`). Asynchronous prefetching can be addressed in a future PR (and probably should be to keep this PR from growing too large).
Remarks
- To handle the case where the main buffer only has part of the requested data, I used the existing `overlap_buf_` (currently used in the async prefetching case) instead of defining a separate buffer. This was discussed in https://github.com/facebook/rocksdb/pull/13118#discussion_r1842839360.
- We use `MultiRead` with a single request to take advantage of the file system buffer. This is consistent with previous work (e.g. https://github.com/facebook/rocksdb/pull/12266).
- Even without the tests I added, there was some code coverage inside in at least `DBIOCorruptionTest.IterReadCorruptionRetry`, since those tests were failing before I addressed a bug in my code for this PR. [Run with failed test](https://github.com/facebook/rocksdb/actions/runs/11708830448/job/32611508818?pr=13118).
- This prefetching code is not too easy to follow, so I added quite a bit of comments to both the code and test case to try to make it easier to understand the exact internal state of the prefetch buffer at every point in time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13118
Test Plan:
I wrote pretty thorough unit tests that cover synchronous prefetching with file system buffer reuse. The flows for partial hits, complete hits, and complete misses are tested. I also parametrized the test to make sure the async prefetching (without file system buffer reuse) still work as expected.
Once we agree on the changes, I will run a long stress test before merging.
Reviewed By: anand1976
Differential Revision: D65559101
Pulled By: archang19
fbshipit-source-id: 1a56d846e918c20a009b83f1371c1791f69849ae
Summary:
**Context/Summary:**
https://github.com/facebook/rocksdb/pull/13117 added check for obsolete SST files that are not cleaned up timely. It caused a infrequent stress test failure `assertion="live_and_quar_files.find(file_number) != live_and_quar_files.end()"` that I haven't repro-ed yet.
This PR prints the file number so we can find out what happens to that file through info logs when encountering the same failure.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13145
Test Plan:
Manually fail the assertion and observe the stderr printing
```
[ RUN ] DBBasicTest.UniqueSession
File 12 is not live nor quarantined
db_basic_test: db/db_impl/db_impl_debug.cc:384: rocksdb::DBImpl::TEST_VerifyNoObsoleteFilesCached(bool) const::<lambda(const rocksdb::Slice&, rocksdb::Cache::ObjectPtr, size_t, const rocksdb::Cache::CacheItemHelper*)>: Assertion `false' failed.
```
Reviewed By: pdillinger
Differential Revision: D66134154
Pulled By: hx235
fbshipit-source-id: 353164c373d3d674cee676b24468dfc79a1d4563
Summary:
Pull in HISTORY for 9.9.0, update version.h for next version, update check_format_compatible.sh, update git hash for folly
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13146
Test Plan: CI
Reviewed By: ltamasi
Differential Revision: D66142259
Pulled By: jowlyzhang
fbshipit-source-id: 90216b2d7cff2e0befb4f56567e3bd074f97c484
Summary:
Follow-up to https://github.com/facebook/rocksdb/issues/13114
This change makes the options mutable in testing only through some internal hooks, so that we can keep the easier mechanics and testing of making the options mutable separate from a more interesting and critical fix needed for the options to be *safely* mutable. See https://github.com/facebook/rocksdb/pull/9964/files#r1024449523 for some background on the interesting remaining problem, which we've added a test for here, with the failing piece commented out (because it puts the DB in a failure state): PrecludeLastLevelTest.RangeTombstoneSnapshotMigrateFromLast.
The mechanics of making the options mutable turned out to be smaller than expected because `RegisterRecordSeqnoTimeWorker()` and `RecordSeqnoToTimeMapping()` are already robust to things like frequently switching between preserve/preclude durations e.g. with new and dropped column families, based on work from
https://github.com/facebook/rocksdb/issues/11920, https://github.com/facebook/rocksdb/issues/11929, and https://github.com/facebook/rocksdb/issues/12253. Mostly, `options_mutex_` prevents races
in applying the options changes, and smart capacity enforcement in `SeqnoToTimeMapping` means it doesn't really matter if the periodic task wakes up too often by being re-scheduled repeatedly.
Functional changes needed other than marking mutable:
* Update periodic task registration (as needed) from SetOptions, with a mapping recorded then also in case it's needed.
* Install SuperVersion(s) with updated mapping when the registration function itself updates the mapping.
Possible follow-up (aside from already mentioned):
* Some FIXME code in RangeTombstoneSnapshotMigrateFromLast is present because Flush does not automatically include a seqno to time mapping entry that puts an upper bound on how new the flushed data is. This has the potential to be a measurable CPU impact so needs to be done carefully.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13124
Test Plan:
updated/refactored tests in tiered_compaction_test to parametrically use dynamic configuration changes (or DB restarts) when changing operating parameters such as these.
CheckInternalKeyRange test got some heavier refactoring in preparation for follow-up, and manually verified that the test still fails when relevant `if (!safe_to_penultimate_level) ...` code is disabled.
Reviewed By: jowlyzhang
Differential Revision: D65634146
Pulled By: pdillinger
fbshipit-source-id: 25c9d00fd5b7fd1b408b5f36d58dc48647970528
Summary:
In PR https://github.com/facebook/rocksdb/issues/13074 , we added a logic to prevent stale OPTIONS file from getting deleted by `PurgeObsoleteFiles()` if the OPTIONS file is being referenced by any of the scheduled the remote compactions.
`PurgeObsoleteFiles()` was not the only place that we were cleaning up the old OPTIONS file. We've been also directly cleaning up the old OPTIONS file as part of `SetOptions()`: `RenameTempFileToOptionsFile()` -> `DeleteObsoleteOptionsFiles()` unless FileDeletion is disabled.
This was not caught by the UnitTest because we always preserve the last two OPTIONS file. A single call of `SetOptions()` was not enough to surface this issue in the previous PR.
To keep things simple, we are just skipping the old OPTIONS file clean up in `RenameTempFileToOptionsFile()` if remote compaction is enabled. We let `PurgeObsoleteFiles()` clean up the old options file later after the compaction is done.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13139
Test Plan:
Updated UnitTest to reproduce the scenario. It's now passing with the fix.
```
./compaction_service_test --gtest_filter="*PreservedOptionsRemoteCompaction*"
```
Reviewed By: cbi42
Differential Revision: D65974726
Pulled By: jaykorean
fbshipit-source-id: 1907e8450d2ccbb42a93084f275e666648ef5b8c
Summary:
I've seen some release notes talking about implementation detail classes, and starting with attempted markdown italics syntax instead of list item syntax. Patched HISTORY.md for existing oddities.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13135
Test Plan:
manual, look at
https://github.com/facebook/rocksdb/blob/main/HISTORY.md
Reviewed By: jowlyzhang
Differential Revision: D65802777
Pulled By: pdillinger
fbshipit-source-id: a1dc2b17709d633352d7e8a275304092dd7be746
Summary:
introduce the class WBWIMemTable that implements ReadOnlyMemTable interface with data stored in a WriteBatchWithIndex object.
This PR implements the main read path: Get, MultiGet and Iterator. It only supports Put, Delete and SingleDelete operations for now. All the keys in the WBWIMemTable will be assigned a global sequence number through WBWIMemTable::SetGlobalSequenceNumber().
Planned follow up PRs:
- Create WBWIMemTable with a transaction's WBWI and ingest it into a DB during Transaction::Commit()
- Support for Merge. This will be more complicated since we can have multiple updates with the same user key for Merge.
- Support for other operations like WideColumn and other ReadOnlyMemTable methods.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13123
Test Plan: * A mini-stress test for the read path is added as a new unit test
Reviewed By: jowlyzhang
Differential Revision: D65633419
Pulled By: cbi42
fbshipit-source-id: 0684fe47260b41f51ca39c300eb72ca5bc9c5a3b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13134
Even though `Transaction` does not currently support the attribute group variants of `PutEntity` / `GetEntity` / `MultiGetEntity`, we can still test the corresponding APIs of the underlying `TransactionDB` or `OptimisticTransactionDB` instance. Note: the multi-operation transaction stress test will be handled separately.
Reviewed By: jaykorean
Differential Revision: D65780384
fbshipit-source-id: e4ef3d0c25bcbde9d6d8410af0b7d9381c6b501a
Summary:
The bug only happens for transaction db with 2pc. The main change is in `MemTableList::TryInstallMemtableFlushResults`. Before this fix, `memtables_to_flush` may not include all flushed memtables, and it causes the min_log_number for the flush to be incorrect. The code path for calculating min_log_number is `MemTableList::TryInstallMemtableFlushResults() -> GetDBRecoveryEditForObsoletingMemTables() -> PrecomputeMinLogNumberToKeep2PC() -> FindMinPrepLogReferencedByMemTable()`. Inside `FindMinPrepLogReferencedByMemTable()`, we need to exclude all memtables being flushed.
The PR also includes some documentation changes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13127
Test Plan: added a new unit that fails before this change.
Reviewed By: ltamasi
Differential Revision: D65679270
Pulled By: cbi42
fbshipit-source-id: 611f34bd6ef4cba51f8b54cb1be416887b5a9c5e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13131
The earlier stress test code did not consider that `PrepareValue()` could fail because of read fault injection, leading to false positives. The patch shuffles the `PrepareValue()` calls around a bit in `TestIterate` / `TestIterateAgainstExpected` in order to prevent this by leveraging the existing code paths that intercept injected faults.
Reviewed By: cbi42
Differential Revision: D65731543
fbshipit-source-id: b21c6584ebaa2ff41cd4569098680b91ff7991d1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13130
The patch changes the stress test code so it always logs the error status to aid debugging when a `PrepareValue` call fails.
Reviewed By: hx235
Differential Revision: D65712502
fbshipit-source-id: da81566a358777b691178f0d0a1b680453d03e7d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13129
The `PrepareValue()` call on an iterator can fail, for example due to our stress tests' read fault injection. Such a failure invalidates the iterator, which makes it illegal to call methods like `key()` on it and leads to assertion violations. The patch fixes this by saving the key before calling `PrepareValue()`, so we can still print it for debugging purposes in case the call fails.
Reviewed By: jowlyzhang
Differential Revision: D65689225
fbshipit-source-id: c2bf298366def0ba3b3c089ee58e28609ecdfab4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13128
Similarly to https://github.com/facebook/rocksdb/pull/13119, the patch adds a new API `Transaction::GetCoalescingIterator` that can be used to create a multi-column-family coalescing iterator over the specified column families, including the data from both the transaction and the underlying database. This API is currently supported for optimistic and write-committed pessimistic transactions.
Reviewed By: jowlyzhang
Differential Revision: D65682389
fbshipit-source-id: faf5dd1de9bce9d403fc34246ecab4c55572a228
Summary:
This PR fixes a few cases where RocksDB was not retrying checksum failure/corruption of file reads with the `verify_and_reconstruct_read` IO option. After fixing these cases, we can almost always successfully open the DB and execute reads even if we see transient corruptions, provided the `FileSystem` supports the `verify_and_reconstruct_read` option. The specific cases fixed in this PR are -
1. CURRENT file
2. IDENTITY file
3. OPTIONS file
4. SST footer
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13122
Test Plan: Unit test in `db_io_failure_test.cc` that injects corruption at various stages of DB open and reads
Reviewed By: jaykorean
Differential Revision: D65617982
Pulled By: anand1976
fbshipit-source-id: 4324b88cc7eee5501ab5df20ef7a95bb12ed3ea7
Summary:
Follow-up to https://github.com/facebook/rocksdb/issues/13106 which revealed that some SST file readers (in addition to blob files) were being essentially leaked in TableCache (until DB::Close() time). Patched sources of leaks:
* Flush that is not committed (builder.cc)
* Various obsolete SST files picked up by directory scan but not caught by SubcompactionState::Cleanup() cleaning up from some failed compactions. Dozens of unit tests fail without the "backstop" TableCache::Evict() call in PurgeObsoleteFiles().
We also needed to adjust the check for leaks as follows:
* Ok if DB::Open never finished (see comment)
* Ok if deletions are disabled (see comment)
* Allow "quarantined" files to be in table_cache because (presumably) they might become live again.
* Get live files from all live Versions.
Suggested follow-up:
* Potentially delete more obsolete files sooner with a FIXME in db_impl_files.cc. This could potentially be high value because it seems to gate deletion of any/all newer obsolete files on all older compactions finishing.
* Try to catch obsolete files in more places using the VersionSet::obsolete_files_ pipeline rather than relying on them being picked up with directory scan, or deleting them outside of normal mechanisms.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13117
Test Plan: updated check used in most all unit tests in ASAN build
Reviewed By: hx235
Differential Revision: D65502988
Pulled By: pdillinger
fbshipit-source-id: aa0795a8a09d9ec578d25183fe43e2a35849209c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13125
The patch adds the new read option `allow_unprepared_value` and the new `Iterator` / `CoalescingIterator` / `AttributeGroupIterator` API `PrepareValue()` to the stress/crash tests. The change affects the batched, non-batched, and CF consistency stress test flavors and the `TestIterate`, `TestPrefixScan`, and `TestIterateAgainstExpected` operations.
Reviewed By: hx235
Differential Revision: D65636380
fbshipit-source-id: fd0caa0e87d03b6206667f07499b0c11847d1bbe
Summary:
This PR adds some missing pieces in order to handle UDT setting toggles while replay WALs for WriteCommitted transactions DB. Specifically, all the transaction markers for no op, prepare, commit, rollback are currently not carried over from the original WriteBatch to the new WriteBatch when there is a timestamp setting difference detected. This PR fills that gap.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13121
Test Plan: Added unit tests
Reviewed By: ltamasi
Differential Revision: D65558801
Pulled By: jowlyzhang
fbshipit-source-id: 8176882637b95f6dc0dad10d7fe21056fa5173d1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13119
The patch adds a new API `Transaction::GetAttributeGroupIterator` that can be used to create a multi-column-family attribute group iterator over the specified column families, including the data from both the transaction and the underlying database. This API is currently supported for optimistic and write-committed pessimistic transactions.
Reviewed By: jowlyzhang
Differential Revision: D65548324
fbshipit-source-id: 0fb8a22129494770fdba3d6024eef72b3e051136
Summary:
This PR does a few misc things for file ingestion flow:
- Add an invalid argument status return for the combination of `allow_global_seqno = false` and external files' key range overlap in `Prepare` stage.
- Add a MemTables status check for when column family is flushed before `Run`.
- Replace the column family dropped check with an assertion after thread enters the write queue and before it exits the write queue, since dropping column family can only happen in the single threaded write queue too and we already checked once after enter write queue.
- Add an `ExternalSstFileIngestionJob::GetColumnFamilyData` API.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13100
Test Plan: Added unit tests, and stress tested the ingestion path
Reviewed By: hx235
Differential Revision: D65180472
Pulled By: jowlyzhang
fbshipit-source-id: 180145dd248a7507a13a543481b135e5a31ebe2d
Summary:
This assertion could fail if the compaction input files were successfully trivially moved. On re-locking db mutex after successful `LogAndApply`, those files could have been picked up again by some other compactions. And the assertion will fail.
Example failure: P1669529213
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13109
Reviewed By: cbi42
Differential Revision: D65308574
Pulled By: jowlyzhang
fbshipit-source-id: 32413bdc8e28e67a0386c3fe6327bf0b302b9d1d
Summary:
This is a small follow-up to https://github.com/facebook/rocksdb/pull/13083.
When we check the `newest_key_time` of files for temperature change compaction, we currently return early if we ever find a file with an unknown `est_newest_key_time`.
However, it is possible for a younger file to have a populated value for `newest_key_time`, since this is a new table property.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13112
Test Plan: The existing unit tests are sufficient.
Reviewed By: cbi42
Differential Revision: D65451797
Pulled By: archang19
fbshipit-source-id: 28e67c2d35a6315f912471f2848de87dd7088d99
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13113
The patch makes some small improvements related to `allow_unprepared_value` and multi-CF iterators as groundwork for further changes:
1) Similarly to `BaseDeltaIterator`'s base iterator, `MultiCfIteratorImpl` gets passed its child iterators by the client. Even though they are currently guaranteed to have been created using the same read options as each other and the multi-CF iterator, it is safer to not assume this and call `PrepareValue` unconditionally before using any child iterator's `value()` or `columns()`.
2) Again similarly to `BaseDeltaIterator`, it makes sense to pass the entire `ReadOptions` structure to `MultiCfIteratorImpl` in case it turns out to require other read options in the future.
3) The constructors of the various multi-CF iterator classes now take an rvalue reference to a vector of column family handle + `unique_ptr` to child iterator pairs and use move semantics to take ownership of this vector (instead of taking two separate vectors of column family handles and raw iterator pointers).
4) Constructor arguments and the members of `MultiCfIteratorImpl` are reordered for consistency.
Reviewed By: jowlyzhang
Differential Revision: D65407521
fbshipit-source-id: 66c2c689ec8b036740bd98641b7b5c0ff7e777f2
Summary:
Move them to MutableCFOptions and perform appropriate refactorings to make that work. I didn't want to mix up refactoring with interesting functional changes. Potentially non-trivial bits here:
* During DB Open or RegisterRecordSeqnoTimeWorker we use `GetLatestMutableCFOptions()` because either (a) there might not be a current version, or (b) we are in the process of applying the desired next options.
* Upgrade some test infrastructure to allow some options in MutableCFOptions to be mutable (should be a temporary state)
* Fix a warning that showed up about uninitialized `paranoid_memory_checks`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13114
Test Plan: existing tests, manually check options are still not settable with SetOptions
Reviewed By: jowlyzhang
Differential Revision: D65429031
Pulled By: pdillinger
fbshipit-source-id: 6e0906d08dd8ddf62731cefffe9b8d94149942b9
Summary:
This PR sets up follow-up changes for large transaction support. It introduces an interface that allows custom implementations of immutable memtables. Since transactions use a WriteBatchWithIndex to index their operations, I plan to add a ReadOnlyMemTable implementation backed by WriteBatchWithIndex. This will enable direct ingestion of WriteBatchWithIndex into the DB as an immutable memtable, bypassing memtable writes for transactions.
The changes mostly involve moving required methods for immutable memtables into the ReadOnlyMemTable class.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13107
Test Plan:
* Existing unit test and stress test.
* Performance: I do not expect this change to cause noticeable performance regressions with LTO and devirtualization. The memtable-only readrandom benchmark shows no consistent performance difference:
```
USE_LTO=1 OPTIMIZE_LEVEL="-O3" DEBUG_LEVEL=0 make -j160 db_bench
(for I in $(seq 1 50);do ./db_bench --benchmarks=fillseq,readrandom --write_buffer_size=268435456 --writes=250000 --num=250000 --reads=500000 --seed=1723056275 2>&1 | grep "readrandom"; done;) | awk '{ t += $5; c++; print } END { print 1.0 * t / c }';
3 runs:
main: 760728, 752727, 739600
PR: 763036, 750696, 739022
```
Reviewed By: jowlyzhang
Differential Revision: D65365062
Pulled By: cbi42
fbshipit-source-id: 40c673ab856b91c65001ef6d6ac04b65286f2882
Summary:
As titled. This flag controls how frequent standalone range deletion file is tested in the file ingestion flow, for better debuggability.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13101
Test Plan: Manually tested in stress test
Reviewed By: hx235
Differential Revision: D65361004
Pulled By: jowlyzhang
fbshipit-source-id: 21882e7cc5918aff45449acaeb33b696ab1e37f0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13111
As a follow-up to https://github.com/facebook/rocksdb/pull/13105, the patch changes `BaseDeltaIterator` so that it honors the read option `allow_unprepared_value`. When the option is set and the `BaseDeltaIterator` lands on the base iterator, it defers calling `PrepareValue` on the base iterator and setting `value()` and `columns()` until `PrepareValue` is called on the `BaseDeltaIterator` itself.
Reviewed By: jowlyzhang
Differential Revision: D65344764
fbshipit-source-id: d79c77b5de7c690bf2deeff435e9b0a9065f6c5c
Summary:
There is a `strict_capacity_limit` option which imposes a hard memory limit on the block cache. When the block cache is enabled, every read request is serviced from the block cache. If the required block is missing, it is first inserted into the cache. If `strict_capacity_limit` is `true` and the limit has been reached, the `Get` and `MultiGet` requests should fail. However, currently this is not happening for `MultiGet`.
I updated `MultiGet` to explicitly check the returned status of `MaybeReadBlockAndLoadToCache`, so the status does not get overwritten later.
Thank you anand1976 for the problem explanation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13104
Test Plan:
Added unit test for both `Get` and `MultiGet` with a `strict_capacity_limit` set.
Before the change, half of my unit test cases failed https://github.com/facebook/rocksdb/actions/runs/11604597524/job/32313608085?pr=13104. After I added the check for the status returned by `MaybeReadBlockAndLoadToCache`, they all pass.
I also ran these tests manually (I had to run `make clean` before):
```
make -j64 block_based_table_reader_test COMPILE_WITH_ASAN=1 ASSERT_STATUS_CHECKED=1
./block_based_table_reader_test --gtest_filter="*StrictCapacityLimitReaderTest.Get*"
./block_based_table_reader_test --gtest_filter="*StrictCapacityLimitReaderTest.MultiGet*"
```
Reviewed By: anand1976
Differential Revision: D65302470
Pulled By: archang19
fbshipit-source-id: 28dcc381e67e05a89fa9fc9607b4709976d6d90e
Summary:
This PR does two things:
1. Adds a new table property `newest_key_time`
2. Uses this property to improve TTL and temperature change compaction.
### Context
The current `creation_time` table property should really be named `oldest_ancestor_time`. For flush output files, this is the oldest key time in the file. For compaction output files, this is the minimum among all oldest key times in the input files.
The problem with using the oldest ancestor time for TTL compaction is that we may end up dropping files earlier than we should. What we really want is the newest (i.e. "youngest") key time. Right now we take a roundabout way to estimate this value -- we take the value of the _oldest_ key time for the _next_ (newer) SST file. This is also why the current code has checks for `index >= 1`.
Our new property `newest_key_time` is set to the file creation time during flushes, and the max over all input files for compactions.
There were some additional smaller changes that I had to make for testing purposes:
- Refactoring the mock table reader to support specifying my own table properties
- Refactoring out a test utility method `GetLevelFileMetadatas` that would otherwise be copy/pasted in 3 places
Credit to cbi42 for the problem explanation and proposed solution
### Testing
- Added a dedicated unit test to my `newest_key_time` logic in isolation (i.e. are we populating the property on flush and compaction)
- Updated the existing unit tests (for TTL/temperate change compaction), which were comprehensive enough to break when I first made my code changes. I removed the test setup code which set the file metadata `oldest_ancestor_time`, so we know we are actually only using the new table property instead.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13083
Reviewed By: cbi42
Differential Revision: D65298604
Pulled By: archang19
fbshipit-source-id: 898ef91b692ab33f5129a2a16b64ecadd4c32432
Summary:
An earlier change (https://github.com/facebook/rocksdb/commit/b34cef57b798520791312f2f40681c4d12d5d33c) removed apparently unused functionality where an obsolete blob file number is passed for removal from TableCache, which manages SST files. This was actually relying on broken/fragile abstractions wherein TableCache and BlobFileCache share the same Cache and using the TableCache interface to manipulate blob file caching. No unit test was actually checking for removal of obsolete blob files from the cache (which is somewhat tricky to check and a second order correctness requirement).
Here we fix the leak and add a DEBUG+ASAN-only check in DB::Close() that no obsolete files are lingering in the table/blob file cache.
Fixes https://github.com/facebook/rocksdb/issues/13066
Important follow-up (FIXME): The added check discovered some apparent cases of leaked (into table_cache) SST file readers that would stick around until DB::Close(). Need to enable that check, diagnose, and fix.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13106
Test Plan:
added a check that is called during DB::Close in ASAN builds (to minimize paying the cost in all unit tests). Without the fix, the check failed in at least these tests:
```
db_blob_basic_test DBBlobBasicTest.DynamicallyWarmCacheDuringFlush
db_blob_compaction_test DBBlobCompactionTest.CompactionReadaheadMerge
db_blob_compaction_test DBBlobCompactionTest.MergeBlobWithBase
db_blob_compaction_test DBBlobCompactionTest.CompactionDoNotFillCache
db_blob_compaction_test DBBlobCompactionTest.SkipUntilFilter
db_blob_compaction_test DBBlobCompactionTest.CompactionFilter
db_blob_compaction_test DBBlobCompactionTest.CompactionReadaheadFilter
db_blob_compaction_test DBBlobCompactionTest.CompactionReadaheadGarbageCollection
```
Reviewed By: ltamasi
Differential Revision: D65296123
Pulled By: pdillinger
fbshipit-source-id: 2276d76482beb2c75c9010bc1bec070bb23a24c0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13105
The `WriteBatchWithIndex::NewIteratorWithBase` interface enables creating a `BaseDeltaIterator` with an arbitrary base iterator passed in by the client, which has potentially been created with the `allow_unprepared_value` read option set. Because of this, `BaseDeltaIterator` has to call `PrepareValue` before using the `value()` or `columns()` from the base iterator. This includes both the case when `BaseDeltaIterator` exposes the `value()` and `columns()` of the base iterator as is and the case when the final `value()` / `columns()` is a result of merging key-values across the base and delta iterators. Note that `BaseDeltaIterator` itself does not support `allow_unprepared_value` yet; this will be implemented in an upcoming patch.
Reviewed By: jowlyzhang
Differential Revision: D65249643
fbshipit-source-id: b0a1ccc0dfd31105b2eef167b463ed15a8bb83b7
Summary:
Follow ups from https://github.com/facebook/rocksdb/issues/13089
- Take `TableProperties` as `const &` instead of `std::shared_ptr<const TableProperties>`
- Move TableProperties OptionsTypeMap definition to another place for other use outside of Remote Compaction
- Add a test verify that the set of field serializations of TableProperties is complete
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13095
Test Plan:
```
./options_settable_test --gtest_filter="*TablePropertiesAllFieldsSettable*"
```
I also intentionally tried adding a new field to `TableProperties`. If it's missed in the OptionsType map, the test detects the missing bytes set and successfully fails.
Reviewed By: pdillinger
Differential Revision: D65077398
Pulled By: jaykorean
fbshipit-source-id: cf10560eb4a467ca523b11fd64945dbc86ac378f
Summary:
Forgot to update after generalizing mutability of BBTO
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13097
Test Plan: no functional change here
Reviewed By: jaykorean
Differential Revision: D65095618
Pulled By: pdillinger
fbshipit-source-id: 6c37cd0e68756c6b56af1c8e15273fae0ca9224d
Summary:
Pull in HISTORY for 9.8.0, update version.h for next version, update check_format_compatible.sh
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13093
Test Plan: CI
Reviewed By: jowlyzhang
Differential Revision: D64987257
Pulled By: pdillinger
fbshipit-source-id: a7cec329e3d245e63767760aa0298c08c3281695
Summary:
In Remote Compactions, the primary host receives the serialized compaction result from the remote worker and deserializes it to build the output. Unlike Local Compactions, where table properties are built by TableBuilder, in Remote Compactions, these properties were not included in the serialized compaction result. This was likely done intentionally since the table properties are already available in the SST files.
Because TableProperties are not populated as part of CompactionOutputs for remote compactions, we were unable to log the table properties in OnCompactionComplete and use them for verification. We are adding the TableProperties as part of the CompactionServiceOutputFile in this PR. By including the TableProperties in the serialized compaction result, the primary host will be able to access them and verify that they match the values read from the actual SST files.
We are also adding the populating `format_version` in table_properties of in TableBuilder. This has not been a big issue because the `format_version` is written to the SST files directly from `TableOptions.format_version`. When loaded from the SST files, it's populated directly by reading from the MetaBlock. This info has only been missing in the TableBuilder's Rep.props.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13089
Test Plan:
```
./compaction_job_test
```
```
./compaction_service_test
```
Reviewed By: pdillinger
Differential Revision: D64878740
Pulled By: jaykorean
fbshipit-source-id: b6f2fdce851e6477ecb4dd5a87cdc62e176b746b
Summary:
Fix a longstanding race condition in SetOptions for `block_based_table_factory` options. The fix is mostly described in new, unified `TableFactoryParseFn()` in `cf_options.cc`. Also in this PR:
* Adds a virtual `Clone()` function to TableFactory
* To avoid behavioral hiccups with `SetOptions`, make the "hidden state" of `BlockBasedTableFactory` shared between an original and a clone. For example, `TailPrefetchStats`
* `Configurable` was allowed to be copied but was not safe to do so, because the copy would have and use pointers into object it was copied from (!!!). This has been fixed using relative instead of absolute pointers, though it's still technically relying on undefined behavior (consistent object layout for non-standard-layout types).
For future follow-up:
* Deny SetOptions on block cache options (dubious and not yet made safe with proper shared_ptr handling)
Fixes https://github.com/facebook/rocksdb/issues/10079
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13082
Test Plan:
added to unit tests and crash test
Ran TSAN blackbox crashtest for hours with options to amplify potential race (see https://github.com/facebook/rocksdb/issues/10079)
Reviewed By: cbi42
Differential Revision: D64947243
Pulled By: pdillinger
fbshipit-source-id: 8390299149f50e2a2b39a5247680f2637edb23c8
Summary:
This PR adds some optimization for compacting standalone range deletion files. A standalone range deletion file is one with just a single range deletion. Currently, such a file is used in bulk loading to achieve something like atomically delete old version of all data with one big range deletion and adding new version of data. These are the changes included in the PR:
1) When a standalone range deletion file is ingested via bulk loading, it's marked for compaction.
