Summary:
* Inefficient block-based filter is no longer customizable in the public
API, though (for now) can still be enabled.
* Removed deprecated FilterPolicy::CreateFilter() and
FilterPolicy::KeyMayMatch()
* Removed `rocksdb_filterpolicy_create()` from C API
* Change meaning of nullptr return from GetBuilderWithContext() from "use
block-based filter" to "generate no filter in this case." This is a
cleaner solution to the proposal in https://github.com/facebook/rocksdb/issues/8250.
* Also, when user specifies bits_per_key < 0.5, we now round this down
to "no filter" because we expect a filter with >= 80% FP rate is
unlikely to be worth the CPU cost of accessing it (esp with
cache_index_and_filter_blocks=1 or partition_filters=1).
* bits_per_key >= 0.5 and < 1.0 is still rounded up to 1.0 (for 62% FP
rate)
* This also gives us some support for configuring filters from OPTIONS
file as currently saved: `filter_policy=rocksdb.BuiltinBloomFilter`.
Opening from such an options file will enable reading filters (an
improvement) but not writing new ones. (See Customizable follow-up
below.)
* Also removed deprecated functions
* FilterBitsBuilder::CalculateNumEntry()
* FilterPolicy::GetFilterBitsBuilder()
* NewExperimentalRibbonFilterPolicy()
* Remove default implementations of
* FilterBitsBuilder::EstimateEntriesAdded()
* FilterBitsBuilder::ApproximateNumEntries()
* FilterPolicy::GetBuilderWithContext()
* Remove support for "filter_policy=experimental_ribbon" configuration
string.
* Allow "filter_policy=bloomfilter:n" without bool to discourage use of
block-based filter.
Some pieces for https://github.com/facebook/rocksdb/issues/9389
Likely follow-up (later PRs):
* Refactoring toward FilterPolicy Customizable, so that we can generate
filters with same configuration as before when configuring from options
file.
* Remove support for user enabling block-based filter (ignore `bool
use_block_based_builder`)
* Some months after this change, we could even remove read support for
block-based filter, because it is not critical to DB data
preservation.
* Make FilterBitsBuilder::FinishV2 to avoid `using
FilterBitsBuilder::Finish` mess and add support for specifying a
MemoryAllocator (for cache warming)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9501
Test Plan:
A number of obsolete tests deleted and new tests or test
cases added or updated.
Reviewed By: hx235
Differential Revision: D34008011
Pulled By: pdillinger
fbshipit-source-id: a39a720457c354e00d5b59166b686f7f59e392aa
Summary:
... seen only in internal clang-analyze runs after https://github.com/facebook/rocksdb/issues/9481
* Mostly, this works around falsely reported leaks by using
std::unique_ptr in some places where clang-analyze was getting
confused. (I didn't see any changes in C++17 that could make our Status
implementation leak memory.)
* Also fixed SetBGError returning address of a stack variable.
* Also fixed another false null deref report by adding an assert.
Also, use SKIP_LINK=1 to speed up `make analyze`
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9515
Test Plan:
Was able to reproduce the reported errors locally and verify
they're fixed (except SetBGError). Otherwise, existing tests
Reviewed By: hx235
Differential Revision: D34054630
Pulled By: pdillinger
fbshipit-source-id: 38600ef3da75ddca307dff96b7a1a523c2885c2e
Summary:
The patch builds on the refactoring done in https://github.com/facebook/rocksdb/issues/9494
and improves the performance of building the hash of file
locations in `VersionStorageInfo` in two ways. First, the hash
building is moved from `AddFile` (which is called under the DB mutex)
to a separate post-processing step done as part of `PrepareForVersionAppend`
(during which the mutex is *not* held). Second, the space necessary
for the hash is preallocated to prevent costly reallocation/rehashing
operations. These changes mitigate the overhead of the file location hash,
which can be significant with certain workloads where the baseline CPU usage
is low (see https://github.com/facebook/rocksdb/issues/9351,
which is a workload where keys are sorted, WAL is turned
off, the vector memtable implementation is used, and there are lots of small
SST files).
Fixes https://github.com/facebook/rocksdb/issues/9351
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9504
Test Plan:
`make check`
```
numactl --interleave=all ./db_bench --benchmarks=fillseq --allow_concurrent_memtable_write=false --level0_file_num_compaction_trigger=4 --level0_slowdown_writes_trigger=20 --level0_stop_writes_trigger=30 --max_background_jobs=8 --max_write_buffer_number=8 --db=/data/ltamasi-dbbench --wal_dir=/data/ltamasi-dbbench --num=800000000 --num_levels=8 --key_size=20 --value_size=400 --block_size=8192 --cache_size=51539607552 --cache_numshardbits=6 --compression_max_dict_bytes=0 --compression_ratio=0.5 --compression_type=lz4 --bytes_per_sync=8388608 --cache_index_and_filter_blocks=1 --cache_high_pri_pool_ratio=0.5 --benchmark_write_rate_limit=0 --write_buffer_size=16777216 --target_file_size_base=16777216 --max_bytes_for_level_base=67108864 --verify_checksum=1 --delete_obsolete_files_period_micros=62914560 --max_bytes_for_level_multiplier=8 --statistics=0 --stats_per_interval=1 --stats_interval_seconds=20 --histogram=1 --bloom_bits=10 --open_files=-1 --subcompactions=1 --compaction_style=0 --level_compaction_dynamic_level_bytes=true --pin_l0_filter_and_index_blocks_in_cache=1 --soft_pending_compaction_bytes_limit=167503724544 --hard_pending_compaction_bytes_limit=335007449088 --min_level_to_compress=0 --use_existing_db=0 --sync=0 --threads=1 --memtablerep=vector --disable_wal=1 --seed=<some_seed>
```
Final statistics before this patch:
```
Cumulative writes: 0 writes, 697M keys, 0 commit groups, 0.0 writes per commit group, ingest: 283.25 GB, 241.08 MB/s
Interval writes: 0 writes, 1264K keys, 0 commit groups, 0.0 writes per commit group, ingest: 525.69 MB, 176.67 MB/s
```
With the patch:
```
Cumulative writes: 0 writes, 759M keys, 0 commit groups, 0.0 writes per commit group, ingest: 308.57 GB, 262.63 MB/s
Interval writes: 0 writes, 1555K keys, 0 commit groups, 0.0 writes per commit group, ingest: 646.61 MB, 215.11 MB/s
```
Reviewed By: riversand963
Differential Revision: D34014734
Pulled By: ltamasi
fbshipit-source-id: acb2703677451d5ccaa7e9d950844b33d240695b
Summary:
Follow-up to https://github.com/facebook/rocksdb/issues/9126
Added new unit tests to validate some of the claims of guaranteed uniqueness
within certain large bounds.
