Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15438
Fix a RocksDB secondary with `open_files_async = true` latching on `PathNotFound` during `TryCatchUpWithPrimary()`.
**Why:** With `open_files_async = true` and `max_open_files = -1`, a secondary's catch-up could fail with `PathNotFound` on a table that was live at open time (most reachably in a non-default column family) but which the primary compacted away before the background async opener pinned it. The error latched in the tailer's status, so the secondary never advanced again and had to be closed and reopened. With `open_files_async = false` the same sequence caught up fine. This blocks enabling async secondary open where it would otherwise cut cold-open time substantially.
**Previous behavior:** In `VersionEditHandlerPointInTime::MaybeCreateVersionBeforeApplyEdit`, a negative-edge `LoadTableHandlers()` failure skipped the unrecoverable point-in-time only when the status was `Corruption`; a `PathNotFound` was returned as-is and latched. During a secondary's catch-up, `open_files_async` leaves recovery-time tables unpinned until the background opener reaches them, so `LoadTableHandlers()` could try to open a file the primary had already deleted.
**New behavior:** That negative-edge failure path now also treats `PathNotFound` as a missing file (skip this point-in-time and continue) when `no_error_if_files_missing_` is set, mirroring the tolerance `VersionEditHandler::LoadTables()` already applies. Catch-up continues past the transient state and installs the current version, so the secondary advances instead of latching.
**Implementation:**
- `db/version_edit_handler.cc`: broaden the negative-edge skip condition in `VersionEditHandlerPointInTime::MaybeCreateVersionBeforeApplyEdit` from `s.IsCorruption()` to also include `s.IsPathNotFound() && no_error_if_files_missing_`.
- `db/db_secondary_test.cc`: add `DBSecondaryTest.OpenFilesAsyncNonDefaultCfCatchUp`, a regression test that opens an async secondary with a non-default column family, holds the background opener back so the open-time table stays unpinned, has the primary flush and `CompactRange(kForce)` to replace and delete that table, then asserts `TryCatchUpWithPrimary()` succeeds and reads the new value.
**Safety boundaries:** The change only affects point-in-time handlers that opted into missing-file tolerance (`no_error_if_files_missing_`, set by secondary/follower tailing and best-efforts recovery); it does not relax error handling for strict recovery. It does not change the `trust_manifest_recovery_` / `defer_sst_file_opening_` paths, which already skip opening.
**Validation and limitations:** The regression test covers async secondary catch-up after deletion of an open-time SST in a non-default column family. Validation covers async opening, catch-up, memtable retention, and repeated scheduling changes. Strict recovery, API and file formats are unchanged. The full supported platform matrix has not been run locally.
Reviewed By: jaykorean
Differential Revision: D125378257
fbshipit-source-id: 3dd23c2f763e4e744c9f38fb77dfdfb3f554ee7f
(cherry picked from commit 81369fc170)
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15481
With blob direct write, each memtable switch moves the column family's open blob files into a new generation, queued in `BlobFilePartitionManager::pending_generations_`. A flush then seals and publishes the generations of the memtables it flushes.
`PrepareFlushAdditions(n)` always used the first `n` queued generations, and `CommitPreparedGenerations(n)` always popped the first `n`. That is only right if flushes pick, and finish, the oldest memtables in order. With more than one flush running for a column family (`max_background_flushes` > 1, or `max_background_jobs` >= 8), a second flush skips the memtable the first one is still flushing and picks a newer one, but prepares the oldest generation, which belongs to the first flush. Both flush edits then add the same blob file, and installing them fails with `Corruption: VersionBuilder: Blob file #N already added`, which stops the database. The newer memtable's own blob files are never published, and committing the second flush pops a generation it did not own. Atomic flush prepares before it waits to install in order, so it has the same problem.
Removing generations by the flush's own memtables is not enough either. A newer flush that finishes first only marks its memtable completed: the older flush installs both, or rolls the newer one back too if it fails. A generation removed by the newer flush is then gone when the retried flush looks for it (`Corruption: Missing blob direct write generation metadata for flush`).
Fix:
- Tag each generation with the ID of the memtable whose switch created it (`RotateCurrentGeneration(memtable_id)`). `PrepareFlushAdditions()` takes the IDs of the memtables the flush picked and seals exactly those generations, in both flush paths.
- Remove generations only once installed: after a successful flush, `CommitFlushedGenerations()` drops the generations of the memtables older than the earliest one still in the immutable list (memtables leave it only by installation, in order), and stops tracking their sealed files as manager-owned.
Reviewed By: xingbowang, archang19
Differential Revision: D126852567
fbshipit-source-id: 910bf466674cf8bc89bc3ccbbcf08c684968ad72
(cherry picked from commit 6d03b3df45)
Backport D126796014 (PR #15451) and D126875720 (PR #15461) onto
11.13.fb to prevent permanent data loss when read-path range tombstone
conversion scans a partial tier or an iterator view predating ingestion.
Retain both fixes' regression tests and release notes.
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15396
## Summary
`ParseEntry()` exposes the components of an internal key, but constructing one with a chosen sequence number and entry type currently requires private RocksDB utilities or duplicating the encoding. Add public `EncodeEntry()` to construct a stored internal key from `ParsedEntryInfo`, including its user key, timestamp, sequence number, and entry type. Also expose `PackSequenceAndType()` and `UnPackSequenceAndType()` to work with the sequence/type footer as a `uint64_t`.
The helpers map all nine representable public entry types to their internal types and use the existing footer encoding; `EncodeEntry()` reuses the public packer. Packing `kEntryOther` or a sequence number above the 56-bit limit returns `InvalidArgument` without changing the output. Unpacking reports internal value types without a public equivalent as `kEntryOther`. Timestamp size remains the caller's responsibility, and input slices must not alias the output string. These are additive C++ utilities using the existing internal-key format.
Reviewed By: anand1976, xingbowang
Differential Revision: D125082298
fbshipit-source-id: 6c09a31695f1d377b9cc208fa5b6edd3ad6638e2
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15428
RocksDB counts per-operation statistics (blocks read, bytes read, mutex waits,
timings) in `perf_context`. It is a `thread_local` variable: each thread has its
own copy, so threads never contend on it. This per-thread copy is "thread-local
storage" (TLS).
**Problem**: `PerfContext` has a non-trivial destructor. For such a variable, C++
must make sure the calling thread's copy is constructed before each use from
another source file, so the compiler emits a call to a small "TLS init"
function. That call goes through the PLT (procedure linkage table, the jump
table used to call functions that may live in another shared library). Every
`PERF_COUNTER_ADD` therefore paid a real function call before incrementing a
counter, and these macros run in hot loops such as key comparisons during block
iteration. `perf_level`, the on/off switch, has a smaller cost: it is
`extern thread_local` without `constinit`, so each read also tests whether it
has been initialized.
**Fix**:
- The perf macros read a `constinit thread_local PerfContext*` that each
thread sets on first use. After that, a counter update is a pointer load,
a null check and the increment.
- `perf_level` is declared and defined `constinit`, so reading it is a plain
thread-local load.
- No public API change. `perf_context`, `get_perf_context()` and macro
behavior are unchanged, and so is thread-exit behavior.
**Results**:
- Microbenchmark (`PERF_COUNTER_ADD` in a loop): 2.10 -> 0.84 ns per update
with counting enabled; unchanged at 0.42 ns with counting disabled.
- Calls to the TLS init functions drop from 40 and 96 to 0 in block.cc.o
and memtable.cc.o.
- In CPU profiles of large production RocksDB deployments, these accessors
cost roughly 0.3-0.6% of total CPU. This change removes most of the
`perf_context` part.
Reviewed By: xingbowang
Differential Revision: D125982209
fbshipit-source-id: f38b4fffda09611d28322e7534b135fb35631389
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15444
**Why the change is required**
`max_bgerror_resume_count <= 0` turns off automatic `DB::Resume()`, so the application is responsible for recovering from background errors. Before automatic recovery was introduced, a background error under `paranoid_checks` latched `bg_error_` and made the DB read-only. Since then, two classes of retryable IO error are either never latched or latched as soft errors that leave the DB accepting writes. When auto resume is disabled, no one recovers from them and the application gets no clear signal.
**Previous behavior** (with `max_bgerror_resume_count <= 0`)
- A retryable or file-scoped IO error during compaction was never set in `bg_error_`. Compaction rescheduled itself and retried indefinitely. A persistent failure was visible only through `EventListener::OnBackgroundError` and the LOG, while compaction debt grew until writes slowed or stopped on L0/pending-compaction triggers.
- A retryable or file-scoped IO error while writing a flush SST or the MANIFEST after a successful WAL sync (`kFlushNoWAL` / `kManifestWriteNoWAL`, which covers almost every retryable flush failure, not only `disableWAL` writes) was a soft error with `soft_error_no_bg_work_` set. Background work stopped, but writes kept succeeding into the memtable. Nothing retried the flush, so writers eventually blocked indefinitely on the memtable limit instead of failing.
**New behavior**
With `max_bgerror_resume_count <= 0`, every retryable or file-scoped background IO error that reaches the retryable path in `ErrorHandler::SetBGErrorImpl` is latched as `kHardError`, whatever the reason (compaction, flush, MANIFEST, WAL). Writes fail immediately and reads continue to work. `DB::Resume()` clears the error, flushes and reschedules compaction as before.
**Implementation**
In `SetBGErrorImpl`, the compaction special case (notify the listener and return without setting `bg_error_`) now applies only when auto resume is enabled, and the severity selection maps everything to `kHardError` when auto resume is disabled. Because `kErrorRecoveryRetryFlush` is no longer selected in that mode, `soft_error_no_bg_work_` is not set. The option documentation in `include/rocksdb/options.h` and the comment above `SetBGError` are updated.
**Safety, compatibility, and limitations**
- This is an intentional behavior change, limited to DBs that set `max_bgerror_resume_count <= 0`. With the default (`INT_MAX`) or any positive value, nothing changes.
- fbcode callers that set the option to 0, and their apparent expectations:
- `unicorn/live_updates/dispatcher/MessageStore.cpp`: the comment says "we do not allow any [background errors] as they should be retried by the underlying FS". Stopping on retryable compaction errors fits that intent better than silent retries.
- `reliable_volumes/client/index_manager/RocksDBIndexStorage.cpp`: a JustKnob (`disableRocksDbBackgroundErrorAutoResume`) disables auto resume alongside other corruption-detection options, which suggests it wants background errors surfaced rather than masked.
- `zippydb/server/store` (`RocksDBStoreResumeWritesTest`): uses 0 with `disableWAL` and drives recovery through its own `ResumeWrites` task calling `DB::Resume()`, so it already expects writes to stop until resume. I read the test and expect it to pass, but did not run it. No production zippydb configuration setting 0 was found.
- Out-of-space errors (`kNoSpace`/`kSpaceLimit`) are untouched: they skip this path and are still recovered through `SstFileManager`, which does not consult `max_bgerror_resume_count`. Errors ignored under `paranoid_checks=false` are also unchanged.
- No API, ABI or on-disk format change.
- Found during testing, pre-existing and not fixed here: `SetBGErrorImpl` releases the DB mutex in `NotifyOnBackgroundError` before calling `StartRecoverFromRetryableBGIOError`. If the last failed auto-resume attempt's flush error is reported in that window, the recovery thread can give up and clear `recovery_in_prog_` first, and the late report then starts a new recovery with a fresh retry budget, so `max_bgerror_resume_count` can be exceeded. This only applies with auto resume enabled. Worth a separate fix.
Reviewed By: joshkang97
Differential Revision: D126453053
fbshipit-source-id: 3f4be1d12a8d44552313b022db5afe200bc1c6ce
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15431
A filesystem failure during column family creation can leave the MANIFEST and OPTIONS snapshots at different points in the column family lifecycle. Recovery must select the snapshot consistent with the MANIFEST without deleting a newer valid legacy snapshot.
Fix: When the MANIFEST committed an older tracked OPTIONS file, recovery previously rejected the highest-numbered candidate without reading its tracking marker. Writable open then deleted a newer legacy file, or an unreadable highest file, as a rejected tracked orphan; read-only open also reported the older file number. Recovery now inspects every higher candidate. A newer legacy snapshot supersedes the older committed snapshot, while an unreadable higher candidate is retained rather than deleted.
Add: Crash-test invocations randomize `track_options_file_number_in_manifest`. Before writable open can replace evidence, db_stress compares the public latest-OPTIONS choice with read-only recovery, using only currently live column family descriptors and running before requested missing families are appended. After open, it verifies the persisted tracking marker and live column family set. TransactionDB still skips the unsupported read-only probe, but it now reads the MANIFEST first so an existing commit enables the post-open verification. The regression test also checks that writable open preserves the newer legacy file before creating another column family.
Change: Reopen MANIFEST-avoidance checks use the loaded `options_` value directly. This is a behavior-preserving simplification, not a separate correctness fix.
Safety and limitations: There is no API, file-format, or production hot-path change. Ordinary writable open still receives the caller options and complete requested column family set. The extra read-only open is stress-only and runs when tracking verification is active. The pre-open comparison remains skipped for best-efforts recovery, TransactionDB, and stacked BlobDB; post-open verification remains skipped for best-efforts recovery, TTL, and stacked BlobDB. Random crash coverage is probabilistic, and this parent diff does not add randomized column family operations; the dependent lifecycle diff adds that profile.
Reviewed By: pdillinger
Differential Revision: D125029493
fbshipit-source-id: 5e291deff735e3cbe6d7df0513f940e84b20f387
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15430
A db_stress blackbox crash test failed in verification with:
`GetLiveFilesStorageInfo failed: IO error: No such file or directory: while stat a file for size: .../OPTIONS-857089`
Since D123104845, `GetLiveFilesStorageInfoImpl` picks the OPTIONS file number while holding the DB mutex, releases the mutex, and only then calls `GetFileSize` on that file. If a `SetOptions()` runs in that window, it writes a new OPTIONS file, and `MaybeDeleteObsoleteOptionsFiles()` keeps only the newest two plus the MANIFEST-committed one. So the file that was just picked can be deleted before it is stat'ed. Checkpoint calls `DisableFileDeletions()` first and is not affected, but a plain `GetLiveFilesStorageInfo()` call, as db_stress makes, gets a spurious IO error.
The fix increments `disable_delete_obsolete_files_` (`DisableFileDeletionsWithLock()`) before releasing the mutex, and decrements it again under the mutex right after the size read. `MaybeDeleteObsoleteOptionsFiles()` already skips cleanup while the counter is non-zero. Any OPTIONS cleanup skipped during that window happens on the next OPTIONS write.
The decrement is done directly rather than through `EnableFileDeletions()`. That method is virtual, and `DBImplReadOnly` and `CompactedDBImpl` override it to return `NotSupported` without decrementing. Read-only DBs do reach this code (their OPTIONS file number is set at open), so going through `EnableFileDeletions()` left the counter stuck at 1 after every call. Calling the non-virtual `DBImpl::EnableFileDeletions()` instead is not safe either: when the counter reaches 0 it runs a forced `FindObsoleteFiles` scan and `PurgeObsoleteFiles`, which a read-only DB must never do. Secondary DBs never take this path, because `ReactiveVersionSet` does not record an OPTIONS file number, so `options_number` is 0 there.
Cost: no extra I/O or scan. Checkpoint already disables deletions before calling, so for it this only bumps the counter. For a caller that has not, such as db_stress, obsolete files that become deletable during the short window are not purged at the decrement. They stay queued and are removed by the next `FindObsoleteFiles` (end of the next flush or compaction, or the periodic full scan). Reading the cached `VersionSet::options_file_size_` under the mutex, as before D123104845, would avoid the stat entirely. I kept the stat because D123104845 chose it deliberately. The cached size is the better option if it is always in sync with `options_file_number_`.
Reviewed By: xingbowang
Differential Revision: D126001916
fbshipit-source-id: 60f4f31c73e48b374f4015ed2847d45e706b85de
Summary:
Before this change, the highest supported block-based table format_version was only available as the internal constant kLatestBbtFormatVersion in table/format.h. Applications that accept a format_version setting from users or configuration (especially for SetOptions()) had no public way to validate it against the linked RocksDB build before applying it.
This adds the public constant
BlockBasedTableOptions::kLatestSupportedFormatVersion (currently 8) and removes kLatestBbtFormatVersion, redirecting all internal and test uses to the new constant. Supported-version checks and option validation are unchanged: format_version above the constant is still rejected as InvalidArgument, and draft format_versions remain usable only under the TEST-only TEST_AllowUnsupportedFormatVersion() hook.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15397
Test Plan: Same value and same checks, so no new tests. Existing format_version validation coverage (e.g. TableTest.UnsupportedFormatVersionConfigTest) and the FormatLatest test instantiations now go through the public constant.
Reviewed By: rban1
Differential Revision: D125117976
Pulled By: pdillinger
fbshipit-source-id: 67e0c5b7eade4bf649957435aa0b28b187793f7c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15416
Why the change is required:
A crash test using `secondary_cache_uri=cachelibwrapper://...` constructs and resets CacheLib/Navy. Sandcastle suppresses CacheLib `XLOG(INFO)` by patching the fbcode XLOG header, but commit `831f00db6a47` enabled the exported xplat header unit. The build therefore consumed an unpatched XLOG header, normal CacheLib startup INFO reached stderr, and `db_crashtest` exited 2 before it could accept the intentional whitebox SIGTERM.
Previous behavior:
Any residual child stderr was fatal. `sanitize_known_stderr()` covered expected SIGTERM and io_uring diagnostics but not CacheLib construction/reset INFO, so an otherwise healthy secondary-cache whitebox iteration failed.
New behavior and implementation:
When `secondary_cache_uri` starts with `cachelibwrapper://`, the sanitizer moves only INFO-severity lines from an explicit CacheLib construction/reset source allowlist to stdout. Warnings, errors, fatal logs, unrelated INFO, and identical lines without the CacheLib URI remain fatal.
Safety and compatibility:
Default and non-CacheLib configurations are unchanged. No API, ABI, persisted format, database behavior, or performance-sensitive engine path changes.
___
landed-with-radar-review
Differential Revision: D123695714
fbshipit-source-id: 844ecd2cbf1b3c1415629d787748b5edeb63d456
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15395
**Why the change is required:**
During `DB::Open`, RocksDB reopens the last WAL file and truncates it to its logical size. This is done even when every column family's log number has already advanced past that WAL, so the file is obsolete and will be deleted later in the same open. The only purpose of the truncation is to release space preallocated beyond the logical end of file (e.g. by `fallocate` with `FALLOC_FL_KEEP_SIZE`); `GetFileSize()` reports the apparent size, so truncating to it can only free preallocated blocks. On a FileSystem that never preallocates, this work is pure overhead: a `GetFileSize`, a `ReopenWritableFile`, a no-op `Truncate` (the default `FSWritableFile::Truncate` returns OK without doing anything), and a `Close` on every open. On remote or append-oriented FileSystems the reopen can be expensive.
**Previous behavior:**
- If recovery flushed all data, or the WALs contained no data (including when all WALs are older than the min log number to keep), `MaybeFlushFinalMemtableOrRestoreActiveLogFiles()` always called `GetLogSizeAndMaybeTruncate()` on the last WAL. In read-only mode it still issued a `GetFileSize()` whose result was discarded.
- With `avoid_flush_during_recovery`, `RestoreAliveLogFiles()` always reopened and truncated the last live WAL.
**New behavior:**
A FileSystem can report that it never preallocates space by returning false from the new `FileSystem::IsPreallocationSupported()`. When it does, DB open skips reopening and truncating the WAL. In the flushed / no-data path it also skips the unused `GetFileSize()`, so no I/O is issued against the old WAL. The read-only no-data path no longer issues the discarded `GetFileSize()` either.
**Implementation:**
- `include/rocksdb/file_system.h`: new `virtual bool IsPreallocationSupported()`, defaulting to `true`. `FileSystemWrapper` forwards it to the target.
- The policy lives in the two callers in `db_impl_open.cc`; `GetLogSizeAndMaybeTruncate()` is unchanged. `RestoreAliveLogFiles()` still calls `GetFileSize()` because the size is needed for `alive_wal_files_` and `wals_total_size_`.
**Safety, compatibility, and limitations:**
- The default is `true`, so every existing FileSystem, including ones that override `SupportedOps()` without calling the base class, keeps the current behavior. This is why it is a separate virtual function instead of an opt-in `FSSupportedOps` bit, which would silently disable truncation for those FileSystems.
- No built-in FileSystem overrides the default: Posix, Windows, `MockFileSystem`, and `LegacyFileSystemWrapper` all report `true`, so their WAL truncation behavior is unchanged. Posix deliberately reports `true` even in builds without `ROCKSDB_FALLOCATE_PRESENT`: WALs may have been preallocated by a different build running on the same machine, and a build-dependent answer would leave that space unreclaimed.
- The function describes what the FileSystem can do, not per-file options such as `allow_fallocate` or build configuration. A WAL may have been preallocated by an earlier process with different options, so gating on the current options could leak space.
- A FileSystem that wrongly returns false can only leak preallocated space, in archived WALs (`WAL_ttl_seconds` / `WAL_size_limit_MB`) or across a crash loop before obsolete WALs are deleted. Data is unaffected.
Reviewed By: joshkang97
Differential Revision: D125049741
fbshipit-source-id: ac9795f2045c46911cd86b8faacef5aa49043496
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15404
The RocksDB Folly checkout rewrites a dead libaio source URL to Debian's mirror. The pinned Folly revision now uses releases.pagure.org, so the rewrite still matched the older pagure.io archive URL and became a no-op. Public Folly builds then downloaded an empty file and failed SHA-256 verification.
Update the rewrite to match Folly's current libaio URL. The Debian archive has the same libaio 0.3.113 contents and SHA-256 already recorded in the Folly manifest, so the checksum and extraction directory remain unchanged. This only affects the public Folly checkout path; RocksDB runtime behavior and system-package users are unaffected.
___
landed-with-radar-review
Differential Revision: D125347462
fbshipit-source-id: 4203b6a0b6037efb23aa1f44306b0ca0cd73a43c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15344
Add the internal TableBuilder::TryAdd() contract so RocksDB-owned table builders can distinguish accepted rows, real builder errors, and clean requests to cut the current output table and retry the same row. Add diagnostic reason codes for retryable table boundaries; existing internal table builders inherit a default TryAdd() implementation that forwards to Add().
Translate the public ExternalTableBuilderBase::TryAdd() result in ExternalTableBuilderAdapter so external table implementations can request a file cut while RocksDB compaction uses the internal TableBuilder contract.
Teach compaction output writing to honor kRequiresNewTable by closing the current output file, opening a new one through the existing callbacks, and retrying the same key without advancing the input iterator. Split OutputValidator validation into CanAdd() and AddValidated() so validator state is not updated until the key is actually accepted by a table builder.
Add a DB compaction test with a custom TableFactory that wraps block-based tables and requests one mid-compaction file cut. The test verifies the cut produces two readable output files and preserves all key/value results.
Reviewed By: jaykorean, xingbowang
Differential Revision: D123932406
fbshipit-source-id: 9ed82c3782787de192bca15e40bfbf014ef28c15
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15394
## Summary
RocksDB's `.clang-tidy` enables several readability checks that existing
code still violates: `readability-isolate-declaration` (476 findings),
`readability-braces-around-statements` (49),
`readability-redundant-string-init` (25), `modernize-use-using` (10),
`misc-definitions-in-headers` (3) and `readability-duplicate-include`
(1). They keep getting flagged whenever nearby lines change. This diff
fixes 553 of the 564; upstream has since fixed the other 11.
What changed:
- Multi-variable declarations such as `std::string a, b;` are split into
one declaration per variable. Pointer and reference declarators are kept
per variable.
- Braces are added around single-statement `if`/`else`/`for`/`while`
bodies, and empty spin loops get an empty `{}` body.
- `std::string s = ""` / `member("")` initializers are removed.
- Remaining `typedef`s in the touched C++ files become `using`. C headers
and `extern "C"` code are untouched.
- `inline` is added to non-template function definitions in test headers.
- A duplicate `#include "util/udt_util.h"` in `db_impl.cc` is removed.
These are clang-tidy's own fix-its, applied mechanically and then
clang-formatted on the changed lines. Three kinds of broken fix-it output
were repaired by hand. Removing a `member("")` initializer left a dangling
comma in constructor initializer lists (`ExternalSstFileInfo`,
`ColumnFamilyMetaData` and others). The function-type typedefs in
`block_test.cc` came out malformed. Empty `while (cond);` spin loops got a
`{;` body. Splitting declarations also exposed two vectors in
`db_etc2_test.cc` to `performance-inefficient-vector-operation`, so a
`reserve()` was added for each.
The lint bot reports every finding in a file a diff touches. This diff
splits declarations in `options/configurable_test.cc`, so it also fixes
that file's 14 older `cert-err58-cpp` and
`cppcoreguidelines-avoid-non-const-global-variables` findings, which come
from seven mutable file-level maps. Six are now `const` and built where
they are used (function-local statics, or locals in their only user).
The unused `imm_struct_option_info` is deleted.
Compatibility: no behavior change. In public headers, the only edits are
removed redundant `""` initializers, split declarations and added braces.
Reviewed By: rban1
Differential Revision: D123186310
fbshipit-source-id: f9c8e88b00e997db0196f8725c7c8d880cf798da
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15337
## Why
RocksDB crash tests repeatedly fail with `MultiGetEntity error: Not implemented: ReadAsync: failed to init io_uring` when synchronous `MultiGet` requests async I/O on a thread that cannot create an io_uring. The same `NotSupported` status can reach users if capability detection succeeds but an individual asynchronous submission later fails.
## Previous behavior
`Version` cached `FSSupportedOps::kAsyncIO` when it was constructed. Posix io_uring instances are thread-local, so that result described the construction thread rather than the thread later executing MultiGet. A cached positive result could route a different thread into `ReadAsync`, where io_uring creation failed and the error reached the caller. A cached negative result could also unnecessarily disable async I/O on a capable read thread.
## New behavior and implementation
Synchronous `Version::MultiGet` now checks async-I/O support on the thread executing the read. Both the cross-level `optimize_multiget_for_io` path and the within-level overlap path enter their coroutine-based I/O only when that thread supports it. Posix `SupportedOps` eagerly creates and retains that thread-local io_uring, so subsequent submissions reuse the verified ring.
The capability remains advisory: if `ReadAsync` nevertheless returns `NotSupported`, `AsyncFileReader` completes the `FSReadRequest` synchronously, including requests that use filesystem-owned buffers. Poll errors are applied only to requests that were actually submitted asynchronously, so synchronous fallbacks in a mixed batch retain their successful status.
Regression coverage forces io_uring capability failure only on a dedicated read thread and exercises both synchronous MultiGet branches. Separate tests cover all-request fallback and a batch mixing submitted and synchronously fallen-back requests under an injected Poll error. A one-line bug-fix release note records the user-visible correction.
## Safety and compatibility
This changes no API, ABI, or persisted format. The support check is only performed when synchronous `MultiGet` receives `ReadOptions::async_io`; the Posix steady-state cost is a virtual call and thread-local ring lookup per such MultiGet. Direct I/O continues to use the existing synchronous `MultiRead` route before `AsyncFileReader`, and native coroutine APIs are unchanged.
Reviewed By: pdillinger
Differential Revision: D117802488
fbshipit-source-id: 0f2991a523607d894436cd53123e7a1b2fe93167
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15388
`InjectedErrorLog::Entry` in `utilities/fault_injection_fs.h` is written to the injected error log in its native layout, and a `static_assert` requires `sizeof(Entry) == 256` so the record size stays stable. That assert fails on i386 builds.
Root cause: the fields add up to only 252 bytes. On targets where `alignof(uint64_t)` is 8 (x86_64, aarch64, armv7), the compiler adds 4 bytes of implicit tail padding, so the size is 256. On i386, `uint64_t` members are 4-byte aligned inside a struct, so there is no tail padding and the size is 252.
Fix: add an explicit trailing `uint8_t padding[4]` field so the size no longer depends on `uint64_t` alignment.
Compatibility: on 64-bit targets the new field takes the place of the old implicit tail padding, so every field offset and the 256-byte record layout are unchanged. `Record()` value-initializes the entry (`Entry entry{}`), so the padding bytes are always written as zero. No decoder in the repo depends on the padding bytes.
___
landed-with-radar-review
Differential Revision: D124873369
fbshipit-source-id: f5c6f43ca85013b0e769853ad13b8d4f3f89b958
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15376
RocksDB `DBIter` polled `ROCKSDB_THREAD_YIELD_CHECK_ABORT()` while scanning ordinary forward entries, but reverse `Prev()` / `SeekForPrev()` traversal and `MergeValuesNewToOld()` could scan many internal keys without polling. Long tombstone, version, or merge-operand runs could therefore keep a MyRocks query running after `KILL QUERY` until the scan finished.
Poll the existing hook before each potentially expensive internal iterator movement in the forward scan, forward merge-operand scan, and four reverse traversal loops. Mark the abort branch `UNLIKELY`; on abort, invalidate the iterator and return `Status::Aborted("Query abort.")`.
MyRocks already defines the hook and maps `Status::Aborted` for a killed THD to `HA_ERR_QUERY_INTERRUPTED`, so no MyRocks change is required. There are no API, ABI, or persisted-format changes. The hook remains Linux-only, and cancellation remains cooperative: this cannot preempt an individual underlying iterator operation that never returns.
The change is rebased onto current master, including D124377137's fix for the unrelated `WriteBufferManagerTest.FlushPolicyRetiresAndRestartsHandoffWorkers` failures seen on V1.
Reviewed By: pdillinger
Differential Revision: D124441004
fbshipit-source-id: 53bf83a00121b42566dbd93de232e973cbe78688
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15387
## Summary
RocksDB's `.clang-tidy` enables `cppcoreguidelines-special-member-functions`,
`cppcoreguidelines-virtual-class-destructor` and the noexcept-move checks,
and clang warns `-Wdeprecated-copy-with-user-provided-dtor`. A full scan
reports 476 findings. In most of them, a class declares a destructor (or
copy operations) but leaves the rest of copy/move implicit. That has two
costs. Readers can't tell whether a class may be copied or moved. And
resource-owning classes keep an implicit copy that would double-free if
anyone ever used it. This diff resolves 453 of the 476.
Approach, picked per class from least to most invasive:
1. An empty user-provided destructor becomes `= default`. This covers most
interface classes, such as `virtual ~X() {}` -> `= default`. If the
class is still copied, that would only rename the deprecated-copy
warning, so the destructor is removed instead when nothing needs it
(`Customizable`, `TableFactory` and `InternalKeyComparator` inherit a
virtual one; `IteratorWrapperBase` isn't polymorphic), and
`MemTableRep::KeyComparator` gets defaulted copy and move.
2. Classes whose copy is already deleted get explicitly deleted move
operations too. Move was already unavailable, so this only documents
it.
3. Resource-owning classes that are never copied get deleted copy and move
operations. Examples: the C API wrapper structs, which run a
`destructor_` callback (edited in `tools/c_api_gen/c_base.cc` and
regenerated into `db/c.cc`); `ShardedCache`, which frees its shards;
iterator wrappers; and RAII handles.
4. Value types that are copied or moved get the missing operations as
`= default`, which keeps their current behavior.
5. Exception specifications of existing functions are unchanged, except
for `WriteBatch`'s move assignment (below): no other `noexcept` is
added or removed. Only the move operations this diff
newly declares (`Comparator`, `FileSampledStats` and the thread-local
`Entry`) are `noexcept`; they copy only scalars or atomics.
`WriteBatch`'s copy and move assignment destroyed `*this` and
placement-new'ed a new `WriteBatch` over it. That is undefined behavior
when `*this` is a subclass, and if the copy constructor threw, `*this`
was left destroyed and was later destroyed again. Move assignment now
assigns member-wise and ends in the same state as the move constructor,
which leaves `has_key_with_ts_`, the timestamp-size tracking and the
other fields it doesn't copy at their defaults. `rep_` is swapped through
a temporary, because a plain string move-assignment can keep the old
buffer (e.g. a large batch reset with `batch = WriteBatch()` would keep
its memory). Nothing in it can throw, so it is now `noexcept`. Copy
assignment copies into a temporary and moves it in, so a throwing copy
leaves `*this` unchanged.
Bugs this exposed and fixed:
- `RestoreAfterCopyOrCreateWorkItem` (backup engine): the hand-written
move constructor and assignment moved only some fields, so `from_file`
and `to_file` were empty after `push_back(std::move(...))`. A restore
that hit a checksum mismatch then reported "While restoring -> : ..."
without file names. It now uses the compiler-generated moves, which
are still `noexcept`.
- `ExternalSstFileIngestionJob` lived in a `std::vector`, and its
destructor frees raw-owned pointers, so the implicit copy used on
vector reallocation would double-free. Its file batches also point into
its own inline file storage, so neither a copy nor a move is safe after
`Prepare()`. Copy and move are now deleted, and `DBImpl` holds jobs
through `std::unique_ptr`.
Compatibility: no public class loses an operation users can call. Public
interfaces mostly only get `virtual ~X() = default;`.
`WriteBatch`'s move assignment becomes `noexcept`, which is not part of
the mangled name, so the ABI doesn't change.
`SubcompactionJobInfo` and `CompactionJobInfo` intentionally stay C++20
aggregates, so users' aggregate and designated initializers keep
compiling. `Configurable`, `CompareInterface`, `WriteBatchBase`,
`ObjectLibrary::Entry` and `MemTableRep::KeyComparator` declare defaulted
copy and move, and `Customizable` no longer declares a destructor. As a
result, `Configurable` and `Customizable` subclasses that declare no
destructor or copy operations of their own now move their
registered-options list instead of copying it. In-tree, these are the
mutable-options adapters, the table factories, the memory allocators and
test classes; all are held through smart pointers or used in place, and
`Clone()` copies, so none is moved by value.
Not changed (23): `IODebugContext` (custom deep copy plus a
`std::shared_mutex`), `DBWithColumnFamilies` in db_bench (stored in a
`std::vector` with non-movable members), `Snapshot`'s protected virtual
destructor (intended), `ManagedSnapshot`, the `SubcompactionJobInfo` and
`CompactionJobInfo` aggregates, and the 16 noexcept-move-constructor and
noexcept-swap findings on existing functions, including `autovector`'s
move assignment (adding `noexcept` would change their exception
specifications). Copying `ManagedSnapshot` already double-releases the
snapshot. Fixing that needs a real move implementation and is a behavior
change, so it is left for a separate diff.
Reviewed By: rban1
Differential Revision: D123186309
fbshipit-source-id: 6f8e57e233e12199a007407130bb9a1f6d52e8d1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15386
Fixes crash-test failure T292562874:
`Checkpoint failed: Corruption: [RocksDBOptionsParser]:failed the verification on BlockBasedTable::: filter_policy.id`
Root cause:
- D123104845 ("Commit authoritative OPTIONS file numbers in MANIFEST", 77d94c989243) made subset checkpoints (`CreateCheckpoint` with `include_cf_ids` that exclude some CFs) write a new OPTIONS file. `DBImpl::GetLiveFilesStorageInfoImpl` captures the snapshot under the DB mutex with `cfd->GetLatestCFOptions()`. That is a shallow copy: `table_factory`, and through it `filter_policy`, are `shared_ptr`s to the live objects. `DBImpl::CreateOptionsFileForSubsetCheckpoint` then calls `PersistRocksDBOptions` without holding `options_mutex_`. `PersistRocksDBOptions` serializes the file, then re-parses it and checks it against the same in-memory options.
- `RibbonFilterPolicy::bloom_before_level_` is a `kMutable` atomic that `SetOptions(table_factory.filter_policy.bloom_before_level=...)` changes in place, and `RibbonFilterPolicy::GetId()` includes it (`ribbonfilter:<bits>:<bloom_before_level>`). A concurrent `SetOptions` between the write and the verify makes the in-memory `filter_policy.id` differ from the file, so verification returns Corruption and the checkpoint fails.
- Every other OPTIONS writer (`WriteOptionsFile`, CF create/drop) holds `options_mutex_`. `SetOptions` holds it for its whole mutate-and-persist sequence. The new subset-checkpoint writer was the only one that skipped it.
Trigger in the crash test: db_stress randomly sets `table_factory.filter_policy.bloom_before_level` through `SetOptions` (when `keepRibbonFilterPolicyOnly`, i.e. `bloom_before_level` != INT_MAX with ribbon filters; see the OPTIONS dump with `ribbonfilter:18:1`, `ribbonfilter:8.9:2147483647`, `ribbonfilter:5.67:-1` across CFs). It also takes subset checkpoints. The two run concurrently on different threads.
Fix: hold `options_mutex_` in `CreateOptionsFileForSubsetCheckpoint` while persisting and verifying, the same as `WriteOptionsFile`. The only caller is `CheckpointImpl::CreateCheckpoint`, which holds no DB locks at that point, so lock ordering (options_mutex_ before mutex_) is kept.
Note: the snapshot's scalar fields are still captured at live-file time. Live pointer sub-options such as the filter policy reflect their state at persist time. This was already true of the shallow snapshot. The change only makes the persisted file self-consistent with what is verified.
Reviewed By: xingbowang
Differential Revision: D124728996
fbshipit-source-id: adb5511c2ddd73ab138fd7fd21109199ff4f5b1e
Summary:
has been neglected for a few months, finally taking care of it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15375
Test Plan: local folly build (from internal wiki instructions) to verify the fixes from claude, CI
Reviewed By: archang19
Differential Revision: D124436669
Pulled By: pdillinger
fbshipit-source-id: 0a0e3f6857cc8ba20941281b9dfd618a6997b86f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15310
Why the change is required:
The committed OPTIONS number in MANIFEST gives persisted-options readers a
precise source of truth. If that file is missing or corrupt, or an operator
must deliberately replace it with a known-good snapshot, persisted-options
loading can fail before a normal writable open can publish a replacement.
RocksDB therefore needs a supported offline recovery path that obeys the
binding and global file-number rules.
Previous behavior:
ldb update_manifest could rewrite file-temperature metadata but could not
repair an OPTIONS binding. Operators could not safely bind a selected snapshot
while accounting for the current MANIFEST boundary, every file-number
allocation root, recoverable WAL output, and crash durability. Blindly
selecting or reusing an existing OPTIONS number could bind the wrong snapshot
or collide with another RocksDB file.
New behavior:
ldb update_manifest --rebind_options_file --options_file=OPTIONS-<number>
performs an offline repair from an explicitly selected trusted snapshot whose
tracking marker is enabled. Optional --file_number_paths arguments identify
external allocation roots that are not serialized in OPTIONS.
The command locks the DB and validates that the selected OPTIONS snapshot
contains every live MANIFEST column family. Extra OPTIONS CFs are accepted as
a safe superset and are filtered by binding-aware persisted-options APIs; a
snapshot missing any live CF is rejected. The command raw-copies the selected
bytes to a newly allocated canonical OPTIONS number, syncs that file and its
directory entry, and commits the new number with one MANIFEST record. It
rewrites the current MANIFEST only through the last completely decoded record,
appends the commit and next-file boundary to a temporary copy, syncs it,
atomically renames it over the original, and syncs the MANIFEST directory.
The crash contract has one publication point. Before MANIFEST replacement,
the previous commit remains authoritative and the copied tracked OPTIONS file
is only an ignorable orphan. After replacement, the new snapshot is
authoritative.
Implementation:
- Parse the selected snapshot for validation but copy it byte-for-byte so
unknown or newer option fields are preserved.
- Replay MANIFEST to recover live CF membership, the committed number, next
file number, and last valid logical-record boundary; discard a torn physical
tail during the bounded-memory rewrite.
- Scan the DB root, WAL/archive directories, serialized roots, and
operator-supplied roots for allocated numbers.
- Parse recoverable WALs and reserve conservative headroom for recovery
outputs, missing-column-family creation, persistent statistics, asynchronous
WAL creation, MANIFEST rollover, and subsequent writable-open allocations.
- Reject file-number overflow, missing allocation roots, corrupt or unreadable
WALs, invalid input names, missing-live-CF snapshots, and simultaneous
temperature rewriting.
- Preserve the existing update_temperatures behavior and result text.
Safety boundaries and unsupported cases:
- This is an explicit offline mutation. It never auto-selects a snapshot and
refuses to run while another process holds the DB lock.
- The selected input must be a canonical OPTIONS-<number> basename in the DB
directory, must parse successfully, must enable MANIFEST tracking, and must
contain all live MANIFEST CFs. Extra CFs are allowed but missing live CFs are
not.
- The command does not repair snapshot contents or alter topology.
- OPTIONS does not serialize all DB/CF data roots. Operators must provide every
external root through --file_number_paths; the command fails closed when a
supplied or inferred root cannot be scanned.
- Conservative allocation headroom can reject a repair near kFileNumberMask
even when a particular future open would allocate fewer files.
- Normal DB open and core publication do not depend on this maintenance
command.
Validation coverage and known limitations:
Coverage includes parsing and incompatible modes, the DB lock, tracking and
membership validation, safe-superset acceptance, missing-live-CF rejection,
byte-preserving copy, torn MANIFEST tails, one-record commit publication,
file-number exhaustion and writable-open headroom, external and missing roots,
WAL recovery bounds, asynchronous WAL and persistent-stat allocations,
MANIFEST rollover, writable reopen, and later CF creation. The end-to-end test
confirms safe-superset rebinding and rejection of a higher tracked snapshot
that omits a live CF, verifies the final committed MANIFEST number, and runs
the real ldb scan --try_load_options subprocess. Cross-platform CI is not part
of local validation.
Reviewed By: pdillinger
Differential Revision: D123104846
fbshipit-source-id: 8cf7c54c1c7a360411034469614eed8aaffdcf8e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15382
Why the change is required:
IteratorCloseWALFile1 is flaky under suite load. Its WAL-count assertions assume the iterator is the final owner of an obsolete SuperVersion, but the test did not prove that the preceding background flush had released its temporary references or that the next flush was unable to start.
Previous behavior:
The test occupied one HIGH-priority Env worker with a sleeping task. The shared Env can have multiple HIGH-priority workers, so another worker could run the flush concurrently. Also, Flush() can return before BackgroundCallFlush() has fully exited. Either race can leave two WAL writers open after iterator deletion. A fatal assertion could then destroy the stack-local SpecialEnv before fixture teardown closed the DB, causing a secondary use-after-return or segmentation fault.
New behavior:
The test waits for DBImpl::BGWorkFlush:done after the iterator-establishing flush, then replaces that dependency with a barrier at DBImpl::BackgroundCallFlush:start for the flush triggered by the final write. The iterator is therefore the final old-SuperVersion owner when it is destroyed, and the test releases the blocked flush only after checking synchronous WAL closure.
Implementation:
Replace the sleeping background task with existing SyncPoint dependencies. Own the iterator with unique_ptr, disable SyncPoint processing on every exit, and close the DB with a scope guard before the local SpecialEnv is destroyed.
Safety, compatibility, and limitations:
This changes only test synchronization and failure cleanup. It does not change production code, RocksDB APIs, persisted formats, locking, or runtime behavior. The test continues to cover foreground iterator cleanup with background_close_inactive_wals enabled; background purge behavior remains covered separately by IteratorCloseWALFile2.
___
landed-with-radar-review
Differential Revision: D124653963
fbshipit-source-id: 65d381128b5013b4d62ad60e42f0e4566928f000
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15305
Why the change is required:
Column-family topology is durable in MANIFEST, while its operational
configuration snapshot is stored in a separate OPTIONS file.
CreateColumnFamily and DropColumnFamily previously committed MANIFEST first
and wrote OPTIONS afterward. An OPTIONS I/O failure could therefore make the
API report failure after the topology change was already durable.
Persisted-options readers also selected the highest file number and could
consume an orphan left by a failed publication.
Previous behavior:
RocksDB had no durable correspondence between a MANIFEST state and the
OPTIONS snapshot describing it. LoadLatestOptions and
GetLatestOptionsFileName treated the highest-numbered OPTIONS file as
authoritative. Column-family creation and deletion returned the later OPTIONS
persistence error even when their MANIFEST edit had already succeeded.
Multi-column-family operations wrote a complete OPTIONS snapshot after every
successful member, producing quadratic OPTIONS bytes.
Ordinary DB::Open consumes caller-supplied DB and column-family options rather
than loading OPTIONS. This change preserves that contract.
New behavior:
Add the default-disabled experimental DB option
track_options_file_number_in_manifest. Its value is serialized into each
OPTIONS file so the selection rule is self-describing and participates in the
normal DB-options compatibility checks.
When enabled, RocksDB first durably creates and renames OPTIONS-N and then
commits N with one safely ignorable MANIFEST field. A normal OPTIONS rotation
uses one MANIFEST commit. Column-family creation and deletion prewrite the
proposed snapshot and commit its number together with a topology edit in the
same VersionEdit. A failure before the OPTIONS file is durable changes no
topology. A failure after the file is durable but before MANIFEST commit leaves
an orphan that binding-aware readers ignore, while the previous committed
snapshot remains authoritative.
Each multi-CF create or drop writes one proposed snapshot, independent of the
batch size. Create commits the new number with the first successful add; drop
commits it with the last successful drop. Existing partial-success semantics
can therefore leave a committed snapshot with extra CFs, but never one missing
a live MANIFEST CF. Persisted-options APIs filter this safe superset against
recovered MANIFEST membership. A snapshot missing a live CF is corruption.
DB::Open with create_missing_column_families uses the same batched path.
If the newest OPTIONS file has tracking disabled, it retains the legacy
highest-numbered-file behavior, including after an old or disabled writer has
run. If it has tracking enabled, readers use the recovered MANIFEST commit and
reject newer tracked orphans. A missing usable commit falls back to the highest
tracked snapshot for RepairDB and legacy-rewrite recovery. DB open logs that
fallback through its configured logger; logger-less utility APIs remain
silent. Writable open removes rejected tracked orphans when local obsolete-file
deletion is enabled. Read-only open never mutates the DB directory.
A DB with no recovered commit and tracking disabled takes the legacy fast path
without parsing the newest OPTIONS file. When inspection is required, RocksDB
reads only the tracking marker. Marker-read failures used for selection do not
fail DB open; a recovered commit remains authoritative and ambiguous newer
candidates are conservatively retained.
The first topology change after enabling tracking first binds the current
topology, so failure of the proposed change cannot make its orphan
authoritative. MANIFEST failure during an ordinary tracked OPTIONS rotation
also enters the existing manifest-write background-error path.
Implementation:
- Add the safely ignorable CommittedOptionsFileNumber VersionEdit field and
carry it through recovery, MANIFEST snapshots, rollover, and subset
checkpoints.
- Persist the tracking marker through C++, C, and Java options, generated
round-trip coverage, compatibility checking, and db_stress configuration.
- Centralize binding-aware selection for DB open and public persisted-options
APIs. Reuse RecoveryContext's directory listing, use a narrow marker scan,
and recover live CF names while replaying MANIFEST.
- Filter manifest-selected safe-superset snapshots to live MANIFEST CFs in
LoadLatestOptions and compatibility verification without discarding their
option maps.
- Prevalidate and sanitize CF batches outside the DB mutex, write one complete
snapshot while options_mutex_ serializes topology changes, and hold the DB
mutex only for the existing individual MANIFEST commits. Batch missing-CF
creation during DB open follows the same locking boundary.
- Keep the committed snapshot live in cleanup and live-file APIs, and make
subset checkpoints write an OPTIONS snapshot and filtered MANIFEST with
identical membership in tracked and legacy modes.
- Preserve the first partial multi-CF drop error while consuming legacy
OPTIONS persistence Status, and sync the checkpoint MANIFEST directory after
replacement.
- Pass checkpoint helper callbacks by reference and group VersionEdit storage
by alignment so the new paths add neither callback copies nor avoidable
object padding.
Safety boundaries and unsupported cases:
- Tracking defaults to false. Disabled writers and never-tracked databases
retain legacy write ordering and highest-numbered selection.
- Older binaries safely ignore the new MANIFEST tag and continue choosing the
highest OPTIONS number. Downgrade is readable but not crash-atomic: an old
binary can consume a tracked orphan. That is the same class of
OPTIONS/MANIFEST inconsistency as the previous behavior.
- A rollback binary must know the serialized option name before the feature is
enabled. Exact compatibility checks intentionally report a mismatch when the
caller's tracking setting differs from the persisted snapshot.
- A newer untracked OPTIONS file deliberately supersedes an older committed
number, allowing an old or disabled writer to become authoritative without
rewriting MANIFEST.
- Reactive secondary and follower instances do not enforce or repair the
binding. Ordinary DB::Open still uses caller-supplied options.
- Writable open does not delete rejected tracked or ambiguous files when a
compaction service owns OPTIONS-file retention. Read-only open performs no
repair writes.
- Direct consumers that scan OPTIONS filenames themselves are unchanged; the
binding is enforced by RocksDB's persisted-options APIs.
- If the committed snapshot is missing and only tracked candidates remain,
selection intentionally falls back to the highest candidate so RepairDB
output remains usable.
Validation coverage and known limitations:
Focused coverage exercises encoding compatibility, tracked and legacy
selection, the never-tracked open fast path, marker-read failures, exact
tracking-marker compatibility, activation and disable/re-enable transitions,
batched create/drop and create-missing behavior, partial-batch safe supersets,
OPTIONS/MANIFEST failure injection, background-error propagation, orphan
cleanup and retention, read-only non-mutation, subset-checkpoint behavior,
stale OPTIONS recovery, RepairDB option sanitization, and transient
read-corruption reconstruction. The assertion-enabled full executable suite
passes. The repository check-format and check-sources wrappers cannot run
because they invoke Git inside this Sapling checkout; fbsource lint is used
instead. Cross-platform CI is not part of local validation.
Reviewed By: pdillinger
Differential Revision: D123104845
fbshipit-source-id: 77d94c989243c3f3f9fedfa81fa364c166250303
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15373
Fix worker retirement when changing `WriteBufferManager` from `kFlushLargestAcrossDBs` to `kFlushOldest`. The coordinator could pass its active-policy check, consume the policy-change notification during refresh, and enter a one-second idle wait. This delayed retirement beyond the 300 ms deadline in `WriteBufferManagerTest.FlushPolicyRetiresAndRestartsHandoffWorkers`. Recheck the policy under `sorter_wait_mu` before waiting so the coordinator promptly enters the retirement path. The setter remains non-blocking, and retirement still waits for executor callbacks to finish.
Also fix startup pressure-wait accounting in `BackgroundFlushCycleTest.FastRejectedCandidateDoesNotAddPressureDelay`. Registering the healthy candidate can invalidate the coordinator's initial ranking snapshot and cause a legitimate pressure wait before either handoff runs. The old assertion counted waits from startup and incorrectly attributed that wait to the rejected candidate. Record the counts at rejected and healthy dispatch and require them to match, preserving the check that rejection adds no pressure interval.
Include both fixes in this diff so its `write_buffer_manager_test` target covers both corrected behaviors. Replace worker retirement and restart polling with sync-point synchronization. The rejection test deterministically invalidates the first snapshot during healthy-candidate registration, verifies that a startup wait occurred, and checks that advancing to the healthy candidate adds no further pressure wait. Retain the clang-format 21 line wrapping required by CI.
Reviewed By: pdillinger
Differential Revision: D124377137
fbshipit-source-id: 1420196e47ccba0268d0427c4ed623925aa3ad5b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15366
## Summary
RocksDB's `.clang-tidy` enables `modernize-use-override`, and clang
reports `-Winconsistent-missing-destructor-override` when a class marks
its other overrides `override` but not its destructor. A full scan finds
190 `modernize-use-override`, 100
`inconsistent-missing-destructor-override` and 2 `suggest-override`
findings in existing code. Each one comes back whenever a nearby line is
edited. They also leave classes where it isn't obvious which methods are
overrides and which introduce new virtuals.
This diff marks every overriding member function and destructor
`override` and drops the redundant `virtual` on them. It also changes
`final override` to plain `final`, following the check's default. Methods
that introduce a new virtual keep `virtual`. Most edits are clang-tidy's
own fix-its, applied mechanically and then clang-formatted.
Compatibility: no signatures change. In public headers the only effect is
`override` on existing overrides (mostly destructors, plus
`Cache::Name()` and `Cache::GetPrintableOptions()`, whose base
declarations are unconditional), so user code that subclasses these types
is unaffected. Every touched class now marks all of its overrides
consistently. This matters because a partially-marked class would make
clang emit `-Winconsistent-missing-override` in user builds.
Reviewed By: rban1
Differential Revision: D123186308
fbshipit-source-id: 1f489ea32d938e3ec0faf9abc4f7ff49e042d37a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15372
Why: keep internal RocksDB main ready for 11.13.0 development while the 11.12 release is cut.
Previous behavior: Conveyor opened a GitHub pull request for this update, which could be missed and was not stacked with the internal release diff.
New behavior: Conveyor owns an internal Phabricator diff that the release-import diff stacks on top of. The release manager exports this diff manually with ExportIt.
Implementation:
- Bump version.h from 11.12.0 to 11.13.0
- Add `11.12.fb` to `check_format_compatible.sh`
- Sync HISTORY.md `## 11.12.` section from `11.12.fb`
- Delete 24 consumed `unreleased_history/` note file(s) from main
Safety and unsupported cases:
- Only version.h, check_format_compatible.sh, HISTORY.md, and consumed release-note files are allowed.
- Open legacy GitHub update-main PRs block this flow; this automation never modifies, exports, or lands them.
- Concurrent direct CLI invocations are unsupported; Conveyor serializes the Release Cut node to one active run.
- The tool refuses unrelated commits, mismatched release metadata, and non-canonical GitHub repositories.
Known limitations:
- ExportIt remains manual. Bot create permission for internal_repo_rocksdb/repo is a rollout prerequisite.
Reviewed By: archang19
Differential Revision: D124370612
fbshipit-source-id: fcd3864695f118bf8be316878d92be5bf8ace43c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15371
`DeleteSchedulerTest.DirSyncFailureAfterTrashFileDeleted` aborts with "Failed to check Status" in builds with `ASSERT_STATUS_CHECKED=1`.
**Why it matters:** this breaks the `ASSERT_STATUS_CHECKED=1` `make check` CI job. Only the test is wrong; `DeleteScheduler` is unchanged.
- **Problem:** the test calls `GetBackgroundErrors().size()`. The returned `std::map<std::string, Status>` holds unchecked `Status` copies, so destroying it aborts.
- **Fix:** keep the map, assert its size, and check each error with `ASSERT_NOK`, as the other tests in this file do.
___
landed-with-radar-review
Differential Revision: D124369694
fbshipit-source-id: 85429b0bc886f6c87af3836bfe34b5da6eb47459
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15325
## Summary
RocksDB's `.clang-tidy` enables the `clang-analyzer-*` and `bugprone-*`
checks. A full scan of the repo still reports 78 findings from these
checks in existing code. Each time a nearby line is edited, they come
back as noise, and a few of them are real defects. This diff resolves 73
of the 78; upstream has since fixed one more, and the four left are
listed under "Intentionally not changed". Some fixes repair real bugs.
The rest make the code's intent explicit to the analyzer without
changing behavior.
Real bugs fixed:
- `port/mmap.cc`: `MemMapping::operator=(MemMapping&&)` ran the
destructor on `*this`, `memcpy`'d over a non-trivially-copyable object,
then placement-new'd the source. That is undefined behavior, and `*this`
can be the base subobject of a `TypedMemMapping`. It now moves the
members one by one, and the previous mapping is released through a
local `MemMapping`.
- `port/mmap.h`: `TypedMemMapping::operator=(MemMapping&&)` had no
`return *this;`, so calling it was undefined behavior.
- `cache/typed_cache.h`: `if constexpr (sizeof(TCreateContext) > 0)` is
always true, so the branch meant for empty create contexts (for example
`BlobContentsCreator`) could never run. The check is now
`!std::is_empty_v<TCreateContext>`. Existing instantiations still call
the same `Create` function.
- `cassandra_row_merge_test.cc`: the test reused a vector after
`std::move`-ing it into `RowValue::Merge(std::vector<RowValue>&&)`. That
only worked because `Merge` happens not to consume its input. The test
now rebuilds the input.
Behavior-preserving changes:
- Virtual calls in constructors and destructors are now explicitly
qualified (for example `DBImpl::CloseImpl()` in `~DBImpl`). During
construction and destruction, dispatch already resolved to these
functions.
- Stale `/*name=*/` argument comments now match the parameter names.
- `unique_ptr::release()` after a successful cache `Insert` (where the
cache takes ownership) uses `[[maybe_unused]] T* released = x.release();`,
the pattern already used in `table_cache.cc`.
- Removed `class Mutex;` and `class StaticMeta;` forward declarations
that sat in the wrong namespace.
- Added default member initializers for the
`TablePropertiesCollectorFactory::Context` fields that its default
constructor left uninitialized.
- Smaller cleanups: a redundant `cond ? 3 : 3` in a test,
`sizeof(std::string)` in a memory estimate, and an explicit
`Cleanable` base move via `up_cast`. Tests that intentionally use
moved-from objects now carry `NOLINT(bugprone-use-after-move)`,
matching the existing convention.