2) When picking input files during compaction picking, we attempt to only pick a standalone range deletion file when oldest snapshot is at or above the file's seqno. To do this, `PickCompaction` API is updated to take existing snapshots as an input. This is only done for the universal compaction + UDT disabled combination, we save querying for existing snapshots and not pass it for all other cases.
3) At `Compaction` construction time, the input files will be filtered to examine if any of them can be skipped for compaction iterator. For example, if all the data of the file is deleted by a standalone range tombstone, and the oldest snapshot is at or above such range tombstone, this file will be filtered out.
4) Every time a snapshot is released, we examine if any column family has standalone range deletion files that becomes eligible to be scheduled for compaction. And schedule one for it.
Potential future improvements:
- Add some dedicated statistics for the filtered files.
- Extend this input filtering to L0 files' compactions cases when a newer L0 file could shadow an older L0 file
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13078
Test Plan: Added unit tests and stress tested a few rounds
Reviewed By: cbi42
Differential Revision: D64879415
Pulled By: jowlyzhang
fbshipit-source-id: 02b8683fddbe11f093bcaa0a38406deb39f44d9e
Summary:
we currently record write operations to tracer before checking callback in PipelinedWriteImpl and WriteImplWALOnly. For optimistic transaction DB, this means that an operation can be recorded to tracer even when it's not written to DB or WAL. I suspect this is the reason some of our optimistic txn crash test is failing. The evidence is that the trace contains some duplicated entry and has more entries compared to the corresponding entry in WAL. This PR moves the tracer logic to be after checking callback status.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13088
Test Plan: monitor crash test.
Reviewed By: hx235
Differential Revision: D64711753
Pulled By: cbi42
fbshipit-source-id: 55fd1223538ec6294ce84a957c306d3d9d91df5f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13079
The patch adds support for the new read option `allow_unprepared_value` to the multi-column-family iterators `CoalescingIterator` and `AttributeGroupIterator`. When this option is set, these iterators populate their value (`value()` + `columns()` or `attribute_groups()`) in an on-demand fashion when `PrepareValue()` is called. Calling `PrepareValue()` on the child iterators is similarly deferred until `PrepareValue()` is called on the main iterator.
Reviewed By: jowlyzhang
Differential Revision: D64570587
fbshipit-source-id: 783c8d408ad10074417dabca7b82c5e1fe5cab36
Summary:
# Summary
There was a [test failure](https://github.com/facebook/rocksdb/actions/runs/11381731053/job/31663774089?fbclid=IwZXh0bgNhZW0CMTEAAR0YJVdnkKUhN15RJQrLsvicxqzReS6y4A14VFQbWu-81XJsSsyNepXAr2c_aem_JyQqNdtpeKFSA6CjlD-pDg) from uninit value in the CompactionServiceInput
```
[ RUN ] CompactionJobTest.InputSerialization
==79945== Use of uninitialised value of size 8
==79945== at 0x58EA69B: _itoa_word (_itoa.c:179)
==79945== by 0x5906574: __vfprintf_internal (vfprintf-internal.c:1687)
==79945== by 0x591AF99: __vsnprintf_internal (vsnprintf.c:114)
==79945== by 0x1654AE: std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > __gnu_cxx::__to_xstring<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, char>(int (*)(char*, unsigned long, char const*, __va_list_tag*), unsigned long, char const*, ...) (string_conversions.h:111)
==79945== by 0x5126C65: to_string (basic_string.h:6568)
==79945== by 0x5126C65: rocksdb::SerializeSingleOptionHelper(void const*, rocksdb::OptionType, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >*) (options_helper.cc:541)
==79945== by 0x512718B: rocksdb::OptionTypeInfo::Serialize(rocksdb::ConfigOptions const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, void const*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >*) const (options_helper.cc:1084)
```
This was due to `options_file_number` value not set in the unit test. However, this value is guaranteed to be set in the normal path. It was just missing in the test path. Setting the 0 as the default value for uninitialized fields in the `CompactionServiceInput` and `CompactionServiceResult` for now.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13080
Test Plan: Existing tests should be sufficient
Reviewed By: cbi42
Differential Revision: D64573567
Pulled By: jaykorean
fbshipit-source-id: 7843a951770c74445620623d069a52ba93ad94d5
Summary:
This is setting up for a fix to a data race in SetOptions on BlockBasedTableOptions (BBTO), https://github.com/facebook/rocksdb/issues/10079
The race will be fixed by replacing `table_factory` with a modified copy whenever we want to modify a BBTO field.
An argument could be made that this change creates more entaglement between features (e.g. BlobSource <-> MutableCFOptions), rather than (conceptually) minimizing the dependencies of each feature, but
* Most of these things already depended on ImmutableOptions
* Historically there has been a lot of plumbing (and possible small CPU overhead) involved in adding features that need to reach a lot of places, like `block_protection_bytes_per_key`. Keeping those wrapped up in options simplifies that.
* SuperVersion management generally takes care of lifetime management of MutableCFOptions, so is not that difficult. (Crash test agrees so far.)
There are some FIXME places where it is known to be unsafe to replace `block_cache` unless/until we handle shared_ptr tracking properly. HOWEVER, replacing `block_cache` is generally dubious, at least while existing users of the old block cache (e.g. table readers) can continue indefinitely.
The change to cf_options.cc is essentially just moving code (not changing).
I'm not concerned about the performance of copying another shared_ptr with MutableCFOptions, but I left a note about considering an improvement if more shared_ptr are added to it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13077
Test Plan:
existing tests, crash test.
Unit test DBOptionsTest.GetLatestCFOptions updated with some temporary logic. MemoryTest required some refactoring (simplification) for the change.
Reviewed By: cbi42
Differential Revision: D64546903
Pulled By: pdillinger
fbshipit-source-id: 69ae97ce5cf4c01b58edc4c5d4687eb1e5bf5855
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13076
The patch makes it possible to construct an `IteratorAttributeGroup` using an `AttributeGroup` instance, and implements `operator==` / `operator!=` for these two classes consistently. It also makes some minor improvements in the related test suites `CoalescingIteratorTest` and `AttributeGroupIteratorTest`.
Reviewed By: jaykorean
Differential Revision: D64510653
fbshipit-source-id: 95d3340168fa3b34e7ef534587b19131f0a27fb7
Summary:
Compaction stats code is not so straightforward to understand. Here's a bit of context for this PR and why this change was made.
- **CompactionStats (compaction_stats_.stats):** Internal stats about the compaction used for logging and public metrics.
- **CompactionJobStats (compaction_job_stats_)**: The public stats at job level. It's part of Compaction event listener and included in the CompactionResult.
- **CompactionOutputsStats**: output stats only. resides in CompactionOutputs. It gets aggregated toward the CompactionStats (internal stats).
The internal stats, `compaction_stats_.stats`, has the output information recorded from the compaction iterator, but it does not have any input information (input records, input output files) until `UpdateCompactionStats()` gets called. We cannot simply call `UpdateCompactionStats()` to fill in the input information in the remote compaction (which is a subcompaction of the primary host's compaction) because the `compaction->inputs()` have the full list of input files and `UpdateCompactionStats()` takes the entire list of records in all files. `num_input_records` gets double-counted if multiple sub-compactions are submitted to the remote worker.
The job level stats (in the case of remote compaction, it's subcompaction level stat), `compaction_job_stats_`, has the correct input records, but has no output information. We can use `UpdateCompactionJobStats(compaction_stats_.stats)` to set the output information (num_output_records, num_output_files, etc.) from the `compaction_stats_.stats`, but it also sets all other fields including the input information which sets all back to 0.
Therefore, we are overriding `UpdateCompactionJobStats()` in remote worker only to update job level stats, `compaction_job_stats_`, with output information of the internal stats.
Baiscally, we are merging the aggregated output info from the internal stats and aggregated input info from the compaction job stats.
In this PR we are also fixing how we are setting `is_remote_compaction` in CompactionJobStats.
- OnCompactionBegin event, if options.compaction_service is set, `is_remote_compaction=true` for all compactions except for trivial moves
- OnCompactionCompleted event, if any of the sub_compactions were done remotely, compaction level stats's `is_remote_compaction` will be true
Other minor changes
- num_output_records is already available in CompactionJobStats. No need to store separately in CompactionResult.
- total_bytes is not needed.
- Renamed `SubcompactionState::AggregateCompactionStats()` to `SubcompactionState::AggregateCompactionOutputStats()` to make it clear that it's only aggregating output stats.
- Renamed `SetTotalBytes()` to `AddBytesWritten()` to make it more clear that it's adding total written bytes from the compaction output.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13071
Test Plan:
Unit Tests added and updated
```
./compaction_service_test
```
Reviewed By: anand1976
Differential Revision: D64479657
Pulled By: jaykorean
fbshipit-source-id: a7a776a00dc718abae95d856b661bcbafd3b0ed5
Summary:
add `IngestExternalFileOptions::fill_cache` to allow users to ingest files without loading index/filter/data and other blocks into block cache during file ingestion. This can be useful when users are ingesting files into a CF that is not available to readers yet.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13067
Test Plan:
* unit test: `ExternalSSTFileTest.NoBlockCache`
* ran one round of crash test with fill_cache disabled: `python3 ./tools/db_crashtest.py --simple blackbox --ops_per_thread=1000000 --interval=30 --ingest_external_file_one_in=200 --level0_stop_writes_trigger=200 --level0_slowdown_writes_trigger=100 --sync_fault_injection=0 --disable_wal=0 --manual_wal_flush_one_in=0`
Reviewed By: jowlyzhang
Differential Revision: D64356424
Pulled By: cbi42
fbshipit-source-id: b380c26f5987238e1ed7d42ceef0390cfaa0b8e2
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13069
Currently, when using range scans with BlobDB, the iterator logic eagerly loads values from blob files when landing on a new entry. This can be wasteful in use cases where the values associated with some keys in the range are not used by the application. The patch introduces a new read option `allow_unprepared_value`; when specified, this option results in the above eager loading getting bypassed. Values needed by the application can be then loaded on an on-demand basis by calling the new iterator API `PrepareValue`. Note that currently, only regular single-CF iterators are supported; multi-CF iterators and transactions will be extended in later PRs.
Reviewed By: jowlyzhang
Differential Revision: D64360723
fbshipit-source-id: ee55502fa15dcb307a984922b9afc9d9da15d6e1
Summary:
In https://github.com/facebook/rocksdb/issues/13025 , we made a change to load the latest options file in the remote worker instead of serializing the entire set of options.
That was done under assumption that OPTIONS file do not get purged often. While testing, we learned that this happens more often than we want it to be, so we want to prevent the OPTIONS file from getting purged anytime between when the remote compaction is scheduled and the option is loaded in the remote worker.
Like how we are protecting new SST files from getting purged using `min_pending_output`, we are doing the same by keeping track of `min_options_file_number`. Any OPTIONS file with number greater than `min_options_file_number` will be protected from getting purged. Just like `min_pending_output`, `min_options_file_number` gets bumped when the compaction is done. This is only applicable when `options.compaction_service` is set.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13074
Test Plan:
```
./compaction_service_test --gtest_filter="*PreservedOptionsLocalCompaction*"
./compaction_service_test --gtest_filter="*PreservedOptionsRemoteCompaction*"
```
Reviewed By: anand1976
Differential Revision: D64433795
Pulled By: jaykorean
fbshipit-source-id: 0d902773f0909d9481dec40abf0b4c54ce5e86b2
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13075
The patch simplifies the iteration logic in `MultiCFIteratorImpl::{Advance,Populate}Iterator` a bit and adds some assertions to uniformly enforce the invariant that any iterators currently on the heap should be valid and have an OK status.
Reviewed By: jaykorean
Differential Revision: D64429566
fbshipit-source-id: 36bc22465285b670f859692a048e10f21df7da7a
Summary:
This PR assigns levels to files in separate batches if they overlap. This approach can potentially assign external files to lower levels.
In the prepare stage, if the input files' key range overlaps themselves, we divide them up in the user specified order into multiple batches. Where the files in the same batch do not overlap with each other, but key range could overlap between batches. If the input files' key range don't overlap, they always just make one default batch.
During the level assignment stage, we assign levels to files one batch after another. It's guaranteed that files within one batch are not overlapping, we assign level to each file one after another. If the previous batch's uppermost level is specified, all files in this batch will be assigned to levels that are higher than that level. The uppermost level used by this batch of files is also tracked, so that it can be used by the next batch.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13064
Test Plan:
Updated test and added new test
Manually stress tested
Reviewed By: cbi42
Differential Revision: D64428373
Pulled By: jowlyzhang
fbshipit-source-id: 5aeff125c14094c87cc50088505010dfd2da3d6e
Summary:
Add a timeout for the blackbox crash test final verification step, and print the db_stress stack trace on a timeout. The crash test occasionally hangs in the verification step and this will help debug.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13070
Reviewed By: hx235
Differential Revision: D64414461
Pulled By: anand1976
fbshipit-source-id: 4629aac01fbe6c788665beddc66280ba446aadbe
Summary:
Checkpoint creation skips flushing the memtable, even if explicitly requested, when the WAL is locked. This can happen if the user calls `LockWAL()`. In this case, db_stress checkpoint verification fails as the checkpoint will not contain keys present in the primary DB's memtable. Sanitize `checkpoint_one_in` and `lock_wal_one_in` so they're mutually exclusive.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13068
Reviewed By: hx235
Differential Revision: D64353998
Pulled By: anand1976
fbshipit-source-id: 7c93563347f033b6008a47a7d71471e59747e143
Summary:
- When `FileChecksumGenFactory` is set, include the `file_checksum` and `file_checksum_func_name` in the output file metadata
- ~~In Remote Compaction, try opening the output files in the temporary directory to do a quick sanity check before returning the result with status.~~
- After offline discussion, we decided to rely on Primary's existing Compaction flow to sanity check the output files. If the output file is corrupted, we will still be able to catch it and not installing it even after renaming them to cf_paths. The corrupted file in the cf_path won't be added to the MANIFEST and will be purged as part of the next `PurgeObsoleteFiles()` call.
- Unit Test has been added to validate above.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13060
Test Plan:
Unit test added
```
./compaction_service_test --gtest_filter="*CorruptedOutput*"
./compaction_service_test --gtest_filter="*TruncatedOutput*"
./compaction_service_test --gtest_filter="*CustomFileChecksum*"
./compaction_job_test --gtest_filter="*ResultSerialization*"
```
Reviewed By: cbi42
Differential Revision: D64189645
Pulled By: jaykorean
fbshipit-source-id: 6cf28720169c960c80df257806bfee3c0d177159
Summary:
In theory, there should be no danger in mutability, as table
builders and readers work from copies of BlockBasedTableOptions.
However, there is currently an unresolved read-write race that
affecting SetOptions on BBTO fields. This should be generally
acceptable for non-pointer options of 64 bits or less, but a fix
is needed to make it mutability general here. See
https://github.com/facebook/rocksdb/issues/10079
This change systematically sets all of those "simple" options (and future
such options) as mutable. (Resurrecting this PR perhaps preferable to
proposed https://github.com/facebook/rocksdb/issues/13063)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/10021
Test Plan: Some unit test updates. XXX comment added to stress test code
Reviewed By: cbi42
Differential Revision: D64360967
Pulled By: pdillinger
fbshipit-source-id: ff220fa778331852fe331b42b76ac4adfcd2d760
Summary:
When user-defined timestamps are not persisted, currently we replace the actual timestamp with min timestamp after an entry is output from compaction iterator. Compaction iterator won't be able to help with removing stale entries this way. This PR adds a wrapper iterator `TimestampStrippingIterator` for `MemTableIterator` that does the min timestamp replacement at the memtable iteration step. It is used by flush and can help remove stale entries from landing in L0 files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13035
Test Plan: Added unit test
Reviewed By: pdillinger, cbi42
Differential Revision: D63423682
Pulled By: jowlyzhang
fbshipit-source-id: 087dcc9cee97b9ea51b8d2b88dc91c2984d54e55
Summary:
When the input files are not overlapping, a.k.a `files_overlap_=false`, it's best to assign them to non L0 levels so that they are not one sorted run each. This can be done regardless of compaction style being leveled or universal without any side effects.
Just my guessing, this special handling may be there because universal compaction used to have an invariant that sequence number on higher levels should not be smaller than sequence number in lower levels. File ingestion used to try to keep up to that promise by doing "sequence number stealing" from the to be assigned level. However, that invariant is no longer true after deletion triggered compaction is added for universal compaction, and we also removed the sequence stealing logic from file ingestion.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13059
Test Plan: Updated existing tests
Reviewed By: cbi42
Differential Revision: D64220100
Pulled By: jowlyzhang
fbshipit-source-id: 70a83afba7f4c52d502c393844e6b3273d5cf628
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13061
As groundwork for further changes, the patch refactors the BlobDB-related parts of `DBIter` by 1) introducing a new internal helper class `DBIter::BlobReader` that encapsulates all members needed to retrieve a blob value (namely, `Version` and the `ReadOptions` fields) and 2) factoring out and cleaning up some duplicate logic related to resolving blob references in the non-Merge (see `SetValueAndColumnsFromBlob`) and Merge (see `MergeWithBlobBaseValue`) cases.
Reviewed By: jowlyzhang
Differential Revision: D64078099
fbshipit-source-id: 22d5bd93e6e5be5cc9ecf6c4ee6954f2eb016aff
Summary:
**Context/Summary:**
A part of this test is to verify compression conditionally happens depending on the shape of the LSM when `options.level_compaction_dynamic_level_bytes = true;`. It uses the total file size to determine whether compression has happened or not. This involves some hard-coded math hard to understand. This PR replaces those with statistics that directly shows whether compression has happened or not.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13044
Test Plan: Existing test
Reviewed By: jaykorean
Differential Revision: D63666361
Pulled By: hx235
fbshipit-source-id: 8c9b1bea9b06ff1e3ed95c576aec6705159af137
Summary:
The write unix time from non L0 files are not surfaced properly because the level's wrapper iterator doesn't have a `write_unix_time` implementation that delegates to the corresponding file. The unit test didn't catch this because it incorrectly destroy the old db and reopen to check write time, instead of just reopen and check. This fix also include a change to support ldb's scan command to get write time for easier debugging.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13057
Test Plan: Updated unit tests
Reviewed By: pdillinger
Differential Revision: D64015107
Pulled By: jowlyzhang
fbshipit-source-id: 244474f78a034f80c9235eea2aa8a0f4e54dff59
Summary:
https://github.com/facebook/rocksdb/blob/263fa15b445935e8229063a080e22a405276df2f/CMakeLists.txt#L44
`HOMEPAGE_URL` is introduced into CMake since 3.12. Compiling RocksDB with CMake ver < 3.12 triggers `CMake Error: Could not find cmake module file: CMakeDetermineHOMEPAGE_URLCompiler.cmake` error.
2 options to fix it:
* Remove `HOMEPAGE_URL`, since it appears to have no practical effect.
* Update RocksDB's minimum required CMake version to 3.12.
This PR chose the second option.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13056
Reviewed By: jaykorean
Differential Revision: D63993577
Pulled By: cbi42
fbshipit-source-id: a6278af6916fcdace19a6c9baaf7986037bff720
Summary:
Stress test detects this variable could potentially overflow, so added some runtime handling to avoid it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13046
Test Plan: Existing tests
Reviewed By: hx235
Differential Revision: D63911396
Pulled By: jowlyzhang
fbshipit-source-id: 7c9abcd74ac9937b211c0ea4bb683677390837c5
Summary:
a small CF can trigger parallel compaction that applies to the entire DB. This is because the bottommost file size of a small CF can be too small compared to l0 files when a l0->lbase compaction happens. We prevent this by requiring some minimum on the compaction debt.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13054
Test Plan: updated unit test.
Reviewed By: hx235
Differential Revision: D63861042
Pulled By: cbi42
fbshipit-source-id: 43bbf327988ef0ef912cd2fc700e3d096a8d2c18
Summary:
This PR added some optimizations for the per key handling for SST file for the user-defined timestamps in Memtable only feature. CPU profiling shows this part is a big culprit for regression. This optimization saves some string construction/destruction/appending/copying. vector operations like reserve/emplace_back.
When iterating keys in a block, we need to copy some shared bytes from previous key, put it together with the non shared bytes and find a right location to pad the min timestamp. Previously, we create a tmp local string buffer to first construct the key from its pieces, and then copying this local string's content into `IterKey`'s buffer. To avoid having this local string and to avoid this extra copy. Instead of piecing together the key in a local string first, we just track all the pieces that make this key in a reused Slice array. And then copy the pieces in order into `IterKey`'s buffer. Since the previous key should be kept intact while we are copying some shared bytes from it, we added a secondary buffer in `IterKey` and alternate between primary buffer and secondary buffer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13031
Test Plan: Existing tests.
Reviewed By: ltamasi
Differential Revision: D63416531
Pulled By: jowlyzhang
fbshipit-source-id: 9819b0e02301a2dbc90621b2fe4f651bc912113c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13052
Currently, `MultiCfIteratorImpl` uses `std::function`s for `reset_func_` and `populate_func_`, which uses type erasure and has a performance overhead. The patch turns `MultiCfIteratorImpl` into a template that takes the two function object types as template parameters, and changes `AttributeGroupIteratorImpl` and `CoalescingIterator` so they pass in function objects of named types (as opposed to lambdas).
Reviewed By: jaykorean
Differential Revision: D63802598
fbshipit-source-id: e202f6d80c9054335e5b2571051a67a9e012c2d0
Summary:
There was a crash test Bus Error crash in `IndexBlockIter::SeekToFirstImpl()` <- .. <-
`BlockBasedTable::~BlockBasedTable()` with `--mmap_read=1`, which suggests some kind of incompatibility that I haven't diagnosed. Bus Error is uncommon these days as CPUs support unaligned reads, but are associated with mmap problems.
Because mmap reads really only make sense without block cache, it's not a concerning loss to essentially disable the combination.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13051
Test Plan: watch crash test
Reviewed By: jowlyzhang
Differential Revision: D63795069
Pulled By: pdillinger
fbshipit-source-id: 6c823c619840086b5c9cff53dbc7470662b096be
Summary:
This PR makes file ingestion job's flush wait a bit further until the SuperVersion is also updated. This is necessary since follow up operations will use the current SuperVersion to do range overlapping check and level assignment.
In debug mode, file ingestion job's second `NeedsFlush` call could have been invoked when the memtables are flushed but the SuperVersion hasn't been updated yet, triggering the assertion.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13045
Test Plan:
Existing tests
Manually stress tested
Reviewed By: cbi42
Differential Revision: D63671151
Pulled By: jowlyzhang
fbshipit-source-id: 95a169e58a7e59f6dd4125e7296e9060fe4c63a7
Summary:
... to note that memory may not be freed when reusing a transaction. This means reusing a large transaction can cause excessive memory usage and it may be better to destruct the transaction object in some cases.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13042
Test Plan: no code change.
Reviewed By: jowlyzhang
Differential Revision: D63570612
Pulled By: cbi42
fbshipit-source-id: f19ff556f76d54831fb94715e8808035d07e25fa
Summary:
The following DBOptions were not being propagated through BuildDBOptions, which could at least lead to settings being lost through `GetOptionsFromString()`, possibly elsewhere as well:
* background_close_inactive_wals
* write_dbid_to_manifest
* write_identity_file
* prefix_seek_opt_in_only
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13038
Test Plan:
This problem was not being caught by
OptionsSettableTest.DBOptionsAllFieldsSettable when the option was omitted from both options_helper.cc and options_settable_test.cc. I have added to the test to catch future instances (and the updated test was how I found three of the four missing options).
The same kind of bug seems to be caught by
ColumnFamilyOptionsAllFieldsSettable, and AFAIK analogous code does not exist for BlockBasedTableOptions.
Reviewed By: ltamasi
Differential Revision: D63483779
Pulled By: pdillinger
fbshipit-source-id: a5d5f6e434174bacb8e5d251b767e81e62b7225a
Summary:
When an item is inserted into the compressed secondary cache, this PR calculates the charge using the malloc_usable_size of the allocated memory, as well as the unique pointer allocation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13032
Test Plan: New unit test
Reviewed By: pdillinger
Differential Revision: D63418493
Pulled By: anand1976
fbshipit-source-id: 1db2835af6867442bb8cf6d9bf412e120ddd3824
Summary:
If the lowest_used_cache_tier DB option is set to kVolatileTier, skip insertion of compressed blocks into the secondary cache. Previously, these were always inserted into the secondary cache via the InsertSaved() method, leading to pollution of the secondary cache with blocks that would never be read.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13030
Test Plan: Add a new unit test
Reviewed By: pdillinger
Differential Revision: D63329841
Pulled By: anand1976
fbshipit-source-id: 14d2fce2ed309401d9ad4d2e7c356218b6673f7b
Summary:
Add the following to the `CompactionServiceJobInfo`
- compaction_reason
- is_full_compaction
- is_manual_compaction
- bottommost_level
Added `is_remote_compaction` to the `CompactionJobStats` and set initial values to avoid UB for uninitialized values.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13029
Test Plan:
```
./compaction_service_test --gtest_filter="*CompactionInfo*"
```
Reviewed By: anand1976
Differential Revision: D63322878
Pulled By: jaykorean
fbshipit-source-id: f02a66ca45e660b9d354a43837d8ec6beb7621fb
Summary:
With some new use cases onboarding to prefix extractors/seek/filters, one of the risks is existing iterator code, e.g. for maintenance tasks, being unintentionally subject to prefix seek semantics. This is a longstanding known design flaw with prefix seek, and `prefix_same_as_start` and `auto_prefix_mode` were steps in the direction of making that obsolete. However, we can't just immediately set `total_order_seek` to true by default, because that would impact so much code instantly.
Here we add a new DB option, `prefix_seek_opt_in_only` that basically allows users to transition to the future behavior when they are ready. When set to true, all iterators will be treated as if `total_order_seek=true` and then the only ways to get prefix seek semantics are with `prefix_same_as_start` or `auto_prefix_mode`.
Related fixes / changes:
* Make sure that `prefix_same_as_start` and `auto_prefix_mode` are compatible with (or override) `total_order_seek` (depending on your interpretation).
* Fix a bug in which a new iterator after dynamically changing the prefix extractor might mix different prefix semantics between memtable and SSTs. Both should use the latest extractor semantics, which means iterators ignoring memtable prefix filters with an old extractor. And that means passing the latest prefix extractor to new memtable iterators that might use prefix seek. (Without the fix, the test added for this fails in many ways.)
Suggested follow-up:
* Investigate a FIXME where a MergeIteratorBuilder is created in db_impl.cc. No unit test detects a change in value that should impact correctness.
* Make memtable prefix bloom compatible with `auto_prefix_mode`, which might require involving the memtablereps because we don't know at iterator creation time (only seek time) whether an auto_prefix_mode seek will be a prefix seek.
* Add `prefix_same_as_start` testing to db_stress
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13026
Test Plan:
tests updated, added. Add combination of `total_order_seek=true` and `auto_prefix_mode=true` to stress test. Ran `make blackbox_crash_test` for a long while.
Manually ran tests with `prefix_seek_opt_in_only=true` as default, looking for unexpected issues. I inspected most of the results and migrated many tests to be ready for such a change (but not all).
Reviewed By: ltamasi
Differential Revision: D63147378
Pulled By: pdillinger
fbshipit-source-id: 1f4477b730683d43b4be7e933338583702d3c25e
Summary:
We've been serializing and deserializing DBOptions and CFOptions (and other CF into) as part of `CompactionServiceInput`. These are all readily available in the OPTIONS file and the remote worker can read the OPTIONS file to obtain the same information. This helps reducing the size of payload significantly.
In a very rare scenario if the OPTIONS file is purged due to options change by primary host at the same time while the remote host is loading the latest options, it may fail. In this case, we just retry once.
This also solves the problem where we had to open the default CF with the CFOption from another CF if the remote compaction is for a non-default column family. (TODO comment in /db_impl_secondary.cc)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13025
Test Plan:
Unit Tests
```
./compaction_service_test
```
```
./compaction_job_test
```
Also tested with Meta's internal Offload Infra
Reviewed By: anand1976, cbi42
Differential Revision: D63100109
Pulled By: jaykorean
fbshipit-source-id: b7162695e31e2c5a920daa7f432842163a5b156d
Summary:
This PR allows a Cache object to be created using the object registry.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13024
Reviewed By: pdillinger
Differential Revision: D63043233
Pulled By: anand1976
fbshipit-source-id: 5bc3f7c29b35ad62638ff8205451303e2cecea9d
Summary:
Per customer request, we should not merge multiple SST files together during temperature change compaction, since this can cause FIFO TTL compactions to be delayed. This PR changes the compaction picking logic to pick one file at a time.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13018
Test Plan: * updated some existing unit tests to test this new behavior.
Reviewed By: jowlyzhang
Differential Revision: D62883292
Pulled By: cbi42
fbshipit-source-id: 6a9fc8c296b5d9b17168ef6645f25153241c8b93
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13022
Currently, `blob_garbage_collection_force_threshold` applies to the oldest batch of blob files, which is typically only a small subset of the blob files currently eligible for garbage collection. This can result in a form of head-of-line blocking: no GC-triggered compactions will be scheduled if the oldest batch does not currently exceed the threshold, even if a lot of higher-numbered blob files do. This can in turn lead to high space amplification that exceeds the soft bound implicit in the force threshold (e.g. 50% would suggest a space amp of <2 and 75% would imply a space amp of <4). The patch changes the semantics of this configuration threshold to apply to the entire set of blob files that are eligible for garbage collection based on `blob_garbage_collection_age_cutoff`. This provides more intuitive semantics for the option and can provide a better write amp/space amp trade-off. (Note that GC-triggered compactions still pick the same SST files as before, so triggered GC still targets the oldest the blob files.)