Also cleaned up the cache_bench -stress-cache-key tool with better comments
and description.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9329
Test Plan: no changes to production code
Reviewed By: mrambacher
Differential Revision: D33269328
Pulled By: pdillinger
fbshipit-source-id: 3a2b684a6b2b15f79dc872e563e3d16563be26de
Summary:
The patch does some cleanup in and around `VersionStorageInfo`:
* Renames the method `PrepareApply` to `PrepareAppend` in `Version`
to make it clear that it is to be called before appending the `Version` to
`VersionSet` (via `AppendVersion`), not before applying any `VersionEdit`s.
* Introduces a helper method `VersionStorageInfo::PrepareForVersionAppend`
(called by `Version::PrepareAppend`) that encapsulates the population of the
various derived data structures in `VersionStorageInfo`, and turns the
methods computing the derived structures (`UpdateNumNonEmptyLevels`,
`CalculateBaseBytes` etc.) into private helpers.
* Changes `Version::PrepareAppend` so it only calls `UpdateAccumulatedStats`
if the `update_stats` flag is set. (Earlier, this was checked by the callee.)
Related to this, it also moves the call to `ComputeCompensatedSizes` to
`VersionStorageInfo::PrepareForVersionAppend`.
* Updates and cleans up `version_builder_test`, `version_set_test`, and
`compaction_picker_test` so `PrepareForVersionAppend` is called anytime
a new `VersionStorageInfo` is set up or saved. This cleanup also involves
splitting `VersionStorageInfoTest.MaxBytesForLevelDynamic`
into multiple smaller test cases.
* Fixes up a bunch of comments that were outdated or just plain incorrect.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9494
Test Plan: Ran `make check` and the crash test script for a while.
Reviewed By: riversand963
Differential Revision: D33971666
Pulled By: ltamasi
fbshipit-source-id: fda52faac7783041126e4f8dec0fe01bdcadf65a
Summary:
In RocksDB, this option was already marked as "NOT SUPPORTED" for a long time, and setting this option does not have any effect on the behavior of RocksDB library. Therefore, we are removing it in the preparations of the upcoming 7.0 release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9446
Reviewed By: ajkr
Differential Revision: D33793048
Pulled By: bjlemaire
fbshipit-source-id: 73316efdb194e90225005246673dae99e65577ae
Summary:
I feel it would be nice if we can fix this spelling error.
In `SizeApproximationOptions`, the `include_memtabtles` should be `include_memtables`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9490
Test Plan: make check
Reviewed By: hx235
Differential Revision: D33949862
Pulled By: riversand963
fbshipit-source-id: b2be67501b65d4aabb6b8df1bf25eb8d54cc1466
Summary:
ajkr reminded me that we have a rule of not including per-kv related data in `WriteOptions`.
Namely, `WriteOptions` should not include information about "what-to-write", but should just
include information about "how-to-write".
According to this rule, `WriteOptions::timestamp` (experimental) is clearly a violation. Therefore,
this PR removes `WriteOptions::timestamp` for compliance.
After the removal, we need to pass timestamp info via another set of APIs. This PR proposes a set
of overloaded functions `Put(write_opts, key, value, ts)`, `Delete(write_opts, key, ts)`, and
`SingleDelete(write_opts, key, ts)`. Planned to add `Write(write_opts, batch, ts)`, but its complexity
made me reconsider doing it in another PR (maybe).
For better checking and returning error early, we also add a new set of APIs to `WriteBatch` that take
extra `timestamp` information when writing to `WriteBatch`es.
These set of APIs in `WriteBatchWithIndex` are currently not supported, and are on our TODO list.
Removed `WriteBatch::AssignTimestamps()` and renamed `WriteBatch::AssignTimestamp()` to
`WriteBatch::UpdateTimestamps()` since this method require that all keys have space for timestamps
allocated already and multiple timestamps can be updated.
The constructor of `WriteBatch` now takes a fourth argument `default_cf_ts_sz` which is the timestamp
size of the default column family. This will be used to allocate space when calling APIs that do not
specify a column family handle.