Intentionally not changed: enum casts in tests that check invalid-value
handling; three `clang-analyzer-core.NullDereference` reports the
analyzer can't prove safe (the C API timestamp out-parameters in
`db/c.cc`, which are only written when the caller asked for a
timestamp, and the waiter-list walk in `util/async_file_reader.cc`);
bit-flag enums in public headers (`FileOpenContract`,
`OptionTypeFlags`), where OR-combined values are valid; ownership
transfers the analyzer can't see across translation units
(`partitioned_index_reader.cc`, `fs_posix.cc`); and the intentional
two-separator check in `file/filename.cc`.
Reviewed By: rban1
Differential Revision: D123186307
fbshipit-source-id: 623e871286b1c73263b1309cb194692194199e0f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15352
DeleteScheduler now releases a trash file's accounting in two more cases where the file is gone but its size stayed counted until restart.
**Why it matters:** Leaked bytes stay in the total trash size. Once they pass `max_trash_db_ratio` of the tracked DB size, every later deletion skips rate limiting and is deleted immediately, which can cause I/O stalls.
- **Immediate retry:** A failed accounted trash deletion is retried through `DeleteExistingTrashFileImpl`. When the retry deletes immediately, for example with rate limiting disabled, and the file is already gone, `DeleteFileImmediately` skipped `OnDeleteFile` and the retry entry was put back. It now untracks the file and subtracts the entry's size. The NotFound status is still returned.
- **Dir sync failure:** If the unlink worked but `NewDirectory` or the dir fsync then failed, `DeleteTrashFile` reported 0 deleted bytes and skipped `OnDeleteFile`, so the accounting stayed. It now records the deleted bytes and untracks the file right after the unlink, and reports `file_gone` so `BackgroundEmptyTrash` releases the rest without parking a retry entry. The sync error is still reported as a background error.
- **Details:** Adds an `IsFileGone` helper for NotFound/PathNotFound and a `DeleteScheduler::DeleteTrashFile:BeforeSyncDir` sync point for tests. Both new tests keep one more tracked file, so they check that only the gone file's bytes are released.
Reviewed By: anand1976
Differential Revision: D124000703
fbshipit-source-id: 54ed8be66d5670709054a46a258ceaf72b10d899
Summary:
DBWriteBufferManagerTest.FlushLargestAcrossDBsPolicy (added in https://github.com/facebook/rocksdb/issues/15047) can hang forever under load, as seen in nightly
build-linux-arm-test-full (db_write_buffer_manager_test shard 4).
The test's first phase waits for other_db to flush via the OnFlushCompleted listener, then immediately starts a second phase that writes 500KB to other_db and calls TEST_ScheduleFlushOnLargestDB(), expecting other_db to be picked. But OnFlushCompleted fires before the flush thread finishes its post-flush work, including refreshing other_db's WriteBufferManager flushable-memory accounting. If the second phase runs in that window, other_db's accounting is stale and the refresh (which only TryLocks the DB mutex) can leave other_db out of the ranking, so db_ is picked and flushed instead. The test then waits in the unbounded FlushedDBRecorder::WaitForFlush(other_db) forever. The same race occasionally fails ASSERT_TRUE(TEST_ScheduleFlushOnLargestDB()) instead, when other_db still has WBM flush work outstanding.
Fix the test by waiting for other_db's background work to finish (including any WBM flush job) after the first-phase flush, before starting the second phase. No production code changes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15365
Test Plan: Reproduced by running db_write_buffer_manager_test shard 4 of 6 (as generated for a 4-core CI machine) many times concurrently on an oversubscribed machine. In an interleaved A/B of 1500 runs each, the unmodified test hung 71 times (all at the second-phase WaitForFlush, confirmed via stack dumps) and failed the line-995 assertion 16 times; with this change there were no hangs or failures of this test. DB LOG thread IDs from hung runs confirmed the test thread had flushed db_ in the second phase. The hang rate was similar before https://github.com/facebook/rocksdb/issues/15331, so the race dates to https://github.com/facebook/rocksdb/issues/15047.
Reviewed By: rban1
Differential Revision: D124196572
Pulled By: pdillinger
fbshipit-source-id: d621c5a0b441634fc339283caf9ad7a0d6eea401
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15349
DeleteScheduler now releases the accounting for a queued trash file that disappears before it is deleted. Before this change, that size leaked until the process restarted.
**Why it matters:** The leaked bytes stay in the total trash size. Once they exceed `max_trash_db_ratio` of the tracked DB size, every later deletion bypasses rate limiting and is deleted immediately, which can cause I/O stalls.
- **Problem:** If a trash file's directory is renamed or removed while the file waits in the delete queue, `DeleteTrashFile` fails with PathNotFound. `BackgroundEmptyTrash` then keeps the file's remaining accounted bytes in `total_trash_size_` and records it in `failed_accounted_trash_deletions_` for a retry. The retry never happens, because nothing schedules that path again. The file also stays tracked in `SstFileManager`, so the total size is inflated too.
- **Fix:** On NotFound/PathNotFound, `DeleteTrashFile` untracks the file via `OnDeleteFile`, and `BackgroundEmptyTrash` subtracts the file's full remaining accounted size and skips the retry map. The error is still reported through `GetBackgroundErrors()`, so that behavior is unchanged.
Reviewed By: anand1976
Differential Revision: D123986893
fbshipit-source-id: 4c084241a8100ae9d28fb92edf02a4362d6f60cb
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15363
as title. we were not accurately reporting block read sizes and counts to the thread local perf context
given that IODispatcher can coalesce requests, we also count any sub coalesce threshold added bytes to these totals
Reviewed By: chaitanyadatye, xingbowang
Differential Revision: D123941253
fbshipit-source-id: eddc33d144afa49a8137fb174008030d22f3bddd
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15048
Adds a validated db_stress flag for selecting the WriteBufferManager flush policy and randomizes it in db_crashtest. Crash-test generation covers all three policies; kFlushLargestAcrossDBs automatically uses two DBs sharing one manager so the cross-DB path is exercised.
Reviewed By: pdillinger
Differential Revision: D114638570
fbshipit-source-id: 00f664445614e083a4c59d4e2bb0896731d4bea7
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15343
Problem
The cross-DB WriteBufferManager coordinator dispatches ranked flush candidates asynchronously. A candidate can reject its handoff immediately—for example because its DB mutex is busy, writes are stopped, or equivalent WBM work is already scheduled—and finish before the sorter reaches its pressure wait.
Before this change, completion called RequestRefresh(), but the sorter’s active-pressure branch did not consume that already-recorded notification. Condition-variable notifications are not retained, so if completion won the race, the sorter still slept for the full 10 ms pressure interval before trying the next ranked DB. With many quickly rejected candidates, that delay accumulated once per candidate; 10,000 rejected DBs ahead of a healthy DB could theoretically delay useful relief by roughly 100 seconds while shared memory remained above its limit.
Solution
Snapshot the existing refresh-request generation immediately before ProcessFlushHandoffRequest(). After processing, if the handoff is already idle, the current ranked candidate cycle remains active, and the generation advanced, the sorter knows that asynchronous work completed during the dispatch window. It consumes that persistent notification and immediately continues to the next candidate instead of entering the pressure backoff.
The generation check is deliberately narrower than checking the refresh Boolean alone: it reacts only to notifications that occurred during this handoff attempt and avoids treating an older unrelated refresh as completion of the current candidate. The existing 10 ms backoff remains after a complete no-progress candidate cycle, and queued or executing handoffs retain their existing lease/wait behavior.
Testing
A deterministic SyncPoint test pauses the sorter only after a handoff has actually been submitted, lets a rejecting executor callback finish first, and verifies that the healthy next candidate is reached without entering the pressure-wait point. The synchronization wait is bounded so a broken ordering fails the test rather than hanging CI.
Reviewed By: xingbowang
Differential Revision: D123926506
fbshipit-source-id: b057e5b9a41ee0abd415f72ef2067b0d3f718384
Summary:
Finish OpenZL, the internal RocksDB source tree, and the released RocksDB copy, removing all three parent AutoHeaders opt-ins. Declare inferred headers while preserving aliases, globs, selectors, and visibility. RocksDB's generator lists library headers and passes test source headers through its template and shared wrapper. Header lists account for the source files compiled by each release, so the older released tree does not gain headers from newer internal test helpers.
Add `release_to_fbcode.py internal_repo_rocksdb --buck-only -y` to refresh the four Buck generator files and regenerate BUCK against the existing release's source lists. The command restores its five outputs if copying or generation fails, applies the existing generated-file warning, and rejects release-commit options. It does not copy runtime sources, change the release version, clone, commit, or deploy. Both RocksDB BUCK files are produced by the maintained generators.
Graft the OpenZL dev declarations into prod. Align the predictor's Folly include paths and dependencies with prod so promotion is reproducible. Add the IPNext controller transition header explicitly.
Authored with Codex.
Reviewed By: IanChilds
Differential Revision: D123482073
fbshipit-source-id: 05ab47e592d710834da56bbea57f5f93a0aea565
Summary:
The workflow_dispatch-only nightly-candidate.yml shared the exact `name: facebook/rocksdb/nightly` with the scheduled nightly.yml. The name collision confuses GitHub's "list workflow runs by file name" REST query, so release tooling querying actions/workflows/nightly.yml/runs got a stale/partial set missing recent scheduled runs. Rename it to facebook/rocksdb/nightly-candidate, matching the pr-jobs-candidate convention, so every workflow name is unique.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15345
Test Plan: rety the integration after fix
Reviewed By: xingbowang
Differential Revision: D123973936
Pulled By: pdillinger
fbshipit-source-id: cde1c238e6a91444f7112848081e161f54f5bf24
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15331
## Motivation
The parent diff already implements cross-DB flushable-memory accounting and a background sorter that ranks DBs outside the write path. This change does not alter how the best candidate is selected.
The remaining liveness problem occurs after the sorter selects a DB and hands it off for execution. Previously, the coordinator tracked one remote handoff as pending and waited for that DB to report completion before advancing. A selected DB could therefore monopolize coordinator progress if its job waited in Env::LOW, blocked on the DB mutex, waited behind an existing write leader or write stall, or otherwise failed to complete promptly. Other ranked DBs could be ready to reclaim memory but would not be attempted while the first handoff remained pending.
## Solution
Make cross-DB flush handoff execution bounded and bypassable.
Cross-DB attempts now run on a small four-worker executor owned by the WriteBufferManager instead of consuming Env::LOW compaction capacity. The executor is created only for kFlushLargestAcrossDBs and its workers are retired when that policy is disabled.
The coordinator distinguishes queued and executing handoffs. A candidate lease begins when an executor worker starts the callback, rather than while the callback is merely waiting in the executor queue. If an executing handoff exceeds its lease, the coordinator releases ownership of that attempt and advances to the next ranked candidate. The original callback is not cancelled and may still complete, so the executor and outstanding-handoff count provide a strict bound on concurrent stuck work.
Each registration has a per-DB outstanding-job reservation, preventing the sorter from dispatching the same DB again while an earlier callback remains alive. The coordinator also records the identity of the active candidate, so late completion from an expired attempt cannot clear or corrupt a newer candidate lease.
DB mutex acquisition, write-thread joining, memtable switching, and flush dispatch continue to happen outside registry and coordinator locks. The executor invokes the existing BackgroundCallWBMFlush path directly; actual immutable-memtable flushing still uses RocksDB normal flush infrastructure.
## Compatibility
This behavior is limited to kFlushLargestAcrossDBs. Existing policies retain their previous local flush behavior, and no WBM executor threads are started for them.
## Validation
Unit coverage exercises lease expiry, advancement to the next candidate, late completion, duplicate suppression, bounded worker capacity, policy changes with blocked callbacks, shutdown, and reranking after progress.
The end-to-end stack was also compared against the original local policy in one matched memory-pressure run with 1,000 DBs and 16 writers. The complete largest-across-DBs policy delivered 61.0% higher throughput, 55.5% fewer WBM flushes, 37.9% lower mean latency, 47.9% lower P99 latency, 5.17x larger average flushes, and 62.8% lower peak RSS. These measurements validate the full policy stack rather than isolating the performance contribution of this liveness follow-up.
Reviewed By: xingbowang
Differential Revision: D123461329
fbshipit-source-id: 4ad86102c2a43beb469f7672d722674dfbda9871
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15342
Add an ExternalTableBuilderBase::TryAdd() contract so external table builders can distinguish accepted rows, real builder errors, and clean requests to cut the current output table and retry the same row.
Add external-table-facing result, reason, and context types. Existing external table builders inherit a default TryAdd() implementation that forwards to Add(), so current implementations keep their historical behavior unless they explicitly opt into retryable file cuts.
Reviewed By: xingbowang
Differential Revision: D123930815
fbshipit-source-id: b5701c704938759db68b7c0bf0ebeba19d176945
Summary:
Motivation
A WriteBufferManager can be shared by many RocksDB instances and column families. Under the existing policy, the DB whose writer first observes memory pressure flushes its own oldest mutable memtable. That DB might own only a small fraction of the reclaimable memory, so workloads with many DBs or CFs can generate many small flushes while the largest memory consumer remains untouched. Scanning every registered DB while holding process-wide or DB mutexes would improve selection accuracy but would serialize unrelated write paths and is not viable for deployments with thousands of DBs.
Previous behavior
kFlushOldest was the only behavior: the write-group leader that noticed WBM pressure synchronously selected the oldest eligible mutable memtable in its own DB. The WBM had no per-DB view of reclaimable memory and could not direct reclamation to another DB sharing the same budget.
New behavior
This change adds configurable WriteBufferFlushPolicy modes:
- kFlushOldest preserves the existing behavior and remains the default.
- kFlushLargest selects the largest eligible flushable column family in the writing DB.
- kFlushLargestAcrossDBs selects the DB advertising the most reclaimable memory across all DBs sharing the WBM, then flushes its largest eligible CF. Atomic-flush DBs advertise and flush the aggregate eligible memory across their CFs.
High-level implementation
Each writable DB registers a lightweight FlushInitiator with the shared WBM. Per-DB accounting publishes two approximate values using atomics: total mutable memory and the largest flushable CF contribution, including eligible immutable memtables waiting for flush. Allocator reporting is batched, so the allocation hot path avoids registry scans and DB mutexes.
A dedicated WBM coordinator thread maintains an approximate ranked snapshot of registered DBs. Refreshes are driven by accounting changes, threshold crossings, and periodic pressure checks. Ranking and O(number of DBs) work happen on that coordinator thread; writers only update counters and signal the coordinator. Registration lifetime is protected independently so DB close can safely drain an in-flight handoff.
Threading and flush handoff
The coordinator reserves and pins one ranked candidate at a time, without holding a registry lock while invoking the DB. The target DB uses TryLock on its DB mutex and rejects the handoff if it is closing, paused, stopped, stalled, already handling a WBM job, or no longer has flushable memory. Rejected candidates do not block consideration of the next DB.
An accepted handoff queues one LOW-priority WBM job. The LOW worker joins the target DB write thread and revalidates eligibility before switching the memtable. The resulting ordinary flush is then scheduled through RocksDB's existing flush machinery on the HIGH-priority flush pool. Keeping the write-thread join off the requesting writer avoids making one DB's write path wait on another DB's mutex or write queue.
Only one manager-wide remote handoff is active at a time. Completion or cancellation clears the handoff and requests a rerank. If memory reaches the hard limit while remote coordination is delayed, a rate-limited local fallback lets a writer seal a local memtable to preserve forward progress. Write stalls, shutdown, WaitForCompact, policy changes, and DB deregistration explicitly drain or reject outstanding handoffs.
Compatibility
The legacy constructor and kFlushOldest behavior remain unchanged unless a new policy is explicitly selected. There are no on-disk format changes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15047
Reviewed By: xingbowang
Differential Revision: D112396106
fbshipit-source-id: 3049adc7034dd4e00d74d7d86e1d19004292d299
Summary:
Follow-up to https://github.com/facebook/rocksdb/issues/15304, which renamed `TableProperties::creation_time` to `oldest_ancestor_time` but left the long-standing "ancester" misspelling everywhere else, so the same concept was spelled two ways (e.g. `TableProperties::oldest_ancestor_time` vs. `FileMetaData::oldest_ancester_time` and `SstFileMetaData::oldest_ancester_time`). This change makes the spelling consistent and addresses remaining review feedback on https://github.com/facebook/rocksdb/issues/15304.
Change:
* All internal identifiers and comments now use "ancestor": `FileMetaData::oldest_ancestor_time`, `CompactionServiceOutputFile::oldest_ancestor_time`, `kUnknownOldestAncestorTime`, `TryGetOldestAncestorTime()`, `Compaction::MinInputFileOldestAncestorTime()`, the `kOldestAncestorTime` manifest tag enum, and related locals and test sync point names.
* Public API: `SstFileMetaData::oldest_ancestor_time` is the new name; `oldest_ancester_time` remains as a deprecated alias sharing storage via an anonymous union (same approach as `TableProperties::creation_time`, so size, layout, and copy semantics are unchanged).
* Public API: `TablePropertiesNames::kOldestAncestorTime` is the new name; `kCreationTime` remains as a deprecated `static const std::string&` alias. A static reference member has no per-object cost and binds at constant initialization, so it adds no static-init-order exposure.
* `TablePropertiesNames` as a whole is marked DEPRECATED for public use. An fbsource-wide BigGrep found no references outside RocksDB and its vendored copies, and it is only meaningful to table format internals; users should read the corresponding `TableProperties` fields.
Fix:
* The `table_file_creation` event log entry had the `oldest_key_time` and `newest_key_time` values swapped since https://github.com/facebook/rocksdb/issues/13083 (v10.0). Now each label is paired with its own value.
Also: tools/c_api_gen/auto_simple_bindings.py resets the pending anonymous union field list after use and skips unnamed members inside the union (review suggestion); generated C API output is unchanged.
Compatibility: names that are persisted or externally parsed intentionally keep the old spelling: the on-disk property name `rocksdb.creation.time`, the `"oldest_ancester_time"` key in remote compaction serialization, and the `oldest_ancester_time` / `OldestAncesterTime` labels in `VersionEdit` DebugString/DebugJSON (ldb manifest_dump output). The manifest tag value is unchanged. HISTORY.md is left as historical record.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15335
Test Plan: DBTablePropertiesTest.CreationTimeAliasesOldestAncestorTime now also verifies `TablePropertiesNames::kCreationTime` and `kOldestAncestorTime` refer to the same object, and new DBTablePropertiesTest.SstFileMetaDataOldestAncesterTimeAlias verifies the two `SstFileMetaData` names share an address and that writes through the old name are visible through the new one. Remote compaction serialization through the unchanged `"oldest_ancester_time"` key is covered by compaction_service_test; manifest encode/decode through the renamed field by version_edit_test. `make check-c-api-gen` confirms the generated C API is up to date and backward compatible.
Reviewed By: jaykorean
Differential Revision: D123559299
Pulled By: pdillinger
fbshipit-source-id: 480106e8d539e654f31db210ef6e1de02325d604
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15334
External table formats cannot currently persist range tombstones, so SST ingestion and full-mode DB flushes or compactions fail when they contain `DeleteRange` entries. This change adds an opt-in contract for storing RocksDB’s canonical range-deletion block. RocksDB retains responsibility for sequence-number handling, fragmentation, and visibility through its native range-tombstone iterators.
Existing factories remain unsupported by default, and existing files without range tombstones are unaffected. Missing or invalid advertised tombstone blocks are treated as corruption. RocksDB also supplies logical entry counts and sizes when a format does not persist the properties block, while preserving format-specific properties.
Reviewed By: xingbowang
Differential Revision: D123475827
fbshipit-source-id: de8fec401bdc311282f5d4a48b248875c682f7c9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15321
`MultiGet` with `read_tier = kBlockCacheTier` now returns `Incomplete` for data blocks that are not in the block cache. Before this fix, it read them from disk.
**Why it matters:** Callers use `kBlockCacheTier` to prevent disk I/O. `Get` obeyed it, but `MultiGet` did not.
**Problem**
- Without mmap reads, `MultiGet` sends cache misses to `RetrieveMultipleBlocks`.
- That path called `MultiRead` with no `kBlockCacheTier` check.
- `RetrieveBlock` (used by `Get`) has the check.
**Fix**
- `RetrieveMultipleBlocks`: if `read_tier == kBlockCacheTier`, set `Incomplete("no blocking io")` for each block and do no I/O.
- `MultiGet`: when a block is `Incomplete`, return that status for the key, the same as `Get`. Before, the key status stayed `OK` with no value.
- Release note in `unreleased_history/bug_fixes`.
**Tests**
- New `DBTest2.MultiGetBlockCacheTierSkipsColdSstData`: `Get` and `MultiGet` both return `Incomplete` for a key whose data block is not in the cache.
- `DBBlobBasicTest.MultiGetBlobsFromCache` relied on the bug. Its inline key passed only because `MultiGet` read the data block from disk. The test now loads the block cache first, so it still checks the blob cache.
Reviewed By: joshkang97
Differential Revision: D122431823
fbshipit-source-id: 262913930cb4b570ea0a4d8f4f92364de08a27f5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15320
Why the change is required:
`DB::ClipColumnFamily(cf, begin_key, end_key)` keeps `[begin_key, end_key)`. It first drops whole non-L0 files outside the range with `DeleteFilesInRanges` on `(-inf, begin_key)` and `[end_key, +inf)`, but called it with the default `include_end = true`. The first range was therefore treated as `(-inf, begin_key]`, so a non-L0 file whose largest key is `begin_key` was deleted even though that key is inside the kept range. Clipping `[b, d)` on L1 files `[a, b]` and `[c, d]` lost `b`.
This showed up while smoke-testing `SplitDB` (D120757697): single-key files trivially moved to the bottom level made the split destination miss its first key.
New behavior:
- Pass `include_end = false`, so `DeleteFilesInRanges` skips files whose largest key equals `begin_key`. The second range has no end, so it is unaffected. The remaining keys below `begin_key` are still removed by the existing `DeleteRange` and compaction steps.
Reviewed By: xingbowang
Differential Revision: D123306376
fbshipit-source-id: 65cb36ae730a55750ea558a6cdcd48d8fc17f0e9
Summary:
The `creation_time` table property has long stored the file's oldest ancestor time (the oldest key time among the file's ancestors, falling back to flush/compaction time), not the file's creation time, which is `file_creation_time`. A TODO in table_properties.h acknowledged the misleading name. This change starts the API migration: `TableProperties::oldest_ancestor_time` is the new name, and `creation_time` remains as a deprecated alias sharing storage via an anonymous union, so existing external reads and writes continue to compile and behave identically.
An anonymous union is used rather than a `uint64_t&` alias member. A reference member would delete TableProperties copy/move assignment, make the implicit copy constructor bind the copy's alias to the source object (dangling), permit writes through a const TableProperties, add 8 bytes, and break the uint64_t-range walk in TEST_SetRandomTableProperties. The union leaves size, layout, copying, constness, and offsetof unchanged. Reading the other union member is formally UB in ISO C++, but both members are uint64_t and this is supported in practice by GCC, Clang, and MSVC.
All in-repo uses of the old name now use `oldest_ancestor_time`. Names that are externally observable or persisted are intentionally unchanged:
* On-disk property name `rocksdb.creation.time` (`TablePropertiesNames::kCreationTime`)
* The `"creation_time"` key in TableProperties serialization (used by remote compaction)
* The event-log JSON key and `TableProperties::ToString()` label
* Java `TableProperties.getCreationTime()` and C `rocksdb_table_properties_creation_time()`
C API: the auto-binding generator (tools/c_api_gen/auto_simple_bindings.py) only considered direct FieldDecls and failed on the anonymous union's implicit unnamed field. It now treats anonymous union members as direct fields. Regenerating adds a new C getter
`rocksdb_table_properties_oldest_ancestor_time()` alongside the existing `rocksdb_table_properties_creation_time()`.
Follow-up: remove the deprecated `creation_time` alias in a future release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15304
Test Plan: New unit test DBTablePropertiesTest.CreationTimeAliasesOldestAncestorTime verifies the two names share an address and that a write through the old name is visible through the new one. options_settable_test and compaction_service_test cover TableProperties serialization round trips through the unchanged "creation_time" key, now mapped to the renamed field.
Reviewed By: jaykorean
Differential Revision: D123079083
Pulled By: pdillinger
fbshipit-source-id: 18b07d91f303a8f6de58da0c730f80d8c371505c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15287
`IngestExternalFileOptions::write_global_seqno` has been deprecated since RocksDB 8.0 and will soon be removed from the API. It requires ingestion to mutate SST files, sync them, regenerate checksums, and expose file-offset plumbing solely for that mutation.
Remove the option and its C, Java, and `ldb` APIs. Assigned sequence numbers now live only in the MANIFEST, leaving newly ingested SST bytes and whole-file checksums unchanged.
## Backward Compatibility
New readers continue to support existing SSTs containing a nonzero global sequence number written by legacy ingestion. Because that old write did not update the properties-block checksum, checksum verification retries against a temporary copy with only the global-seqno value restored to its original zero placeholder. The parsed sequence number and file contents remain unchanged.
This fallback only accepts mismatches explained by the legacy rewrite. Other checksum failures continue through normal corruption handling, and `verify_checksums=false` continues to bypass verification.
Reviewed By: xingbowang
Differential Revision: D122405856
fbshipit-source-id: 934e9bc755d58ea8d84e1c085e9769015bd8dfd7
Summary:
CONTEXT: The AI review workflow had generic concurrency and performance checks, but it did not require reviewers to determine whether changed code or transitive callees execute while DBImpl::mutex_ is held. Reviews could therefore miss added contention, blocking work, lifetime errors, or snapshot and lock-ordering violations.
PREVIOUS: Review agents could examine DB-mutex changes through general correctness and performance lenses, but there was no explicit call-path audit, before/after mutex-held work comparison, or evidence requirement for regressions.
NEW: Every non-trivial review must state whether the DB mutex is affected. When it is affected, reviewers must inspect lock boundaries and correctness invariants, compare CPU and I/O before and after the change, keep unavoidable locked work highly optimized, and require measurements for material regressions.
IMPLEMENTATION: Add a DB Mutex Review Perspective covering locking contracts, object lifetime, snapshot boundaries, lock ordering, hidden callee costs, blocking work, validation, and performance evidence. Assign its checks to the correctness and performance reviewers and add required context and review checklist entries.
SAFETY: This changes only the AI code-review instructions. It does not change RocksDB runtime behavior, APIs, file formats, or compatibility. Changes that cannot execute under the DB mutex only need to state that the mutex is not affected.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15324
Test Plan:
- `git show --check HEAD`
- `git diff --check upstream/main...2026_10_05_main`
- Documentation-only change; no runtime tests were run.
Reviewed By: joshkang97
Differential Revision: D123415306
Pulled By: xingbowang
fbshipit-source-id: 0f78e0b1d677edf003cd5dc6ed5ff8123aff4f98
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15285
Why the change is required:
MyRocks bulk-load v2 creates a temporary RocksDB and calls DestroyDB when the load is committed, rolled back, abandoned, or recovered. Successfully ingested SSTs are moved into the main DB, so leftover SST cleanup is primarily an error, abort, or recovery case; WALs, obsolete files, and compaction-created trash can still remain. All physical deletion performed by DestroyDB should honor the configured SstFileManager rate, including rocksdb_sst_mgr_rate_bytes_per_sec in MyRocks.
Previous behavior:
DestroyDB sent recognized table, blob, and WAL files through SstFileManager, but ignored filenames ending in .trash because ParseFileName rejects that suffix. The final best-effort DeleteDir then silently left the directory behind. DB::Open handled the same files through CleanupDirectory, forcing MyRocks cleanup code to discover and register trash itself. A global WaitForEmptyTrash workaround would also drain unrelated compaction, DROP TABLE, and DROP INDEX deletion work from every DB sharing the manager.
New behavior:
DestroyDB waits only for pending deletion work associated with each DB, data, WAL, and archive directory. It then schedules any remaining .trash file through the existing rate-limited scheduler and waits only for the private bucket created for this DestroyDB call. Directory keys are normalized for bookkeeping, so equivalent path spellings identify the same pending work while filesystem operations retain the original path.
Accounting and failure handling:
A failed accounted trash deletion retains its original file path and remaining trash-byte charge. If DB open or DestroyDB later discovers the file, the retry preserves the accounted classification, avoids charging the trash bytes twice, and clears SstFileManager tracked-size and trash-size accounting after successful deletion.
Scope and safety:
There is no new sysvar, public RocksDB API, or persisted-format change. Existing rate settings, hard-link handling, partial deletion, and separate-WAL foreground deletion remain unchanged. The bookkeeping is confined to file enqueue/completion and DestroyDB cleanup. A shared FIFO can still delay this DB behind unrelated work already ahead of its files, but unrelated work queued behind this DB no longer gates DestroyDB.
Reviewed By: xingbowang
Differential Revision: D122221503
fbshipit-source-id: ec106f57efa70e24827234b4173366e5dbcbda5c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15293
Why the change is required:
Full background-error recovery flushes all column families because WAL durability can be uncertain. A later recovery-flush failure or process crash must not persist different WAL replay boundaries for different column families. Otherwise reopen can recover only part of a write batch spanning those column families.
The risk includes non-file-scoped WAL write or sync errors. A failed write can have reached the WAL without reaching the memtables. An earlier unsynced write can instead remain in the memtables while its WAL bytes are not durable; retrying the fsync after resetting the writer error does not by itself prove those bytes survived. Sequential recovery can make either case visible in only some column families after reopen.
Previous behavior:
File-scoped WAL-error recovery already set FlushOptions::force_atomic_flush. Other full kErrorRecovery paths flushed column families sequentially when DBOptions::atomic_flush was false. Each successful per-column-family flush independently advanced its MANIFEST log number, so a later flush failure or process crash could leave mismatched replay boundaries.
New behavior:
Every full kErrorRecovery atomically commits the flush results and WAL replay boundaries for all participating column families. A successful recovery commits every column family; a failed recovery commits none. If an atomic recovery attempt fails, all of its flush output is discarded and recreated by a later attempt instead of retaining output from column families that finished first.
Implementation:
Set FlushOptions::force_atomic_flush for every full error-recovery FlushAllColumnFamilies call. Remove the now-unreachable recovery-only behavior from FlushMemTable and assert that recovery does not enter the non-atomic path. Keep the two-write-queue sequence fence at the atomic memtable-switch point.
Add an end-to-end fault-injection test for a retryable, non-file-scoped WAL sync error after a multi-column-family record is appended. It covers successful and failed recovery flushes, including a column family with no other memtable data, and verifies all-or-none behavior after reopen. Update the manual_wal_flush comment and add a release note.
Safety, compatibility, and limitations:
Only full kErrorRecovery changes. Normal, manual, shutdown, and kErrorRecoveryRetryFlush behavior is unchanged. File-scoped WAL-error recovery was already atomic. There are no API, ABI, option-default, or file-format changes.
The additional work is limited to the cold recovery path. A failed atomic recovery can redo more flush work than the previous sequential path, but this avoids partial multi-column-family recovery. Automatic recovery with manual_wal_flush remains disabled as fatal pending separate analysis and testing; this change does not remove that workaround. Validation covers local Linux builds and focused recovery tests; platform CI for this version is pending.
Reviewed By: pdillinger, rban1
Differential Revision: D122469648
fbshipit-source-id: f72845903d9b6bff70f5d881e02a4cb2e920ed72
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15294
Why the change is required:
`open_files_async` reduces startup latency by moving SST validation and table-reader opening to background work, but `DB::OpenAsSecondary()` still performed that work synchronously.
Previous behavior:
A secondary using `open_files_async=true` replayed its MANIFEST through `ReactiveVersionSet`, synchronously verified and opened every current SST, and only then returned. The option therefore provided no startup benefit for secondary instances.
New behavior:
`DB::OpenAsSecondary()` can now return after installing its initial read view and start initial SST validation and table-reader pinning through the existing asynchronous opener. With `max_open_files=-1`, the opener processes all current SSTs in the background; a finite table cache can stop the eager pass at capacity and validate remaining files on demand. Later `TryCatchUpWithPrimary()` rounds remain synchronous. Async-open failures notify listeners and are latched according to the existing severity and `paranoid_checks` policy. If an error is not latched, a later read retries the file access and reports a persistent failure. A latched error requires fixing the underlying problem and reopening the secondary. Read-only instances do not start write-oriented recovery.
Implementation:
Initial reactive MANIFEST replay can defer SST verification and handler loading. The deferral is immutable constructor state on `ReactiveVersionSet`, is cleared before catch-up mode, and is restricted to normal secondary recovery. `VersionBuilder::GetAndClearIntermediateFiles()` now moves out and clears its own historical add-then-delete bookkeeping so callers cannot accidentally retain stale state. Secondary open schedules the background opener after publishing the read view. Its context pins the captured Versions and column families, including retained secondary handles whose column family was subsequently dropped by the primary. Existing table-reader opening performs SST size and unique-ID validation, avoiding a separate serial metadata pass. Closing the DB or canceling background work safely cancels a queued opener or waits for a running opener. Read-only async-open errors use an ErrorHandler path that notifies listeners without automatic recovery. Secondary async open explicitly rejects FIFO compaction.
Safety, compatibility, and limitations:
The default `open_files_async=false` path and existing primary async-open behavior are unchanged. WAL and blob recovery, as well as later secondary catch-up, remain synchronous. Follower DB and `OpenAndCompact()` do not use this deferral. There are no file-format, ABI, or public API signature changes. Secondary async open requires `skip_stats_update_on_db_open=true` and does not support FIFO compaction. Applications must keep files accessible by coordinating deletion with the primary or by using an Env/FileSystem that delays physical deletion until every instance releases a file. With asynchronous opening, `max_open_files=-1` cannot guarantee every SST is pinned before `OpenAsSecondary()` returns; the table-files-only `max_open_files=-1` workaround applies only when asynchronous opening is disabled. No completion API is added.
Validation covers initial deferral and pinning, synchronous catch-up, table-reader detection of physical-size mismatch, FIFO rejection, listener-adjusted errors and synchronous read retry, retained handles after primary column-family deletion, close and explicit cancellation with no HIGH-priority worker, stopping new table opens during shutdown, primary async open, follower isolation, and remote compaction. Linux Buck builds and focused tests passed. Cross-platform CI is pending, and `make check-sources` cannot run successfully in this Sapling checkout because its script requires Git.
Reviewed By: pdillinger
Differential Revision: D122417547
fbshipit-source-id: 91d408469c35c4ea21165c5e7db75068e65c2389
Summary:
The persistent injected-error log writes `InjectedErrorLog::Entry` objects using their native in-memory layout, while the log format's consumer expects little-endian multi-byte fields. The existing `static_assert(port::kLittleEndian)` protected that format by rejecting every big-endian build that included the fault-injection filesystem, even though the log is an optional diagnostic.
This change keeps RocksDB and the fault-injection filesystem available on supported big-endian platforms while disabling only the incompatible persistent log. Log-file creation now returns an inactive descriptor when `port::kLittleEndian` is false, which makes `Record()` and `Flush()` remain safe no-ops. The fixed 256-byte entry-size assertion remains in place, and format-dependent tests are bypassed on big-endian targets.
The change also adds a release note and documents big-endian support plus the preferred capability-check pattern for optional architecture-specific features.
## Compatibility and limitations
- Little-endian behavior and the on-disk log layout are unchanged.
- On big-endian platforms, configuring an injected-error log path does not create or populate a log file; the rest of fault injection remains available.
- Windows already leaves this POSIX-backed diagnostic disabled.
- This does not introduce a portable log encoding. Supporting persistent injected-error logs on big-endian platforms would require a format change.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15297
Test Plan:
- `build_tools/rockstest.sh fault_injection_fs_test '--gtest_filter=InjectedErrorLogTest.*'` (little-endian macOS): all 5 tests passed.
- `make check-sources`
- `git diff --check upstream/main...2026_10_01_endian_fix`
Native big-endian build and test execution were not available locally; that path relies on the compile-time `port::kLittleEndian` capability and should be covered by big-endian CI.
Reviewed By: jaykorean
Differential Revision: D122880760
Pulled By: xingbowang
fbshipit-source-id: 59dae9661a47608aa12f41078ad1b3ef0fa7231e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15269
Why the change is required:
Transaction entity reads must enforce the same commit visibility and read-your-own-write rules as value-only reads. Without the transaction-specific read callback, GetEntity-family APIs can return a value that is not visible to the caller.
Previous behavior:
TransactionBaseImpl handled GetEntity and MultiGetEntity through non-virtual helpers that called WriteBatchWithIndex without a ReadCallback. WritePrepared transactions could expose a prepared but uncommitted update. WriteUnprepared transactions could miss their own write after it had been flushed as an unprepared batch, although that write remained hidden from other transactions.
New behavior:
GetEntity, MultiGetEntity, and GetEntityForUpdate now follow the same snapshot visibility and read-your-own-write semantics as Get and MultiGet for WritePrepared and WriteUnprepared transactions.
Implementation:
Make the internal entity-read helper hooks virtual. Override them in WritePreparedTxn and WriteUnpreparedTxn to construct the existing policy-specific ReadCallback and pass it through WriteBatchWithIndex. The WritePrepared batched path validates the callback refreshed read sequence so reads without an explicit snapshot do not bypass eviction validation. Invalid callback or snapshot state returns TryAgain. Add shared transaction entity-read assertions and parameterized regression coverage for prepared visibility and unprepared read-your-own-write.
Safety boundaries and unsupported cases:
There is no public API or persisted-format change. Other transaction policies keep the existing base implementation. The change adds only the callback and validation already used by value-only reads. Validation covers synchronous snapshot and non-snapshot entity reads, prepared visibility, and unprepared read-your-own-write in normal and checked-Status builds. Coroutine-specific and named wide-column payload cases were not separately tested, and the full repository suite was not run.
Reviewed By: rban1
Differential Revision: D121793013
fbshipit-source-id: e300642ed360b640814056e4d95cced5f5c180d8
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15274
Expose `SerializePreparedFileInfo()` and `DeserializePreparedFileInfo()` so opaque metadata produced by SST writers or DBs can be persisted and transferred across process boundaries. The representation uses RocksDB `OptionTypeInfo` adapters for `InternalKey` and `TableProperties` and includes a CRC32C integrity check.
Deserialization now rejects malformed nested string-map entries, including missing key/value separators and invalid hex, with `InvalidArgument` instead of asserting in debug builds or accepting malformed input in release builds. Valid serialized data and the wire format are unchanged. The `PreparedFileInfo` type metadata is also initialized on first use to avoid throwable namespace-scope initialization.
Reviewed By: xingbowang
Differential Revision: D121897852
fbshipit-source-id: 59a3d221bfd20e6c257bd60477823a12d05a35c5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15296
**Why:** trace replay (`DB::NewDefaultReplayer()`, db_bench `replay`) cannot replay a trace of a DB with only the default column family, and multi-threaded replay reports success when records fail. Together these make db_bench replay look successful while replaying no reads.
**Previous behavior:**
- `TraceExecutionHandler` resolves a record's column family ID only against the `handles` passed to `NewDefaultReplayer()`. Records on a column family missing from `handles`, including the default one, fail with `Corruption: Invalid Column Family ID.`. An empty `handles` also trips `assert(!handles.empty())` in debug builds.
- db_bench fills its handle list (`db_.cfh`) only when `--num_column_families` > 1, so with the default of 1 it passes an empty list. Every Get, MultiGet and iterator record fails. Write records are not affected, because `DB::Write()` resolves the column family from the batch itself.
- In multi-threaded `Replay()` (`num_threads` > 1), `BackgroundWork()` sends only decode errors to the error callback. Execution errors are dropped, so `Replay()` returns OK. Single-threaded replay stops and returns the error, including a NotSupported from execution (for example `DB::Write()` of a DeleteRange with `row_cache` set). It skips only records whose trace type it cannot decode.
- Result in db_bench on a single column family DB: `--trace_replay_threads=1` prints `Replay failed. Error: Corruption: Invalid Column Family ID.`. `--trace_replay_threads=4` prints `Replay completed` with `rocksdb.number.keys.read` = 0.
**New behavior:**
- The replay handler always includes the DB's default column family, so `handles` may be empty or leave it out. A handle the caller passes for a column family is still the one used.
- Multi-threaded `Replay()` reports execution errors through the same callback as decode errors. It stops scheduling new records and returns the error of the failed record with the earliest trace timestamp, as the existing code already does for decode errors. Get records that find no key still count as success, since the handler treats NotFound as OK.
- As in single-threaded replay, NotSupported is skipped only when decoding returns it (a trace type the replayer does not support). NotSupported from executing a record stops the replay.
- The scheduling loop no longer reads the shared status `bg_s` while worker threads write it. It already did so for decode errors; with execution errors reported, TSAN flags it in the new test (see Test Plan).
**Implementation:**
- `TraceExecutionHandler` constructor: drop the non-empty assert; after the caller's handles, insert `db->DefaultColumnFamily()`. `unordered_map::insert` does not overwrite, so a caller's handle wins.
- `TraceExecutionHandler` constructor: skip the default column family if `DefaultColumnFamily()` returns nullptr (some test `DB` mocks do), so such DBs behave as before.
- `ReplayerImpl::BackgroundWork()`: call `error_cb` when `Accept()` returns an error. On a decode error, skip `error_cb` for NotSupported, matching single-threaded replay.
- `ReplayerImpl::Replay()`: `error_cb` no longer filters out NotSupported (the decode path now does that); it still keeps the error of the earliest record.
- `ReplayerImpl::Replay()`: `error_cb` also sets an `std::atomic<bool> bg_failed` while holding `mtx`, and the scheduling loop checks that flag instead of `bg_s.ok()`. `bg_s` is read only after `WaitForJobsAndJoinAllThreads()`.
- Document both contracts in `db.h` (`NewDefaultReplayer`), `trace_record.h` (`NewExecutionHandler`) and `replayer.h` (`Replay`).
- Release notes in `unreleased_history/bug_fixes/`.
**Safety, compatibility, limitations:**
- No API or ABI change.
- Callers that already pass the default column family handle see no change from the first fix.
- Behavior change: multi-threaded replay callers that got OK while records failed now get the error, including NotSupported from execution. That matches single-threaded replay and the new `Replay()` comment.
- A record read by the scheduling loop while a failing worker is setting `bg_failed` may still be scheduled. This is allowed by the `Replay()` comment (records already scheduled still run) and does not change the returned status, which is that of the earliest failed record.
- Cost: one extra map entry per replayer, and one branch per record in threaded replay. The loop's `bg_s.ok()` read becomes a relaxed atomic load.
- Read records on column families that are neither passed nor the default still fail with Corruption, as before. Write records never used `handles` and are unchanged.
Reviewed By: xingbowang
Differential Revision: D122626982
fbshipit-source-id: 9a0973448710bac25233feedbaf264696ac85035
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15291
Add experimental coroutine interfaces for implementing asynchronous basic and full `ExternalTableReader`s. Implementations opt in through `CoroExternalTableReaderBase<Mode>` by providing `GetCoroutine()` and may override `MultiGetCoroutine()`; the default implementation runs the individual Get operations concurrently. The external-table adapter uses these APIs when supported and otherwise retains the existing synchronous behavior.
As a smaller supporting fix, Single Get coroutine reads using non-mmap, non-direct I/O on a kFSBuffer-capable `FileSystem` now use `FSRandomAccessFile::SubmitReadAsync()` instead of falling back to synchronous `Read()`.
Reviewed By: xingbowang
Differential Revision: D122187491
fbshipit-source-id: 217e2953756938bbd3f30358a997a7ca3ed5543e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15268
Why this is required: remote file systems need an immutable blob-file checksum and checksum-function name when RocksDB opens a finalized blob file. This lets the file system distinguish file generations and safely validate or key cached extents.
Previous behavior: integrated BlobDB opened normal read-path blob files with the column family's ordinary FileOptions. Although finalized checksum metadata was stored in BlobFileMetaData and BlobFileAddition, it was not propagated through point reads, range reads, MultiGet, retry opens, or recovery verification. Delayed reads of direct-write files could also cache a finalized file reader opened without metadata.
New behavior: every finalized integrated BlobDB file open receives its checksum value and checksum-function name through FileOptions. Active direct-write files continue to use empty checksum fields until finalization. Delayed reads pinned to a pre-flush SuperVersion use the finalized metadata once the file has been sealed.
Implementation: add internal BlobFileOpenInfo transport across Version, BlobSource, and BlobFileCache. Populate it from version metadata and recovery additions. At the actual cache miss/open, copy the base FileOptions and set both checksum fields. Direct-write sealing registers owned checksum metadata with BlobFileCache, atomically evicts footer-less readers, and prevents a racing pre-seal refresh from repopulating the cache without metadata. Metadata is released after write-path ownership and old-memtable protection end.
Safety boundaries: this does not change public APIs, persisted formats, or blob cache keys. Cache hits do not copy checksum strings. Files with no generated checksum, including active direct-write files, retain empty checksum fields. Existing footer-skip, corruption, and reader-refresh behavior remains in place. Integration in downstream file systems is outside this change.
Tests and known limitations: filesystem-capture tests cover regular blob reads and delayed direct-write reads. After rebasing onto the latest master, blob_file_cache_test, blob_source_test, db_blob_basic_test, and db_blob_direct_write_test passed both normally and with ASSERT_STATUS_CHECKED=1 (95 tests in each configuration). Fast arc lint passed. make format-auto and make check-sources could not operate because their scripts assume a Git checkout while this checkout uses Sapling; all files in the patch were formatted directly with clang-format.
Reviewed By: chaitanyadatye, anand1976
Differential Revision: D121630172
fbshipit-source-id: 601d1097e4a3caa60f0d9ba1794e78e81855980c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14924
Widen the serialized "counts" array in CompactionServiceResult from kNumOfReasons - 1 back to the full kNumOfReasons, so the count for the newest CompactionReason is included in the remote compaction result.
The width was previously reduced by one to match readers that expected the narrower array; restoring it changes what is written. This is safe only because deserialization of "counts" is now size-tolerant (a reader expecting fewer elements drops the extra, one expecting more zero-fills the missing). **It must not land until that size-tolerant read D106321320 has been deployed to every primary, including rollback targets, so should be >= 2 releases ahead; otherwise a primary can still or fallback to running the strict reader would reject the wider result and fail deserialization of this field.**
== Test plan ==
**Legend**
CompactionReason is an enum whose last entry is kNumOfReasons (currently 21). CompactionStats::counts is an int[kNumOfReasons] holding one count per reason; it is serialized into the remote compaction result and read back by the primary.
- "narrow" = kNumOfReasons - 1 counts (omits the newest reason); what current code writes.
- "wide" = kNumOfReasons counts (includes the newest reason).
- strict reader: accepts only its exact expected element count; any other count fails deserialization (the plain OptionTypeInfo::Array behavior).
- tolerant reader: accepts any count -- more elements than expected are dropped, fewer are zero-filled; never fails on a count mismatch (this change, SizeTolerantDeserializeArray).
**Unit test**
CompactionServiceTest.CompatCheckCountsWidthTolerance -- verifies that deserializing "counts" tolerates a serialized element count equal to, one greater than, or one less than the reader's expected width (extras dropped, missing zero-filled), and that the binary serializes exactly the expected number of elements.
**Manual cross-version verification (ldb remote_compaction)**
Test 1: expect-narrow tolerant-read primary reads wide counts a worker write (extra reason dropped)
```
$ tolerant_wide remote_compaction_worker --db=$DB --job_dir=$JOB &
$ tolerant_narrow remote_compaction_primary --db=$DB --job_dir=$JOB
remote_compaction_worker: OK (4915 bytes result)
remote_compaction_primary: OK
$ grep 'remote compaction result' $DB/LOG
[default] [JOB 3] Received remote compaction result, output path: /tmp/cs_run/output, files: 000022.sst
```
=> PASS: the primary parsed and installed the remote result.
Test 2: expect-wide tolerant-read primary reads narrow counts a worker writes (missing reason zero-filled)
```
$ strict_narrow remote_compaction_worker --db=$DB --job_dir=$JOB &
$ tolerant_wide remote_compaction_primary --db=$DB --job_dir=$JOB
remote_compaction_worker: OK (4911 bytes result)
remote_compaction_primary: OK
$ grep 'remote compaction result' $DB/LOG
[default] [JOB 3] Received remote compaction result, output path: /tmp/cs_run/output, files: 000022.sst
```
=> PASS: the primary parsed and installed the remote result.
Test 3: expect-wide tolerant-read primary reads wide counts a worker writes
```
$ tolerant_wide remote_compaction_worker --db=$DB --job_dir=$JOB &
$ tolerant_wide remote_compaction_primary --db=$DB --job_dir=$JOB
remote_compaction_worker: OK (4917 bytes result)
remote_compaction_primary: OK
$ grep 'remote compaction result' $DB/LOG
[default] [JOB 3] Received remote compaction result, output path: /tmp/cs_run/output, files: 000022.sst
```
=> PASS: the primary parsed and installed the remote result.
Reviewed By: hx235
Differential Revision: D111068809
fbshipit-source-id: d78dd4b7501cb74e60bcb27dccd869b73052ac6d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15193
Keep `DB::Write`'s existing immediate `IOError` contract while allowing RocksDB to recover safely when a remote file system reports that a WAL handle is permanently unusable. The public API documentation makes the persisted-state boundary, reopen behavior, and safe post-resumption retry point explicit. Normal operation remains unchanged; the additional protocol is entered only after a supported file-scoped WAL error.
### Motivation
A WAL write crosses an acknowledgement boundary: storage can accept the complete record and then fail before the caller receives confirmation. The resulting `IOError` therefore cannot tell RocksDB or the application whether the record is absent, partial, or complete. At the same time, the remote file handle can become permanently poisoned, so retrying that same handle can never succeed.
The previous recovery path retried the poisoned handle. Simply replacing it is not sufficient: RocksDB must also prevent reuse of an ambiguous sequence range, keep every column family on one durable boundary, and prevent WAL tailers from publishing a physically complete record that successful live resumption deliberately abandoned. The new protocol quarantines the affected WALs, reserves the ambiguous sequence range, creates a replacement WAL, atomically persists acknowledged state, retires the quarantined handles, and only then resumes writes.
### Behavior changes
#### 1. Recovering from a poisoned remote WAL handle
- **Why required:** A remote file system can permanently poison a WAL handle after an I/O failure. Retrying that handle cannot succeed and leaves the database write-stopped.
- **Previous behavior:** RocksDB retried the same WAL handle during recovery.
- **New behavior:** The remote file system marks the failure as file-scoped. RocksDB creates a replacement WAL, atomically persists all state that could depend on the failed WAL, and then removes the quarantined WAL handles from active use. Physical file deletion remains part of normal cleanup.
#### 2. Preserving an indeterminate WAL write outcome
- **Why required:** A WAL record can reach storage before the append returns `IOError`; neither RocksDB nor the application can know the outcome from that error alone.
- **Previous behavior:** Recovery could reuse a file-scoped failed write group's sequence numbers, allowing a later acknowledged write to overlap a record that might already be in the WAL. The public write API did not explain the acknowledgement ambiguity or distinguish live resumption from close/reopen recovery.
- **New behavior:** `DB::Write` still returns the original `IOError` immediately and does not retry the logical batch. Normal DB reads can continue, but the failed write is not visible in that DB session. The persisted DB remains indeterminate with respect to the failed write until supported file-scoped live resumption succeeds without closing or reopening the DB. On reopen, or for any other WAL error without the durable-abandonment guarantee, applications must resolve consistency within RocksDB's documented guarantees, including `WALRecoveryMode`, through idempotency, deduplication, reconciliation, or an equivalent mechanism. After successful supported file-scoped live resumption, RocksDB has durably abandoned the failed batch and reserved its sequence range; retrying the same logical write is then recommended to preserve one linear logical history.
#### 3. Keeping TransactionLogIterator consistent with live resumption
- **Why required:** A physically complete error-returned record can remain readable from its WAL even though supported file-scoped recovery excludes that batch from the database. Returning it from `GetUpdatesSince()` would make replication or change-data-capture consumers diverge from the recovered DB.
- **Previous behavior:** An iterator opened before recovery could return the abandoned batch. A new iterator could also return it while the physical WAL remained visible.
- **New behavior:** For file-scoped failures, RocksDB records the exact sequence range and WAL number of each indeterminate write. For the lifetime of that open DB, existing and new WAL iterators stop with `Status::NotFound` before the range, whether the physical record is complete or absent. A caller can begin a new iterator after the reserved range. Iterator creation snapshots the recovery-gap generation together with its WAL-file view so a failure on a later, unseen WAL returns `TryAgain` for the old iterator instead of skipping its valid prefix.
#### 4. Keeping all column families on one recovery boundary
- **Why required:** A write batch can update multiple column families. Retiring its WAL after independently persisting only some column families can expose a partially applied batch after reopen.
- **Previous behavior:** With `atomic_flush=false`, recovery could persist column families independently before abandoning the WAL.
- **New behavior:** File-scoped WAL recovery force-flushes every live column family atomically and verifies that each one advanced beyond the failed WAL before writers are retired or writes resume.
#### 5. Merging concurrent WAL failures
- **Why required:** Another WAL write can fail while an earlier recovery attempt temporarily releases the DB mutex.
- **Previous behavior:** A later failed WAL cutoff or ambiguous sequence could be overwritten or cleared by the earlier recovery attempt.
- **New behavior:** Each error episode monotonically merges the largest failed WAL cutoff and the sequence range of the latest ambiguous write group. Recovery clears the quarantine cutoff only if no concurrent failure extended it; otherwise the newer failure remains write-stopping for the next attempt. Iterator fencing tracks the actual ambiguous write range separately, so an older failed `SyncWAL()` and a newer ambiguous write are not incorrectly paired.
#### 6. Retaining file scope through `NoSpace` recovery
- **Why required:** `NoSpace` can still identify one permanently unusable WAL handle.
- **Previous behavior:** The `SstFileManager` recovery path entered generic recovery without carrying the failed WAL context.
- **New behavior:** File-scoped `NoSpace` carries the same cutoff and sequence evidence through the `SstFileManager` path as other file-scoped WAL errors.
#### 7. Scoping recovery context to one error episode
- **Why required:** A completed recovery must not affect a later unrelated background error.
- **Previous behavior:** WAL recovery context could remain latched after `ClearBGError` succeeded.
- **New behavior:** Successful background-error clearing resets the context. A failed recovery retains it so a later retry has the same safety information. The separate iterator gap record remains for the lifetime of the open DB because later WAL readers must continue to reject the abandoned physical record.
#### 8. Preserving successful `SyncWAL` metadata
- **Why required:** A healthy WAL can finish syncing while a separate recovery error already exists.
- **Previous behavior:** `SyncWAL` could return the existing background error before applying the successful WAL's `VersionEdit`, dropping its MANIFEST update.
- **New behavior:** `SyncWAL` skips quarantined writers, applies metadata for successfully synced healthy WALs, and only then returns any existing recovery error.
#### 9. Classifying only the intended remote-file-system failure
- **Why required:** File-scoped recovery is valid only for an actual WAL `IOError` from a handle whose ownership is still valid.
- **Previous behavior:** Poison state could be queried before ownership fencing was resolved, and file scope could be stamped onto a non-I/O status.
- **New behavior:** Ownership fencing is resolved first, poison state is queried afterward, and only an actual `IOError` can be classified as file-scoped. `IOFenced` remains fatal.
#### 10. Preserving `paranoid_checks=false` retry behavior
- **Why required:** Non-paranoid mode historically cleared a writer's sticky error so a later operation could retry after storage recovered.
- **Previous behavior:** Avoiding reclassification of a cached synthetic writer error also bypassed `WALIOStatusCheck`, so later `SyncWAL` calls kept returning the stale error.
- **New behavior:** A cached error still fails the current `SyncWAL` call, but the non-paranoid path resets the writer for a later retry without setting a background error or entering file-scoped recovery.
#### 11. Preserving async WAL precreation across recovery
- **Why required:** A precreated WAL is an unpublished replacement and is not the active WAL. Recovery can also consume the prepared slot while background work is stopped.
- **Previous behavior:** A file-scoped failure while creating or starting the unpublished replacement was attributed to the active WAL, unnecessarily quarantining it. After successful recovery consumed a prepared WAL, the next slot was not replenished, so the following rotation created its WAL synchronously.
- **New behavior:** Only an active WAL `WriteBuffer()` failure attaches the active-WAL cutoff. A failed unpublished candidate is discarded while the current WAL remains active, and successful recovery schedules the next async precreation after `ClearBGError()` re-enables background work.
### Implementation
- Carry `DBRecoverContext` through generic, `NoSpace`, and `SstFileManager` recovery, merging WAL cutoffs and failed write-group sequence ranges independently of error severity. Generic errors retain the range only so a later retry that establishes file scope can activate the stronger protocol; they do not persist a reservation or install an iterator gap on their own.