Reviewed By: jowlyzhang
Differential Revision: D62977860
fbshipit-source-id: a999f31fe9cdda313de513f0e7a6fc707424d4a3
Summary:
* Set write_dbid_to_manifest=true by default
* Add new option write_identity_file (default true) that allows us to opt-in to future behavior without identity file
* Refactor related DB open code to minimize code duplication
_Recommend hiding whitespace changes for review_
Intended follow-up: add support to ldb for reading and even replacing the DB identity in the manifest. Could be a variant of `update_manifest` command or based on it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13019
Test Plan: unit tests and stress test updated for new functionality
Reviewed By: anand1976
Differential Revision: D62898229
Pulled By: pdillinger
fbshipit-source-id: c08b25cf790610b034e51a9de0dc78b921abbcf0
Summary:
Add an option `--only_print_seqno_gaps` for wal dump to help with debugging. This option will check the continuity of sequence numbers in WAL logs, assuming `seq_per_batch` is false. `--walfile` option now also takes a directory, and it will check all WAL logs in the directory in chronological order.
When a gap is found, we can further check if it's related to operations like external file ingestion.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13014
Test Plan: Manually tested
Reviewed By: ltamasi
Differential Revision: D62989115
Pulled By: jowlyzhang
fbshipit-source-id: 22e3326344e7969ff9d5091d21fec2935770fbc7
Summary:
There was a subtle design/contract bug in the previous version of range filtering in experimental.h If someone implemented a key segments extractor with "all or nothing" fixed size segments, that could result in unsafe range filtering. For example, with two segments of width 3:
```
x = 0x|12 34 56|78 9A 00|
y = 0x|12 34 56||78 9B
z = 0x|12 34 56|78 9C 00|
```
Segment 1 of y (empty) is out of order with segment 1 of x and z.
I have re-worked the contract to make it clear what does work, and implemented a standard extractor for fixed-size segments, CappedKeySegmentsExtractor. The safe approach for filtering is to consume as much as is available for a segment in the case of a short key.
I have also added support for min-max filtering with reverse byte-wise comparator, which is probably the 2nd most common comparator for RocksDB users (because of MySQL). It might seem that a min-max filter doesn't care about forward or reverse ordering, but it does when trying to determine whether in input range from segment values v1 to v2, where it so happens that v2 is byte-wise less than v1, is an empty forward interval or a non-empty reverse interval. At least in the current setup, we don't have that context.
A new unit test (with some refactoring) tests CappedKeySegmentsExtractor, reverse byte-wise comparator, and the corresponding min-max filter.
I have also (contractually / mathematically) generalized the framework to comparators other than the byte-wise comparator, and made other generalizations to make the extractor limitations more explicitly connected to the particular filters and filtering used--at least in description.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13005
Test Plan: added unit tests as described
Reviewed By: jowlyzhang
Differential Revision: D62769784
Pulled By: pdillinger
fbshipit-source-id: 0d41f0d0273586bdad55e4aa30381ebc861f7044
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13015
`Close()`ing a database now releases tracked files in `SstFileManager`. Previously this space would be leaked until the database was later reopened.
Reviewed By: jowlyzhang
Differential Revision: D62590773
fbshipit-source-id: 5461bd253d974ac4967ad52fee92e2650f8a9a28
Summary:
A recent crash test failure shows that auto recovery from WAL write failure can cause CFs to be inconsistent. A unit test repro in P1569398553. The following is an example sequence of events:
```
0. manual_wal_flush is true. There are multiple CFs in a DB.
1. Submit a write batch with updates to multiple CF
2. A FlushWAL or a memtable swtich that will try to write the buffered WAL data. Fail this write so that buffered WAL data is dropped: https://github.com/facebook/rocksdb/blob/4b1d595306fae602b56d2aa5128b11b1162bfa81/file/writable_file_writer.cc#L624
The error needs to be retryable to start background auto recovery.
3. One CF successfully flushes its memtable during auto recovery.
4. Crash the process.
5. Reopen the DB, one CF will have the update as a result of successful flush. Other CFs will miss all the updates in the write batch since WAL does not have them.
```
This can happen if a users configures manual_wal_flush, uses more than one CF, and can hit retryable error for WAL writes. This PR is a short-term fix that upgrades WAL related errors to fatal and not trigger auto recovery.
A long-term fix may be not drop buffered WAL data by checking how much data is actually written, or require atomically flushing all column families during error recovery from this kind of errors.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12995
Test Plan:
added unit test to check error severity and if recovery is triggered. A crash test repro command that fails in a few runs before this PR:
```
python3 ./tools/db_crashtest.py blackbox --interval=60 --metadata_write_fault_one_in=1000 --column_families=10 --exclude_wal_from_write_fault_injection=0 --manual_wal_flush_one_in=1000 --WAL_size_limit_MB=10240 --WAL_ttl_seconds=0 --acquire_snapshot_one_in=10000 --adaptive_readahead=1 --adm_policy=1 --advise_random_on_open=1 --allow_data_in_errors=True --allow_fallocate=1 --async_io=0 --auto_readahead_size=0 --avoid_flush_during_recovery=1 --avoid_flush_during_shutdown=1 --avoid_unnecessary_blocking_io=0 --backup_max_size=104857600 --backup_one_in=0 --batch_protection_bytes_per_key=0 --bgerror_resume_retry_interval=100 --block_align=1 --block_protection_bytes_per_key=0 --block_size=16384 --bloom_before_level=2147483647 --bottommost_compression_type=none --bottommost_file_compaction_delay=0 --bytes_per_sync=0 --cache_index_and_filter_blocks=1 --cache_index_and_filter_blocks_with_high_priority=1 --cache_size=33554432 --cache_type=auto_hyper_clock_cache --charge_compression_dictionary_building_buffer=0 --charge_file_metadata=1 --charge_filter_construction=1 --charge_table_reader=0 --check_multiget_consistency=0 --check_multiget_entity_consistency=0 --checkpoint_one_in=0 --checksum_type=kxxHash64 --clear_column_family_one_in=0 --compact_files_one_in=0 --compact_range_one_in=0 --compaction_pri=1 --compaction_readahead_size=1048576 --compaction_ttl=0 --compress_format_version=1 --compressed_secondary_cache_size=8388608 --compression_checksum=0 --compression_max_dict_buffer_bytes=0 --compression_max_dict_bytes=0 --compression_parallel_threads=4 --compression_type=none --compression_use_zstd_dict_trainer=1 --compression_zstd_max_train_bytes=0 --continuous_verification_interval=0 --daily_offpeak_time_utc= --data_block_index_type=0 --db_write_buffer_size=0 --decouple_partitioned_filters=1 --default_temperature=kCold --default_write_temperature=kWarm --delete_obsolete_files_period_micros=30000000 --delpercent=4 --delrangepercent=1 --destroy_db_initially=0 --detect_filter_construct_corruption=0 --disable_file_deletions_one_in=1000000 --disable_manual_compaction_one_in=1000000 --disable_wal=0 --dump_malloc_stats=1 --enable_checksum_handoff=1 --enable_compaction_filter=0 --enable_custom_split_merge=0 --enable_do_not_compress_roles=0 --enable_index_compression=0 --enable_memtable_insert_with_hint_prefix_extractor=0 --enable_pipelined_write=1 --enable_sst_partitioner_factory=0 --enable_thread_tracking=1 --enable_write_thread_adaptive_yield=1 --error_recovery_with_no_fault_injection=1 --fail_if_options_file_error=1 --fifo_allow_compaction=1 --file_checksum_impl=big --fill_cache=1 --flush_one_in=1000000 --format_version=6 --get_all_column_family_metadata_one_in=1000000 --get_current_wal_file_one_in=0 --get_live_files_apis_one_in=10000 --get_properties_of_all_tables_one_in=1000000 --get_property_one_in=100000 --get_sorted_wal_files_one_in=0 --hard_pending_compaction_bytes_limit=274877906944 --index_block_restart_interval=4 --index_shortening=1 --index_type=0 --ingest_external_file_one_in=0 --initial_auto_readahead_size=16384 --inplace_update_support=0 --iterpercent=10 --key_len_percent_dist=1,30,69 --key_may_exist_one_in=100000 --last_level_temperature=kWarm --level_compaction_dynamic_level_bytes=0 --lock_wal_one_in=10000 --log_file_time_to_roll=0 --log_readahead_size=0 --long_running_snapshots=0 --lowest_used_cache_tier=2 --manifest_preallocation_size=5120 --mark_for_compaction_one_file_in=10 --max_auto_readahead_size=0 --max_background_compactions=20 --max_bytes_for_level_base=10485760 --max_key=100000 --max_key_len=3 --max_log_file_size=0 --max_manifest_file_size=1073741824 --max_sequential_skip_in_iterations=16 --max_total_wal_size=0 --max_write_batch_group_size_bytes=16777216 --max_write_buffer_number=10 --max_write_buffer_size_to_maintain=2097152 --memtable_insert_hint_per_batch=1 --memtable_max_range_deletions=0 --memtable_prefix_bloom_size_ratio=0.001 --memtable_protection_bytes_per_key=2 --memtable_whole_key_filtering=0 --memtablerep=skip_list --metadata_charge_policy=1 --metadata_read_fault_one_in=0 --min_write_buffer_number_to_merge=1 --mmap_read=1 --mock_direct_io=False --nooverwritepercent=1 --num_file_reads_for_auto_readahead=2 --open_files=100 --open_metadata_read_fault_one_in=0 --open_metadata_write_fault_one_in=0 --open_read_fault_one_in=0 --open_write_fault_one_in=0 --optimize_filters_for_hits=0 --optimize_filters_for_memory=0 --optimize_multiget_for_io=0 --paranoid_file_checks=1 --paranoid_memory_checks=0 --partition_filters=0 --partition_pinning=2 --pause_background_one_in=10000 --periodic_compaction_seconds=0 --prefix_size=8 --prefixpercent=5 --prepopulate_block_cache=0 --preserve_internal_time_seconds=0 --progress_reports=0 --promote_l0_one_in=0 --read_amp_bytes_per_bit=0 --read_fault_one_in=0 --readahead_size=524288 --readpercent=45 --recycle_log_file_num=0 --reopen=0 --report_bg_io_stats=0 --reset_stats_one_in=10000 --sample_for_compression=5 --secondary_cache_fault_one_in=0 --secondary_cache_uri= --set_options_one_in=10000 --skip_stats_update_on_db_open=1 --snapshot_hold_ops=100000 --soft_pending_compaction_bytes_limit=1048576 --sqfc_name=bar --sqfc_version=1 --sst_file_manager_bytes_per_sec=0 --sst_file_manager_bytes_per_truncate=0 --stats_dump_period_sec=600 --stats_history_buffer_size=1048576 --strict_bytes_per_sync=1 --subcompactions=2 --sync=0 --sync_fault_injection=1 --table_cache_numshardbits=6 --target_file_size_base=524288 --target_file_size_multiplier=2 --test_batches_snapshots=0 --top_level_index_pinning=3 --uncache_aggressiveness=8 --universal_max_read_amp=-1 --unpartitioned_pinning=2 --use_adaptive_mutex=1 --use_adaptive_mutex_lru=0 --use_attribute_group=1 --use_delta_encoding=0 --use_direct_io_for_flush_and_compaction=0 --use_direct_reads=0 --use_full_merge_v1=0 --use_get_entity=0 --use_merge=1 --use_multi_cf_iterator=1 --use_multi_get_entity=0 --use_multiget=0 --use_put_entity_one_in=1 --use_sqfc_for_range_queries=0 --use_timed_put_one_in=0 --use_write_buffer_manager=0 --user_timestamp_size=0 --value_size_mult=32 --verification_only=0 --verify_checksum=1 --verify_checksum_one_in=1000000 --verify_compression=1 --verify_db_one_in=100000 --verify_file_checksums_one_in=1000000 --verify_iterator_with_expected_state_one_in=5 --verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=0 --wal_compression=none --write_buffer_size=4194304 --write_dbid_to_manifest=0 --write_fault_one_in=50 --writepercent=35 --ops_per_thread=100000 --preserve_unverified_changes=1
```
Reviewed By: hx235
Differential Revision: D62888510
Pulled By: cbi42
fbshipit-source-id: 308bdbbb8d897cc8eba950155cd0e37cf7eb76fe
Summary: I came across this code while buckifying parts of folly and fizz in open source. This is pretty hacky code and cleaning it up doesn't seem that hard, so I did it.
Reviewed By: zertosh, pdillinger
Differential Revision: D62781766
fbshipit-source-id: 43714bce992c53149d1e619063d803297362fb5d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13010
The OnAddFile cur_compactions_reserved_size_ accounting causes wraparound when re-opening a database with an unowned SstFileManager and during recovery. It was introduced in #4164 which addresses out of space recovery with an unclear purpose. Compaction jobs do this accounting via EnoughRoomForCompaction/OnCompactionCompletion and to my understanding would never reuse a sst file name.
Reviewed By: anand1976
Differential Revision: D62535775
fbshipit-source-id: a7c44d6e0a4b5ff74bc47abfe57c32ca6770243d
Summary:
For SST checksum mismatch corruptions in the read path, RocksDB retries the read if the underlying file system supports verification and reconstruction of data (`FSSupportedOps::kVerifyAndReconstructRead`). There were a couple of places where the retry was missing - reading the SST footer and the properties block. This PR fixes the retry in those cases.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13007
Test Plan: Add new unit tests
Reviewed By: jaykorean
Differential Revision: D62519186
Pulled By: anand1976
fbshipit-source-id: 50aa38f18f2a53531a9fc8d4ccdf34fbf034ed59
Summary:
in ReFitLevel(), we were not setting being_compacted to false after ReFitLevel() is done. This is not a issue if refit level is successful, since new FileMetaData is created for files at the target level. However, if there's an error during RefitLevel(), e.g., Manifest write failure, we should clear the being_compacted field for these files. Otherwise, these files will not be picked for compaction until db reopen.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13009
Test Plan:
existing test.
- stress test failure in T200339331 should not happen anymore.
Reviewed By: hx235
Differential Revision: D62597169
Pulled By: cbi42
fbshipit-source-id: 0ba659806da6d6d4b42384fc95268b2d7bad720e
Summary:
Prepare this internal API to be used by atomic data replacement. The main purpose of this API is to get a `VersionEdit` to mark the entire current `MemTableListVersion` as dropped. Flush needs the similar functionality when installing results, so that logic is refactored into a util function `GetDBRecoveryEditForObsoletingMemTables` to be shared by flush and this internal API.
To test this internal API, flush's result installation is redirected to use this API when it is flushing all the immutable MemTables in debug mode. It should achieve the exact same results, just with a duplicated `VersionEdit::log_number` field that doesn't upsets the recovery logic.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13001
Test Plan: Existing tests
Reviewed By: pdillinger
Differential Revision: D62309591
Pulled By: jowlyzhang
fbshipit-source-id: e25914d9a2e281c25ab7ee31a66eaf6adfae4b88
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/13004
The patch extends the buckifier script so it generates a target for `db_bench` as well.
Reviewed By: cbi42
Differential Revision: D62407071
fbshipit-source-id: 0cb98a324ce0598ad84a8675aa77b7d0f91bf40c
Summary:
Add the `ApplyToHandle` method to the `Cache` interface to allow a caller to request the invocation of a callback on the given cache handle. The goal here is to allow a cache that manages multiple cache instances to use a callback on a handle to determine which instance it belongs to. For example, the callback can hash the key and use that to pick the correct target instance. This is useful to redirect methods like `Ref` and `Release`, which don't know the cache key.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12987
Reviewed By: pdillinger
Differential Revision: D62151907
Pulled By: anand1976
fbshipit-source-id: e4ffbbb96eac9061d2ab0e7e1739eea5ebb1cd58
Summary:
`Compaction` is already creating its own ref for the input Version: https://github.com/facebook/rocksdb/blob/4b1d595306fae602b56d2aa5128b11b1162bfa81/db/compaction/compaction.cc#L73
And properly Unref it during destruction:
https://github.com/facebook/rocksdb/blob/4b1d595306fae602b56d2aa5128b11b1162bfa81/db/compaction/compaction.cc#L450
This PR redirects compaction's access of `cfd->current()` to this input `Version`, to prepare for when a column family's data can be replaced all together, and `cfd->current()` is not safe to access for a compaction job. Because a new `Version` with just some other external files could be installed as `cfd->current()`. The compaction job's expectation of the current `Version` and the corresponding storage info to always have its input files will no longer be guaranteed.
My next follow up is to do a similar thing for flush, also to prepare it for when a column family's data can be replaced. I will make it create its own reference of the current `MemTableListVersion` and use it as input, all flush job's access of memtables will be wired to that input `MemTableListVersion`. Similarly this reference will be unreffed during a flush job's destruction.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12992
Test Plan: Existing tests
Reviewed By: pdillinger
Differential Revision: D62212625
Pulled By: jowlyzhang
fbshipit-source-id: 9a781213469cf366857a128d50a702af683a046a
Summary:
The `SchedulePending*` API is a bit confusing since it doesn't immediately schedule the work and can be confused with the actual scheduling. So I have changed these to be `EnqueuePending*` and added some documentation for the corresponding state transitions of these background work.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12994
Test Plan: existing tests
Reviewed By: cbi42
Differential Revision: D62252746
Pulled By: jowlyzhang
fbshipit-source-id: ee68be6ed33070cad9a5004b7b3e16f5bcb041bf
Summary:
* https://github.com/facebook/rocksdb/issues/12936 was insufficient to fix the std::optional false positives. Making a fix validated in CI this time (see https://github.com/facebook/rocksdb/issues/12991)
* valgrind grinds to a halt on startup on my dev machine apparently because it expects internet access. Disable its attempts to access the internet when git is using a proxy.
* Move PORTABLE=1 from CI job to the Makefile. Without it, valgrind complains about illegal instructions (too new)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12990
Test Plan: manual, watch nightly valgrind job
Reviewed By: ltamasi
Differential Revision: D62203242
Pulled By: pdillinger
fbshipit-source-id: a611b08da7dbd173b0709ed7feb0578729553a17
Summary:
It appears the arm testsuite is failing because it is building without snappy, which is causing the SST files not to be compressed, which somehow causes these tests to fail. Manually setting LZ4 which is already required.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12993
Test Plan: reproduced and verified fix on ARM laptop
Reviewed By: anand1976
Differential Revision: D62216451
Pulled By: pdillinger
fbshipit-source-id: 3f21fcd9be0edaa66c7eca0cb7d56b998171e263
Summary:
`ignore_unknown_options=true` had an undocumented behavior of having no effect (disallow unknown options) if reading from the same or older major.minor version. Presumably this was intended to catch unintentional addition of new options in a patch release, but there is no automated version compatibility testing between patch releases. So this was a bad choice without such testing support, because it just means users would hit the failure in case of adding features to a patch release.
In this diff we respect ignore_unknown_options when reading a file from any newer version, even patch versions, and document this behavior in the API.
I don't think it's practical or necessary to test among patch releases in check_format_compatible.sh. This seems like an exceptional case of applying a *different semantics* to patch version updates than to minor/major versions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12989
Test Plan: unit test updated (and refactored)
Reviewed By: jaykorean
Differential Revision: D62168738
Pulled By: pdillinger
fbshipit-source-id: fb3c3ef30f0bbad0d5ffcc4570fb9ef963e7daac
Summary:
`check_format_compatible` script was broken due to extra comma added in 5b8f5cbcf4
e.g. https://github.com/facebook/rocksdb/actions/runs/10505042711/job/29101787220
```
...
2024-08-23T11:44:15.0175202Z == Building 9.5.fb, debug
2024-08-23T11:44:15.0190592Z fatal: ambiguous argument '_tmp_origin/9.5.fb,': unknown revision or path not in the working tree.
...
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12988
Test Plan:
```
tools/check_format_compatible.sh
```
```
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 8.6.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/8.6.fb to /tmp/rocksdb_format_compatible_jewoongh/db/8.6.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 8.7.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/8.7.fb to /tmp/rocksdb_format_compatible_jewoongh/db/8.7.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 8.8.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/8.8.fb to /tmp/rocksdb_format_compatible_jewoongh/db/8.8.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 8.9.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/8.9.fb to /tmp/rocksdb_format_compatible_jewoongh/db/8.9.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 8.10.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/8.10.fb to /tmp/rocksdb_format_compatible_jewoongh/db/8.10.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 8.11.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/8.11.fb to /tmp/rocksdb_format_compatible_jewoongh/db/8.11.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.0.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.0.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.0.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.1.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.1.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.1.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.2.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.2.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.2.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.3.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.3.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.3.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.4.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.4.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.4.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.5.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.5.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.5.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
== Use HEAD (48339d2a65670211bc9c204364a2127ba9b2a460) to open DB generated using 9.6.fb...
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/9.6.fb to /tmp/rocksdb_format_compatible_jewoongh/db/9.6.fb/db_dump.txt
== Dumping data from /tmp/rocksdb_format_compatible_jewoongh/db/current to /tmp/rocksdb_format_compatible_jewoongh/db/current/db_dump.txt
==== Compatibility Test PASSED ====
```
Reviewed By: pdillinger
Differential Revision: D62162454
Pulled By: jaykorean
fbshipit-source-id: 562225c6cb27e0eb66f241a6f9424dc624d8c837
Summary:
Met the following error while compiling the project.
```
build_tools/check-sources.sh
utilities/fault_injection_fs.cc:509: // If there<E2><80><99>s no injected error, then cb will be called asynchronously when
utilities/fault_injection_fs.cc:510: // target_ actually finishes the read. But if there<E2><80><99>s an injected error, it
utilities/fault_injection_fs.cc:512: // isn<E2><80><99>t invoked at all.
^^^^ Use only ASCII characters in source files
make[1]: *** [Makefile:1291: check-sources] Error 1
make[1]: Leaving directory '/home/janus/Github/symious/rocksdb'
make: *** [Makefile:1084: check] Error 2
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12972
Reviewed By: hx235
Differential Revision: D61923865
Pulled By: cbi42
fbshipit-source-id: 63af0a38fea15e09a860895bdd5ed0a57700e447
Summary:
Add option `IngestExternalFileOptions::link_files` that hard links input files and preserves original file links after ingestion, unlike `move_files` which will unlink input files after ingestion. This can be useful when being used together with `allow_db_generated_files` to ingest files from another DB. Also reverted the change to `move_files` in https://github.com/facebook/rocksdb/issues/12959 to simplify the contract so that it will always unlink input files without exception.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12980
Test Plan: updated unit test `ExternSSTFileLinkFailFallbackTest.LinkFailFallBackExternalSst` to test that input files will not be unlinked.
Reviewed By: pdillinger
Differential Revision: D61925111
Pulled By: cbi42
fbshipit-source-id: eadaca72e1ae5288bdd195d57158466e5656fa62
Summary:
There are several crash test failures due to DB verification failure. Retain some trace history in the expected state directory to make debugging easier.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12978
Reviewed By: cbi42
Differential Revision: D61864921
Pulled By: anand1976
fbshipit-source-id: 9f3f37b7e1e958bc89a3cf0373182354c2c1aa3b
Summary:
Followed instruction per https://docs.github.com/en/actions/writing-workflows/workflow-syntax-for-github-actions#defining-access-for-the-github_token-scopes
It turns out that we did not need any of these except `Metadata: read`.
Before
```
GITHUB_TOKEN Permissions
Actions: write
Attestations: write
Checks: write
Contents: write
Deployments: write
Discussions: write
Issues: write
Metadata: read
Packages: write
Pages: write
PullRequests: write
RepositoryProjects: write
SecurityEvents: write
Statuses: write
```
After
```
GITHUB_TOKEN Permissions
Metadata: read
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12973
Test Plan: GitHub Actions triggered by this PR
Reviewed By: cbi42
Differential Revision: D61812651
Pulled By: jaykorean
fbshipit-source-id: 4413756c93f503e8b2fb77eb8b684ef9e6a6c13d
Summary:
so `IngestExternalFileOptions::move_files` and `IngestExternalFileOptions::allow_db_generated_files` are now compatible. The original file links won't be removed if `allow_db_generated_files` is true. This is to prevent deleting files from another DB.
There was a [comment](https://github.com/facebook/rocksdb/pull/12750#discussion_r1684509620) in https://github.com/facebook/rocksdb/issues/12750 about how exactly-once ingestion would work with `move_files`. I've discussed with customer and decided that it can be done by reading the target DB to see if it contains any ingested key.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12959
Test Plan: updated unit tests `IngestDBGeneratedFileTest*` to enable `move_files`.
Reviewed By: jowlyzhang
Differential Revision: D61703480
Pulled By: cbi42
fbshipit-source-id: 6b4294369767f989a2f36bbace4ca3c0257aeaf7
Summary:
.. so that appropriate implementations can return temperature information from GetChildrenFileAttributes
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12965
Test Plan: just an API placeholder for now
Reviewed By: anand1976
Differential Revision: D61748199
Pulled By: pdillinger
fbshipit-source-id: b457e324cb451e836611a0bf630c3da0f30a8abf
Summary:
We have a request to use the cold tier as primary source of truth for the DB, and to best support such use cases and to complement the existing options controlling SST file temperatures, we add two new DB options:
* `metadata_write_temperature` for DB "small" files that don't contain much user data
* `wal_write_temperature` for WALs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12957
Test Plan: Unit test included, though it's hard to be sure we've covered all the places
Reviewed By: jowlyzhang
Differential Revision: D61664815
Pulled By: pdillinger
fbshipit-source-id: 8e19c9dd8fd2db059bb15f74938d6bc12002e82b
Summary:
Disabling the job temporarily. We will re-enable this when ready again
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12964
Reviewed By: ltamasi
Differential Revision: D61740941
Pulled By: jaykorean
fbshipit-source-id: 167e50c4f5e38d508a8e56633261611467f30690
Summary:
Issue: MultiGet(PinnableSlice) can't read out all timestamps.
Fixed the impl, and added an UT as well. In the original impl, if MultiGet reads multiple column families, a later column family would clean up timestamps of previous column family.
Fix: https://github.com/facebook/rocksdb/issues/12950#issue-2476996580
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12943
Reviewed By: anand1976
Differential Revision: D61729257
Pulled By: pdillinger
fbshipit-source-id: 55267c26076c8a59acedd27e14714711729a40df
Summary:
When merged into internal code base we see the following error. This should fix it.
```
Actions failed:
[2024-08-20T07:45:53.879-07:00] Action failed: fbcode//rocksdb/src:rocksdb_lib (cfg:macos-arm64-macosx-clang17-no-san#e5847010950663ca) (cxx_compile util/write_batch_util.cc)
[2024-08-20T07:45:53.879-07:00] Remote command returned non-zero exit code 1
[2024-08-20T07:45:53.879-07:00] Remote action, reproduce with: `frecli cas download-action 2fe3749f2d3ea6107cce103d4e2be1dcc76a9df797bae308cde5eaccc65201b7:145`
fbcode/rocksdb/src/include/rocksdb/write_batch.h:460:14: error: no template named 'unordered_map' in namespace 'std'; did you mean 'unordered_set'?
const std::unordered_map<uint32_t, size_t>& GetColumnFamilyToTimestampSize() {
~~~~~^~~~~~~~~~~~~
fbcode/rocksdb/src/include/rocksdb/write_batch.h:540:8: error: no template named 'unordered_map' in namespace 'std'; did you mean 'unordered_set'?
std::unordered_map<uint32_t, size_t> cf_id_to_ts_sz_;
~~~~~^~~~~~~~~~~~~
/paragon/pods/259551525/home/execution/3/202ac945754041b6bc424b0c35e42c9d/work/buck-out/v2/gen/fbsource/a90614bbe22ec1d7/xplat/toolchains/minimal_xcode/__clang_genrule__/out/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/__memory/compressed_pair.h:113:3: error: static_assert failed due to requirement '!is_same<unsigned long, unsigned long>::value' "__compressed_pair cannot be instantiated when T1 and T2 are the same type; The current implementation is NOT ABI-compatible with the previous implementation for this configuration"
static_assert((!is_same<_T1, _T2>::value),
^ ~~~~~~~~~~~~~~~~~~~~~~~~~
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12949
Test Plan: CI
Reviewed By: jowlyzhang, cbi42
Differential Revision: D61577604
Pulled By: jaykorean
fbshipit-source-id: 3584a2cd550a303346d80ccc5cc90f4a9b3e2da2
Summary:
This pull request transitions the benchmarking process from CircleCI to GitHub Actions. The benchmarking jobs will now be executed on a self-hosted runner. Unlike the previous CircleCI configuration, where jobs were queued due to the long execution time (nearly 60 minutes per job), the new setup schedules the benchmarking tasks to run every two hours.
Closes https://github.com/facebook/rocksdb/issues/12615
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12885
Reviewed By: pdillinger
Differential Revision: D61422468
Pulled By: jaykorean
fbshipit-source-id: 10535865c849797825f9652e4e9ef367b3d73599
Summary:
# Summary
Mistakenly double-updated the HISTORY.md file by running `unreleased_history/release.sh` after the first commit in https://github.com/facebook/rocksdb/issues/12945. Manually fixing the file to reflect the correct content
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12947
Test Plan: N/A. History file change.