Also, updated `DB::Get()`, `DB::MultiGet()`, `DB::NewIterator()`, `DB::NewIterators()` methods, replacing
some assertions about timestamp to returning Status code.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8946
Test Plan:
make check
./db_bench -benchmarks=fillseq,fillrandom,readrandom,readseq,deleterandom -user_timestamp_size=8
./db_stress --user_timestamp_size=8 -nooverwritepercent=0 -test_secondary=0 -secondary_catch_up_one_in=0 -continuous_verification_interval=0
Make sure there is no perf regression by running the following
```
./db_bench_opt -db=/dev/shm/rocksdb -use_existing_db=0 -level0_stop_writes_trigger=256 -level0_slowdown_writes_trigger=256 -level0_file_num_compaction_trigger=256 -disable_wal=1 -duration=10 -benchmarks=fillrandom
```
Before this PR
```
DB path: [/dev/shm/rocksdb]
fillrandom : 1.831 micros/op 546235 ops/sec; 60.4 MB/s
```
After this PR
```
DB path: [/dev/shm/rocksdb]
fillrandom : 1.820 micros/op 549404 ops/sec; 60.8 MB/s
```
Reviewed By: ltamasi
Differential Revision: D33721359
Pulled By: riversand963
fbshipit-source-id: c131561534272c120ffb80711d42748d21badf09
Summary:
Note: rebase on and merge after https://github.com/facebook/rocksdb/pull/9349, https://github.com/facebook/rocksdb/pull/9345, (optional) https://github.com/facebook/rocksdb/pull/9393
**Context:**
(Quoted from pdillinger) Layers of information during new Bloom/Ribbon Filter construction in building block-based tables includes the following:
a) set of keys to add to filter
b) set of hashes to add to filter (64-bit hash applied to each key)
c) set of Bloom indices to set in filter, with duplicates
d) set of Bloom indices to set in filter, deduplicated
e) final filter and its checksum
This PR aims to detect corruption (e.g, unexpected hardware/software corruption on data structures residing in the memory for a long time) from b) to e) and leave a) as future works for application level.
- b)'s corruption is detected by verifying the xor checksum of the hash entries calculated as the entries accumulate before being added to the filter. (i.e, `XXPH3FilterBitsBuilder::MaybeVerifyHashEntriesChecksum()`)
- c) - e)'s corruption is detected by verifying the hash entries indeed exists in the constructed filter by re-querying these hash entries in the filter (i.e, `FilterBitsBuilder::MaybePostVerify()`) after computing the block checksum (except for PartitionFilter, which is done right after each `FilterBitsBuilder::Finish` for impl simplicity - see code comment for more). For this stage of detection, we assume hash entries are not corrupted after checking on b) since the time interval from b) to c) is relatively short IMO.
Option to enable this feature of detection is `BlockBasedTableOptions::detect_filter_construct_corruption` which is false by default.
**Summary:**
- Implemented new functions `XXPH3FilterBitsBuilder::MaybeVerifyHashEntriesChecksum()` and `FilterBitsBuilder::MaybePostVerify()`
- Ensured hash entries, final filter and banding and their [cache reservation ](https://github.com/facebook/rocksdb/issues/9073) are released properly despite corruption
- See [Filter.construction.artifacts.release.point.pdf ](https://github.com/facebook/rocksdb/files/7923487/Design.Filter.construction.artifacts.release.point.pdf) for high-level design
- Bundled and refactored hash entries's related artifact in XXPH3FilterBitsBuilder into `HashEntriesInfo` for better control on lifetime of these artifact during `SwapEntires`, `ResetEntries`
- Ensured RocksDB block-based table builder calls `FilterBitsBuilder::MaybePostVerify()` after constructing the filter by `FilterBitsBuilder::Finish()`
- When encountering such filter construction corruption, stop writing the filter content to files and mark such a block-based table building non-ok by storing the corruption status in the builder.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9342
Test Plan:
- Added new unit test `DBFilterConstructionCorruptionTestWithParam.DetectCorruption`
- Included this new feature in `DBFilterConstructionReserveMemoryTestWithParam.ReserveMemory` as this feature heavily touch ReserveMemory's impl
- For fallback case, I run `./filter_bench -impl=3 -detect_filter_construct_corruption=true -reserve_table_builder_memory=true -strict_capacity_limit=true -quick -runs 10 | grep 'Build avg'` to make sure nothing break.
- Added to `filter_bench`: increased filter construction time by **30%**, mostly by `MaybePostVerify()`
- FastLocalBloom
- Before change: `./filter_bench -impl=2 -quick -runs 10 | grep 'Build avg'`: **28.86643s**
- After change:
- `./filter_bench -impl=2 -detect_filter_construct_corruption=false -quick -runs 10 | grep 'Build avg'` (expect a tiny increase due to MaybePostVerify is always called regardless): **27.6644s (-4% perf improvement might be due to now we don't drop bloom hash entry in `AddAllEntries` along iteration but in bulk later, same with the bypassing-MaybePostVerify case below)**
- `./filter_bench -impl=2 -detect_filter_construct_corruption=true -quick -runs 10 | grep 'Build avg'` (expect acceptable increase): **34.41159s (+20%)**
- `./filter_bench -impl=2 -detect_filter_construct_corruption=true -quick -runs 10 | grep 'Build avg'` (by-passing MaybePostVerify, expect minor increase): **27.13431s (-6%)**
- Standard128Ribbon
- Before change: `./filter_bench -impl=3 -quick -runs 10 | grep 'Build avg'`: **122.5384s**
- After change:
- `./filter_bench -impl=3 -detect_filter_construct_corruption=false -quick -runs 10 | grep 'Build avg'` (expect a tiny increase due to MaybePostVerify is always called regardless - verified by removing MaybePostVerify under this case and found only +-1ns difference): **124.3588s (+2%)**
- `./filter_bench -impl=3 -detect_filter_construct_corruption=true -quick -runs 10 | grep 'Build avg'`(expect acceptable increase): **159.4946s (+30%)**
- `./filter_bench -impl=3 -detect_filter_construct_corruption=true -quick -runs 10 | grep 'Build avg'`(by-passing MaybePostVerify, expect minor increase) : **125.258s (+2%)**
- Added to `db_stress`: `make crash_test`, `./db_stress --detect_filter_construct_corruption=true`
- Manually smoke-tested: manually corrupted the filter construction in some db level tests with basic PUT and background flush. As expected, the error did get returned to users in subsequent PUT and Flush status.
Reviewed By: pdillinger
Differential Revision: D33746928
Pulled By: hx235
fbshipit-source-id: cb056426be5a7debc1cd16f23bc250f36a08ca57
Summary:
Crash test recently started showing failures as in https://github.com/facebook/rocksdb/issues/9118 but
for files created by compaction. This change applies a similar fix.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9480
Test Plan:
Updated / extended unit test. (Some re-arranging to do the
simpler compaction testing before this special case.)