- Document the WAL acknowledgement ambiguity, persisted-state boundary, reopen behavior, application consistency requirements, and post-resumption retry guidance on the public `DB::Write` API.
- Publish a lock-free quarantine cutoff used by write and sync paths; replace even an empty failed current WAL; persist file-scoped ambiguous sequence reservations; force and verify an atomic all-column-family flush; and detach quarantined writers only after that durable boundary.
- Record each file-scoped ambiguous write group's exact WAL number and sequence range in a DB-lifetime tracker shared by `WalManager` and all `WalIteratorImpl` instances. Stop an iterator before a batch that contains or crosses the range, and use a generation snapshot to preserve the existing `TryAgain` behavior when an iterator's fixed WAL-file view predates the failure.
- Keep successful healthy-WAL tracking edits in `SyncWAL`, reset completed recovery context, and preserve legacy non-paranoid writer reset behavior.
- Distinguish failures on an unpublished replacement WAL from failures flushing the active WAL. Log the failed replacement identity, keep the active WAL out of quarantine, and replenish the async-precreation slot after successful recovery.
- Classify poisoned remote-file-system handles after ownership fencing and only for `IOError`.
- Add lifecycle logging and a state-machine comment covering quarantine, sequence reservation, iterator fencing, replacement, atomic persistence, retirement, and concurrent-error handoff.
- Add an opt-in `db_stress` mode that injects file-scoped faults only into WAL writes, disables WAL tracking in that mode, validates unsupported direct invocations, and sanitizes crash-test parameters to the supported recovery envelope. A small fraction of ordinary non-specialized crash-test runs select this mode automatically; existing stress behavior is unchanged when the flag is off.
- Make the concurrent-error regression test hold the initial background flush until `SyncWAL` records the intended older-WAL error, preventing scheduler-dependent retirement of the test WAL before fault injection.
### Safety boundaries and unsupported cases
- Automatic file-scoped WAL recovery fails closed with `NotSupported` for manual WAL flush, 2PC/transactions, two write queues, pipelined or unordered writes, WAL recycling, WAL TTL or size retention, either WAL-tracking mode, or blob direct writes.
- Unsupported recovery leaves the DB write-stopped. `DB::Resume()` returns `NotSupported`; after correcting the I/O condition, operational recovery requires close and reopen.
- The failed logical batch is never retried. Only supported file-scoped recovery durably reserves its sequence range and fences `GetUpdatesSince()`. Generic WAL errors retain the existing indeterminate outcome and can replay after close/reopen; applications must use idempotency or reconciliation unless successful file-scoped live resumption has provided the stronger guarantee.
- Failed recovery attempts retain their context. Only a successful `ClearBGError` resets it.
- Physical WAL deletion is not part of the recovery correctness boundary; normal obsolete-file cleanup removes files later.
- The TransactionLogIterator ambiguity fence is kept for the lifetime of the DB instance that observed the file-scoped error and is not persisted as new MANIFEST metadata. After a later close/reopen, normal WAL recovery and obsolete-WAL retention/deletion semantics apply.
- The in-memory iterator-gap tracker can grow with the number of recovered file-scoped failures during one DB lifetime. Safe pruning must account for retained WALs and existing iterator snapshots; it is left for follow-up.
- Async WAL precreation remains disabled with WAL recycling. A prepared WAL is unpublished until its start records succeed; failure before installation does not make the active WAL file-scoped.
- The new stress mode is test-only and defaults off, so normal `db_stress` and database operation without an error are unchanged.
- Automatic stress-profile selection is limited to fully parsed, supported configurations. Explicit targeted runs reject options files and other unsupported combinations.
### Tests and known limitations
- The affected recovery, WAL-manager, WAL-iterator, fault-injection, async-WAL-precreation, and secondary-close binaries build successfully.
- Checked-status suites pass for all 70 `error_handler_fs_test` cases, all 7 `wal_manager_test` cases, all 11 `db_log_iter_test` cases, all 8 `fault_injection_fs_test` cases, and all 5 `DBWALTest.AsyncWalPrecreate*` cases. `DBSecondaryTest.SecondaryCloseFiles` passes 20 consecutive repetitions.
- Focused regressions cover file-scoped gap fencing after an iterator snapshot, generic errors not installing a file-scoped sequence gap, and later recovery discovering that a generic error is file-scoped.
- Earlier validation on this diff also passed crash-test unit coverage, remote-file-system adapter tests, repeated concurrency tests, two bounded multi-column-family stress runs, and bounded black-box crash/reopen dry runs.
- The exact internal RocksDB format check and `arc f --take-formatters --enforce-lint-clean` report no formatting or lint issues.
- The full crash/stress matrix and TSAN were not run.
Reviewed By: pdillinger
Differential Revision: D119324616
fbshipit-source-id: 05779d2bff7018f0ae69c4cbbe5b6455a45675d6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15292
D120637416 introduced a range-for whose `const std::string&` loop variable binds to temporary strings created from a string-literal initializer list. GCC diagnoses this with `-Wrange-loop-construct`, breaking warnings-as-errors builds. Iterate by value to make the intended string construction explicit without changing test behavior.
___
Differential Revision: D122474818
fbshipit-source-id: c7f30a374dc7569ceb1a4ecc6f7f8c3c3637be00
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15271
## Summary
Extend the experimental external-table API with a full mode that preserves RocksDB internal keys, sequence numbers, and entry types. Unlike the existing ingestion-oriented mode, full external tables can represent normal flush and compaction output and participate in live databases with puts, merges, deletions, snapshots, `Get`, `MultiGet`, iteration, multiple LSM levels, and external-file ingestion.
The existing `ExternalTable*` interfaces remain aliases for the original sequence-zero, Put-only mode, preserving compatibility for current implementations.
## Implementation details
- Introduce `ExternalTableMode` and mode-specific reader and factory interfaces. Full-mode builders and iterators receive complete internal keys, while the original mode continues to receive user keys.
- Adapt full-mode point lookups to RocksDB’s `GetContext`, allowing readers to return multiple versions of a user key in internal-key order so RocksDB can resolve snapshots, merges, and deletions correctly.
- Add a batched `MultiGet` context that preserves per-request lookup state and maps external-reader results back to RocksDB’s original request range.
- Translate iterator seek boundaries and returned keys while ensuring file-wide global sequence numbers are applied exactly once.
- Track sequence-number bounds, deletion and merge counts, user-collected properties, file size, and unique IDs when building tables. Unsupported entries fail and abandon the builder.
- Forward checksum verification to the external reader while retaining compatibility with implementations that do not override it.
- Add a standalone test implementation shared by basic and full modes. It uses a simple length-prefixed file format with a RocksDB properties block and checksum, avoiding dependencies on native table implementations.
## Crash testing
Add `use_full_external_table` support to `db_stress` and a dedicated blackbox `db_crashtest.py` profile. The profile exercises randomized reads, writes, merges, ingestion, flushes, compactions, snapshots, checksum verification, and crash recovery.
The profile disables unsupported combinations, including range deletions, user timestamps, sequence-number time tracking, backward scans, multiple column families, backup reconstruction, remote compaction, dynamic table-factory replacement, randomized aggregate/all-table property probes, and global-sequence-number rewriting during ingestion.
Because the simple reader materializes each open table in memory, the profile limits open files and generated file sizes to bound total database memory usage.
Reviewed By: xingbowang
Differential Revision: D120637416
fbshipit-source-id: aa2cbd06999e4325a234210d9c45987addd37550
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15288
Follow-up to D122242630 (https://github.com/facebook/rocksdb/pull/15286), addressing its review.
Previous behavior:
Before D122242630, when `VersionSet::ProcessManifestWrites` failed to build a batch's new `Version`, it returned without removing the batch from `manifest_writers_` (the two `FIXME? manifest_writers_ still has requested updates` comments). The writers are destroyed when their `LogAndApply` call returns, so later MANIFEST updates, including flushes and compactions, could hang behind the dangling entries or corrupt memory. It needs a fatal LSM consistency-check failure, or a `MemoryLimit` from charging `CacheEntryRole::kFileMetadata` (off by default) to a full block cache with `strict_capacity_limit`. It was found by reading the code.
D122242630 fixed that, but found the end of a failed batch with a second scan that copied the grouping loop's stop conditions, which can drift apart, and added no release note.
New behavior:
A failed batch completes the same writers as with D122242630, but only the grouping loop decides where it ends: `ProcessManifestWrites` records the first `LogAndApplyHelper` failure in `batch_apply_status`, later iterations skip version construction but still run the loop's end-of-batch checks, and the batch then completes through `finish_batch`. This also labels the inner-loop `break`s and adds a release note.
Safety, compatibility, and limitations:
No API, option, or format change. The success path and the `SaveTo` failure path are unchanged.
Reviewed By: pdillinger
Differential Revision: D122417768
fbshipit-source-id: c6d474ace32446d6b413b1a1b3cf2fa49a02447d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15289
Why the change is required:
Secondary catch-up must release reader-free retired read views before scanning
for obsolete files. T290043099 and duplicate T290043100 exposed a race where
that ordering was lost, leaving two obsolete table files open and failing
DBSecondaryTest.SecondaryCloseFiles. The assertion failure also exposed unsafe
test cleanup because the secondary DB could outlive the test-local Env.
Previous behavior:
Retiring a read view always woke the background reclaimer. The worker could
remove a reader-free view from the retirement queue and pause before destroying
its SuperVersions. Synchronous cleanup then saw no queued work, ignored the
in-flight reclamation, and continued to obsolete-file discovery while the old
files were still referenced. The test also used a static lambda that captured
the first fixture across repeated runs and closed the secondary only on the
success path.
New behavior:
Catch-up synchronously reclaims reader-free views, or waits when a competing
background worker has already claimed them. Views still held by active readers
remain non-blocking and are reclaimed asynchronously after the last reader
releases them. The test now closes the secondary before its local Env on every
exit path and creates its verification lambda per test invocation.
Implementation:
Publishing a replacement view no longer wakes the background reclaimer; the
immediately following synchronous cleanup owns that work. Reader release still
wakes the worker. Cleanup waits on the existing in-flight counter when a worker
won an earlier race. The test uses a Defer guard for lifetime ordering and
removes the repeat-unsafe static lambda.
Safety, compatibility, and limitations:
This changes no public API, ABI, file format, recovery behavior, or read
consistency. Cleanup does not wait for views retained by active readers, so the
non-blocking multi-column-family iterator behavior is preserved. It waits only
for reclamation already in progress. Validation covered the internal Linux
ASAN/UBSAN target; the external GitHub Actions matrix was not run. Repository
make-based format/source checks require a Git worktree and were unavailable in
this Sapling checkout.
Reviewed By: jaykorean
Differential Revision: D121816425
fbshipit-source-id: a86cfff43a0cedd10b90e526a56e862f18d2a58d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15286
**What**: `LogAndApply` completes every MANIFEST writer in a batch when version construction fails before persistence.
**How**: `VersionSet::ProcessManifestWrites` centralizes queue completion in `finish_batch`, invokes it for `VersionBuilder::Apply` and `SaveTo` failures, and extends assembly failures through the normal group boundary. Completion propagates the failure through callbacks, signals writers owned by other callers, and wakes the next leader.
**Why**: Stack-backed writers otherwise remain queued after their caller returns, leaving dangling pointers, blocked writes, and flush or compaction state that never receives its failure callback.
Reviewed By: xingbowang
Differential Revision: D122242630
fbshipit-source-id: bab4ac8f665269ba57dd86579299c1d5935fa907
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15273
**What**: `ClipColumnFamily` keeps boundary keys valid across unlocked writes, and atomic-flush waits pair every column family with the correct memtable bound.
**How**: `DBImpl::ClipColumnFamily` copies boundary keys from the current `Version` while holding the DB mutex, with a sync point that lets the regression test replace that version before deletion. `DBImpl::AtomicFlushMemTables` and `DBImpl::RetryFlushesForErrorRecovery` construct wait bounds in `cfds` order and supply no-bound fallbacks when a column family is absent from the flush-request map. Both wait sites gain a debug invariant asserting each id matches the map entry for the cfd at the same index.
**Why**: Background version replacement could leave `ClipColumnFamily` reading freed metadata, while unordered or missing atomic-flush map entries could cause early completion or out-of-bounds access for idle or duplicate column families.
Reviewed By: xingbowang
Differential Revision: D121840843
fbshipit-source-id: 8e49b5c953a1aa29a0244fc0aae2ca3ab84be74f
Summary:
GitHub Actions container jobs previously used a fixed 16 GB `/dev/shm` allocation regardless of runner size, while the clang-tidy container did not set one explicitly. This over-allocated shared memory on smaller runners and gave larger, more highly parallel runners no additional shared-memory capacity.
Scale container shared memory to 2 GB per runner core across the PR, nightly, nightly-candidate, weekly, and clang-tidy workflows:
- 4-core jobs use 8 GB.
- 16-core jobs use 32 GB.
- 32-core jobs use 64 GB.
- The nightly-candidate ARM job uses 8 GB, matching its four-way build.
The clang-tidy job now explicitly receives 8 GB. Existing unrelated container options, including the mini crash test's `nofile` limit, remain unchanged.
## Safety and compatibility
This only changes container resource configuration for GitHub Actions jobs. It does not change RocksDB code, APIs, persistent formats, or non-containerized jobs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15284
Test Plan:
- `make check-workflow-yaml`
- `git diff --check upstream/main...2026_09_28_shm`
Both checks pass. The CI workloads themselves were not executed locally, so the new shared-memory allocations will be exercised by PR and scheduled workflow runs.
Reviewed By: omkarhgawde
Differential Revision: D122218094
Pulled By: xingbowang
fbshipit-source-id: aa577c37683022ba01c4a073a011dbe0032abbf6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15283
**What**: `DBSecondaryTest.SecondaryCloseFiles` no longer calls into a destroyed `Env` when the test body returns early, fixing a ThreadSanitizer heap-use-after-free at secondary DB close.
**How**: The test opens `db_secondary_` (a fixture member) on a stack-local `traced_env`. A `Defer` now closes the secondary before `traced_env` is destroyed on every exit path, so the later `~DBSecondaryTestBase` close never reaches `DBImpl::CloseHelper`'s `env_->UnSchedule` on a freed Env. This mirrors the sibling `DBSecondaryCatchUpFaultTest` fixture, which already closes the secondary in its destructor before its Env member for the same reason.
**Why**: On any early `ASSERT_*` return before the explicit close, `traced_env` (a body local) was freed first while `db_secondary_` (a fixture member) was closed later in the base destructor, which made a virtual call on the freed Env. ThreadSanitizer flagged this as a land-blocking heap-use-after-free in the `build-linux-clang21-mini-tsan` CI job.
Failed CI run: https://fburl.com/jo6gtllf
___
Differential Revision: D122195601
fbshipit-source-id: 3d8b427ba50759c40f15d2b512379594b6057234
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15265
This change adds column family name reflection to an external table reader and allows the user to extract an external table factory from a top level table factory, if it exists
Motivation
Adding column family metadata to the reader allows the reader to handle CF specific metadata, also add external table reflection
Previous/New Behavior
No change in behavior. Added a defaulted parameter threaded through many constructors to preserve existing API contract while introducing new capability. Added a function which allows unwrapping of an external table factory
Nothing much to test
Reviewed By: xingbowang
Differential Revision: D121335729
fbshipit-source-id: 74f9b1be6a8f301bdbe4540de0656b83e26f88ab
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15270
D121318547 added SecondaryDB batch-iterator coverage to CfConsistencyStressTest::ContinuouslyVerifyDb(), but the automated CF-consistency crash-test profile never selected test_secondary. Automated runs therefore exercised only the primary DB path while the targeted manual validation used test_secondary=true.
Randomize test_secondary between 0 and 1 in cf_consistency_params. Each generated run can now exercise either the existing primary-only path or the SecondaryDB path, preserving both configurations in the continuous matrix. The regression test deterministically verifies both random outcomes. Existing sanitization may still disable SecondaryDB when another selected option is incompatible, such as remote compaction or primary UDI mode.
Reviewed By: xingbowang
Differential Revision: D121813002
fbshipit-source-id: 4080e2131c1f063d590d24cc0bcee9b9096de63a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15199
## Why
PrefixScan and MultiScan verification failures are difficult to reconstruct from the terminal assertion alone. The iterator options, bounds, seeks, preparation state, and immediately preceding movements are often the evidence needed to distinguish a database defect from an invalid test assumption.
## Why the existing recovery trace is not enough
The expected-state trace exists to reconstruct the expected database state after recovery. It preserves ordered write batches and deliberately filters out reads and iterator seeks, then recovery consumes a sequence-number-selected prefix. Consequently it contains none of the scan-specific evidence needed here: read options, bounds, seek targets, iterator validity, lazy-value preparation, movements, or the exact mismatch state. Scan witnesses capture that bounded human-readable context only when a verification failure needs it; they are best effort, are never replayed, and do not alter the recovery trace or its failure semantics.
## Previous behavior
The scan paths emitted generic failure output but did not retain a focused bounded witness of iterator setup and state transitions. Adding instrumentation naively would also format diagnostic strings on the normal path even when disabled and could create one file per failing operation.
## New behavior
An opt-in bounded ScanWitnessLog records scan configuration and recent iterator events for PrefixScan and MultiScan variants. On a qualifying failure it writes the latest complete witness for that DB, PID, thread, and scan type. The operation ordinal remains in the file contents. Repeated failures atomically replace the fixed witness rather than creating unbounded files.
## Implementation
The witness captures sanitized read options, input ranges, bounds, seeks, movement, PrepareValue results, iterator status, keys, value size and hash, and wide-column state. MultiScan tracks whether the current primary iterator value was actually prepared so failure snapshots report accurate state. Disabled witnesses avoid scan-label ownership, range formatting, expected-count formatting in release builds, and numeric iterator-label conversion. Witness writes share the per-thread diagnostic I/O-disable latch with breadcrumbs, write to a fixed temporary path, close, and rename over the final file. Entry counts are capped at 65,536. A formatting-only follow-up normalized all touched C++ lines with the repository clang-format rules.
## Safety and unsupported cases
The feature is stress-tool-only and raw db_stress keeps it disabled by default. It observes public iterator state and does not change database contents, expected-state tracing, recovery replay, iterator decisions, or fault-injection outcomes. Diagnostic I/O remains best effort and cannot fail the workload. Files are bounded per retained PID, thread, and scan type. The witness covers the instrumented PrefixScan and MultiScan implementations only and is not a general RocksDB trace. Files are not fsynced and therefore are not a power-loss durability mechanism.
## Tests and limitations
Validated with 47 tools/db_crashtest_test.py tests, exact-candidate debug and release db_stress builds, arc lint over the complete stack, oversized-entry rejection, and clean independent correctness reviews of the final integrated stack. make check-sources is unavailable because it invokes Git in this Sapling checkout; equivalent added-line hygiene checks pass. Only the newest complete witness per DB, PID, thread, and scan type is retained, so later failures intentionally replace earlier witnesses.
## Rebase update
Rebased onto fbsource master dc59f986b964. The logical diff is unchanged; the rebased stack passes debug and release db_stress builds, all 47 db_crashtest Python tests, and arc lint.
Reviewed By: jaykorean
Differential Revision: D119160911
fbshipit-source-id: 0e9b040c3bbb482e0a3b1b11070a7c5cc519f060
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15241
Native `TableFactory` implementations receive the RocksDB `WritableFileWriter` in `NewTableBuilder()` and `RandomAccessFileReader` in `NewTableReader()`. These wrappers provide buffering, checksums, rate limiting, statistics, listeners, tracing, and direct-I/O handling. External table implementations previously received a limited writable-file adapter and reopened readers from a path, bypassing parts of this wrapper logic.
This change makes `WritableFileWriter` and `RandomAccessFileReader` implement `FSWritableFile` and `FSRandomAccessFile`, respectively, so the same RocksDB wrappers can be passed to external tables. `ExternalTableFactory::NewTableReader()` now receives the already-opened reader and file size. The filesystem interfaces expose standard append, flush, sync, fsync, file metadata, synchronous reads, asynchronous reads, and batched reads. `RandomAccessFileReader::SubmitReadAsync()` schedules its wrapped coroutine read on the configured filesystem read executor, and the DB table-cache and `SstFileReader` construction paths pass that filesystem through. `SstFileReader` also supplies its configured statistics, listeners, rate limiter, and file temperature, removing the need for a separate external-table metrics filesystem wrapper.
For direct I/O, `MultiRead()` checks whether requests are already sorted. When alignment and merging are needed, it constructs a sorted index only for unsorted input, preserving caller request order and result mapping. Fully aligned requests using aligned caller buffers bypass this path and may be supplied in any order.
Reviewed By: xingbowang
Differential Revision: D121035539
fbshipit-source-id: 27cb897726c31cb1a4dc811c65c8d79e690d82b1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15264
Why the change is required:
LargeSizeSstFileWriter creates approximately 12 GiB of SST output in the shared /dev/shm used by build-linux. With 32 concurrent test jobs in a 16 GiB tmpfs, this can exhaust the filesystem and make unrelated fault-injection and repair shards fail with ENOSPC.
Previous behavior:
The test wrote an approximately 8 GiB large-key SST and an approximately 4 GiB large-value SST even though it never read or ingested either file. It validated the size limits and successful Finish calls, but the physical output competed with every other test process.
New behavior:
The test still runs the real SstFileWriter and table builder for over-limit rejection and maximum-size acceptance, including Finish, sync, and close. The generated byte stream is counted and discarded instead of materialized in the shared tmpfs.
Implementation:
Add a test-local FileSystem wrapper whose writable file counts appended bytes and records sync and close calls. Inject it only into LargeSizeSstFileWriter and verify that the writer-reported logical size matches the observed append stream, the lifecycle completes, and no physical SST is created.
Safety, compatibility, and limitations:
Production code, shared filesystem implementations, serialized formats, and other tests are unchanged. The fake does not validate SST contents, but the previous test did not read the generated files either. Table construction still consumes substantial memory, so the existing HasBigMem guard remains; this change addresses tmpfs exhaustion only.
Reviewed By: rban1
Differential Revision: D121483663
fbshipit-source-id: 055372bc07e3a1f7289e34ebc35694bc3c4bfad9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15263
Why: keep internal RocksDB main ready for 11.12.0 development while the 11.11 release is cut.
Previous behavior: Conveyor opened a GitHub pull request for this update, which could be missed and was not stacked with the internal release diff.
New behavior: Conveyor owns an internal Phabricator diff that the release-import diff stacks on top of. The release manager exports this diff manually with ExportIt.
Implementation:
- Bump version.h from 11.11.0 to 11.12.0
- Add `11.11.fb` to `check_format_compatible.sh`
- Sync HISTORY.md `## 11.11.` section from `11.11.fb`
- Delete 19 consumed `unreleased_history/` note file(s) from main
Safety and unsupported cases:
- Only version.h, check_format_compatible.sh, HISTORY.md, and consumed release-note files are allowed.
- Open legacy GitHub update-main PRs block this flow; this automation never modifies, exports, or lands them.
- Concurrent direct CLI invocations are unsupported; Conveyor serializes the Release Cut node to one active run.
- The tool refuses unrelated commits, mismatched release metadata, and non-canonical GitHub repositories.
Known limitations:
- ExportIt remains manual. Bot create permission for internal_repo_rocksdb/repo is a rollout prerequisite.
Reviewed By: xingbowang
Differential Revision: D121434642
fbshipit-source-id: 3c6c90f425e043ff9c13318ecc671aedbdef0347
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15259
Publish immutable SecondaryDB read views after open and successful catch-up so `NewIterators()` can acquire one cross-CF sequence and SuperVersion set without waiting for `DBImpl::mutex_`. Keep failed catch-ups on the previous view, reject refresh for batch iterators, and exercise concurrent catch-up plus batch scans in the CF-consistency stress profile.
Reviewed By: xingbowang
Differential Revision: D121318547
fbshipit-source-id: 3198f4766c90ab0bb39400963df3c329b4ecf6bf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15254
**What**: `IODispatcher` verifies prefetched data blocks before serving or caching them and safely falls back when block construction fails.
**How**: `CreateAndPinBlockFromBuffer` validates every returned span and, when `ReadOptions::verify_checksums` is enabled, verifies its checksum and records the standard counters. `PollAndProcessAsyncIO` handles coalesced blocks independently and retires shared request state before failed entries fall back through `SyncRead`; `ReleaseFailedSyncPrefetchMemory` returns limiter capacity for unmaterialized synchronous blocks.
**Why**: Unverified prefetches could serve or cache bit rot, while newly detected failures could otherwise leave async state referencing released provider buffers or permanently consume bounded prefetch capacity.
Reviewed By: xingbowang
Differential Revision: D121277687
fbshipit-source-id: 65cc1d5b9dcc53bdc64f8ee5d672bfe2cda7b40b
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15261
Why the change is required:
Internal Phabricator diffs and external GitHub pull requests need the same substantive explanation of a RocksDB change. Existing local instructions focus on code generation and review but do not define what a summary must tell reviewers, so summaries can omit prior behavior, safety boundaries, compatibility, or validation evidence and can drift between the two review surfaces.
Previous behavior:
There was no CLAUDE.md at the internal_repo_rocksdb wrapper root, the RocksDB CLAUDE.md did not point authors to summary guidance, and there was no shared document covering both internal diff and external pull request summaries.
New behavior:
The wrapper-level CLAUDE.md requires authors to consult a shared summary guideline before writing or updating an internal diff summary. The repository CLAUDE.md applies the same guideline to external pull request summaries. The guideline requires motivation, previous and new behavior, implementation, safety and compatibility boundaries, and test evidence, with RocksDB-specific advice for correctness, persisted formats, concurrency, and performance changes. Examples are separated by access boundary: the internal wrapper contains Phabricator diff examples, while the repository guideline contains publicly accessible GitHub pull request examples.
Implementation:
Add a concise wrapper-level CLAUDE.md, add claude_md/pull_request_summary_guideline.md as the shared source of truth, and link it from repo/CLAUDE.md. Keep annotated internal diff references in the wrapper-level file and corresponding public pull request references in the repository guideline, including a recent change from Peter Dillinger.
Safety boundaries and unsupported cases:
This is documentation-only and changes no runtime, API, build, test, or CI behavior. The repository guideline contains no internal links, so external readers do not need Meta access. The guidance is manually followed by authors and agents; it does not mechanically validate summaries, replace reviewer judgment, or require a rigid section layout for small changes.
___
Differential Revision: D121388476
fbshipit-source-id: bb09ab7ba8ef6dc2b2830a42735322890536622f
Summary:
`BlockBasedTableIterator::SeekImpl()` sets
`seek_key_prefix_for_readahead_trimming_` only when given a seek key,
and `ResetBlockCacheLookupVar()` does not clear it. After a `Seek()`, a
`SeekToFirst()` on the same iterator trims readahead against the
previous seek key's prefix. `DBIter::SeekToFirst()` under
`prefix_same_as_start` takes its prefix from the first key found, so the
old prefix does not describe the keys being scanned.
With the stale prefix, `IsNextBlockOutOfReadaheadBound()` is true at
every block instead of only at the prefix boundary, and each prefetch is
clamped to the current block. The same predicate sets
`is_index_out_of_bound_`, which `FindBlockForward()` converts to
`is_out_of_bound_` and ends the scan. Under `ReadOptions::async_io` this
drops keys.
The bug needs `auto_readahead_size` and a preceding `Seek()` that reads
no data block. A `Seek()` that lands on a key leaves the prefetch buffer
anchored above the scan, and the trimming callback is not called. The
bug does not occur under `allow_mmap_reads`, where `FilePrefetchBuffer`
is constructed disabled.
The fix clears the prefix in `SeekToFirst()`, so prefix-based trimming
is disabled for that positioning. The `auto_readahead_size`
documentation in `include/rocksdb/options.h` already says prefix-based
trimming requires `Seek(key)`.
On buffered reads with `prefix_same_as_start`, `auto_readahead_size` and
`readahead_size=8192`, a cold-cache `SeekToFirst()` scan of 2M keys in a
single 222 MB file, preceded on the same iterator by a `Seek()` past the
last key, took 391 ms before and 308 ms after. The same scan with no
preceding `Seek()` takes 307 ms. Under direct IO on a local disk it went
from 33.1 s to 11.8 s. Data blocks read are unchanged; file reads drop
by 3x.
Also fixes a pre-existing test bug: `SeekParallelizationTestWithPosix`
asserts on `number_async_seek` without resetting that cumulative
thread-local, unlike its three siblings. The new tests move
`prefetch_test` from 11 shards to 13, so different tests share a
process, and the assertion fails at the parent commit under that
grouping.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15216
Test Plan:
New regression test in `PrefetchTrimReadaheadTestParam`, in sync and
`async_io` variants over the fixture's three `index_shortening` modes.
It runs two scans of the same keys from an equally cold block cache,
differing only in a preceding `Seek()`. Reverting the `clear()` fails 3
of 6 sync parameterizations on `READAHEAD_TRIMMED` (6 vs 40) and 3 of 6
`async_io` parameterizations on key count (40 vs 1).
`make check` and `ASSERT_STATUS_CHECKED=1 make check` show no new
failures. `prefetch_test` 122/122, `db_iterator_test` 249/249,
`table_test` 7629/7629, `block_based_table_reader_test` 5347/5347,
`db_basic_test` 300/300. New cases ran 1200 times under
`COERCE_CONTEXT_SWITCH=1` without failure.
Added db_stress coverage for `SeekToFirst()` after `Seek()` under
`prefix_same_as_start`. With the fix reverted and io_uring available,
the configuration in the second commit fails 8 of 8 seeds with
"Iterator diverged from control iterator"; with the fix, the same seeds
pass.
Reviewed By: rban1
Differential Revision: D120228466
Pulled By: xingbowang
fbshipit-source-id: a4935f03a0080cc8dd4c7a8c1438404e9b8f5c3a
Summary:
Make `IndexFactory` the public interface for custom indexes. Keep the old `UserDefinedIndex` names, header, and `NewBuilder()` / `NewReader(Slice&)` overloads so existing implementations can still compile.
UDI and options tests pass on the full stack, both normally and with `ASSERT_STATUS_CHECKED=1` and `COERCE_CONTEXT_SWITCH=1`. The C API test and generation/compatibility checks pass. Full `make check` runs had test timeouts and hit the 30-minute local limit. These test results are from before the rebase.
Part 5 of 9. Previous: https://github.com/facebook/rocksdb/issues/14961 (landed). Next: https://github.com/facebook/rocksdb/issues/14963.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14962
Reviewed By: jaykorean
Differential Revision: D119960591
Pulled By: xingbowang
fbshipit-source-id: 9703950bd48723c2c524da89cbea63bf00254cc1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15218
Why the change is required:
Public and serialized RocksDB inputs can change observable performance or resource behavior without a signature or default change. These changes require deliberate human judgment, clear documentation, and research into downstream production impact. The automated CI review also needs an explicit perspective for identifying gaps in that analysis.
Previous behavior:
CLAUDE.md described source and ABI compatibility but did not explicitly guide authors or reviewers through behavioral changes to already accepted inputs. The CI review prompt had only a brief API-compatibility checklist item and no shared instructions for evaluating intentionality, release notes, downstream usage, or regression coverage.
New behavior:
CLAUDE.md states that behavior changes are human judgment calls that must be intentional, documented, and researched. It gives concrete guidance for existing accepted inputs while asking agents and reviewers to call out gaps or ambiguity. The CI review prompt applies the same perspective to every review agent and makes clear that a behavior change is not automatically a defect.
Implementation:
Revise the Behavioral Compatibility policy in CLAUDE.md, reduce its workflow entry to a concise pointer, add a dedicated Behavioral Compatibility Review Perspective to ci_review_prompt.md, and point the API and compatibility reviewer to that shared section.
Safety boundaries and unsupported cases:
This is documentation-only and changes no runtime, API, build, or CI execution behavior. The guidance supports human judgment; it does not automatically decide whether a behavior change is acceptable, guarantee exhaustive discovery of downstream consumers, or replace release-note, testing, and production-impact review.
Reviewed By: pdillinger
Differential Revision: D120344151
fbshipit-source-id: aa5bbb00fbbe167ef21b91b3551868c4d1b4046d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15237
Route `DB::MultiGet` batches through `ExternalTableReader::MultiGet` instead of `TableReader`'s per-key fallback, preserving per-key status and pinned-value handling.
Allow basic external tables to honor DB-assigned global sequence numbers across overlapping levels, snapshots, `Get`, `MultiGet`, and iteration. Preserve ingestion-only behavior by rejecting compaction output, and support both `allow_global_seqno` and deprecated `write_global_seqno` files.
Reviewed By: xingbowang
Differential Revision: D120629061
fbshipit-source-id: 366f48660e97a300afb59627b9a1a41fe3805590
Summary:
These changes improve the block-based SST table properties available for compression benchmarking and optimization.
- Fix: `uncompressed_data_size` was computed while building a file but was never written to the properties block, so it always read back as 0. It is now persisted (written in AddTableProperty and read back in ParsePropertiesBlock).
- Add: a new packed `lsm_info_at_creation` property records the LSM level and bottommost status at the file's creation. A new `LsmInfoAtCreation` struct owns the (extensible) uint64 encoding via symmetric Encode()/DecodeFrom(), keeping TableProperties a bare data struct. It distinguishes three states: "unknown" (nothing recorded, e.g. older files), "not applicable" (no LSM position, e.g. SstFileWriter/misc output), and "applicable" (level + bottommost, with an unknown-level sentinel for the DB repairer). The level is stored as its natural value.
- Change: the `_compression_manager` detail in the `compression_options` property now records the manager's full serialized configuration (via ToString) instead of only its GetId().
New properties are additive and backward compatible: older files without them read back as "unknown"/0.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15231
Test Plan:
Added coverage in db/db_table_properties_test.cc: LsmInfoAtCreation encode/decode round-trips (natural levels, unknown level, not-applicable vs unknown); end-to-end persistence of uncompressed_data_size and lsm_info_at_creation after flush (L0, not bottommost) and a forced bottommost compaction; not-applicable recorded for an SstFileWriter file; and full compression manager configuration present in compression_options. Updated options/options_settable_test.cc for the new field.
Also made table_test's BlockBasedTableTest.CompressionRatioThreshold resilient to per-file overhead changes (such as these added table properties): it now verifies the compression decision via the persisted data_size vs uncompressed_data_size rather than modeling total file size against a fixed overhead constant, which had become too tight. Full table_test passes.
Reviewed By: rban1
Differential Revision: D120553038
Pulled By: pdillinger
fbshipit-source-id: 07ede62dbcbe31c040fd9b037c7636a49558c4d9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15194
## Why
Intermittent db_stress and db_crashtest failures often report only the terminal assertion, after the operations that established the failing state have scrolled out of logs. The diagnostics must retain the latest operations even when the stress process is killed before it can run a failure handler.
## Why the existing recovery trace is not enough
The existing expected-state trace is a correctness oracle for recovery: after an expected-state checkpoint it records ordered write batches, and recovery replays the prefix implied by the database sequence number. Reads and iterator seeks are deliberately filtered out, trace-write failures are fatal, and the trace is machine-consumed rather than intended as a concise failure narrative. It therefore cannot show which read, scan, maintenance operation, or operation phase immediately preceded an assertion. Breadcrumbs fill that observability gap with a bounded, human-readable, per-thread history. They are best effort, never replayed, and remain independent of the recovery-validation contract.
## Previous behavior
db_stress did not retain a bounded per-thread operation history. db_crashtest did not automatically provision a local directory for these artifacts. The first version of this diff kept the ring in process-local memory and periodically wrote a complete temp-file snapshot, which could lose every operation since the last snapshot when the process crashed.
## New behavior
An opt-in per-thread memory-mapped breadcrumb ring records recent operation begin, context, and end events directly in a local file. Committed records therefore remain in the kernel page cache when only the db_stress process exits or is killed; no periodic flush is required. db_crashtest enables breadcrumbs by default, assigns a local diagnostics directory, retains the newest eight process groups after failed child runs, and removes managed artifacts after success.
## Implementation
Each ThreadState lazily creates a uniquely named, preallocated file and maps it through Env::NewMemoryMappedFileBuffer(). The file contains a fixed header and 512-byte text records. A record slot is marked I before its body is replaced and V only after the complete body is published; E denotes an unused slot. Each valid record carries a monotonically increasing sequence number so wrapped records can be ordered. Oversized context is bounded and ends with details_truncated=1. Lazy initialization avoids empty files for auxiliary ThreadState objects, and a process-wide file ID prevents collisions between ThreadState objects that share a logical thread ID.
The periodic-flush flag, temp-file/rename snapshot path, and verification-failure fallback flush are removed. Enabled operation-context formatting continues to reuse a per-thread string and uses std::to_chars for integers; disabled call sites avoid formatting. Diagnostic directory setup and mapping failures are reported but do not change the workload result. Auto-directory pruning and cleanup accept only generated filename schemas, reject symlink roots, preserve unrelated files and explicit user directories, and retain at most eight PID groups.
## Safety and unsupported cases
This is stress-tool-only diagnostic behavior. Raw db_stress keeps breadcrumbs disabled by default. The expected-state recovery trace and scan-witness contracts are unchanged. The mapping deliberately does not use msync or fsync: it is intended to survive a process crash while the kernel remains running, not power loss, host reset, or kernel failure. Filesystems or Env implementations without shared writable mappings disable breadcrumbs non-fatally. Replacing or truncating a live mapped file is unsupported. Records are fixed width, so unusually large operation context is explicitly truncated. Concurrent live readers are not promised a consistent snapshot; post-termination readers must ignore I and E records and order V records by sequence number. Nested operation scopes remain unsupported.
## Tests and limitations
Validated with all 47 tools/db_crashtest_test.py tests, exact integrated debug and release db_stress builds, complete-stack arc lint, lower/upper/combined clang-format checks, normal ring wrap, distinct files for four worker threads, an oversized 256-column-family context, and process-crash tests. A release process killed with SIGKILL retained all 16 committed ring records without an explicit flush. A debugger-forced SIGKILL after the I store but before V publication left the interrupted slot marked I. In five CPU-pinned alternating release pairs of 50,000 operations, enabled and disabled medians were both 2.15 seconds. This remains a bounded latest history rather than a complete log and provides no host-crash or power-loss durability.
## Rebase update
Rebased onto fbsource master dc59f986b964. The logical diff is unchanged; the rebased stack passes debug and release db_stress builds, all 47 db_crashtest Python tests, and arc lint.
## Review follow-up
The mapped-slot invalidation now uses an acquire-release atomic exchange so later body stores cannot become visible before the I marker on weakly ordered architectures.
Reviewed By: anand1976
Differential Revision: D119160910
fbshipit-source-id: ffe172af9ebc75c617196592806dc9641a539a25
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15229
Add `db_session_id` to obsolete file deletion logs and structured table/blob deletion event logs. Also include `db_session_id` when logging new MANIFEST creation so split-brain DBImpl sessions can be correlated without relying on process-local pointers.
Reviewed By: joshkang97
Differential Revision: D120413139
fbshipit-source-id: c175b70d9b1f93e3242ee9e3191c274dc51ef3bf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15232
**What**: Make `SstFileWriter::Open()` return `InvalidArgument` without replacing the active writer when a file is already open.
**How**: Check `rep_->builder` before constructing new output resources, matching `OpenWithEmbeddedBlobs()` and leaving the original file finishable.
**Why**: A second open otherwise destroys an unfinished `TableBuilder`, violating its lifecycle contract and triggering an assertion.
Reviewed By: pdillinger
Differential Revision: D120469611
fbshipit-source-id: 8daa4607725eea574fe80012f9df98132160ce3e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15230
Add a raw C++ `DB::NewIterators()` overload accepting parallel vectors of `ReadOptions` and column family handles. Each iterator receives its own options while all iterators share one consistent sequence snapshot. Require identical snapshot pointers and tailing modes across the options.
Implement the overload for primary, secondary, read-only, stackable, and write-prepared DBs. Building on D120048299, both secondary overloads share one helper that captures all SuperVersions and the sequence number under the DB mutex.
Preserve read-only `table_filter` support when range-tombstone conversion is configured, and reject that unsafe combination in write-prepared read-write iterators before allocating any iterator.
Add validation, per-CF prefix-seek coverage, a secondary catch-up regression test, table-filter contract tests, and a public API release note.
Reviewed By: xingbowang
Differential Revision: D120296578
fbshipit-source-id: 3775884537ea29a0d32bb3fe10330cf62ac5a406
Summary:
Follow-up (Phase 3) to the experimental lazy wide-column read API (DB::GetEntityLazy / DB::MultiGetEntityLazy). A MultiGetEntityLazy batch now resolves blob references across keys with coalesced I/O instead of key-by-key.
- MultiGetEntityLazy acquires one SuperVersion and one consistent (implicit) sequence number for the whole batch via MultiCFSnapshot (as multi-CF iterators do) and transfers a single shared per-column-family pin into the LazyWideColumnsBatch, replacing the former per-key SuperVersion + explicit-snapshot loop. Single column family for now; a cross-CF ColumnFamily** overload remains a follow-up (the batch already models a per-CF pin map).
- LazyWideColumnsBatch::MultiResolve groups the batch's reads by (Version, blob file) for separate-file references and by SST for embedded references, and issues one coalesced read per group -- for whole-column and byte-range reads alike -- caching whole values in each entity's resolver and pinning partial reads directly.
- Adds the coalesced blob-read primitives this needs (none existed before): range-aware BlobFileReader::MultiGetBlobRange / BlobSource::MultiGetBlobRange / Version::MultiGetBlobLazy for separate-file references, and an embedded (SimpleGen2) batch read path (blob_gen2_format batch readers + BlobSource::MultiGetSimpleGen2Blob[Range] + BlockBasedTable::MultiGetSameFileBlob). Whole-value separate-file coalescing reuses the existing MultiGetBlob. The batch readers sort their requests by file offset before issuing the underlying MultiRead (which requires ascending offsets), and gen2 checksum verification is factored into one helper shared by the scalar and batch readers.
- The hot non-lazy Get / MultiGet / iterator / compaction paths are untouched: the new multis are lazy-only, and the GetImpl change is gated on a new opt-in GetImplOptions field (an injected shared SuperVersion + sequence number) used only by the batched lazy read.
- Blob read failures on these paths (short reads, invalid offsets, out-of-range requests, compression mismatches) now carry file/offset/size and expected-vs-actual context, mirroring the block-checksum diagnostics.
- db_bench gains a multireadrandomentitylazy benchmark for measuring the coalesced batch path.
No public API change, so this is a performance follow-up to the still-recent lazy API rather than a new feature. EXCEPT that `LazyColumnReadRequest::status == nullptr is now explicitly disallowed`.
Left for later follow-up:
- When a whole fetch is planned for an (entity, column), bind ordinary range requests for that column to the same fetch and slice them after adoption instead of issuing separate range I/O.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15178
Test Plan:
New unit tests in db/wide/db_lazy_entity_test.cc assert cross-key coalescing via a blob-file / SST MultiRead counter added to the test FileSystem wrapper:
- N whole-column separate-file reads across keys in one blob file collapse to a single coalesced MultiRead (and the fetched values are cached, so a repeat read does no further blob I/O);
- N byte-range separate-file reads collapse to one MultiRead, save the un-read bytes (rocksdb.blobdb.lazy.partial.bytes.saved), and never fill the blob cache;
- embedded (same-file) reads across keys in one SST collapse to one MultiRead over that SST;
- an out-of-order embedded batch (reads issued in descending record offset) still resolves correctly in one coalesced read, covering the batch readers' internal offset sort;
- a mixed batch (inline + whole + byte-range across keys) returns the correct bytes for each read;
- a batched MultiGetEntityLazy + one MultiResolve matches resolving each key via a separate GetEntityLazy. The existing db_stress lazy-vs-eager differential (MaybeTestMultiGetEntityLazy) exercises the new batched read and cross-key MultiResolve unchanged (same public API).
Reviewed By: xingbowang
Differential Revision: D118660682
Pulled By: pdillinger
fbshipit-source-id: 6a832175e730917eddeedab8e0b9cb2b8d6ae7b4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15211
Capture the secondary DB sequence number and all requested column-family SuperVersions while holding the DB mutex, then construct the iterators after releasing it. This keeps every child on one state even when `TryCatchUpWithPrimary()` concurrently installs file-backed SuperVersions.
Add a file-only catch-up regression test and release note.
Reviewed By: xingbowang
Differential Revision: D120048299
fbshipit-source-id: 89f4de533c4fbc1c6e86656845bda4bbb9da4d2c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15215
**What**: Stop `SstFileDumper` from dereferencing a null file handle on two error paths, both of which crash `sst_dump` today.
**How**: Both sites create a file and then use the handle without checking whether creation succeeded. `Env::NewWritableFile()` and `FileSystem::NewRandomAccessFile()` each reset the caller's handle before they can fail, so on failure the handle is null.
- `DumpTable()` fell through to a shared error path that closed the handle. It now returns as soon as `NewWritableFile()` fails.
- `GetTableReader()` re-opens plain and cuckoo tables with mmap enabled and discarded the returned status entirely, then wrapped the null handle in a `RandomAccessFileReader` that the next property read dereferenced. It now propagates the status.
**Why**: Both are reachable from the `sst_dump` tool. The output file cannot be created on an unwritable directory, a missing parent, or a full filesystem; the mmap re-open fails when mmap is unavailable or the file changed between the two opens. In each case the tool crashed instead of reporting the error. The plain and cuckoo branch had no test coverage at all, which is why its unchecked status never tripped an `ASSERT_STATUS_CHECKED` build.
Reviewed By: xingbowang
Differential Revision: D120182509
fbshipit-source-id: dbe9b7f84140ed11dea395f4192b895e0a7bc70d
Summary:
Extends `sst_dump --command=recompress` from a basic evaluation tool to a serious compression benchmarking tool, also able to evaluate interactions with table format options affecting uncompressed sizes and compressibility.
For each input file it now emits a benchmark "baseline" entry for the file as stored, followed by one entry per compression type / strategy / level, reporting compressed and uncompressed data-block sizes, index sizes, compression ratio, and both wall-clock and CPU time for the write and read tests. All data/index sizes are block payload excluding block trailers, so compressed vs uncompressed figures are on a consistent basis. The input's uncompressed data size is inferred from decompression statistics, so it is available even for files that don't persist it. That inference is exact when the input's index blocks are stored uncompressed (the default); with enable_index_compression it can be slightly overstated because index-block decompression is also counted (see the CAVEAT in GetInputBenchmarkMeasurement).
New capabilities:
* `--json` for machine-readable output. It is self-contained: each measurement carries the compression manager id and the full CompressionOptions, and every baseline/measurement entry is tagged with its input file path so multiple input files produce one valid document.
* `--compression_strategy=` to sweep a list of CompressionOptions strategy values (multiplied with levels), useful for side-channel configuration of custom compressors.
* `--show_properties` additionally dumps the input and last recompressed output TableProperties.
To keep measurements meaningful, a quiet throw-away priming build runs whenever the compression type changes (warming up the compressor), CPU time is measured process-wide so parallel compression worker threads (compression_parallel_threads > 1) are included rather than undercounted, and the timed windows are kept tight: the write build is fed from an in-memory copy of the input so its timing excludes reading and decompressing the input (and is independent of block-cache state), and the read timing covers only the block reads. The timings for compression and decompression do include BlockBasedTable overheads but this is considered reasonable--if not preferable--to the intended application of finding the overall most efficient settings related to compression.
Also fixes a latent bug exposed by measuring several compression types in one run: the recompressed table was read back through the shared block cache under a common (unknown) cache-key identity, so a later type could get a cache hit on a stale block from an earlier, differently-compressed table and fail with "Corruption: bad entry in block". The read-back now bypasses the block cache.
Also fixes a typo in CLAUDE.md.
Intended follow-up: persist TableProperties::uncompressed_data_size. It is already computed in memory at build time but is never written to the properties block (PropertyBlockBuilder::AddTableProperty has no entry for it), so it reads back as 0 for on-disk files. Persisting it (backward compatible; old files read 0) would fix `data uncompressed size: 0` for all property readers and let this tool read the value directly, removing the decompression-ticker inference above and its index-compression caveat for all but pre-existing files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15195
Test Plan:
Added sst_dump_test cases exercising recompress with --show_properties and with --json.
Ad hoc manual testing against real SST files: verified the text benchmark table, and piped `--json` output through `json_pp` to confirm well-formed JSON, for a single file, a directory of multiple SST files, and combined with --show_properties, --output_hex, --compression_strategy, and --block_based_table_options. Confirmed process CPU time exceeds wall-clock write time with --compression_parallel_threads=4 (and matches at =1).
Reviewed By: joshkang97
Differential Revision: D119536444
Pulled By: pdillinger
fbshipit-source-id: 30dbabc14872b9c6f7d8ca11227e762674dcef02
Summary:
Compressor::GetRecommendedParallelThreads() was a narrow hook for the one CompressionOptions field (parallel_threads) that the SST builder, rather than the Compressor, is responsible for applying. Other such options exist (max_compressed_bytes_per_kb, auto_skip, auto_skip_min_sample_every), and more may be added. This replaces the narrow hook with an omnibus MaybeOverrideCompressionOptions(CompressionOptions*) that a Compressor can override to steer any of them. GetRecommendedParallelThreads() is now deprecated; the default implementation of the new method still honors an overridden GetRecommendedParallelThreads(), so existing custom Compressors keep working. CompressorWrapper forwards the new method; built-in compressors are unchanged (picked up via the default bridge).
The block-based table builder now consults the data-block compressor (falling back to the base compressor before the data-block compressor exists, e.g. during dictionary sampling, where parallel compression does not start anyway) and applies the possibly-overridden options. While here, the builder records the effective (post-override) options in the rocksdb.compression_options table property instead of the pre-override configured options, and the two affected Rep members (compression_parallel_threads, max_compressed_bytes_per_kb) are set exactly once from the effective options -- with poisoned defaults so any use before they are computed surfaces -- rather than initialized in the member list and overwritten later.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15177
Test Plan:
Added DBCompressionTest.CompressorOverridesCompressionOptions, which installs a Compressor that overrides MaybeOverrideCompressionOptions() to force parallel_threads and verifies via the builder's sync point that parallel compression starts with the overridden thread count, then reads the data back. The existing GetRecommendedParallelThreads and
ConfiguredCompressionTypeRecordedInProperties tests continue to cover the deprecated bridge and the table-property recording respectively.
Reviewed By: joshkang97
Differential Revision: D118656406
Pulled By: pdillinger
fbshipit-source-id: 8a5d7cde90ca38913525a7ac11650607fa0bce58
Summary:
Let the built-in binary, hash, and partitioned indexes be created through `IndexFactory`. The wrappers preserve the existing internal-key handling and multi-block finish path.
Pass the common-prefix flag to every builder, including both levels of a partitioned index. The regression test checks the prefix encoding and decoded keys with the flag on and off.
The factory tests pass on the original PR plus the fix. Focused tests on the full stack also pass with `ASSERT_STATUS_CHECKED=1` and `COERCE_CONTEXT_SWITCH=1`. Full `make check` runs had test timeouts and hit the 30-minute local limit.
Part 4 of 9. Previous: https://github.com/facebook/rocksdb/issues/14960. Next: https://github.com/facebook/rocksdb/issues/14962. The first three parts have landed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14961
Reviewed By: jaykorean
Differential Revision: D119493650
Pulled By: xingbowang
fbshipit-source-id: 7d43cc456de86e8bd35e4b0689fb3d256608cf2f
Summary:
Why this change is required:
Block iterators parse persisted SST bytes. Truncated headers, malformed restart arrays, and invalid key or value lengths are runtime corruption, not private control-flow failures. They must be rejected safely in release builds before any out-of-range read, pointer arithmetic, or Slice construction.
Previous behavior:
DecodeEntry relied on an assertion before reading its three-byte fast-path header. Strict entry-span validation was enabled only for selected meta-block paths. Restart offsets were not all validated up front, and several separated-key/value bounds assumptions were asserted only after pointers or slices had already been constructed. Because assertions disappear under NDEBUG, malformed blocks could trigger out-of-bounds access or incorrect iteration in release builds.
New behavior:
All block iterators now reject truncated entry headers, keys or inline values extending past the key section, invalid separated-value offsets or lengths, and malformed restart arrays with Status::Corruption. Block construction validates every restart offset for section bounds and strict ordering before an iterator can use it.
Implementation:
Remove the StrictCheck template distinction and apply the span checks to every BlockIter. Validate restart keys before constructing their Slice, validate separated values before pointer arithmetic, and scan the restart array once during Block construction. Mark corruption-only branches UNLIKELY and combine the inline key/value span into one overflow-safe uint64_t comparison so valid data remains on the compiler hot path.
Safety boundaries and unsupported cases:
These checks are unconditional because they are required for safe parsing of persisted bytes; they are not gated by paranoid_checks or checksum verification. Existing assertions covering private iterator state and post-validation relationships remain. The change does not alter the SST format or repair corrupted blocks; it detects malformed blocks and reports corruption.
Performance:
Both revisions were explicitly cleaned and rebuilt with DEBUG_LEVEL=0, then measured on one pinned AMD EPYC Genoa CPU using 2 million 16-byte keys, 32-byte values, 4 KiB blocks, restart interval 16, and no compression. Hardware cycle counts are candidate versus parent, with lower being better:
- Cached random reads: 47.68B versus 47.69B cycles (-0.03%).
- Random seeks plus forward/reverse scans and batched MultiGet: 70.26B versus 69.96B cycles (+0.43%).
- mmap random reads without block cache: 16.92B versus 17.19B cycles (-1.55%).
- Separated-KV random plus sequential reads: 30.44B versus 31.20B cycles (-2.43%).
No tested workload regressed. These are targeted single-thread CPU microbenchmarks rather than a guarantee for every production workload.
Tests performed and known limitations:
The corruption tests cover truncated headers, entries extending past key or value sections, invalid separated-value offsets, and out-of-range or non-increasing restart offsets in both inline and separated-KV layouts. They verify detection rather than recovery or repair of a corrupted block.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15200
Test Plan:
Built block_test and ran the 14 BlockEntryCorruption and separated-KV corruption tests under ASSERT_STATUS_CHECKED=1 and for 100 repetitions with COERCE_CONTEXT_SWITCH=1. Ran make check-sources.
Ran make clean and built db_bench with DEBUG_LEVEL=0 for both this commit and its parent. Compared cached random reads, random seeks, forward and reverse scans, batched MultiGet, mmap reads without block cache, and separated-KV variants using the configuration documented above.
Reviewed By: pdillinger
Differential Revision: D119668019
Pulled By: xingbowang
fbshipit-source-id: 1594379acefa047f9598df341e675340831066ee
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15188
In the pointer branch of `lossless_cast`, the destination type check tested
`std::is_enum_v<To>` instead of `std::is_enum_v<ToDeref>`. Since `To` is a
pointer type in that branch, `std::is_enum_v<To>` is always false, so the
check collapsed to `std::is_integral_v<ToDeref>` and rejected casting to a
pointer-to-enum, e.g. `lossless_cast<MyEnum8*>(&some_uint8)`.
Reviewed By: xingbowang
Differential Revision: D119422904
fbshipit-source-id: b1f4e205d536ba3d10563fa62adb04867fae02ac
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15086
Why: An application built on top of RocksDB could use cache population reads through explicit snapshots and needs to know when the returned value has already been superseded, without a second point read.
Previous behavior: Get and MultiGet returned the snapshot-visible value and optional timestamp but could not report a newer point update or covering range deletion.
New behavior: experimental C++, C, and Java metadata-read APIs report newer committed puts, deletes, merges, and covering range tombstones; ordinary primary-DB writes completed before the read are guaranteed to be reported, racing writes remain best-effort, and true has no false positives.
Implementation: metadata reads widen lookup behind read-callback hooks while keeping metadata-only state off passive common object layouts; per-key state stays in KeyContext/GetContext, ordinary MultiGet row-cache replay semantics are preserved, BlobDB/TTL use shared output helpers, dead single-CF plumbing and an invalid cross-key stress assertion are removed, and the range-tombstone ordering invariant is documented without changing behavior.
Safety and unsupported cases: tracking is opt-in and requires an explicit snapshot; persisted-tier reads, custom transactional visibility callbacks, secondary DBs, and unsupported subclasses return NotSupported, while read-only/compacted DBs return false because their in-process views are static.
Tests and limitations: 1,265 Buck tests plus focused metadata, 100x coerced-context-switch row-cache, write-committed/write-prepared crash tests, builds, and lint passed. Current-parent release db_bench found no material passive regression: readrandom +0.65%, MultiGet batch 4 -1.40%, and batch 32 +0.48%.
Reviewed By: pdillinger
Differential Revision: D104182397
fbshipit-source-id: 36f7be1cd50fe2383689adaae31a6bb7d7e462dc
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15197
**What**: Compare the excluded-file lookup iterator against the engine it was obtained from, in `BackupEngineImpl::RestoreDBFromBackup()`.
**How**: The loop walks `locked_restore_from_dirs` calling `find()` on each engine's `backuped_file_infos_`, but tested the result against `this` engine's `end()`. Test against the same engine's `end()`.