Reviewed By: cbi42
Differential Revision: D61512756
Pulled By: jaykorean
fbshipit-source-id: 50dc7e92a945fa80c7dfd01cc89243fd5eaf0548
Summary:
Main branch cut at defd97bc9.
Updated HISTORY.md, version and format compatibility test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12945
Test Plan: CI
Reviewed By: jowlyzhang
Differential Revision: D61482149
Pulled By: jaykorean
fbshipit-source-id: 4edf7c0a8c6e4df8fcc938bc778dfd02981d0c55
Summary:
add a new CF option `paranoid_memory_checks` that allows additional data integrity validations during read/scan. Currently, skiplist-based memtable will validate the order of keys visited. Further data validation can be added in different layers. The option will be opt-in due to performance overhead.
The motivation for this feature is for services where data correctness is critical and want to detect in-memory corruption earlier. For a corrupted memtable key, this feature can help to detect it during during reads instead of during flush with existing protections (OutputValidator that verifies key order or per kv checksum). See internally linked task for more context.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12889
Test Plan:
* new unit test added for paranoid_memory_checks=true.
* existing unit test for paranoid_memory_checks=false.
* enable in stress test.
Performance Benchmark: we check for performance regression in read path where data is in memtable only. For each benchmark, the script was run at the same time for main and this PR:
* Memtable-only randomread ops/sec:
```
(for I in $(seq 1 50);do ./db_bench --benchmarks=fillseq,readrandom --write_buffer_size=268435456 --writes=250000 --num=250000 --reads=500000 --seed=1723056275 2>&1 | grep "readrandom"; done;) | awk '{ t += $5; c++; print } END { print 1.0 * t / c }';
Main: 608146
PR with paranoid_memory_checks=false: 607727 (- %0.07)
PR with paranoid_memory_checks=true: 521889 (-%14.2)
```
* Memtable-only sequential scan ops/sec:
```
(for I in $(seq 1 50); do ./db_bench--benchmarks=fillseq,readseq[-X10] --write_buffer_size=268435456 --num=1000000 --seed=1723056275 2>1 | grep "\[AVG 10 runs\]"; done;) | awk '{ t += $6; c++; print; } END { printf "%.0f\n", 1.0 * t / c }';
Main: 9180077
PR with paranoid_memory_checks=false: 9536241 (+%3.8)
PR with paranoid_memory_checks=true: 7653934 (-%16.6)
```
* Memtable-only reverse scan ops/sec:
```
(for I in $(seq 1 20); do ./db_bench --benchmarks=fillseq,readreverse[-X10] --write_buffer_size=268435456 --num=1000000 --seed=1723056275 2>1 | grep "\[AVG 10 runs\]"; done;) | awk '{ t += $6; c++; print; } END { printf "%.0f\n", 1.0 * t / c }';
Main: 1285719
PR with integrity_checks=false: 1431626 (+%11.3)
PR with integrity_checks=true: 811031 (-%36.9)
```
The `readrandom` benchmark shows no regression. The scanning benchmarks show improvement that I can't explain.
Reviewed By: pdillinger
Differential Revision: D60414267
Pulled By: cbi42
fbshipit-source-id: a70b0cbeea131f1a249a5f78f9dc3a62dacfaa91
Summary:
Add an optional callback function upon remote compaction temp output installation. This will be internally used for setting the final status in the Offload Infra.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12940
Test Plan:
Unit Test added
```
./compaction_service_test
```
_Also internally tested by manually merging into internal code base_
Reviewed By: anand1976
Differential Revision: D61419157
Pulled By: jaykorean
fbshipit-source-id: 66831685bc403949c26bfc65840dd1900d2a5a67
Summary:
This PR make best efforts recovery more permissive by allowing it to recover incomplete Version that presents a valid point in time view from the user's perspective. Currently, a Version is only valid and saved if all files consisting that Version can be found. With this change, if only a suffix of L0 files (and their associated blob files) are missing, a valid Version is also available to be saved and recover to. Note that we don't do this if the column family was atomically flushed. Because atomic flush also need a consistent view across the column families, we cannot guarantee that if we are recovering to incomplete version.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12938
Test Plan: Existing tests and added unit tests.
Reviewed By: anand1976
Differential Revision: D61414381
Pulled By: jowlyzhang
fbshipit-source-id: f9b73deb34d35ad696ab42315928b656d586262a
Summary:
Partitioned metadata blocks were introduced back in 2017 to deal more gracefully with large DBs where RAM is relatively scarce and some data might be much colder than other data. The feature allows metadata blocks to compete for memory in the block cache against data blocks while alleviating tail latencies and thrash conditions that can arise with large metadata blocks (sometimes megabytes each) that can arise with large SST files. In general, the cost to partitioned metadata is more CPU in accesses (especially for filters where more binary search is needed before hashing can be used) and a bit more memory fragmentation and related overheads.
However the feature has always had a subtle limitation with a subtle effect on performance: index partitions and filter partitions must be cut at the same time, regardless of which wins the space race (hahaha) to metadata_block_size. Commonly filters will be a few times larger than indexes, so index partitions will be under-sized compared to filter (and data) blocks. While this does affect fragmentation and related overheads a bit, I suspect the bigger impact on performance is in the block cache. The coupling of the partition cuts would be defensible if the binary search done to find the filter block was used (on filter hit) to short-circuit binary search to an index partition, but that optimization has not been developed.
Consider two metadata blocks, an under-sized one and a normal-sized one, covering proportional sections of the key space with the same density of read queries. The under-sized one will be more prone to eviction from block cache because it is used less often. This is unfair because of its despite its proportionally smaller cost of keeping in block cache, and most of the cost of a miss to re-load it (random IO) is not proportional to the size (similar latency etc. up to ~32KB).
## This change
Adds a new table option decouple_partitioned_filters allows filter blocks and index blocks to be cut independently. To make this work, the partitioned filter block builder needs to know about the previous key, to generate an appropriate separator for the partition index. In most cases, BlockBasedTableBuilder already has easy access to the previous key to provide to the filter block builder.
This change includes refactoring to pass that previous key to the filter builder when available, with the filter building caching the previous key itself when unavailable, such as during compression dictionary training and some unit tests. Access to the previous key eliminates the need to track the previous prefix, which results in a small SST construction CPU win in prefix filtering cases, regardless of coupling, and possibly a small regression for some non-prefix cases, regardless of coupling, but still overall improvement especially with https://github.com/facebook/rocksdb/issues/12931.
Suggested follow-up:
* Update confusing use of "last key" to refer to "previous key"
* Expand unit test coverage with parallel compression and dictionary training
* Consider an option or enhancement to alleviate under-sized metadata blocks "at the end" of an SST file due to no coordination or awareness of when files are cut.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12939
Test Plan:
unit tests updated. Also did some unit test runs with "hard wired" usage of parallel compression and dictionary training code paths to ensure they were working. Also ran blackbox_crash_test for a while with the new feature.
## SST write performance (CPU)
Using the same testing setup as in https://github.com/facebook/rocksdb/issues/12931 but with -decouple_partitioned_filters=1 in the "after" configuration, which benchmarking shows makes almost no difference in terms of SST write CPU. "After" vs. "before" this PR
```
-partition_index_and_filters=0 -prefix_size=0 -whole_key_filtering=1
923691 vs. 924851 (-0.13%)
-partition_index_and_filters=0 -prefix_size=8 -whole_key_filtering=0
921398 vs. 922973 (-0.17%)
-partition_index_and_filters=0 -prefix_size=8 -whole_key_filtering=1
902259 vs. 908756 (-0.71%)
-partition_index_and_filters=1 -prefix_size=8 -whole_key_filtering=0
917932 vs. 916901 (+0.60%)
-partition_index_and_filters=1 -prefix_size=8 -whole_key_filtering=0
912755 vs. 907298 (+0.60%)
-partition_index_and_filters=1 -prefix_size=8 -whole_key_filtering=1
899754 vs. 892433 (+0.82%)
```
I think this is a pretty good trade, especially in attracting more movement toward partitioned configurations.
## Read performance
Let's see how decoupling affects read performance across various degrees of memory constraint. To simplify LSM structure, we're using FIFO compaction. Since decoupling will overall increase metadata block size, we control for this somewhat with an extra "before" configuration with larger metadata block size setting (8k instead of 4k). Basic setup:
```
(for CS in 0300 1200; do TEST_TMPDIR=/dev/shm/rocksdb1 ./db_bench -benchmarks=fillrandom,flush,readrandom,block_cache_entry_stats -num=5000000 -duration=30 -disable_wal=1 -write_buffer_size=30000000 -bloom_bits=10 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters=1 -statistics=1 -cache_size=${CS}000000 -metadata_block_size=4096 -decouple_partitioned_filters=1 2>&1 | tee results-$CS; done)
```
And read ops/s results:
```CSV
Cache size MB,After/decoupled/4k,Before/4k,Before/8k
3,15593,15158,12826
6,16295,16693,14134
10,20427,20813,18459
20,27035,26836,27384
30,33250,31810,33846
60,35518,32585,35329
100,36612,31805,35292
300,35780,31492,35481
1000,34145,31551,35411
1100,35219,31380,34302
1200,35060,31037,34322
```
If you graph this with log scale on the X axis (internal link: https://pxl.cl/5qKRc), you see that the decoupled/4k configuration is essentially the best of both the before/4k and before/8k configurations: handles really tight memory closer to the old 4k configuration and handles generous memory closer to the old 8k configuration.
Reviewed By: jowlyzhang
Differential Revision: D61376772
Pulled By: pdillinger
fbshipit-source-id: fc2af2aee44290e2d9620f79651a30640799e01f
Summary:
This is in part a refactoring / simplification to set up for "decoupled" partitioned filters and in part to fix an intentional regression for a correctness fix in https://github.com/facebook/rocksdb/issues/12872. Basically, we are taking out some complexity of the filter block builders, and pushing part of it (simultaneous de-duplication of prefixes and whole keys) into the filter bits builders, where it is more efficient by operating on hashes (rather than copied keys).
Previously, the FullFilterBlockBuilder had a somewhat fragile and confusing set of conditions under which it would keep a copy of the most recent prefix and most recent whole key, along with some other state that is essentially redundant. Now we just track (always) the previous prefix in the PartitionedFilterBlockBuilder, to deal with the boundary prefix Seek filtering problem. (Btw, the next PR will optimize this away since BlockBasedTableReader already tracks the previous key.) And to deal with the problem of de-duplicating both whole keys and prefixes going into a single filter, we add a new function to FilterBitsBuilder that has that extra de-duplication capabilty, which is relatively efficient because we only have to cache an extra 64-bit hash, not a copied key or prefix. (The API of this new function is somewhat awkward to avoid a small CPU regression in some cases.)
Also previously, there was awkward logic split between FullFilterBlockBuilder and PartitionedFilterBlockBuilder to deal with some things specific to partitioning. And confusing names like Add vs. AddKey. FullFilterBlockBuilder is much cleaner and simplified now.
The splitting of PartitionedFilterBlockBuilder::MaybeCutAFilterBlock into DecideCutAFilterBlock and CutAFilterBlock is to address what would have been a slight performance regression in some cases. The split allows for more intruction-level parallelism by reducing unnecessary control dependencies.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12931
Test Plan:
existing tests (with some minor updates)
Also manually ported over the pre-broken regression test described in
https://github.com/facebook/rocksdb/issues/12870 and ran it (passed).
Performance:
Here we validate that an entire series of recent related PRs are a net improvement in aggregate. "Before" is with these PRs reverted: https://github.com/facebook/rocksdb/issues/12872#12911https://github.com/facebook/rocksdb/issues/12874#12867https://github.com/facebook/rocksdb/issues/12903#12904. "After" includes this PR (and all
of those, with base revision 16c21af). Simultaneous test script designed to maximally depend on SST construction efficiency:
```
for PF in 0 1; do for PS in 0 8; do for WK in 0 1; do [ "$PS" == "$WK" ] || (for I in `seq 1 20`; do TEST_TMPDIR=/dev/shm/rocksdb2 ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=30000000 -memtablerep=vector -allow_concurrent_memtable_write=0 -bloom_bits=10 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -partition_index_and_filters=$PF -prefix_size=$PS -whole_key_filtering=$WK 2>&1 | grep micros/op; done) | awk '{ t += $5; c++; print } END { print 1.0 * t / c }'; echo "Was -partition_index_and_filters=$PF -prefix_size=$PS -whole_key_filtering=$WK"; done; done; done) | tee results
```
Showing average ops/sec of "after" vs. "before"
```
-partition_index_and_filters=0 -prefix_size=0 -whole_key_filtering=1
935586 vs. 928176 (+0.79%)
-partition_index_and_filters=0 -prefix_size=8 -whole_key_filtering=0
930171 vs. 926801 (+0.36%)
-partition_index_and_filters=0 -prefix_size=8 -whole_key_filtering=1
910727 vs. 894397 (+1.8%)
-partition_index_and_filters=1 -prefix_size=0 -whole_key_filtering=1
929795 vs. 922007 (+0.84%)
-partition_index_and_filters=1 -prefix_size=8 -whole_key_filtering=0
921924 vs. 917285 (+0.51%)
-partition_index_and_filters=1 -prefix_size=8 -whole_key_filtering=1
903393 vs. 887340 (+1.8%)
```
As one would predict, the most improvement is seen in cases where we have optimized away copying the whole key.
Reviewed By: jowlyzhang
Differential Revision: D61138271
Pulled By: pdillinger
fbshipit-source-id: 427cef0b1465017b45d0a507bfa7720fa20af043
Summary:
`VersionEditHandlerPointInTime` is tracking found files, missing files, intermediate files in order to decide to build a `Version` on negative edge trigger (transition from valid to invalid) without applying the current `VersionEdit`. However, applying `VersionEdit` and check completeness of a `Version` are specialization of `VersionBuilder`. More importantly, when we augment best efforts recovery to recover not just complete point in time Version but also a prefix of seqno for a point in time Version, such checks need to be duplicated in `VersionEditHandlerPointInTime` and `VersionBuilder`.
To avoid this, this refactor move all the file tracking functionality in `VersionEditHandlerPointInTime` into `VersionBuilder`. To continue to let `VersionEditHandlerPIT` do the edge trigger check and build a `Version` before applying the current `VersionEdit`, a suite of APIs to supporting creating a save point and its associated functions are added in `VersionBuilder` to achieve this.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12928
Test Plan: Existing tests
Reviewed By: anand1976
Differential Revision: D61171320
Pulled By: jowlyzhang
fbshipit-source-id: 604f66f8b1e3a3e13da59d8ba357c74e8a366dbc
Summary:
Add a couple of ticker stats for corruption retry count and successful retries. This PR also eliminates an extra read attempt when there's a checksum mismatch in a block read from the prefetch buffer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12923
Test Plan: Update existing tests
Reviewed By: jowlyzhang
Differential Revision: D61024687
Pulled By: anand1976
fbshipit-source-id: 3a08403580ab244000e0d480b7ee0f5a03d76b06
Summary:
**Context/Summary:** .... since it won't work in the PrepareDelete() path
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12932
Test Plan: CI
Reviewed By: cbi42
Differential Revision: D61155155
Pulled By: hx235
fbshipit-source-id: 99b0784f6c903d70c7b3b88b53ae8e2c885de96f
Summary:
this testcase set syncpoint function which reference this test case heap variable "enable_per_key_placement_" and this sync point function will be triggered by another testcase, so asan will report asan heap use after free error
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12908
Reviewed By: hx235
Differential Revision: D60973363
Pulled By: cbi42
fbshipit-source-id: df4f488f51e7741784d5a92fc0a5fc538c5d5b1a
Summary:
**Context:**
https://github.com/facebook/rocksdb/pull/12838 allows a write thread encountered certain injected error to release the lock and sleep before retrying write in order to reduce performance cost. This requires adding checks like [this](https://github.com/facebook/rocksdb/blob/b26b395e0a15255d322be08110db551976188745/db_stress_tool/expected_value.cc#L29-L31) to prevent writing to the same key from another thread.
The added check causes a false-positive failure when delete range + file ingestion + backup is used. Consider the following scenario:
(1) Issue a delete range covering some key that do not exist and a key does exist (named as k1). k1 will have "pending delete" state while the keys that does not exit will have whatever state they already have since we don't delete a key that does not exist already.
(2) After https://github.com/facebook/rocksdb/pull/12838, `PrepareDeleteRange(... &prepared)` will return `prepared = false`. So below logic will be executed and k1's "pending delete" won't get roll-backed nor committed.
```
std::vector<PendingExpectedValue> pending_expected_values =
shared->PrepareDeleteRange(rand_column_family, rand_key,
rand_key + FLAGS_range_deletion_width,
&prepared);
if (!prepared) {
for (PendingExpectedValue& pending_expected_value :
pending_expected_values) {
pending_expected_value.PermitUnclosedPendingState();
}
return s;
}
```
(3) Issue an file ingestion covering k1 and another key k2. Similar to (2), we will have `shared->PreparePut(column_family, key, &prepared)` return `prepared = false` for k1 while k2 will have a "pending put" state. So below logic will be executed and k2's "pending put" state won't get roll-backed nor committed.
```
for (int64_t key = key_base;
s.ok() && key < shared->GetMaxKey() &&
static_cast<int32_t>(keys.size()) < FLAGS_ingest_external_file_width;
++key)
PendingExpectedValue pending_expected_value =
shared->PreparePut(column_family, key, &prepared);
if (!prepared) {
pending_expected_value.PermitUnclosedPendingState();
for (PendingExpectedValue& pev : pending_expected_values) {
pev.PermitUnclosedPendingState();
}
return;
}
}
```
(4) Issue a backup and verify on k2. Below logic decides that k2 should exist in restored DB since it has a pending write state while k2 is never ingested into the original DB as (3) returns early.
```
bool Exists() const { return PendingPut() || !IsDeleted(); }
TestBackupRestore() {
...
Status get_status = restored_db->Get(
read_opts, restored_cf_handles[rand_column_families[i]], key,
&restored_value);
bool exists = thread->shared->Exists(rand_column_families[i], rand_keys[0]);
if (get_status.ok()) {
if (!exists && from_latest && ShouldAcquireMutexOnKey()) {
std::ostringstream oss;
oss << "0x" << key.ToString(true)
<< " exists in restore but not in original db";
s = Status::Corruption(oss.str());
}
} else if (get_status.IsNotFound()) {
if (exists && from_latest && ShouldAcquireMutexOnKey()) {
std::ostringstream oss;
oss << "0x" << key.ToString(true)
<< " exists in original db but not in restore";
s = Status::Corruption(oss.str());
}
}
...
}
```
So we see false-positive corruption like `Failure in a backup/restore operation with: Corruption: 0x000000000000017B0000000000000073787878 exists in original db but not in restore`
A simple fix is to remove `PendingPut()` from `bool Exists() ` since it's called under a lock and should never see a pending write. However, in order for "under a lock and should never see a pending write" to be true, we need to remove the logic of releasing the lock during sleep in the write thread, which expose pending write to other thread that can call Exists() like back up thread.
The downside of holding lock during sleep is blocking other write thread of the same key to proceed cuz they need to wait for the lock. This should happen rarely as the key of a thread is selected randomly in crash test like below.
```
void StressTest::OperateDb(ThreadState* thread) {
for (uint64_t i = 0; i < ops_per_open; i++) {
...
int64_t rand_key = GenerateOneKey(thread, i);
...
}
}
```
**Summary:**
- Removed the "lock release" part and related checks
- Printed recovery time if the write thread waited more than 10 seconds
- Reverted regression in testing coverage when deleting a non-existent key
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12917
Test Plan:
Below command repro-ed frequently before the fix and not after.
```
./db_stress --WAL_size_limit_MB=1 --WAL_ttl_seconds=60 --acquire_snapshot_one_in=0 --adaptive_readahead=0 --adm_policy=1 --advise_random_on_open=1 --allow_concurrent_memtable_write=0 --allow_data_in_errors=True --allow_fallocate=0 --allow_setting_blob_options_dynamically=1 --async_io=0 --auto_readahead_size=1 --avoid_flush_during_recovery=0 --avoid_flush_during_shutdown=0 --avoid_unnecessary_blocking_io=0 --backup_max_size=104857600 --backup_one_in=100000 --batch_protection_bytes_per_key=0 --bgerror_resume_retry_interval=100 --blob_cache_size=8388608 --blob_compaction_readahead_size=1048576 --blob_compression_type=none --blob_file_size=1073741824 --blob_file_starting_level=1 --blob_garbage_collection_age_cutoff=0.0 --blob_garbage_collection_force_threshold=0.75 --block_align=0 --block_protection_bytes_per_key=8 --block_size=16384 --bloom_before_level=2147483647 --bloom_bits=16.216959977115277 --bottommost_compression_type=xpress --bottommost_file_compaction_delay=600 --bytes_per_sync=262144 --cache_index_and_filter_blocks=1 --cache_index_and_filter_blocks_with_high_priority=1 --cache_size=8388608 --cache_type=lru_cache --charge_compression_dictionary_building_buffer=1 --charge_file_metadata=0 --charge_filter_construction=0 --charge_table_reader=1 --check_multiget_consistency=0 --check_multiget_entity_consistency=0 --checkpoint_one_in=1000000 --checksum_type=kXXH3 --clear_column_family_one_in=0 --column_families=1 --compact_files_one_in=1000 --compact_range_one_in=0 --compaction_pri=3 --compaction_readahead_size=0 --compaction_ttl=10 --compress_format_version=2 --compressed_secondary_cache_size=8388608 --compression_checksum=0 --compression_max_dict_buffer_bytes=2097151 --compression_max_dict_bytes=16384 --compression_parallel_threads=1 --compression_type=zlib --compression_use_zstd_dict_trainer=0 --compression_zstd_max_train_bytes=0 --continuous_verification_interval=0 --daily_offpeak_time_utc=04:00-08:00 --data_block_index_type=0 --db=/dev/shm/rocksdb_test/rocksdb_crashtest_whitebox --db_write_buffer_size=0 --default_temperature=kUnknown --default_write_temperature=kWarm --delete_obsolete_files_period_micros=21600000000 --delpercent=0 --delrangepercent=5 --destroy_db_initially=0 --detect_filter_construct_corruption=1 --disable_file_deletions_one_in=10000 --disable_manual_compaction_one_in=1000000 --disable_wal=0 --dump_malloc_stats=0 --enable_blob_files=0 --enable_blob_garbage_collection=1 --enable_checksum_handoff=1 --enable_compaction_filter=1 --enable_custom_split_merge=1 --enable_do_not_compress_roles=0 --enable_index_compression=1 --enable_memtable_insert_with_hint_prefix_extractor=0 --enable_pipelined_write=1 --enable_sst_partitioner_factory=1 --enable_thread_tracking=0 --enable_write_thread_adaptive_yield=0 --error_recovery_with_no_fault_injection=1 --exclude_wal_from_write_fault_injection=1 --expected_values_dir=/dev/shm/rocksdb_test/rocksdb_crashtest_expected --fail_if_options_file_error=0 --fifo_allow_compaction=1 --file_checksum_impl=big --fill_cache=1 --flush_one_in=1000000 --format_version=2 --get_all_column_family_metadata_one_in=10000 --get_current_wal_file_one_in=0 --get_live_files_apis_one_in=1000000 --get_properties_of_all_tables_one_in=100000 --get_property_one_in=100000 --get_sorted_wal_files_one_in=0 --hard_pending_compaction_bytes_limit=2097152 --high_pri_pool_ratio=0.5 --index_block_restart_interval=1 --index_shortening=2 --index_type=0 --ingest_external_file_one_in=1000 --initial_auto_readahead_size=0 --inplace_update_support=0 --iterpercent=0 --key_len_percent_dist=1,30,69 --key_may_exist_one_in=100 --last_level_temperature=kUnknown --level_compaction_dynamic_level_bytes=0 --lock_wal_one_in=10000 --log2_keys_per_lock=10 --log_file_time_to_roll=0 --log_readahead_size=0 --long_running_snapshots=1 --low_pri_pool_ratio=0.5 --lowest_used_cache_tier=1 --manifest_preallocation_size=0 --manual_wal_flush_one_in=0 --mark_for_compaction_one_file_in=10 --max_auto_readahead_size=16384 --max_background_compactions=1 --max_bytes_for_level_base=67108864 --max_key=100000 --max_key_len=3 --max_log_file_size=1048576 --max_manifest_file_size=1073741824 --max_sequential_skip_in_iterations=16 --max_total_wal_size=0 --max_write_batch_group_size_bytes=16 --max_write_buffer_number=10 --max_write_buffer_size_to_maintain=8388608 --memtable_insert_hint_per_batch=1 --memtable_max_range_deletions=1000 --memtable_prefix_bloom_size_ratio=0.001 --memtable_protection_bytes_per_key=4 --memtable_whole_key_filtering=1 --memtablerep=skip_list --metadata_charge_policy=1 --metadata_read_fault_one_in=0 --metadata_write_fault_one_in=0 --min_blob_size=16 --min_write_buffer_number_to_merge=2 --mmap_read=0 --mock_direct_io=False --nooverwritepercent=1 --num_file_reads_for_auto_readahead=0 --open_files=-1 --open_metadata_read_fault_one_in=0 --open_metadata_write_fault_one_in=0 --open_read_fault_one_in=0 --open_write_fault_one_in=0 --ops_per_thread=20000000 --optimize_filters_for_hits=1 --optimize_filters_for_memory=0 --optimize_multiget_for_io=1 --paranoid_file_checks=1 --partition_filters=0 --partition_pinning=1 --pause_background_one_in=10000 --periodic_compaction_seconds=10 --prefix_size=8 --prefixpercent=0 --prepopulate_blob_cache=1 --prepopulate_block_cache=1 --preserve_internal_time_seconds=0 --progress_reports=0 --promote_l0_one_in=0 --read_amp_bytes_per_bit=0 --read_fault_one_in=0 --readahead_size=524288 --readpercent=60 --recycle_log_file_num=1 --reopen=20 --report_bg_io_stats=0 --reset_stats_one_in=1000000 --sample_for_compression=5 --secondary_cache_fault_one_in=0 --secondary_cache_uri= --skip_stats_update_on_db_open=1 --snapshot_hold_ops=100000 --soft_pending_compaction_bytes_limit=68719476736 --sqfc_name=foo --sqfc_version=1 --sst_file_manager_bytes_per_sec=0 --sst_file_manager_bytes_per_truncate=0 --stats_dump_period_sec=10 --stats_history_buffer_size=1048576 --strict_bytes_per_sync=1 --subcompactions=2 --sync=0 --sync_fault_injection=0 --table_cache_numshardbits=0 --target_file_size_base=16777216 --target_file_size_multiplier=1 --test_batches_snapshots=0 --top_level_index_pinning=3 --uncache_aggressiveness=118 --universal_max_read_amp=-1 --unpartitioned_pinning=0 --use_adaptive_mutex=0 --use_adaptive_mutex_lru=1 --use_attribute_group=0 --use_blob_cache=0 --use_delta_encoding=1 --use_direct_io_for_flush_and_compaction=0 --use_direct_reads=0 --use_full_merge_v1=0 --use_get_entity=0 --use_merge=0 --use_multi_cf_iterator=0 --use_multi_get_entity=0 --use_multiget=1 --use_put_entity_one_in=0 --use_shared_block_and_blob_cache=1 --use_sqfc_for_range_queries=1 --use_timed_put_one_in=0 --use_write_buffer_manager=0 --user_timestamp_size=0 --value_size_mult=32 --verification_only=0 --verify_checksum=1 --verify_checksum_one_in=1000000 --verify_compression=0 --verify_db_one_in=10000 --verify_file_checksums_one_in=1000000 --verify_iterator_with_expected_state_one_in=5 --verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=0 --wal_compression=none --write_buffer_size=33554432 --write_dbid_to_manifest=0 --write_fault_one_in=0 --writepercent=35
```
Reviewed By: cbi42
Differential Revision: D60890580
Pulled By: hx235
fbshipit-source-id: 401f90d6d351c7ee11088cad06fb00e54062d416
Summary:
I was investigating a crash test failure with "Corruption: SST file is ahead of WALs" which I haven't reproduced, but I did reproduce a data loss issue on recovery which I suspect could be the same root problem. The problem is already somewhat known (see https://github.com/facebook/rocksdb/issues/12403 and https://github.com/facebook/rocksdb/issues/12639) where it's only safe to recovery multiple recycled WAL files with trailing old data if the sequence numbers between them are adjacent (to ensure we didn't lose anything in the corrupt/obsolete WAL tail).
However, aside from disableWAL=true, there are features like external file ingestion that can increment the sequence numbers without writing to the WAL. It is simply unsustainable to worry about this kind of feature interaction limiting where we can consume sequence numbers. It is very hard to test and audit as well. For reliable crash recovery of recycled WALs, we need a better way of detecting that we didn't drop data from one WAL to the next.
Until then, let's disable WAL recycling in the crash test, to help stabilize it.
Ideas for follow-up to fix the underlying problem:
(a) With recycling, we could always sync the WAL before opening the next one. HOWEVER, this potentially very large sync could cause a big hiccup in writes (vs. O(1) sized manifest sync).
(a1) The WAL sync could ensure it is truncated to size, or
(a2) By requiring track_and_verify_wals_in_manifest, we could assume that the last synced size in the manifest is the final usable size of the WAL. (It might also be worth avoiding truncating recycled WALs.)
(b) Add a new mechanism to record and verify the final size of a WAL without requiring a sync.