Reviewed By: ltamasi
Differential Revision: D33909835
Pulled By: pdillinger
fbshipit-source-id: 58e4b44e4ecc2d21e4df2c2d8440ec0633aa1f6c
Summary:
* Add more micro-benchmark tests
* Expose an API in DBImpl for waiting for compactions (still not visible to the user)
* Add argument name for ribbon_bench
* remove benchmark run from CI, as it runs too long.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9436
Test Plan: CI
Reviewed By: riversand963
Differential Revision: D33777836
Pulled By: jay-zhuang
fbshipit-source-id: c05de3bc082cc05b5d019f00b324e774bf4bbd96
Summary:
Apparently setting total_order_seek=true for DB::Get was
intended to allow accurate read semantics if the current prefix
extractor doesn't match what was used to generate SST files on
disk. But since prefix_extractor was made a mutable option in 5.14.0, we
have been able to detect this case and provide the correct semantics
regardless of the total_order_seek option. Since that time, the option
has only made Get() slower in a reasonably common case: prefix_extractor
unchanged and whole_key_filtering=false.
So this change primarily removes unnecessary effect of
total_order_seek on Get. Also cleans up some related comments.
Also adds a -total_order_seek option to db_bench and canonicalizes
handling of ReadOptions in db_bench so that command line options have
the expected association with library features. (There is potential
for change in regression test behavior, but the old behavior is likely
indefensible, or some other inconsistency would need to be fixed.)
TODO in follow-up work: there should be no reason for Get() to depend on
current prefix extractor at all.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9427
Test Plan:
Unit tests updated.
Performance (using db_bench update)
Create DB with `TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=10000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=12 -whole_key_filtering=0`
Test with and without `-total_order_seek` on `TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -use_existing_db -readonly -benchmarks=readrandom -num=10000000 -duration=40 -disable_wal=1 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=12`
Before this change, total_order_seek=false: 25188 ops/sec
Before this change, total_order_seek=true: 1222 ops/sec (~20x slower)
After this change, total_order_seek=false: 24570 ops/sec
After this change, total_order_seek=true: 25012 ops/sec (indistinguishable)
Reviewed By: siying
Differential Revision: D33753458
Pulled By: pdillinger
fbshipit-source-id: bf892f34907a5e407d9c40bd4d42f0adbcbe0014
Summary:
**Context/Summary:**
AdvancedColumnFamilyOptions::rate_limit_delay_max_milliseconds has been marked as deprecated and it's time to actually remove the code.
- Keep `soft_rate_limit`/`hard_rate_limit` in `cf_mutable_options_type_info` to prevent throwing `InvalidArgument` in `GetColumnFamilyOptionsFromMap` when reading an option file still with these options (e.g, old option file generated from RocksDB before the deprecation)
- Keep `soft_rate_limit`/`hard_rate_limit` in under `OptionsOldApiTest.GetOptionsFromMapTest` to test the case mentioned above.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9455
Test Plan: Rely on my eyeball and CI
Reviewed By: ajkr
Differential Revision: D33811664
Pulled By: hx235
fbshipit-source-id: 866859427fe710354a90f1095057f80116365ff0
Summary:
In RocksDB DBOptions::skip_log_error_on_recovery is marked as
"NOT SUPPORTED" for a long time, and setting this option does not have
any effect on the behavior of RocksDB library. Therefore, we are removing it
in the upcoming 7.0 release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9434
Test Plan: CircleCI
Reviewed By: ajkr
Differential Revision: D33763015
Pulled By: akankshamahajan15
fbshipit-source-id: 11f09643298da6c02d3dcdb090b996f4c3cfdd76
Summary:
The API is deprecated long time ago. Clean up the codebase by
removing it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9462
Test Plan: CI, fake release: D33835220
Reviewed By: riversand963
Differential Revision: D33835103
Pulled By: jay-zhuang
fbshipit-source-id: 6d2dc12c8e7fdbe2700865a3e61f0e3f78bd8184
Summary:
Disallow `immutable_db_opts.use_direct_io_for_flush_and_compaction == true` and
`mutable_db_opts.writable_file_max_buffer_size == 0`, since it causes `WritableFileWriter::Append()`
to loop forever and does not make much sense in direct IO.
This combination of options itself does not make much sense: asking RocksDB to do direct IO but not allowing
RocksDB to allocate a buffer. We should detect this false combination and warn user early, no matter whether
the application is running on a platform that supports direct IO or not. In the case of platform **not** supporting
direct IO, it's ok if the user learns about this and then finds that direct IO is not supported.
One tricky thing: the constructor of `WritableFileWriter` is being used in our unit tests, and it's impossible
to return status code from constructor. Since we do not throw, I put an assertion for now. Fortunately,
the constructor is not exposed to external applications.
Closing https://github.com/facebook/rocksdb/issues/7109
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9348
Test Plan: make check
Reviewed By: ajkr
Differential Revision: D33371924
Pulled By: riversand963
fbshipit-source-id: 2a3701ab541cee23bffda8a36cdf37b2d235edfa
Summary:
This also removes the obsolete names BackupableDBOptions
and UtilityDB. API users must now use BackupEngineOptions and
DBWithTTL::Open. In C API, `rocksdb_backupable_db_*` is replaced
`rocksdb_backup_engine_*`. Similar renaming in Java API.
In reference to https://github.com/facebook/rocksdb/issues/9389
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9438
Test Plan: CI
Reviewed By: mrambacher
Differential Revision: D33780269
Pulled By: pdillinger
fbshipit-source-id: 4a6cfc5c1b4c78bcad790b9d3dd13c5fdf4a1fac
Summary:
MemTable::MultiGet was not considering range tombstones before
querying Bloom filter. This means range tombstones would be skipped for
keys (or prefixes) with no other entries in the memtable. This could cause
old values for a key (in SST files) to still show up until the range tombstone
covering it has been flushed.