**Why**: Comparing iterators obtained from different containers is undefined. It is inert here rather than a live bug: on libstdc++ and libc++ an `unordered_map` end iterator is a null sentinel that is not container-specific, so a miss compares equal and a hit does not. On an implementation with a container-specific end iterator - MSVC, a required CI axis - a miss would read as found and dereference the sentinel node. No test can pin this; see the test plan.
Reviewed By: xingbowang
Differential Revision: D119585045
fbshipit-source-id: 9ff0e770b426f6a2c4a0fdad261910a927d1bfa7
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15189
Add `CoroDB::CoGet` and `CoMultiGet` overloads for standalone `SstFileReader` instances. Reuse a shared sync/coroutine implementation on the private reader `Rep`, configure and validate the read executor at `Open`, and retain synchronous fallback when unavailable.
Reviewed By: xingbowang
Differential Revision: D119431444
fbshipit-source-id: 1ac1926b6a2ffc11cd3e91a438a8c62cff5d57ec
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15196
**What**: Fix heap corruption and out-of-bounds reads in `SstFileReader::MultiGet()` when the request exceeds `MultiGetContext::MAX_BATCH_SIZE` keys.
**How**:
- Batch the lookup in `MAX_BATCH_SIZE` slices over the sorted keys, with a fresh `MultiGetContext` and `TableReader::MultiGet()` call per slice. Caller ordering survives because each `KeyContext` already points into the caller's status and value vectors.
- `reserve()` the `key_context`, `get_ctx` and `merge_ctx` autovectors, as `DBImpl::MultiGetCommon` does: the setup loop takes addresses of their elements, and past its inline capacity an `autovector` spills into a `std::vector` that reallocates as later keys are appended.
- Only the slicing is borrowed from `DBImpl::MultiGetImpl`. Its cross-batch value-size accumulation and per-batch deadline check are deliberately not replicated: value size is accumulated only under `db/` and never by a `TableReader`, and the deadline already reaches every read through the table reader's I/O options.
**Why**: `MultiGetContext` caps at `MAX_BATCH_SIZE` keys but only asserts it, so release builds hit a bad `delete[]` from its mismatched `LookupKey` construct/destroy bounds, an out-of-bounds read of its fixed-size `sorted_keys_` array, and an out-of-range shift once the key count exceeds the width of its range mask; `include/rocksdb/sst_file_reader.h` documents no key-count limit, so any caller of this public API could reach it.
Reviewed By: xingbowang
Differential Revision: D119401930
fbshipit-source-id: 3bbafd72badaa944dbb94f4ab6ecb7f14a11f6d0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15101
This diff contains all changes under fbcode/internal_repo_rocksdb/repo for the RocksDB regression benchmark observability work.
It adds opt-in per-interval latency percentile reporting to db_bench. When --report_interval_percentiles is enabled with histogram and a positive stats interval, each worker records into pre-created per-operation interval histograms and counters. A background reporter periodically merges and resets those thread-owned stats, skips excluded background-writer stats, performs a final partial-interval flush, and emits IntervalPercentiles lines with sample count, operation count, and P50/P75/P99/P99.9/P99.99 latency. Worker updates and reporter snapshots use a double-buffer handoff, so the reporter never reads or clears a histogram still being updated. The legacy histogram timing path remains unchanged when interval reporting is disabled.
The interval implementation uses RocksDB UnorderedMap aliases for the hot lookup maps and UnownedPtr for non-owning reporter references.
It also aggregates each worker thread PerfContext and IOStatsContext into the final merged benchmark Stats and emits final PERF_CONTEXT and IOSTATS_CONTEXT blocks, so multi-thread benchmark metrics represent all workers rather than one thread.
For options-file runs, explicit benchmark-critical CLI overrides are applied after loading the file. This includes --disable_auto_compactions, fixing an existing precedence bug where disable_auto_compactions=false in the options file silently overrode an explicit CLI true value. The regression harness relies on that CLI setting, so the bug allowed automatic compactions during origin construction and write-heavy benchmarks. The regression test persists false in an options file, passes --disable_auto_compactions=true, reloads the opened DB options, and verifies the effective value is true.
One shared WriteBufferManager is created after option loading and reused by all DBs, enforcing the intended aggregate memory cap for multi-DB benchmarks.
The compactall benchmark now accepts --compact_all_parallelism. It compacts independent DBs through a bounded worker set, applies --subcompactions to each DB, emits per-DB progress and elapsed time, and propagates the first CompactRange failure after joining all workers. Its default parallelism remains 1 for backward compatibility.
The dependent harness diff enables these features, parses their output, and publishes the resulting interval and final metrics.
Reviewed By: pdillinger
Differential Revision: D116025321
fbshipit-source-id: d44185ba2110aacbe9fadc65a59799bf8c1c54f0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15179
A remote compaction worker (`DB::OpenAndCompact()`) opens the *live* primary's DB directory read-only, and that open reads `CURRENT`. The primary publishes each MANIFEST rotation by writing a new `CURRENT` under a temp name and renaming it over the old `CURRENT`. POSIX keeps an already-opened `CURRENT` readable after that replacement, but file systems without read-after-unlink semantics (remote or object stores) do not: the replaced object is reported as gone, surfacing in RocksDB as `IO error: No such file or directory` (`Status::PathNotFound`). The secondary open was not retried, so this transient race failed the whole compaction. There is an equivalent, already-recognized race one step later for the MANIFEST file (`ReactiveVersionSet::MaybeSwitchManifest()` converts it into `Status::TryAgain`), which was fatal here for the same reason.
`DB::OpenAndCompact()` now retries the secondary open (up to 3 attempts) when it fails with `TryAgain` / `PathNotFound` / `NotFound` -- exactly the statuses a concurrent `CURRENT`/MANIFEST replacement produces. The retry limit is configurable via `OpenAndCompactOptions::max_secondary_open_retries`. Genuine failures are still returned after the bounded retries, and no safety check is weakened or skipped.
Found by a `db_stress` crash test failure. Two non-default options in that run make this reachable and fatal:
* `remote_compaction_worker_threads=8` with `remote_compaction_failure_fall_back_to_local=0` -- the latter was only recently randomized to 0; before that a remote compaction failure was always silently retried locally. Now `DbStressCompactionService::Wait()` maps the failure to `kFailure`, which becomes a sticky background error on the primary, so every subsequent `Flush()` returns it and `StressTest::ProcessStatus()` aborts with `assert(false)`.
* `max_manifest_file_size=8192` -- the primary rotates MANIFEST (and therefore replaces `CURRENT`) constantly, which is what makes the race window hit in practice.
Full chain: primary renames a new `CURRENT` over the old one -> the file system reports the replaced object as gone for the reader that already resolved it -> `GetCurrentManifestPath()` inside the worker's secondary open fails with `IO error: No such file or directory` -> `DB::OpenAndCompact()` returns that error with no retry -> `kFailure` (no local fallback) -> primary background error -> `Flush failed: IO error: No such file or directory: ... /CURRENT ...` -> assertion failure in `db_stress`.
Reviewed By: joshkang97
Differential Revision: D118480192
fbshipit-source-id: a2618ee24e582e2e713af04464ee1e4038925ccb
Summary:
**What**: Fix an out-of-bounds read and a forged end timestamp in the timestamped `SstFileWriter::DeleteRange()` overload.
**How**:
- `Rep::DeleteRange()` now builds each boundary through a new `Rep::AppendTimestamp()` helper, which tests one key's adjacency to `timestamp` and either aliases the two or materializes `key + timestamp` into a caller-owned buffer.
- Drops the fast path's `assert`. It was never a guard: it performed the same out-of-bounds read it was meant to police, and `DeleteRangeImpl()` already asserts the two boundary timestamps match.
- Behavior change for callers whose two keys each carry their own trailing timestamp: they now take one extra copy. Aliasing is still used for whichever key is genuinely adjacent.
**Why**: A single `timestamp` buffer can be byte-adjacent to at most one of two distinct keys, but the old branch tested `end_key` using `begin_key`'s length and then widened both slices whenever either test passed - so an asymmetric caller, one key contiguous with the timestamp and the other bare, read past an allocation and persisted a range tombstone carrying foreign heap bytes as its end timestamp. Nothing forbids that layout: the public contract requires only that the keys carry no timestamp.
Reviewed By: xingbowang
Differential Revision: D119401932
fbshipit-source-id: 4c842d4afbe7bd7c4eabc68c6aee0543cdee2c9c
Summary:
`ShardedCache::ApplyToAllEntries()` clamped `average_entries_per_lock` with `std::min(aepl, 1)`, forcing every positive value (including the default 256) down to 1. The intent was a floor of 1 (`std::max`); `ApplyToSomeEntries` even asserts the value is > 0. As a result a full cache traversal advanced a single hash-table slot per iteration step instead of batching ~256, so the documented 256 default never actually took effect. The bug dates to the original API (https://github.com/facebook/rocksdb/issues/8225) and was carried through the ShardedCache refactor (https://github.com/facebook/rocksdb/issues/10801) unchanged, so the intended default was never exercised: in hindsight the "gather stats" latency in the https://github.com/facebook/rocksdb/issues/8225 test plan is a batch-size-1 measurement.
Despite its name, `average_entries_per_lock` is really a batch size for iterating the cache, not something specific to locking: it bounds the work done per uninterrupted step. For a lock-based cache (LRUCache) a shard mutex is held for each batch, so single-entry batches mean many extra lock acquire/release cycles; but lock-free implementations (HyperClockCache) also pay fixed per-batch overhead and benefit from the fix as well. This is a control-path cost (e.g. periodic block cache stats via DumpStats -> CollectCacheEntryStats), not hot-path read/write, so the practical impact is usually minor. The option's doc comment is updated to describe it as a batch size; the field is not renamed given how obscure the option is.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15186
Test Plan:
No unit test: asserting on batch/lock-acquisition counts would be intrusive/awkward, and LRUCache is disfavored (though not deprecated).
Measured with release-mode cache_bench, comparing the binary built before (`std::min`) vs after (`std::max`); 4 GiB cache, 1 KiB values, ~3.7M entries, default gather_stats_entries_per_lock=256. Average full ApplyToAllEntries traversal latency:
| cache_type | before | after | speedup |
|-------------------------|---------|---------|---------|
| lru_cache | 2.69 s | 0.64 s | ~4.2x |
| fixed_hyper_clock_cache | 0.441 s | 0.138 s | ~3.2x |
| auto_hyper_clock_cache | 0.264 s | 0.136 s | ~1.9x |
All implementations improve; LRUCache most, due to per-batch mutex acquire/release. HCC traversals are already ~6-10x cheaper than LRU in absolute terms, and entries visited are identical before and after.
Reviewed By: rban1
Differential Revision: D119256456
Pulled By: pdillinger
fbshipit-source-id: e58f7f97d47d8970ebc8c3ab786199132ce6340d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15181
A crash test failed when `use_async_db_api=1` created read-executor workers on a host unable to allocate io_uring rings. Folly logged the setup failure and threw; RocksDB caught it and successfully fell back to synchronous reads, but the harness failed because stderr was non-empty.
Representative stderr:
```
WARNING: Logging before InitGoogleLogging() is written to STDERR
I0903 14:05:39.470695 286445 IoUringBackend.cpp:527] io_uring_queue_init_params(512,1024) failed errno = 0:"Success" 0x7d2e06295280 retrying with capacity = 512
E0903 14:05:39.470759 286445 IoUringBackend.cpp:537] io_uring_queue_init_params(512,512) failed ret = -12:"Unknown error -12" 0x7d2e06295280
```
Initialize glog in the Folly-enabled OSS `db_stress` entry point and centralize known stderr sanitization. Only `IoUringBackend` ring-initialization failures are moved to stdout when the async DB API is enabled; existing expected post-SIGTERM errors use the same sanitizer. All unrelated stderr remains fatal.
___
Differential Revision: D118836521
fbshipit-source-id: c9ce05896f417c7723c5cd73d293fbb1cd14ee1e
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15142
Coroutine statistics use TLS, but coroutine reads can interleave on the same executor thread. Previously, when a stats-enabled request suspended, it saved its counters without disabling the executor's TLS configuration. A stats-disabled request running next could inherit those enabled settings and collect statistics unexpectedly. Stats setup also lived in individual DB implementations, so wrapper early returns and future stackable DB implementations could bypass it.
Move statistics ownership to the public `CoroDB` and callback-based async read boundaries, before virtual dispatch. Each operation captures its caller's configuration, enabled requests preserve their counters across suspensions, and every suspension or completion leaves executor TLS disabled. Because the scope wraps stackable DB dispatch, current and future `CoroStackableDB` implementations automatically inherit the correct behavior without adding their own stats reset or scope.
For consistency between coroutine and callback-based reads, the experimental callback API now uses TLS for both configuring and consuming statistics. `AsyncCallback::OnComplete()` no longer receives context arguments; RocksDB publishes the completed per-operation counters to TLS before invoking it.
```
Before
time -------------------------------------------------------------->
stats-on A: onSet(enable A) -- work -- onUnset(save A) ...... onSet(A)
executor TLS: [A enabled] ------------> [still enabled] ------> [A enabled]
stats-off B: run
^ inherits A's enabled TLS
After
time -------------------------------------------------------------->
stats-on A: onSet(enable A) -- work -- onUnset(save A, disable) ... onSet(A)
executor TLS: [A enabled] ------------> [disabled] ----------------> [A enabled]
stats-off B: run
^ remains stats-disabled
```
Reviewed By: xingbowang
Differential Revision: D117017089
fbshipit-source-id: 8a85f7ba43e3b73225995f5d0b79da3892baafb9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15170
`warm_storage_tsan_liveness_crash_test` fails with
```
error : Get() returns IO error: Failed to read 1032 bytes from file
ws://ws.zdb.ldc0zdbtest1/.../000093.sst from offset 5461, message
[0 code=DEADLINE_EXCEEDED isRetryable=1 ... error=WSE_BLOCK_READ_TIMEOUT
classification=TRANSIENT_ERROR ...] for key: 0000...78 (30280).
Verification failed :(
```
The Warm Storage error is explicitly `classification=TRANSIENT_ERROR isRetryable=1`, and the DB artifact confirms there is no RocksDB problem: `000093.sst` is intact (37760 bytes, downloaded successfully after the run) and the DB `LOG` reaches `Shutdown complete` with zero IO-error or corruption entries. This is exactly the class of failure `--tolerate_non_injected_io_errors_for_remote_dbs` was added for (D112136438), but that mechanism never actually took effect. Two independent defects:
**1. The triggering non-default option: `tolerate_non_injected_io_errors_for_remote_dbs=0` on a remote DB.**
`db_crashtest.py` hard-coded the parameter to `0` in `default_params` and only ever *force-disabled* it (for local DBs), leaving each remote job to opt in on its own command line. `rocks/sandcastle/test-driver.sh`'s `warm_storage_*crash*` branch never passed it, so every Warm Storage crash-test job ran with the flag off and a single transient read timeout failed the run. Fix: the parameter now defaults to `1` whenever `--env_uri` / `--fs_uri` names a remote backend, so no remote job has to remember to opt in. A caller can still pass `0` explicitly, and `finalize_and_sanitize()` still force-disables it for local DBs so genuine local IO errors (e.g. io_uring failures) are never masked.
**2. `injected_error_count == 0` made the flag a no-op in `no_batched_ops_stress.cc`.**
All five read-path guards were
`injected_error_count == 0 || !IsErrorInjectedAndRetryable(s)`.
`IsErrorInjectedAndRetryable()` covers *non-injected* remote IO errors, which by definition have no injected fault to correlate with — and the `liveness` profile forces every `*_fault_one_in` to `0`, pinning `injected_error_count` to `0` for the whole run. So the left operand short-circuited to true and the error was reported as a verification failure regardless of the flag. Fix: extract `StressTest::IsTolerableNonInjectedRemoteIOError()` and add `StressTest::IsRetryableOperationError(injected_error_count, s)`, which gates the injected-error branch on `injected_error_count > 0` but lets a tolerated non-injected remote IO error through unconditionally. The five call sites now use the helper.
With the tolerate flag off (all local runs), the new guard reduces algebraically to the old expression, so local behaviour is unchanged. The genuine consistency-check gate at `no_batched_ops_stress.cc:1146` (`do_consistency_check && injected_error_count == 0`) is intentionally left alone.
Fixes T286575055.
[Session trajectory link](https://www.internalfb.com/intern/devai/devmate/inspector/?id=9646c050-6791-4ac3-9abb-3ee110dcb54e)
Reviewed By: pdillinger
Differential Revision: D117886823
fbshipit-source-id: d7591472db8ee6e03c9d9f018480409b6d802838
Summary:
Remote compaction writes output SSTs into caller-selected job directories. If the primary crashes before installation or falls back to local compaction, those directories can outlive the DB session and accumulate under the DB directory.
This change defines `kSessionTmpDirName` as `session_tmp`, the canonical name for a DB-local temporary directory whose contents have the lifetime of a DB session. The immutable boolean option `use_session_tmp_dir_for_remote_compaction` opts remote compaction into this directory. When enabled, an `OpenAndCompact()` directory name is resolved as `<dbname>/session_tmp/<job>`.
During DB open, RocksDB creates the session temporary directory, completes recovery, and then synchronously deletes contents left by the previous DB session before scheduling compactions. `DestroyDB()` removes it as part of database deletion, and normal completion, failure, and cancellation paths clean the resolved per-job directory.
The option defaults to false. Disabled callers retain the existing `<dbname>/<job>` layout and no session-directory cleanup. For rollout compatibility, legacy callers that already pass a DB-rooted output path are accepted unchanged only while the option is false. Directory creation and invalid-path errors are delegated to the configured filesystem rather than prevalidated by RocksDB.
After D118301531 landed, per-job directory creation and resumption cleanup are kept together in `InitializeCompactionWorkspace()`. `OpenAndCompact()` resolves the client-provided name and expects DB open to have initialized the session temporary directory when the feature is enabled.
The boolean is wired through C++ and generated C APIs, option serialization and reconstruction, `OpenAndCompact()`, ldb, db_bench, and db_stress. Tests cover path resolution, cross-session deletion, ownership boundaries, disabled behavior, `DestroyDB()`, configured compaction output, resumption fallback, and option round trips.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15169
Reviewed By: pdillinger
Differential Revision: D118227112
fbshipit-source-id: 01aa834b0b3f75638102130a86d95d537a2797bd
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15173
Add an opaque, non-owning `ReadOptions::custom_context` pointer that is passed to table implementations so external table readers can consume application-defined read options. Document its lifetime and synchronization contract, and keep the unsupported pointer out of generated C bindings.
Reviewed By: xingbowang
Differential Revision: D118341098
fbshipit-source-id: e51185f58b5ce80d8fd07efab02a05dbc8c7bd12
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15143
## Summary
Adds a RocksDB blog post describing the native async/coroutine read APIs introduced in RocksDB 11.10.0. It covers the callback and `folly::coro::Task` interfaces, differences from `ReadOptions::async_io`, executor and I/O-completion handoffs, filesystem integration, request-local statistics across suspension, and API lifetime constraints.
The post includes three explanatory diagrams and benchmark results comparing synchronous and coroutine point reads across task counts and `MultiGet` batch sizes.
https://www.internalfb.com/intern/px/p/cDGqv/
Reviewed By: xingbowang
Differential Revision: D117159178
fbshipit-source-id: 5452e6b07e29abb5594505021995e9b5bac9d198
Summary:
Follow-up to the "UPDATE" note on D116738823 (PR https://github.com/facebook/rocksdb/issues/15133, "Faster and safer MANIFEST recovery for remote compaction (OpenAndCompact)"). That change observed that the crash test does not really cover remote compaction because `remote_compaction_failure_fall_back_to_local` is pinned to 1 in db_crashtest.py, so any remote compaction failure is silently retried locally instead of failing the test -- masking existing failures. Addressing that was called out as a high-priority follow-up (tracked under T232588967, "Rocksdb Crash Test failed: Remote Compaction").
This does the cited enablement and fixes the first masked failure it surfaces:
* Crash test: randomize `remote_compaction_failure_fall_back_to_local` between 0 and 1 (the TODO(jaykorean)), so remote compaction failures actually surface instead of always being hidden by a local fallback.
* Fix a real OpenAndCompact bug the enablement immediately exposed. When a compaction is scheduled with `allow_resumption=true` but resumption is internally disabled because it is incompatible with output-hash verification (`paranoid_file_checks=true`, or `verify_output_flags` requesting iteration verification), the worker started a fresh compaction but did NOT clean the output directory. The caller, having requested resumption, does not empty the directory either, so a compaction that was canceled after writing some output and then retried reused the same output file numbers and collided with the leftover files -- surfacing under the crash test's fault-injection FS as `NewWritableFile violates no-reopen-for-write contract`, which became a background error and failed subsequent writes. The fix honors the documented `OpenAndCompactOptions::allow_resumption=true` fallback contract: when resumption is requested but cannot be fulfilled, clean any leftover progress and output files before starting fresh (reusing the existing Scan/Cleanup helpers), so the fresh compaction always begins from a clean directory. When `allow_resumption=false`, the caller still owns emptying the directory (unchanged).
No known production exposure: this bug can only bite a caller that requests `allow_resumption=true` on a column family where output-hash verification is enabled (`paranoid_file_checks=true`, or `verify_output_flags` including `kVerifyIteration`), because only that combination disables resumption while the (uncleaned) output directory is left to the caller. Auditing how the option is used across internal deployments found nothing in that state: the deployments that enable resumption for remote compaction do not enable those iteration/paranoid verification modes, and where remote-compaction output verification is enabled at all it is block-checksum-only, which keeps resumption active and never hits the buggy path. So this is effectively a latent-bug fix -- it prevents a failure that would appear if iteration/paranoid output verification were later turned on for a resumption-enabled remote-compaction fleet.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15168
Test Plan:
Added compaction_service_test.cc unit test
`CleansOutputDirWhenResumptionDisabled`, which drives OpenAndCompact with `allow_resumption=true` and `paranoid_file_checks=true` after seeding a leftover output SST, and asserts the leftover is cleaned and the compaction succeeds. It was confirmed to FAIL without the fix (leftover retained) and pass with it.
Local crash test runs (blackbox, `remote_compaction_worker_threads=8`, `remote_compaction_failure_fall_back_to_local=0`):
- Without the fix, the collision-trigger config (`allow_resumption_one_in=1 --paranoid_file_checks=1`) fails within the first ~60s iteration on the no-reopen-for-write contract violation.
- With the fix, that same config ran ~10 min clean, and a full-random config (everything else randomized) ran ~15 min clean, both with 10-15 kill/reopen cycles plus a final crash-recovery verification. More local crash test runs are planned to shake out any remaining, rarer masked failures now that the fallback no longer hides them.
Reviewed By: jaykorean
Differential Revision: D118301531
Pulled By: pdillinger
fbshipit-source-id: ceb9683ae06ce930ea78bafa3585ad616012243f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15163
Reintroduces the `optimize_manifest_for_recovery` optimization for `OPTIONS-*` and temp files without letting `VersionSet::NewFileNumber()` collide with files already present in the DB directory.
Recovery now distinguishes between files that require a persisted `SetNextFile` edit and files that only need the in-memory allocator reserved. If only `OPTIONS-*` or temp files are above the recovered `next_file_number_`, recovery advances `next_file_number_` in memory and still skips the MANIFEST edit.
Reviewed By: jaykorean
Differential Revision: D117887276
fbshipit-source-id: f211fa90303904befdf331242ba0d97553f84169
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15155
SST block writing now hands checksum coverage across the table-builder to file-writer boundary when the block checksum algorithm is CRC32C.
The block builder passes the raw CRC32C of the block contents to `WritableFileWriter::Append()`, which validates the copied destination-buffer contents and accumulates that checksum into the filesystem handoff checksum instead of recomputing it after the copy. The trailer is still appended separately; table building computes and passes the trailer handoff checksum independently, while preserving the existing persisted block-checksum encoding. Non-CRC32C block checksum algorithms continue to use the existing file-writer checksum path.
Destination-buffer validation failures return a stage-specific corruption `Status` that includes the file path. Flush and compaction propagate that status through their existing background error handling, so the same message is logged without adding local duplicate log sites.
The SST tests record the builder append stream through sync-point callbacks while writing to a null sink, then replay those exact append calls into a file-backed writer to cover successful handoff, block-content corruption, compression-type corruption, trailer-checksum corruption, non-CRC32C fallback behavior, and a small randomized boundary-injection loop.
Reviewed By: xingbowang
Differential Revision: D116969191
fbshipit-source-id: f75dcb37936106b7f6f14cf1ed999c81a0879d7b
Summary:
Instrument partitioned Bloom-filter block retrievals with read_filter_block_nanos so PerfContext includes time spent looking up these blocks in the block cache or reading them from SST files.
Root cause:
The top-level filter-partition index is read through ReadFilterBlock, which already owns the timing guard. Individual filter partitions bypass that helper and call RetrieveBlock directly. Once the top-level index is cached, workloads can therefore report filter block reads and cache hits while read_filter_block_nanos remains zero.
Implementation:
Place the timer at the partition retrieval boundary, after the pinned filter-map fast path and immediately around RetrieveBlock. This keeps the metric aligned with actual cache or SST retrieval work without charging accesses to already-pinned partition objects.
Testing:
Extend the partitioned MultiGet test to enable time-level PerfContext collection and verify that partition reads populate read_filter_block_nanos across whole-key and prefix-filter configurations and table format versions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15164
Test Plan:
- `make check-sources`
- `build_tools/rocksptest.sh db_bloom_filter_test`
- `ASSERT_STATUS_CHECKED=1 build_tools/rockstest.sh db_bloom_filter_test --gtest_filter=*PartitionedMultiGet*`
Reviewed By: rban1
Differential Revision: D118052826
Pulled By: xingbowang
fbshipit-source-id: e4467b107ba13e5cfe4826a8a93834b319a131a5
Summary:
Remote compaction passes snapshot sequence numbers through serialized `CompactionServiceInput`. Downstream compaction logic assumes the sequence numbers are strictly increasing when selecting the earliest snapshot and using `std::lower_bound()` for snapshot visibility. Previously, malformed input could bypass the debug-only ordering assertion in release builds and select an incorrect visibility boundary.
- Validate decoded snapshot sequence numbers before remote compaction setup and return `InvalidArgument` for duplicate or descending values.
- Document the strict ordering contract while retaining the `CompactionIterator` assertion as defense in depth for internally constructed inputs.
- Canonicalize randomized serialization-test snapshots and add coverage for malformed snapshot lists through `DB::OpenAndCompact()`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15160
Test Plan:
- Built `compaction_job_test` and ran `CompactionJobTest.OpenAndCompactRejectsUnorderedSnapshots` in the normal configuration.
- Ran the focused test with `ASSERT_STATUS_CHECKED=1`.
- Ran the focused test for 100 repetitions.
Reviewed By: rban1
Differential Revision: D117888826
Pulled By: xingbowang
fbshipit-source-id: 2312205acfe17271c2083808d372c93a5914114c
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15136
`WritePreparedTxnDB` overrides `Get` and `MultiGet` to install a
`WritePreparedTxnReadCallback` that performs visibility checking for
prepared-but-uncommitted data. However, `GetEntity` (both the single-CF
`PinnableWideColumns` variant and the multi-CF `PinnableAttributeGroups`
variant) was never overridden, causing it to fall through to
`StackableDB::GetEntity` which delegates to the inner `DBImpl` without
any WritePrepared visibility callback.
This means `GetEntity` could read prepared-but-uncommitted data, violating
snapshot isolation. The bug surfaces in the crash test with the combination:
`--use_txn=1 --txn_write_policy=1 --use_attribute_group=1 --use_get_entity=1
--unordered_write=1 --two_write_queues=1 --test_batches_snapshots=1`
The fix adds `GetEntity` overrides to `WritePreparedTxnDB` that create a
`WritePreparedTxnReadCallback` and pass it via `GetImplOptions::callback`,
matching the pattern used by the existing `Get` override.
Reviewed By: xingbowang
Differential Revision: D116324596
fbshipit-source-id: faf636cfbf093f47a645c1a5d5037baa02ff594a
Summary:
Failures in GitHub Actions CI workflows attempting to download assets from releases.pagure.org occur because the Pagure platform has been decommissioned. Maintainers must update their configurations to retrieve packages from active mirrors, alternative repositories.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15159
Reviewed By: rban1
Differential Revision: D117863179
Pulled By: xingbowang
fbshipit-source-id: e54e80aa6f1e2e0503dbd7ec16e67b86c4bad7ad
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15158
Reverts `https://github.com/facebook/rocksdb/pull/15106`, which ignored `OPTIONS-*` and temp files when deciding whether recovery must advance `next_file_number_`.
Crash-test failures show backup/checkpoint failing to copy the current `OPTIONS-*` file after options cleanup, because `DeleteObsoleteOptionsFiles()` ignores `next_file_number_` and `options_file_number_` when choosing which files to keep. The failing commands use `optimize_manifest_for_recovery=1`, so restore the previous conservative file-number reservation behavior while we investigate a narrower fix.
Reviewed By: xingbowang
Differential Revision: D117807052
fbshipit-source-id: 89f8656f045be4ac36d2ec851980573ed83370fe
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15154
**What**: Let `IngestExternalFileArg::atomic_replace_range` replace a half-open `[start, limit)` range that only partially overlaps existing SSTs, by committing a generated range tombstone alongside the replacement files.
**How**:
- Tombstone: `NumFilesToPrepare` reserves one extra file number and `PrepareAtomicReplaceRangeTombstone` writes the SST only when the pinned `Version` holds a straddling file (`HasPartialOverlap`), so the no-straddle case pays no extra write; run rescans and returns `TryAgain` if a straddler appeared since. On activation it is rotated to the front of `files_to_ingest_` so one `VersionEdit` lands `old < tombstone < data` seqnos, straddlers are retained, contained files metadata-deleted, and `generated_for_ingestion` keeps `move_files` from unlinking it.
- Half-open plumbing: `RangesOverlapWithMemtables`, `RangeOverlapWithCompaction`, `OverlapProximalLevelOutputRange`, and `IsRangeOverlapped` take a required limit-exclusivity argument, and `largest_internal_key` now comes from `RangeTombstone::SerializeEndKey()`. `OverlapWithLevelIterator` stays inclusive: over-counting can only push the tombstone one level shallower, which is safe.
- Eligibility: `SupportsAtomicReplaceRangeTombstone` requires universal compaction, a delete-range-capable table factory, `allow_global_seqno`, `!allow_db_generated_files`, `!snapshot_consistency`, and no UDT. `fail_if_not_bottommost_level=true` rejects partial overlap because the retained file forces the tombstone and replacements above Lmax. The `atomic_replace_range` doc comment and a new `unreleased_history/public_api_changes/` note state the half-open contract.
- `ResetTableReader` passes `kNoFileChecksumFuncName` in `FileOptions` for external files, making "no checksum metadata" explicit rather than empty.
- Stress: new `ingest_external_file_atomic_replace_one_in` gflag. `PickAtomicReplaceRange` picks a run of overwriteable keys strictly inside an existing SST so every replacement straddles, writes even-indexed keys and marks the interleaved odd-indexed keys deleted so the tombstone must clear the holes, and rolls back expected values on `TryAgain`. `db_stress_tool.cc` and `db_crashtest.py` gate the flag on the same eligibility set and force `acquire_snapshot_one_in=0`.
**Why**: Atomic replacement previously required range bounds to align with SST bounds, and the public doc admitted the upper bound "may be interpreted as inclusive or exclusive". The tombstone overlay removes the alignment restriction and pins the contract to half-open without slowing the Nimble path or the no-straddle case.
Reviewed By: pdillinger
Differential Revision: D117155307
fbshipit-source-id: 65f45463e8082a7db7a9d44c8dc50debe95e36e7
Summary:
`periodic_compaction_seconds` softly guarantees every SST is rewritten at least
once per interval. Across a fleet of similarly-configured DBs, many events line
those intervals up in lockstep -- a shared enable, a coordinated restart (e.g.
DC power-outage recovery), or just similar write history -- so the periodic
compactions all come due together, a recurring "thundering herd" that is
especially painful with remote compaction (SEV S698635).
The obvious fixes trade one problem for another. Spreading the herd out means
some files get compacted off their natural schedule, which is wasted work; and a
clumsy spread just relocates the herd or spawns a new one. So the real goal is
narrow: **de-herd while keeping "badness" -- extra and/or late compaction --
small, and without inducing fresh herding at any step.**
This change adds an EXPERIMENTAL, opt-in mechanism that gives each `(DB, column
family)` a stable, well-spread *preferred phase* within the interval and steers
periodic compactions toward it. It is randomized and *uncoordinated* (DBs never
talk to each other), which both keeps it simple and, as it turns out, de-herds
far more cheaply than adding jitter to the interval. It is controlled by a single
mutable DBOption and is **off by default**:
- `periodic_compaction_phase_recovery_percent` (default `0` = disabled / kill
switch; recommended `33`; clamped to `[0, 100]`) -- how aggressively each
trigger converges toward the phase, as a percent of the remaining gap, only
ever pulling the trigger *earlier*. `0` reproduces the classic scheduler
exactly; `100` jumps to the phase in one cycle; lower values approach gradually
with much less extra work.
The preferred phase itself needs no tuning: it is derived from the DB ID (stable
across re-open and physical cloning/migration, but distinct across DBs) and each
column family is offset from the DB's base phase, so the CFs of one DB don't fire
together. (An earlier draft exposed a `compaction_schedule_seed` string option
for choosing the phase source; it was over-engineered for the one behavior anyone
needed, and as a mutable string it also broke OPTIONS-file round-trip
verification in crash testing. It has been removed in favor of always using the
DB ID.)
## The two kinds of badness
"Badness" here is the fraction of compaction work that is ideally unnecessary:
over-compaction (rewriting a file earlier than the interval required) plus a
penalty for lateness (going overdue past the interval). Steady state is 0; both
"never compact" and "compact twice as often as needed" approach 1. De-herding
inherently spends a little badness to spread load out; every design choice below
exists to bound one of those two costs *without* re-clustering the fleet.
The scenario names cited in the design bullets below (`upgrade_dynamic_r33`,
`naive_upgrade_r25`, `burst_r*`, `turndown_r33`, `turnup_r33`, ...) all come from
the out-of-tree simulator; see its results page (the "Simulator and
visualization" section below, with an interactive plot per scenario) for the
evidence behind each claim.
## Design
- **A stable per-(DB, CF) preferred phase = one fixed attractor per DB.** Each DB
gets a base phase from its DB ID; each CF is offset from it. Because the target
is fixed (not re-rolled each cycle) and unique per DB, the fleet converges to a
uniform spread and then compacts *on time* -- badness decays toward ~0. This is
the crucial contrast with naive per-cycle jitter, which de-herds but never stops
jittering and so pays a permanent over-compaction cost (see
`naive_upgrade_r25`: flat ~12-14%, only diffusive spreading) vs.
`upgrade_dynamic_r33`, whose badness decays to ~0 for the same anti-herding.
- **Converge gradually, and only ever earlier.** Each trigger closes a
`recovery_percent` *fraction* of the gap to the phase, reaching it over a few
cycles rather than in one burst, and never later than
`periodic_compaction_seconds` in steady state. A lower rate does less extra
work, which matters when the phase keeps moving (write-driven / bursty
workloads), where a high rate wastefully "chases" it -- the burst sweep shows
`burst_r25 < burst_r33 < burst_r50 < burst_r100` in sustained badness. `33` is
the default balance: quick spreading, low extra work.
- **Fractional recovery (not "snap") is latency-tolerant and doesn't re-herd.**
Closing a fraction of the gap gives a single stable convergence point, so
real-world scheduling/queueing latency -- the delay between a file becoming due
and its compaction actually running -- merely shifts the whole fleet uniformly
instead of re-clustering it. A tried "snap onto the phase when in range" variant
converged a bit faster with no latency but re-herded around the snap
threshold under latency; it was rejected. Graceful degradation beat peak speed.
- **Spread a suddenly-due cohort instead of bursting it (bounded lateness).**
Enabling the feature, turning `periodic_compaction_seconds` down, or catching up
after downtime can leave a whole cohort past due at the same instant. Firing it
all at once would just be the herd again, so such a cohort is spread over the
first quarter-interval (~N/4) after the phasing anchor on an *absolute*,
epoch-aligned per-(DB, CF) grid. This trades a bounded lateness (<= ~N/4 past
the deadline, so the interval is a *soft* upper bound during these transients)
for anti-herding, and the absolute grid keeps a rapidly restarting DB from being
perpetually re-deferred and starved. `turndown_r33` shows the past-due cohort
smeared over N/4 rather than bursting; `turnup_r33` confirms a turn-up creates
no past-due cohort (deadlines only grow) and so no burst. For immediate
enforcement at such a transition, trigger a manual compaction.
- **Off by default, with a kill switch.** Inert unless
`periodic_compaction_phase_recovery_percent > 0`; at `0` behavior is identical
to the classic scheduler.
## Implementation notes
- Phase math is integer-only (`Hash64` + `FastRange64`); floating point is avoided
for its pitfalls. The per-CF offset is a golden-ratio (Weyl/Kronecker) additive
sequence -- low-discrepancy for any number of CFs.
- Logic lives in `db/periodic_compaction_phaser.{h,cc}`
(`PeriodicCompactionPhaser`), which owns the recovery percent and the phasing
anchor and exposes the stateless trigger-time math; `VersionSet` holds one and
caches the per-CF result on each `Version`. The anchor is (re)set to now on DB
open, on a recovery-percent change, and on any `periodic_compaction_seconds`
change -- the three events that can create a past-due cohort.
- Plumbed through the C API, `db_bench`, `db_stress`, and `db_crashtest`.
## Simulator and visualization (out-of-tree, not part of this change)
A standalone simulator drives 40 real RocksDB instances through a real
periodic-task scheduler, records every compaction (periodic vs write-driven), and
scores the badness metric above across ~17 scenarios: dynamic/restart enable,
interval turn-down/turn-up via `SetOptions`, on-cadence writes, random write
bursts with a recovery-rate sweep, and the naive-jitter baseline. It is the basis
for the numbers cited above. Interactive HTML (**NOTE** fb-internal URL sanitization might require you to copy-paste the URL yourself):
https://htmlpreview.github.io/?https://github.com/facebook/rocksdb/blob/5759c6a62d9ff36cecf030b80453cc819b2f66ae/pc_phase_sim.html
Reviewed By: jaykorean
Differential Revision: D117487278
fbshipit-source-id: a62439548b7dd6fe7dfdbbe21b69531d8b57a1e8
Summary:
Leveled compaction combined with preserve_internal_time_seconds or preclude_last_level_data_seconds is a known-broken combination, but it is easy to enable by accident -- for example a fleet-wide config rollout that turns on the preserve time for CFs that happen to be leveled. Rather than degrading gracefully, it triggers an infinite kBottommostFiles compaction loop: a bottommost file gets marked, rewritten with no progress, and immediately re-marked, continuously rewriting the file and growing the MANIFEST so the file set never quiesces. This change makes such a misconfiguration a harmless no-op instead.
Before this change, the marking logic
(ComputeBottommostFilesMarkedForCompaction) marked a bottommost file whenever its largest seqno was below the oldest snapshot, without checking whether a compaction could actually zero that seqno. Recently written seqnos stay within the seqno->time "preserve" window and cannot be zeroed, so the rewrite never made progress. The same futile marking also occurred with user-defined timestamps when full_history_ts_low is unset. The underlying issue is that the "can this bottommost seqno be zeroed?" question was answered independently by the marking logic and by the compaction output path (CompactionIterator::PrepareOutput), and the two had drifted apart.
With this change, both paths answer that question through one shared predicate (BottommostSeqnoCanBeZeroed), and the preserve-window boundary that marking needs is plumbed into VersionStorageInfo from DBImpl, mirroring the existing oldest-snapshot plumbing. Snapshot visibility stays a per-site check. On CFs without preserve/preclude the predicate is a no-op. Confirmed that caller invariants ensure that relocating the `ikey_.sequence <= preserve_seqno_after_` check in CompactionIterator::PrepareOutput() will not trip the ROCKSDB_LOG_FATAL.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15146
Test Plan:
- compaction_picker_test: new CompactionPickerTest BottommostFileNotMarkedWithinPreserveWindow / BottommostFileMarkedBelowPreserveWindow cover the preserve-window boundary; new CompactionPickerU64TsTest BottommostNotMarkedWithEmptyFullHistoryTsLowAndUnknownMaxTs covers the UDT edge case.
- tiered_compaction_test: new PrecludeBottommostLoopTest LeveledPreserveTimeNoBottommostLoop is an end-to-end regression for the loop and the DBImpl plumbing: a bottommost file with a nonzero largest seqno inside the preserve window, made a marking candidate via a snapshot release, must not be marked. It reproduces the loop (hangs) when the plumbing is neutralized and passes with the fix.
Reviewed By: xingbowang
Differential Revision: D117299979
Pulled By: pdillinger
fbshipit-source-id: 51e8e2d4e2116984d16d3c7ef54a9635d5aded74
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15145
## Summary
RocksDB coroutine statistics are stored in `RequestContext`, but their `RequestData` is thread-affine. A filesystem implementation may propagate the context while completing `SubmitReadAsync` on another thread. Installing that context invokes `CoroutineStatsRequestData::onSet()`, whose `AssertSameThread()` detects the thread hop and aborts.
```
Before:
Thread A Thread B
RequestContext
├── caller tracing/attribution data
└── RocksDB coroutine stats ── propagated ──> setContext()
│
▼
CoroutineStatsRequestData::onSet()
│
▼
AssertSameThread(A != B)
│
▼
abort
```
Keep propagating the complete `RequestContext`, including the coroutine-stats token, so filesystem callbacks retain caller tracing and attribution. `CoroutineStatsRequestData` now records its owner using `Env::GetThreadID()` and ignores `onSet()` and `onUnset()` on foreign threads. When `ViaIfAsync` schedules the continuation back onto the RocksDB read EventBase, it restores the saved context on the owner thread before the coroutine continues, so the stats are safely loaded back into RocksDB TLS.
```
After:
Thread A Thread B
caller RequestContext
├── caller data ───────── propagated ───────> caller data
└── RocksDB coroutine stats ─ propagated ───> onSet()/onUnset()
│
▼
owner thread != current thread
│
▼
no-op
continuation queued back onto Thread A
│
▼
saved RequestContext restored on owner thread
│
▼
RocksDB coroutine stats reinstalled in TLS
```
Reviewed By: xingbowang
Differential Revision: D117274942
fbshipit-source-id: 9995cc3053a0a1d3a63d0cb2f4dd9320ab553005
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15123
Subset checkpoint post-processing appends column-family drop records to a copied MANIFEST. Recovery can tolerate an incomplete atomic group at the tail while retaining the reader high-water size, so appending at that size can produce a checkpoint that fails recovery.
Track the valid logical append boundary together with its MANIFEST file number, refresh it after recovery and successful writes, reset it across reactive MANIFEST switches, and trim only subset-checkpoint copies before appending drop records. Whole-DB checkpoints and the source MANIFEST remain unchanged.
Reviewed By: anand1976
Differential Revision: D116449394
fbshipit-source-id: a162dd6a2558f0f9ec40d878ee83956a89df698b
Summary:
Previously, during WAL replay, a secondary stored each 2PC prepare in
`recovered_transactions_`. Replaying the matching commit or rollback marker
deletes it. If the primary purges that marker's WAL before the secondary reads
it, the prepared `WriteBatch` and its tracker entry remain until the secondary
closes. This retention case was flagged by xingbowang in https://github.com/facebook/rocksdb/issues/15066. Even when
the marker is replayed, the secondary records completion but previously did not
prune the completed prefix from `logs_with_prep_tracker_`.
This change deletes recovered transactions whose prepare WALs are all below
`min_log_number_to_keep` after opening every discovered WAL reader but before
replaying records. The primary keeps that watermark at or below every
unresolved prepare and every prepare with unflushed committed data, so an entry
below it is committed and flushed, or rolled back. Deleting before replay lets
a reused transaction name start a new generation instead of merging into stale
state. If a reader cannot be opened, no recovered transaction is deleted.
After a round successfully replays every discovered WAL, the secondary also
prunes the completed prefix from `logs_with_prep_tracker_`. This covers
completions from normal commit or rollback replay and from the watermark-based
cleanup. The oldest outstanding prepare stops the pruning scan.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15121
Test Plan:
Added coverage for missed-marker cleanup, transaction-name reuse before replay,
the exact watermark boundary, a catch-up round that reports success after
replaying no WALs, and tracker pruning after normal commit replay. The
missed-marker case verifies cleanup of both retained structures.
Reviewed By: pdillinger
Differential Revision: D116620460
Pulled By: xingbowang
fbshipit-source-id: baca7faf53f414507957f2afdcc4ca67133c5669
Summary:
Building on D116105558, this change finishes and publishes `format_version = 8` as
production-ready.
The headline feature is a tweaked block encoding that stores the common user-key
prefix shared by all keys in a block **once**, in the block's leading bytes
`[0, restarts[0])`, instead of repeating it at every restart point. This shrinks
prefix-heavy blocks (and their block-cache footprint) and, for bytewise-ordered
keys, speeds up `Seek` by comparing prefix-stripped suffixes. It is applied to
**data blocks** and **every index tier** (leaf/partition index, the
partitioned-index top-level, and the partitioned-filter top-level).
Motivation: MyRocks/ZippyDB/timeseries-style keys carry a long shared prefix
(table/index id, partition, etc.). Repeating it at each restart wastes space and
cache, and every Seek re-compares those bytes. Profiles show index-block seek CPU
is comparable to data-block seek CPU (partitioned index does two index seeks per
data seek), so index blocks are a first-class target, not just data blocks.
### Format (self-describing, no new footer bit)
- The common prefix occupies the block's leading bytes; a non-zero `restarts[0]`
(the offset of the first entry, i.e. the prefix length) self-signals its
presence. `restarts[0]` is 0 in every block that does not use the feature (all
versions, all block types), so a single fv8 reader transparently handles any
mix of stripped and unstripped blocks -- no footer bit is consumed.
- Previous versions would misinterpret this encoding by quietly ignoring the
common prefix, which (along with the lack of extensibility in data-block
metadata) is why it is gated behind format_version 8.
- Only restart-point entries change on disk; non-restart entries are byte
identical to fv7, so the writer produces the exact same bytes regardless of
implementation.
### Public option
New `BlockBasedTableOptions::optimize_key_common_prefix`
(`enum class OptimizeKeyCommonPrefix`), default `kIfFastSeek`:
- `kDisabled` -- never use it.
- `kIfFastSeek` (default) -- use it only where it also enables the Seek speedup:
format_version >= 8 with a built-in (reverse-)bytewise comparator.
- `kEnabled` -- use it for all comparators at fv8; (reverse-)bytewise get space +
fast seek, other comparators get the space savings only (the reader
reconstructs full keys and compares with the real comparator).
Expressed as a criteria so it need not be toggled per CF based on the comparator
(which can vary by CF). It additionally requires `use_delta_encoding` (the
feature builds on delta encoding) and `ts_sz == 0`; when `use_delta_encoding` is
false -- the setting that also preserves zero-copy key pinning -- it is a no-op.
Default format_version is unchanged (still 7); the feature is opt-in via fv8.
### Scope: every index tier
The reader handles all index-like blocks uniformly (a non-zero `restarts[0]`
self-signals a stripped block regardless of block type), so the writer strips the
common prefix from every tier:
- **Data blocks** and **leaf / partition index blocks**: enabled.
- **Partitioned-index top-level block**: enabled. Under the default
`index_shortening = kShortenSeparatorsAndSuccessor` the file's last separator
would otherwise be a short successor sharing no leading bytes, collapsing the
single-block top-level's common prefix to 0. When common-prefix is on the
builder suppresses that one successor (keeps the full last key -- i.e.
`kShortenSeparators` behavior for that one entry), so the top-level strips
uniformly under every shortening mode; the full last key is also a tighter file
upper bound, so it is not worse.
- **Partitioned-filter top-level index**: enabled (same uniform treatment). Its
space/CPU effect is small at default partitioning (the top-level filter index
is tiny) but grows with many small partitions -- a filter miss short-circuits
at, and is dominated by, the top-level filter-index seek (see Performance).
Included for a uniform on-disk surface across tiers.
Bloom filter *contents* still hash full user keys, so prefix/whole-key filtering
is unaffected.
### Implementation
- Writer (`block_builder.{h,cc}`): keys are encoded incrementally in the fv7
layout while the running common prefix is tracked cheaply (reusing the delta
`shared`); at `Finish()` only the restart-point entries are rewritten in place
with the prefix stripped, and the prefix is written once at the block start.
The rewrite is generalized to the value-delta (fv4) index layout used by index
blocks (no `value_size` header; bulk-copy the inline BlockHandle and
non-restart entries verbatim).
- Reader (`block.{h,cc}`, `dbformat.h`): `Block` decodes the prefix from
`restarts[0]`; `BlockIter::ParseNextKey` prepends it on a restart point
(`shared == 0`). The suffix-seek helpers (strip target once, compare shortened
suffixes) live in the templated `BlockIter` base and are shared by data and
index iterators (binary search, templated on `DecodeKey`/`DecodeKeyV4`); for
non-bytewise comparators (data) and the hash `PrefixSeek` refine step, the
reader reconstructs the probed restart key and compares full keys.
- Gating (`block_based_table_builder.cc`, `index_builder.{h,cc}`,
`partitioned_filter_block.{h,cc}`): the criteria is threaded to the data-block
builder and to every index tier -- leaf/partition index, the partitioned-index
top-level (`use_common_prefix_top`), and the partitioned-filter top-level --
together with the last-separator successor suppression
(`ShortenedIndexBuilder::GetSeparatorWithSeq`) that lets the top-level strip
under the default shortening mode. The index base gate also requires
`super_block_alignment_size == 0` (block alignment can trigger
`skip_delta_encoding`, producing non-restart `shared == 0` entries the reader
could not distinguish from stripped restart keys).
- Memory: the per-block/per-iterator prefix is stored as a `uint32_t` length
(the prefix is always at offset 0) rather than a full `Slice`, and `Block` /
`BlockBuilder` members were repacked so the feature adds little or no struct
size.
- Java: a new `OptimizeKeyCommonPrefix` enum and a `BlockBasedTableConfig`
getter/setter, plumbed through JNI in both directions (the
`newTableFactoryHandle` writer and the options-reconstruction constructor).
### Whole-file (table-level) prefix: investigated and rejected
A natural extension is to store the whole-file common prefix once per *file*
(when an `SstPartitioner` guarantees every key in the file shares a fixed prefix)
instead of once per *block* -- e.g. the MyRocks index/table id, or a future
128-bit global id. It was prototyped end-to-end (partitioner-provided prefix
captured at file open, stored in a `user_key_common_prefix` table property, and
stripped from every data/index block; the reader re-prepends it) and then
dropped. The `user_key_common_prefix` table property reserved in the
previous diff is dropped in this one. The rationale, in case it is revisited:
- Marginal benefit over per-block. Per-block already strips the full shared
prefix from every key; the only thing whole-file removes on top is the single
per-block copy of that prefix, i.e. `~= prefix_len / block_size` of the
uncompressed (block-cache) footprint, with no meaningful read-CPU change
(per-block already delivers strip-once + suffix compares). Measured at
prefix_len=12, 4 KB blocks this was ~-0.3% incremental space; at the 8-16 KB
data blocks typical of the space-constrained target (MyRocks/UDB) it is
~0.1-0.2% even with a 16-byte global id -- a fraction of a percent, seen only
by partitioner users. Small block-size workloads are becoming more
niche, and even where they exist, they are also CPU-sensitive (see next
paragraph).
- Real, if small, CPU tax on the *per-block* path. Reconstructing a key from two
non-contiguous parts (file prefix + block residual) cost ~0.5-1.0% additional
retired instructions on the per-block read path (deterministic across gcc and
clang), wall-clock-neutral in cache-resident tests but present even when no
whole-file prefix is configured. Attempts to remove it did not pan out: a
compile-time specialization of the reconstruction *increased* instructions
(+2.9%) by defeating inlining; a per-block combined-prefix buffer trades the
CPU cost for a memory cost on every (default) block; and a per-iterator buffer
was the original larger regression. Only invasive iterator-level
monomorphization would truly zero it, which is not justified on this hot path.
- Decision. A fraction-of-a-percent, partitioner-only space win does not justify
a permanent per-block-read-path CPU tax plus additional on-disk format surface
(a table property that changes whole-file interpretation). Dropped. The generic
`SstPartitioner::ShouldPartitionByPrefix` hook is kept (cheap, and useful
independently), and the reserved `user_key_common_prefix` table property is
removed since no writer will emit it.
### Performance (db_bench, release/native, cache-warm, best-of-3, pinned)
Three arms: (A) fv7 on the base/main branch, (B) fv7 with this change (should be
neutral / byte-identical writing), (C) fv8 with this change and
`optimize_key_common_prefix=enabled` (the optimization is active). Partitioned
index + filters; `compression_type=none`, so SST bytes are the uncompressed
block-cache footprint. `int` = db_bench default keys (variation-early, near worst
case); `pfxN` = an N-byte shared prefix + varying suffix (MyRocks-like).
Setup: `num=3,000,000`, `reads=1,500,000`, single thread,
`metadata_block_size=4096`, cache holds the working set, pinned to one core.
| shape (key/val) | arm | SST bytes | readrandom us/op | seekrandom us/op |
|---|---|---:|---:|---:|
| int (32/8) | A fv7 main | 138,983,658 | 3.484 | 4.995 |
| | B fv7 ours | 138,983,146 | 3.518 | 5.024 |
| | C fv8 ours | 137,681,420 | 3.490 | 4.986 |
| pfx24 (32/8) | A fv7 main | 71,697,531 | 3.417 | 5.561 |
| | B fv7 ours | 71,698,043 | 3.448 | 5.579 |
| | C fv8 ours | 65,578,236 | 3.284 | 5.386 |
| pfx56 (64/16) | A fv7 main | 102,942,266 | 3.624 | 5.830 |
| | B fv7 ours | 102,942,266 | 3.678 | 5.835 |
| | C fv8 ours | 90,554,642 | 3.444 | 5.626 |
fv8 vs fv7-main (C vs A) -- space / readrandom / seekrandom:
- int (32/8): **-0.9%** / ~neutral / ~neutral
- pfx24 (32/8): **-8.5%** / **-3.9%** / **-3.1%**
- pfx56 (64/16): **-12.0%** / **-5.0%** / **-3.5%**
fv7-ours vs fv7-main (B vs A): SST bytes byte-identical (within 512 B; the fv8
paths are gated off), read/seek us/op within run-to-run noise (~1%, best-of-3 on
a shared host). Net: fv8 is a space **and** read/seek-CPU win on prefix-heavy
data and neutral on variation-early (`int`) data, so `kIfFastSeek` is safe on by
default at fv8.
### Space by tier (compression=none, num=3M, mdb=4096; fv7 -> fv8)
Where the win comes from -- the index tiers strip the most:
| shape | data | leaf/partition index | top-level index | filter |
|---|---:|---:|---:|---:|
| int | -0.8% | -29% | -45% | -0.2% |
| pfx24 | -8.3% | -69% | -89% | -0.8% |
| pfx56 | -11.1% | -83% | -96% | -1.6% |
Data bytes dominate the absolute total, but every index tier is stripped:
leaf/partition index -29% to -83% and the partitioned-index top-level -45% to
-96%. Filter *blocks* barely move (-0.2% to -1.6%) because filter bits hash full
user keys; only the small top-level filter index is stripped.
### Point-miss lookups (readmissing)
A partitioned-filter miss short-circuits at the filter (no data-block read), so
it is dominated by the top-level filter-index seek plus the partition bloom. At a
typical `metadata_block_size=4096` (few partitions, tiny top-level filter index)
fv8 is neutral on misses (int +0.9%, pfx24 -1.0%, pfx56 -0.7% us/op -- all
noise). The top-level-filter stripping pays only with many small partitions: at
`metadata_block_size=256` it is pfx24 1.166->1.138 (**-2.4%**) and pfx56
1.273->1.188 (**-6.7%**) us/op.
### Backward / forward compatibility
- Default `format_version` remains 7; the feature is inert unless a CF opts into
fv8. Older RocksDB rejects fv8 files at footer decode (fail-safe).
- Within fv8, the stripped-block set a writer can emit is fixed and every fv8
reader can read it (the `restarts[0]` signal plus uniform reader handling), so
the on-disk surface is self-consistent across builds/downgrade.