(b1) By requiring track_and_verify_wals_in_manifest, this could be new WAL metadata recorded in the manifest (at the time of switching WALs). Note that new fields of WalMetadata are not forward-compatible, but a new kind of manifest record (next to WalAddition, WalDeletion; e.g. WalCompletion) is IIRC forward-compatible.
(b2) A new kind of WAL header entry (not forward compatible, unfortunately) could record the final size of the previous WAL.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12918
Test Plan: Added disabled reproducer for non-linear data loss on recovery
Reviewed By: hx235
Differential Revision: D60917527
Pulled By: pdillinger
fbshipit-source-id: 3663d79aec81851f5cf41669f84a712bb4563fd7
Summary:
Ahead of a "decoupled" variant of partitioned filters, refactoring this unit test file to make it easier to incorporate that new variant.
* bool test param to new enum class FilterPartitioning
* Some cases of iterating over that bool to new parameterized test
* Combine some common functionality for configuring parameterized options
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12911
Test Plan: no production changes, and no intentional changes to scope or conditions of tests
Differential Revision: D60701287
fbshipit-source-id: 3497e3230e29a4f62c934bcb75693965a2df41d8
Summary:
**Context/Summary:**
`CleanStagingDirectory()` is called when the temporary .tmp folder we use to create checkpoint is not empty to begin with.
Expanded fault injection can make this call fail e.g, `Delete file /dev/shm/rocksdb_test/rocksdb_crashtest_blackbox/.checkpoint17.tmp/012393.sst -- IO error: injected metadata write error`.
But The result of `CleanStagingDirectory()` is ignored in `CreateCheckpoint()`. So the injected IO error can't be propagated to db stress test and handled correctly. Hence we see `While mkdir: /dev/shm/rocksdb_test/rocksdb_crashtest_blackbox/.checkpoint17.tmp: File exists` when we try to re-use a non-empty .tmp folder for new snapshots.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12894
Test Plan: Monitor CI
Reviewed By: ltamasi
Differential Revision: D60422849
Pulled By: hx235
fbshipit-source-id: 6f735c98eaa05d2b97ba4f781e0928357a50377a
Summary:
In normal use cases, meta info like column family's timestamp size is tracked at the transaction layer, so it's not necessary and even detrimental to track such info inside the internal WriteBatch because it may let anti-patterns like bypassing Transaction write APIs and directly write to its internal WriteBatch like this:
https://github.com/facebook/mysql-5.6/blob/9d0a754dc9973af0508b3ba260fc337190a3218f/storage/rocksdb/ha_rocksdb.cc#L4949-L4950
Setting this option to true will keep aforementioned use case continue to work before it's refactored out. This option is only for this purpose and it will be gradually deprecated after aforementioned MyRocks use case are refactored.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12864
Test Plan: Added unit tests
Reviewed By: cbi42
Differential Revision: D60194094
Pulled By: jowlyzhang
fbshipit-source-id: 64a98822167e99aa7e4fa2a60085d44a5deaa45c
Summary:
https://github.com/facebook/rocksdb/issues/12891 updated this deletion rate in the test to be much higher, which makes the test flaky. The rate is being intentionally set to very low to maximize the retention of a ".log.trash" file after DB closes. This PR just change it back.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12915
Reviewed By: ltamasi
Differential Revision: D60776312
Pulled By: jowlyzhang
fbshipit-source-id: d193557a042c65816fcc337cceb09905e042e9f6
Summary:
Make `DestroyDB` slowly delete files if it's configured and enabled via `SstFileManager`.
It's currently not available mainly because of DeleteScheduler's logic related to tracked total_size_ and total_trash_size_. These accounting and logic should not be applied to `DestroyDB`. This PR adds a `DeleteUnaccountedDBFile` util for this purpose which deletes files without accounting it. This util also supports assigning a file to a specified trash bucket so that user can later wait for a specific trash bucket to be empty. For `DestroyDB`, files with more than 1 hard links will be deleted immediately.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12891
Test Plan: Added unit tests, existing tests.
Reviewed By: anand1976
Differential Revision: D60300220
Pulled By: jowlyzhang
fbshipit-source-id: 8b18109a177a3a9532f6dc2e40e08310c08ca3c7
Summary:
Was checking == a desired number of entries added to a filter, when the combination of whole key and prefix filtering could add more than one entry per table internal key. This could lead to unnecessarily large filter partitions, which could affect performance and block cache fairness.
Also (only somewhat related because of other work in progress):
* Some variable renaming and a new assertion in BlockBasedTableBuilder, to add some clarity.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12904
Test Plan:
If you add assertion logic to the base revision checking that the partition cut is requested whenever `keys_added_to_partition_ >= keys_per_partition_`, it fails on a number of db_bloom_filter_test tests. However, such an assertion in the revised code would be essentially redundant with the new logic.
If I added a regression test for this, it would be tricky and fragile, so I don't think it's important enough to chase and maintain. (Open to suggestions / input.)
Reviewed By: jowlyzhang
Differential Revision: D60557827
Pulled By: pdillinger
fbshipit-source-id: 77a56097d540da6e7851941a26d26ced2d944373
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12910
There is currently a call to `GetBGError()` in `DBImpl::WriteImplWALOnly()` where the DB mutex is (incorrectly) not held, leading to a data race. Technically, we could acquire the mutex here but instead, the patch removes the affected check altogether, since the same check is already performed (in a thread-safe manner) in the subsequent call to `PreprocessWrite()`.
Reviewed By: cbi42
Differential Revision: D60682008
fbshipit-source-id: 54b67975dcf57d67c068cac71e8ada09a1793ec5
Summary:
This is ahead of some related changes/enhancements. Refactorings here:
* Restructure some state of PartitionedFilterBlockBuilder to reduce redundancy in state tracking, improve clarity.
* Changed some function signatures to better match standard practice (return Status)
* Improve comments, arrange related fields
* Discourage/prevent production use of Finish without status (now TEST_Finish)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12903
Test Plan: existing tests
Reviewed By: jowlyzhang
Differential Revision: D60548613
Pulled By: pdillinger
fbshipit-source-id: d7dbc79951fcc3b837877227d58f713698ad2596
Summary:
In leveled compaction, we pick intra-L0 compaction instead of L0->Lbase whenever L0 size is small. When L0 files contain many deletions, it makes more sense to compact then down instead of accumulating tombstones in L0. This PR uses compensated_file_size when computing L0 size for determining intra-L0 compaction. Also scale down the limit on total L0 size further to be more cautious about accumulating data in L0.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12878
Test Plan: updated unit test.
Reviewed By: hx235
Differential Revision: D59932421
Pulled By: cbi42
fbshipit-source-id: 9de973ac51eb7df81b38b8c68110072b1aa06321
Summary:
As titled. The `emplace_back` below will add the same collector factory again during Reopen.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12907
Reviewed By: pdillinger
Differential Revision: D60614170
Pulled By: jowlyzhang
fbshipit-source-id: a79498d209e4910a5e94a5cb742935015277918c
Summary: https://github.com/facebook/rocksdb/pull/12801 updated the version of `folly` used in RocksDB builds to a revision that requires `g++` version 10 when built with a GNU toolchain. This shouldn't really matter for this nightly GitHub Actions job, since we're supposed to be building with `clang++-13`; however, due to the way the compilers had been set, seems like we were historically only building RocksDB with `clang` (and `folly` with `gcc-9`, which led to a broken build after the update). Attempt to fix this by setting `CC` / `CXX` to `clang` / `clang++` in the job's environment.
Reviewed By: pdillinger
Differential Revision: D60534452
fbshipit-source-id: c7b5a02409fb1ea50e4524731237f7bc8d3f7ca6
Summary:
Crash test encountered this failure:
```file ingestion error: Corruption: properties unsorted under specified IngestExternalFileOptions: move_files: 0, verify_checksums_before_ingest: 1, verify_checksums_readahead_size: 1048576 (Empty string or missing field indicates default option or value is used```
Further inspection showed out of order table properties in an external file created by `SstFileWriter` for ingestion, and the file is likely created like this because it passed the initial checksum check. This change added some assertions to check invariant at the properties creation and collecting side.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12898
Test Plan: Existing tests
Reviewed By: hx235
Differential Revision: D60459817
Pulled By: jowlyzhang
fbshipit-source-id: 91474943d2f9d7795f00b6031c08a13ab91e2470
Summary:
A crash test failure in log sync in DBImpl::WriteToWAL is due to a missed case in https://github.com/facebook/rocksdb/issues/12734. Just need to apply similar logic from DBImpl::SyncWalImpl to check for an already closed WAL (nullptr writer). This is extremely rare because it only comes from failed Sync on a closed WAL.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12899
Test Plan: watch crash test
Reviewed By: cbi42
Differential Revision: D60481652
Pulled By: pdillinger
fbshipit-source-id: 4a176bb6a53dcf077f88344710a110c2f946c386
Summary:
The `PessimisticTransaction::SetName()` code checks for an existing txn of the given name before registering the new txn. However, this is not atomic, which could result in a race condition if two txns try to register with the same name. Both might succeed and lead to unpredictable behavior. This PR makes the test and set atomic.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12895
Reviewed By: pdillinger
Differential Revision: D60460482
Pulled By: anand1976
fbshipit-source-id: e8afeb2356e1b8f4e8df785cb73532739f82579d
Summary:
By reusing an object that owns a vector. The vector allocation/sizing was substantial in a CPU profile.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12893
Test Plan: existing tests
Reviewed By: jowlyzhang
Differential Revision: D60405139
Pulled By: pdillinger
fbshipit-source-id: 8bfbc07cd9b4829f2ac9015e90f2b4eba61fd984
Summary:
As titled. This PR adds a `TransactionDBOptions` field `enable_udt_validation` to allow user to toggle the timestamp based validation behavior across the whole DB. When it is true, which is the default value and the existing behavior. A recap of what this behavior is: `GetForUpdate` does timestamp based conflict checking to make sure no other transaction has committed a version of the key tagged with a timestamp equal to or newer than the calling transaction's `read_timestamp_` the user set via `SetReadTimestampForValidation`. When this field is set to false, we disable timestamp based validation for the whole DB. MyRocks find it hard to find a read timestamp for this validation API, so we added this flexibility.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12857
Test Plan: Added unit test
Reviewed By: ltamasi
Differential Revision: D60194134
Pulled By: jowlyzhang
fbshipit-source-id: b8507f8ddc37fc7a2948cf492ce5c599ae646fef
Summary:
**Context/Summary:**
We discovered the following false positive in our crash test lately:
(1) PUT() writes k/v to WAL but fails in `ApplyWALToManifest()`. The k/v is in the WAL
(2) Current stress test logic will rollback the expected state of such k/v since PUT() fails
(3) If the DB crashes before recovery finishes and reopens, the WAL will be replayed and the k/v is in the DB while the expected state have been roll-backed.
We decided to leave those expected state to be pending until the loop-write of the same key succeeds.
Bonus: Now that I realized write to manifest can also fail the write which faces the similar problem as https://github.com/facebook/rocksdb/pull/12797, I decided to disable fault injection on user write per thread (instead of globally) when tracing is needed for prefix recovery; some refactory
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12838
Test Plan:
Rehearsal CI
Run below command (varies on sync_fault_injection=1,0 to verify ExpectedState behavior) for a while to ensure crash recovery validation works fine
```
python3 tools/db_crashtest.py --simple blackbox --interval=30 --WAL_size_limit_MB=0 --WAL_ttl_seconds=0 --acquire_snapshot_one_in=10000 --adaptive_readahead=1 --adm_policy=1 --advise_random_on_open=0 --allow_concurrent_memtable_write=0 --allow_data_in_errors=True --allow_fallocate=0 --async_io=0 --auto_readahead_size=0 --avoid_flush_during_recovery=0 --avoid_flush_during_shutdown=0 --avoid_unnecessary_blocking_io=0 --backup_max_size=104857600 --backup_one_in=0 --batch_protection_bytes_per_key=0 --bgerror_resume_retry_interval=1000000 --block_align=1 --block_protection_bytes_per_key=4 --block_size=16384 --bloom_before_level=4 --bloom_bits=56.810257702625165 --bottommost_compression_type=none --bottommost_file_compaction_delay=0 --bytes_per_sync=262144 --cache_index_and_filter_blocks=1 --cache_index_and_filter_blocks_with_high_priority=1 --cache_size=8388608 --cache_type=auto_hyper_clock_cache --charge_compression_dictionary_building_buffer=1 --charge_file_metadata=1 --charge_filter_construction=1 --charge_table_reader=0 --check_multiget_consistency=0 --check_multiget_entity_consistency=1 --checkpoint_one_in=10000 --checksum_type=kxxHash --clear_column_family_one_in=0 --column_families=1 --compact_files_one_in=1000 --compact_range_one_in=1000 --compaction_pri=4 --compaction_readahead_size=1048576 --compaction_ttl=10 --compress_format_version=1 --compressed_secondary_cache_ratio=0.0 --compressed_secondary_cache_size=0 --compression_checksum=0 --compression_max_dict_buffer_bytes=0 --compression_max_dict_bytes=0 --compression_parallel_threads=1 --compression_type=none --compression_use_zstd_dict_trainer=0 --compression_zstd_max_train_bytes=0 --continuous_verification_interval=0 --daily_offpeak_time_utc=04:00-08:00 --data_block_index_type=1 --db_write_buffer_size=0 --default_temperature=kWarm --default_write_temperature=kCold --delete_obsolete_files_period_micros=30000000 --delpercent=20 --delrangepercent=20 --destroy_db_initially=0 --detect_filter_construct_corruption=0 --disable_file_deletions_one_in=10000 --disable_manual_compaction_one_in=1000000 --disable_wal=0 --dump_malloc_stats=0 --enable_checksum_handoff=1 --enable_compaction_filter=0 --enable_custom_split_merge=0 --enable_do_not_compress_roles=0 --enable_index_compression=1 --enable_memtable_insert_with_hint_prefix_extractor=0 --enable_pipelined_write=0 --enable_sst_partitioner_factory=0 --enable_thread_tracking=0 --enable_write_thread_adaptive_yield=0 --error_recovery_with_no_fault_injection=1 --exclude_wal_from_write_fault_injection=0 --fail_if_options_file_error=1 --fifo_allow_compaction=0 --file_checksum_impl=crc32c --fill_cache=1 --flush_one_in=1000000 --format_version=3 --get_all_column_family_metadata_one_in=1000000 --get_current_wal_file_one_in=0 --get_live_files_apis_one_in=1000000 --get_properties_of_all_tables_one_in=1000000 --get_property_one_in=100000 --get_sorted_wal_files_one_in=0 --hard_pending_compaction_bytes_limit=274877906944 --high_pri_pool_ratio=0.5 --index_block_restart_interval=4 --index_shortening=2 --index_type=0 --ingest_external_file_one_in=0 --initial_auto_readahead_size=16384 --inplace_update_support=0 --iterpercent=10 --key_len_percent_dist=1,30,69 --key_may_exist_one_in=100 --last_level_temperature=kWarm --level_compaction_dynamic_level_bytes=1 --lock_wal_one_in=10000 --log_file_time_to_roll=60 --log_readahead_size=16777216 --long_running_snapshots=1 --low_pri_pool_ratio=0 --lowest_used_cache_tier=0 --manifest_preallocation_size=0 --manual_wal_flush_one_in=0 --mark_for_compaction_one_file_in=10 --max_auto_readahead_size=16384 --max_background_compactions=1 --max_bytes_for_level_base=67108864 --max_key=100000 --max_key_len=3 --max_log_file_size=1048576 --max_manifest_file_size=32768 --max_sequential_skip_in_iterations=1 --max_total_wal_size=0 --max_write_batch_group_size_bytes=16 --max_write_buffer_number=10 --max_write_buffer_size_to_maintain=8388608 --memtable_insert_hint_per_batch=1 --memtable_max_range_deletions=0 --memtable_prefix_bloom_size_ratio=0.01 --memtable_protection_bytes_per_key=1 --memtable_whole_key_filtering=1 --memtablerep=skip_list --metadata_charge_policy=1 --metadata_read_fault_one_in=0 --metadata_write_fault_one_in=8 --min_write_buffer_number_to_merge=1 --mmap_read=1 --mock_direct_io=False --nooverwritepercent=1 --num_file_reads_for_auto_readahead=1 --open_files=-1 --open_metadata_read_fault_one_in=0 --open_metadata_write_fault_one_in=8 --open_read_fault_one_in=0 --open_write_fault_one_in=8 --ops_per_thread=100000000 --optimize_filters_for_hits=1 --optimize_filters_for_memory=1 --optimize_multiget_for_io=1 --paranoid_file_checks=0 --partition_filters=0 --partition_pinning=3 --pause_background_one_in=1000000 --periodic_compaction_seconds=2 --prefix_size=7 --prefixpercent=0 --prepopulate_block_cache=0 --preserve_internal_time_seconds=0 --progress_reports=0 --promote_l0_one_in=0 --read_amp_bytes_per_bit=0 --read_fault_one_in=1000 --readahead_size=524288 --readpercent=10 --recycle_log_file_num=1 --reopen=0 --report_bg_io_stats=0 --reset_stats_one_in=1000000 --sample_for_compression=0 --secondary_cache_fault_one_in=0 --set_options_one_in=0 --skip_stats_update_on_db_open=1 --snapshot_hold_ops=100000 --soft_pending_compaction_bytes_limit=68719476736 --sqfc_name=foo --sqfc_version=0 --sst_file_manager_bytes_per_sec=104857600 --sst_file_manager_bytes_per_truncate=0 --stats_dump_period_sec=10 --stats_history_buffer_size=0 --strict_bytes_per_sync=1 --subcompactions=4 --sync=1 --sync_fault_injection=0 --table_cache_numshardbits=6 --target_file_size_base=16777216 --target_file_size_multiplier=1 --test_batches_snapshots=0 --top_level_index_pinning=2 --uncache_aggressiveness=239 --universal_max_read_amp=-1 --unpartitioned_pinning=1 --use_adaptive_mutex=1 --use_adaptive_mutex_lru=1 --use_attribute_group=0 --use_delta_encoding=0 --use_direct_io_for_flush_and_compaction=0 --use_direct_reads=0 --use_full_merge_v1=0 --use_get_entity=0 --use_merge=0 --use_multi_cf_iterator=0 --use_multi_get_entity=0 --use_multiget=0 --use_put_entity_one_in=0 --use_sqfc_for_range_queries=1 --use_timed_put_one_in=0 --use_write_buffer_manager=0 --user_timestamp_size=0 --value_size_mult=32 --verification_only=0 --verify_checksum=1 --verify_checksum_one_in=1000000 --verify_compression=0 --verify_db_one_in=100000 --verify_file_checksums_one_in=1000000 --verify_iterator_with_expected_state_one_in=5 --verify_sst_unique_id_in_manifest=1 --wal_bytes_per_sync=0 --wal_compression=none --write_buffer_size=33554432 --write_dbid_to_manifest=0 --write_fault_one_in=8 --writepercent=40
```
Reviewed By: cbi42
Differential Revision: D59377075
Pulled By: hx235
fbshipit-source-id: 91f602fd67e2d339d378cd28b982095fd073dcb6
Summary:
This PR fix `VersionSet`'s `manifest_number_` could be pointing to an invalid number intermediately. This happens when a new manifest roll is attempted but fast failed after loading table handlers and before the new manifest file creation/writing is actually attempted.
In theory, a later manifest roll effort will overthrow this intermediate invalid in memory state. There is on harm when the DB crashes in this invalid state either. But efforts that takes a file snapshot of the DB like backup will incorrectly try to copy a non existing manifest file.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12882
Reviewed By: cbi42
Differential Revision: D60204956
Pulled By: jowlyzhang
fbshipit-source-id: effbdb124b582f879d114988af06ac63867fc549
Summary:
MultiCfIterators (`CoalescingIterator` and `AttributeGroupIterator`) are not yet compatible with write-prepared/write-unprepared transactions, yet (write-committed is fine). This fix includes the following.
- Properly return `ErrorIterator` if the user attempts to use the `CoalescingIterator` or `AttributeGroupIterator` in WritePreparedTxnDB (and WriteUnpreparedTxnDB)
- Set `use_multi_cf_iterator = 0` if `use_txn=1` and `txn_write_policy != 0 (WRITE_COMMITTED)` in stress test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12883
Test Plan:
Works
```
./db_stress ... --use_txn=1 --txn_write_policy=0 --use_multi_cf_iterator=1
```
Fails
```
./db_stress ... --use_txn=1 --txn_write_policy=1 --use_multi_cf_iterator=1
```
Reviewed By: cbi42
Differential Revision: D60190784
Pulled By: jaykorean
fbshipit-source-id: 3bc1093e81a4ef5753ba9b32c5aea997c21bfd33
Summary:
Something I am working on is going to expand usage of `BlockBasedTableBuilder::Rep::last_key`, but the existing code contract for `IndexBuilder::AddIndexEntry` makes that difficult because it modifies its `last_key` parameter to be the separator value recorded in the index, often something between the two boundary keys.
This change primarily changes the contract of that function and related functions to separate function inputs and outputs, without sacrificing efficiency. For efficiency, a reusable scratch string buffer is provided by the caller, which the callee can use (or not) in returning a result Slice. That should yield a performance improvement as we are reusing a buffer for keys rather than copying into a new one each time in the FindShort* functions, without any additional string copies or conditional branches.
Additional improvements in PartitionedIndexBuilder specifically:
* Reduce string copies by eliminating `sub_index_last_key_` and instead tracking the key for the next partition in a placeholder Entry.
* Simplify code and improve code quality by changing `sub_index_builder_` to unique_ptr.
* Eliminate unnecessary NewFlushBlockPolicy call/object.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12867
Test Plan: existing tests, crash test. Will validate performance along with the change this is setting up.
Reviewed By: anand1976
Differential Revision: D59793119
Pulled By: pdillinger
fbshipit-source-id: 556da75cf13b967511f84702b2713d152f536a07
Summary:
**Context/Summary:**
We recently discovered a case where write of the same key right after error recovery of a previous failed write of the same key finishes causes two same WAL entries, violating our assertion. This is because we don't advance seqno on failed write and reuse the same WAL containing the failed write for the new write if the memtable at the time is empty.
This PR reuses the flush path for an empty memtable to switch WAL and update min WAL to keep in error recovery flush
as well as updates the INFO log message for clarity.
```
2024/07/17-15:01:32.271789 327757 (Original Log Time 2024/07/17-15:01:25.942234) [/flush_job.cc:1017] [default] [JOB 2] Level-0 flush table https://github.com/facebook/rocksdb/issues/9: 0 bytes OK It's an empty SST file from a successful flush so won't be kept in the DB
2024/07/17-15:01:32.271798 327757 (Original Log Time 2024/07/17-15:01:32.269954) [/memtable_list.cc:560] [default] Level-0 commit flush result of table https://github.com/facebook/rocksdb/issues/9 started
2024/07/17-15:01:32.271802 327757 (Original Log Time 2024/07/17-15:01:32.271217) [/memtable_list.cc:760] [default] Level-0 commit flush result of table https://github.com/facebook/rocksdb/issues/9: memtable https://github.com/facebook/rocksdb/issues/1 done
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/12873
Test Plan:
New UT that failed before this PR with following assertion failure (i.e, duplicate WAL entries) and passes after
```
db_wal_test: db/write_batch.cc:2254: rocksdb::Status rocksdb::{anonymous}::MemTableInserter::PutCFImpl(uint32_t, const rocksdb::Slice&, const rocksdb::Slice&, rocksdb::ValueType, RebuildTxnOp, const ProtectionInfoKVOS64*) [with RebuildTxnOp = rocksdb::{anonymous}::MemTableInserter::PutCF(uint32_t, const rocksdb::Slice&, const rocksdb::Slice&)::<lambda(rocksdb::WriteBatch*, uint32_t, const rocksdb::Slice&, const rocksdb::Slice&)>; uint32_t = unsigned int; rocksdb::ProtectionInfoKVOS64 = rocksdb::ProtectionInfoKVOS<long unsigned int>]: Assertion `seq_per_batch_' failed.
```
Reviewed By: anand1976
Differential Revision: D59884468
Pulled By: hx235
fbshipit-source-id: 5d854b719092552c69727a979f269fb7f6c39756
2024-07-22 12:40:25 -07:00
1139 changed files with 236481 additions and 35918 deletions
# The image configuration is build_tools/ubuntu20_image/Dockerfile
# To update and build the image:
# $ cd build_tools/ubuntu20_image
# $ docker build -t zjay437/rocksdb:0.5 .
# $ docker push zjay437/rocksdb:0.5
# `zjay437` is the account name for zjay@meta.com which readwrite token is shared internally. To login:
# $ docker login --username zjay437
# Or please feel free to change it to your docker hub account for hosting the image, meta employee should already have the account and able to login with SSO.
# To avoid impacting the existing CI runs, please bump the version every time creating a new image
# to run the CI image environment locally:
# $ docker run --cap-add=SYS_PTRACE --security-opt seccomp=unconfined -it zjay437/rocksdb:0.5 bash
# option `--cap-add=SYS_PTRACE --security-opt seccomp=unconfined` is used to enable gdb to attach an existing process
- image:zjay437/rocksdb:0.6
linux-java-docker:
docker:
# This is the Docker Image used for building RocksJava releases, see: https://github.com/evolvedbinary/docker-rocksjava
- image:evolvedbinary/rocksjava:centos6_x64-be
jobs:
build-macos:
macos:
xcode:14.3.1
resource_class:macos.m1.medium.gen1
environment:
ROCKSDB_DISABLE_JEMALLOC:1# jemalloc cause env_test hang, disable it for now
steps:
- increase-max-open-files-on-macos
- install-gflags-on-macos
- pre-steps-macos
- run:ulimit -S -n `ulimit -H -n` && make V=1 J=16 -j16 all
- post-steps
build-macos-cmake:
macos:
xcode:14.3.1
resource_class:macos.m1.medium.gen1
parameters:
run_even_tests:
description:run even or odd tests, used to split tests to 2 groups
- run:ASSERT_STATUS_CHECKED=1 TEST_UINT128_COMPAT=1 ROCKSDB_MODIFY_NPHASH=1 LIB_MODE=static OPT="-DROCKSDB_NAMESPACE=alternative_rocksdb_ns" make V=1 -j24 check
- post-steps
build-linux-release:
executor:linux-docker
resource_class:2xlarge
steps:
- checkout# check out the code in the project directory
- run:make V=1 -j32 LIB_MODE=shared release
- run:ls librocksdb.so# ensure shared lib built
- run:./db_stress --version# ensure with gflags
- run:make clean
- run:make V=1 -j32 release
- run:ls librocksdb.a# ensure static lib built
- run:./db_stress --version# ensure with gflags
- run:make clean
- run:apt-get remove -y libgflags-dev
- run:make V=1 -j32 LIB_MODE=shared release
- run:ls librocksdb.so# ensure shared lib built
- run:if ./db_stress --version; then false; else true; fi# ensure without gflags
- run:make clean
- run:make V=1 -j32 release
- run:ls librocksdb.a# ensure static lib built
- run:if ./db_stress --version; then false; else true; fi# ensure without gflags
- post-steps
build-linux-release-rtti:
executor:linux-docker
resource_class:xlarge
steps:
- checkout# check out the code in the project directory
- run:USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench
- run:./db_stress --version# ensure with gflags
- run:make clean
- run:apt-get remove -y libgflags-dev
- run:USE_RTTI=1 DEBUG_LEVEL=0 make V=1 -j16 static_lib tools db_bench
- run:if ./db_stress --version; then false; else true; fi# ensure without gflags
build-linux-clang-no_test_run:
executor:linux-docker
resource_class:xlarge
steps:
- checkout# check out the code in the project directory
- run:CC=clang CXX=clang++ USE_CLANG=1 PORTABLE=1 make V=1 -j16 all
- post-steps
build-linux-clang10-asan:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:COMPILE_WITH_ASAN=1 CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check# aligned new doesn't work for reason we haven't figured out
- post-steps
build-linux-clang10-mini-tsan:
executor:linux-docker
resource_class:2xlarge+
steps:
- pre-steps
- run:COMPILE_WITH_TSAN=1 CC=clang-13 CXX=clang++-13 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 check
- post-steps
build-linux-clang10-ubsan:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:COMPILE_WITH_UBSAN=1 OPT="-fsanitize-blacklist=.circleci/ubsan_suppression_list.txt" CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 USE_CLANG=1 make V=1 -j32 ubsan_check# aligned new doesn't work for reason we haven't figured out
- post-steps
build-linux-valgrind:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:PORTABLE=1 make V=1 -j32 valgrind_test
- post-steps
build-linux-clang10-clang-analyze:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:CC=clang-10 CXX=clang++-10 ROCKSDB_DISABLE_ALIGNED_NEW=1 CLANG_ANALYZER="/usr/bin/clang++-10" CLANG_SCAN_BUILD=scan-build-10 USE_CLANG=1 make V=1 -j32 analyze# aligned new doesn't work for reason we haven't figured out. For unknown, reason passing "clang++-10" as CLANG_ANALYZER doesn't work, and we need a full path.