This is fixed by essentially disabling the memtable Bloom filter when there
are any range tombstones. (This could be better optimized in the future, but
good enough for now.)
Did some other cleanup/optimization in the same code to (more than) offset
the cost of checking on range tombstones in more cases. There is now
notable improvement when memtable_whole_key_filtering and prefix_extractor
are used together (unusual), and this makes MultiGet closer to the Get
implementation.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9453
Test Plan:
new unit test added. Added memtable Bloom to crash test.
Performance testing
--------------------
Build WAL-only DB (recovers to memtable):
```
TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=1000000 -write_buffer_size=250000000
```
Query test command, to maximize sensitivity to the changed code:
```
TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -use_existing_db -readonly -benchmarks=multireadrandom -num=10000000 -write_buffer_size=250000000 -memtable_bloom_size_ratio=0.015 -multiread_batched -batch_size=24 -threads=8 -memtable_whole_key_filtering=$MWKF -prefix_size=$PXS
```
(Note -num here is 10x larger for mostly memtable misses)
Before & after run simultaneously, average over 10 iterations per data point, ops/sec.
MWKF=0 PXS=0 (Bloom disabled)
Before: 5724844
After: 6722066
MWKF=0 PXS=7 (prefixes hardly unique; Bloom not useful)
Before: 9981319
After: 10237990
MWKF=0 PXS=8 (prefixes unique; Bloom useful)
Before: 12081715
After: 12117603
MWKF=1 PXS=0 (whole key Bloom useful)
Before: 11944354
After: 12096085
MWKF=1 PXS=7 (whole key Bloom useful in new version; prefixes not useful in old version)
Before: 9444299
After: 11826029
MWKF=1 PXS=7 (whole key Bloom useful in new version; prefixes useful in old version)
Before: 11784465
After: 11778591
Only in this last case is the 'before' *slightly* faster, perhaps because hashing prefixes is slightly faster than hashing whole keys. Otherwise, 'after' is faster.
Reviewed By: ajkr
Differential Revision: D33805025
Pulled By: pdillinger
fbshipit-source-id: 597523cae4f4eafdf6ae6bb2bc6cb46f83b017bf
Summary:
**Context/Summary:**
AdvancedColumnFamilyOptions::soft_rate_limit/hard_rate_limit have been marked as deprecated and it's time to actually remove the code.
- Keep `soft_rate_limit`/`hard_rate_limit` in `cf_mutable_options_type_info` to prevent throwing `InvalidArgument` in `GetColumnFamilyOptionsFromMap` when reading an option file still with these options (e.g, old option file generated from RocksDB before the deprecation)
- Keep `soft_rate_limit`/`hard_rate_limit` in under `OptionsOldApiTest.GetOptionsFromMapTest` to test the case mentioned above.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9452
Test Plan: Rely on my eyeball and CI
Reviewed By: ajkr
Differential Revision: D33804938
Pulled By: hx235
fbshipit-source-id: 133d49f7ec5238d7efceeb0a3122a5792a2b9945
Summary:
In RocksDB, this option was already marked as "NOT SUPPORTED" for a long time, and setting this option does not have any effect on the behavior of RocksDB library. Therefore, we are removing it in the preparations of the upcoming 7.0 release.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9450
Reviewed By: ajkr
Differential Revision: D33802466
Pulled By: bjlemaire
fbshipit-source-id: 97570985f1400525304053476450f7ef504c0cd5
Summary:
Add an option to set the WAL compression algorithm - wal_compression.
TODO: WAL compression is not implemented and will only support zstd initially. Will be added in subsequent diffs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9432
Reviewed By: pdillinger
Differential Revision: D33797275
Pulled By: sidroyc
fbshipit-source-id: 8db81d9c9cea5e2e4f1445d3aecad8106137b8e7
Summary:
Fixes a major performance regression in 6.26, where
extra CPU is spent in SliceTransform::AsString when reads involve
a prefix_extractor (Get, MultiGet, Seek). Common case performance
is now better than 6.25.
This change creates a "fast path" for verifying that the current prefix
extractor is unchanged and compatible with what was used to
generate a table file. This fast path detects the common case by
pointer comparison on the current prefix_extractor and a "known
good" prefix extractor (if applicable) that is saved at the time the
table reader is opened. The "known good" prefix extractor is saved
as another shared_ptr copy (in an existing field, however) to ensure
the pointer is not recycled.
When the prefix_extractor has changed to a different instance but
same compatible configuration (rare, odd), performance is still a
regression compared to 6.25, but this is likely acceptable because
of the oddity of such a case. The performance of incompatible
prefix_extractor is essentially unchanged.
Also fixed a minor case (ForwardIterator) where a prefix_extractor
could be used via a raw pointer after being freed as a shared_ptr,
if replaced via SetOptions.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9407
Test Plan:
## Performance
Populate DB with `TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -benchmarks=fillrandom -num=10000000 -disable_wal=1 -write_buffer_size=10000000 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=12`
Running head-to-head comparisons simultaneously with `TEST_TMPDIR=/dev/shm/rocksdb ./db_bench -use_existing_db -readonly -benchmarks=seekrandom -num=10000000 -duration=20 -disable_wal=1 -bloom_bits=16 -compaction_style=2 -fifo_compaction_max_table_files_size_mb=10000 -fifo_compaction_allow_compaction=0 -prefix_size=12`
Below each is compared by ops/sec vs. baseline which is version 6.25 (multiple baseline runs because of variable machine load)
v6.26: 4833 vs. 6698 (<- major regression!)
v6.27: 4737 vs. 6397 (still)
New: 6704 vs. 6461 (better than baseline in common case)
Disabled fastpath: 4843 vs. 6389 (e.g. if prefix extractor instance changes but is still compatible)
Changed prefix size (no usable filter) in new: 787 vs. 5927
Changed prefix size (no usable filter) in new & baseline: 773 vs. 784
Reviewed By: mrambacher
Differential Revision: D33677812
Pulled By: pdillinger
fbshipit-source-id: 571d9711c461fb97f957378a061b7e7dbc4d6a76
Summary:
Wasn't able to easily reproduce error, but easy to see a race
condition between TestFlushListener::OnFlushCompleted and
DBTestBase::Close(), which frees CF handles before closing DB.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9400
Test Plan: CI etc.