### Follow-ups (not in this change)
- Suffix-aware interpolation / hash index seek (currently correct via the
reconstruction fallback, not perf-tuned).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15117
Test Plan:
- New focused unit tests (`table_test`), run at the full `format_version` matrix
including the now-supported fv8:
`GeneralTableTest.OptimizeKeyCommonPrefix` (data-block space+read+Seek
correctness: existing and missing keys straddling the prefix, before/after-all,
SeekForPrev), `OptimizeKeyCommonPrefixNonBytewise` (reconstruction path), and
`IndexBlockCommonPrefixEquivalence` across `{kBinarySearch,
kBinarySearchWithFirstKey, kHashSearch, kTwoLevelIndexSearch}` x
`{partitioned filters}` x `{bytewise, reverse-bytewise}` (so both top-level
tiers -- partitioned-index and partitioned-filter -- are exercised), asserting
identical reads and a shrunken fv8 index.
- The randomized `ParameterizedHarnessTest` now runs at fv8 (reverse-bytewise,
restart intervals `{1,16,1024}`, special/empty keys fuzzed against the feature).
- `db_crashtest.py` randomizes `format_version` including 8 and
`optimize_key_common_prefix` in {0,1,2}.
- Removed the reserved-property carve-out: dropped the `ReservedUserKeyCommonPrefix`
reader test and the C API getter for the property; C API regenerated via
`tools/c_api_gen/` (`make regen-c-api`), and the intentional public-C-API
removal is recorded in `tools/c_api_gen/api_compatibility_allowlist.json`.
- Java: `BlockBasedTableConfigTest.optimizeKeyCommonPrefix` (default
`kIfFastSeek`, plus set/get of every value).
- `table_test` `FooterTests` updated for the now-published fv8 footer, whose
metaindex-gap encoding stores the metaindex block size in 16 bits (the test
caps that size for fv >= 8).
- Manual verification: the db_bench runs in Performance above (three-arm
space + read/seek CPU, per-tier byte breakdown via `sst_dump`, and
`readmissing`); and `readrandom`/`seekrandom` with
`--block_protection_bytes_per_key=8` find 100% of keys at both fv7 and fv8
(reconstructed keys pass per-key checksum).
Reviewed By: xingbowang
Differential Revision: D116365016
Pulled By: pdillinger
fbshipit-source-id: 66f4b5c60ea4fe0ff48d34c98d191ba8ebdf0665
Summary:
Faster and safer MANIFEST recovery for remote compaction (OpenAndCompact)
`DB::OpenAndCompact` (remote/distributed compaction) opens the target DB as a
secondary and reconstructs the LSM state from the MANIFEST before running the
requested compaction. Previously, point-in-time recovery would stat/open the
SST (and blob) files referenced by the MANIFEST to classify them found/missing
and materialize candidate versions, and eagerly load their table handlers.
That is (a) a lot of unnecessary file operations and open latency, especially
on remote/distributed storage, and (b) fragile: when a file referenced by the
MANIFEST but not an input to the compaction was momentarily unavailable to the
worker -- e.g. an obsolete file the primary had already deleted -- recovery
could fail the job (surfacing as an incomplete "no valid result") or roll back
to an earlier version, even though the compaction only needs its own input
files, which the primary keeps referenced for the job's duration. This was seen
as a flaky failure of CompactionServiceTest.BasicCompactions on aarch64, and
reproduces on x86 under COERCE_CONTEXT_SWITCH.
Efficiency + reliability: recovery for the OpenAndCompact path now trusts the
MANIFEST. It reconstructs the current version from FileMetaData without
stat-ing/opening files to classify them and without eagerly opening
candidate-version table handlers, so it does far fewer file operations and
never fails or rolls back over a file the compaction does not need. The
compaction's input files are opened (and unique-id verified) on demand as it
reads them.
Correctness hardening: trusting the MANIFEST means the worker's own (possibly
eventually-consistent) view of CURRENT/MANIFEST drives the reconstructed LSM
shape, which feeds correctness-critical decisions (Compaction::IsBottommostLevel
/ KeyNotExistsBeyondOutputLevel). A worker that read a stale CURRENT (older
MANIFEST), a truncated MANIFEST, or a MANIFEST with a corrupt/torn tail could
reconstruct against an older LSM shape and silently drop keys that should be
kept. To prevent this, the primary records the MANIFEST position (file number
and size) it scheduled the compaction from and passes it in
CompactionServiceInput as a floor. After recovery the worker compares the size
of the maximal *valid* prefix it reconstructed -- which stops at any corrupt or
torn record and at a partial atomic group at EOF, so tolerated tail garbage
cannot inflate it -- against that floor, and refuses to compact against an older
view (an older file number, or the same file number whose valid prefix is
shorter), returning Status::Incomplete so the CompactionService can fall back to
a local compaction. Accept-equal-or-later is safe by monotonicity: while the
inputs stay locked (being_compacted), a later view can only have moved data
further down, never resurrecting a shadowed value, so only an older view is
unsafe.
Single kill switch: both behaviors are bundled behind one signal and one
option. Presence of the floor on the request (min_manifest_file_number != 0)
couples on both trust-manifest recovery and the floor check in the worker;
absence reverts the worker to the original point-in-time recovery with no floor
check. A single mutable DBOption, remote_compaction_manifest_floor (default
on), gates whether the primary populates the floor, so it is the one escape
hatch that disables the entire new recovery path on a running DB via
SetDBOptions if it ever misbehaves. The new CompactionServiceInput fields are
wire-compatible in both directions (0 = not provided), so an old worker or an
old primary simply behaves as if the switch were off.
Follow-up: recovery is not yet strictly "open only the compaction's inputs". A
limited number of non-input files are still opened by the bounded file-property
sampling in Version::PrepareAppend -> UpdateAccumulatedStats (tolerating any
that are missing), which today populates input FileMetaData::num_entries for
compaction input-record-count verification. A natural follow-up is to record
the needed per-file info (e.g. num_entries) in the MANIFEST so the worker can
skip these auxiliary opens entirely; alternatively, source input num_entries
from the worker's own read of its input table properties. Relatedly, the floor
comparand is the recovered "maximal valid prefix" size, which today equals the
MANIFEST file size but is defined to stay correct if a future MANIFEST format
writes trailing padding/footer past the last record.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15133
Test Plan:
Added tests in compaction_service_test.cc:
- RemoteCompactionMissingLowerFileTrustsManifest: a non-input file referenced
by the MANIFEST is unavailable during a remote compaction; the job completes
with correct results (fails without trust-manifest recovery). The underlying
race also reproduces under COERCE_CONTEXT_SWITCH.
- RemoteCompactionManifestFloorRejectsStaleView: inflate the floor so the
worker's recovered MANIFEST is below it; assert the worker rejects, the job
falls back to local, and a below-output deleted key stays deleted (no silent
resurrection).
- RemoteCompactionManifestFloorSendsPosition: a non-zero floor is attached and
the worker accepts its equal-or-later view.
- RemoteCompactionManifestFloorKillSwitchOff: with the option off, no floor is
sent, the worker uses the original recovery, and a transiently-missing
non-input file makes the job fail/roll back (the escape hatch reverts
behavior).
- RemoteCompactionManifestFloorMutable: SetDBOptions flips the option; the
floor stops being attached and the worker skips the check.
options_settable_test covers the new mutable DBOption.
Crash test: remote compaction is already exercised (db_crashtest randomly
enables remote_compaction_worker_threads, driving DB::OpenAndCompact), so with
remote_compaction_manifest_floor defaulting on, the new recovery path is
covered by the crash test by default -- no new crash-test plumbing is needed.
The kill switch is intentionally left un-randomized (always on): turning it off
re-enables the original, occasionally-broken recovery, which would only add
noisy failures for no benefit.
UPDATE: crash test is not well covered because of `"remote_compaction_failure_fall_back_to_local": 1` which is masking other existing failures. Addressing this is a high-priority follow-up item. (Not practical to fix in this diff.)
Reviewed By: jaykorean
Differential Revision: D116738823
Pulled By: pdillinger
fbshipit-source-id: c02e0a437d62fc526dbbfcf70edf43434dfc7c1f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15129
When `persist_user_defined_timestamps=false` and `long_running_snapshots=1`, `AssertSame` in `MaybeReleaseSnapshots` reuses the timestamp saved at snapshot acquisition time. That timestamp can become stale because concurrent flushes advance `full_history_ts_low`, and long-running snapshots can be held much longer than the 10-second headroom added by `GetReadTimestamp()`.
The snapshot validation is intended to check that a held snapshot still exposes the same sequence-number view. It should not require the acquisition-time read timestamp to remain valid after history below `full_history_ts_low` has been collapsed.
The fix keeps the timestamped read API contract intact. Column families with user-defined timestamps still require `ReadOptions.timestamp`, even when timestamps are memtable-only. For `persist_user_defined_timestamps=false`, `AssertSame` now uses a fresh validation timestamp so the read remains above `full_history_ts_low` while the snapshot sequence number continues to pin data visibility. For persisted timestamps, it continues reusing the original timestamp to validate the exact timestamped snapshot view.
Reviewed By: joshkang97
Differential Revision: D116110124
fbshipit-source-id: 9cf6cae6e7123f8453b7065565e9e2e0eb34321a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15130
## Summary
Fix a corruption error in remote compaction where `FilterInputsForCompactionIterator` produces different results on the primary host vs. the remote worker.
**Root Cause:**
When a standalone range tombstone file completely shadows another input file, `FilterInputsForCompactionIterator()` removes the shadowed file from `input_levels_`. This filtering relies on `FileIsStandAloneRangeTombstone()`, which checks `num_entries == num_range_deletions == 1`.
On the primary host, these `FileMetaData` fields are always populated (set during flush/compaction when the file was created). On the remote worker, however, these fields are NOT persisted in the MANIFEST and are only lazily loaded from table properties via `UpdateAccumulatedStats()` during DB open, which is capped at 20 files (`kMaxInitCount`).
If the standalone range tombstone file is not among those first 20 initialized files, its `num_entries` remains 0 on the remote worker. This causes `FileIsStandAloneRangeTombstone()` to return false, so the remote worker does NOT filter the shadowed file and processes all its keys. Meanwhile the primary's expected record count is computed from its (filtered) view, leading to:
```
Corruption: Compaction number of input keys does not match number of keys
processed. Expected 0 but processed 12.
```
**Fix:**
Before constructing the `Compaction` object on the remote worker, explicitly initialize `FileMetaData` stats from table properties for all input files. This ensures `FilterInputsForCompactionIterator` produces consistent filtering results on both sides.
**Introduced by:** D83375779 (2025-09-29), which enabled `FilterInputsForCompactionIterator` on the remote worker by passing `earliest_snapshot` to `PickCompactionForCompactFiles`.
Reviewed By: xingbowang
Differential Revision: D116252973
fbshipit-source-id: 2cb2d61dae1f3982edeaf41f9a0f39d5aef1e789
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15132
This fixes a liveness bug in background-error recovery. Retryable soft errors allow foreground writes to continue, but recovery must wait for every scheduled background flush worker to exit before it can submit the recovery flush.
While recovery was active, normal flush requests could still enter `flush_queue_`. Atomic flush requests bypass per-column-family deduplication, so writes could accumulate a backlog of normal requests.
For each request, the background worker:
1. Pops and drops the normal request because BG work is stopped
2. Sleeps for the one-second background-error backoff
3. Decrements `bg_flush_scheduled_`: 1 -> 0
4. Calls `MaybeScheduleFlushOrCompaction()`
5. Schedules the next queued request: 0 -> 1
6. Only then signals the recovery thread
The rescheduling happens under the DB mutex before recovery is signaled:
Queued requests: R1, R2, R3, ...
Flush worker Recovery thread
------------ ---------------
drop R1
sleep 1 second waiting for scheduled == 0
scheduled: 1 -> 0
schedule R2: 0 -> 1
signal -------------------------> wake, observe scheduled == 1
wait again
drop R2
sleep 1 second
scheduled: 1 -> 0
schedule R3: 0 -> 1
signal -------------------------> wake, observe scheduled == 1
wait again
The dropped request is not requeued. Instead, the worker exit path guarantees that the next queued request is scheduled while `unscheduled_flushes_ > 0`.
A finite backlog therefore delays recovery by roughly one second per request. If writes keep producing requests as quickly as workers discard them, the queue may never drain and recovery is starved indefinitely. Recovery can complete after writes become quiescent, so this is a liveness failure rather than a lock deadlock or data-corruption bug.
There is also a queue-accounting edge. Scheduling consumes `unscheduled_flushes_`, but a worker does not own its `FlushRequest` until it pops `flush_queue_`. A worker can exit on the background error before popping, leaving:
flush_queue_.size() > 0
unscheduled_flushes_ == 0
Clearing only unscheduled requests therefore misses scheduled-but-unclaimed requests.
The fix:
- Reject new non-recovery requests while background work is stopped.
- Allow error-recovery flush reasons to bypass that guard.
- Clear the entire flush queue before recovery waits for scheduled workers.
- Retain the worker-side reason check for requests queued before the error.
- Rebuild recovery flushes and schedule a catch-up flush for writes made during recovery.
Reviewed By: pdillinger
Differential Revision: D116708546
fbshipit-source-id: 6e29fd3653ef72d047c64ca89d311a7bee51ab84
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15131
ASAN and other sanitizer runtimes can disable coredumps or terminate without a core after the test driver sets its shell core limit. This leaves sanitizer stress-test failures with an LLDB stack but no core for the DB artifact.
Configure ASAN_OPTIONS, TSAN_OPTIONS, and UBSAN_OPTIONS during sanitizer crash-test steps so reports terminate with a core-generating abort and core dumping remains enabled. Preserve caller-provided options and the existing Make fallback semantics. Keep the behavior scoped to test steps and crash targets.
Enabling sanitizer cores raises a concern that the generated core-dump artifacts can be large because they include the process working set. Reduce RocksDB's cache and memtable contribution to the core size without replacing the normal randomized test configuration. Add a db_crashtest cache/write-buffer size multiplier that is applied after each iteration chooses its values, and pass 0.5 from ASAN, TSAN, and UBSAN crash lanes. For the simple profile, cache_size remains randomized with 8/32 MiB becoming 4/16 MiB, while the 32 MiB write buffer becomes 16 MiB. Preserve RocksDB's 64 KiB minimum write-buffer size and leave non-sanitizer lanes unchanged.
Reviewed By: rban1
Differential Revision: D116666171
fbshipit-source-id: aa157cfa255bb637d4a68e54ea2039db779fdc53
Summary:
When a column family configures an SstPartitioner, compaction paid a per-key cost for every key-value written to an output SST: a full user-key copy (into `last_key_for_partitioner_`) plus a virtual `SstPartitioner::ShouldPartition()` call (which also builds a PartitionerRequest and truncates/compares two Slices).
This is motivated by an internal user whose custom SstPartitioner parses a variable-length prefix (a prefix code) out of each key to decide partition boundaries; today that prefix parsing runs on every key written during compaction, on top of the copy and virtual call above.
This wires up the `SstPartitioner::ShouldPartitionByPrefix()` API (added, but unused, alongside the format_version=8 groundwork) so compaction can skip that per-key work. When a partitioner can express its next boundary as a key prefix, `CompactionOutputs` caches that prefix once per output file (re-derived from the file's first key) and replaces the per-key copy + virtual call with an inlined `Slice::starts_with()` check. That also lets a partitioner like the one above do its prefix parsing once per output file instead of once per key. When the partitioner returns no prefix (ambiguous, e.g. `SstPartitionerFixedPrefix` for keys shorter than its length), compaction falls back to the previous per-key `ShouldPartition()` path. Output file boundaries are byte-for-byte identical to before. `SstPartitionerFixedPrefix` implements `ShouldPartitionByPrefix()`; the base-class default returns "no prefix", so existing custom partitioners are unaffected until they opt in.
Also:
- db_bench: new `--sst_partitioner_fixed_prefix_len` to configure a fixed-prefix partitioner (enables benchmarking; no default behavior change).
- `ShouldPartitionByPrefix()` API doc updated now that it is used and recommended for compaction CPU efficiency (any prefix code works, not just fixed-size).
No new option and no behavior or format change.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15118
Test Plan:
Added db_compaction_test cases covering the new fast path and its fallbacks: many keys sharing a prefix (one output file per prefix, no spurious cuts), keys shorter than the fixed prefix length (fallback to ShouldPartition), the zero-length-prefix "never partition" case, and short<->long key transitions across output-file boundaries. Existing SstPartitioner tests and the range tombstone + partitioner tests (db_range_del_test) provide regression coverage that file boundaries are unchanged, including the fast path with range tombstones. The crash test already exercises the partitioner via `enable_sst_partitioner_factory`.
Benchmark (db_bench `compact` of a pre-built 50M-key DB; 16B keys / 8B values, 8-byte shared key prefixes with FixedPrefix length 8, no compression, single subcompaction). A/B compares this change against the same binary with only the compaction hot-path change reverted. Means of 6 trials, optimized faster on every trial and every metric; both variants produced identical output (50 files, i.e. identical partition boundaries):
| Metric | Baseline | Optimized | Delta |
|---------------------------------|----------|-----------|--------|
| Instructions (whole process) | 92.46 B | 90.09 B | -2.6% |
| CPU cycles | 46.20 B | 45.28 B | -2.0% |
| CPU time (task-clock) | 18.74 s | 18.39 s | -1.9% |
| `compact` wall time | 20.58 s | 20.08 s | -2.4% |
DB-internal compaction stats corroborate: CompMergeCPU 15.84s -> 15.42s (-2.6%), compaction write throughput 38.9 -> 39.9 MB/s. The relative gain shrinks with larger values or compression enabled, where the partitioner is a smaller fraction of total compaction work; it is a pure win whenever a prefix partitioner is configured.
Reviewed By: rban1
Differential Revision: D116372363
Pulled By: pdillinger
fbshipit-source-id: 63de190329112aa2ce5b48da9fc77ef386494558
Summary:
PR https://github.com/facebook/rocksdb/issues/15066 fixed the bug that has caused this combination fails in the past. Re-enable this combination in the stress test.
db_crashtest.py currently disables continuous_verification_interval whenever test_secondary=1. This change removes that quarantine and enables continuous verification for secondary DB stress tests.
The disable predates auto_refresh_iterator_with_snapshot. When auto-refresh was later added, its stress-test integration mechanically added snapshot setup to several iterator paths, including the secondary-
only NonBatchedOpsStressTest::ContinuouslyVerifyDb() path:
snapshot = std::make_unique<ManagedSnapshot>(db_);
read_opts.snapshot = snapshot->snapshot();
This creates a snapshot from the primary db_ and passes it to an iterator created from secondary_db_. That was never supported: a ReadOptions snapshot must belong to the DB being read, and secondary DB
iterators explicitly reject non-null snapshots.
The incompatibility went unnoticed because the existing sanitizer made this combination unreachable in recurring crash tests. It was therefore a stress-test integration bug, not evidence that snapshot-
preserving auto-refresh was supported by secondary DBs.
This change:
- Stops forcing continuous_verification_interval=0 for secondary tests.
- Avoids attaching a primary-owned snapshot to secondary iterators.
- Continues exercising auto_refresh_iterator_with_snapshot through primary iterator paths.
Secondary iterators still support ordinary Iterator::Refresh(), which advances them to the latest caught-up secondary state. Snapshot-preserving automatic refresh remains unsupported.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15114
Test Plan:
- python3 tools/db_crashtest_test.py
- make check-sources
- AUTO_CLEAN=1 make -j192 db_stress
- Targeted blackbox crash test with:
- test_secondary=1
- continuous_verification_interval=1000
- auto_refresh_iterator_with_snapshot=1
Reviewed By: jaykorean
Differential Revision: D116381630
Pulled By: xingbowang
fbshipit-source-id: a162ab357a537c2e5047c1eda85d0c2138b56ed1
Summary:
`ExpectedState::Restore()` always attempted to open and replay the post-`saved_seqno_` trace. In the crash-test failure, recovery had already returned the DB exactly to `saved_seqno_`, so `replay_write_ops == 0` and the trace was legitimately empty. `FileTraceReader` reported `Status::Incomplete()` on that empty trace, which incorrectly failed restore.
Skip the trace replay work when `replay_write_ops == 0`, but still run the normal state promotion, cleanup, and `saved_seqno_` reset path so later saves continue to work.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14840
Reviewed By: xingbowang
Differential Revision: D107968797
fbshipit-source-id: b4024e1ff329aa3bd47f0307e67d9a4f224b66d4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15106
When `optimize_manifest_for_recovery` is enabled, `MaybeUpdateNextFileNumber` uses on-disk files to decide whether recovery must append a `SetNextFile` edit. `OPTIONS-*` and temp options files are written as sidecar option snapshots during open, after recovery, so their presence alone should not force MANIFEST growth on a later clean reopen.
Exclude `kOptionsFile` and `kTempFile` from the trigger that requires a MANIFEST file-number advance. If any real file requires an advance, the code still advances past the largest observed file number, including OPTIONS files.
Also tighten db_stress strict MANIFEST verification so fatal zero-growth checks only run when all fault injection is disabled. Open write/sync fault injection can legitimately make `reuse_manifest_on_open` fall back to a fresh MANIFEST.
Reviewed By: xingbowang
Differential Revision: D116092705
fbshipit-source-id: 65b57b0b7f57b2cef42efc4df353fb2218688d34
Summary:
Add an opaque external_table_config option to the external table adapter. RocksDB persists the string in OPTIONS files and passes it to ExternalTableFactory::Configure during bootstrap, allowing implementations such as Nimble to parse implementation-specific settings without adding RocksDB-specific fields. The option is immutable after DB open, and attempted runtime updates fail safely for non-clonable table factories. A reopen test verifies that callers can supply a different valid configuration between DB opens. Initialize the adapter option metadata lazily as const to avoid namespace-scope dynamic initialization and mutable-global lint issues.
PR https://github.com/facebook/rocksdb/pull/15119
Reviewed By: anand1976
Differential Revision: D116362444
fbshipit-source-id: 225a4aedc455f320b5c90ac2f0862917da477ccb
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15102
Add a dedicated Skycastle signal that invokes liveness through the existing remote-storage test-driver path.
For remote DB runs, double default wrapper/no-progress timeouts while preserving explicit overrides. Default remote liveness `expected_values_dir` to local storage so `UniqueIdVerifier` never treats a remote DB path as a local path; local liveness behavior remains unchanged.
Normalize both equals and spaced `--env_uri`/`--fs_uri` forms once so remote detection, workload execution, and cleanup share the same arguments. Track the crash harness and internal release scripts as CI source triggers, scoped so release-script changes select only the remote-storage signal. Add command-level regression coverage and failure-safe expected-values cleanup. Use generic remote DB/storage terminology throughout public RocksDB code and comments.
Reviewed By: pdillinger
Differential Revision: D115936164
fbshipit-source-id: e4997bf19eec18e4914626d479690b78c6cd3ad2
Summary:
Follow-ups to the lazy wide-column read API (DB::GetEntityLazy / MultiGetEntityLazy). Three related, low-risk changes under one theme; no public API change.
1. Internal read-API cleanup (retires TODO(lazy-blob-resolution-cleanup)). The lazy byte-range work had left a whole/range/force-verify method fan-out on the blob read path: VersionBlobFetcherBase::FetchBlobRange plus a separate FetchBlobForceVerify, and SameFileBlobReader::GetSameFileBlob paired with GetSameFileBlobRange, with the (verify_checksums, force_verify) decision duplicated across branches. This folds them into a single 3-valued BlobVerifyPolicy (kVerifyIfPresent / kVerifyIfNoAmplification / kSkip, added to db/blob/blob_constants.h) derived once at the resolver boundary and threaded down. FetchBlobRange and GetSameFileBlob now take a range descriptor (range_length == kWholeBlobLength means the whole value) plus the policy; FetchBlobForceVerify and the *Range twins are removed. The public LazyColumnReadRequest::force_verify bool is unchanged. The widely-called lower-layer whole-value primitives (BlobFileReader / BlobSource / Version GetBlob) are intentionally left as delegated primitives to bound blast radius; folding those is a later cleanup.
2. Embedded block-cache-tier quirk fix (retires TODO(lazy-blob-resolution-quirks)). ValidateEmbeddedBlobIndex rejected read_tier == kBlockCacheTier before the blob-cache probe, so a same-file (embedded) column already in the blob cache returned Incomplete under kBlockCacheTier -- unlike the separate-file path, which probes the cache first. The up-front rejection moves to the two no-cache paths (ResolveEmbeddedBlob / ResolveEmbeddedBlobPinned); the cache-probing paths now serve a cached record under kBlockCacheTier, matching separate-file behavior. This is a shared lower-layer change: the non-lazy Get / GetEntity and iterator embedded read paths (get_context / MaybeResolveEmbeddedValue -> ResolveEmbeddedBlobCached) get the same corrected behavior, not just the lazy resolver.
3. Coverage. db_bench gains a readrandomentitylazy benchmark (GetEntityLazy + MultiResolve of --lazy_entity_read_length bytes per column) for measuring byte-range (partial) blob read savings; it requires open_files == -1. The crash test now also fuzzes LazyColumnReadRequest::force_verify in the lazy-vs-eager differential (force_verify affects only verification, not the returned bytes, so the comparison is unchanged).
This also sets up the next planned step -- cross-key coalescing for MultiGetEntityLazy / LazyWideColumnsBatch::MultiResolve. The internal BlobReadRequest already carries a {offset, len} range descriptor, and with the verify decision now a single BlobVerifyPolicy rather than a bool plus a separate force-verify method, a batched resolve can group a vector of range+policy requests per (CF, Version, blob file) and coalesce them onto the existing MultiGetBlob machinery -- without re-introducing whole/range/force-verify method twins in the multi-read path.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15115
Test Plan:
- New unit test DBLazyEntityTest.EmbeddedBlockCacheTierServedFromCachedValue: a cached embedded column serves whole- and sub-range reads under kBlockCacheTier instead of returning Incomplete. Existing force_verify / BlockCacheTier tests are unchanged (public behavior preserved).
- New unit test DBBlobIndexTest.EmbeddedBlobWideColumnGetEntityBlockCacheTierServesCached: covers the non-lazy GetEntity path -- a cached embedded column is served under kBlockCacheTier; the existing ...GetEntityBlockCacheTier test still asserts a genuine miss yields Incomplete.
- Crash test: extended the existing lazy differential to fuzz force_verify (reuses --lazy_entity_read_one_in).
- db_bench: readrandomentitylazy over 2000 entities x 2 uncompressed embedded blob columns (4000 B each), reading 100 B per column:
| metric | value |
|---|---|
| rocksdb.blobdb.lazy.partial.read.count | 4000 |
| rocksdb.blobdb.lazy.read.bytes | 400000 |
| rocksdb.blobdb.lazy.partial.bytes.saved | 15600000 |
i.e. 100 B read per column instead of the 4000 B whole-column value.
Reviewed By: xingbowang
Differential Revision: D116334969
Pulled By: pdillinger
fbshipit-source-id: cebe90456b23be0f5be60cdc4bff85c8a37b5442
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15103
The crash artifact showed `MANIFEST-000005` was physically well-formed but contained an incomplete atomic group that had been tolerated at EOF. `reuse_manifest_on_open` then appended ordinary records after that tail, turning the tolerated crash tail into mid-file `corrupted atomic group` corruption seen by OpenAndCompact and backup restore.
Record whether MANIFEST recovery ended with a pending atomic group. When that marker is set, skip `ReopenManifestForAppend()` so the next manifest write creates a fresh MANIFEST instead of appending after the partial group. The existing physical-size guard remains as a separate torn-tail defense.
This replaces the earlier keep-size preallocation hypothesis with the artifact-backed root cause.
Reviewed By: xingbowang
Differential Revision: D116062618
fbshipit-source-id: fb55ce2891a6fd50fa2c670dc720e91bad27e94d
Summary:
The point_lock_manager_stress_test (via PointLockValidationTestRunner) was seen flaky under USE_CLANG=1 COMPILE_WITH_ASAN=1 COMPILE_WITH_UBSAN=1, manifesting as a heap-use-after-free plus "terminate called without an active exception" / SIGABRT. There were two distinct problems.
Cleanup on failure: run() uses gtest ASSERT_* macros, which return early from run() on failure. That early return skipped the shutdown flag set and the join of the worker threads, so the worker threads kept running and accessed members of the (now destroyed) test runner object (heap-use-after-free), and destroying the still-joinable threads_ vector called std::terminate. Fixed by adding a destructor plus a StopAndJoinThreads() helper that always signals shutdown and joins the worker threads before any member is destroyed, no matter how run() exits. Copy/move are deleted since the object owns worker threads.
Underlying flaky failure: the per-second "No locks were acquired in the last 1 second" liveness assertion was too strict. Under sanitizer builds combined with very short lock timeout/expiration and post-acquisition sleeps, all threads can occasionally fail to acquire a single lock within a given one-second window without the lock manager actually being stuck. The per-thread variant of this check was already disabled for this reason; the aggregate one is now relaxed to tolerate a few consecutive no-progress seconds, which still reliably detects a genuinely stuck (dead/livelocked) lock manager. Also fixed the end-of-run check from `>= 0` (always true) to `> 0` so it actually catches a fully stuck manager.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15116
Test Plan:
Verified the cleanup path by temporarily forcing the liveness assertion to fail while worker threads were active: the test now reports a normal gtest failure with clean teardown, with no heap-use-after-free and no std::terminate/SIGABRT.
I don't have a consistent reproducer for the failure but ran
* 3x `USE_CLANG=1 COMPILE_WITH_ASAN=1 COMPILE_WITH_UBSAN=1 make check`
* `USE_CLANG=1 COMPILE_WITH_ASAN=1 COMPILE_WITH_UBSAN=1 rocksptest point_lock_manager_stress_test -r100`
Reviewed By: xingbowang
Differential Revision: D116349420
Pulled By: pdillinger
fbshipit-source-id: 104784c5c59770414224ac4751eabf69f65e0958
Summary:
Fix https://github.com/facebook/rocksdb/issues/15051: a secondary serves stale reads because entries replayed from a WAL the primary has since flushed are never evicted from the active memtable. During TryCatchUpWithPrimary(), seal the active memtable once the installed Version provably covers everything it holds, so the following RemoveOldMemTables() call drops it. Every memtable-dropping decision takes that same watermark, so a file set the secondary cannot read never turns the stale read into a vanished key.
## Why and How
A secondary replays the primary's WALs into its active memtable and stops replaying a WAL once the primary flushes it. Entries already replayed stayed behind, and since reads consult the memtable first, they shadowed the newer flushed values indefinitely. Iterators go stale too once bottommost compaction rewrites the flushed entry's sequence number to 0; until then Get() and an iterator disagree.
Seal the active memtable during catch-up, gated on the installed Version: each point-in-time Version now carries the column family's log number it reflects, exposed via the new
ReactiveVersionSet::GetInstalledVersionLogNumber(). Sealing requires cf_id_to_current_log_[cfd->GetID()] < that number, which guarantees the primary flushed everything the memtable holds (unflushed WAL data is kept) and that the flushed data is readable through the installed Version. The gate asks the Version directly because proxies fail: the log number alone advances even when no Version was installed, and "no missing files" misses atomic_flush groups parked by a sibling's missing file and corrupt-but-present files with
verify_sst_unique_id_in_manifest=false. In those cases the memtable holds the only readable copy, so dropping it would turn a stale read into a vanished key; the gate fails closed and keeps it.
Reconcile every initialized column family, not just cfds_changed: a round that failed after advancing the log number leaves a stale memtable that later rounds see no new MANIFEST record for. It can also leave a collectible immutable memtable, so the loop consults the new O(1) MemTableList::HasOldMemTablesToRemove() before skipping a column family. The predicate keeps the skip precise: reconciling any column family with a non-empty immutable list would be correct but wasteful, installing a fresh SuperVersion every catch-up call for memtables legitimately waiting on a primary flush and costing readers a mutex acquisition each time. With it, no extra super versions are installed in the steady state. That predicate and the RemoveOldMemTables() cutoff take the installed Version's log number as well, not cfd->GetLogNumber(): that advances as soon as the flush record is read, so an immutable memtable sealed on a WAL switch could otherwise be freed while the flushed file superseding it is unreadable. A sealed memtable is stamped with cf_id_to_current_log_[cfd->GetID()] + 1, the log number following its contents.
cf_id_to_current_log_ is keyed by column family id rather than ColumnFamilyData*, because ids are never reused while a dropped column family's pointer can be recycled, and it is recorded before WriteBatchInternal::InsertInto() rather than after: a batch stops at its first failure with earlier entries already inserted, so recording afterwards could leave the map naming an older WAL than the memtable holds. While the watermark lags and uncollectible memtables accumulate, the secondary warns, since it has no write path to enforce max_write_buffer_number.
The seal sequence, shared with RecoverLogFiles() and the follower's TryCatchUpWithLeader(), is extracted into
DBImplSecondary::SealActiveMemtable(). The follower is refactored onto it with no behaviour change: it keeps cfd->GetLogNumber() as its cutoff because that catch-up path never replays WALs, so its memtables hold nothing that only a not-yet-installed Version could make readable.
## Testing
New tests in db_secondary_test cover: the stale read via both Get() and an iterator, before and after bottommost compaction; unflushed WAL-only data surviving sealing, including two column families sharing a WAL; no data loss with a missing, atomic_flush-sibling-missing, or corrupt flushed file; an immutable memtable sealed on a WAL switch surviving an unreadable flushed file; a batch rejected partway through replay leaving the recorded WAL no older than the memtable's contents; and reconciliation after a failed round (injected via FaultInjectionTestFS at a sync point). Each mechanism was individually neutralized and confirmed to fail its test.
Verified with build_tools/rocksptest.sh over db_secondary_test, db_follower_test, memtable_list_test, version_set_test and version_edit_test (all pass, also under ASSERT_STATUS_CHECKED=1), plus db_basic_test, corruption_test, repair_test and db_flush_test; the secondary tests also ran 20x under COERCE_CONTEXT_SWITCH=1 with no flakes. make format-auto and make check-sources are clean.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15066
Reviewed By: jaykorean, anand1976
Differential Revision: D115317523
Pulled By: xingbowang
fbshipit-source-id: a1d4615eb951d75e69ed242731336485c0dfa137
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15098
Compute the base block checksum in `CompressAndVerifyBlock()` before `verify_compression` decompresses the compressed output. Parallel compression stores that base checksum on the block rep so worker threads can keep the CPU-heavy checksum and verification work off the ordered write path; the writer only applies the offset context modifier before writing the trailer.
Update the regression test to cover both serial and parallel compression paths.
Reviewed By: pdillinger
Differential Revision: D115950785
fbshipit-source-id: b5e47f7d7c41680e7e7907443fc805bc0d043d6e
Summary:
format_version background: SST files are shared across RocksDB versions. format_version is a critical tool in ensuring prior (and later) versioned readers do not misinterpret a file created by some version. Opt-in features can sometimes rely on previously-rejected corners of the schema space, but not always.
This change: lands much of the format_version=8 (fv8) format surface as an internal-only draft -- readable and testable but not user-writable -- so upcoming features can build on a stable base. fv8 is NOT published: kLatestBbtFormatVersion stays 7, and reading or writing fv8 is gated behind the existing TEST-only TEST_AllowUnsupportedFormatVersion() hook. Publishing fv8 (raising the ceiling) is deferred to a later change.
The pieces of fv8 added here, and what each unlocks:
- Index value-delta escape. Existing behavior: index blocks (format_version >= 4) store each block handle as a delta against the previous entry's handle, and signal "this value is a full handle instead" by reusing the key sentinel `shared == 0`. Because the two are coupled, an index entry whose handle is not contiguous with the previous one must be forced to a restart, costing a full key as well as a full value. fv8 decouples them with an in-value escape codepoint: varint 0 means "a full BlockHandle follows," and genuine size deltas are remapped to zigzag(delta)+1 so that 0 stays free. A non-contiguous entry can then remain an ordinary key-delta-encoded, on-cadence entry. The escape applies to every index-shaped block, which includes the top-level index of index partitions and the index of filter partitions, not just the index over data blocks. Unlocks: the upcoming optimize_key_common_prefix index feature (removes its need to special-case super_block_alignment) and value-delta encoding for embedded-blob SSTs (previously disabled outright because interleaved blob records break handle contiguity). Guarded strictly on fv8; format_version <= 7 encoding and decoding are byte-for-byte unchanged. Note that restart points must not delta-encode, so they still have full values without a sentinel.
- Data block footer bit 30 is reserved as an "extended metadata present" escape. Decode rejects a set bit as unsupported/corrupt at all versions. Unlocks: adding block-level metadata in the future through the one shared footer decoder, without another format_version bump or plumbing format_version into every block parser. (Older reader versions relied on this bit being zero for some integrity/consistency checking; format_version bump now makes it safe to use.)
- Footer metaindex gap. For fv8, the top 16 bits of the footer's 4-byte metaindex-size field record the gap (in bytes) between the end of the metaindex block and the start of the footer; the low 16 bits hold the metaindex block size. Unlocks: a space very near the end of the file to save full file checksum state, for a "files carry their own checksums" feature that doesn't require SST parsing to extract the checksum states (to run them to completion on the trivial file tail).
- Reserved "user_key_common_prefix" table property. The reader returns NotSupported when it is present and non-empty. Unlocks: a future whole-file optimization that stores the common prefix of all user keys once instead of in every restart key; until implemented, fv8 readers refuse such a file rather than mis-read it.
- Because of the problem of inferring such a prefix at the start of constructing an SST file, we expect it will rely on SstPartitioner. Here we've added SstPartitioner::ShouldPartitionByPrefix() (not yet used) that will enable this optimization, and avoid the overhead of a virtual callback to SstPartitioner on every key written to an SST file.
Supporting cleanups: the Footer's "unset" format_version sentinel is named (kInvalidFormatVersion) and the FormatVersionUses*() predicates assert against it, as each is a `version >= N` test that an unpopulated Footer would silently satisfy. Index block reader entry points (Block::NewIndexIterator() and friends) drop their defaulted trailing arguments so each caller states which codec it expects, and BlockBuilder::Add() takes delta_value by const reference rather than a pointer whose null case no caller used.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15104
Test Plan:
New unit tests, all exercising fv8 through TEST_AllowUnsupportedFormatVersion():
- block_test: value-delta escape codec round-trip (delta 0, negative, large, and the escape; asserts format_version <= 7 bytes match the legacy encoding and a legacy 0x00 still decodes as delta 0), and rejection of footer bit 30 at all versions.
- db_flush_test: fv7-vs-fv8 read equivalence for super_block_alignment, including the separate_key_value_in_data_block intersection, confirming padding events (hence escape entries) occur and fv8 index values are delta encoded.
- sst_file_reader_test: fv7-vs-fv8 read equivalence for embedded-blob SSTs, confirming fv8 (unlike fv7) delta-encodes their index values and actually emits escape entries (counted through a sync point).
- block_based_table_reader_test: a builder+reader test that injects a footer metaindex gap via a sync point (verifies the gap bytes are written and the reader locates the metaindex via the recorded gap, and that too large a gap surfaces as a builder error), and that a non-empty reserved user_key_common_prefix property makes the reader return NotSupported.
- table_test: two-level index with partitioned filters and index_block_restart_interval > 1, the configuration in which both top-level indexes delta-encode their values; verifies scan and point lookup round trips at each tested format_version including fv8.
- (Not strictly fv8) Basic SstPartitionerFixedPrefix tests in slice_test (ShouldPartition, ShouldPartitionByPrefix, their mutual consistency, and CanDoTrivialMove), which had no non-DB coverage before.
fv8 is also added to the table_test harness (GenerateArgList) and the shared footer version set (test::kFooterFormatVersionsToTest), so the existing table-level and DB-level property/checksum suites now also run at fv8. And added to crash test (minor revisions to db_stress).
## Performance details
Release db_bench (PORTABLE=0, -march=native) A/B/C on the read/write fast path,
comparing (a) parent, (b) this commit @ fv7, (c) this commit @ fv8. All DBs on
tmpfs, 20M keys, block_size=2048, compression=none. Note index_block_restart_
interval=16 (not the default 1) is required to actually exercise the fv7<->fv8
index value-delta codec -- with restart interval 1 every index entry is a full-
handle restart and the changed path never runs (confirmed delta-value=1 via
sst_dump). Reads single-thread + core-pinned, warm cache, interleaved/rotated
rounds, paired deltas.
Reads (readrandom + seekrandom, non-partitioned single ~1.9MB index block and
partitioned index+filters): fv7->fv8 in the *same* binary is within noise on all
four workloads (largest -1.0%, CIs cross zero). before->after (fv7 path) is also
within noise; perf shows a deterministic +0.4-0.6% instructions/op from the added
branches (the format_version>=8 value_delta_escape guard + refactored data-block
footer decode), which is below the throughput noise floor.
Writes (fillseq + compact total CPU): fv7->fv8 null. compact (which rewrites all
index blocks) is null before->after. fillseq showed +3.5% CPU before->after, but
perf attributes it entirely to cycles/IPC with *identical* instructions/op
(+0.01-0.04%) -- a code-layout effect on the untouched memtable-insert loop, not
this change.
Sanity: any single throughput/CPU delta <~1% is within the build-to-build noise
floor, measured directly by comparing two byte-identical binaries (swings up to
~0.9% reads; the pre-native/PORTABLE=1 build showed a spurious -1.85% that
vanished under -march=native).
Net: no measurable regression from fv8 draft; fv7 encoding/decoding byte-for-byte
unchanged, only the noted sub-1% instruction bump on the shared read path.
Reviewed By: xingbowang
Differential Revision: D116105558
Pulled By: pdillinger
fbshipit-source-id: ead776cdd396d497d47872c0956ea8c651c0d13f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15111
T284514290 exposed an online-recovery race between the automatic recovery coordinator and a foreground synchronous write. A retryable no-WAL flush error selected lightweight retry-flush recovery. While that recovery attempt waited for the failed background flush, `Delete(sync=true)` appended and synced sequence N to the active WAL, then failed while persisting WAL-tracking metadata to the MANIFEST. Because memtable insertion and sequence publication happen after that MANIFEST step, the record was durable but logically unpublished.
OLD: the concurrent error triggered another attempt, but the attempt reused a stale recovery plan. That preserved both ingredients needed for a duplicate: the first `Delete(key, N)` remained in the active WAL, while the published sequence remained `N-1`.
```
Recovery coordinator Foreground sync-write thread
------------------------------------ --------------------------------
[db_mutex held]
plan = RetryFlush
(copied once before the retry loop)
attempt 1: ResumeImpl(plan)
WaitForBackgroundWork()
bg_cv_.Wait()
[atomically releases db_mutex] ---------> assign sequence N
append + sync Delete(key, N)
[db_mutex held]
MANIFEST WAL edit fails
no memtable insert
no LastSequence publication
write returns I/O error
recovery_error_ = Hard
(requests another attempt)
bg_error_ = Hard
recover_context_ eventually
becomes FullRecovery
[db_mutex released]
[db_mutex reacquired] <---------
ClearBGError() sees recovery_error_
attempt 1 cannot clear the error
attempt 2:
recovery_error_ = OK
ResumeImpl(plan)
^
+-- still the cached RetryFlush plan
RetryFlushesForErrorRecovery()
retries the failed immutable flush only
does not call SwitchMemtable()
does not switch the active WAL
ClearBGError() succeeds
caller retries Delete(key)
LastSequence is still N-1
=> assigns sequence N again
appends Delete(key, N) to same WAL
memtable insert succeeds
publishes LastSequence = N
Reopen / WAL replay:
first (key, N, deletion) -> memtable insert succeeds
second (key, N, deletion) -> TryAgain("key+seq exists")
normal WAL recovery has seq_per_batch_ == false
assert(seq_per_batch_) fails
```
The cached argument to `ResumeImpl()` is therefore the direct bridge to the assertion: lightweight retry-flush recovery can report success without retiring the WAL that contains the unpublished record. Since it also does not publish sequence N, the successful caller retry writes the identical internal key into that same WAL.
NEW: the same recovery coordinator consumes the concurrent error through a mutex-ordered handoff.
```
Recovery coordinator Foreground sync-write thread
------------------------------------ --------------------------------
[db_mutex held]
attempt 1:
plan = recover_context_ // RetryFlush
ResumeImpl(plan)
WaitForBackgroundWork()
bg_cv_.Wait()
[atomically releases db_mutex] ---------> append + sync Delete(key, seq=N)
[db_mutex held]
MANIFEST WAL edit fails
CheckAndSet...(Hard, FullRecovery):
recovery_error_ = Hard
bg_error_ = Hard
recover_context_ = FullRecovery
[db_mutex released]
[db_mutex reacquired] <---------
ClearBGError() sees recovery_error_
current attempt cannot clear the error
next retry, with db_mutex held:
recovery_error_ = OK
plan = recover_context_ // FullRecovery
ResumeImpl(plan)
SwitchMemtable()
switch to a new WAL
clear the background error
```
The foreground unlock of `db_mutex_` happens before the recovery coordinator reacquires it and snapshots the next plan. Publishing `bg_error_` and `recover_context_` together under that mutex prevents the next attempt from observing a new severity with an old recovery requirement. `recovery_error_` is the per-attempt latch: a concurrent I/O error cannot be cleared as part of the in-flight attempt; it either stops recovery or forces another attempt to consume the updated context. If another error arrives while `ResumeImpl()` releases the mutex again, the same handoff repeats. No second recovery coordinator is needed.
Manual `Resume()` now uses the retained recovery context instead of reconstructing one from the background-work scheduling flag, and the soft-error clear-only shortcut is allowed only when that flag shows there is no pending recovery work. The write ordering and `assert(seq_per_batch_)` detector are unchanged.
Reviewed By: anand1976, xingbowang
Differential Revision: D116106177
fbshipit-source-id: a78450c0924a326b5302ae6709e367f07bd3d5a6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15100
Fix the TSAN race in T284274705 where `CoroutineStatsRequestData::onSet()` can access its borrowed `Env*` after `AsyncCallback::OnComplete()` allows the caller to destroy that environment.
When `co_await baton` suspends, the Task's implicit `ViaIfAsync` saves the read's statistics `RequestContext` (`S`) as the context needed by the real RocksDB continuation. The old explicit `executor->add()` also ran the Baton bridge under `S`. Consequently, `baton.post()` resumed the Via trampoline under `S`, and the trampoline's own `executor_->add()` caused EventBase to capture `S` a second time as the queued continuation's outer context. After the read removed its stats scope and called `OnComplete()`, the Via guard restored that outer `S`, triggering a late `onSet()` that dereferenced the potentially destroyed `Env*`.
Remove the redundant outer executor hop and call `baton.post()` directly under an explicitly null `RequestContext`. If the read is suspended, `post()` resumes only the Via trampoline; the trampoline still queues the real continuation on the bound read executor. The Via promise retains its saved `S`, so `executeContinuation()` installs `S` while RocksDB executes, but EventBase now captures null as the outer context restored afterward. If filesystem completion happens before `co_await baton`, `post()` only marks the Baton ready and the read continues inline under its existing context.
Thread Race:
`E` = temporary backup `Env`
`S` = statistics `RequestContext` containing `CoroutineStatsRequestData{env_ = E}`
`P` = parent context without RocksDB statistics
`B` = old explicit Baton bridge
`C` = `ViaIfAsync` callback that resumes the real read
```
Caller / stress thread RocksDB read EventBase thread
---------------------- -----------------------------
BackupInfo owns shared_ptr<Env> E
Call MultiGetAsync()
Wait for OnComplete() Start the RocksDB read coroutine
Install S:
P -> S
S.onSet()
Call SubmitReadAsync()
No io_uring is available, so:
Read() executes synchronously
filesystem callback runs inline
callback is still under S
Old code calls executor->add(B)
EventBase records:
B.outer_context = S
Reach co_await baton
B has not executed, so Baton is not ready
ViaIfAsync:
saves execution_context = S
gives trampoline V to Baton
Suspend the real read
Leave S:
S -> P
S.onUnset()
Later, EventBase runs B
Install B.outer_context:
P -> S
S.onSet()
B calls baton.post()
-> V resumes inline
-> V queues callback C
-> EventBase records:
C.outer_context = S
Later, EventBase runs C
Install C.outer_context:
P -> S
S.onSet()
Via executeContinuation():
remembers previous context = S
installs saved execution context S
resumes the real RocksDB read
Read finishes
Statistics scope removes S:
S -> P
S.onUnset()
Call OnComplete()
Wait returns <--------------- OnComplete signals caller
================ PUBLIC READ IS COMPLETE ================
Destroy restored DB C is still unwinding
Release BackupInfo Via guard restores previous context:
Last shared_ptr<Env> disappears P -> S
Begin Env E destruction S.onSet()
-> E->GetThreadID()
Env::~Env() <--------------- TSAN race ---------------> S.onSet()
```
The same context was captured in two roles:
```
Via execution_context = S required while the real read executes
C.outer_context = S accidentally restored after the read completes
```
The fix removes bridge `B` and posts under null. For an outstanding asynchronous read, the Via trampoline therefore queues `C` with `outer_context = null`, while retaining `execution_context = S`. The real read still executes under `S`, but after `OnComplete()` the Via guard restores null instead of reactivating `S`.
Reviewed By: anand1976
Differential Revision: D115967513
fbshipit-source-id: 87d744ecdbad3421c4d2a1a7be6423a3f0373573
Summary:
The parallel-compression auto-skip feature (https://github.com/facebook/rocksdb/issues/15079) introduced a data race that TSAN reports in block_based_table_reader_test. In StateTransition(), the auto-skip loop reads
ring_buffer[cslot].auto_skip_bypassed speculatively, from a not-yet-committed atomic_state snapshot, before the CAS that would validate that snapshot. Unlike every other BlockRep field (touched only after a successful CAS grants exclusive slot ownership), a stale snapshot can point that read at a physical ring-buffer slot the emit thread is concurrently recycling for a new block in EmitBlockForParallel, racing the plain write there. The speculative result is discarded when the CAS fails, so there is no logical corruption, but the plain read/write is a data race (UB) and is flagged by TSAN.
Make auto_skip_bypassed a RelaxedAtomic<bool> so the speculative read and the emit-thread write are atomic. Relaxed ordering suffices: correctness of the auto-skip decision is still gated by the CAS, and when the CAS succeeds the value was published by the emit thread through the existing acquire/release ordering on atomic_state and is not modified elsewhere.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15094
Test Plan: Covered by existing block_based_table_reader_test parameterizations that build with parallel compression (parallel_threads > 1) under TSAN, which is where the race was originally reported.
Reviewed By: joshkang97
Differential Revision: D115777206
Pulled By: pdillinger
fbshipit-source-id: 7ee969ec2d95890255d2e8b76eceaa5876b82170
Summary:
Custom filter and compression strategies decide what to build per SST from a FilterBuildingContext (shared by FilterPolicy::GetBuilderWithContext and CompressionManager::GetCompressorForSST). This adds more context so those strategies can adapt to where and how a file is being produced.
FilterBuildingContext gains two fields:
* is_remote_compaction: true only when the file is produced by an offloaded CompactionService worker. Detected with a new virtual CompactionJob::IsRemoteCompaction() (default false) overridden by CompactionServiceCompactionJob, mirroring the existing GetTableFileName hook. Local/fallback compactions and flushes are false.
* db_name: a non-owning Slice naming the DB, populated for flush, compaction, recovery, and repair, and empty only where there is no DB, as with SstFileWriter. Under remote compaction it is the DB as opened by the worker, which can differ from the primary's.
Both are threaded through TableBuilderOptions and populated in BlockBasedTableBuilder::Rep, whose "is this file in the LSM" test becomes a switch with no default case, so that a future TableFileCreationReason has to be classified there.
Separately, TableFileCreationReason is now backed by uint8_t and gains a kSstFileWriter value, which SstFileWriter reports instead of the catch-all kMisc. It is appended after kMisc so existing enumerator values are unchanged; those values are public API, surfaced to EventListener and mirrored by the Java binding (updated here). Note that SstFileWriter neither notifies listeners nor consults CompactionFilterFactory, so for now the new value is only observable through FilterBuildingContext.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15085
Test Plan:
* New CompactionServiceTest.IsRemoteCompactionInCompressorContext installs a custom CompressionManager and asserts GetCompressorForSST sees is_remote_compaction == true for the offloaded worker's output files and false for the primary's flushes in the same run, so the flag is shown to be per-file rather than global. IsRemoteCompactionFalseForLocalCompaction covers the same manager with no compaction_service configured.
* DBBloomFilterTest.ContextCustomFilterPolicy now reports db_name as a token (expected / none / UNEXPECTED:<value>), so the value rather than just its presence is checked for flush and compaction, and expects kSstFileWriter instead of kMisc for the external SstFileWriter file.
* bloom_test's "Like SST file writer" case uses kSstFileWriter so that it keeps modelling SstFileWriter.
* Java EventListenerTest references TableFileCreationReason.MISC.getValue() instead of a hardcoded enum byte, so it stays correct across enum changes.
Reviewed By: jaykorean
Differential Revision: D115603616
Pulled By: pdillinger
fbshipit-source-id: ad73fbc391a5377ace406725238ae6eede46c8d9
Summary:
Add a db_crashtest liveness mode that runs db_stress with a watchdog for
foreground progress. The profile disables fault injection and uses longer runs
so failures are more likely to represent real forward-progress bugs instead of
expected injected I/O errors. The watchdog tracks completed operations, active
operation type/start time, and per-operation completion counts to distinguish
stuck writes from long admin operations.
Bug: The new mode exposed a latent flush scheduling bug. A flush request can be queued
with an effective max memtable id, then a newer immutable memtable can be added
before the background flush picks work. PickMemtablesToFlush() could pick only
the older memtables and still clear flush_requested_. A later queued flush for
the newer memtable would then be skipped because the column family no longer
looked flush-pending, leaving the immutable memtable permanently unflushed. In
stress this surfaced as IngestExternalFile() waiting in WaitForFlushMemTables(),
with writes eventually blocked behind the same unflushed memtable.
Fix: Keep flush_requested_ set until all unstarted immutable memtables have been
picked. Add a direct MemTableList regression for the partial-pick case and a
DB-level regression covering the ingest/get-live-files ordering that reproduced
the lost request.
## Testing
- CI
- Local liveness mode 32 full 20-minute runs passed, 0 failures
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14830
Reviewed By: joshkang97
Differential Revision: D107895052
Pulled By: xingbowang
fbshipit-source-id: ca67c5f9376899eefdc9af7bad4af06200b52446
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15061
Adds mutable ColumnFamilyOptions field `blob_file_writable_file_max_buffer_size` to cap blob-file WritableFileWriter buffers independently from DBOptions::writable_file_max_buffer_size.
The default value is 0, which preserves existing behavior by inheriting the DB-wide writer buffer cap. Non-zero values apply to both flush/compaction blob writers and direct-write blob writers. Direct-write updates are propagated through DBImpl::SetOptions to newly opened blob files without requiring a DB restart.
Also wires option parsing, dumping, db_bench, db_stress, crashtest, and release-note coverage.
Reviewed By: xingbowang
Differential Revision: D114979265
fbshipit-source-id: 3f764b797628eee0ce379110c1a1f8a16beb3fa2
Summary:
The Claude/Codex CI reviewer was posting PR comments like "the review has been written to review-findings.md" instead of the review itself, and the comment footer cited claude_md/code_review.md as the review methodology even though CI stopped feeding that file to the model in https://github.com/facebook/rocksdb/issues/14507 (it switched to claude_md/ci_review_prompt.md). code_review.md had been orphaned from CI for months while still advertising itself as the CI methodology, so edits to it (e.g. the "post directly in your response" note) had no effect on CI. Separately, an over-long review exceeded GitHub's per-comment character limit and was rejected outright, losing the whole review.
Changes:
- ci_review_prompt.md: make the final response the review -- it becomes the PR comment verbatim -- and forbid ending the turn with a meta-summary such as "written to review-findings.md". review-findings.md is demoted to a recovery backstop; the per-phase incremental writes and the turn-limit recovery path are unchanged, so both the normal and recovery comment paths still produce a real review.
- Repoint the "Review methodology" footer in parse-claude-review.js and parse-codex-review.js to claude_md/ci_review_prompt.md (the file CI actually uses), so the citation no longer points at a deleted file.
- build-ai-review-comment.js: cap the assembled comment at GitHub's 65536-char limit, trimming only the model response (the sole unbounded input) and appending a pointer to the execution-log artifact, so an oversized review is truncated cleanly instead of rejected by GitHub. The header, footer, and <details> scaffolding are always preserved.
- Delete claude_md/code_review.md (CI-dead since https://github.com/facebook/rocksdb/issues/14507; superseded locally).
- Add claude_md/manual_code_review.md: a lightweight, single-reviewer local wrapper around ci_review_prompt.md (no multi-agent team/debate, no scratch files, inline output) for running a modest review by hand. It sources its methodology from ci_review_prompt.md so it cannot drift.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15087
Test Plan:
- The workflow scripts have no CI unit coverage; verified both parse-*-review.js still parse, and that no reference to code_review.md remains anywhere.
- Dry-ran manual_code_review.md by reviewing the current commit locally; it produced a right-sized single-reviewer report without spawning agents or writing files, and helped find+fix the comment limit issue.