- run:make V=1 -j8 -k check-headers# could be moved to a different build
- post-steps
build-linux-gcc-7-with-folly:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- setup-folly
- build-folly
- run: USE_FOLLY=1 LIB_MODE=static CC=gcc-7 CXX=g++-7 V=1 make -j32 check # TODO:LIB_MODE only to work around unresolved linker failures
- post-steps
build-linux-gcc-7-with-folly-lite-no-test:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- setup-folly
- run:USE_FOLLY_LITE=1 CC=gcc-7 CXX=g++-7 V=1 make -j32 all
- post-steps
build-linux-gcc-8-no_test_run:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:CC=gcc-8 CXX=g++-8 V=1 make -j32 all
- post-steps
build-linux-cmake-with-folly-coroutines:
executor:linux-docker
resource_class:2xlarge
environment:
CC:gcc-10
CXX:g++-10
steps:
- pre-steps
- setup-folly
- build-folly
- run:(mkdir build && cd build && cmake -DUSE_COROUTINES=1 -DWITH_GFLAGS=1 -DROCKSDB_BUILD_SHARED=0 .. && make V=1 -j20 && ctest -j20)
- post-steps
build-linux-gcc-10-cxx20-no_test_run:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:CC=gcc-10 CXX=g++-10 V=1 ROCKSDB_CXX_STANDARD=c++20 make -j32 all
- post-steps
build-linux-gcc-11-no_test_run:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run: LIB_MODE=static CC=gcc-11 CXX=g++-11 V=1 make -j32 all microbench # TODO:LIB_MODE only to work around unresolved linker failures
- post-steps
build-linux-clang-13-no_test_run:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- run:CC=clang-13 CXX=clang++-13 USE_CLANG=1 make -j32 all microbench
- post-steps
# Ensure ASAN+UBSAN with folly, and full testsuite with clang 13
build-linux-clang-13-asan-ubsan-with-folly:
executor:linux-docker
resource_class:2xlarge
steps:
- pre-steps
- setup-folly
- build-folly
- run: CC=clang-13 CXX=clang++-13 LIB_MODE=static USE_CLANG=1 USE_FOLLY=1 COMPILE_WITH_UBSAN=1 COMPILE_WITH_ASAN=1 make -j32 check # TODO:LIB_MODE only to work around unresolved linker failures
- post-steps
# This job is only to make sure the microbench tests are able to run, the benchmark result is not meaningful as the CI host is changing.
// The workflow runs a recovery session when this happens, so this
// branch is typically only hit if recovery wasn't attempted (e.g.,
// no findings file was written). Extract what we can.
constpartial=getLastAssistantText(executionLog);
if(partial){
responseBody=
`⚠️ **Review incomplete — Claude hit the turn limit.**\n\nBelow is the last partial output. You can request a fresh review with \`/claude-review\`.\n\n---\n\n${
partial}`;
}else{
responseBody=
'⚠️ **Review incomplete — Claude hit the turn limit before producing output.** You can request a fresh review with `/claude-review`.';
This document provides guidance for generating and reviewing code in the RocksDB project, derived from analysis of code review feedback across hundreds of complex merged Pull Requests. Use this as a reference when writing code with AI assistants or conducting code reviews.
---
## General Best Practices
### Code Quality and Maintainability
**Clarity and Readability:** Write clear, self-documenting code. Use meaningful variable names, add comments for complex logic, and structure code to minimize cognitive load. Avoid clever tricks that sacrifice readability for marginal performance gains unless absolutely necessary.
**Consistent Style:** Follow existing code style conventions. RocksDB uses `.clang-format` for formatting, specific naming conventions, and structural patterns. Deviations from these patterns are frequently flagged in reviews.
**Error Handling:** Ensure robust error handling throughout the codebase. Use RocksDB's `Status` type consistently, propagate errors appropriately, and avoid silently ignoring failures. Reviewers pay close attention to edge cases and failure modes.
### Testing Philosophy
**Comprehensive Coverage:** Every change should include appropriate test coverage. This includes unit tests for isolated functionality, integration tests for component interactions, and stress tests for concurrency and performance validation. Reviewers will ask for additional tests if coverage is insufficient.
**Edge Cases and Failure Modes:** Tests should explicitly cover edge cases, boundary conditions, and potential failure scenarios. This is especially important for changes affecting core database operations, compaction, or recovery logic.
**Platform-Specific Testing:** RocksDB supports multiple platforms (Linux, Windows, macOS) and compilers (GCC, Clang, MSVC). Changes should be tested across relevant platforms, particularly when touching platform-specific code or using compiler-specific features.
### Performance Considerations
**⚠️ PERFORMANCE IS CRITICAL:** RocksDB is a high-performance storage engine where every CPU cycle and memory access matters. When writing code, always evaluate from a performance perspective. This is not optional—performance-aware coding is a fundamental requirement for all contributions.
**Benchmarking and Profiling:** Performance claims should be backed by empirical evidence. Use RocksDB's benchmarking tools (e.g., `db_bench`) to validate improvements. Reviewers will request benchmark results for changes that could impact performance.
**Memory Allocation:** Minimize dynamic memory allocations, especially in hot paths. Prefer stack allocation over heap allocation. Reuse buffers when possible. Consider using arena allocators or memory pools for frequent small allocations. Every `new`, `malloc`, or container resize has a cost.
**Memory Copy:** Avoid unnecessary memory copies. Use move semantics, `std::string_view`, `Slice`, and pass-by-reference where appropriate. Be aware of implicit copies in STL containers and function returns. Prefer in-place operations over copy-and-modify patterns.
**CPU Cache Efficiency:** Design data structures and access patterns to be cache-friendly. Keep frequently accessed data together (data locality). Prefer sequential memory access over random access. Be mindful of cache line sizes (typically 64 bytes) and avoid false sharing in concurrent code. Consider struct packing and field ordering to improve cache utilization.
**Loop Optimization:** Look for opportunities to collapse nested loops, reduce loop overhead, and minimize branch mispredictions. Hoist invariant computations out of loops. Consider loop unrolling for tight inner loops. Batch operations when possible to amortize per-operation overhead.
**SIMD and Vectorization:** Leverage SIMD instructions (SSE, AVX) for data-parallel operations when appropriate. Structure data to enable auto-vectorization by the compiler. Consider explicit SIMD intrinsics for critical hot paths like checksum computation, encoding/decoding, and bulk data processing.
**Branch Prediction:** Minimize unpredictable branches in hot paths. Use `LIKELY`/`UNLIKELY` macros to hint branch prediction, e.g. for error cases and other rare or otherwise costly cases, but NOT for predicting popular configurations. Consider branchless alternatives for simple conditionals. Order switch cases and if-else chains by frequency.
**Memory and Resource Management:** Be mindful of memory allocations, especially in hot paths. Use RAII patterns, smart pointers, and RocksDB's memory management utilities appropriately.
**Hot Path Analysis:** When deciding how aggressively to optimize code, consider whether it's on a hot path:
- **Hot path** (executed thousands+ times, e.g., data access, iteration, compaction loops): Performance is paramount. Apply all optimization techniques—loop collapsing, SIMD, cache optimization, pre-allocation, etc. The cost of each operation is multiplied by execution frequency.
- **Cold path** (executed rarely, e.g., DB open, configuration parsing, error handling): Maintainability and clarity are more important. Prefer readable code over micro-optimizations. Complex optimizations here add maintenance burden with negligible performance benefit.
- **Warm path** (moderate frequency): Balance both concerns. Use profiling data to guide optimization decisions.
**Avoid Premature Optimization:** While performance is critical, focus on correctness first, then optimize based on profiling data. However, be performance-aware from the start—choosing the right algorithm and data structure upfront is not premature optimization. Use the hot path analysis above to decide how much optimization effort is warranted.
### API Design and Compatibility
**Backwards Compatibility:** RocksDB maintains strong backwards compatibility guarantees. Breaking changes are rare and require extensive justification. When deprecating features, follow the project's deprecation policy (typically spanning multiple releases).
**API Consistency:** New APIs should be consistent with existing patterns. Use similar naming conventions, parameter ordering, and return types. Reviewers will suggest changes to improve consistency with the broader codebase.
**Documentation:** Public APIs must be thoroughly documented. Include usage examples, parameter descriptions, and notes on thread safety, performance characteristics, and compatibility considerations.
---
## Component-Specific Guidance
### Database Core (`db`)
The database core handles write-ahead logging (WAL), memtables, compaction, and recovery. This component receives the most scrutiny in code reviews.
**Concurrency and Thread Safety:** Database operations are highly concurrent. Reviewers carefully examine locking strategies, atomic operations, and memory ordering. Document synchronization assumptions clearly. Use appropriate memory ordering semantics (`acquire`/`release` vs. `seq_cst`).
**Compaction Logic:** Changes to compaction are complex and high-risk. Ensure that compaction logic respects configured parameters, handles edge cases (empty databases, single-file compactions), and maintains correctness under concurrent operations.
**Error Propagation:** Database operations can fail in many ways (I/O errors, corruption, resource exhaustion). Ensure that errors are properly propagated, logged, and handled. Avoid assertions in production code paths.
**Testing:** Database core changes require extensive testing, including unit tests, integration tests, and stress tests. Test with various configurations, compaction styles, and concurrent workloads.
### Public Headers (`include`)
Public headers define RocksDB's API surface. Changes here have the highest compatibility impact.
**API Design:** New APIs should be intuitive, consistent with existing patterns, and well-documented. Consider how the API will be used in practice and avoid adding unnecessary complexity.
**Backwards Compatibility:** Breaking changes to public APIs require extensive justification and a deprecation plan. Maintain ABI compatibility for bug fixes and patch releases.
**Documentation:** Every public API must be thoroughly documented with usage examples, parameter descriptions, and notes on thread safety and performance characteristics.
**Deprecation:** When deprecating APIs, follow the project's policy. Mark deprecated APIs clearly, provide migration guidance, and maintain support for at least one major release.
### Internal Utilities (`util`)
Internal utilities provide common functionality used throughout the codebase.
**Code Reuse:** Utilities should be general-purpose and reusable. Avoid duplicating functionality that already exists elsewhere in the codebase.
**Error Handling:** Utility functions should handle errors robustly and propagate them appropriately. Consider edge cases like overflow, underflow, and invalid inputs.
**Testing:** Utility functions should have comprehensive test coverage, including edge cases and failure modes. Consider adding death tests for assertions.
**Performance:** Utilities are often used in hot paths. Ensure that implementations are efficient and avoid unnecessary allocations or copies.
### Table Management (`table`)
Table management handles SST file format, block-based tables, and table readers/writers.
**Block Format and Checksums:** Changes to block format require extreme care. Ensure that checksums are computed and verified correctly. Test with various compression algorithms and block sizes.
**Iterator Correctness:** Table iterators are used throughout the codebase. Ensure that iterator semantics (Seek, Next, Prev) are correct, especially at boundaries and with deletions.
**Caching and Prefetching:** Table readers interact with the block cache and prefetching logic. Ensure that cache keys are unique and that prefetching respects configured limits.
**Performance:** Table operations are performance-critical. Benchmark changes that could impact read or write performance.
### Utilities (`utilities`)
Utilities include optional features like transactions, backup engine, and checkpoint.
**Feature Isolation:** Utilities should be self-contained and not introduce unnecessary dependencies on core database internals.
**Deprecation and Cleanup:** Legacy features are being phased out. When removing deprecated code, ensure that migration paths are documented and that users have sufficient warning.
**Cross-Platform Compatibility:** Utilities often interact with OS-specific APIs. Ensure that code works on all supported platforms.
### Options and Configuration (`options`)
Options define RocksDB's configuration system.
**Type Safety:** Use appropriate types for options (e.g., `uint32_t` for flags, scoped enums for enumerated values).
**Deprecation Policy:** When deprecating options, follow the project's policy. Document the deprecation, provide migration guidance, and maintain support for at least one major release.
**Dynamic Configuration:** Some options can be changed dynamically. Ensure that dynamic changes are thread-safe and take effect correctly.
**Validation:** Validate option values and provide clear error messages for invalid configurations.
### Cache (`cache`)
Cache management is critical for RocksDB's performance.
**Concurrency:** Cache operations are highly concurrent. Ensure that implementations are thread-safe and use appropriate synchronization primitives.
**Performance:** Cache operations are in the hot path. Optimize for low latency and high throughput. Benchmark changes carefully.
**Memory Management:** Cache implementations must manage memory carefully to avoid leaks and excessive allocations.
**Eviction Policies:** Changes to eviction policies should be well-tested and benchmarked to ensure they improve overall performance.
---
## Code Review Checklist
When reviewing RocksDB code (or preparing code for review), use this checklist:
### Contract Boundaries
- [ ] Is each behavior owned by the right layer? High-level policy (for example,
"compaction wants this I/O mode") should live at the caller/policy layer, while
lower layers should expose generic mechanisms (for example, "open a fresh
reader", "skip shared cache insertion", or "use these FileOptions").
- [ ] Do comments and names describe local contracts rather than leaking a
specific caller's rationale into reusable APIs? Generic code should not need to
know about one current use case unless the API itself is intentionally
use-case-specific.
- [ ] Does each flag or parameter control one coherent behavior? If one boolean
starts implying ownership, cache policy, I/O mode, prefetching, and caller
identity, split it into explicit flags or an options struct.
- [ ] Could a future caller use this lower-level API without accidentally
inheriting assumptions from compaction, backup, user reads, or a particular
table format? If not, tighten the contract with assertions, clearer names, or
a narrower API.
- [ ] Are implementation details not being used as policy signals? Prefer an
explicit contract over inferring behavior from incidental fields such as file
options, cache handles, or current table-reader state.
### Correctness
- [ ] Does the change preserve database semantics (e.g., snapshot isolation, key ordering)?
- [ ] Are all error cases handled appropriately?
- [ ] Is the change thread-safe? Are synchronization primitives used correctly?
- [ ] Are there any potential data races or deadlocks?
### Testing
- [ ] Does the change include appropriate test coverage?
- [ ] Are edge cases and failure modes tested?
- [ ] Have the tests been run on all supported platforms?
- [ ] Are stress tests passing?
### Performance
- [ ] Are there benchmark results for performance-sensitive changes?
- [ ] Does the change avoid unnecessary allocations or copies?
- [ ] Are hot paths optimized appropriately?
### API and Compatibility
- [ ] Is the change backwards compatible?
- [ ] Are new APIs consistent with existing patterns?
- [ ] Is the public API documented?
- [ ] Are deprecated features handled according to policy?
### Code Quality
- [ ] Does the code follow RocksDB's style conventions?
- [ ] Is the code clear and maintainable?
- [ ] Are comments and documentation sufficient?
- [ ] Are there any code smells or anti-patterns?
---
## Common Review Feedback Patterns
The following patterns emerged as frequent sources of review feedback:
1.**Test Coverage:** Reviewers frequently request additional tests for edge cases, platform-specific behavior, and failure modes. Complex changes require comprehensive test coverage including unit tests, integration tests, and stress tests.
2.**Error Handling:** Ensure proper error propagation using RocksDB's `Status` type. Avoid silent failures and provide clear error messages that include context about what failed and why.
3.**API Design:** New APIs should be consistent with existing patterns. Use descriptive names that follow established conventions. Avoid breaking changes without strong justification and a clear deprecation plan.
4.**Documentation:** Public APIs must be documented with usage examples and notes on thread safety, performance characteristics, and compatibility considerations. Complex internal logic should also be well-commented.
5.**Performance:** Performance-sensitive changes require benchmark results to validate improvements. Use `db_bench` and other profiling tools to measure impact. Avoid premature optimization that adds complexity without measurable benefit.
6.**Concurrency:** Thread safety is critical in RocksDB. Document synchronization assumptions clearly. Use appropriate memory ordering semantics. Consider potential race conditions and deadlocks.
7.**Code Style:** Follow existing conventions for naming, formatting, and structure. Use `.clang-format` for consistent formatting. Prefer scoped enums (`enum class`) over unscoped enums.
8.**Backwards Compatibility:** RocksDB maintains strong compatibility guarantees. Breaking changes require extensive justification. When deprecating features, provide migration guidance and maintain support across multiple releases.
9.**Refactoring:** Reviewers appreciate refactoring that improves code readability and maintainability. Look for opportunities to deduplicate code and simplify complex logic.
10.**Platform Compatibility:** Ensure changes work correctly on all supported platforms (Linux, Windows, macOS) and with all supported compilers (GCC, Clang, MSVC).
11.**Contract Boundary Leaks:** When a change plumbs a new option or use-case
specific behavior through multiple subsystems, review the call chain for
contract leaks. Caller-specific rationale belongs at the call site or public API
documentation; reusable layers should expose precise, layer-local capabilities.
Watch especially for comments mentioning one caller in generic code, booleans
that silently bundle several behaviors, and downstream code inferring policy
from an implementation detail instead of an explicit option.
---
## Important tips
### Build system
* There are 3 build system. Make for git clones, BUCK (meta internal) for hg
clones, and CMake for some special cases.
* When a new .cc file is added, update Makefile, CMakeLists.txt, src.mk, BUCK.
* Don't manually edit BUCK file, after updating src.mk, run
/usr/local/bin/python3 buckifier/buckify_rocksdb.py to update it
* For -j in make command, use the number of CPU cores to decide it.
* When searching for references to something (a symbol, library, etc.), do not
restrict or truncate your search based on presumed relevance or scope. It is
important and time-saving to keep the repo reasonably consistent across
different build systems, programming languages, and even between
documentation and implementation.
### Avoiding mixed build modes with Make (use `AUTO_CLEAN=1`)
Object files are written to the same paths regardless of build flags, so
reusing objects from a prior build with different flags causes confusing
linker errors, etc. This problem is essentially avoidable by ALWAYS using
`AUTO_CLEAN=1 make -j<n> <something>` for manual make invocations. This
will automatically clean object files if the build parameters/flavor have
changed. The `build_tools/rockstest.sh` / `rocksptest.sh` helpers described
below set `AUTO_CLEAN=1` for you.
### Source checks
* Run `make check-sources` before committing. This catches non-ASCII
characters in source files and other source-level issues that CI will
reject. In particular, **do not use Unicode characters** (em dashes,
smart quotes, etc.) in comments or strings -- use ASCII equivalents
(`--` instead of em dash, `'` instead of smart quote, etc.).
### License headers
* Every new source file needs a license header. For a file that does **not**
carry an outside/third-party copyright, use the standard Meta dual-licensed
header (the dual-license designation is required -- a bare
"All Rights Reserved" copyright is not an acceptable open-source header):
```
// Copyright (c) Meta Platforms, Inc. and affiliates.
// This source code is licensed under both the GPLv2 (found in the
// COPYING file in the root directory) and Apache 2.0 License
// (found in the LICENSE.Apache file in the root directory).
```
Use a `#` comment prefix instead of `//` for shell, Python, and Makefile
fragments.
* Files derived from an external source (e.g. LevelDB) keep their original
upstream copyright line in addition to the header above.
### RTTI and dynamic_cast
* Production code and `db_stress` must build in **release mode
(`-fno-rtti`)**. Do not use `dynamic_cast` anywhere except unit tests.
Use `static_cast_with_check` from `util/cast_util.h` (validates with
`dynamic_cast` in debug builds, plain `static_cast` in release).
* Unit tests (`*_test.cc`) are built in debug mode with RTTI enabled.
### Cross-platform / portability
Local `make` only exercises Linux with GCC/Clang, but CI
(`.github/workflows/pr-jobs.yml` and `nightly.yml`) gates on a much wider
matrix, so portability breaks are invisible locally until CI fails. Code must
> NOTE: Entries for next release do not go here. Follow instructions in `unreleased_history/README.txt`
## 11.6.0 (07/02/2026)
### New Features
* Added EXPERIMENTAL embedded blob SST support through `SstFileWriter::OpenWithEmbeddedBlobs()`, storing eligible large values as same-file blob records in block-based SST files and resolving them transparently for reads. This niche feature currently supports uncompressed embedded blobs only; compression options are placeholders and compression support is deferred to follow-up work.
### Public API Changes
* Expanded the C API (`include/rocksdb/c.h`) with a large set of new `rocksdb_*` functions, mostly option getters/setters plus table-properties, job/event-listener, and metadata accessors, a WAL filter, a ReadOptions table filter, and a backup exclude-files callback. Many are now produced by a new semi-automated generator (`tools/c_api_gen/`) from the C++ headers; `include/rocksdb/c.h` remains a single self-contained header and the signatures of pre-existing functions are unchanged.
### Bug Fixes
* Reverted PR14831 that made range_lock_manager aware of reverse-order CF
* Fixed a bug in `RandomAccessFileReader::ReadAsync` where an already-aligned direct-IO read request with a null `scratch` and a caller-provided `aligned_buf` would take the "already aligned" fast path and submit the null buffer to the underlying async read (e.g. a null iovec base to io_uring, failing with EFAULT). This could surface as spurious iterator failures during async prefetch (MultiScan with async IO) on direct-IO databases. The async path now allocates a backing buffer in this case, matching the synchronous `Read` path.
* Fixed a bug where closing a read-only DB instance could delete live SST files created by a concurrent read-write DB sharing the same directory.
## 11.5.0 (06/16/2026)
### New Features
* External table readers that open files through `ExternalTableOptions::fs` now update RocksDB SST/file-read statistics and file IO listener callbacks, making external file IO activity visible in existing read metrics.
* Added `DB::PrepareFileIngestion()`, a two-phase form of `IngestExternalFile`/`IngestExternalFiles`. `PrepareFileIngestion()` performs all of the work that does not require the DB mutex (validating the arguments, reading each external file's metadata, and linking/copying the files into the DB) and returns an opaque `FileIngestionHandle`. `DB::CommitFileIngestionHandle()` (or `DB::CommitFileIngestionHandles()` for several handles at once) then makes the prepared files visible; multiple handles are committed atomically in a single MANIFEST write, or none are. `FileIngestionHandle::Abort()` (or simply destroying the handle) cancels a prepared ingestion and rolls it back. This gives flexibility to the application to Prepare a file separately from when its committed, and may shorten an applications critical path on `IngestExternalFile`.
* Added two stats histograms reporting the per-call latency (in microseconds) of each successful `IngestExternalFile`/`IngestExternalFiles` call, split by phase: `rocksdb.ingest.external.file.prepare.micros` (argument validation and reading/validating/linking the external files, which does not block live writes) and `rocksdb.ingest.external.file.run.micros` (the ingestion performed under the DB mutex while live writes are blocked). Failed ingestions are not recorded.
* Added `DBOptions::use_direct_io_for_compaction_reads` (default false). When enabled, compaction-input SST reads use `O_DIRECT` while user reads remain buffered, avoiding page-cache eviction of hot user-read data by sequential compaction scans. Pair with `use_direct_io_for_flush_and_compaction = true` on write-heavy workloads for direct I/O on both compaction inputs and outputs. Rejected at Open when combined with `allow_mmap_reads = true`. No-op when `use_direct_reads = true` is already set (since all reads are already direct).
### Public API Changes
* `DB::GetPreparedFileInfoForExternalSstIngestion()` can now prepare metadata for a live DB-generated SST file so it can be passed to `IngestExternalFileArg::file_infos`. The input path must exactly match a live table file owned by the source DB, and the returned metadata handle must outlive the ingestion.
* `ExternalSstFileInfo` gained a `prepared_file_info` member produced by `SstFileWriter::Finish`, and `IngestExternalFileArg` gained a `file_infos` field: a vector of borrowed `const PreparedFileInfo*` pointers, parallel to `external_files`. The owning `ExternalSstFileInfo::prepared_file_info` handle must outlive the ingestion. When set, `IngestExternalFiles()` reuses the supplied per-file metadata instead of re-opening and scanning each file to recompute it, avoiding that extra I/O.
* Corrected the public C++ option spelling from `memtable_veirfy_per_key_checksum_on_seek` to `memtable_verify_per_key_checksum_on_seek`. RocksDB continues to accept the old misspelled OPTIONS-file key for compatibility, while newly serialized OPTIONS files use the corrected key.
* Added experimental read-scoped block buffer provider API, configured through `ReadOptions::read_scoped_block_buffer_provider`, for supported block-based table iterator scans and MultiScan reads to use caller-provided read-scoped storage for final data-block contents. When configured, supported provider-backed scan data-block reads bypass the data-block cache. The provider is ignored when mmap reads are enabled. RocksDB may still use ordinary temporary scratch for serialized block bytes, such as when a block may be compressed. Get and MultiGet do not currently provide API guarantees for this provider.
* Added `FileSystem::SyncFile()` and `Env::SyncFile()` public APIs for syncing or fsyncing a file by name without requiring callers to reopen it as writable.
### Behavior Changes
* The default `compression` for column families is now `kLZ4Compression` rather than `kSnappyCompression`. This affects only column families that do not explicitly set `compression`, and only newly-written SST files. The change is fully compatible: existing data remains readable with no migration needed, as RocksDB selects the decompressor per block. LZ4 offers slightly better compression ratios and decompression CPU efficiency vs. Snappy and similar compression CPU, tested across various server CPUs. When support is not compiled in, the fallback path for default compression is LZ4 -> Snappy -> NoCompression.
* Disable parallel compression (ignore `CompressionOptions::parallel_threads`) for fast built-in compressors: Snappy, LZ4 (accelerated, not LZ4HC), and accelerated levels of ZSTD (level < 0). These do not generally benefit from parallel compression. This behavior can be overridden with a custom Compressor from a custom CompressionManager.
* `PosixFileSystem::OptimizeForCompactionTableRead` now delegates to the base `FileSystem::OptimizeForCompactionTableRead` implementation before applying its Linux-only compaction-readahead clamp (added for #12038). The returned `FileOptions::use_direct_reads` therefore reflects `DBOptions::use_direct_reads`, consistent with the base class and with other `FileSystem` implementations. Previously the override ignored the base implementation and keyed both the returned options and the Linux readahead clamp off the incoming `FileOptions` rather than `DBOptions`. At the in-tree call site the incoming `FileOptions::use_direct_reads` already matches `DBOptions::use_direct_reads`, so this is effectively a no-op for existing configurations; it removes a latent inconsistency where a caller passing `FileOptions` whose `use_direct_reads` disagreed with `DBOptions` would have had the global flag silently ignored for compaction-input reads on Linux. (Note: the `#12038` readahead clamp still keys off the global `use_direct_reads`; it is not skipped for the compaction-only `use_direct_io_for_compaction_reads` path, since that flag enables O_DIRECT after this hook runs.)
* Brought more unity to `kLZ4Compression` and `kLZ4HCCompression` by giving each access to the other's compression levels. Negative (fast) values previously only available to LZ4 and positive (slow) values previously only available to LZ4HC are now available to both, so configuring a non-default `compression_opts.level` now selects the LZ4 compressor variant. This is a behavior change for previously tolerated but dubious configurations such as positive compression level with `kLZ4Compression`. `kLZ4Compression` and `kLZ4HCCompression` keep their effective default levels, equivalent to -1 and 9 respectively.
* Removed a discontinuity in `kZSTD` compression levels at `compression_opts.level == 0` by mapping that setting to `level = -1` rather than to `level = 3` as the ZSTD library does internally. This improves the compression auto-tuning landscape.
### Bug Fixes
* Fixed a bug where the range lock manager would crash with an assertion failure when using a reverse comparator column family.
### Performance Improvements
* Reduced commit latency when ingesting many external files by allowing `IngestExternalFileOptions::file_opening_threads` to open table readers for committed ingested files using multiple threads.
* Reduced commit latency for large external file ingestions into the last level by adding `IngestExternalFileOptions::prefetch_lmax_index_and_filter_blocks`, which can skip commit-time index and filter block prefetching for cache-backed table metadata.
## 11.4.0 (06/02/2026)
### Public API Changes
* Added `rocksdb_options_set_memtable_batch_lookup_optimization()` and `rocksdb_options_get_memtable_batch_lookup_optimization()` to the C API, exposing the existing `AdvancedColumnFamilyOptions::memtable_batch_lookup_optimization` field. This allows C API users (and downstream language bindings) to enable the skip-list memtable's batch-lookup optimization for `MultiGet`, which caches the search path between consecutive keys and reduces per-key cost from O(log N) to O(log d) where d is the distance between consecutive keys.
* Added `rocksdb_readoptions_set_optimize_multiget_for_io()` and `rocksdb_readoptions_get_optimize_multiget_for_io()` to the C API, exposing the existing `ReadOptions::optimize_multiget_for_io` field. This allows C API users (and downstream language bindings) to opt out of the multi-level parallel MultiGet path, which only takes effect when the library is built with `USE_COROUTINES`.
* Added `rocksdb_block_based_table_index_block_search_type_auto` enum constant and the `rocksdb_block_based_options_set_uniform_cv_threshold()` setter to the C API. The constant exposes `BlockSearchType::kAuto`, and the setter exposes `BlockBasedTableOptions::uniform_cv_threshold`. Both are required for `kAuto` index-block search to take effect from the C API: without setting `uniform_cv_threshold >= 0`, the per-block `is_uniform` footer bit is never written, and `kAuto` falls back to binary search at read time.
* Added `EventListener::OnDBShutdownBegin`, a callback that fires once when a DB begins shutdown, including when RocksDB cleans up a failed `DB::Open()` attempt.
* Added a new PerfContext counter `blob_cache_read_byte` for blob cache bytes read.
### Behavior Changes
* Remote compaction now falls back to local compaction when the primary cannot deserialize a successful `CompactionServiceResult` before installing any remote output files.
* StringToMap (rocksdb/convenience.h) now preserves the outer braces of nested values so each map entry is in self-contained form and can be embedded directly into another `key=value;` string (e.g. SetOptions) without losing inner ';' delimiters. A new symmetric MapToString utility is provided.