Reviewed By: riversand963
Differential Revision: D33645134
Pulled By: pdillinger
fbshipit-source-id: d0ec914cc43c9e14f53da633876b95b61995138d
Summary:
In response to https://github.com/facebook/rocksdb/issues/9354, this PR adds a way for users to "opt out"
of extra checks that can impact peak write performance, which
currently only includes force_consistency_checks. I considered including
some other options but did not see a db_bench performance difference.
Also clarify in comment for force_consistency_checks that it can "slow
down saturated writing."
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9363
Test Plan:
basic coverage in unit tests
Using my perf test in https://github.com/facebook/rocksdb/issues/9354 comment, I see
force_consistency_checks=true -> 725360 ops/s
force_consistency_checks=false -> 783072 ops/s
Reviewed By: mrambacher
Differential Revision: D33636559
Pulled By: pdillinger
fbshipit-source-id: 25bfd006f4844675e7669b342817dd4c6a641e84
Summary:
Fix https://github.com/facebook/rocksdb/issues/8046 : FlushMemTable return ok but memtable does not synchronize flush. The way to fix it is to expose RecoveryError.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8173
Reviewed By: ajkr
Differential Revision: D31674552
Pulled By: jay-zhuang
fbshipit-source-id: 9d16b69ba12a196bb429332ec8224754de97773d
Summary:
As title.
This is part of an fb-internal task.
First, remove all `using namespace` statements if applicable.
Next, utilize multiple build platforms and see if anything is broken.
Should anything become broken, fix the compilation errors with as little extra change as possible.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9369
Test Plan:
internal build and make check
make clean && make static_lib && cd examples && make all
Reviewed By: pdillinger
Differential Revision: D33517260
Pulled By: riversand963
fbshipit-source-id: 3fc4ce6402a073421dfd9a9b2d1c79441dca7a40
Summary:
Compatible change, more natural (especially in generated Rust bindings), no risk that the API will ever need mutable access because it has to make a copy anyway.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9376
Reviewed By: ajkr
Differential Revision: D33541435
Pulled By: pdillinger
fbshipit-source-id: 15c512a0d70b6e8694fa99d598b7d022751c1e59
Summary:
In order to support old-style regex function registration, restored the original "Register<T>(string, Factory)" method using regular expressions. The PatternEntry methods were left in place but renamed to AddFactory. The goal is to allow for the deprecation of the original regex Registry method in an upcoming release.
Added modes to the PatternEntry kMatchZeroOrMore and kMatchAtLeastOne to match * or +, respectively (kMatchAtLeastOne was the original behavior).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9362
Reviewed By: pdillinger
Differential Revision: D33432562
Pulled By: mrambacher
fbshipit-source-id: ed88ab3f9a2ad0d525c7bd1692873f9bb3209d02
Summary:
This change adds the filename of the offending filen to several place that produce Status objects with code `kCorruption`.
This is not an attempt to have every Corruption message in the codebase extended with the filename, but it is a start.
The motivation for the change was to quickly diagnose which file is corrupted when a large database is openend and there is not option to copy it offsite for analysis, run strace or install the ldb tool.
In the particular case in question, the error message improved from a mere
```
Corruption: checksum mismatch
```
to
```
Corruption: checksum mismatch in file /path/to/db/engine-rocksdb/MANIFEST-000171
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9239
Reviewed By: jay-zhuang
Differential Revision: D33237742
Pulled By: riversand963
fbshipit-source-id: bd42559cfbf786a0a674d091671d1a2bf07bdd31
Summary:
Function `Version::UpdateFilesByCompactionPri()` is never called and not implemented.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8724
Reviewed By: ajkr
Differential Revision: D30643943
Pulled By: riversand963
fbshipit-source-id: 174b2d9a2a42e286222909a035cc74a7b5602335
Summary:
Closing https://github.com/facebook/rocksdb/issues/5006
Calling `DB::DestroyColumnFamilyHandle(column_family)` with `column_family` being the return value of
`DB::DefaultColumnFamily()` will return `Status::InvalidArgument()`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9347
Test Plan: make check
Reviewed By: akankshamahajan15
Differential Revision: D33369675
Pulled By: riversand963
fbshipit-source-id: a8266a4daddf2b7a773c2dc7f3eb9a4adfb6b6dd
Summary:
The LastSequence field in the MANIFEST file is the baseline seqno for a recovered DB. Recovering WAL entries might cause the recovered DB's seqno to advance above this baseline, but the recovered DB will never use a smaller seqno.
Before this PR, we were writing the DB's seqno at the time of LogAndApply() as the LastSequence value. This works in the sense that it is a large enough baseline for the recovered DB that it'll never overwrite any records in existing SST files. At the same time, it's arbitrarily larger than what's needed. This behavior comes from LevelDB, where there was no tracking of largest seqno in an SST file.