- Exercised the end-to-end GitHub CI path (per https://github.com/facebook/rocksdb/issues/14696, review-workflow changes not yet on main can only be tested from a branch on facebook/rocksdb): pushed these changes to the fix_hidden_review_details_test branch and ran the Claude Code Review workflow (claude-review.yml) via "Run workflow" against PR https://github.com/facebook/rocksdb/issues/15079, confirming the posted comment carries the full review inline (not a review-findings.md summary) with the ci_review_prompt.md footer. https://github.com/facebook/rocksdb/actions/runs/31537781229https://github.com/facebook/rocksdb/pull/15079#issuecomment-5259127037
Reviewed By: xingbowang
Differential Revision: D115635558
Pulled By: pdillinger
fbshipit-source-id: d3ca7188162a086cd42d7b8bd8e60f3fc3df3bad
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15071
__Why__: Give operators visibility into currently running bottom-priority compactions. Low-priority running compactions can be derived from `rocksdb.num-running-compactions - rocksdb.num-running-bottom-compactions`.
__What__: Expose `rocksdb.num-running-bottom-compactions` as a new statistics property.
__How__: We already record `num_running_bottom_compactions_`, so this feeds that value through the DB property path.
Reviewed By: xingbowang
Differential Revision: D114989532
fbshipit-source-id: efa19ef3a69eaecf4cbbbc9e4fc9cd55fce09ca5
Summary:
Auto-skip compression (stop attempting compression on data that isn't compressing well, to save CPU) was previously an experimental CompressionManager wrapper (CreateAutoSkipCompressionManager) whose predictor lived in each per-thread working area. Besides being fragmented across parallel workers and reset on a tumbling window, it decided based on how OFTEN blocks failed to compress, ignoring how MUCH they compress -- so on mixed data it could give up substantial compression to save a little CPU (see performance results).
This makes auto-skip a first-class, optional, ratio-aware feature of the block-based table builder, controlled by two new CompressionOptions fields:
- auto_skip (default false): the on/off switch. There is no separate threshold knob; the decision reuses max_compressed_bytes_per_kb (the existing per-block "minimum compression worth keeping" bar) as the bar for whether attempting compression on the stream is worthwhile.
- auto_skip_min_sample_every: sampling interval that bounds the local impact around a transition back to compressible data. 0 selects an internal default (currently 64 skipped data blocks between forced samples).
The emit thread makes the skip/attempt decision once per data block (serial and parallel paths alike; data blocks only), so a skipped block is never dispatched to a compression worker. It is driven by a continuously updated, shared estimate q_hat of the aggregate stored/uncompressed ratio the builder would achieve by attempting -- each block counted at its achieved ratio, or 1.0 when stored raw: attempt while q_hat is at or below the bar, otherwise skip.
Because the estimate is payoff-weighted, a stream whose blocks each save little -- or where only a few blocks compress well -- does not clear the bar and is skipped, while a stream that compresses substantially is attempted. The compression given up per KB is bounded by (1024 - max_compressed_bytes_per_kb).
q_hat uses an asymmetric exponentially weighted moving average (EWMA), slightly faster toward better compression than toward worse; the fast side is deliberately gentle so a single marginally-compressible sample does not flip a whole file back to attempting. It is kept on its own cache line to avoid false sharing. Under parallel compression q_hat is updated with relaxed atomics, and the emit thread can decide from a value that lags the learning it drives by the in-flight depth -- tolerable concessions of accuracy and responsiveness for a coarse heuristic.
The first data block of each file is always attempted (the sample countdown starts at 0), and while skipping the builder keeps sampling (one block after a randomized gap) so a return to compressible data is detected. q_hat is carried across the files a flush/compaction thread emits via thread_local state keyed by preferred compression type, so a sustained region does not re-learn on every file.
The old CompressionManager-based AutoSkip is removed (CostAware is retained); db_stress, db_bench, and db_crashtest are migrated to the new option. db_bench is extended so it can generate the relevant regimes: --compressible_data_fraction and --compressible_data_run_length produce a mix of compressible/incompressible blocks -- i.i.d. per value (run length 1) or in alternating regions (large run length). The default (fraction 1.0) reproduces the historical single-pool behavior.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15079
Test Plan:
New/updated unit tests in `util/compression_test.cc` (`DBAutoSkipTest`):
- skips sustained incompressible data;
- never skips compressible data;
- disabled by default never bypasses a block;
- 4-thread parallel compression round-trips all data across alternating compressible/incompressible regions.
**Performance** (`db_bench fillseq`, release, zstd level 3; 100k values of 16 KiB each, roughly one value per data block; WAL and auto-compaction off; median of 5 runs on a quiet host). Auto-skip trades one cost for another, so each cell reports two costs, both lower-is-better: CPU cost as process CPU time (user+sys) per GB of uncompressed input (s/GB; ~1.64 GB uncompressed here), and storage cost as the resulting DB size. Auto-skip does not appear to change parallelizability, so wall-clock s/GB tracks CPU s/GB (the two agree here); CPU s/GB is reported. Percentages are relative to the no-auto-skip baseline: negative CPU = cheaper, positive size = more storage (shown only where storage grows).
| Workload | baseline (off) | (a) old wrapper | (b) draft | (c) this version |
|---|---|---|---|---|
| Incompressible | 2.33 s/GB, 1568 MB | 0.96 (-59%), 1568 MB | 0.82 (-65%), 1568 MB | 0.82 (-65%), 1568 MB |
| Compressible | 3.55 s/GB, 710 MB | 3.54 (~0%), 710 MB | 3.56 (~0%), 710 MB | 3.55 (~0%), 710 MB |
| Marginal-payoff mix | 3.80 s/GB, 1376 MB | 1.54 (-59%), 1523 MB (+11%) | 3.81 (~0%), 1376 MB | 1.43 (-62%), 1531 MB (+11%) |
| High-payoff mix | 2.20 s/GB, 929 MB | 1.14 (-48%), 1417 MB (+53%) | 2.16 (~0%), 929 MB | 2.17 (~0%), 929 MB |
A strategy wins by cutting CPU cost without paying much storage cost. On incompressible data all three cut CPU ~60% at no storage cost; on compressible data none skip (within run-to-run noise). The two mixes (50% incompressible, 50% compressible per block) separate the strategies:
- Marginal-payoff mix: the compressible half barely clears the per-block bar, so the stream is not worth compressing -- skipping is right. (c) and (a) cut CPU ~60% for ~11% more storage. That extra storage is compression (c) intentionally declines because its aggregate benefit is below the configured worthwhile threshold (max_compressed_bytes_per_kb, ~12.5% per KB); it stays within the (1024 - max_compressed_bytes_per_kb) bound. The ratio-unaware draft (b) cannot see the low payoff, keeps compressing, and saves no CPU.
- High-payoff mix: the compressible half compresses hard, so compressing is right. (c) and (b) keep full compression at baseline storage; the old wrapper (a) cuts CPU 48% but inflates storage +53% -- far beyond that bound -- giving up worthwhile compression.
Only (c) makes the right call on both mixes: it never gives up worthwhile compression (unlike (a)) and does not waste CPU on low-payoff data (unlike (b)).
Reviewed By: xingbowang
Differential Revision: D115444567
Pulled By: pdillinger
fbshipit-source-id: 05e122d25de95ddcd0edeb0dd07c9d25bead7293
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15084
MaybeOpenReadOnlyOnPrimary is intended to exercise opening and closing a read-only DB on the primary live directory, then verify that read-only close does not delete files still live in the primary.
D115278311 diagnosed T283967366 as a race between GetEntityLazy on that read-only handle and the primary deleting SST files. After checking the call path, the ENOENT is from GetEntityLazy's initial point lookup opening an SST through TableCache, not from deferred lazy column resolution. Other read APIs are not exercised on this read-only-on-primary handle, so adding GetEntityLazy there made this old open/close harness test also perform a live-directory point read.
Remove the read-only-handle GetEntityLazy exercise and keep GetEntityLazy coverage in the normal stress read paths, consistent with the other read operations.
Reviewed By: pdillinger
Differential Revision: D115569968
fbshipit-source-id: fcef30d09903525c2a7897b9c52bc9ef951bed6a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15074
Why: We want to be able to monitor how many remote compaction jobs are running on a given server. Currently we know the total number of running compactions but not the local vs remote breakdown.
What: Add a new `rocksdb.num-running-remote-compaction-jobs` statistic. This measures in-progress remote compaction jobs.
Note that remote compaction work is scheduled per subcompaction, so one RocksDB compaction may contribute more than one running remote compaction job.
Reviewed By: jaykorean
Differential Revision: D115324149
fbshipit-source-id: 9397e39acd696c88ce7499ecb0da68de215bf1f3
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15076
Fix the `io_activity == Env::IOActivity::kUnknown || io_activity == options.io_activity`
assertion failure in the crash test. The crash occurs because `TestMultiGetEntity` in
`no_batched_ops_stress.cc` changes the thread operation to `OP_GETENTITY` during its
consistency-check phase (the `check_results` lambda calls `SetThreadOperation(OP_GETENTITY)`
before calling `GetEntity` for comparison). When `MaybeTestMultiGetEntityLazy` is called
afterward, it invokes `MultiGetEntityLazy` which correctly sets
`read_options.io_activity = kMultiGetEntity`, but the thread operation is still stale at
`OP_GETENTITY`. The debug assertion in `DbStressRandomAccessFileWrapper::Read` then sees
`TEST_GetExpectedIOActivity(OP_GETENTITY) = kGetEntity != kMultiGetEntity` and aborts.
The fix restores `SetThreadOperation(OP_MULTIGETENTITY)` before each call to
`MaybeTestMultiGetEntityLazy` in `TestMultiGetEntity`.
Triggered by commit cb73fc845fe0 which added the `MaybeTestMultiGetEntityLazy` calls.
The `--lazy_entity_read_one_in=100` flag (non-default) enables this code path.
Reviewed By: pdillinger
Differential Revision: D115021723
fbshipit-source-id: e3eed600db4986cab27d7b1924282b1dfb8b0a7a
Summary:
## Summary
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15080
Introduce reusable `CoroStackableDBBase` forwarding and public `CoroDB::CoGet()` / `CoMultiGet()` entry points matching the `DB::Get()` / `MultiGet()` overloads. These helpers schedule native coroutine reads on the read executor and use the synchronous DB API when coroutine support or a read executor is unavailable.
Use the stackable coroutine hooks for TTL and WritePrepared transaction reads. TTL preserves expiration decoding and timestamp behavior, while WritePrepared preserves transaction visibility.
TTL keeps a single `DB` inheritance path while adding `CoroDB`:
```
DBWithTTLImpl
`-- CoroStackableDBBase<DBWithTTL>
|-- DBWithTTL -> StackableDB -> DB
`-- CoroDB
```
`CoroDB::CoGet()` dispatches through the stackable coroutine hook and schedules the operation on the read EventBase. `CoMultiGet()` follows the same path:
```
CoroDB::CoGet(db, ...)
|-- db->GetCoroDB() -> DBWithTTLImpl
|-- no CoroDB or read executor -> db->Get()
`-- schedule on read EventBase
`-- virtual CoroDB::GetCoroutine()
`-- DBWithTTLImpl::GetCoroutine()
|-- CoroStackableDBBase<DBWithTTL>::GetCoroutine()
| `-- wrapped CoroDB::GetCoroutine()
`-- strip TTL timestamp
```
Reviewed By: xingbowang
Differential Revision: D115232510
fbshipit-source-id: 59e5090db28d15bcff1a8b9b8c0b30ca7eeb742d
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15078
The legacy regression test runner attempted to discover multi-DB origin directories by running find through run_remote, then assigned dirs=$?, which captured only find's exit code. For NUM_MULTI_DB > 1 this meant the checkpoint loop operated on the value 0 instead of the discovered sub-DB directories.
For NUM_MULTI_DB > 1, build checkpoints for the deterministic numeric sub-DB directories 0..NUM_MULTI_DB-1 and explicitly fail if any expected origin DB is missing. This avoids rediscovering DBs through find output, so checkpoint creation no longer depends on pipeline exit status or whitespace-sensitive word splitting.
Reviewed By: joshkang97
Differential Revision: D115380183
fbshipit-source-id: 24adfa37f872dd7b8ee4ee986cf4a7a4e58562d0
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15077
__Why__: Allow offload compaction workers to configure `max_open_files` via `options_map`.
__Problem__: `OpenAndCompact` hard-codes `max_open_files` to -1. This overwrites any value set via the `options_map` and prevents configuration in the remote compaction workers.
__Fix__: Remove the hard-coding assignment.
The value now comes from:
1. OPTIONS file via `LoadOptionsFromFile`
2. `override_options.options_map` via `GetDBOptionsFromMap`
The default value of `max_open_files` was already -1, so existing behavior is preserved.
Reviewed By: jaykorean
Differential Revision: D115373592
fbshipit-source-id: fa742397246092544ab6f9004678be7ae938d051
Summary:
## Summary
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15073
The callback-based async read APIs make coroutine-native implementations and StackableDB composition awkward because wrappers must bridge callbacks back into tasks. This change introduces `CoroDB` as a side capability exposing lazy `GetCoroutine` and `MultiGetCoroutine` tasks. `DBImpl` and `CompactedDBImpl` implement this capability directly, while `DB::GetAsync` and `DB::MultiGetAsync` remain available as compatibility entry points with synchronous fallback.
The native interface should also be more efficient because callers directly await the DB coroutine. It avoids the callback/Baton adapter, callback context handoff, and explicit rescheduling required for each request. Native tasks must run on the DB filesystem read executor's EventBase; debug builds verify this affinity.
Public API documentation, a release note, and native-coroutine support in `db_bench` and `db_stress` are included.
## Dispatch
DB::GetAsync
-> GetCoroDB() on the concrete DB
-> virtual CoroDB::GetCoroutine(...)
-> DBImpl::GetCoroutine or CompactedDBImpl::GetCoroutine
-> AsyncCallback::OnComplete(...)
Virtual dispatch selects the coroutine implementation belonging to the concrete DB while preserving the existing callback API.
## Statistics
Coroutine reads preserve RocksDB's TLS statistics behavior while keeping concurrent requests isolated. Each top-level task copies only the caller's statistics configuration—`PerfLevel`, per-level perf context enablement, and IO statistics enablement—into fresh request-local `PerfContext` and `IOStatsContext` objects. Existing counters are not inherited. If both perf and IO statistics are disabled, no request statistics context is installed.
Folly `RequestContext` preserves these contexts across suspension and resumption on the read executor's EventBase. At completion, RocksDB publishes the request's collected values to TLS, allowing the awaiting caller to use `get_perf_context()` and `get_iostats_context()` as after a synchronous read. Callers must consume or copy them before leaving that EventBase. Statistics scopes are installed only by top-level coroutine reads and cannot be nested.
## Benchmarks
Compared the previous `GetAsync` callback bridge with native `GetCoroutine`. Both variants used optimized builds, the same compacted DB, a fixed seed, eight executor threads, ten jobs per thread, and 2.56M reads per invocation. Results are the mean of three independent invocations using the default RocksDB block cache. A sequential read warmed the OS page cache before the non-direct-read measurements.
| Workload | `GetAsync` bridge | Native `GetCoroutine` | Change |
| Direct I/O | 237,710 ops/s | 243,588 ops/s | +2.47% |
| Warm OS page cache | 783,859 ops/s | 891,531 ops/s | +13.74% |
The direct-I/O difference is within observed run-to-run variation. When reads are served from the OS page cache and API overhead is a larger portion of latency, the native interface improves throughput by 13.74%. Every invocation completed all requested reads.
```bash
db_bench \
--benchmarks=readrandomcoroutine \
--db=$CORO_BENCH_DB \
--use_existing_db=true \
--num=2000000 \
--value_size=1024 \
--compression_type=none \
--threads=8 \
--coro_jobs_per_thread=10 \
--reads=320000 \
--use_direct_reads=<true|false> \
--perf_level=1 \
--statistics=true \
--seed=20260807
```
Reviewed By: xingbowang
Differential Revision: D115232511
fbshipit-source-id: 30b48da2a798094d9080178f5430aae26e6c5954
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15060
Expose the `rocksdb.num-unscheduled-compactions` DB statistic to help troubleshoot high L0 counts.
The metric captures the raw `unscheduled_compactions_` count. It measures compaction work waiting in the DB internal compaction queue, but not yet assigned to a background job.
This complements `rocksdb.compaction-pending`, which tells us "some compaction work is needed," and `rocksdb.num-running-compactions`, which tells us how many background compactions are executing.
`rocksdb.compaction-pending` is just a boolean value so we can't infer too much from it. `rocksdb.num-unscheduled-compactions` can be used to indicate that we may need to increase the max number of background jobs.
Reviewed By: xingbowang
Differential Revision: D114539448
fbshipit-source-id: 3657951bd9d8a4f4a7bc884513f9a06a87892b50
Summary:
Builds on the lazy wide-column read API (DB::GetEntityLazy / DB::MultiGetEntityLazy) to serve true byte-range reads: resolving a sub-range of an uncompressed blob column now reads only the requested bytes from storage instead of fetching and slicing the whole column. This covers blobs in separate blob files and embedded (same-file) blobs in the SST, so a caller that knows a large value's layout can pull just the parts it needs.
A partial read skips whole-record checksum verification (a strict sub-range cannot cover the checksum) and does not populate the blob cache (the cache entry holds the whole record). A compressed column, a whole-column read, a cache hit, or a read with LazyColumnReadRequest::force_verify still resolves (and verifies) the whole column; force_verify forces verification even when ReadOptions::verify_checksums is off.
Mechanics: new range primitives mirror the whole-value read paths -- BlobFileReader::GetBlobRange, BlobSource::GetBlobRange, Version::GetBlobRange, and VersionBlobFetcherBase::FetchBlobRange for separate-file blobs; ReadSimpleGen2BlobRange, BlobSource::GetSimpleGen2BlobRange, and SameFileBlobReader::GetSameFileBlobRange (implemented by BlockBasedTable) for embedded blobs. ReadPathBlobResolver::ResolveColumnRange owns the partial-vs-whole decision and routes same-file references through the current SST's reader; the whole-column path also gained
VersionBlobFetcherBase::FetchBlobForceVerify to honor force_verify.
Observability: deferred resolve reads are attributed to a new Env::IOActivity::kLazyResolve (distinct from the kGetEntity / kMultiGetEntity of the initial entity read), and lazy resolution's storage reads are counted by new statistics rocksdb.blobdb.lazy.read.count / .read.bytes (all lazy reads) and rocksdb.blobdb.lazy.partial.read.count / .partial.bytes.saved (the byte-range subset and the bytes it avoided). The whole-column read sites gate recording on kLazyResolve so non-lazy reads are excluded.
Also fixes a latent bug in LazyWideColumnsBatch::MultiResolve where an empty/default-constructed batch (null rep_) would accept a foreign standalone column (null owning_batch_rep_) through a nullptr == nullptr comparison instead of returning InvalidArgument.
Follow-ups (in a local plan): unifying the blob-read API surface -- folding the whole/range method twins and the verify-policy flags (verify_checksums x force_verify) into one internal BlobReadRequest primitive (a range descriptor plus a 3-valued verify policy), which retires FetchBlobForceVerify and the whole/range method pairs; this is deferred for feature velocity and is not public-API affecting (the public force_verify bool stays). Marked in code with TODO(lazy-blob-resolution-cleanup) in db/blob/blob_fetcher.h and db/wide/read_path_blob_resolver.cc. Also cross-key coalescing (Phase 3) and async execution (Phase 4).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15067
Test Plan:
Extended db/wide/db_lazy_entity_test.cc: separate-file and embedded partial reads return the exact sub-span and read only the requested bytes; a partial read does not fill the blob cache and a cache hit slices with no I/O; a compressed column resolves whole and slices; force_verify does a full verified read and is detected via a tampered record even with verify_checksums off; the lazy statistics distinguish whole vs partial reads and are not touched by non-lazy reads; deferred reads carry kLazyResolve (verified with a FileSystem wrapper); and a regression test for the empty-batch ownership check. Added an IOActivityToString case in env/env_test.cc and registered the new tickers in the statistics name map.
Crash test: extended the db_stress lazy-vs-eager differential (db_stress_tool/db_stress_test_base.cc) to fuzz per-column byte ranges. Each resolved column now randomly picks a whole-column read or a partial (offset, length) sub-range -- occasionally with an offset past the end or a length beyond the remainder to exercise clamping -- for both GetEntityLazy and MultiGetEntityLazy; when an eager reference is available the resolved bytes are compared against the same clamped sub-range of it. No new flags (reuses --lazy_entity_read_one_in); force_verify fuzzing remains a follow-up.
Reviewed By: xingbowang
Differential Revision: D115130388
Pulled By: pdillinger
fbshipit-source-id: 454368dea8fae069d9de649c2c51611469d3c809
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15059
**What**:
1. Extends the parallel `CheckpointEngine` so it can also checkpoint a subset of column families.
2. Adds `db_stress` and `db_crashtest.py` coverage for subset checkpoints, over both the serial and the parallel API.
**How**:
- **Engine API.** `CreateCheckpointOptions` gains a `std::vector<ColumnFamilyHandle*> column_families` field (empty, the default, still means all column families). This goes on the existing per-operation options
truct rather than adding a second pure virtual to `CheckpointEngineBase`: it avoids overload-resolution hazards with the defaulted `sequence_number_ptr`, and it is the pattern the engine API already uses.
`CreateCheckpointOptions` has no users outside the engine API, so adding a field breaks nothing.
- **Shared validation.** The handle checks that were inline in `CheckpointImpl::CreateCheckpoint` -- reject null handles, handles belonging to another DB, and handles for already-dropped families; coalesce ids;
always add the default CF -- move to `static CheckpointImpl::ResolveIncludeColumnFamilyIds()`, now called by both APIs. Rejecting a foreign handle matters because `handle->GetID()` alone would silently coerce a
foreign id onto whichever family in this DB happens to share it. Validation runs before any staging directory is created, so a rejected request leaves nothing behind.
- **Almost no new plumbing.** `CheckpointImpl::CreateCheckpointImpl()` already accepted `include_cf_ids` and is shared by the serial and parallel movers, and the MANIFEST `kColumnFamilyDrop` rewrite runs after
`mover->Finish()`, so it was already mover-agnostic. The engine change is just resolving the ids and forwarding them.
- **db_stress.** New `subset_cf_checkpoint_one_in` knob (mirrors `parallel_checkpoint_one_in`): each checkpoint operation covers a random proper subset with probability 1/N. It is a no-op when the DB has only the
default column family, and it is independent of `parallel_checkpoint_one_in`, so subset checkpoints get exercised through both the serial API and the engine. Subset selection includes each non-default family
with probability 1/2 and then force-excludes one, so every subset checkpoint exercises the MANIFEST rewrite instead of degenerating into a whole-DB checkpoint; the default handle is passed in only half the time,
covering the "default is always included whether or not you pass it" contract on both sides.
- **db_stress validation.** `DB::ListColumnFamilies()` on the checkpoint must return exactly the requested set -- the direct check on the MANIFEST post-processing, catching both an excluded family that was not
recorded as dropped and a requested family that went missing. The checkpoint is then opened read-only with exactly those descriptors, proving the reduced MANIFEST and reduced file set are self-consistent under
the default `paranoid_checks`. The pre-existing per-key comparison against `SharedState::Exists()` still runs, restricted to the families actually present.
- **Incompatibility.** `clear_column_family_one_in > 0` is rejected together with the new knob, the same way D113103899 did for `abort_and_resume_cf_compactions_one_in`: a subset checkpoint hands live handles
from `column_families_` to the API and validates against `column_family_names_`, and neither is stable while families are dropped and recreated underneath. `db_crashtest.py` already pins
`clear_column_family_one_in` to 0.
- **Test cleanup.** The subset tests shared a lot of setup and verification, so `PopulateFourCfsForSubsetCheckpoint()`, `VerifyFourCfsIntact()` and `VerifyDefaultAndTwoSubsetCheckpoint()` were extracted into the
test file's anonymous namespace and are reused by the pre-existing `CheckpointSubsetOfCF` as well as the four new engine tests.
This extends `StressTest::TestCheckpoint` with a mode rather than adding a separate operation, which is how `parallel_checkpoint_one_in` was integrated.
Reviewed By: rban1
Differential Revision: D114832267
fbshipit-source-id: f3f843819614477497cc1466910547922cafc51d
Summary:
- make the ARM nightly crash-test swap setup skip `/swapfile` creation when it is already active
- keep the existing 4 GiB swap creation path for hosts without active `/swapfile`
- print the active swap devices after setup for easier workflow diagnostics
## Testing
- `git diff --check upstream/main...2026_08_06_arm_failure`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15065
Reviewed By: archang19
Differential Revision: D115045941
Pulled By: xingbowang
fbshipit-source-id: ff2248b8501c82f5f9e7561bd77937d4823bc67f
Summary:
Adds an experimental **lazy wide-column read API** that lets callers enumerate a
wide-column entity and read back only the columns (and, in a later phase, the
byte ranges) they actually need, instead of eagerly materializing every
blob-backed column like `GetEntity()` / `MultiGetEntity()`.
New public entry points (`include/rocksdb/db.h`):
- `DB::GetEntityLazy(read_options, cf, key, LazyWideColumns* result)`
- `DB::MultiGetEntityLazy(read_options, cf, num_keys, keys, LazyWideColumnsBatch* result, statuses, sorted_input)`
Both leave a wide-column entity's **inline** columns materialized up front but
its **blob-backed** columns as *unresolved references*, so blobs for columns
that are never pulled are never read.
### Result types (`include/rocksdb/lazy_wide_columns.h`)
- **`LazyWideColumn`** — the lazy analogue of `WideColumn`: `name()`, `index()`,
`inline_value()` (an `OptSlice`; empty ⇒ unresolved reference),
`is_reference()`, and `logical_size()` (`std::optional<uint64_t>`; known for
inline and uncompressed references, empty for compressed ones). All no-I/O.
- **`LazyWideColumns`** — a move-only result that is **indexable and iterable
like a vector** of `LazyWideColumn` (`size()`, `operator[]`, range `begin/end`).
Unlike `PinnableWideColumns`, it may be used after the producing call returns:
it **pins the SuperVersion** (like an iterator) so referenced blob files / SST
readers stay valid for deferred reads.
- **`LazyWideColumnsBatch`** — the `MultiGetEntityLazy` result; vector-like over
its per-key `LazyWideColumns`.
### Resolution
- **`LazyColumnReadRequest { const LazyWideColumn* column; uint64_t offset;
size_t length; bool force_verify; PinnableSlice* result; Status* status; }`** —
a read names a **typed column pointer**, not an untyped index. Because a
column identifies its owning result (and, for a batch, its owning entity),
`MultiResolve` routes each read correctly and **rejects a null or foreign
column** with `InvalidArgument` on that read's per-read status — cross-result
misuse can't be silently mis-indexed.
- **`LazyWideColumns::MultiResolve` / `LazyWideColumnsBatch::MultiResolve`** take
the same request type (pointer+count and `std::vector` sugar overloads); the
batch form resolves reads spanning any entities of the batch in one call.
- **`ResolveColumn(column, result)` / `ResolveColumnRange(column, offset, length,
result)`** — per-column sugar over a one-entry `MultiResolve`.
- A column's blob is fetched **at most once** over the result's lifetime (bytes
are cached in the result), so repeated reads of the same column reuse that
fetch. Resolution honors the `ReadOptions` captured at `GetEntityLazy()` time
(e.g. `read_tier`, `verify_checksums`); the resolve methods intentionally take
no per-resolve `ReadOptions`.
### Where it works
- Primary, **read-only**, and **secondary** DB instances (each `GetImpl`
override forwards the lazy signal and transfers a SuperVersion pin via the
shared `DBImpl::TransferSuperVersionPin` helper).
- Forwarded through `StackableDB` wrappers (e.g. `TransactionDB`) rather than
falling through to the `NotSupported` base default.
- Requires `max_open_files == -1` (immortal table readers) so same-file
("embedded") references stay resolvable lazily; otherwise the call returns
`InvalidArgument`.
## Implementation notes
- Read path: `GetContext` gains a lazy mode that leaves blob references
unresolved and captures the SST's `SameFileBlobReader`; `Version::Get` forwards
the new (defaulted) `lazy_columns_same_file_reader` out-param.
- `ReadPathBlobResolver` performs on-demand, cached blob resolution bound to the
pinned `Version` + decoded blob references.
- Internal identity is by heap-stable `Rep` pointers, so column/result/batch
pointers stay valid across moves with no move-time fix-ups.
## Scope / current limitations
- Resolves **whole columns**; byte ranges are served by slicing the whole column
for now (so `length`/`offset`/`force_verify` don't yet save I/O or skip
checksums). Single column family. Synchronous. `kBlockCacheTier` on the
originating call yields `Incomplete` on a blob-cache miss.
- User-defined timestamps: entities aren't writable under UDT, so `GetEntityLazy`
is at parity with `GetEntity` (same read-timestamp validation; plain values
return a single default column).
- Planned follow-ups: partial (byte-range) blob reads that skip checksum/cache-
fill, cross-key coalescing, and asynchronous execution.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15012
Test Plan:
- **`db/wide/db_lazy_entity_test.cc` (15 tests, all passing):** enumeration with
no blob I/O; single range and whole-column resolution; `MultiResolve` of many
blobs and mixed ranges/columns (incl. the `std::vector` overload); cross-key
batch resolution; result-outlives-call / pin survives move; unpulled columns
never read; `max_open_files == -1` requirement; `kBlockCacheTier` ⇒
`Incomplete`; foreign/null column ⇒ `InvalidArgument` (single-entity and
batch); UDT parity with `GetEntity`; `StackableDB` forwarding; and read-only
and secondary instances.
- **db_stress:** `GetEntityLazy` / `MultiGetEntityLazy` run as a differential
against eager `GetEntity` / `MultiGetEntity` (gated by `--lazy_entity_read_one_in`,
which only takes effect under `--open_files=-1`), including coverage on
read-only (`--open_read_only_one_in`) and secondary (`--test_secondary`)
instances. Smoke-run (`EXIT=0`, "Verification successful") with integrated blob
files + `--use_put_entity_one_in`.
- `make format-auto` clean; builds `ASSERT_STATUS_CHECKED`-clean.
Reviewed By: xingbowang
Differential Revision: D113792365
Pulled By: pdillinger
fbshipit-source-id: cb73fc845fe0e20ed352d697e552e7b47b2eb9cf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15054
`db_stress` often terminates on raw `assert(status.ok())` checks, which discard the underlying error and make crash-test failures difficult to diagnose.
Log the expression and full status before terminating. A message variant preserves operation-specific context and variadic formatting. In debug builds, assert after logging to retain assertion diagnostics and stack traces. In optimized builds, where assertions are disabled, exit with status 1.
Reviewed By: jaykorean
Differential Revision: D114756034
fbshipit-source-id: 5a0d4409bac69820b6560c8200cc401b7bfc65d6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15052
When `persist_user_defined_timestamps=false`, `full_history_ts_low` is automatically advanced by flushes to just above the newest timestamp in the flushed memtable. This creates the following race:
```
Reader Concurrent writer / flush
------ -------------------------
T0: Sample read timestamp:
892779959679761
T1: Write with a newer timestamp:
892779963289806
T2: Flush the memtable.
T3: Advance full_history_ts_low:
892779963289807
T4: Validate the read:
892779959679761
< 892779963289807
Result: InvalidArgument
```
This is expected RocksDB behavior because history below `full_history_ts_low` is no longer guaranteed to be complete. However, the stress test intends these operations to read the latest value.
This change adds ten seconds of headroom to read timestamps when user-defined timestamps are not persisted. This makes it substantially less likely that a concurrent flush overtakes a read timestamp. Unlike tolerating every `InvalidArgument`, this preserves reporting for genuine timestamp errors. Persisted read timestamps and write timestamps remain unchanged.
The triggering non-default options are `persist_user_defined_timestamps=false` and `user_timestamp_size=8`.
Reviewed By: xingbowang
Differential Revision: D114565978
fbshipit-source-id: 6939df9163fd97fefec2fa185bc2ca72231ab04f
Summary:
`TransactionLogIterator` iterates the write-ahead log. It does not return transactions, does not require `TransactionDB`, and in fact returns `Status::NotSupported()` for two of the three `TransactionDB` write policies (WritePrepared and WriteUnprepared). The name predates `TransactionDB` and comes from "transaction log" being an old synonym for WAL. Everything else in this header was already renamed away from that terminology -- `LogFile` -> `WalFile`, `VectorLogPtr` -> `VectorWalPtr` -- and the C API has long used `rocksdb_wal_iterator_t` and `rocksdb_wal_iter_*`. This finishes the job.
The public class is now `WalIterator` in `include/rocksdb/wal_iterator.h`. `include/rocksdb/transaction_log.h` remains as a forwarding header and `TransactionLogIterator` remains as an alias, both marked deprecated by comment only (not `[[deprecated]]`) so existing users compiling with -Werror are unaffected, matching how `LogFile` and `VectorLogPtr` were retired. `DB::GetUpdatesSince()` keeps its name.
More importantly, the documentation was incomplete in ways that could cost a user data, so this states plainly what the API has always done:
- It only sees the WAL. Writes made with `WriteOptions::disableWAL`, and sequence numbers consumed by `IngestExternalFile()`, advance the DB sequence number without producing a WAL record. They leave permanent holes that end iteration. This was not documented anywhere, so it was unreasonable to expect a user to know those features silently break WAL tailing.
- The three iterator states, in particular that `!Valid()` with an OK `status()` means caught up rather than finished, and that calling `Next()` again is the intended way to tail a DB. The old comment on `Next()` said "REQUIRES: Valid() to be true", which forbids the primary use case.
- `Status::TryAgain`, which a caller must handle but which appeared nowhere in the public headers.
- That `GetUpdatesSince(seq)` silently starts later when `seq` is unavailable. Combined with the above this is the sharp edge: recovering from a spent iterator means calling `GetUpdatesSince()` again, so the recovery step is exactly where data can be skipped unnoticed. A consumer applying these batches to a copy of the DB diverges permanently and undetectably if it misses one, so the header now says such consumers must verify continuity themselves using `BatchResult::sequence` and `WriteBatch::Count()`.
Almost all of this diff is mechanical: renaming the type and its files, updating `#include` paths, rewriting comments, and updating build lists (`src.mk`, `CMakeLists.txt`, `BUCK`, `java/Makefile`, `java/CMakeLists.txt`). `db/c.cc` is regenerated from `tools/c_api_gen/c_base.cc`. Reviewers should focus on the four places that are not:
1. `WalIteratorImpl::UpdateCurrentWriteBatch()` in `db/wal_iterator_impl.cc` (renamed from `db/transaction_log_impl.cc`) drops dead code. On a sequence discontinuity it set `current_status_` to `NotFound` and then called `SeekToStartSequence()`, which returns immediately when `current_status_` is not OK. That guard and the early return in `Next()` were both added by fe63899d1 ("Add checks to GetUpdatesSince", https://github.com/facebook/rocksdb/issues/9459), which made the reseek unreachable without mentioning it. The surviving behavior -- a discontinuity ends the run with `NotFound` -- is what the class has always documented, so this deletes the unreachable reseek and the comment claiming it can restore an OK status rather than resurrecting it. There is no observable change: `Valid()` is `started_ && is_valid_`, `is_valid_` is false either way, and a non-OK `current_status_` already makes `Next()` a no-op.
2. `WalIteratorImpl::IsBatchExpected()` in the same file drops "Log iterator will reseek the correct batch." from its info log message, since it no longer does.
3. `org.rocksdb.WalIterator` now holds the implementation and the nested `BatchResult`, and `org.rocksdb.TransactionLogIterator` becomes a subclass of it. Existing Java code keeps compiling, `RocksDB.getUpdatesSince()` keeps its return type, and `TransactionLogIterator.BatchResult` still resolves through inheritance. Because the native methods are now declared on `WalIterator`, the JNI entry points in `java/rocksjni/wal_iterator.cc` (renamed from `transaction_log.cc`) become `Java_org_rocksdb_WalIterator_*`.
4. `BatchResultJni::getJClass()` in `java/rocksjni/portal.h` looks up `org/rocksdb/WalIterator$BatchResult` to match.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15056
Test Plan:
Adds three tests to `db_log_iter_test.cc` that pin the semantics now being documented, none of which had coverage:
- `CaughtUpThenResumes` -- an iterator that has run out of records is `!Valid()` with an OK `status()`, and a later write to the same WAL is picked up by calling `Next()` again. This is the tailing pattern the old `Next()` comment prohibited.
- `GapFromDisableWAL` -- a `disableWAL` write between two normal writes ends the run with `NotFound` instead of skipping to the later sequence number, and the iterator stays spent across further `Next()` calls.
- `SilentlySkipsUnavailableStart` -- `GetUpdatesSince(1)` where sequence 1 never reached the WAL returns OK and positions at sequence 2, documenting the silent skip.
`IngestExternalFile()` is documented as the other source of holes but is not covered by a test; it needs `SstFileWriter` setup and is worth a follow-up.
The Java rename is verified by `WalIteratorTest`, which declares `WalIterator` while `RocksDB.getUpdatesSince()` still returns `TransactionLogIterator`, so the compatibility subclass is exercised on every use. It is otherwise an identifier-only rename of the previous test.
Renames the C++ test fixture to `DBWalIteratorTest`, drops the now-redundant `TransactionLogIterator` prefix from its test names, and updates the comments in `wal_manager.cc`, `sync_point.h` and `fs_posix.cc` that refer to those tests by name. Those comments already pointed at the wrong fixture and file (`DBTest` in `db_test.cc`) before this change.
Reviewed By: xingbowang
Differential Revision: D114809792
Pulled By: pdillinger
fbshipit-source-id: d3cfd23867b4738b62a096db1dbc895bada069b7
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14982
**What**: Add reference-counted `DB::AbortCompactions(cf)` / `DB::ResumeCompactions(cf)` to abort and resume compactions for a single column family without stopping unrelated column families. They compose with the DB-wide `AbortAllCompactions()` / `ResumeAllCompactions()` and are no-ops once DB shutdown has begun.
**How**:
- **Per-CF abort flag.** `ColumnFamilyData` gains `compaction_aborted_`, wrapped in `CacheAlignedWrapper` so the lock-free reads done by compaction workers do not false-share with the DB-mutex-written fields packed around it. A new `IsCompactionAborted(cfd)` helper checks it alongside the DB-wide counter at every entry point (`CompactRange()`, `RunManualCompaction()`, `CompactFilesImpl()`), and `CompactionJob::ProcessKeyValue()` checks both on entry and in its periodic cron check.
- **Parking.** Automatic work for an aborted CF is preserved rather than dropped. `PickCompactionFromQueue()` now returns `std::variant<CompactionQueuePicked, CompactionQueueThrottled, CompactionQueueParked>` and moves an aborted CF into `parked_compaction_cfds_` only once it reaches the front of the queue, so CFs behind it keep their place and the scheduler credit is not restored (which would cause a hot re-pick loop). `EnqueuePendingCompaction()` parks newly needed work for an aborted CF and now early-returns for dropped CFs. `RestoreParkedCompaction()` re-enqueues on the final resume; a CF dropped while parked is handled in `DropColumnFamily()`, and one dropped before its aborted queue entry is reached is handled by a fast path in `PickCompactionFromQueue()`.
- **Knowing when it is safe to return.** `ColumnFamilyData::num_in_flight_compactions_` counts actually-picked work, and `Defer` balances `BeginInFlightCompaction()` / `EndInFlightCompaction()` on every `BackgroundCompaction()` and `CompactFiles()` exit path. That covers the `NotifyOnCompactionCompleted()` window that `CompactionPicker::compactions_in_progress_` does not, since the picker set is cleared during `Install`'s manifest write. `MustWaitForCompaction()` combines the counter with the picker set (filtered to skip `kExternalSstIngestion` / `kRefitLevel` reservations, which are not real jobs) because only the picker set covers the LOW-to-BOTTOM priority handoff, where the counter briefly dips to 0.
- **Waking the waiter.** `AbortCompactions()` cancels matching `manual_compaction_dequeue_` entries and then waits on `bg_cv_`; `per_cf_compaction_abort_waiters_` forces the compaction completion paths to signal, because the picker's `UnregisterCompaction()` does not. `CloseHelper()` drains `parked_compaction_cfds_` and releases their references.
- **API surface.** The two methods are new pure virtuals on `DB`, so `StackableDB` forwards them and the `db_test.cc` DB wrapper stubs them. `db.h` documents how the per-CF API composes with DB-wide abort, `DisableManualCompaction()`, `PauseBackgroundWork()`, shutdown, ingestion reservations, level refit and remote compaction, plus the parked-work caveat on `WaitForCompact()`. Comments in `listener.h`, `statistics.h` and `status.h` now name both abort APIs.
- **Coverage.** 10 new cases in `db_compaction_abort_test.cc` (CF scoping, counter composition, parked-work restore, CF drop both while parked and before parking, trivial-move rejection, waiting for the completion callback, post-close no-op). `db_stress` gains `abort_and_resume_cf_compactions_one_in` and `TestAbortAndResumeCfCompactions()`, rejected in combination with `clear_column_family_one_in` because that would drop the column family out from under an active abort. Also a `db_crashtest.py` knob and one `new_features` release note.
- **Known gap.** A `TODO` records the pre-existing post-pick stale compaction-score problem: aborting after `PickCompaction()` releases the inputs without recomputing the score, so the resume can observe a stale sub-trigger score and re-enqueue nothing. `ResumeAllCompactions()` has the same gap and carries a pointer to the same TODO; fixing it is left to a separate change covering both APIs.
**Why**: Atomic-replace bulk loads only conflict with their target column family, but the DB-wide abort suppresses compaction for every column family colocated in the same RocksDB instance. Scoping the abort keeps unrelated column families compacting for the duration of a bulk load.
Reviewed By: pdillinger
Differential Revision: D113103899
fbshipit-source-id: 9bef924132676ed118a37aee86da6ee8bcefef09
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15046
`BlobFetcher` is the policy boundary for single-blob reads, but
`Version::Get()` and primary direct-write postprocessing still called
lower-level backends directly. This duplicated backend selection and could
bypass future fetcher policy or decorators.
Route those paths through their existing `VersionBlobFetcher` and remove the
redundant Slice-based `Version::GetBlob()` overload. This is behavior-preserving:
each fetcher delegates to the same Version or direct-write backend with the same
`ReadOptions`, cache, and lifetime context.
Reviewed By: xingbowang
Differential Revision: D114616978
fbshipit-source-id: 0d94cc389138f27791de636e2cca585e51cb142a
Summary:
Adds db_stress / crash-test coverage for the parallel `CheckpointEngine` so it
is exercised alongside existing features (atomic flush, sync fault injection,
cf_consistency, backups).
- Opens one shared `CheckpointEngine` per db_stress process, configured by
`--checkpoint_engine_max_background_operations` and
`--checkpoint_engine_use_link_file_when_available`.
- `--parallel_checkpoint_one_in=N`: each checkpoint independently uses the
parallel engine with probability 1/N, otherwise the legacy serial
`Checkpoint` API, so serial and parallel paths run concurrently across threads.
- `db_crashtest.py` randomizes all three per run.
This coverage surfaced the missing destination fsync in the parallel copy path
fixed by the parent diff.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15037
Reviewed By: xingbowang
Differential Revision: D114414578
Pulled By: rban1
fbshipit-source-id: 48e7920a62b9b663a1c6054fc60b5765667bcbe5
Summary:
Parallel checkpoint copies were never fsynced. `CopyEngine::CopyOrCreateFile`
gated the destination sync on `use_fsync` (wired to `db_options.use_fsync`) --
but `use_fsync` means "fsync vs fdatasync", not "whether to sync". With the
default `use_fsync=false`, the parallel path skipped the sync, so a copied
MANIFEST/SST stayed in the unsynced buffer and was dropped by
`TestFSWritableFile::Close` under `sync_fault_injection`, leaving a 0-byte file
and checkpoint corruption:
Corruption: no log_file_number, next_file_number, last_sequence entry in MANIFEST
Corruption: Sst file size mismatch between expected N and file system 0
Serial `CopyFile` always syncs (using `use_fsync` only to pick the flavor), so
it never hit this.
Fix: split `WorkItem` / `CopyOrCreateFile` into `sync` (whether to sync) and
`use_fsync` (fsync vs fdatasync). CheckpointEngine now passes `sync=true,
use_fsync=db_options.use_fsync`, matching serial. BackupEngine is unchanged --
`sync=options_.sync` with the new `use_fsync=false` default reproduces the prior
behavior.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15038
Reviewed By: xingbowang
Differential Revision: D114414620
Pulled By: rban1
fbshipit-source-id: e0f1f3dbadcd48c5a5c2a11111a266dbed623f86
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15034
`db_stress` startup validation had accumulated multiple error-reporting paths, including direct `fprintf`/`exit(1)` pairs. This made invalid configurations inconsistent and bypassed normal unwinding.
Centralize validation failures through `ReturnValidationError`, including diagnostics with runtime values, while leaving operational I/O failures on their existing paths. The variadic stream interface keeps the existing messages without `printf` casts.
Reviewed By: jaykorean
Differential Revision: D114384751
fbshipit-source-id: f396f8e7d099ef87c72aaf19455b777f2125f4df
Summary:
The C API compatibility job always fetched `main` as its baseline. On release branches, that can report false API removals for symbols that were added to `main` after the release branch was created.
Select the compatibility baseline from the triggering event instead:
- Pull requests use the merge target commit.
- Pushes use the branch's pre-push commit.
- Newly created branches or tags use `HEAD`, since they have no prior ref state to preserve.
Fetch the selected commit by SHA, fail explicitly for unsupported events or missing event data, and pass the resolved baseline to `check-c-api-gen`. Also update the Makefile documentation to describe the PR and push behavior.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15039
Test Plan: CI
Reviewed By: pdillinger
Differential Revision: D114437175
Pulled By: xingbowang
fbshipit-source-id: ee66f54bcabba238477d07a69ec7dcb42219c088
Summary:
- add `rocksdb_batched_multi_get_pinned_cf` for batched reads using `rocksdb_slice_t` keys
- return one `rocksdb_pinnable_multi_get_t` owner for all values and per-key errors
- preserve input ordering, duplicates, misses, and sorted-input behavior
- expose indexed borrowed results plus count and destroy functions
## Why
`rocksdb_batched_multi_get_cf_slice` returns one separately allocated `rocksdb_pinnableslice_t` wrapper for every found key. Language bindings that want to retain a full batch have to manage and destroy each wrapper independently.
The new API keeps successful `PinnableSlice` values in one owner. It stores only compact value and error payloads, while a result index maps them back to the original key order. Found values and error messages remain borrowed until the batch is destroyed.
The batch must be destroyed before closing the DB that created it because pinned values can retain DB-owned cache state.
## Compatibility
This is additive. Existing MultiGet APIs and ownership rules are unchanged.
## Testing
`c_test` covers:
- found and missing keys
- duplicate keys and input ordering
- empty values
- per-key merge errors
- sorted and unsorted input
- empty batches
- out-of-bounds access
- borrowed value and error lifetimes
Validation run locally:
- `make -j8 check-c-api-c_test`
- `CXX=clang++ make check-c-api-gen CHECK_C_API_GEN_STRICT=1 CLANG_FORMAT_BINARY=clang-format`
- C API link-completeness and backward-compatibility checks
This API is intended to replace a temporary compatibility extension in `rust-rocksdb` once it is available in a RocksDB release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14988
Reviewed By: jaykorean
Differential Revision: D114357453
Pulled By: xingbowang
fbshipit-source-id: a35785f7bb4d097b391932e2b9f96fc177d1ba1c
Summary:
CI/build downloaded pinned third-party binaries (compression source tarballs, Java test jars, and clang-format-diff.py) from an S3 bucket. Move the mirror to assets on an unlisted "third-party-deps" GitHub release, so the files live alongside the code with no separate bucket to maintain and no per-repo bandwidth quota (unlike Git LFS). Recent releases on the RocksDB release page are not polluted (https://github.com/facebook/rocksdb/releases/, far down the list probably from tagging a very old release).
The compression download variables are refactored from *_DOWNLOAD_BASE (a directory URL) to *_DOWNLOAD_PREFIX (directory + "<project>-", or "...archive/v"), so each release asset can carry a project-name prefix in the otherwise flat release namespace while the upstream defaults still resolve to byte-identical URLs. The Java DEPS_URL defaults (Makefile and CMake) and the clang-format-diff.py download are repointed at the release. A comment at each dependency site links to the release edit page for adding/updating assets.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15041
Test Plan:
CI and...
Downloaded all 11 assets from the release and confirmed their SHA256 sums match the files previously served from S3 (tarball/jar sums are also enforced by the existing Makefile checksum checks). Verified the refactored Makefile upstream defaults expand to the same URLs as before, and that the CI *_DOWNLOAD_PREFIX overrides plus DEPS_URL resolve to the uploaded asset names. Covered by the normal PR/nightly/weekly jobs that build compression from source and run the Java tests (Make and CMake).
Reviewed By: xingbowang
Differential Revision: D114443886
Pulled By: pdillinger
fbshipit-source-id: 8c59435e7064f0d9931cba103c8120fe113b0d6d
Summary:
ReadOptions is copied frequently on read paths, and its only non-trivially-copyable member, table_filter, is a std::function that is almost always empty yet still gets deep-copied on every ReadOptions copy (and can heap-allocate). This changes table_filter to a pointer to a caller-owned std::function instead of storing one by value: when non-nullptr, the caller is responsible for keeping the pointed-to function alive for the duration of the read (like the other pointer fields such as iterate_upper_bound). With this change every ReadOptions member is trivially copyable, so copying ReadOptions no longer touches a std::function at all. That trivial-copyability is locked in with a static_assert in options_test.cc (kept out of the public header).
This is inspired by D113109635, which targeted the same ReadOptions-copy hotspot by relocating table_filter to the end of the struct so the trivial members memcpy in one contiguous block. This is intended to be a superior solution: rather than only defragmenting the copy, it removes the std::function copy entirely.
This is a breaking API change, tolerated because table_filter is a relatively obscure option. A nullptr, or a pointer to an empty std::function, means no filtering (preserving the previous empty-function behavior). A shared internal helper HasTableFilter(const ReadOptions&) in options_helper.h centralizes the nullptr + empty-function check used on the read paths. The C API and Java bindings now own the std::function and hand ReadOptions a pointer to it; their public signatures are unchanged, and the C API rocksdb_readoptions_t clears the owned function when the callback is null.
Making ReadOptions trivially copyable also let
-Werror=unused-but-set-variable fire on a pre-existing dead local in db_bench's MultiScan: a ReadOptions that set async_io/readahead_size but was never passed to NewMultiScan (the warning had been suppressed while ReadOptions had a non-trivial member). That local was in fact inert for multiscan -- BlockBasedTableIterator::Prepare overrides ReadOptions::async_io with MultiScanArgs::use_async_io for the coalesced prefetch, so multiscan async I/O is driven by --multiscan_use_async_io, not ReadOptions::async_io. Removed the dead local; MultiScan now uses read_options_ directly, consistent with MultiScanRandom.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15005
Test Plan:
Existing table_filter coverage was adapted to the new pointer API and still exercises the behavior end to end: db_iterator_test.TableFilter (callback invoked once per table; returning false hides a table) and its range-conversion rejection case; db_bloom_filter_test SstQueryFilter/FixedWidthSegments range queries; and trie_index_db_test's range-query iterator (which returns iterators outliving the filter, exercising the caller-owned lifetime). db/c_test covers the C API set/clear/has_table_filter path. options_test asserts std::is_trivially_copyable_v<ReadOptions>.
Manually verified the db_bench MultiScan async behavior with a ~200MB DB and --use_direct_reads: with ReadOptions::async_io held true, async reads tracked only --multiscan_use_async_io (rocksdb.read.async.micros count 329511 with use_async_io=true vs 0 with false), confirming ReadOptions::async_io does not drive multiscan async and the removed local had no effect.
Reviewed By: anand1976
Differential Revision: D113613525
Pulled By: pdillinger
fbshipit-source-id: 38d73b607500a5c7ca2b89dd8f4187ef43940aa4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15019
RocksDB's Checkpoint API currently copies all column families.
For use cases that only need a subset, this is wasteful.
This diff adds a new `CreateCheckpoint(dir, column_families, ...)` overload that:
- Validates input handles (rejects null/foreign handles, coalesces duplicates, always includes default CF).
- Filters GetLiveFilesStorageInfo to capture only the requested CFs' table/blob files.
- Post-processes the copied MANIFEST by appending kColumnFamilyDrop records for excluded CFs, so the checkpoint opens without referencing missing files.
- An empty column_families vector falls through to the existing whole-DB path.
Reviewed By: xingbowang, pdillinger
Differential Revision: D113963616
fbshipit-source-id: d1becbbb9528dd13c2dc314165880b76123757f0
Summary:
`make check` ran a series of extra checks serially, one after another, after the GNU-parallel unit tests: the Python syntax check, ldb/db_crashtest/ rocksdb_dump tests, and the format/buck/sources/workflow/c-api-gen source checks. These had accumulated into a non-trivial single-threaded tail on every `make check`.
Each extra check is now a make target collected in a single CHECK_EXTRA_TARGETS list, and gen_check_scripts emits a `t/run-<target>` wrapper per enabled check. check_0 already runs every `t/run-*` as a gnu_parallel job, so the checks now run alongside the unit tests, hiding their cost behind the (dominant) test phase. The slow ones (check-c-api-gen, check-format) are front-loaded via slow_test_regexp, and under CI sharding the checks are distributed round-robin so each runs on exactly one shard. The J=1 / no-gnu-parallel fallback runs them serially. New check-all-python and rocksdb_dump_test targets wrap the two steps that were previously inlined; the guards match the old serial tail exactly.
Each wrapper frames its result with a gtest-style
`[ RUN ]` / `[ OK ]` / `[ FAILED ] >>> make <check> <<<` line so a non-unit-test failure is easy to distinguish and re-run.
The wrappers spawn sub-makes concurrently, and each make invocation otherwise re-runs build_detect_platform, which rewrites the shared make_config.mk non-atomically -- concurrently that corrupts it ("missing separator"). The wrappers pass a new ROCKSDB_RECURSIVE_MAKE_SAME_CONFIG=1, which reuses the parent's already-generated make_config.mk (guarded on it existing, so normal builds still detect) and also silences the parse-time banners for clean output.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15028
Test Plan: `make check` runs the same set of auxiliary checks as before, now as parallel jobs. Verified that gen_check_scripts emits one wrapper per enabled check, that the SKIP_FORMAT_BUCK_CHECKS / ASSERT_STATUS_CHECKED / AIX guards still gate the set, that CI_TOTAL_SHARDS distributes the checks one-per-shard (union across shards is the full set), and that many concurrent wrappers no longer corrupt make_config.mk. A seeded check failure surfaces framed as `[ FAILED ] >>> make <check> <<<`.
Reviewed By: xingbowang
Differential Revision: D114245857
Pulled By: pdillinger
fbshipit-source-id: 519f78ccdcfbfdd88509980f78944e527a572331
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15027
`BackgroundPurgeCFDropTest` blocks `DBImpl::BGWorkPurge:start` to verify dropped-CF files remain until background purge is released. Runtime OPTIONS-file cleanup can also schedule a purge on the HIGH pool when `avoid_unnecessary_blocking_io` is enabled, so installing the sync-point dependency before test setup can park that OPTIONS purge before the foreground `Flush()` has a chance to run.
This does not need a broader product change. Runtime OPTIONS-file writes are rare, and the hard deadlock is a test artifact caused by the sync-point dependency. Outside the test, the observable impact would be limited to a foreground `Flush()` API call waiting longer if its flush is queued behind the OPTIONS-file purge.
Scope the sync-point dependency to the behavior under test by installing it only for the background-purge case, after the setup flush has produced the SST and just before `DropColumnFamily()`. This keeps setup-side OPTIONS-file purge work outside the blocked window while still verifying that dropped-CF file deletion waits for background purge.
Reviewed By: joshkang97
Differential Revision: D114241457
fbshipit-source-id: 21c5f4242a50590e161b3720077a9dfe5e77aa08
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15020
Each read executor worker creates its own `IoUringBackend`, and Folly's I/O thread pool creates its worker threads and EventBases lazily. A successful calling-thread probe or `DB::Open()` therefore does not validate every worker. A later read can start a new worker; previously, an `IoUringBackend` initialization exception on that worker could escape startup and terminate the process. The previous 512-submit setup also allowed Folly's capacity retry to reduce the completion queue to 512 while retaining a 1024-entry submission queue, which the kernel rejects with `EINVAL`.
Use a 256-submit limit so the 512-entry retry remains valid. Configure the shared read executor during `DB::Open()` through `DBOptions::read_io_executor_threads`, defaulting to one thread. The backend factory catches per-worker io_uring initialization failures and uses Folly's default backend. Each worker's thread-local availability marker starts false and is set true only after `IoUringBackend` construction succeeds. When construction fails, the marker remains false and `SubmitReadAsync()` performs the read synchronously. The executor's maximum thread capacity can be raised or lowered; `db_stress` requests eight threads for async DB APIs.