### Bug Fixes
* Fixed a bug where `AbortAllCompactions()` could leave automatic compaction work unscheduled when aborting before the background worker picked it, causing compaction and write stalls until DB restart.
* Fix a bug where db had a false positive compaction corruption error due to remote compaction service result with `has_accurate_num_input_records=false` not being serialized.
* Fixed WritePrepared TransactionDB cleanup after retryable commit or rollback write failures so prepared transactions remain rollbackable instead of leaving unresolved prepared state.
* Fix a rare corruption bug for tiered compaction that incorrectly moved last level range tombstones into proximal level. This corruption error is surfaced when force_consistency_checks is enabled.
### Performance Improvements
* Reduce the likelihood of user-facing metadata operations blocking on MANIFEST rotation when file creation is slow, such as on remote storage. MANIFEST write batches containing foreground edits from `CreateColumnFamily()`, `DropColumnFamily()`, `CreateColumnFamilyWithImport()`, `IngestExternalFile()`, and `DeleteFilesInRanges()` now get a relaxed file size threshold for triggering MANIFEST rotation; background-only MANIFEST write batches, such as compaction and flush, continue to use the normal threshold. This change should make auto-tuning manifest file size more attractive (see `max_manifest_file_size` and `max_manifest_space_amp_pct` options).
## 11.3.0 (05/15/2026)
### New Features
* Add experimental DB option `async_wal_precreate` to precreate the next WAL file in a background thread and reduce foreground WAL rotation latency. The option is sanitized to false when WAL recycling is enabled.
* Added a new `EventListener::OnCompactionPreCommit` callback that fires after a compaction job finishes but before its input files are released (i.e. while `FileMetaData::being_compacted` is still true). Listeners that maintain bookkeeping of which files are currently being compacted can clean up such state in this new callback to avoid races with concurrent compaction picking, where another thread might pick up the same files for a new compaction immediately after `being_compacted` is flipped back to false but before `OnCompactionCompleted` fires. The default implementation is a no-op so this is not a breaking change.
* Add mutable DBOption `optimize_manifest_for_recovery` (default false). When enabled, RocksDB can reduce recovery work after a clean shutdown, which may lower DB::Open latency on warm reopens.
* Added public utility APIs `ParseCompressionNameForDisplay()` to convert `TableProperties::compression_name` into a human-readable compression name for both legacy and format_version 7+ SST metadata, including custom `CompressionManager`-provided display names for custom compression types.
* Add `reuse_manifest_on_open` DBOption (default false). When enabled, DB::Open reuses the existing MANIFEST file for append instead of creating a fresh one, avoiding the cost of serializing the entire database state into a new MANIFEST on the first post-open write. To prevent this feature from interfering with manifest file size auto-tuning, an extra forward-compatible field is now always added to the MANIFEST (to track the last "compacted" size).
### Behavior Changes
* Read-only open with `error_if_wal_file_exists=true` now tolerates empty WAL files so empty precreated WALs do not prevent inspection.
* WriteCommitted TransactionDB now matches WritePrepared and WriteUnprepared compaction filtering in both single- and two-write-queue modes: a compaction filter's FilterMergeOperand will not be invoked on merge operands at or below the latest published sequence number.
### Bug Fixes
* Fixed blob-backed wide-column merge reads to preserve correct status
propagation and resolution across memtable, read-only, and secondary DB
paths.
* Fixed a bug where `DB::GetCreationTimeOfOldestFile()` could return inaccurate results instead of the real creation time when called shortly after opening a legacy DB (one whose manifest lacks `file_creation_time`) with `open_files_async = true`. The API now waits for background SST file loading to complete only when needed; modern DBs are unaffected.
* Fixed merge reads against wide-column/blob-backed base values to preserve precise failure statuses, including `GetMergeOperands()` and direct-write memtable reads.
* Fix bug in range tombstone synthesis that covers live keys added during an IngestExternalFile
* Reject the empty string as a column family name in `DB::CreateColumnFamily` / `DB::CreateColumnFamilies`. Previously such calls returned OK and a usable handle, but the column family was not persisted in the manifest, so any data written to it was silently lost on DB reopen.
* Fixed a bug where a WriteCommitted TransactionDB using commit-bypass WBWI ingestion could drop an entry that is still visible at the published sequence boundary.
## 11.2.0 (04/18/2026)
### New Features
* Added experimental `DBOptions::fast_sst_open` option. When enabled, RocksDB retrieves opaque file system metadata for SST files after flush, compaction, and external file ingestion, persists it in the MANIFEST, and passes it back to the file system on subsequent file opens to accelerate DB open time.
* Added new option `min_tombstones_for_range_conversion` in `AdvancedColumnFamilyOptions`. When set to a non-zero value N, forward or reverse iteration will convert N or more contiguous point tombstones into a range tombstone in the mutable memtable. Future read operations will then be able to benefit from range tombstone optimizations. There are some limitations when it comes to table_filters, prefix_filters, and UDTs. See header comments for more details.
* Added read-triggered compaction: a new column family option `read_triggered_compaction_threshold` (default 0, disabled) that marks SST files for compaction when their read frequency (`num_collapsible_entry_reads_sampled / file_size`) exceeds the threshold. This helps reduce read amplification for frequently-read ("hot") keys. A new DB option `max_compaction_trigger_wakeup_seconds` (default 43200s / 12 hours) controls the maximum interval for periodic compaction score re-evaluation, which is necessary for this feature to work on quiet (no-write) databases.
### Public API Changes
* Added new `WideColumnBlobResolver` interface and `CompactionFilter::FilterV4()` method, including `ResolveColumn()` / `ResolveColumns()` helpers for lazy loading blob column values in wide-column entities during compaction. This allows compaction filters to resolve blob values on-demand, avoiding unnecessary I/O for blob columns they don't need to access.
* Changed experimental feature `ExternalTableReader::Get` and `ExternalTableReader::MultiGet` to use `PinnableSlice` instead of `std::string` for output values, enabling zero-copy pinning. This will break existing implementations.
* Added `SstFileReader::Get` and `SstFileReader::MultiGet` overloads that accept `PinnableSlice`/`std::vector<PinnableSlice>*`, enabling zero-copy reads when the underlying `TableReader` supports pinning.
### Behavior Changes
* Prefix filter changes - when seeking to a key that is out of domain, and total_order_seek is false, total_order_seek is treated as if it were true. When prefix_same_as_start = true, now iterating past a key that is out of domain invalidates the iterator. The existing behavior does not check for InDomain in the DBIter, so Transform() can produce undefined behavior (e.g. key of size 3 on FixedPrefixTransform(4)).
* Wide-column entities with blob-backed columns now use a new V2 on-disk encoding; older RocksDB versions that do not support wide-column blob separation will reject DBs or SSTs containing those entities.
### Bug Fixes
* Fix blob garbage accounting for blob direct-write flushes so flush-time filtering and overwrite elision correctly register obsolete blob bytes in blob metadata.
* Fix a memory accounting leak in IODispatcher where ReadIndex() moved block values out of ReadSet without releasing the associated prefetch memory, causing subsequent prefetches to be blocked when max_prefetch_memory_bytes was set.
* Fix MultiScan to fall back to synchronous coalesced reads when async I/O is unsupported at runtime.
## 11.1.0 (03/23/2026)
### New Features
* Add a new option `open_files_async`. The existing behavior is on DB open, we open all sst files and do basic validations. For very large DBs on remote filesystems with many ssts, this may take very long. This option performs these validations instead in the background. Open errors found by this async background task are surfaced as a new background error kAsyncFileOpen.
* Added `BlockBasedTableOptions::kAuto` index block search type that automatically selects between binary and interpolation search on a per-index-block basis. During SST construction, each index block's key distribution uniformity is analyzed using the coefficient of variation of key gaps, and index blocks with uniform keys use interpolation search while others fall back to binary search. The uniformity threshold is configurable via `BlockBasedTableOptions::uniform_cv_threshold` (default: 0.2).
* Introduced enforce_write_buffer_manager_during_recovery option to allow WriteBufferManager to be enforced during WAL recovery. (Default: true)
* Add `memtable_batch_lookup_optimization` option to use batch lookup optimization for memtable MultiGet. For skip list memtables, after each key lookup, the search path is cached and reused for the next key, reducing per-key cost from O(log N) to O(log d) where d is the distance between consecutive keys. Benchmarks show ~7% improvement in memtable-resident MultiGet throughput.
* Added `BlockBasedTableOptions::PrepopulateBlockCache::kFlushAndCompaction` to prepopulate the block cache during both flush and compaction. Compaction-warmed blocks are inserted at `BOTTOM` priority (vs `LOW` for flush) so they are evicted first under cache pressure. Recommended only for use cases where most or all of the database is expected to reside in cache (e.g., tiered or remote storage where the working set fits in cache).
* Added new mutable DB option `verify_manifest_content_on_close` (default: false). When enabled, on DB close the MANIFEST file is read back and all records are validated (CRC checksums and logical content). If corruption is detected, a fresh MANIFEST is written from in-memory state.
### Behavior Changes
* num_reads_sampled now factors in re-seeks and next/prev() on file iterators for files in L1+. next/prev() is discounted by 64x compared to seek due to being a much cheaper call.
* Remote compaction workers now skip WAL recovery when opening the secondary DB instance, since only the LSM state from MANIFEST is needed for compaction. This reduces I/O and speeds up remote compaction startup.
### Bug Fixes
* Fix a bug in round-robin compaction that missed selecting input files that are needed to guarantee data correctness and cause crashing in debug builds or silent data corruption in release builds for Get().
## 11.0.0 (02/23/2026)
### New Features
* Added support for storing wide-column entity column values in blob files. When `min_blob_size` is configured, large column values in wide-column entities will be stored in blob files, reducing SST file size and improving read performance.
* Added `CompactionOptionsFIFO::max_data_files_size` to support FIFO compaction trimming based on combined SST and blob file sizes. Added `CompactionOptionsFIFO::use_kv_ratio_compaction` to enable a capacity-derived intra-L0 compaction strategy optimized for BlobDB workloads, producing uniform-sized compacted files for predictable FIFO trimming.
* Include interpolation search as an alternative to binary search, which typically performs better when keys are uniformly distributed. This is exposed as a new table option `index_block_search_type`. The default is `binary_search`.
### Public API Changes
* Added new virtual methods `AbortAllCompactions()` and `ResumeAllCompactions()` to the `DB` class. Added new `Status::SubCode::kCompactionAborted` to indicate a compaction was aborted. Added `Status::IsCompactionAborted()` helper method to check if a status represents an aborted compaction.
* Drop support for reading (and writing) SST files using `BlockBasedTableOptions.format_version` < 2, which hasn't been the default format for about 10 years. An upgrade path is still possible with full compaction using a RocksDB version >= 4.6.0 and < 11.0.0 and then using the newer version.
* Remove deprecated raw `DB*` variants of `DB::Open` and related functions. Some other minor public APIs were updated as a result
* Remove deprecated DB option `skip_checking_sst_file_sizes_on_db_open`. The option was deprecated in 10.5.0 and has been a no-op since then. File size validation is now always performed in parallel during DB open.
* Remove deprecated `SliceTransform::InRange()` virtual method and the `in_range` callback parameter from `rocksdb_slicetransform_create()` in the C API. `InRange()` was never called by RocksDB and existed only for backward compatibility.
* Remove deprecated, unused APIs and options: `ReadOptions::managed` and `ColumnFamilyOptions::snap_refresh_nanos`. Corresponding C and Java APIs are also removed.
* Remove deprecated `SstFileWriter::Add()` method (use `Put()` instead) and the deprecated `skip_filters` parameter from `SstFileWriter` constructors (use `BlockBasedTableOptions::filter_policy` set to `nullptr` to skip filter generation instead).
### Behavior Changes
* Change the default value of `CompactionOptionsUniversal::reduce_file_locking` from `false` to `true` to improve write stall and reduce read regression
### Bug Fixes
* Fix longstanding failures that can arise from reading and/or compacting old DB dirs with range deletions (likely from version < 5.19.0) in many newer versions.
* Fix a bug where WritePrepared/WriteUnprepared TransactionDB with two_write_queues=true could experience "sequence number going backwards" corruption during recovery from a background error, due to allocated-but-not-published sequence numbers not being synced before creating new WAL files.
### Performance Improvements
* Add a new table option `separate_key_value_in_data_block`. When set to true keys and values will be stored separately in the data block, which can result in higher cpu cache hit rate and better compression. Works best with data blocks with sufficient restart intervals and large values. Previous versions of RocksDB will reject files written using this option.
## 10.11.0 (01/23/2026)
### Public API Changes
* New SetOptions API that allows setting options for multiple CFs, avoiding the need to reserialize OPTIONS file for each CF
* Remove remaining pieces of Lua integration
### Behavior Changes
* The new default for `BlockBasedTableOptions::format_version` is 7, which has been supported since RocksDB 10.4.0 and is required in order to use CompressionManagers supporting custom compression types.
### Bug Fixes
* Fixed a small performance bug with `format_version=7` when decompressing formats other than Snappy and ZSTD.
* Fixed an infinite compaction loop bug with User-Defined Timestamps (UDT) where bottommost files were repeatedly marked for compaction even though their timestamp could not be collapsed.
* Bugfix for persisted UDT record sequence number zeroing logic.
## 10.10.0 (12/16/2025)
### Bug Fixes
* Fixed a bug in best-efforts recovery that causes use-after-free crashes when accessing SST files that were cached during the recovery.
* Fix resumable compaction incorrectly allowing resumption from a truncated range deletion that is not well handled currently.
* Fixed a bug in `PosixRandomFileAccess` IO uring submission queue ownership & management. Fix eliminates the false positive 'Bad cqe data' IO errors in `PosixRandomFileAccess::MultiRead` when interleaved with `PosixRandomFileAccess::ReadAsync` on the same thread.
## 10.9.0 (11/21/2025)
### New Features
* Added an auto-tuning feature for DB manifest file size that also (by default) improves the safety of existing configurations in case `max_manifest_file_size` is repeatedly exceeded. The new recommendation is to set `max_manifest_file_size` to something small like 1MB and tune `max_manifest_space_amp_pct` as needed to balance write amp and space amp in the manifest. Refer to comments on those options in `DBOptions` for details. Both options are (now) mutable.
* Added a new API to support option migration for multiple column families
* Added new option target_file_size_is_upper_bound that makes most compaction output SST files come close to the target file size without exceeding it, rather than commonly exceeding it by some fraction (current behavior). For now the new behavior is off by default, but we expect to enable it by default in the future.
* Add a new option allow_trivial_move in CompactionOptions to allow CompactFiles to perform trivial move if possible. By default the flag of allow_trivial_move is false, so it preserve the original behavior.
### Public API Changes
* To reduce risk of ODR violations or similar, `ROCKSDB_USING_THREAD_STATUS` has been removed from public headers and replaced with static `const bool ThreadStatus::kEnabled`. Some other uses of conditional compilation have been removed from public API headers to reduce risk of ODR violations or other issues.
### Behavior Changes
* PosixWritableFile now repositions the seek pointer to the new end of file after a call to Truncate.
* Updated standalone range deletion L0 file compaction behavior to avoid compacting with any newer L0 files (which is expensive and not useful).
### Bug Fixes
* Fix a bug where compaction with range deletion can persist kTypeMaxValid in MANIFEST as file metadata. kTypeMaxValid is not supposed to be persisted and can change as new value types are introduced. This can cause a forward compatibility issue where older versions of RocksDB don't recognize kTypeMaxValid from newer versions. A new placeholder value type kTypeTruncatedRangeDeletionSentinel is also introduced to replace kTypeMaxValid when reading existing SST files' metadata from MANIFEST. This allows us to strengthen some checks to avoid using kTypeMaxValid in the future.
* Fixed a bug where `DB::GetSortedWalFiles()` could hang when waiting for a purge operation that found nothing to do (potentially triggered by iterator release, flush, compaction, etc.).
* Fixed a bug in MultiScan where `max_sequential_skip_in_iterations` could cause the iterator to seek backward to already-unpinned blocks when the same user key spans multiple data blocks, leading to assertion failures or seg fault.
* Fixed a bug for `WAL_ttl_seconds > 0` use cases where the newest archived WAL files could be incorrectly deleted when the system clock moved backwards.
### Performance Improvements
* Added optimization that allowed for the asynchronous prefetching of all data outlined in a multiscan iterator. This optimization was applied to the level iterator, which prefetches all data through each of the block-based iterators.
## 10.8.0 (10/21/2025)
### New Features
* Add kFSPrefetch to FSSupportedOps enum to allow file systems to indicate prefetch support capability, avoiding unnecessary prefetch system calls on file systems that don't support them.
* Added experimental support `OpenAndCompactOptions::allow_resumption` for resumable compaction that persists progress during `OpenAndCompact()`, allowing interrupted compactions to resume from the last progress persitence. The default behavior is to not persist progress.
### Public API Changes
* Allow specifying output temperature in CompactionOptions
* Added `DB::FlushWAL(const FlushWALOptions&)` as an alternative to `DB::FlushWAL(bool sync)`, where `FlushWALOptions` includes a new `rate_limiter_priority` field (default `Env::IO_TOTAL`) that allows rate limiting and priority passing of manual WAL flush's IO operations.
* The MultiScan API contract is updated. After a multi scan range got prepared with Prepare API call, the following seeks must seek the start of each prepared scan range in order. In addition, when limit is set, upper bound must be set to the same value of limit before each seek
### Behavior Changes
* `kChangeTemperature` FIFO compaction will now honor `compaction_target_temp` to all levels regardless of `cf_options::last_level_temperature`
* Allow UDIs with a non BytewiseComparator
### Bug Fixes
* Fix incorrect MultiScan seek error status due to bugs in handling range limit falling between adjacent SST files key range.
* Fix a bug in Page unpinning in MultiScan
### Performance Improvements
* Fixed a performance regression in LZ4 compression that started in version 10.6.0
## 10.7.0 (09/19/2025)
### New Features
* Add the fail_if_no_udi_on_open flag in BlockBasedTableOption to control whether a missing user defined index block in a SST is a hard error or not.
* A new flag memtable_verify_per_key_checksum_on_seek is added to AdvancedColumnFamilyOptions. When it is enabled, it will validate key checksum along the binary search path on skiplist based memtable during seek operation.
* Introduce option MultiScanArgs::use_async_io to enable asynchronous I/O during MultiScan, instead of waiting for I/O to be done in Prepare().
* Add new option `MultiScanArgs::max_prefetch_size` that limits the memory usage of per file pinning of prefetched blocks.
* Improved `sst_dump` by allowing standalone file and directory arguments without `--file=`. Also added new options and better output for `sst_dump --command=recompress`. See `sst_dump --help`
### Public API Changes
* HyperClockCache with no `estimated_entry_charge` is now production-ready and is the preferred block cache implementation vs. LRUCache. Please consider updating your code to minimize the risk of hitting performance bottlenecks or anomalies from LRUCache. See cache.h for more detail.
* RocksDB now requires a C++20 compatible compiler (GCC >= 11, Clang >= 10, Visual Studio >= 2019), including for any code using RocksDB headers.
* MultiScanArgs used to have a default constructor with default parameter of BytewiseComparator. Now it always requires Comparator in its constructor.
### Behavior Changes
* The default provided block cache implementation is now HyperClockCache instead of LRUCache, when `block_cache` is nullptr (default) and `no_block_cache==false` (default). We recommend explicitly creating a HyperClockCache block cache based on memory budget and sharing it across all column families and even DB instances. This change could expose previously hidden memory or resource leaks.
### Bug Fixes
* Reported numbers for compaction and flush CPU usage now include time spent by parallel compression worker threads. This now means compaction/flush CPU usage could exceed the wall clock time.
* Fix a race condition in FIFO size-based compaction where concurrent threads could select the same non-L0 file, causing assertion failures in debug builds or "Cannot delete table file from LSM tree" errors in release builds.
* Fix a bug in RocksDB MultiScan with UDI when one of the scan ranges is determined to be empty by the UDI, which causes incorrect results.
### Performance Improvements
* Add a new table property "rocksdb.key.smallest.seqno" which records the smallest sequence number of all keys in file. It makes ingesting DB generated files faster by
avoiding scanning the whole file to find the smallest sequence number.
* Add a new experimental PerKeyPointLockManager to improve efficiency under high lock contention. PointLockManager was not efficient when there is high write contention on same key, as it uses a single conditional variable per lock stripe. PerKeyPointLockManager uses per thread conditional variable supporting fifo order. Although this is an experimental feature. By default, it is disabled. A new boolean flag TransactionDBOptions::use_per_key_point_lock_mgr is added to optionally enable it. Search the flag in code for more info.
Together, a new configuration TransactionOptions::deadlock_timeout_us is added, which allows the transaction to wait for a short period before perform deadlock detection. When the workload has low lock contention, the deadlock_timeout_us can be configured to be slightly higher than average transaction execution time, so that transaction would likely be able to take the lock before deadlock detection is performed when it is waiting for a lock. This allows transaction to reduce CPU cost on performing deadlock detection, which could be expensive in CPU time. When the workload has high lock contention, the deadlock_timeout_us can be configured to 0, so that transaction would perform deadlock detection immediately. By default the value is 0 to keep the behavior same as before.
* Majorly improved CPU efficiency and scalability of parallel compression (`CompressionOptions::parallel_threads` > 1), though this efficiency improvement makes parallel compression currently incompatible with UserDefinedIndex and with old setting of `decouple_partitioned_filters=false`. Parallel compression is now considered a production-ready feature. Maximum performance is available with `-DROCKSDB_USE_STD_SEMAPHORES` at compile time, but this is not currently recommended because of reported bugs in implementations of `std::counting_semaphore`/`binary_semaphore`.
## 10.6.0 (08/22/2025)
### New Features
* Introduce column family option `cf_allow_ingest_behind`. This option aims to replace `DBOptions::allow_ingest_behind` to enable ingest behind at the per-CF level. `DBOptions::allow_ingest_behind` is deprecated.
* Introduce `MultiScanArgs::io_coalesce_threshold` to allow a configurable IO coalescing threshold.
### Public API Changes
* `IngestExternalFileOptions::allow_db_generated_files` now allows files ingestion of any DB generated SST file, instead of only the ones with all keys having sequence number 0.
* `decouple_partitioned_filters = true` is now the default in BlockBasedTableOptions.
* GetTtl() API is now available in TTL DB
* Minimum supported version of LZ4 library is now 1.7.0 (r129 from 2015)
* Some changes to experimental Compressor and CompressionManager APIs
* A new Filesystem::SyncFile function is added for syncing a file that was already written, such as on file ingestion. The default implementation matches previous RocksDB behavior: re-open the file for read-write, sync it, and close it. We recommend overriding for FileSystems that do not require syncing for crash recovery or do not handle (well) re-opening for writes.
### Behavior Changes
* When `allow_ingest_behind` is enabled, compaction will no longer drop tombstones based on the absence of underlying data. Tombstones will be preserved to apply to ingested files.
### Bug Fixes
* Files in dropped column family won't be returned to the caller upon successful, offline MANIFEST iteration in `GetFileChecksumsFromCurrentManifest`.
* Fix a bug in MultiScan that causes it to fall back to a normal scan when dictionary compression is enabled.
* Fix a crash in iterator Prepare() when fill_cache=false
* Fix a bug in MultiScan where incorrect results can be returned when a Scan's range is across multiple files.
* Fixed a bug in remote compaction that may mistakenly delete live SST file(s) during the cleanup phase when no keys survive the compaction (all expired)
* Allow a user defined index to be configured from a string.
* Make the User Defined Index interface consistently use the user key format, fixing the previous mixed usage of internal and user key.
### Performance Improvements
* Small improvement to CPU efficiency of compression using built-in algorithms, and a dramatic efficiency improvement for LZ4HC, based on reusing data structures between invocations.
## 10.5.0 (07/18/2025)
### Public API Changes
* DB option skip_checking_sst_file_sizes_on_db_open is deprecated, in favor of validating file size in parallel in a thread pool, when db is opened. When DB is opened, with paranoid check enabled, a file with the wrong size would fail the DB open. With paranoid check disabled, the DB open would succeed, the column family with the corrupted file would not be read or write, while the other healthy column families could be read and write normally. When max_open_files option is not set to -1, only a subset of the files will be opened and checked. The rest of the files will be opened and checked when they are accessed.
### Behavior Changes
* PessimisticTransaction::GetWaitingTxns now returns waiting transaction information even if the current transaction has timed out. This allows the information to be surfaced to users for debugging purposes once it is known that the timeout has occurred.
* A new API GetFileSize is added to FSRandomAccessFile interface class. It uses fstat vs stat on the posix implementation which is more efficient. Caller could use it to get file size faster. This function might be required in the future for FileSystem implementation outside of the RocksDB code base.
* RocksDB now triggers eligible compactions every 12 hours when periodic compaction is configured. This solves a limitation of the compaction trigger mechanism, which would only trigger compaction after specific events like flush, compaction, or SetOptions.
### Bug Fixes
* Fix a bug in BackupEngine that can crash backup due to a null FSWritableFile passed to WritableFileWriter.
* Fix DB::NewMultiScan iterator to respect the scan upper bound specified in ScanOptions
### Performance Improvements
* Optimized MultiScan using BlockBasedTable to coalesce I/Os and prefetch all data blocks.
## 10.4.0 (06/20/2025)
### New Features
* Add a new CF option `memtable_avg_op_scan_flush_trigger` that supports triggering memtable flush when an iterator scans through an expensive range of keys, with the average number of skipped keys from the active memtable exceeding the threshold.
* Vector based memtable now supports concurrent writers (DBOptions::allow_concurrent_memtable_write) #13675.
* Add new experimental `TransactionOptions::large_txn_commit_optimize_byte_threshold` to enable optimizations for large transaction commit by transaction batch data size.
* Add a new option `CompactionOptionsUniversal::reduce_file_locking` and if it's true, auto universal compaction picking will adjust to minimize locking of input files when bottom priority compactions are waiting to run. This can increase the likelihood of existing L0s being selected for compaction, thereby improving write stall and reducing read regression.
* Add new `format_version=7` to aid experimental support of custom compression algorithms with CompressionManager and block-based table. This format version includes changing the format of `TableProperties::compression_name`.
### Public API Changes
* Change NewExternalTableFactory to return a unique_ptr instead of shared_ptr.
* Add an optional min file size requirement for deletion triggered compaction. It can be specified when creating `CompactOnDeletionCollectorFactory`.
### Behavior Changes
* `TransactionOptions::large_txn_commit_optimize_threshold` now has default value 0 for disabled. `TransactionDBOptions::txn_commit_bypass_memtable_threshold` now has no effect on transactions.
### Bug Fixes
* Fix a bug where CreateColumnFamilyWithImport() could miss the SST file for the memtable flush it triggered. The exported CF then may not contain the updates in the memtable when CreateColumnFamilyWithImport() is called.
* Fix iterator operations returning NotImplemented status if disallow_memtable_writes and paranoid_memory_checks CF options are both set.
* Fixed handling of file checksums in IngestExternalFile() to allow providing checksums using recognized but not necessarily the DB's preferred checksum function, to ease migration between checksum functions.
## 10.3.0 (05/17/2025)
### New Features
* Add new experimental `CompactionOptionsFIFO::allow_trivial_copy_when_change_temperature` along with `CompactionOptionsFIFO::trivial_copy_buffer_size` to allow optimizing FIFO compactions with tiering when kChangeTemperature to move files from source tier FileSystem to another tier FileSystem via trivial and direct copying raw sst file instead of reading thru the content of the SST file then rebuilding the table files.
* Add a new field to Compaction Stats in LOG files for the pre-compression size written to each level.
* Add new experimental `TransactionOptions::large_txn_commit_optimize_threshold` to enable optimizations for large transaction commit with per transaction threshold. `TransactionDBOptions::txn_commit_bypass_memtable_threshold` is deprecated in favor of this transaction option.
* [internal team use only] Allow an application-defined `request_id` to be passed to RocksDB and propagated to the filesystem via IODebugContext
### Bug Fixes
* Fix a bug where transaction lock upgrade can incorrectly fail with a Deadlock status. This happens when a transaction has a non-zero timeout and tries to upgrade a shared lock that is also held by another transaction.
* Pass wrapped WritableFileWriter pointer to ExternalTableBuilder so that the file checksum can be correctly calculated and returned by SstFileWriter for external table files.
* Fix an infinite-loop bug in transaction locking. This can happen if a transaction reaches lock limit and its time out expires before it attempts to wait for it.
* Fixed a potential data race with `CompressionOptions::parallel_threads > 1` and a `TablePropertiesCollector` overriding `BlockAdd()`.
## 10.2.0 (04/21/2025)
### New Features
* Provide histogram stats `COMPACTION_PREFETCH_BYTES` to measure number of bytes for RocksDB's prefetching (as opposed to file
system's prefetch) on SST file during compaction read
* A new API DB::GetNewestUserDefinedTimestamp is added to return the newest user defined timestamp seen in a column family
* Introduce API `IngestWriteBatchWithIndex()` for ingesting updates into DB while bypassing memtable writes. This improves performance when writing a large write batch to the DB.
* Add a new CF option `memtable_op_scan_flush_trigger` that triggers a flush of the memtable if an iterator's Seek()/Next() scans over a certain number of invisible entries from the memtable.
### Public API Changes
* AdvancedColumnFamilyOptions.max_write_buffer_number_to_maintain is deleted. It's deprecated since introduction of a better option max_write_buffer_size_to_maintain since RocksDB 6.5.0.