Now we know the largest seqno of newly written SST files, so we can write an exact value in LastSequence that actually reflects the largest seqno in any file referred to by the MANIFEST. This is primarily useful for correctness testing with unsynced data loss, where the recovered DB's seqno needs to indicate what records were recovered.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9305
Test Plan:
- https://github.com/facebook/rocksdb/issues/9338 adds crash-recovery correctness testing coverage for WAL disabled use cases
- https://github.com/facebook/rocksdb/issues/9357 will extend that testing to cover file ingestion
- Added assertion at end of LogAndApply() for `VersionSet::descriptor_last_sequence_` consistency with files
- Manually tested upgrade/downgrade compatibility with a custom crash test that randomly picks between a `db_stress` built with and without this PR (for old code it must run with `-disable_wal=0`)
Reviewed By: riversand963
Differential Revision: D33182770
Pulled By: ajkr
fbshipit-source-id: 0bfafaf685f347cc8cb0e1d62e0186340a738f7d
Summary:
Allows the Env to have options (Configurable) and loads like other Customizable classes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9293
Reviewed By: pdillinger, zhichao-cao
Differential Revision: D33181591
Pulled By: mrambacher
fbshipit-source-id: 55e823886c654d214eda9eedd45ccdc54dac14d7
Summary:
Fixes https://github.com/facebook/rocksdb/issues/9339
When writing SST file, the name, computed as `prefix_extractor->GetId()` will be written to the properties block.
When the SST is opened again in the future, `CreateFromString()` will take the name as argument and try
to create a prefix extractor object. Without this fix, the C API will pass a `Wrapper` pointer to the underlying
DB's `prefix_extractor`. `Wrapper::GetId()`, in this case, will be missing the prefix length component, causing a
prefix extractor of length 0 to be silently created and used.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9343
Test Plan:
```
make c_test
./c_test
```
Reviewed By: mrambacher
Differential Revision: D33355549
Pulled By: riversand963
fbshipit-source-id: c92c3acd8be262c3bff8794b4229e42b9ee31203
Summary:
This option causes trace records to be written in the serialized write thread. That way, the write records in the trace must follow the same order as writes that are logged to WAL and writes that are applied to the DB.
By default I left it disabled to match existing behavior. I enabled it in `db_stress`, though, as that use case requires order of write records in trace matches the order in WAL.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9334
Test Plan:
- See if below unsynced data loss crash test can run for 24h straight. It used to crash after a few hours when reaching an unlucky trace ordering.
```
DEBUG_LEVEL=0 TEST_TMPDIR=/dev/shm /usr/local/bin/python3 -u tools/db_crashtest.py blackbox --interval=10 --max_key=100000 --write_buffer_size=524288 --target_file_size_base=524288 --max_bytes_for_level_base=2097152 --value_size_mult=33 --sync_fault_injection=1 --test_batches_snapshots=0 --duration=86400
```
Reviewed By: zhichao-cao
Differential Revision: D33301990
Pulled By: ajkr
fbshipit-source-id: 82d97559727adb4462a7af69758449c8725b22d3
Summary:
As (https://github.com/facebook/rocksdb/issues/9210) discussed, the **full_history_ts_low** is a member of CompactRangeOptions currently, which means a CF's fullHistoryTsLow is advanced only when users submit a CompactRange request.
However, users may want to advance the fllHistoryTsLow without an immediate compact.
This merge make IncreaseFullHistoryTsLow to a public API so users can advance each CF's fullHistoryTsLow seperately.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9221
Reviewed By: akankshamahajan15
Differential Revision: D33201106
Pulled By: riversand963
fbshipit-source-id: 9cb1d013ba93260f72e16353e693ffee167b47ee
Summary:
The failure looked like this:
```
utilities/backupable/backupable_db_test.cc:3161: Failure
Value of: db_chroot_env_->FileExists(prev_manifest_path).IsNotFound()
Actual: false
Expected: true
```
The failure could be coerced consistently with the following patch:
```
diff --git a/db/db_impl/db_impl_compaction_flush.cc b/db/db_impl/db_impl_compaction_flush.cc
index 80410f671..637636791 100644
--- a/db/db_impl/db_impl_compaction_flush.cc
+++ b/db/db_impl/db_impl_compaction_flush.cc
@@ -2772,6 +2772,8 @@ void DBImpl::BackgroundCallFlush(Env::Priority thread_pri) {
if (job_context.HaveSomethingToClean() ||
job_context.HaveSomethingToDelete() || !log_buffer.IsEmpty()) {
mutex_.Unlock();
+ bg_cv_.SignalAll();
+ sleep(1);
TEST_SYNC_POINT("DBImpl::BackgroundCallFlush:FilesFound");
// Have to flush the info logs before bg_flush_scheduled_--
// because if bg_flush_scheduled_ becomes 0 and the lock is
```
The cause was a familiar problem, which is manual flush/compaction may
return before files they obsoleted are removed. The solution is just to
wait for "scheduled" work to complete, which includes all phases
including cleanup.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9327
Test Plan:
after this PR, even the above patch to coerce the bug cannot
cause the test to fail.
Reviewed By: riversand963
Differential Revision: D33252208
Pulled By: ajkr
fbshipit-source-id: 720a7eaca58c7247d221911fffe3d5e1dbf581e9
Summary:
Fixes a problem where the iterator for metadata was being treated as a non-user key when in fact it was a user key. This led to a problem where the property keys could not be searched for correctly.
The main exposure of this problem was that the HashIndexReader could not get the "prefixes" property correctly, resulting in the failure of retrieval/creation of the BlockPrefixIndex.
Added BlockBasedTableTest.SeekMetaBlocks test to validate this condition.
Fixing this condition exposed two other tests (SeekWithPrefixLongerThanKey, MultiGetPrefixFilter) that passed incorrectly previously and now failed. Updated those two tests to pass. Not sure if the tests are functionally correct/still appropriate, but made them pass...