To expose this degradation, `FSRandomAccessFile::SubmitReadAsync()` now reports whether the asynchronous path was used. `RandomAccessFileReader` increments the `FILE_SUBMIT_ASYNC_READ_FALLBACK` ticker for every submission that falls back to synchronous I/O, making the fallback visible through the DB's `Statistics`.
Reviewed By: xingbowang
Differential Revision: D114031018
fbshipit-source-id: 1b7f6c1cea693e5dd743ea1ada0fca1cf757b3f9
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15021
The warm storage crash failure in T282626155 exposed a RocksDB bookkeeping bug, not a transient WS error. The regression was introduced by D91480994 / upstream RocksDB PR https://github.com/facebook/rocksdb/pull/14227 (“Support abort background compaction jobs”). That change routed `BlobFileBuilder` output tracking through `CompactionOutputs::output_file_paths_`, which contains both blob and SST paths, while blob completion still assumed `.back()` identified the current blob.
Failure timeline:
1. Compaction job 225 opens `000455.blob`, producing `[000455.blob]`.
2. It opens `000456.sst`, producing `[000455.blob, 000456.sst]`.
3. Blob 455 finishes, but `OnBlobFileCompleted()` receives `.back()`, which is `000456.sst`.
4. `SstFileManager` stats the still-open SST. WS correctly returns non-retryable `NOT_FOUND` because the output is not pathname-visible until close.
5. Compaction fails before installing its outputs in the MANIFEST. Cleanup then deletes `000455.blob` and `000456.sst`; those deletions are consequences of the failure, not its cause.
This change derives callback paths from the blob file number instead of the shared cleanup vector, renames the vector accordingly, and adds a deterministic regression test verifying that `SstFileManager` tracks every blob created during compaction.
Reviewed By: pdillinger
Differential Revision: D114002145
fbshipit-source-id: 472b563f3fb731753f286a40feece7428928bc20
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15004
This change reduces DB open latency on remote storage by avoiding foreground deletion of obsolete `OPTIONS-*` files during open when `avoid_unnecessary_blocking_io` is enabled. DB open now avoids foreground deletion of obsolete `OPTIONS-*` files on the pre-open `WriteOptionsFile()` path. `RenameTempFileToOptionsFile()` records the new OPTIONS file and chooses an explicit cleanup mode: skip while open is still in progress, schedule deletion when background work is available, or delete immediately when blocking IO is allowed or background jobs are rejected.
`DBImpl::Open()` makes the deferred cleanup decision once final open status is known. A successful open schedules obsolete OPTIONS files on the purge queue, while a failed open deletes them synchronously before tearing down the partially opened DB. The default synchronous behavior remains unchanged when `avoid_unnecessary_blocking_io` is disabled, and the latest-two OPTIONS retention policy is preserved.
Reviewed By: pdillinger
Differential Revision: D113593680
fbshipit-source-id: 8aee99b3f38a729647d0bcb644d212181fe86895
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15017
## Summary
RocksDB async `GetAsync` and `MultiGetAsync` callbacks now default to a completion-only path. Callbacks can override `EnableStats()` when they need request-specific `PerfContext` or `IOStatsContext`; otherwise RocksDB avoids creating a coroutine stats context and returns null stats pointers. Callbacks must not initiate another async read from `OnComplete()`.
RocksDB stats normally use thread-local storage, which does not compose with interleaved coroutine reads. A stats-enabled request captures only the caller's configuration: perf level, per-level perf enablement, and whether io stats are disabled. It then starts with empty request contexts. `onSet()` loads that request's contexts into TLS, while `onUnset()` saves the accumulated contexts back into request data. On resume, the same contexts are loaded again. This isolates request counters without preserving or restoring pre-existing executor TLS state. The synchronous fallback similarly resets TLS before executing a stats-enabled read.
`PerfContext` and `IOStatsContext` gain `Merge()` support for caller-side aggregation. `PerfContext` move construction and assignment now transfer ownership of per-level state rather than deep-copying it, avoiding allocations during coroutine context handoffs. A fbcode-wide audit found no caller that reads a moved-from `PerfContext`.
`db_bench` opts in only for coroutine jobs that print per-job perf context and merges every completed request's callback stats into the job aggregate.
## Example: Two Async Gets on One Thread
Both callbacks enable stats. Each request starts with empty counters and carries its contexts across suspensions.
```text
Shared executor thread
|
| onSet(A): load A's empty contexts into TLS
| A records 3 reads
| onUnset(A): save TLS contexts back into A
|
| onSet(B): load B's empty contexts into TLS
| B records 4 reads
| onUnset(B): save TLS contexts back into B
|
| onSet(A): load A's contexts containing 3 reads
| A records 2 more reads
| callback A receives block_read_count = 5
|
| onSet(B): load B's contexts containing 4 reads
| callback B receives block_read_count = 4
```
Isolation comes from replacing TLS with the active request's saved contexts on every `onSet()`.
### Timers
Existing wall-clock timers remain in coroutine frames, so timers spanning `co_await` include suspended time and measure end-to-end latency.
Request-level `get_cpu_nanos` is suspension-aware: it starts in `onSet()` and stops in `onUnset()` or `Finalize()`, charging only time while the request is active.
Nested CPU timers remain disabled because normal RAII CPU timers would also charge executor work performed while their coroutine is suspended. `MultiGetAsync` fanout additionally requires per-child timer state to avoid double counting.
## Benchmarks
Benchmarked to confirm no performance regressions when perf is disabled, and also to see the overhead of enabling perf.
```bash
db_bench --benchmarks=<readrandomcoroutine|multireadrandomcoroutine> --db=<db_path> --coro_jobs_per_thread=10 --use_existing_db=true --num=2000000 --value_size=1024 --compression_type=none --threads=8 --batch_size=16 --multiread_batched=true --use_direct_reads=true --cache_size=1428076626 --reads=320000 --perf_level=<1|3> --seed=<seed>
```
| Workload | Before change | Perf level 1 | Before -> 1 | Perf level 3 | 1 -> 3 |
| -------- | ------------- | ------------ | ----------- | ------------ | ------ |
| Async Get | 475.6K ops/s | 476.7K ops/s | +0.2% | 470.5K ops/s | -1.3% |
| Async MultiGet | 567.3K ops/s | 567.4K ops/s | +0.0% | 552.2K ops/s | -2.7% |
Reviewed By: xingbowang
Differential Revision: D112634874
fbshipit-source-id: 94e240aeb1496fa12dac995b5aa5a9c3543cc5fc
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14997
`CheckpointEngine`'s shared `CopyEngine` recorded copied bytes under the
`BACKUP_READ_BYTES` / `BACKUP_WRITE_BYTES` tickers (the `CopyEngineOptions`
defaults), conflating checkpoint I/O with backup I/O in statistics. This adds
dedicated `CHECKPOINT_READ_BYTES` / `CHECKPOINT_WRITE_BYTES` tickers and points
`CheckpointEngineImpl`'s `CopyEngineOptions` at them, so checkpoint copies report
under their own metric names. `BackupEngine` is unchanged (still uses the
backup-ticker defaults).
Addresses review feedback that the shared `CopyEngine` should not hardcode
backup metric names for non-backup callers.
The new tickers are intentionally not mirrored in the Java bindings
(`TickerType.java` / `portal.h`): the signed-`jbyte` id space there is exhausted,
so they remain C++-only, documented inline in `statistics.h`.
Reviewed By: xingbowang
Differential Revision: D113469742
fbshipit-source-id: 2a6a7ef04c939a1240f1c48c7bcaba8b2e7a1c10
Summary:
- Gate the `MultiScanPrepare` test's async I/O mode on the filesystem's reported `kAsyncIO` support.
- Use the same gated async-I/O flag for `MultiScanArgs::use_async_io` and read-count expectations.
- Match production DB-level MultiScan behavior so the test falls back consistently after io_uring initialization fails.
## Background
- Recent Sandcastle failures in T265255062 all hit the MultiScanPrepare read-count assertion after PosixFileSystem::SupportedOps() logged "CreateIOUring failed: Cannot allocate memory (errno=12)" and returned without kAsyncIO support.
- The test already probed SupportedOps(), but it only used that result when computing expected reads. It still copied the raw use_async_io test parameter into MultiScanArgs. That let the lower-level table-reader path call ReadAsync() and retry CreateIOUring() after the support probe had failed. If the retry succeeded, Prepare queued async reads and recorded zero sync reads while the assertion expected the sync fallback count.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15016
Test Plan:
- `make format-auto`
- `AUTO_CLEAN=1 make -j$(nproc) block_based_table_reader_test`
- `timeout 60s ./block_based_table_reader_test --gtest_filter="*MultiScanPrepare*"`
- `make check-sources`
Reviewed By: archang19
Differential Revision: D113830233
Pulled By: xingbowang
fbshipit-source-id: 38460f24fd44b33eb139a8f59bd87c637983e503
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15014
Implements coroutine-backed execution behind the callback-based `DB::GetAsync` and `DB::MultiGetAsync` APIs that issue SST block reads asynchronously, so a single application thread can keep many point-lookup reads in flight at once instead of paying one blocking read per thread. The motivation is point-lookup throughput on disk-bound workloads (working set larger than the block cache), where the win comes from queue depth rather than per-read latency.
The async capability rests on a small, generic FileSystem API that a `FileSystem` must implement to participate. Every method is virtual with a working synchronous default, so existing and custom `FileSystem` implementations compile and run unchanged (they simply stay synchronous):
- `FSRandomAccessFile::SubmitReadAsync(req, opts, cb, dbg)` -- a fire-and-forget read: it starts the read and returns without blocking, then invokes `cb` once `req.result`/`req.status` are filled in (the callback may run on a different thread). The default implementation is synchronous -- a blocking `Read` followed immediately by `cb`.
- `FileSystem::GetReadExecutor()` -- returns the `folly::Executor*` that async queries and read completions run on. The default returns `nullptr`, which the query layer treats as "async reads unavailable" and falls back to the synchronous path. It is exposed to users, so they may add additional application logic to run on the same thread pool, thus avoiding expensive context switches.
- `FileSystem::SetReadIOExecutorThreads(int)` -- sizes that executor. The default is a no-op.
`RandomAccessFileReader::Read`/`MultiRead` are `DEFINE_SYNC_AND_ASYNC` coroutines (new `file/random_access_file_reader_sync_and_async.h`), and `DBImpl::GetAsync`/`MultiGetAsync` bind the query to `GetReadExecutor()` with `co_withExecutor`. For a MultiRead, `SubmitMultiReadAsync` issues all of the block reads through `SubmitReadAsync` back-to-back and then `co_await`s a `folly::coro::Baton`: awaiting the baton suspends the query and releases the worker thread while the reads are in flight, and the callback for the last outstanding read posts the baton to resume it. A MultiGet's per-file `MultiGetFromSST` reads additionally fan out across the current eventbase so overlapping-level lookups run in parallel. This layer works against any executor and any `SubmitReadAsync` implementation.
The Posix `FileSystem` is the concrete async implementation, and is where io_uring enters. `GetReadExecutor()` is a `folly::IOThreadPoolExecutor` whose workers are EventBases that each drive an io_uring ring (sized by `SetReadIOExecutorThreads`). `PosixRandomAccessFile::SubmitReadAsync` -- which must be invoked on one of those EventBase workers -- places the read on that worker's submission queue via `IoUringBackend::queueRead` and returns immediately: no blocking `pread`, no parked thread. When the kernel posts the completion the backend runs the callback on that same worker, and a short read re-queues its remainder from inside the completion callback, so even multi-syscall reads stay on the non-blocking path. The net effect is that one worker thread keeps many reads outstanding at once, bounded by io_uring queue depth rather than thread count.
## Thread-local state handling
A coroutine suspends at every `co_await` and can resume on a different read-executor worker, and another coroutine can run on that worker in between. Any thread-local state held across a suspension point is therefore unsafe -- it can be observed or mutated by an interleaved coroutine, restored onto the wrong thread, or trip an in-use assertion. The read path handles each thread-local it touches explicitly, and introduces no new thread-local storage of its own:
- SuperVersion cache: the thread-local SuperVersion slot is marked `kSVInUse` for the duration of a read, so a second coroutine resuming on the same worker would trip that assertion. The coroutine path instead takes an independent heap reference that releases the slot before any suspension -- `cfd->GetReferencedSuperVersion()` for single Get, and `kExtraSvRef = true` threaded through `MultiCFSnapshot` for MultiGet -- and releases it with `CleanupSuperVersion` (via a `Defer` for single Get). The synchronous path is unchanged and keeps using the thread-local slot (`GetAndRefSuperVersion` / `ReturnAndCleanupSuperVersion`).
- perf_context / iostats_context / perf level: the timing guards that save-and-restore or accumulate into these thread-locals across an IO -- `GetPerfLevel`/`SetPerfLevel`, `IOSTATS_TIMER_GUARD(read_nanos)`, `IOSTATS_CPU_TIMER_GUARD(cpu_read_nanos)` -- are compiled out on the coroutine path (`#if defined(WITHOUT_COROUTINES)`), so none of that thread-local timing state spans a `co_await`. The tradeoff is that these perf/IO counters are not populated for coroutine reads (a known limitation of this path).
- TLS-free helpers are left as-is: `StopWatch`, for instance, holds only clock/statistics pointers and a start time, so it is safe across a suspension (it can over-count wall time under oversubscription, which is a metrics-accuracy issue, not corruption).
## Benchmarks
Benchmarked using 8 threads for both multiget (with/without async_io enabled), and single get on both direct-io enabled and disabled on a ~11 GB DB. Non direct-io benchmarks are not a good representation because my dev machine is capable of paging the entire DB into memory. But, it is still useful to compare against async_io=true, since that feature only benefits non direct-io.
Jobs-per-thread (jpt) is the multiplier for the number of tasks being submitted to the thread pool. E.g. a threadpool of 8 threads with jpt = 4 has 32 running tasks on the thread pool. It is important to oversubscribe so that when a task suspends, another one can resume on the thread pool.
MultiGet (batch 16):
| IO | cache hit % | sync | async_io | coro jpt=1 | coro jpt=2 | coro jpt=5 | coro jpt=10 |
| -------- | --- | ---- | -------- | ---------- | ---------- | ---------- | ----------- |
| direct | 10% | 65K | 64K | 145K | 151K | 161K | 166K |
| direct | 50% | 88K | 80K | 239K | 287K | 327K | 340K |
| direct | 90% | 435K | 399K | 732K | 991K | 1,093K | 1,119K |
| buffered | 10% | 449K | 561K | 555K | 593K | 572K | 566K |
| buffered | 50% | 495K | 682K | 656K | 724K | 700K | 701K |
| buffered | 90% | 600K | 878K | 828K | 1,021K | 1,048K | 1,060K |
single Get:
| IO | cache hit % | sync | coro jpt=1 | coro jpt=2 | coro jpt=5 | coro jpt=10 |
| -------- | --- | ---- | ---------- | ---------- | ---------- | ----------- |
| direct | 10% | 41K | 37K | 69K | 125K | 144K |
| direct | 50% | 74K | 66K | 127K | 238K | 281K |
| direct | 90% | 390K | 330K | 540K | 767K | 849K |
| buffered | 10% | 403K | 291K | 369K | 447K | 447K |
| buffered | 50% | 441K | 337K | 443K | 554K | 527K |
| buffered | 90% | 544K | 433K | 606K | 753K | 810K |
Example command (MultiGet coroutine, ~50% hit):
```
db_bench -benchmarks=multireadrandomcoroutine -coro_jobs_per_thread=10 -use_existing_db=true -num=2000000 -value_size=1024 -compression_type=none -threads=8 -batch_size=16 -use_direct_reads=true -cache_size=402653184 -reads=320000 -statistics
```
### Coroutine vs context switch overhead
Coroutines are not zero overhead, so we also compare it against the context switching of using many sync threads. Both variants were pinned to the same 8 CPUs and ran 80 concurrent MultiGet operations.
| variant | throughput |
| ---------------------------------------- | ------------ |
| sync, 80 threads | 330.6K ops/s |
| coroutine, 8 executor threads / 80 jobs | 446.0K ops/s |
Reviewed By: xingbowang
Differential Revision: D111116635
fbshipit-source-id: bd0ac052efe79261a552e9e66137e5e70f7cd8a1
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15013
Add `--use_async_db_api` to db_stress and route direct `DB::Get` and value `DB::MultiGet` reads through the callback async APIs (`DB::GetAsync` / `DB::MultiGetAsync`) when coroutine support is available. The db_stress helpers block on callback completion and notify waiters while holding the callback lock, so existing stress-test control flow and callback lifetime are preserved. Builds without coroutine support silently keep the existing synchronous path.
Centralize the dispatch in db_stress helpers and use them for workload, verification, secondary, and consistency/debug reads while leaving transaction and entity APIs on their existing paths.
Randomize `use_async_db_api` in `tools/db_crashtest.py` one run in six.
Reviewed By: jaykorean
Differential Revision: D111132477
fbshipit-source-id: 2aeac267867d8501131161c9fdbaf1da2b0c51d5
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15011
Root cause: `open_read_only_one_in` can create a read-only DB instance on the same live directory as the primary DB. When the stress run also enables `mmap_read=1`, that read-only instance can hold mmap-backed slices into SST files while the primary continues compacting and deleting obsolete files. If `SstFileManager` uses chunked truncation (`sst_file_manager_bytes_per_truncate > 0`), Linux can raise SIGBUS when the read-only path later touches a page whose backing SST was truncated or removed. The range tombstone fragmenter/comparator is where the invalid mmap-backed key slice was touched, not necessarily the root cause.
Evidence from T281873950 / Sandcastle job `63050395314610170`: the crashing command had `--mmap_read=1`, `--open_read_only_one_in=16`, and `--sst_file_manager_bytes_per_truncate=1048576`. The crash stack was `BlockBasedTable::ReadRangeDelBlock()` -> `FragmentedRangeTombstoneList::FragmentTombstones()` -> `std::set<ParsedInternalKey>::emplace()` -> comparator -> `memcmp`, ending in signal 7/SIGBUS. The crash DB LOG shows `allow_mmap_reads: 1`; table `104662.sst` was created cleanly at size `6257069` with range tombstones and was later deleted by the primary. The artifact does not show persistent SST corruption or checksum failure, and injected read faults would normally surface as RocksDB statuses rather than an OS SIGBUS.
Why this started surfacing: D111995911 / `rFBS0f7a2ef93e63cfa6677b6fa32c1bd34eb818c978` added randomized read-only-on-primary coverage via `open_read_only_one_in` on 2026-07-20. Before that change, blackbox crash test did not randomly keep a separate read-only DB open on the primary live directory, so the existing `mmap_read + SstFileManager truncation` hazard lacked the second reader needed to trigger SIGBUS. The failure is therefore a probabilistic exposure of the new stress mode when it combines with mmap reads and chunked SST truncation.
Fix: disable `open_read_only_one_in` in the existing `mmap_read=1` sanitizer path. This preserves read-only-on-primary stress coverage for non-mmap runs while avoiding an OS-level mmap/truncation hazard that RocksDB cannot convert into a normal `Status`.
Reviewed By: jaykorean
Differential Revision: D113774892
fbshipit-source-id: f82ec2924c12f4512aeeb7a913659dc9a72a4c2f
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15006
Block::GetCorruptionStatus() uses restart_offset_ in error state to decide whether it can re-decode the data block footer for a more specific corruption status: nonzero means the footer decode failed and restart_offset_ stores the original block size; zero means the corruption was detected after the footer decoded.
For separated-KV data blocks, Block::Block() first sets restart_offset_ to the restart-array offset, then validates footer.values_section_offset. If that offset is past the restart array, the block is marked corrupt with size = 0, but restart_offset_ was left nonzero. NewDataIterator() then called GetCorruptionStatus(), which treated the stale restart-array offset as an original block size, re-decoded a syntactically valid footer, and hit DEBUG_FAIL("ok status on presumed bad block contents").
Reset restart_offset_ to 0 on this semantic-corruption path so GetCorruptionStatus() reports generic block corruption instead of trying to re-decode a footer that already decoded successfully.
The regression test builds a separated-KV hash-index data block, corrupts only the encoded values_section_offset to point past the block, and verifies NewDataIterator() returns "Corruption: bad block contents". Removing the fix line makes the test reproduce the original assertion.
This update also applies the clang-format layout required by check-format for the new regression test.
Reviewed By: joshkang97
Differential Revision: D113561303
fbshipit-source-id: 8d989a38761063c62694837976aad98f1868f884
Summary:
Follow-up cleanup to the embedded (same-file) blob wide-column work (PR https://github.com/facebook/rocksdb/issues/14990): the "ReadOptions usage cleanup/optimization" item.
Split the Version-backed blob fetcher into two variants sharing a common VersionBlobFetcherBase (which implements FetchBlob via a virtual read_options()):
- VersionBlobFetcher now holds a `const ReadOptions&` for transient/stack fetchers (the common case: point Get/MultiGet, DBIter, version_set), so the growing ReadOptions is no longer copied per fetcher on the hot path. A deleted rvalue-ref constructor rejects binding to a temporary.
- OwningVersionBlobFetcher takes `ReadOptions&&` and owns a copy, for fetchers whose lifetime is independent of the caller (the read-path lazy resolver and compaction), making ownership transfer explicit.
DBIter::BlobReader now holds a single blob fetcher instead of decomposed scalar fields (read_tier/verify_checksums/fill_cache/io_activity). This removes the decompose-then-recompose that the old CreateBlobFetcher and RetrieveAndSetBlobValue performed, and closes the field-loss gap: all ReadOptions fields now flow through to blob reads rather than just those four. CreateBlobFetcher is gone; the merge path uses the reader's owned fetcher directly. A CanResolve() predicate on the fetcher preserves the prior "unexpected blob index" Corruption instead of dereferencing a null Version, and the now-redundant allow_blob_write_path_fallback_ member/ parameter (equal to the fetcher's own flag) was dropped.
ReadPathBlobResolver keeps owning its fetcher (now OwningVersionBlobFetcher) so it stays self-contained for the planned lazy entity read path, where the result outlives the read call. A TODO in db_iter.h records the remaining duplication (BlobReader and ReadPathBlobResolver each own a fetcher with the same {version, read_options, blob_file_cache, allow_write_path_fallback}), deferred to that work.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15003
Test Plan:
Pure refactoring with no intended behavior change; Version::GetBlob's Slice overload just decodes and calls the BlobIndex overload, so routing whole-value blob reads through the fetcher is equivalent, and the direct-write fallback flag/read options are unchanged. Covered by existing tests:
- db/blob/db_blob_index_test.cc: embedded/entity blob GetEntity/MultiGetEntity zero-copy, plain Get of an embedded default column, merge over an embedded-blob entity, read_tier == kBlockCacheTier Incomplete, corrupt record Corruption.
- db/blob/db_blob_basic_test.cc and db/wide/db_wide_blob_direct_write_test.cc: separate-file and direct-write blob resolution (the fallback branch).
- db/db_iterator_test.cc (blob) and db/compaction/compaction_iterator_test.cc: DBIter blob/merge paths and the compaction (owning) fetcher.
Reviewed By: xingbowang
Differential Revision: D113582247
Pulled By: pdillinger
fbshipit-source-id: 231c9364e266b826d12a8479cbb7f026a0d26bf6
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14998
T280925646 reported an ASAN double-free in db_stress' batched prefix scan. The visible free was in VerifyWideColumns(), but WideColumn values are non-owning Slice views, so that destructor should only free the vector element storage. The stack points to prior ownership corruption rather than an intentional verifier free.
The concrete bug is in PinnableSlice's move constructor: it delegates to move assignment without first initializing the target object. When moving from a pinned slice, move assignment does not set buf_ because the target is still pinned. If that moved-to slice is later Reset() and reused with PinSelf(), it writes through an uninitialized buf_ pointer.
This became reachable after D112212266 changed PinnableWideColumns to keep backing buffers in std::forward_list<PinnableSlice> and emplace moved PinnableSlice values into backing nodes.
Fix the move constructor by delegating to the default constructor first, so buf_ is initialized before move assignment. Add a regression that poisons placement-new storage, move-constructs from a pinned slice, resets it, and reuses it with PinSelf(); this crashed before the fix.
Reviewed By: pdillinger
Differential Revision: D113480016
fbshipit-source-id: 38ce9030f392ff9f0438d38d53d19e356e2d2fbf
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/15001
Add IODispatcher-specific histograms for async read observed completion latency, `FileSystem::Poll` wait time, and submit-to-consumer-poll prefetch lead time. Record submit timestamps on async read state, report completion latency from the callback, and report poll wait/prefetch lead when the block consumer polls the handle.
Keep Java histogram byte mappings stable by assigning the new histogram IDs after the existing range. Add a controlled async filesystem test that verifies one deferred async read records all three histograms.
Address Infer feedback by iterating `async_state->block_indices` directly when removing async IO map entries. Keep `PollAndProcessAsyncIO()` holding an owning `shared_ptr<AsyncIOState>` while erasing those entries so the coalesced request state cannot be destroyed mid-loop.
Format `util/io_dispatcher_imp.cc` and `util/io_dispatcher_test.cc` with clang-format 21.1.2 to fix the format signal.
Reviewed By: pdillinger, anand1976
Differential Revision: D113549401
fbshipit-source-id: dafc272c9252e1baccf404f681d33f76890ce0b2
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14993
A corrupted RocksDB SST data block whose footer sets the hash-index bit
(`kDataBlockBinaryAndHash`) could drive an out-of-bounds heap read while the
block is being parsed. `DataBlockHashIndex::Initialize` guarded its
`num_buckets` / `map_offset` arithmetic only with `assert()`, which is compiled
out in release and ASAN builds. When the encoded `num_buckets` exceeds the
block, the `uint16_t` `map_offset` computation underflows; that value then
flows into `Block`'s ctor at `input.remove_suffix(input.size() - map_offset)`,
inflating `restart_offset_` far past the real allocation. If `size` is later
zeroed as an error marker (e.g. a corrupted separated-KV footer),
`Block::GetCorruptionStatus` builds a `Slice(data(), restart_offset_)` and
`DataBlockFooter::DecodeFrom` reads `data() + restart_offset_ - 4`, over-reading
the block's heap buffer (reported by agentic fuzzing, T279933950).
Fix:
- `DataBlockHashIndex::Initialize` now returns `bool` and rejects a
`num_buckets` of 0 or one whose bucket array plus the `NUM_BUCKETS` field does
not fit within `size`, instead of relying on debug-only asserts.
- `Block`'s ctor checks that return value and also rejects blocks too large for
the 64KiB `uint16_t`-offset hash-index format, marking the block corrupt
(`size = 0`) rather than under-flowing `map_offset` and over-reading.
Reviewed By: pdillinger
Differential Revision: D113429299
fbshipit-source-id: 3167106e793252c09fd6a558e3ac96de7a2169c0
Summary:
DB::Flush() with wait=true only waits until the flush result is committed (memtables removed from the immutable list). The EventListener::OnFlushCompleted callbacks run afterward, on the background flush thread, with the DB mutex released, and the waiter is only woken by a shared condition variable that many unrelated events also signal. As a result a Flush(wait=true) caller can return before -- or in the middle of -- its OnFlushCompleted callbacks. This surfaced as a flaky failure in the c_test table_filter_listener_metadata case, which asserts listener_state.flush_count == 3 immediately after Flush() returns (https://github.com/facebook/rocksdb/actions/runs/29976431449). It is also long-standing behavior that other tests have worked around with TEST_WaitForBackgroundWork().
This adds an opt-in FlushOptions::listener_wait (default false). When set together with wait=true, Flush() does not return until the OnFlushCompleted callbacks for the flushed memtables have finished. Default behavior is unchanged.
Implementation: MemTableList gains a per-column-family num_pending_flush_notifications_ counter (guarded by the DB mutex). The background flush thread reserves a pending notification at flush-commit time -- before the mutex is released to run the callbacks -- and releases it after they return, in both the single-CF (FlushMemTableToOutputFile) and atomic (AtomicFlushMemTablesToOutputFiles) paths. WaitForFlushMemTables treats a CF as finished only when its normal predicate holds and the counter is zero. Because the memtables are removed and the counter incremented under one continuous mutex hold (and, for atomic flush, all CFs are reserved before the per-CF notify loop starts releasing the mutex), a listener_wait waiter can never observe the flush as finished before the callbacks complete.
The counter is per-CF rather than per-flush, so a listener_wait waiter can wait somewhat longer than strictly necessary: under rapid flushing it may block until an unrelated, later flush's OnFlushCompleted on the same CF also drains (e.g. a waiter satisfied by GetEarliestMemTableID() advancing still waits for the whole CF's pending notifications). This can only over-wait, never return early, and is tolerated in favor of implementation simplicity; a per-memtable-id watermark could tighten it if it ever matters.
To keep the counter balanced against future edits, the release uses an RAII Defer (runs on every exit path) and ~MemTableList asserts the counter has drained to zero; SubPendingFlushNotifications asserts against underflow. A reservation is taken only when a callback will actually fire (committed flush info non-empty, not a mempurge, CF not dropped, listeners registered).
Also adds the C API rocksdb_flushoptions_set/get_listener_wait (regenerated via tools/c_api_gen) and a DBImpl::TEST_FlushMemTableWithListenerWait() helper.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14992
Test Plan:
New db_flush_test cases ListenerWaitWaitsForOnFlushCompleted and ListenerWaitAtomicFlushWaitsForAllOnFlushCompleted use a listener that blocks inside OnFlushCompleted plus spurious bg_cv_ wakeups to deterministically verify that Flush(wait=true, listener_wait=true) does not return until the callback (all CFs, for atomic flush) has finished; the atomic case guards the per-CF mutex-release gap in the notify loop.
Converted existing racy/worked-around cases to the new option: c_test table_filter_listener_metadata, listener_test OnSingleDBFlushTest and MultiDBMultiListeners, and db_flush_test FireOnFlushCompletedAfterCommittedResult (the last previously relied on TEST_WaitForBackgroundWork()). The auto-generated C API roundtrip test covers the new flushoptions field.
db_stress StressTest::TestFlush now sets listener_wait on a hard-coded 1 in 4 of its manual flushes (both the atomic and non-atomic branches), so the crash test exercises the new path with the DbStressListener registered.
Reviewed By: jaykorean
Differential Revision: D113425407
Pulled By: pdillinger
fbshipit-source-id: 6fd0f5bc93fabaccd8010c2fead1d93c4c44ccfb
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14948
Adds a reusable `CheckpointEngine` (opened once, works for any DB) that creates checkpoints via the shared `CopyEngine` pool, copying and hard-linking data files across `max_background_operations` threads. `CheckpointImpl::CreateCheckpoint` is refactored into a shared `CreateCheckpointImpl`, with the serial/parallel split behind a `CheckpointFileMover`; files the FileSystem cannot hard-link fall back to copies. The legacy `Checkpoint` API stays serial and unchanged.
Reviewed By: xingbowang
Differential Revision: D111387993
fbshipit-source-id: 036d375da9c9b8f051cfc3da0551b035e29d80c4
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14947
Teaches the shared `CopyEngine` to hard-link files through its background pool: adds `WorkItemType::Link` and `CopyEngine::LinkFile(src, dst, dst_env)`, which returns the FileSystem's `NotSupported` when links are unavailable so callers can fall back to copying. This is what CheckpointEngine (next diff) needs for warm storage, where linking is supported but benefits from parallelism.
No existing caller links through the engine yet (BackupEngine never hard-links), so no behavior change; the path is exercised by a new `copy_engine_test.cc` (parallel copy, parallel link, single-thread copy, per-item error reporting).
Reviewed By: xingbowang
Differential Revision: D111387972
fbshipit-source-id: ec0b0e9b80749a3f51e62066ceef2df21a1811d2
Summary:
Extracts BackupEngine's parallel file-copy/checksum thread pool into a standalone `CopyEngine` (`utilities/copy_engine/copy_engine.{h,cc}`) so `CheckpointEngine` (later in the stack) can reuse it. Behavior-preserving; all `TEST_SYNC_POINT` names are unchanged.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14946
Reviewed By: xingbowang
Differential Revision: D111387950
Pulled By: rban1
fbshipit-source-id: b5046adb98735a606fd1ba471731fd316ce6f036
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14952
`DBImpl::ResumeImpl()` constructed its local `ReadOptions`/`WriteOptions`
with defaults and carried a `// TODO: plumb Env::IOActivity, Env::IOPriority`
comment. Those options flow into `VersionSet::LogAndApply()` (the MANIFEST
write during error recovery), so the rate limiter could not attribute or
prioritize that I/O correctly.
This resolves the TODO for `ResumeImpl` (1 of 91 identical TODOs across the
codebase) by accepting an `Env::IOActivity` parameter and constructing the
options with it, following the pattern already established in `Get()`,
`MultiGet()`, and `NewIterator()`, and in the `kDBOpen` paths in
`db_impl_open.cc`.
Error recovery is driven by flush operations (as the existing comment in
`ErrorHandler::RecoverFromRetryableBGIOError` notes), so both production
callers and the `TEST_ResumeImpl` hook pass `Env::IOActivity::kFlush`. Setting
`io_activity` on these options is safe: `LogAndApply` does not assert
`io_activity == kUnknown`.
Reviewed By: pdillinger
Differential Revision: D112537885
fbshipit-source-id: 59045fd2f0a6e5bbacef6e0381e03ff95cd28c51
Summary:
This is a follow-up to D112212266 /
https://github.com/facebook/rocksdb/pull/14943 (multi-buffer PinnableWideColumns), which added the zero-copy read path for separate-blob-file references and explicitly left extending it to SST embedded-blob references as a follow-up.
Point GetEntity/MultiGetEntity of a wide-column entity whose columns are stored as embedded (same-file) blob records previously deserialized the entity, copied every column name/value, fetched each embedded blob (copying it), and re-serialized the whole entity before GetContext ever saw it (BlockBasedTable::MaybeResolveEmbeddedValue's wide-column branch). This extends the existing multi-buffer PinnableWideColumns zero-copy machinery (previously only for separate blob files / direct-write) to embedded blob columns.
Changes:
- New SameFileBlobReader interface (db/blob/same_file_blob_reader.h), a narrow seam for resolving a same-file BlobIndex against the current SST. BlockBasedTable implements it (GetSameFileBlob), delegating to the existing blob-cache-backed / pinned embedded reads.
- BlobFetcher is now an abstract interface; the Version-backed implementation moved to VersionBlobFetcher. Added an EmbeddedAwareBlobFetcher decorator that routes same-file references to a SameFileBlobReader and delegates all others to a base fetcher (EffectiveFetcher() selects the decorator only when a same-file reader is present).
- The point-Get and MultiGet loops now hand the raw entity to GetContext with the table as SameFileBlobReader, instead of pre-resolving/re-serializing. This is gated on not building a row-cache replay log, so a row-cache entry still captures a fully resolved entity (a later row-cache hit never reopens the SST to resolve same-file references).
- GetContext composes the decorator and resolves same-file and separate-file blob columns zero-copy across the columns / default-value / merge consumers.
- The iterator (DBIter) path is unchanged (still re-serializes via MaybeResolveEmbeddedValue).
- Also fixes latent method hiding in the test-only MockBlobFetcher.
Intended follow-ups:
- Lazy/deferred embedded resolution: return a LazyWideColumns result whose blob columns stay unresolved and are resolved on demand (whole or by byte range). The SameFileBlobReader seam and the polymorphic BlobFetcher introduced here are the mechanism for embedded support there.
- ReadOptions usage cleanup/optimization: have DBIter/BlobReader hold a single ReadOptions instead of decomposed scalar fields (removing the decompose-then-recompose in CreateBlobFetcher and closing the field-loss gap), and split VersionBlobFetcher into a ReadOptions-referencing variant for transient/stack fetchers plus an owning variant for lifetime-independent ones (the lazy resolver and compaction), to avoid copying the (growing) ReadOptions on the hot path.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14990
Test Plan:
Added embedded wide-column tests to db/blob/db_blob_index_test.cc:
- GetEntity/MultiGetEntity of an embedded-blob wide-column entity return the correct columns and resolve them zero-copy (blob columns served from the blob cache: identical data() across reads and in separate backing buffers; pointer-stable across a PinnableWideColumns move).
- Plain Get of an entity whose default column is an embedded blob.
- Merge base and GetMergeOperands over an embedded-blob entity.
- read_tier == kBlockCacheTier yields Incomplete for an unresolved same-file reference.
- A corrupt embedded record surfaces Corruption, not a raw BlobIndex.
# Performance validation with db_bench
Measured with db_bench benchmarks added in https://github.com/facebook/rocksdb/issues/14995 / D113455538, temporarily stacked
on this diff for A/B testing.
## Setup
- **Baseline**: this diff's parent + db_bench benchmark commit.
- **Treatment**: this diff, incl. its test/lint fixups + the same
db_bench benchmark commit.
- Both are release builds (`DEBUG_LEVEL=0 make db_bench`). The diff touches only
the read path, so both binaries produce byte-identical SSTs; the db_bench commit
is additive (`tools/db_bench_tool.cc` only) and identical in both.
- Host: AMD EPYC-Genoa, 166 vCPU, 224 GiB RAM. Single reader thread.
## Method
- Datasets are created **once** and read by **both** binaries via
`--use_existing_db`, so the A/B compares identical on-disk data.
- Each read runs twice in-process (warmup + measured); the blob cache (1 GiB) and
block cache (512 MiB) hold the whole ~400 MiB dataset, so the reported (2nd) run
is fully warm — isolating the column-resolution cost the diff changes rather than
I/O.
- 100k keys, `value_size=1024`, `embedded_blob_min_size=256`,
`compression_type=none`, `format_version=7`, `--duration=10` per measured run.
- Datasets:
- **A** — wide-column entity, `num_short_wide_columns=1` + `num_wide_columns=4`
(small inline default column + 4 embedded-blob columns). Read via
`readrandomentity` (GetEntity) and `multireadentity` (MultiGetEntity, batch 32).
- **B** — wide-column entity, `num_short_wide_columns=0` + `num_wide_columns=4`
(the default column itself is an embedded blob). Read via plain `readrandom`
(Get returns the resolved default column).
- **C** — whole-value embedded blobs (`fillembeddedblob`). Read via plain
`readrandom`. This path is **not** changed by the diff — a regression check.
Example commands:
```
# create (once)
db_bench --benchmarks=fillembeddedentity --db=/tmp/perfA --use_existing_db=0 \
--num=100000 --value_size=1024 --embedded_blob_min_size=256 --format_version=7 \
--num_wide_columns=4 --num_short_wide_columns=1 \
--use_blob_cache=true --blob_cache_size=1073741824 --cache_size=536870912 \
--compression_type=none
# read (each binary; warmup + measured)
db_bench --benchmarks=readrandomentity,readrandomentity --db=/tmp/perfA \
--use_existing_db=1 --threads=1 --duration=10 <same options>
```
## Results (warm, 2 runs each)
| Benchmark (read path) | Baseline µs/op | Treatment µs/op | Latency | Baseline Kops/s | Treatment Kops/s | Throughput |
|---|---|---|---|---|---|---|
| `readrandomentity` — GetEntity, inline default + 4 embedded cols | 7.16 | 5.00 | **-30%** | 140 | 200 | **1.43×** |
| `multireadentity` — MultiGetEntity (batch 32) | 6.46 | 4.54 | **-30%** | 155 | 220 | **1.42×** |
| `readrandom` — Get of embedded default column (dataset B) | 6.55 | 2.12 | **-68%** | 153 | 471 | **3.09×** |
| `readrandom` — whole-value embedded blob (dataset C, unchanged path) | 1.92 | 1.91 | ~0% | 521 | 524 | neutral |
Both runs were within ~1–3% of each other (raw numbers below).
## Interpretation
- **GetEntity / MultiGetEntity of embedded-blob wide-column entities improve ~30%
in latency (~1.4× throughput)** — the expected win from resolving embedded blob
columns zero-copy into the `PinnableWideColumns` backing buffers instead of the
old deserialize + per-column copy + re-serialize round trip.
- **Plain `Get` of an entity whose default column is an embedded blob improves
~3×** — the largest win, because the old path re-serialized the entire entity
even when only the default column is returned, whereas the new path resolves just
that column.
- **The whole-value embedded blob `Get` path is unchanged (~0% delta), confirming
no regression** on the pre-existing zero-copy path.
No unexpected regressions observed; all deltas match the diff's intended scope.
## Raw measured numbers (both runs)
```
readrandomentity BASELINE 7.183 / 7.137 µs/op 139225 / 140118 ops/s
readrandomentity TREATMENT 4.992 / 5.008 µs/op 200333 / 199681 ops/s
multireadentity BASELINE 6.458 / 6.458 µs/op 154837 / 154843 ops/s
multireadentity TREATMENT 4.509 / 4.577 µs/op 221798 / 218463 ops/s
readrandom (Bdef) BASELINE 6.497 / 6.609 µs/op 153920 / 151307 ops/s
readrandom (Bdef) TREATMENT 2.126 / 2.120 µs/op 470317 / 471805 ops/s
readrandom (Cblob) BASELINE 1.891 / 1.947 µs/op 528760 / 513554 ops/s
readrandom (Cblob) TREATMENT 1.895 / 1.920 µs/op 527767 / 520807 ops/s
```
Reviewed By: xingbowang
Differential Revision: D113347622
Pulled By: pdillinger
fbshipit-source-id: c51bc187085695d559411d9c0555bb52c8e0a61a
Summary:
db_bench had no way to exercise SST embedded (same-file) blob reads or wide-column entity reads (GetEntity/MultiGetEntity), which are needed to measure the zero-copy embedded-blob wide-column entity read path.
This adds:
- Two setup benchmarks, fillembeddedentity and fillembeddedblob, that build an SST via SstFileWriter::OpenWithEmbeddedBlobs and ingest it (embedded blobs can only be produced this way, not by fillrandom/flush/compaction). Entity mode writes wide-column entities whose large columns are embedded blobs; blob mode writes whole-value embedded blobs. Both require format_version >= 7.
- readrandomentity (GetEntity) and multireadentity (MultiGetEntity) read benchmarks; plain readrandom is reused for the whole-value embedded blob baseline and for entity default-column Get.
- Flags embedded_blob_min_size, num_wide_columns, and num_short_wide_columns to control the embedded threshold and the inline/embedded column mix per entity.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14995
Test Plan: Manual db_bench runs of the new benchmarks, used to validate D113347622.
Reviewed By: xingbowang
Differential Revision: D113455538
Pulled By: pdillinger
fbshipit-source-id: c6b62f5a2d6a61ae0b7cc32ac1b445b566ea8842
Summary:
`USE_CLANG=1 make check-c-api-gen` failed with "clang++ not found" because the staleness check and the Python generator detected clang inconsistently: the Makefile inspected the `$(CXX)` make variable (the fbcode clang path) while the generator read the `CXX` environment variable, which make never exports. More broadly, clang++ was detected in several ad hoc places with fragile heuristics (grepping the compiler path, duplicated fallback lists in shell and Python).
Centralize clang++ detection in build_detect_platform as a single `CLANG_CXX`:
- `CLANG_CXX` honors `$CXX` when it is itself clang (so CI's `CXX=clang++-21` is used), else takes the first clang++ on PATH (bare, versioned, or the Meta-internal clang++.par). `CXX` (when building with clang) and `CLANG_ANALYZER` now derive from `CLANG_CXX` instead of re-detecting.
- Infer `USE_CLANG=1` when the compiler is clang, detected via the compiler's own `__clang__` predefined macro rather than matching the binary path (which false-positives, e.g. a home dir containing "clang"), and emit it so clang-specific Makefile logic is consistent.
- Emit `CLANG_CXX` to make_config.mk and pass it explicitly to the codegen, so `make check-c-api-gen` works without requiring `USE_CLANG=1` and the generator never guesses.
Add a first-class `make regen-c-api` target -- now the preferred regenerate command referenced by the generated banners and user-facing messages. The Python generators no longer detect clang themselves; `--clang` defaults to `clang++` for manual runs and is overridden by the Makefile with `CLANG_CXX`. Regeneration verifies the clang runs before writing any file, so a missing/broken toolchain aborts cleanly instead of leaving a partially regenerated tree.
Make field discovery clang-version-independent: older clang (e.g. 15) spells field types fully qualified (`rocksdb::ReadTier`,
`rocksdb::BlockBasedTableOptions::IndexType`) while the fbcode/CI clang (21) uses the bare `ReadTier` / `IndexType` the family tables expect. Strip the ROCKSDB namespace and enclosing-struct qualifiers so both agree; this is a no-op on newer clang, so checked-in output and CI are unaffected.
Also escape the literal "generated" in the generator .py sources (they emit the marker but are not themselves generated) so tooling stops treating them as read-only generated files.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14994
Test Plan:
Manual verification (not covered by CI):
- `make check-c-api-gen` now runs the staleness sub-check (instead of skipping) and passes in a default gcc build as well as with `USE_CLANG=1` on an fbcode host, since `CLANG_CXX` is always populated.
- `ROCKSDB_NO_FBCODE=1 make regen-c-api` using a system clang 15 (clang++.par) produces output byte-identical to the checked-in files (generated with clang 21), and `check-c-api-gen` reports "up to date" -- confirming the namespace/struct stripping is version-robust and a no-op on clang 21.
- `regen_all.py --clang /nonexistent` exits non-zero with a clear message and leaves the generated files untouched.
- `make regen-c-api` is idempotent (no diff on a clean tree).
- Confirmed the escaped `\x40generated` still emits `// generated` byte-for-byte in the regenerated c.h/c.cc/.inc.
- Reviewed .github/workflows/pr-jobs.yml: the build-linux-clang-21-no_test_run job's `CXX=clang++-21 make check-c-api-gen` resolves `CLANG_CXX` to clang++-21 (honoring `$CXX`) rather than the container's stray bare clang++.
Reviewed By: xingbowang
Differential Revision: D113447462
Pulled By: pdillinger
fbshipit-source-id: 91ca92d21a71a8b92f9ff57500a8cec91a884078
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14945
First diff in a stack that lets `Checkpoint` and `BackupEngine` share one parallel copy/link infrastructure. Pure options refactor, no behavior change.
Adds two shared option bases in `rocksdb/utilities/checkpoint.h`: `CheckpointOrBackupEngineOptions` (engine-level config) and `CreateCheckpointOrBackupOptions` (per-operation options). `CheckpointEngineOptions` / `CreateCheckpointOptions` derive from them for checkpoint; `BackupEngineOptions` / `CreateBackupOptions` derive from them for backup and keep their backup-only fields.
Reviewed By: xingbowang
Differential Revision: D111387916
fbshipit-source-id: d7efa528ac91fb00090130c9d2a671b13e934a82
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14991
D91705195 ("Replace Prefetch Logic in BlockBasedTableIterator with IODispatcher") moved MultiScan block prefetching out of BlockBasedTableIterator and into the IODispatcher. The IO-level RecordTick/RecordInHistogram calls lived in the deleted prefetch code, and the IODispatcher was not given equivalent instrumentation, so these tickers were still defined but never populated (they published as absent):
- MULTISCAN_BLOCKS_PREFETCHED
- MULTISCAN_BLOCKS_FROM_CACHE
- MULTISCAN_PREFETCH_BYTES
- MULTISCAN_IO_REQUESTS
- MULTISCAN_IO_COALESCED_NONADJACENT
- MULTISCAN_BLOCKS_PER_PREPARE (histogram)
The PREPARE_*, SEEK_ERRORS, and PREFETCH_BLOCKS_WASTED tickers were unaffected (recorded on the prepare/teardown path, which was not replaced).
Fix:
- ReadSet now exposes generic, caller-agnostic prefetch counters (GetNumBlocksPrefetched, GetNumIORequests, GetPrefetchBytes, GetNumCoalescedNonadjacent), incremented once per block in DispatchPrefetch -- the single choke point for both immediate and memory-deferred dispatch. Counting at dispatch time (not completion) keeps async and deferred prefetches accounted for without double counting the async->sync fallback.
- PrepareIORequests now outputs prefetch_bytes (sum of request len) and num_coalesced_nonadjacent (within-threshold gap count) while building coalesced requests, per review feedback to avoid the extra pass over read_reqs/coalesced_block_indices at former line 836. DispatchPrefetch now gets stats directly from PrepareIORequests.
- The MULTISCAN_* ticker names stay at the caller: MultiScanIndexIterator's destructor records the aggregate counters at end-of-scan (alongside the existing PREFETCH_BLOCKS_WASTED), so async/deferred reads that complete during iteration are included. MULTISCAN_BLOCKS_PER_PREPARE is recorded in BlockBasedTableIterator::Prepare. This keeps the IODispatcher a generic mechanism rather than leaking a caller's metric names into it.
- Also plumb the same metrics into PerfContext for per-operation visibility.
Reviewed By: xingbowang
Differential Revision: D113356436
fbshipit-source-id: 660e9755d6b1b9aecccd112df94eb6838061b298
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14985
The Open Source Automated Checkup (T280493780) flagged the 27 generated
`c_api_gen/c_generated_*.inc` fragments for missing the required copyright
header ("Meta Platforms, Inc. and affiliates"). The fragments are emitted by
the C API code generators, none of which wrote a header, so every
regeneration reproduced the violation.
Root-cause fix in the generators so regenerated output is compliant:
- `tools/c_api_gen/generate_c_api.py` (`HEADER_BANNER`)
- `tools/c_api_gen/auto_simple_bindings.py` (`auto_header_banner`)
- `tools/c_api_gen/gen_roundtrip_tests.py`
Each now emits the standard Meta dual-license block (matching `c_base.h` /
`include/rocksdb/c.h`) ahead of the `// generated` marker, and the 27
checked-in `.inc` fragments are regenerated accordingly.
The assembled `include/rocksdb/c.h` and `db/c.cc` already carry a top-level
header, so `regen_all.py`'s inliner now strips each fragment's license block
when inlining to avoid repeating it once per fragment; those two files are
therefore unchanged.
Also adds the header to `tools/gtest_parallel_repro.py`, the one non-`.inc`
source file flagged by the checkup.
Reviewed By: pdillinger
Differential Revision: D113242908
fbshipit-source-id: 4d81d1355c8d8a35cbc4377dd9b21df901ad3cb7
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14934
First diff in the async read stack: pure plumbing, no behavior change. Both the point `Get` path and the batched `MultiGet` path gain synchronous/coroutine twins generated by `DEFINE_SYNC_AND_ASYNC`; `DB::GetAsync` and `DB::MultiGetAsync` become the public callback-based async entry points with synchronous default implementations. For `Get`, the existing `DBImpl::Get` / `GetImpl` bodies are relocated into the sync/async wrapper. For `MultiGet`, the same wiring continues through the DBImpl, version, and table-cache layers. The generated coroutine variants still run the existing synchronous logic.
This keeps the refactor mechanical while preserving dispatch semantics: the coroutine side remains virtual at the same `GetImpl` boundary as the synchronous path, read-only and secondary DBs use generated sync/coroutine overrides instead of falling back to base `DBImpl`, and `StackableDB` leaves async `Get` / `MultiGet` to the base wrappers so wrapper-specific `Get` / `MultiGet` behavior is preserved.
For ABI compatibility and avoiding MACRO blocks in public headers, the asynchronous APIs are callback based instead of coroutine based. Internally it will be developed using coroutines, and applications can also develop their own coroutine wrappers.
This is meant to make the later changes to actually enable coroutine-backed async queries much easier to review.
Benchmarks:
Compared `db_bench` with and without this diff using:
`db_bench --benchmarks=fillseq,readrandom --num=1000000 --reads=5000000 --key_size=16 --value_size=100 --seed=12345 --db=<tmp>`
| Benchmark | Without diff avg | With diff avg | Delta |
| `fillseq` | 4.076 us/op | 4.165 us/op | +2.2% |
| `readrandom` | 3.939 us/op | 3.947 us/op | +0.2% |
No meaningful local `db_bench` regression was observed.
Reviewed By: xingbowang
Differential Revision: D109346179
fbshipit-source-id: 0dee1239f49099b67c11c63426204466e963f903
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14987
The `fast_sst_open` optimization speeds up DB open times by persisting file metadata information in the MANIFEST.
When MANIFESTs are rotated however, this metadata is currently getting dropped, since it is not fed through `VersionEdit::AddFile`. There were also a few additional missing `AddFile` call sites.
Fix: Add an optional `file_open_metadata` to `AddFile`. Feed this through wherever we call `AddFile`.
Reviewed By: anand1976
Differential Revision: D113265655
fbshipit-source-id: c8371d66f6555bb9c372eea095bfa30a5f56a88a
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14986
Why: We want to measure the DB open time during offloaded compactions.
What: Add a histogram for the DB open time in `DB::OpenAndCompact()`.
Reviewed By: anand1976
Differential Revision: D113136652
fbshipit-source-id: f8b2d74bcda3506f7b9ac80fa3e50a172deafc5e
Summary:
An audit of the wide-column blob-reference read paths (from the preceding "multiple backing buffers" change, D112212266) surfaced several inconsistencies in how Status is used: control-flow-via-Status, a status classification that diverged from the rest of the codebase, the same "no blob fetcher" Corruption repeated at many call sites, a swallowed non-OK status (bug), and read statistics advanced on failed reads. This change cleans those up. It is internal only -- no public API or on-disk format change -- and stacks on that diff.
Concretely, this resolves two of the intended follow-ups called out in D112212266:
* "Resolve FIXME about pre-existing swallowed non-OK status in db_impl.cc"
* the error-status/API cleanup (classify an unrecognized wide-column version as Corruption, not NotSupported, at all entry points; replace the NotSupported sentinel from GetValueOfDefaultColumn with an is_blob_reference out-param, removing GetValueOfDefaultColumnResolvingBlobs; and centralize the "blob-backed value without a blob fetcher" Corruption into the resolvers, resolving the ResolveEntityForMerge FIXME, rather than repeating it at each caller). The other two follow-ups from that diff -- the value_size_soft_limit TODO in db_impl.cc and extending the zero-copy optimization to SST embedded-blob references -- are deferred to separate diffs.
Unrecognized version -> Corruption. A wide-column version newer than kVersion2 now returns Corruption instead of NotSupported, consistent with how the rest of the codebase classifies an unrecognized serialized version/type (BlobIndex, the blob log header, the table footer). The guard is applied at every entry point (Deserialize, GetValueOfDefaultColumn, HasBlobColumns, ForEachBlobFileNumber); the latter two previously lacked it and would misparse a future/corrupt version.
Reject trailing data after an empty entity. Deserialize() previously returned OK early for a zero-column entity, silently ignoring any trailing bytes; it now dispatches zero-column entities to the V1/V2 parser like the non-empty case, so the "exactly one serialized entity" trailing-data check the preceding diff added for non-empty entities now covers the empty case too.
Default-column extraction no longer signals via Status. now returns (value, is_blob_reference): for a blob-referenced default column it returns OK with the raw serialized BlobIndex bytes and the flag set, instead of a NotSupported sentinel that every caller had to detect and pair with a second, redundant full deserialization. GetValueOfDefaultColumnResolvingBlobs and the memtable GetDefaultColumnBlobIndexSlice helper are removed; a single ResolveDefaultColumnBlobReference() resolves the raw BlobIndex (handling an inlined index with no fetch). This also fixes a case where a non-default column 0 that happened to be a blob reference was misreported as a blob-backed default. The one remaining NotSupported is the honest, explicit base-DB case: reading a blob-index default column when is_blob_index was not requested.
Null-blob-fetcher handling centralized. "A blob-backed value with no fetcher to read it" (reachable via a reader without blob support such as SstFileReader, or corrupt/legacy data) was checked -- and reported as Corruption -- at ~6 call sites with three slightly different messages, forced by the resolvers asserting the fetcher was non-null. The asserts are gone; each resolver now owns the check in one commented place: ResolveEntityBlobColumnsMultiBuffer (whole entity), ResolveDefaultColumnBlobReference (default column), and the MaybeResolveMemtableBlobValue plain-value path (which open-codes its fetch). Callers just propagate the status. A side benefit: an entity/column whose only blob references are inlined now resolves without a fetcher instead of being wrongly rejected.
Return Status from the memtable resolve helpers, fixing a swallowed error. MaybeResolveMemtableBlobValue (was bool + Status*) and PostprocessMemtableValue- Read (was void + Status*) now return Status, matching codebase convention; MaybeResolveMemtableBlobValue reports "there was a reference to resolve" via a bool* did_resolve out-param (true means it has finalized value/columns). This resolves a pre-existing bug: on the blob-resolve path a non-OK status was returned early without the output cleanup the plain path performed, leaving a half-populated result behind an error. PostprocessMemtableValueRead now has a single success/error finalization block reached by every path (self-pin on success, clear outputs on any error), and returns the finalized status. Also switched the memtable "no blob fetcher" case from NotSupported to Corruption (an internal helper whose sole caller treats all non-OK statuses identically).