* Added arbitrary string map for additional options to be overridden for remote compactions
* The fail_if_options_file_error option in DBOptions has been removed. The behavior now is to always return failure in any API that fails to persist the OPTIONS file.
### Behavior Changes
* Make stats `PREFETCH_BYTES_USEFUL`, `PREFETCH_HITS`, `PREFETCH_BYTES` only account for prefetching during user initiated scan
### Bug Fixes
* Fix a bug in Posix file system that the FSWritableFile created via `FileSystem::ReopenWritableFile` internally does not track the correct file size.
* Fix a bug where tail size of remote compaction output is not persisted in primary db's manifest
## 10.1.0 (03/24/2025)
### New Features
* Added a new `DBOptions.calculate_sst_write_lifetime_hint_set` setting that allows to customize which compaction styles SST write lifetime hint calculation is allowed on. Today RocksDB supports only two modes `kCompactionStyleLevel` and `kCompactionStyleUniversal`.
* Add a new field `num_l0_files` in `CompactionJobInfo` about the number of L0 files in the CF right before and after the compaction
* Added per-key-placement feature in Remote Compaction
* Implemented API DB::GetPropertiesOfTablesByLevel that retrieves table properties for files in each LSM tree level
### Public API Changes
* `GetAllKeyVersions()` now interprets empty slices literally, as valid keys, and uses new `OptSlice` type default value for extreme upper and lower range limits.
* `DeleteFilesInRanges()` now takes `RangeOpt` which is based on `OptSlice`. The overload taking `RangePtr` is deprecated.
* Add an unordered map of name/value pairs, ReadOptions::property_bag, to pass opaque options through to an external table when creating an Iterator.
* Introduced CompactionServiceJobStatus::kAborted to allow handling aborted scenario in Schedule(), Wait() or OnInstallation() APIs in Remote Compactions.
* format\_version < 2 in BlockBasedTableOptions is no longer supported for writing new files. Support for reading such files is deprecated and might be removed in the future. `CompressedSecondaryCacheOptions::compress_format_version == 1` is also deprecated.
### Behavior Changes
* `ldb` now returns an error if the specified `--compression_type` is not supported in the build.
* MultiGet with snapshot and ReadOptions::read_tier = kPersistedTier will now read a consistent view across CFs (instead of potentially reading some CF before and some CF after a flush).
* CreateColumnFamily() is no longer allowed on a read-only DB (OpenForReadOnly())
### Bug Fixes
* Fixed stats for Tiered Storage with preclude_last_level feature
## 10.0.0 (02/21/2025)
### New Features
* Introduced new `auto_refresh_iterator_with_snapshot` opt-in knob that (when enabled) will periodically release obsolete memory and storage resources for as long as the iterator is making progress and its supplied `read_options.snapshot` was initialized with non-nullptr value.
* Added the ability to plug-in a custom table reader implementation. See include/rocksdb/external_table_reader.h for more details.
* Experimental feature: RocksDB now supports FAISS inverted file based indices via the secondary indexing framework. Applications can use FAISS secondary indices to automatically quantize embeddings and perform K-nearest-neighbors similarity searches. See `FaissIVFIndex` and `SecondaryIndex` for more details. Note: the FAISS integration currently requires using the BUCK build.
* Add new DB property `num_running_compaction_sorted_runs` that tracks the number of sorted runs being processed by currently running compactions
* Experimental feature: added support for simple secondary indices that index the specified column as-is. See `SimpleSecondaryIndex` and `SecondaryIndex` for more details.
* Added new `TransactionDBOptions::txn_commit_bypass_memtable_threshold`, which enables optimized transaction commit (see `TransactionOptions::commit_bypass_memtable`) when the transaction size exceeds a configured threshold.
### Public API Changes
* Updated the query API of the experimental secondary indexing feature by removing the earlier `SecondaryIndex::NewIterator` virtual and adding a `SecondaryIndexIterator` class that can be utilized by applications to find the primary keys for a given search target.
* Added back the ability to leverage the primary key when building secondary index entries. This involved changes to the signatures of `SecondaryIndex::GetSecondary{KeyPrefix,Value}` as well as the addition of a new method `SecondaryIndex::FinalizeSecondaryKeyPrefix`. See the API comments for more details.
* Minimum supported version of ZSTD is now 1.4.0, for code simplification. Obsolete `CompressionType``kZSTDNotFinalCompression` is also removed.
### Behavior Changes
* `VerifyBackup` in `verify_with_checksum`=`true` mode will now evaluate checksums in parallel. As a result, unlike in case of original implementation, the API won't bail out on a very first corruption / mismatch and instead will iterate over all the backup files logging success / _degree_of_failure_ for each.
* Reversed the order of updates to the same key in WriteBatchWithIndex. This means if there are multiple updates to the same key, the most recent update is ordered first. This affects the output of WBWIIterator. When WriteBatchWithIndex is created with `overwrite_key=true`, this affects the output only if Merge is used (#13387).
* Added support for Merge operations in transactions using option `TransactionOptions::commit_bypass_memtable`.
### Bug Fixes
* Fixed GetMergeOperands() API in ReadOnlyDB and SecondaryDB
* Fix a bug in `GetMergeOperands()` that can return incorrect status (MergeInProgress) and incorrect number of merge operands. This can happen when `GetMergeOperandsOptions::continue_cb` is set, both active and immutable memtables have merge operands and the callback stops the look up at the immutable memtable.
## 9.11.0 (01/17/2025)
### New Features
* Introduce CancelAwaitingJobs() in CompactionService interface which will allow users to implement cancellation of running remote compactions from the primary instance
* Experimental feature: RocksDB now supports defining secondary indices, which are automatically maintained by the storage engine. Secondary indices provide a new customization point: applications can provide their own by implementing the new `SecondaryIndex` interface. See the `SecondaryIndex` API comments for more details. Note: this feature is currently only available in conjunction with write-committed pessimistic transactions, and `Merge` is not yet supported.
* Provide a new option `track_and_verify_wals` to track and verify various information about WAL during WAL recovery. This is intended to be a better replacement to `track_and_verify_wals_in_manifest`.
### Public API Changes
* Add `io_buffer_size` to BackupEngineOptions to enable optimal configuration of IO size
* Clean up all the references to `random_access_max_buffer_size`, related rules and all the clients wrappers. This option has been officially deprecated in 5.4.0.
* Add `file_ingestion_nanos` and `file_ingestion_blocking_live_writes_nanos` in PerfContext to observe file ingestions
* Offer new DB::Open and variants that use `std::unique_ptr<DB>*` output parameters and deprecate the old versions that use `DB**` output parameters.
* The DB::DeleteFile API is officially deprecated.
### Behavior Changes
* For leveled compaction, manual compaction (CompactRange()) will be more strict about keeping compaction size under `max_compaction_bytes`. This prevents overly large compactions in some cases (#13306).
* Experimental tiering options `preclude_last_level_data_seconds` and `preserve_internal_time_seconds` are now mutable with `SetOptions()`. Some changes to handling of these features along with long-lived snapshots and range deletes made this possible.
### Bug Fixes
* Fix a longstanding major bug with SetOptions() in which setting changes can be quietly reverted.
## 9.10.0 (12/12/2024)
### New Features
* Introduce `TransactionOptions::commit_bypass_memtable` to enable transaction commit to bypass memtable insertions. This can be beneficial for transactions with many operations, as it reduces commit time that is mostly spent on memtable insertion.
### Public API Changes
* Deprecated Remote Compaction APIs (StartV2, WaitForCompleteV2) are completely removed from the codebase
### Behavior Changes
* DB::KeyMayExist() now follows its function comment, which means `value` parameter can be null, and it will be set only if `value_found` is passed in.
### Bug Fixes
* Fix the issue where compaction incorrectly drops a key when there is a snapshot with a sequence number of zero.
* Honor ConfigOptions.ignore_unknown_options in ParseStruct()
### Performance Improvements
* Enable reuse of file system allocated buffer for synchronous prefetching.
* In buffered IO mode, try to align writes on power of 2 if checksum handoff is not enabled for the file type being written.
## 9.9.0 (11/18/2024)
### New Features
* Multi-Column-Family-Iterator (CoalescingIterator/AttributeGroupIterator) is no longer marked as experimental
* Adds a new table property "rocksdb.newest.key.time" which records the unix timestamp of the newest key. Uses this table property for FIFO TTL and temperature change compaction.
### Public API Changes
* Added a new API `Transaction::GetAttributeGroupIterator` that can be used to create a multi-column-family attribute group iterator over the specified column families, including the data from both the transaction and the underlying database. This API is currently supported for optimistic and write-committed pessimistic transactions.
* Added a new API `Transaction::GetCoalescingIterator` that can be used to create a multi-column-family coalescing iterator over the specified column families, including the data from both the transaction and the underlying database. This API is currently supported for optimistic and write-committed pessimistic transactions.
### Behavior Changes
* `BaseDeltaIterator` now honors the read option `allow_unprepared_value`.
### Bug Fixes
* `BaseDeltaIterator` now calls `PrepareValue` on the base iterator in case it has been created with the `allow_unprepared_value` read option set. Earlier, such base iterators could lead to incorrect values being exposed from `BaseDeltaIterator`.
* Fix a leak of obsolete blob files left open until DB::Close(). This bug was introduced in version 9.4.0.
* Fix missing cases of corruption retry during DB open and read API processing.
* Fix a bug for transaction db with 2pc where an old WAL may be retained longer than needed (#13127).
* Fix leaks of some open SST files (until `DB::Close()`) that are written but never become live due to various failures. (We now have a check for such leaks with no outstanding issues.)
* Fix a bug for replaying WALs for WriteCommitted transaction DB when its user-defined timestamps setting is toggled on/off between DB sessions.
### Performance Improvements
* Fix regression in issue #12038 due to `Options::compaction_readahead_size` greater than `max_sectors_kb` (i.e, largest I/O size that the OS issues to a block device defined in linux)
## 9.8.0 (10/25/2024)
### New Features
* All non-`block_cache` options in `BlockBasedTableOptions` are now mutable with `DB::SetOptions()`. See also Bug Fixes below.
* When using iterators with BlobDB, it is now possible to load large values on an on-demand basis, i.e. only if they are actually needed by the application. This can save I/O in use cases where the values associated with certain keys are not needed. For more details, see the new read option `allow_unprepared_value` and the iterator API `PrepareValue`.
* Add a new file ingestion option `IngestExternalFileOptions::fill_cache` to support not adding blocks from ingested files into block cache during file ingestion.
* The option `allow_unprepared_value` is now also supported for multi-column-family iterators (i.e. `CoalescingIterator` and `AttributeGroupIterator`).
* When a file with just one range deletion (standalone range deletion file) is ingested via bulk loading, it will be marked for compaction. During compaction, this type of files can be used to directly filter out some input files that are not protected by any snapshots and completely deleted by the standalone range deletion file.
### Behavior Changes
* During file ingestion, overlapping files level assignment are done in multiple batches, so that they can potentially be assigned to lower levels other than always land on L0.
* OPTIONS file to be loaded by remote worker is now preserved so that it does not get purged by the primary host. A similar technique as how we are preserving new SST files from getting purged is used for this. min_options_file_numbers_ is tracked like pending_outputs_ is tracked.
* Trim readahead_size during scans so data blocks containing keys that are not in the same prefix as the seek key in `Seek()` are not prefetched when `ReadOptions::auto_readahead_size=true` (default value) and `ReadOptions::prefix_same_as_start = true`
* Assigning levels for external files are done in the same way for universal compaction and leveled compaction. The old behavior tends to assign files to L0 while the new behavior will assign the files to the lowest level possible.
### Bug Fixes
* Fix a longstanding race condition in SetOptions for `block_based_table_factory` options. The fix has some subtle behavior changes because of copying and replacing the TableFactory on a change with SetOptions, including requiring an Iterator::Refresh() for an existing Iterator to use the latest options.
* Fix under counting of allocated memory in the compressed secondary cache due to looking at the compressed block size rather than the actual memory allocated, which could be larger due to internal fragmentation.
* `GetApproximateMemTableStats()` could return disastrously bad estimates 5-25% of the time. The function has been re-engineered to return much better estimates with similar CPU cost.
* Skip insertion of compressed blocks in the secondary cache if the lowest_used_cache_tier DB option is kVolatileTier.
* Fix an issue in level compaction where a small CF with small compaction debt can cause the DB to allow parallel compactions. (#13054)
* Several DB option settings could be lost through `GetOptionsFromString()`, possibly elsewhere as well. Affected options, now fixed:`background_close_inactive_wals`, `write_dbid_to_manifest`, `write_identity_file`, `prefix_seek_opt_in_only`
## 9.7.0 (09/20/2024)
### New Features
* Make Cache a customizable class that can be instantiated by the object registry.
* Add new option `prefix_seek_opt_in_only` that makes iterators generally safer when you might set a `prefix_extractor`. When `prefix_seek_opt_in_only=true`, which is expected to be the future default, prefix seek is only used when `prefix_same_as_start` or `auto_prefix_mode` are set. Also, `prefix_same_as_start` and `auto_prefix_mode` now allow prefix filtering even with `total_order_seek=true`.
* Add a new table property "rocksdb.key.largest.seqno" which records the largest sequence number of all keys in file. It is verified to be zero during SST file ingestion.
### Behavior Changes
* Changed the semantics of the BlobDB configuration option `blob_garbage_collection_force_threshold` to define a threshold for the overall garbage ratio of all blob files currently eligible for garbage collection (according to `blob_garbage_collection_age_cutoff`). This can provide better control over space amplification at the cost of slightly higher write amplification.
* Set `write_dbid_to_manifest=true` by default. This means DB ID will now be preserved through backups, checkpoints, etc. by default. Also add `write_identity_file` option which can be set to false for anticipated future behavior.
* In FIFO compaction, compactions for changing file temperature (configured by option `file_temperature_age_thresholds`) will compact one file at a time, instead of merging multiple eligible file together (#13018).
* Support ingesting db generated files using hard link, i.e. IngestExternalFileOptions::move_files/link_files and IngestExternalFileOptions::allow_db_generated_files.
* Add a new file ingestion option `IngestExternalFileOptions::link_files` to hard link input files and preserve original files links after ingestion.
* DB::Close now untracks files in SstFileManager, making available any space used
by them. Prior to this change they would be orphaned until the DB is re-opened.
### Bug Fixes
* Fix a bug in CompactRange() where result files may not be compacted in any future compaction. This can only happen when users configure CompactRangeOptions::change_level to true and the change level step of manual compaction fails (#13009).
* Fix handling of dynamic change of `prefix_extractor` with memtable prefix filter. Previously, prefix seek could mix different prefix interpretations between memtable and SST files. Now the latest `prefix_extractor` at the time of iterator creation or refresh is respected.
* Fix a bug with manual_wal_flush and auto error recovery from WAL failure that may cause CFs to be inconsistent (#12995). The fix will set potential WAL write failure as fatal error when manual_wal_flush is true, and disables auto error recovery from these errors.
## 9.6.0 (08/19/2024)
### New Features
* Best efforts recovery supports recovering to incomplete Version with a clean seqno cut that presents a valid point in time view from the user's perspective, if versioning history doesn't include atomic flush.
* New option `BlockBasedTableOptions::decouple_partitioned_filters` should improve efficiency in serving read queries because filter and index partitions can consistently target the configured `metadata_block_size`. This option is currently opt-in.
* Introduce a new mutable CF option `paranoid_memory_checks`. It enables additional validation on data integrity during reads/scanning. Currently, skip list based memtable will validate key ordering during look up and scans.
### Public API Changes
* Add ticker stats to count file read retries due to checksum mismatch
* Adds optional installation callback function for remote compaction
### Behavior Changes
* There may be less intra-L0 compaction triggered by total L0 size being too small. We now use compensated file size (tombstones are assigned some value size) when calculating L0 size and reduce the threshold for L0 size limit. This is to avoid accumulating too much data/tombstones in L0.
### Bug Fixes
* Make DestroyDB supports slow deletion when it's configured in `SstFileManager`. The slow deletion is subject to the configured `rate_bytes_per_sec`, but not subject to the `max_trash_db_ratio`.
* Fixed a bug where we set unprep_seqs_ even when WriteImpl() fails. This was caught by stress test write fault injection in WriteImpl(). This may have incorrectly caused iteration creation failure for unvalidated writes or returned wrong result for WriteUnpreparedTxn::GetUnpreparedSequenceNumbers().
* Fixed a bug where successful write right after error recovery for last failed write finishes causes duplicate WAL entries
* Fixed a data race involving the background error status in `unordered_write` mode.
* Fix a bug where file snapshot functions like backup, checkpoint may attempt to copy a non-existing manifest file. #12882
* Fix a bug where per kv checksum corruption may be ignored in MultiGet().
* Fix a race condition in pessimistic transactions that could allow multiple transactions with the same name to be registered simultaneously, resulting in a crash or other unpredictable behavior.
## 9.5.0 (07/19/2024)
### Public API Changes
* Introduced new C API function rocksdb_writebatch_iterate_cf for column family-aware iteration over the contents of a WriteBatch
* Add support to ingest SST files generated by a DB instead of SstFileWriter. This can be enabled with experimental option `IngestExternalFileOptions::allow_db_generated_files`.
### Behavior Changes
* When calculating total log size for the `log_size_for_flush` argument in `CreateCheckpoint` API, the size of the archived log will not be included to avoid unnecessary flush
### Bug Fixes
* Fix a major bug in which an iterator using prefix filtering and SeekForPrev might miss data when the DB is using `whole_key_filtering=false` and `partition_filters=true`.
* Fixed a bug where `OnErrorRecoveryBegin()` is not called before auto recovery starts.
* Fixed a bug where event listener reads ErrorHandler's `bg_error_` member without holding db mutex(#12803).
* Fixed a bug in handling MANIFEST write error that caused the latest valid MANIFEST file to get deleted, resulting in the DB being unopenable.
* Fixed a race between error recovery due to manifest sync or write failure and external SST file ingestion. Both attempt to write a new manifest file, which causes an assertion failure.
### Performance Improvements
* Fix an issue where compactions were opening table files and reading table properties while holding db mutex_.
* Reduce unnecessary filesystem queries and DB mutex acquires in creating backups and checkpoints.
## 9.4.0 (06/23/2024)
### New Features
* Added a `CompactForTieringCollectorFactory` to auto trigger compaction for tiering use case.
@@ -57,7 +585,7 @@
* Added a new API `GetEntityFromBatchAndDB` to `WriteBatchWithIndex` that can be used for wide-column point lookups with read-your-own-writes consistency. Similarly to `GetFromBatchAndDB`, the API can combine data from the write batch with data from the underlying database if needed. See the API comments for more details.
* [Experimental] Introduce two new cross-column-family iterators - CoalescingIterator and AttributeGroupIterator. The CoalescingIterator enables users to iterate over multiple column families and access their values and columns. During this iteration, if the same key exists in more than one column family, the keys in the later column family will overshadow the previous ones. The AttributeGroupIterator allows users to gather wide columns per Column Family and create attribute groups while iterating over keys across all CFs.
* Added a new API `MultiGetEntityFromBatchAndDB` to `WriteBatchWithIndex` that can be used for batched wide-column point lookups with read-your-own-writes consistency. Similarly to `MultiGetFromBatchAndDB`, the API can combine data from the write batch with data from the underlying database if needed. See the API comments for more details.
* *Adds a `SstFileReader::NewTableIterator` API to support programmatically read a SST file as a raw table file.
* Adds a `SstFileReader::NewTableIterator` API to support programmatically read a SST file as a raw table file.
* Add an option to `WaitForCompactOptions` - `wait_for_purge` to make `WaitForCompact()` API wait for background purge to complete
### Public API Changes
@@ -86,24 +614,24 @@ the whole DB to be dropped right after migration if the migrated data is larger
* Fixed a bug where wrong padded bytes are used to generate file checksum and `DataVerificationInfo::checksum` upon file creation
* Correctly implemented the move semantics of `PinnableWideColumns`.
* Fixed a bug when the recycle_log_file_num in DBOptions is changed from 0 to non-zero when a DB is reopened. On a subsequent reopen, if a log file created when recycle_log_file_num==0 was reused previously, is alive and is empty, we could end up inserting stale WAL records into the memtable.
* *Fix a bug where obsolete files' deletion during DB::Open are not rate limited with `SstFilemManager`'s slow deletion feature even if it's configured.
* Fix a bug where obsolete files' deletion during DB::Open are not rate limited with `SstFilemManager`'s slow deletion feature even if it's configured.
## 9.1.0 (03/22/2024)
### New Features
* Added an option, `GetMergeOperandsOptions::continue_cb`, to give users the ability to end `GetMergeOperands()`'s lookup process before all merge operands were found.
* *Add sanity checks for ingesting external files that currently checks if the user key comparator used to create the file is compatible with the column family's user key comparator.
* Add sanity checks for ingesting external files that currently checks if the user key comparator used to create the file is compatible with the column family's user key comparator.
*Support ingesting external files for column family that has user-defined timestamps in memtable only enabled.
* On file systems that support storage level data checksum and reconstruction, retry SST block reads for point lookups, scans, and flush and compaction if there's a checksum mismatch on the initial read.
* Some enhancements and fixes to experimental Temperature handling features, including new `default_write_temperature` CF option and opening an `SstFileWriter` with a temperature.
* `WriteBatchWithIndex` now supports wide-column point lookups via the `GetEntityFromBatch` API. See the API comments for more details.
* *Implement experimental features: API `Iterator::GetProperty("rocksdb.iterator.write-time")` to allow users to get data's approximate write unix time and write data with a specific write time via `WriteBatch::TimedPut` API.
* Implement experimental features: API `Iterator::GetProperty("rocksdb.iterator.write-time")` to allow users to get data's approximate write unix time and write data with a specific write time via `WriteBatch::TimedPut` API.
### Public API Changes
* Best-effort recovery (`best_efforts_recovery == true`) may now be used together with atomic flush (`atomic_flush == true`). The all-or-nothing recovery guarantee for atomically flushed data will be upheld.
* Remove deprecated option `bottommost_temperature`, already replaced by `last_level_temperature`
* Added new PerfContext counters for block cache bytes read - block_cache_index_read_byte, block_cache_filter_read_byte, block_cache_compression_dict_read_byte, and block_cache_read_byte.
* Deprecate experimental Remote Compaction APIs - StartV2() and WaitForCompleteV2() and introduce Schedule() and Wait(). The new APIs essentially does the same thing as the old APIs. They allow taking externally generated unique id to wait for remote compaction to complete.
* *For API `WriteCommittedTransaction::GetForUpdate`, if the column family enables user-defined timestamp, it was mandated that argument `do_validate` cannot be false, and UDT based validation has to be done with a user set read timestamp. It's updated to make the UDT based validation optional if user sets `do_validate` to false and does not set a read timestamp. With this, `GetForUpdate` skips UDT based validation and it's users' responsibility to enforce the UDT invariant. SO DO NOT skip this UDT-based validation if users do not have ways to enforce the UDT invariant. Ways to enforce the invariant on the users side include manage a monotonically increasing timestamp, commit transactions in a single thread etc.
* For API `WriteCommittedTransaction::GetForUpdate`, if the column family enables user-defined timestamp, it was mandated that argument `do_validate` cannot be false, and UDT based validation has to be done with a user set read timestamp. It's updated to make the UDT based validation optional if user sets `do_validate` to false and does not set a read timestamp. With this, `GetForUpdate` skips UDT based validation and it's users' responsibility to enforce the UDT invariant. SO DO NOT skip this UDT-based validation if users do not have ways to enforce the UDT invariant. Ways to enforce the invariant on the users side include manage a monotonically increasing timestamp, commit transactions in a single thread etc.
* Defined a new PerfLevel `kEnableWait` to measure time spent by user threads blocked in RocksDB other than mutex, such as a write thread waiting to be added to a write group, a write thread delayed or stalled etc.
* `RateLimiter`'s API no longer requires the burst size to be the refill size. Users of `NewGenericRateLimiter()` can now provide burst size in `single_burst_bytes`. Implementors of `RateLimiter::SetSingleBurstBytes()` need to adapt their implementations to match the changed API doc.
* Add `write_memtable_time` to the newly introduced PerfLevel `kEnableWait`.
* Provide support for FSBuffer for point lookups. Also added support for scans and compactions that don't go through prefetching.
* *Make `SstFileWriter` create SST files without persisting user defined timestamps when the `Option.persist_user_defined_timestamps` flag is set to false.
* Make `SstFileWriter` create SST files without persisting user defined timestamps when the `Option.persist_user_defined_timestamps` flag is set to false.
* Add support for user-defined timestamps in APIs `DeleteFilesInRanges` and `GetPropertiesOfTablesInRange`.
* Mark wal\_compression feature as production-ready. Currently only compatible with ZSTD compression.
* *Remove the default `WritableFile::GetFileSize` and `FSWritableFile::GetFileSize` implementation that returns 0 and make it pure virtual, so that subclasses are enforced to explicitly provide an implementation.
* Remove the default `WritableFile::GetFileSize` and `FSWritableFile::GetFileSize` implementation that returns 0 and make it pure virtual, so that subclasses are enforced to explicitly provide an implementation.
* `sst_dump --command=check` now compares the number of records in a table with `num_entries` in table property, and reports corruption if there is a mismatch. API `SstFileDumper::ReadSequential()` is updated to optionally do this verification. (#12322)
* `rocksdb.blobdb.blob.file.write.micros` expands to also measure time writing the header and footer. Therefore the COUNT may be higher and values may be smaller than before. For stacked BlobDB, it no longer measures the time of explictly flushing blob file.
* `rocksdb.blobdb.blob.file.write.micros` expands to also measure time writing the header and footer. Therefore the COUNT may be higher and values may be smaller than before. For stacked BlobDB, it no longer measures the time of explicitly flushing blob file.
* Files will be compacted to the next level if the data age exceeds periodic_compaction_seconds except for the last level.
* Reduced the compaction debt ratio trigger for scheduling parallel compactions
* For leveled compaction with default compaction pri (kMinOverlappingRatio), files marked for compaction will be prioritized over files not marked when picking a file from a level for compaction.
@@ -240,7 +768,7 @@ want to continue to use force enabling, they need to explicitly pass a `true` to
### Behavior Changes
* During off-peak hours defined by `daily_offpeak_time_utc`, the compaction picker will select a larger number of files for periodic compaction. This selection will include files that are projected to expire by the next off-peak start time, ensuring that these files are not chosen for periodic compaction outside of off-peak hours.
* If an error occurs when writing to a trace file after `DB::StartTrace()`, the subsequent trace writes are skipped to avoid writing to a file that has previously seen error. In this case, `DB::EndTrace()` will also return a non-ok status with info about the error occured previously in its status message.
* If an error occurs when writing to a trace file after `DB::StartTrace()`, the subsequent trace writes are skipped to avoid writing to a file that has previously seen error. In this case, `DB::EndTrace()` will also return a non-ok status with info about the error occurred previously in its status message.
* Deleting stale files upon recovery are delegated to SstFileManger if available so they can be rate limited.
* Make RocksDB only call `TablePropertiesCollector::Finish()` once.
* When `WAL_ttl_seconds > 0`, we now process archived WALs for deletion at least every `WAL_ttl_seconds / 2` seconds. Previously it could be less frequent in case of small `WAL_ttl_seconds` values when size-based expiration (`WAL_size_limit_MB > 0 `) was simultaneously enabled.
@@ -1028,7 +1556,7 @@ Note: The next release will be major release 7.0. See https://github.com/faceboo
### Public API change
* Extend WriteBatch::AssignTimestamp and AssignTimestamps API so that both functions can accept an optional `checker` argument that performs additional checking on timestamp sizes.
* Introduce a new EventListener callback that will be called upon the end of automatic error recovery.
* Add IncreaseFullHistoryTsLow API so users can advance each column family's full_history_ts_low seperately.
* Add IncreaseFullHistoryTsLow API so users can advance each column family's full_history_ts_low separately.
* Add GetFullHistoryTsLow API so users can query current full_history_low value of specified column family.
The Bing search engine from Microsoft uses RocksDB as the storage engine for its web data platform: https://blogs.bing.com/Engineering-Blog/october-2021/RocksDB-in-Microsoft-Bing
## LinkedIn
Two different use cases at Linkedin are using RocksDB as a storage engine:
1. [Venice](https://venicedb.org/) is a derived data platform using RocksDB as its storage engine. It is LinkedIn's ML feature store, powering thousands of recommender use cases, including the Feed, Video recommendations, and People You May Know.
2. LinkedIn's follow feed for storing user's activities. Check out the blog post: https://engineering.linkedin.com/blog/2016/03/followfeed--linkedin-s-feed-made-faster-and-smarter
3. Apache Samza, open source framework for stream processing.
1. LinkedIn's follow feed for storing user's activities. Check out the blog post: https://engineering.linkedin.com/blog/2016/03/followfeed--linkedin-s-feed-made-faster-and-smarter
2. Apache Samza, open source framework for stream processing
Learn more about those use cases in a Tech Talk by Ankit Gupta and Naveen Somasundaram: http://www.youtube.com/watch?v=plqVp_OnSzg
Learn more about LinkedIn's follow feed and Apache Samza in a Tech Talk by Ankit Gupta and Naveen Somasundaram: http://www.youtube.com/watch?v=plqVp_OnSzg
## Yahoo
Yahoo is using RocksDB as a storage engine for their biggest distributed data store Sherpa. Learn more about it here: http://yahooeng.tumblr.com/post/120730204806/sherpa-scales-new-heights
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Peter Dillinger