Read-only/secondary read stats mirror the primary path on error. DBImpl::GetImpl gates byte/throughput accounting (BYTES_READ, BYTES_PER_READ, get_read_bytes) and merge-operand output on s.ok(); the read-only and secondary Get paths did so unconditionally, logging spurious samples (e.g. a 0-byte BYTES_PER_READ sample) on NotFound/error. They now gate on s.ok() too, while keeping NUMBER_KEYS_READ, MEMTABLE_HIT, and super-version cleanup unconditional.
Header docs updated to match (Deserialize/DeserializeSimple return Corruption for an unexpected blob reference; unknown per-column value types are rejected as Corruption).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14980
Test Plan:
Extended db/wide/wide_column_serialization_test.cc: DeserializeUnsupportedVersion now expects Corruption; new FutureVersionRejectedConsistently verifies that Deserialize, GetValueOfDefaultColumn, HasBlobColumns, and ForEachBlobFileNumber all reject a future version as Corruption; V2GetValueOfDefaultColumnBlobRef now checks the is_blob_reference flag and raw-BlobIndex bytes and that resolving without a fetcher is Corruption; DeserializeRejectsTrailingData now also covers empty (zero-column) V1 and V2 entities.
db/blob/db_blob_index_test.cc: MaybeResolveMemtableBlobValueWithoutFetcherFails- Closed updated to the new signature (asserts did_resolve + Corruption); new PostprocessMemtableValueReadClearsOutputsOnError pins the swallowed-status fix, covering both an already-non-OK incoming status and (the actual FIXME scenario) an OK incoming status whose blob-resolve path fails -- both clear outputs and return the error. The StackableDB-style DBBlobIndexTest.Get contract is unchanged (its NotSupported comes from the base-DB blob-index read path, not these helpers).
Reviewed By: xingbowang
Differential Revision: D113099114
Pulled By: pdillinger
fbshipit-source-id: 4bc374dd74f8ba83bdf49bd8753c1a8d88529cde
Summary:
Resolves the value_size_soft_limit TODO left in db_impl.cc by the PinnableWideColumns multi-buffer change (D112212266 / PR https://github.com/facebook/rocksdb/issues/14943).
ReadOptions::value_size_soft_limit is meant to bound how much a MultiGet buffers, but it was enforced in the batch loop on pre-resolution sizes. Pre-flush blob direct write references (a plain value encoded as a BlobIndex, or a wide-column entity with unresolved blob-valued columns) are only resolved later, in MultiGetImpl's post-processing loop, so their (possibly large) blob values were fetched with no limit applied -- read unbounded. Separate-file blob columns do not have this gap because they are resolved eagerly in GetContext and thus already counted.
Now the post-processing loop enforces the limit on direct-write resolution using the "always make progress" soft-limit semantics: a key is aborted only once the bytes already returned by this MultiGet exceed the limit, so at least one key is always read even if its value alone exceeds the limit. Once the returned total is over the limit, subsequent keys that need fetching get Status::Aborted. This bounds the reads while ensuring a caller retrying the aborted keys cannot loop forever on a single oversized value.
Pieces that benefited from human input and refinement:
* Enforcement semantics: settled on "always make progress" -- checking the bytes already returned (rather than a projected pre-fetch size, which in foreseeable user integrations could starve a lone key that by itself exceeds the limit) -- matching the existing separate-file/batch soft-limit behavior.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14983
Test Plan:
Added to db/wide/db_wide_blob_direct_write_test.cc (memtable / pre-flush direct-write reads, using the ResolveDirectWriteWideColumns:BlobFetched sync point to confirm which blobs are fetched):
* MultiGetEntitySoftLimitBoundsDirectWriteBlobResolution -- two over-limit entities: first read (progress), second aborted; one blob fetch.
* MultiGetEntitySoftLimitAlwaysReadsOversizedKey -- a single entity whose value alone far exceeds the limit is returned OK (no starvation).
* MultiGetSoftLimitBoundsDirectWritePlainValue -- same for plain values stored as direct-write blob references: first read, second aborted.
Reviewed By: xingbowang
Differential Revision: D113159249
Pulled By: pdillinger
fbshipit-source-id: 5af443b86bdfc3ecd0112d3c5af541a77f0d3d67
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14979
`MaybeOpenReadOnlyOnPrimary` (added in D111995911) opens a read-only DB on the primary's live directory while the primary keeps mutating it. Two separate stress-test false positives were observed there; this diff fixes both and consolidates the earlier one-off fixes into a single diff.
1) TSAN deadlock-detector crash in the CF-name snapshot (`fbcode_tsan_blackbox_crash_test`):
ThreadSanitizer: CHECK failed: sanitizer_deadlock_detector.h:67
"((n_all_locks_)) < (...)" (0x80, 0x80)
`n_all_locks_` is the number of mutexes a thread holds *simultaneously*; TSAN caps it at 128 (0x80). The snapshot loop called `LockColumnFamily(cf)` to read `column_family_names_[cf]`, and that helper `.Lock()`s ALL per-key mutexes for the CF at once (`ceil(max_key / 2^log2_keys_per_lock)` -- e.g. 25,000 in blackbox with max_key=100000, log2_keys_per_lock=2) without releasing any, so a single call blows past 128. The writer (`MaybeClearOneColumnFamily`) holds all key locks while renaming a CF, so a single key lock is enough to synchronize the read. Fix: take one `MutexLock` on key-lock index 0 per CF instead of `LockColumnFamily`, keeping the simultaneously-held count at 1.
(`LockColumnFamily` is a latent hazard for any caller; the pre-existing caller `MaybeClearOneColumnFamily` never runs in the crash-test config -- `clear_column_family_one_in=0` -- so only this new, ungated caller exposed it.)
2) False-positive read-only Get (folded in from D112987468). After flushing the primary, the test performed a `Get(Key(0))` on each CF of the read-only instance. That Get races against concurrent primary compaction, which can delete SST files the read-only's frozen MANIFEST still references, surfacing as IOError/PathNotFound/Corruption ("truncated block read"). The Get adds no unique coverage -- the scenario under test (the read-only close wrongly deleting primary live files) is detected on the primary's own verification paths -- so remove the Get block entirely instead of widening a tolerated-status list.
Note on D112936376: that diff tried to fix crash (1) by setting `table_cache_numshardbits=0` on the read-only instance, on the theory that the primary's and reader's table caches contribute "128 unique lock objects." That is a misdiagnosis: the failing CHECK counts locks held *simultaneously*, and sharded-cache shard mutexes are held transiently (one at a time; read-path nesting is a constant depth of 2), so shard count does not affect `n_all_locks_`. The real cause is `LockColumnFamily`, fixed in (1), so `table_cache_numshardbits=0` is unnecessary and is not included here.
This diff subsumes and replaces D112936376 and D112987468.
Reviewed By: joshkang97
Differential Revision: D112980148
fbshipit-source-id: 012ff852109223f3d19d8dbbc065fde9340f9c66
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14971
Follow-up to https://github.com/facebook/rocksdb/pull/14942. That PR made db_bench read ops end the run with an INVALID result (`SetFatal` -> exit 1) on a tolerable remote IO error and left write ops on unconditional `SetFatal`. In practice that still failed the remote benchmark TW jobs (exit 1) on transient WS infra errors, and did not cover write ops or `DB::Open`.
Change the toleration to skip-and-continue so a transient remote IO error no longer fails the run:
- `HandleBenchmarkIOError` now just logs and returns on a tolerable error (was: `SetFatal`); the caller skips the failed op and keeps going. Non-tolerable, data-loss, and local-DB errors still `abort()` (core dump for debugging).
- Applied uniformly to all worker op error paths: `Get` / `MultiGet`, put / merge, delete / delete-range, `DoDelete`, `RandomWithVerify` put/delete-many, and the `BGScan` iterator (stops the scan and continues).
- `DB::Open`: on a tolerable remote IO error, exit cleanly (0) -- the benchmark cannot run, but it is not a db_bench / RocksDB failure.
Gated by `--tolerate_io_errors_for_remote_dbs` plus a remote `--env_uri` / `--fs_uri` backend, so local runs and non-IO / data-loss errors are unaffected. The result is best-effort (skipped ops are not counted); the goal is to keep remote benchmark jobs green through transient WS infrastructure errors instead of failing or crashing.
db_stress is intentionally NOT changed: it is a correctness (crash-recovery) test and must keep failing on real errors.
Differential Revision: D112749712
fbshipit-source-id: a8ccd46ef66e6c3c97a1baffd35d27830879d428
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14940
A few improvements to the CPU corruption injector:
- Seed handling: `--seed` is now validated against db_stress's accepted range and no longer silently wraps, so an out-of-range seed fails fast instead of producing misleading NO_INJECTION runs. Omit `--seed` for a fresh random base seed.
- Per-run reproducibility: each run's full db_stress command is recorded at the top of its `gdb.log`, so any run can be reproduced by copy-pasting one line.
- README: clarified the `--seed` semantics and examples.
**Test plan:**
All db_stress runs use a debug build: `make DEBUG_LEVEL=0 EXTRA_CXXFLAGS="-g -fno-inline" db_stress`.
## Seed handling: out-of-range seeds fail fast
Command:
```
python3 tools/cpu_corruption_injector/runner.py --op write --runs 2 --stress_cmd ./db_stress --report_dir /tmp/tp_badseed --seed 4294967295; echo "exit=$?"
python3 tools/cpu_corruption_injector/runner.py --op write --runs 1 --stress_cmd ./db_stress --report_dir /tmp/tp_badseed0 --seed 0; echo "exit=$?"
```
Output:
```
2026-07-15 01:26:14,408 ERROR --seed range [4294967295, 4294967296] is outside db_stress's accepted [1, 4294967295]; pass a smaller --seed
exit=2
2026-07-15 01:26:14,558 ERROR --seed range [0, 0] is outside db_stress's accepted [1, 4294967295]; pass a smaller --seed
exit=2
```
Both are rejected before any run starts (previously the seed silently wrapped and every run came back a misleading NO_INJECTION).
## Per-run command logged (full db_stress command in gdb.log)
Command:
```
python3 tools/cpu_corruption_injector/runner.py --op write --runs 2 --stress_cmd ./db_stress --report_dir /tmp/tp_seed --seed 424242
head -6 /tmp/tp_seed/run_00000/gdb.log
```
Output (the first line is the full command; the run's gdb/db_stress output follows it):
```
# /usr/bin/gdb --batch --nx -iex 'py import sys; sys.argv=["injector.py", "--op", "write", "--op_index", "259", "--entry_fn", "rocksdb::MemTable::Add", "--target_fn", "rocksdb::MemTable::Add", "--seed", "424242", "--dir", "/tmp/tp_seed/run_00000"]' -x /data/users/huixiao/fbsource/fbcode/internal_repo_rocksdb/repo/tools/cpu_corruption_injector/injector.py --args /data/users/huixiao/fbsource/fbcode/internal_repo_rocksdb/repo/db_stress --db=/tmp/tp_seed/run_00000/db --expected_values_dir=/tmp/tp_seed/run_00000/exp --seed=424242 --verify_cpu_corruption_dir=/tmp/tp_seed/run_00000 --ops_per_thread=1000 --max_key=1000 --value_size_mult=8 --column_families=1 --nooverwritepercent=0 --writepercent=50 --delpercent=40 --delrangepercent=10 --readpercent=0 --iterpercent=0 --prefixpercent=0 --flush_one_in=20 --compact_range_one_in=10 --compact_files_one_in=10 --target_file_size_base=65536 --reopen=0 --destroy_db_initially=1 --clear_column_family_one_in=0 --disable_wal=0 --threads=1 --num_dbs=1 --disable_auto_compactions=1 --write_buffer_size=1073741824 --remote_compaction_worker_threads=0 --subcompactions=1 --batch_protection_bytes_per_key=8 --memtable_protection_bytes_per_key=8 --block_protection_bytes_per_key=8 --paranoid_file_checks=1 --paranoid_memory_checks=1 --memtable_verify_per_key_checksum_on_seek=1 --detect_filter_construct_corruption=true --verify_checksum=1 --compression_checksum=1 --file_checksum_impl=crc32c --verify_checksum_one_in=1 --verify_file_checksums_one_in=1000 --read_fault_one_in=0 --write_fault_one_in=0 --metadata_read_fault_one_in=0 --metadata_write_fault_one_in=0 --open_read_fault_one_in=0 --open_write_fault_one_in=0 --open_metadata_read_fault_one_in=0 --open_metadata_write_fault_one_in=0 --secondary_cache_fault_one_in=0 --sync_fault_injection=0
This GDB supports auto-downloading debuginfo from the following URLs:
<https://debuginfod.centos.org/>
Debuginfod has been disabled.
```
The first line of `gdb.log` is the complete `gdb` + `db_stress` command, including `--seed=424242` and `--max_key=1000`; the run's own output follows below it.
## The logged command reflects real flag values
Temporarily set `max_key` to `12345` in `runner_build.py`, then:
Command:
```
python3 tools/cpu_corruption_injector/runner.py --op write --runs 1 --stress_cmd ./db_stress --report_dir /tmp/tp_maxkey --seed 424242
head -1 /tmp/tp_maxkey/run_00000/gdb.log | tr ' ' '
' | grep -- --max_key
```
Output:
```
--max_key=12345
```
The logged command tracked the changed preset value. Reverted the change afterward.
Reviewed By: archang19
Differential Revision: D112075800
fbshipit-source-id: 0d2be1741ede806a84a1a0f9ca6b2a32147c4e9c
Summary:
Wide-column read paths that materialize blob-referenced columns used to re-serialize the entire entity into a fresh contiguous buffer, copying the (potentially large) resolved column values byte-for-byte. This was forced by PinnableWideColumns holding a single backing buffer with all column Slices pointing into it.
PinnableWideColumns now holds its columns as zero-copy Slices into one or more stable backing buffers, kept in a std::forward_list<PinnableSlice> whose nodes never relocate on move, move-assignment, or splice. Resolved blob columns can be spliced in as additional backing buffers instead of re-serializing the entity.
Using address-stable nodes also cleans up the old Slice-invalidation handling. The previous single-buffer representation had to cope with a self-pinned PinnableSlice whose small (SSO) std::string relocates when the PinnableSlice is moved: on every move it detected the relocated backing pointer and re-derived the column Slices by re-parsing. With forward_list nodes that never move, the column Slices stay valid across move / move-assignment / splice, so that handling is gone and the move operations are now the defaulted member-wise moves.
The multi-buffer plumbing is internal only: a new PinnableWideColumnsHelper (a friend) performs the splice and the little bit of private-member access the read paths need, so there is no new public API surface and the previous `friend class DBImpl` is gone. The public view is unchanged: columns() still returns a single, in-order set.
Read paths converted to resolve without re-serializing:
- GetContext::SaveWideColumnEntityToColumns (eager GetEntity SST reads, separate-blob-file references)
- DBImpl::ResolveDirectWriteWideColumns (direct-write blob references)
- DBImpl::MaybeResolveMemtableBlobValue (blob-backed memtable reads)
Merge output still serializes (the MergeOperator API consumes/produces a serialized entity). Same-file ("embedded") blob columns are resolved -- and still re-serialized -- earlier in the block-based table reader (MaybeResolveEmbeddedValue / ResolveEmbeddedValueForGet) before the entity reaches GetContext, so the GetEntity path above only avoids re-serialization for separate-blob-file references. Compaction's separate entity machinery is unchanged.
API/accounting: deprecated PinnableWideColumns::serialized_size() (comment-only, per repo convention) in favor of the new payload_size() (sum of column name + value sizes; equals the value size for a plain value). Read statistics for entities (BYTES_READ, BYTES_PER_READ, per-key GetEntity/MultiGetEntity byte counts) and the ReadOptions::value_size_soft_limit threshold now measure entities by payload_size() instead of the serialized size. This drops framing bytes (entity measurements are slightly smaller), thus making plain-value and wide-column accounting consistent and reducing unnecessary exposure to implementation details.
Serialization cleanup enabled by the above:
- Consolidated Deserialize/DeserializeV2 into a single Deserialize(const Slice&, WideColumns&, blob_columns*) plus a DeserializeSimple alias for the no-blob-references case; both take const Slice&, require the input to be exactly one serialized entity, and reject any trailing data uniformly (V1 previously ignored trailing bytes). An unexpected blob reference where a fully-resolved entity is required now returns Corruption.
- ResolveEntityBlobColumns (serialized output, used only by merge) is now a thin wrapper over the new multi-buffer resolver plus Serialize, removing the duplicated fetch loop and the SerializeResolvedEntity helper.
- Consistently use `const Slice& input` rather than `Slice& input` for wide column functions expecting an input slice with no trailing data.
Design/polish from human review:
* The initial plan called for more complex handling around a vector<PinnableSlice> backing store (unique_ptr indirection and re-deriving the column Slices on move to cope with reallocation and SSO relocation); switching the backing store to a std::forward_list of address-stable nodes was a human suggestion that removed that complexity and the move-time Slice re-derivation entirely.
* Keep resolve plumbing off the public API via PinnableWideColumnsHelper and removing the DBImpl friend
* Various pieces of cleanup and polish (more left to do)
* The eager-GetEntity optimization
* The memtable inline-copy NOTE/TODO.
Intended follow-up:
* Resolve FIXME about pre-existing swallowed non-OK status in db_impl.cc
* Resolve TODO about value_size_soft_limit in db_impl.cc
* Extend this optimization to SST embedded-blob references
* Error-status/API cleanup: classify an unrecognized wide-column version as Corruption (not NotSupported) at all entry points; replace the NotSupported sentinel from GetValueOfDefaultColumn with an explicit is_blob_reference out-param (removing GetValueOfDefaultColumnResolvingBlobs); and centralize the "blob-backed value without a blob fetcher" Corruption into the resolvers (resolving the ResolveEntityForMerge FIXME) rather than repeating it at each caller
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14943
Test Plan:
Extended db/wide/wide_column_serialization_test.cc with PinnableWideColumns unit tests: set/payload_size for plain, V1, and resolved entities; move and move-assign of a multi-buffer result preserve column data pointers (the node-stability guarantee); Reset; ResolveColumns splices blob buffers in zero-copy (resolved values' data() point into the spliced-in buffers, not a re-serialized entity); serialized_size() == SerializedSizeV1(columns()); and Deserialize rejects trailing data after a serialized entity for both V1 and V2 layouts.
Built and ran (debug): wide_column_serialization_test, db_wide_basic_test, db_wide_blob_direct_write_test, db_blob_basic_test, compaction_iterator_test, db_merge_operator_test, db_iterator_test, udt_util_test, write_batch_with_index_test -- all pass. librocksdb and db_stress build. make format-auto and build_tools/check-sources.sh are clean.
Reviewed By: xingbowang
Differential Revision: D112212266
Pulled By: pdillinger
fbshipit-source-id: 13b89e2bb7d8371f4c1ff18b234a549bea9c9e77
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14938
Adds db_stress coverage for the topology where a read-only DB (`DB::OpenForReadOnly`) is opened on the same directory as a live read-write primary. This interaction was previously untested: `FLAGS_read_only` runs a DB by itself in read-only mode (no live writer), and `FLAGS_test_secondary` co-locates a secondary that tails the MANIFEST and never owns/deletes files, so neither exercises a read-only instance holding a stale live-file snapshot while the primary keeps creating files. That gap allowed a bug where a read-only DB's close ran obsolete-file cleanup against its stale snapshot and deleted the primary's live SST files.
New flag `open_read_only_one_in=N`: on thread 0, with probability 1/N per operation, open a read-only instance on the primary's live directory, flush the primary (so new SSTs exist that are absent from the reader's frozen snapshot), do a few reads, then close the reader. If the read-only close wrongly purges obsolete files, the primary's live files are deleted and surface through db_stress's existing missing-file/corruption detection (deterministically on the next reopen).
The read-only reader is opened with the primary's options (so it inherits the table factory, e.g. UDI) but with `compaction_service` and `listeners` cleared, since a read-only reader never compacts and must not share the primary's background-coordination services -- otherwise the reader's open/close would abort the primary's in-flight remote compactions.
`db_crashtest.py` enables the mode randomly and excludes configurations it does not support (transactions, BlobDB, TTL, and the blob-direct-write profile). Startup validation in `db_stress_tool.cc` rejects those same combinations when the flag is set directly.
Reviewed By: joshkang97
Differential Revision: D111995911
fbshipit-source-id: 0f7a2ef93e63cfa6677b6fa32c1bd34eb818c978
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14942
Remote storage reached over `--env_uri` / `--fs_uri` (e.g. Warm Storage) can surface transient infrastructure IO errors that are unrelated to any RocksDB bug. Previously such an error would take down the test/benchmark process, making remote runs unusable. This diff lets both `db_stress` and `db_bench` tolerate non-data-loss IO errors, behind default-off flags that only take effect against a remote backend, so local behavior is unchanged and local IO errors are never masked.
**db_stress:** Add a default-off flag `--tolerate_non_injected_io_errors_for_remote_dbs` that treats non-injected, non-data-loss IO errors as retryable, so remote infrastructure errors do not fail the stress harness through `SharedState::SafeTerminate()`. `IsErrorInjectedAndRetryable` enforces at the binary level that this only applies when a remote backend (`--env_uri` / `--fs_uri`) is in use; `db_crashtest.py` additionally force-disables it for local DBs. Remote crash-test jobs opt in by passing the flag on the command line (see the companion Tupperware config change).
**db_bench:** Add a default-off flag `--tolerate_io_errors_for_remote_dbs`. When set and a remote backend (`--env_uri` / `--fs_uri`) is in use, a non-data-loss IO error surfaced by a benchmark operation ends the run gracefully with an INVALID result through the existing cooperative shutdown path (`SharedState::SetFatal`) instead of calling `abort()` (which previously produced a `Received signal 6 (Aborted)` process crash). All read `Get` / `MultiGet` abort sites and the `BGScan` iterator-error site are routed through a shared `HandleBenchmarkIOError` helper; IO errors on local DBs and data-loss errors are never tolerated. With the flag off (default), the original `abort()` behavior is preserved so local runs still produce a core dump for debugging.
Reviewed By: anand1976
Differential Revision: D112136438
fbshipit-source-id: 510dbc644da78c4315a476ec766e26b2711dc060
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14930
Fix a TSAN-detected data race on the vptr between `MultiOpsTxnsStressTest::VerifyPkSkFast` making virtual calls through `db_` and `WritePreparedTxnDB::~WritePreparedTxnDB` updating the vptr during destruction.
The race occurs because `VerifyPkSkFast` (called from a compaction notification callback on a background thread) accesses `db_` which is a non-atomic raw pointer. Meanwhile, the main thread destroys the TransactionDB wrapper via `db_owner_.reset()` in `StressTest::CleanUp()`.
The previous fix (commit 768838ddcd99, `OnDBShutdownBegin`) reduced the race window by adding a `shutting_down_` check in the listener, but did not fully eliminate it because a background thread can pass the check before shutdown begins and then proceed to use `db_` while the destructor runs.
This fix addresses the root cause in two ways:
1. In `VerifyPkSkFast`: Use the atomically loaded `db` local variable (from `db_aptr_.load(memory_order_acquire)`) for ALL virtual calls instead of the non-atomic `db_` member. This ensures the pointer used for virtual dispatch was safely acquired before any shutdown signaling.
2. In `StressTest::CleanUp()`: Store nullptr to `db_aptr_` before calling `Close()`. This prevents any NEW entries into `VerifyPkSkFast` from background threads that have not yet loaded the pointer. `Close()` then waits for all in-flight background work (including any `VerifyPkSkFast` that already loaded a non-null pointer) to complete before proceeding to destroy the DB.
Together, these changes ensure that no background thread performs a virtual call through the TransactionDB pointer while the destructor is modifying the vptr.
Reviewed By: xingbowang
Differential Revision: D111480308
fbshipit-source-id: 0c998888d07712236c28eee60d04c6f169949a80
Summary:
Pull Request resolved: https://github.com/facebook/rocksdb/pull/14951
The `build-macos-cmake` CI job intermittently but persistently fails during `make` with:
GoogleTestAddTests.cmake:... (parse_tests_from_json)
* Line 1, Column 1
Syntax error: value, object or array expected.
make: *** [all] Error 2
This is `gtest_discover_tests` running in its default POST_BUILD mode: it executes each freshly-built test binary during the parallel `-j` build to enumerate test cases as JSON. On the resource-constrained arm64 macOS runners, that discovery run competes with the in-flight compile jobs and can produce empty output, which CMake's JSON parser then rejects, aborting the whole build. The failing targets (e.g. `cache_reservation_manager_test`, `lru_cache_test`) are unrelated to whatever change is under test -- other test binaries in the same run discover fine -- confirming it is a build-phase discovery race, not a code defect.
Switch discovery to `DISCOVERY_MODE PRE_TEST` so enumeration happens at ctest time (serially, after the build) instead of during the parallel build. This keeps the build green while preserving full test discovery. `PRE_TEST` requires CMake >= 3.18, so it is guarded by a version check; on older CMake (the project's declared minimum is 3.12) behavior is unchanged.
Reviewed By: xingbowang
Differential Revision: D112478854
fbshipit-source-id: 0d5edea416610f9cc1e9ae806d7d521165abe656
2026-07-17 09:09:51 -07:00
834 changed files with 78590 additions and 14809 deletions
@@ -14,6 +14,12 @@ This document provides guidance for generating and reviewing code in the RocksDB
**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.
**Refactoring traps:** At least in production code, avoid constructs that could allow existing code to unexpectedly, quietly change meaning.
Specifically:
* Avoid new defaulted parameters. This is the #1 trap on refactoring!
* Avoid static_cast, reinterpret_cast, and C-style casts; static_cast_with_check, up_cast, and lossless_cast from cast_util.h are preferred.
* Avoid declared type `auto` (`auto&`, `auto*` OK).
### 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.
@@ -32,7 +38,7 @@ This document provides guidance for generating and reviewing code in the RocksDB
**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.
**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 (CACHE_LINE_SIZE, typically 64 bytes) and avoid false sharing in concurrent code (see CacheAlignedWrapper). Especially when adding new data members, be mindful of ordering within structs and classes to reduce unnecessary padding (special care with bool members) and avoid larger-than-necessary types (e.g. prefer OptSlice to std::optional<Slice>).
**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.
@@ -47,15 +53,17 @@ This document provides guidance for generating and reviewing code in the RocksDB
- **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.
**Avoid Premature Optimization:**Performance is known critical in places like inner loops of compaction, read path and write path. Elsewhere, try to find a "holy trinity" solution of *correct, simple, and reasonably efficient*. More complex solutions are only appropriate when profiling/benchmarking data shows a measurable payoff.
### 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).
**Behavioral Compatibility:** Behavior changes are human judgment calls that must be intentional, well-documented (including release notes), and well researched for potential downstream production impact. Agents and reviewers must call out potential gaps, ambiguity, or adverse impacts in intentionality, documentation, and production impact. For an existing accepted input—including an undocumented, discouraged, or historically ignored value—the analysis should compare old and new observable behavior, including performance and resource impact, identify known direct callers and stored configurations, and add or update regression coverage as appropriate.
**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.
**Documentation:** Public APIs must be thoroughly documented, without unnecessary embelishment nor dwelling on implementation details nor project planning. When non-obvious, include usage examples, parameter descriptions, known bugs or limitations, notes on thread safety, performance characteristics, and compatibility considerations. Re-read comments for ambiguous terminology and phrasing, such as ambiguously re-purposed programming jargon.
---
@@ -73,7 +81,7 @@ The database core handles write-ahead logging (WAL), memtables, compaction, and
**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 / "Public API" (`include/rocksdb/`)
Public headers define RocksDB's API surface. Changes here have the highest compatibility impact.
@@ -236,6 +244,11 @@ from an implementation detail instead of an explicit option.
## Important tips
### Pull request and diff summaries
* Before writing or updating an external RocksDB pull request summary or its
corresponding internal Phabricator diff summary, read and follow
`claude_md/pull_request_summary_guideline.md`.
### Build system
* There are 3 build system. Make for git clones, BUCK (meta internal) for hg
clones, and CMake for some special cases.
@@ -259,6 +272,10 @@ 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.
`AUTO_CLEAN=1` does not fix Make failures associated with stale .d files
referring to removed files. Resolve that manually or with `make clean`
without complaining to the user.
### Source checks
* Run `make check-sources` before committing. This catches non-ASCII
characters in source files and other source-level issues that CI will
@@ -298,6 +315,7 @@ build (and where noted, run tests) across:
| Axis | Must support |
|------|--------------|
| OS | Linux (x86_64 + ARM), macOS, Windows |
| Architecture | Little-endian and big-endian targets, including Linux on s390x |
> NOTE: Entries for next release do not go here. Follow instructions in `unreleased_history/README.txt`
## 11.13.2 (10/28/2026)
### Bug Fixes
* Fixed a secondary instance with `open_files_async = true` getting stuck after `TryCatchUpWithPrimary()` returned `PathNotFound` for a table file that the primary deleted before the background opener reached it.
* Fixed blob direct write with more than one concurrent flush of a column family: a later flush sealed and published the blob files of the oldest pending memtable instead of its own, so flushes failed with `Corruption: VersionBuilder: Blob file #N already added` and the database stopped.
## 11.13.1 (10/23/2026)
### Bug Fixes
* Fixed data loss with `min_tombstones_for_range_conversion` and external file ingestion. An iterator whose view predated the ingestion could convert the deleted keys around the ingested keys into a range tombstone that deleted them. This happened when the file got no new sequence number (sequence number zero, `ingest_behind`, or `allow_db_generated_files`), and when an iterator was created during the ingestion.
* Fixed a data loss bug where an iterator with `ReadOptions::read_tier = kMemtableTier` could, with `min_tombstones_for_range_conversion` enabled, insert a range tombstone that deleted live keys stored only in SST files. Read-path range tombstone conversion now runs only for iterators with `read_tier = kReadAllTier`.
## 11.13.0 (10/21/2026)
### New Features
* Added `FileSystem::IsPreallocationSupported()`. A `FileSystem` that never preallocates file space can return false so that DB open skips reopening and truncating the last WAL file.
### Public API Changes
* Added the experimental `DBOptions::track_options_file_number_in_manifest`
rollout option (with C and Java bindings). The option is persisted in each
OPTIONS file. When enabled, RocksDB durably creates an OPTIONS file and then
commits its file number in MANIFEST. Column-family create/drop commits the
configuration snapshot and membership in one MANIFEST record. A newer tracked
OPTIONS file without a commit is ignored after a crash, while a newer OPTIONS
file written with tracking disabled retains the legacy highest-numbered-file
behavior. Batched column-family changes publish one snapshot; after partial
success it can be a safe superset that persisted-options APIs filter against
live MANIFEST membership. Ordinary `DB::Open` continues to use caller-supplied
options and does not depend on the operational snapshot.
* Added `EncodeEntry()`, `PackSequenceAndType()`, and `UnPackSequenceAndType()` in `rocksdb/utilities/types_util.h` to construct internal keys and pack or unpack their sequence numbers and entry types. `kInternalKeyFooterSize` exposes the footer size in bytes.
* Added `ldb update_manifest --rebind_options_file --options_file=OPTIONS-<number>` for offline repair of a committed MANIFEST OPTIONS file number. Operators must select a trusted snapshot that has MANIFEST tracking enabled and provide any external DB/CF allocation roots that are not serialized in OPTIONS.
### Behavior Changes
* With `max_bgerror_resume_count <= 0` (auto resume disabled), a retryable or file-scoped IO error during compaction or flush is now a hard background error that stops writes until `DB::Resume()` is called. Previously, compaction errors were silently retried, and flush errors with a healthy WAL were soft errors that let writes continue while flushes were stopped.
### Bug Fixes
* Fixed two more `SstFileManager` trash accounting leaks: retrying a failed trash deletion immediately (for example with rate limiting disabled) after the file had disappeared, and a trash file that was deleted but whose directory sync then failed. In both cases the file's size stayed in the total trash size and tracked size until restart.
* Fixed `SstFileManager` trash accounting leaking the size of a queued trash file that disappeared before deletion (for example because its directory was renamed). The leaked bytes stayed in the total trash size until restart, which could push every later deletion past `max_trash_db_ratio` and make it immediate instead of rate limited.
* Fixed `MultiGet` with `ReadOptions::async_io` returning `NotSupported` when io_uring initialization or an asynchronous read submission fails. It now checks async I/O support on the reading thread and falls back to synchronous reads when needed.
* Fixed subset checkpoints so their newest OPTIONS file lists only the column
families retained by the checkpoint. This applies with both legacy
highest-numbered OPTIONS selection and MANIFEST-tracked OPTIONS selection, and
prevents `LoadLatestOptions` from recreating column families that the subset
checkpoint intentionally dropped from its copied MANIFEST.
### Performance Improvements
* Perf context counters and timers no longer call the thread-local init function of `perf_context` on every update. With `kEnableCount`, a counter update costs about 60% less.
## 11.12.0 (10/07/2026)
### New Features
* Added `FullExternalTableFactory` and related full-mode external table APIs that preserve RocksDB internal keys, enabling external tables to support nonzero sequence numbers, point and single deletions, merge operands, blob indexes, wide-column entities, snapshots, live DB writes, and multi-level compactions.
### Public API Changes
* Added experimental coroutine capability interfaces for basic and full `ExternalTableReader` implementations. Coroutine-capable external readers can initiate file reads through `FSRandomAccessFile::SubmitReadAsync()`.
* `ExternalTableFactory::NewTableReader()` now receives an already-opened `FSRandomAccessFile` and its file size instead of requiring implementations to reopen `file_path`. The supplied file preserves RocksDB's direct-I/O alignment handling, rate limiting, read statistics and listener notifications, prefetching, and synchronous, asynchronous, and batched reads.
* Added `PutRangeDeletionBlock()`, `GetRangeDeletionBlock()`, and `IsDeleteRangeSupported()` to the external table interfaces so formats can opt in to range deletions using RocksDB's raw range-deletion metadata block.
* Opaque `PreparedFileInfo` metadata can now be serialized to a checksummed representation with `SerializePreparedFileInfo()` and restored with `DeserializePreparedFileInfo()` for reuse across processes.
* Removed the deprecated `IngestExternalFileOptions::write_global_seqno` option and its C, Java, and `ldb` APIs, `TableProperties::external_sst_file_global_seqno_offset`, and the `file_offset` output from `ExternalTableReader::GetPropertiesBlock()`. Existing external SST files whose global sequence number was rewritten without updating the properties-block checksum remain readable.
* `TablePropertiesNames` has no apparent use in the public API so has been deprecated there.
* `TableProperties::creation_time` is renamed to `oldest_ancestor_time`, which better describes its meaning, along with `TablePropertiesNames::kCreationTime` to `kOldestAncestorTime`. The misspelled `SstFileMetaData::oldest_ancester_time` is renamed to `oldest_ancestor_time`. The old names remain available as deprecated aliases and will be removed in a future release.
* Added `WriteBufferFlushPolicy` and `WriteBufferManager` policy selection for flushing the oldest or largest column family, including configurable batched flushing of the largest column families across DBs sharing a manager, with C and Java bindings.
### Behavior Changes
* When `ExternalTableBuilderBase::PutPropertiesBlock()` returns `NotSupported`, RocksDB now overrides `num_entries`, `raw_key_size`, and `raw_value_size` from `ExternalTableBuilderBase::GetTableProperties()` with the logical totals tracked by the external-table adapter. Format-specific properties such as `data_size` remain unchanged.
* The `FSWritableFile` passed to `ExternalTableFactory::NewTableBuilder()` now reports logical file size including RocksDB-buffered data and exposes the underlying file's direct-I/O mode and required buffer alignment.
* `Iterator::Refresh()` now returns `NotSupported` on iterators from a secondary DB's `DB::NewIterators()` only, because those iterators are pinned to one cross-column-family read view. Iterators from `DB::NewIterator()` are unaffected; to advance a batch, call `DB::NewIterators()` again after `TryCatchUpWithPrimary()`.
* `SstFileReader` now reports standard SST read statistics and file I/O listener notifications, and honors the configured read rate limiter.
### Bug Fixes
* Fixed full recovery from retryable background errors to atomically advance WAL replay boundaries for all column families, preventing partial recovery of multi-column-family write batches after a failed recovery flush.
* Fixed RocksDB builds on big-endian architectures such as s390x by disabling the optional persistent fault-injection log, whose binary format is little-endian-only.
* Fixed `DB::ClipColumnFamily()` deleting a non-L0 SST file whose largest key equals `begin_key`, which lost keys at the start of the kept range.
* Fixed swapped `oldest_key_time` and `newest_key_time` values in the `table_file_creation` event log entry.
* Fixed `ExternalTableReader::VerifyChecksum()` results not being honored by the external table adapter.
* Fixed a bug in manifest handling that could lead to a hang or memory corruption in the case of LSM sanity checks being fatally violated (exceptional; see `force_consistency_checks`).
* Fixed `MultiGet()` with `ReadOptions::read_tier == kBlockCacheTier` reading data blocks from disk when `allow_mmap_reads` is false. Keys whose data block is not in the block cache now return `Status::Incomplete()`, the same as `Get()`.
* Fixed trace replay (`DB::NewDefaultReplayer()`, db_bench `replay`) failing every read record (Get, MultiGet, iterator seek) on the default column family with `Corruption: Invalid Column Family ID` when the default column family handle was not passed, which is always the case in db_bench with `--num_column_families=1`.
* Fixed multi-threaded `Replayer::Replay()` ignoring errors from executing trace records and returning OK. It now stops and returns the error, including `Status::NotSupported()` from executing a record, as single-threaded replay does.
* Fixed secondary DB `NewIterators()` returning iterators from different database states or blocking for `TryCatchUpWithPrimary()` by serving the last completed catch-up view.
### Performance Improvements
* Single Get coroutine reads using non-mmap, non-direct I/O on a `kFSBuffer`-capable `FileSystem` now use `FSRandomAccessFile::SubmitReadAsync()` instead of falling back to synchronous `Read()`. When rate-limited, these reads acquire their full token budget before submission to preserve a single filesystem-owned buffer, which can make physical I/O burstier than reads using caller-provided buffers.
## 11.11.0 (09/21/2026)
### New Features
* External tables now support DB-assigned global sequence numbers when `IngestExternalFileOptions::allow_global_seqno` is enabled, allowing overlapping external tables to be ingested across multiple levels when compactions are disabled. This includes the deprecated `write_global_seqno` path when the reader supplies the properties block's file offset.
* Added checksum handoff support for SST file writes when `BlockBasedTableOptions::checksum` is `kCRC32c`, allowing block and trailer CRC32C checksums to be passed from the table builder to the file writer for buffered-data validation and filesystem in-flight write protection.
* Enhanced `sst_dump --command=recompress` to be more powerful for benchmarking, including through automation with a new `--json` option for machine-readable output.
### Public API Changes
* Changed the experimental async read statistics API by removing `DB::AsyncCallback::EnableStats()` and changing `OnComplete(const PerfContext*, const IOStatsContext*)` to `OnComplete()`. Callers must configure TLS statistics before each async or coroutine read. `DB::GetAsync()` and `DB::MultiGetAsync()` callbacks must copy per-operation statistics from TLS during `OnComplete()`; coroutine users must copy them from inside a `CoroStackableDB` hook before suspending again. Each operation leaves TLS statistics collection disabled.
* Added `Compressor::MaybeOverrideCompressionOptions()` (EXPERIMENTAL), an omnibus hook letting a custom compressor adjust the subset of `CompressionOptions` that the SST builder applies rather than the compressor itself: `parallel_threads`, `max_compressed_bytes_per_kb`, `auto_skip`, and `auto_skip_min_sample_every` (and potentially more in the future). This deprecates `Compressor::GetRecommendedParallelThreads()`, which it subsumes; the default implementation still honors an overridden `GetRecommendedParallelThreads()` for compatibility.
* Added experimental `CoroDB::CoGet()` and `CoroDB::CoMultiGet()` overloads for point reads from an `SstFileReader`.
* Added experimental `DB::GetWithMetadata()` and `DB::MultiGetWithMetadata()` APIs with opt-in output metadata for key timestamps and explicit-snapshot newer-version detection.
* Added a `DB::NewIterators()` overload that accepts parallel vectors of `ReadOptions` and column family handles, allowing each returned iterator to use independent read settings while sharing a consistent database state. The vectors must have the same size, all snapshots must be null or identical, and all `tailing` settings must match.
* Added `ReadOptions::custom_context`, an opaque, non-owning pointer passed through table reads so external table implementations can consume application-defined read options.
### Behavior Changes
* `ExternalTableReader::MultiGet()` is now used by RocksDB's `DB::MultiGet()` path. Previously, the external table adapter used `TableReader`'s fallback, which called `ExternalTableReader::Get()` separately for each key.
* `DB::OpenAndCompact()` now requires `output_directory` to be a simple name for a new subdirectory when the new `use_session_tmp_dir_for_remote_compaction` DB option is enabled. The option places that subdirectory under the canonical `<dbname>/session_tmp/` root, and `DB::Open()` deletes stale contents under the root after recovery. The disabled default preserves the existing `<dbname>/<output_directory>` layout without automatic cleanup. Previously, remote-compaction output directories were created directly under `<dbname>`, which made it challenging to safely identify and delete obsolete files.
* Several block-based SST table properties now better support compression benchmarking and optimization: `uncompressed_data_size` is now persisted in the file (previously it was computed only at write time and read back as 0); a new `lsm_info_at_creation` property records the LSM level and "bottommost" status at the file's creation, distinguishing "not applicable" (e.g. SstFileWriter files) from "unknown" (older files); and the `compression_options` property now records the full compression manager configuration instead of only its name.
### Bug Fixes
* Fixed a usually-minor performance bug where `Cache::ApplyToAllEntries()` used single-entry batches instead of the intended 256-entry default for `ApplyToAllEntriesOptions::average_entries_per_lock`. All of the Cache implementations (LRU and HCC) benefit from the fix, but the disfavored LRU cache was especially impacted by repeated mutex acquisition and release.
* Fixed stats-disabled coroutine and asynchronous reads inheriting enabled TLS statistics state when interleaved with stats-enabled reads on the same executor. Each read now uses its captured per-call configuration, while enabled request counters remain isolated across suspensions.
* Fixed a bug where a remote compaction (`CompactionService` / `DB::OpenAndCompact`) requested with `OpenAndCompactOptions::allow_resumption=true` but for which resumption was internally disabled (incompatible with `paranoid_file_checks` or output-iteration verification) could fail a retried attempt, because the output directory was not cleaned before starting fresh and output files left by a previously interrupted attempt collided with the reused file numbers.
* Fixed a bug where a remote compaction (`CompactionService` / `DB::OpenAndCompact`) could fail when the primary rotated its MANIFEST while the worker was opening the primary's directory read-only. The open reads `CURRENT`, and the primary publishes a new MANIFEST by renaming a freshly written `CURRENT` over the old one; on file systems without POSIX read-after-unlink semantics (remote/object stores) the replaced object is then reported as gone (`PathNotFound`/`NotFound`, or `TryAgain` for the MANIFEST equivalent) instead of remaining readable. `DB::OpenAndCompact()` now retries the secondary open a bounded, configurable number of times for those transient statuses, so a concurrent rotation no longer fails the compaction (which, for a `CompactionService` that does not fall back to a local compaction, became a background error on the primary).
* Fixed secondary DB `NewIterators()` returning iterators from different database states when `TryCatchUpWithPrimary()` runs concurrently.
* Fixed a bug where a `SeekToFirst()` on an iterator previously used for a `Seek()` could end the scan early and silently omit keys with `ReadOptions::async_io`, or otherwise run with readahead effectively disabled. Affects scans using `ReadOptions::prefix_same_as_start` together with `ReadOptions::auto_readahead_size` (enabled by default); memory-mapped reads are unaffected.
### Performance Improvements
* `DB::MultiGetEntityLazy()` now reads all keys of a batch under a single shared per-column-family SuperVersion pin, and a single `LazyWideColumnsBatch::MultiResolve()` coalesces the resulting blob reads across keys: one read per blob file (and one per SST for embedded blobs), for whole-column and byte-range reads alike, instead of reading each key's blobs separately.
## 11.10.0 (08/28/2026)
### New Features
* Added `BlockBasedTableOptions::optimize_key_common_prefix` as part of new `format_version=8`. When enabled, each data block and index block stores the common user-key prefix of its keys once instead of at every restart point, reducing block size and block-cache footprint for prefix-heavy data and speeding up Seek for bytewise-ordered keys.
* A new experimental feature spreads out periodic (time-based) compaction trigger times across a fleet of similarly-configured DBs to avoid recurring "thundering herd" triggers, especially with remote compaction. See the new experimental DB option `periodic_compaction_phase_recovery_percent`.
### Public API Changes
* `IngestExternalFileArg::atomic_replace_range` now supports a range that only partially overlaps existing files, instead of failing. Requires universal compaction and a table factory that supports range deletion; other configurations still fail on partial overlap. The range is also now specified as half-open (`start` inclusive, `limit` exclusive), resolving a previously documented ambiguity in the upper bound.
* Added an immutable `external_table_config` option that is persisted in OPTIONS files and passed to external table factories during bootstrap.
### Bug Fixes
* Fixed a liveness bug where RocksDB could remain indefinitely in recovery from a soft background error when concurrent writes continuously generated normal flush requests.
* Fixed a race in online error recovery where a concurrent higher-severity I/O error could leave a stale retry-flush plan active, prevent the required WAL switch, and create duplicate internal WAL history detected on DB reopen.
* Fix "bottommost files" infinite compaction loops (`kBottommostFiles`). This happens with the unsupported combination of leveled compaction and `preserve_internal_time_seconds` / `preclude_last_level_data_seconds`, and in an edge case with user-defined timestamps when `full_history_ts_low` is unset. The underlying logic is now more unified to reduce risk of a similar divergence in logic in the future.
* Fixed `reuse_manifest_on_open` to create a fresh MANIFEST instead of appending after recovery sees an incomplete atomic group at the MANIFEST tail.
* Starting remote compaction (`CompactionService` / `DB::OpenAndCompact`) is now more efficient and more reliable, by avoiding unnecessary file operations (latency) while using more accurate consistency / precondition checks, both avoiding some spurious failures and rejecting more legitimate inconsistencies. New DBOption `remote_compaction_manifest_floor` can be disabled in case of unexpected problems with the change.
* Fixed unbounded retention of recovered transaction state by a long-lived secondary opened with `allow_2pc`.
* Fixed stale reads on secondary instances, where entries replayed from a WAL that the primary had already flushed could shadow the newer values in the flushed files indefinitely. Also fixed a case where a secondary could stop returning such a key altogether while the primary's flushed file holding it was unreadable.
* Fixed subset checkpoints to discard tolerated MANIFEST tail corruption before appending column-family drop records.
### Performance Improvements
* Improved compaction CPU efficiency for column families using an `SstPartitioner` (such as the built-in `SstPartitionerFixedPrefix`): when the partitioner can express its next boundary as a key prefix via the new `SstPartitioner::ShouldPartitionByPrefix()` API, compaction now checks that cached prefix inline instead of copying every user key and making a virtual `ShouldPartition()` call for each key-value written to an SST.
## 11.9.0 (08/14/2026)
### New Features
* Added an integrated and improved compression auto-skip feature (new CompressionOptions, still experimental) that saves CPU by skipping compression when and where the overall ratio is inadequate. This improves on the previous auto-skip compression manager by (a) taking achieved compression ratios into account, (b) using a continuously evolving payoff estimate, and (c) combining natively with parallel compression.
* Added the `blob_file_writable_file_max_buffer_size` mutable column family option. When set to a non-zero value, blob-file writers use it as their `WritableFileWriter` max buffer size instead of inheriting `DBOptions::writable_file_max_buffer_size`.
* Added an experimental lazy wide-column read API, `DB::GetEntityLazy()` / `DB::MultiGetEntityLazy()` (see `include/rocksdb/lazy_wide_columns.h`). The returned `LazyWideColumns` leaves blob-backed columns as unresolved references and, like an iterator, stays valid after the call returns, so the caller resolves only the columns -- and only the byte ranges -- it needs; a byte-range read of an uncompressed blob (separate-file or embedded) reads just those bytes from storage. Adds `Env::IOActivity::kLazyResolve` and `rocksdb.blobdb.lazy.*` statistics. Requires `max_open_files == -1`.
* Added DB property `rocksdb.num-running-remote-compactions`, reporting the number of compaction service jobs currently waiting in `CompactionService::Wait()`.
* Added DB property `rocksdb.num-unscheduled-compactions`, reporting compactions waiting in the DB's internal compaction queue but not yet assigned to a background job.
* Added `DB::AbortCompactions()` and `DB::ResumeCompactions()` for per-column-family compaction abort and resume, with the same active-abort semantics as the DB-wide `AbortAllCompactions()`/`ResumeAllCompactions()` but scoped to a single column family. Calls are reference-counted independently per column family and compose with the DB-wide abort.
### Public API Changes
* Add IndexFactory compatibility names for the existing user-defined index API.
* Added the experimental `CoroDB` capability, exposed by `DB::GetCoroDB()`, with lazy native coroutine variants of `Get` and `MultiGet`. The existing `DB::GetAsync()` and `DB::MultiGetAsync()` APIs dispatch through it.
* Added `rocksdb_batched_multi_get_pinned_cf()` to return batch-owned pinned MultiGet values and per-key errors without allocating one wrapper per found value.
* Add `DBOptions::read_io_executor_threads` to increase the maximum thread count of the shared filesystem read I/O executor used for asynchronous reads when opening a DB. Opening a DB never reduces the shared executor thread count. Change the experimental `FSRandomAccessFile::SubmitReadAsync()` API to return whether the asynchronous path was used, and add the `FILE_SUBMIT_ASYNC_READ_FALLBACK` ticker to count calls that fall back to synchronous reads.
* `ReadOptions::table_filter` is now a *pointer* to a caller-owned `std::function` instead of a `std::function` value, to improve efficiency in copying and destroying `ReadOptions` objects (which RocksDB does internally). This is a breaking change for users of this relatively obscure option, who will now need to declare a `std::function` that outlives the read. As a side effect of `ReadOptions` being trivially copyable, new but legitimate compiler warnings may surface.
* Removed the experimental `CreateAutoSkipCompressionManager()`. Auto-skip compression is now a first-class feature of the block-based table builder, controlled by the new experimental `CompressionOptions::auto_skip`.
* Added `CoroStackableDB` for implementing wrappers over native coroutine reads and used it to add coroutine `Get` and `MultiGet` support to `DBWithTTL` and WritePrepared `TransactionDB`. A stackable wrapper advertises coroutine support only when its wrapped DB supports it; TTL value decoding and transaction visibility semantics are preserved.
* Added a `kSstFileWriter` value to `TableFileCreationReason`, which `SstFileWriter` now reports instead of the catch-all `kMisc`. This, along with new `db_name` and `is_remote_compaction` fields in `FilterBuildingContext`, improves the contextual information available to custom filter and compression strategies.
* Renamed `TransactionLogIterator` to `WalIterator` (and `include/rocksdb/transaction_log.h` to `include/rocksdb/wal_iterator.h`), because it iterates the write-ahead log and is unrelated to `TransactionDB`. The old names still work. `DB::GetUpdatesSince()` is unchanged. In Java, `WalIterator` is the new name and `TransactionLogIterator` remains as a subclass. Also documented existing behavior that was previously unstated: only writes that reached the WAL are visible (so `WriteOptions::disableWAL` and `IngestExternalFile()` leave holes that end iteration), what each iterator state means, and that `GetUpdatesSince()` silently starts later when the requested sequence number is unavailable.
### Bug Fixes
* Fixed a bug where `BlobFileBuilder` passed an SST path to `OnBlobFileCompleted()` instead of the completed blob path during BlobDB garbage-collection compactions, causing incorrect `SstFileManager` accounting and compaction failures on filesystems where the SST was not yet pathname-visible.
### Performance Improvements
* When `DBOptions::avoid_unnecessary_blocking_io` is true, obsolete `OPTIONS-*` files found during DB open are deleted by background purge instead of synchronously on the opening thread.
## 11.8.0 (07/28/2026)
### Public API Changes
* Add callback-based asynchronous read APIs, `DB::GetAsync()` and `DB::MultiGetAsync()`. The idea is that when IO is required, RocksDB can suspend its internal coroutine read path, allowing the read executor thread to do other work. When the IO is complete, RocksDB invokes the user callback. This requires filesystem support for full performance benefits. A new filesystem API `FSRandomAccessFile::SubmitReadAsync()` is introduced for this. Unlike `ReadAsync`, the filesystem is responsible for eventually calling the callback. `FileSystem::GetReadExecutor()` returns the IO executor whose EventBases run coroutine read processing. In the Posix filesystem, an IO-uring based event loop is used to complete the IO.
* Added `FlushOptions::listener_wait` (default `false`). When set together with `FlushOptions::wait == true`, `DB::Flush()` will not return until the registered `EventListener::OnFlushCompleted` callbacks for the flushed memtables have finished running. By default (`false`), `Flush(wait=true)` may return as soon as the flush result is committed, which can be before (or while) the `OnFlushCompleted` callbacks execute on the background flush thread. Also added the corresponding C API `rocksdb_flushoptions_set_listener_wait()` / `rocksdb_flushoptions_get_listener_wait()`.
* Deprecated `PinnableWideColumns::serialized_size()` in favor of the new `PinnableWideColumns::payload_size()`, which returns the total size of the columns' names and values (for a plain value this equals the value size). Relatedly, wide-column read accounting changed subtly: read statistics for entities (`BYTES_READ`, `BYTES_PER_READ`, and the per-key byte counts for `GetEntity`/`MultiGetEntity`) and the `ReadOptions::value_size_soft_limit` threshold now measure entities by this payload size instead of the serialized entity size. This drops serialization framing bytes from those measurements (so reported/limited sizes for entity reads are slightly smaller than before) and makes plain-value and wide-column accounting consistent.
### Behavior Changes
* `ReadOptions::value_size_soft_limit` now also bounds the blob values resolved from pre-flush blob direct write references during `MultiGet`/`MultiGetEntity` (previously these were resolved after the limit check and read without bound). Enforcement is "always make progress": at least one key is read even if its value alone exceeds the limit, and only once the returned size exceeds the limit do subsequent keys get `Status::Aborted` -- so a caller retrying the aborted keys cannot loop forever on a single value that by itself exceeds the limit.
### Performance Improvements
* Reduced copying when reading wide-column entities that have blob-referenced columns, whether stored in separate blob files or embedded (same-file) as blob records in the same SST. On the point-lookup (`GetEntity`/`MultiGetEntity`) and blob-backed memtable/direct-write read paths, resolving these columns no longer re-serializes the whole entity into a fresh buffer: inline columns are referenced in place in the pinned entity and each resolved blob value is referenced in place from its fetched buffer. This is enabled by an internal change to `PinnableWideColumns` that lets its columns reference multiple backing buffers.
## 11.7.0 (07/16/2026)
### Public API Changes
* Added `RangeLockManagerHandle::SetIsKilledCallback()`, a `std::function<bool()>` predicate polled during range-lock waits. When it returns true, a blocked `GetRangeLock()` wait returns promptly with `Status::Aborted()` instead of waiting for the lock timeout. Range-lock users that install no callback keep the original wait-until-grant-or-timeout behavior.
* Added `MultiScanArgs::reverse` to support reverse MultiScan reads over bounded scan ranges.
### Behavior Changes
* When using `LockWAL()`, live-file capture APIs such as checkpoint and backup now flush WAL-disabled unpersisted data; they can block until `UnlockWAL()` or return `Aborted` when called from the same thread that holds `LockWAL()` to avoid deadlock. `UnlockWAL()` must now also be called from the same thread that called `LockWAL()`.
### Performance Improvements
* Fixed slow DB open with `best_efforts_recovery`, and slow catch-up of secondary and follower (read-only) instances, in which time and memory grew quadratically with the number of files in the DB (so the slowdown was worst for large DBs with long history).
## 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.
export CC CXX AR AS CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE CLANG_ANALYZER CLANG_SCAN_BUILD
export CC CXX AR AS CFLAGS CXXFLAGS EXEC_LDFLAGS EXEC_LDFLAGS_SHARED VALGRIND_VER JEMALLOC_LIB JEMALLOC_INCLUDE CLANG_CXX CLANG_ANALYZER CLANG_SCAN_BUILD
Loaded 100 of 834 files, more 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.
Xingbo Wang
cm-wang
Josh Kang
meta-rocksdb-release-automation[bot]
Andrew Chang
Anand Ananthabhotla
Jay Huh
generatedunixname3846135475516776
Peter Dillinger
Ranjan Banerjee
Yoshinori Matsunobu
Davis Rollman
xingbowang
Omkar Gawde
Steph Pontikes
Andrei Preda
zaidoon
Denis Yaroshevskiy
rocksdb-release-bot
Kunal Pathak
Michael C Huang
Darkhan Imangaliyev
anand1976
generatedunixname1395027625275998
Anusha Mukka
Hui Xiao