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8692
Reviewed By: riversand963
Differential Revision: D33119539
Pulled By: mrambacher
fbshipit-source-id: 658969fe9265f73dc184dab97cc3f4eaed2d881a
Summary:
We saw the below assertion failure in `error_handler_fs_test`:
```
db/error_handler_fs_test.cc:2471: Failure
Expected equality of these values:
listener->new_bg_error()
Which is: 16-byte object <00-00 00-00 00-00 00-00 00-00 00-00 00-00 00-00>
Status::Aborted()
Which is: 16-byte object <0A-00 00-00 60-61 00-00 00-00 00-00 00-00 00-00>
terminate called after throwing an instance of 'testing::internal::GoogleTestFailureException'
what(): db/error_handler_fs_test.cc:2471: Failure
Expected equality of these values:
listener->new_bg_error()
Which is: 16-byte object <00-00 00-00 00-00 00-00 00-00 00-00 00-00 00-00>
Status::Aborted()
Which is: 16-byte object <0A-00 00-00 60-61 00-00 00-00 00-00 00-00 00-00>
Received signal 6 (Aborted)
```
The problem was completing `OnErrorRecoveryCompleted()` would
wake up the main thread and allow it to proceed to that assertion. But
that assertion assumes `OnErrorRecoveryEnd()` has completed since
only `OnErrorRecoveryEnd()` affects `new_bg_error()`.
The fix is just to make `OnErrorRecoveryCompleted()` not wake up the
main thread, by means of not implementing it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9325
Test Plan:
- ran `while TEST_TMPDIR=/dev/shm ./error_handler_fs_test ; do : ; done` for a while
- injected sleep between `OnErrorRecovery{Completed,End}()` callbacks, which guaranteed repro before this PR
Reviewed By: anand1976
Differential Revision: D33249200
Pulled By: ajkr
fbshipit-source-id: 1659ee183cd09f90d4dbd898f65103473fcf84a8
Summary:
- Make MemoryAllocator and its implementations into a Customizable class.
- Added a "DefaultMemoryAllocator" which uses new and delete
- Added a "CountedMemoryAllocator" that counts the number of allocs and free
- Updated the existing tests to use these new allocators
- Changed the memkind allocator test into a generic test that can test the various allocators.
- Added tests for creating all of the allocators
- Added tests to verify/create the JemallocNodumpAllocator using its options.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8980
Reviewed By: zhichao-cao
Differential Revision: D32990403
Pulled By: mrambacher
fbshipit-source-id: 6fdfe8218c10dd8dfef34344a08201be1fa95c76
Summary:
This change standardizes on a new 16-byte cache key format for
block cache (incl compressed and secondary) and persistent cache (but
not table cache and row cache).
The goal is a really fast cache key with practically ideal stability and
uniqueness properties without external dependencies (e.g. from FileSystem).
A fixed key size of 16 bytes should enable future optimizations to the
concurrent hash table for block cache, which is a heavy CPU user /
bottleneck, but there appears to be measurable performance improvement
even with no changes to LRUCache.
This change replaces a lot of disjointed and ugly code handling cache
keys with calls to a simple, clean new internal API (cache_key.h).
(Preserving the old cache key logic under an option would be very ugly
and likely negate the performance gain of the new approach. Complete
replacement carries some inherent risk, but I think that's acceptable
with sufficient analysis and testing.)
The scheme for encoding new cache keys is complicated but explained
in cache_key.cc.
Also: EndianSwapValue is moved to math.h to be next to other bit
operations. (Explains some new include "math.h".) ReverseBits operation
added and unit tests added to hash_test for both.
Fixes https://github.com/facebook/rocksdb/issues/7405 (presuming a root cause)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9126
Test Plan:
### Basic correctness
Several tests needed updates to work with the new functionality, mostly
because we are no longer relying on filesystem for stable cache keys
so table builders & readers need more context info to agree on cache
keys. This functionality is so core, a huge number of existing tests
exercise the cache key functionality.
### Performance
Create db with
`TEST_TMPDIR=/dev/shm ./db_bench -bloom_bits=10 -benchmarks=fillrandom -num=3000000 -partition_index_and_filters`
And test performance with
`TEST_TMPDIR=/dev/shm ./db_bench -readonly -use_existing_db -bloom_bits=10 -benchmarks=readrandom -num=3000000 -duration=30 -cache_index_and_filter_blocks -cache_size=250000 -threads=4`
using DEBUG_LEVEL=0 and simultaneous before & after runs.
Before ops/sec, avg over 100 runs: 121924
After ops/sec, avg over 100 runs: 125385 (+2.8%)
### Collision probability
I have built a tool, ./cache_bench -stress_cache_key to broadly simulate host-wide cache activity
over many months, by making some pessimistic simplifying assumptions:
* Every generated file has a cache entry for every byte offset in the file (contiguous range of cache keys)
* All of every file is cached for its entire lifetime
We use a simple table with skewed address assignment and replacement on address collision
to simulate files coming & going, with quite a variance (super-Poisson) in ages. Some output
with `./cache_bench -stress_cache_key -sck_keep_bits=40`:
```
Total cache or DBs size: 32TiB Writing 925.926 MiB/s or 76.2939TiB/day
Multiply by 9.22337e+18 to correct for simulation losses (but still assume whole file cached)
```
These come from default settings of 2.5M files per day of 32 MB each, and
`-sck_keep_bits=40` means that to represent a single file, we are only keeping 40 bits of
the 128-bit cache key. With file size of 2\*\*25 contiguous keys (pessimistic), our simulation
is about 2\*\*(128-40-25) or about 9 billion billion times more prone to collision than reality.
More default assumptions, relatively pessimistic:
* 100 DBs in same process (doesn't matter much)
* Re-open DB in same process (new session ID related to old session ID) on average
every 100 files generated
* Restart process (all new session IDs unrelated to old) 24 times per day
After enough data, we get a result at the end:
```
(keep 40 bits) 17 collisions after 2 x 90 days, est 10.5882 days between (9.76592e+19 corrected)
```
If we believe the (pessimistic) simulation and the mathematical generalization, we would need to run a billion machines all for 97 billion days to expect a cache key collision. To help verify that our generalization ("corrected") is robust, we can make our simulation more precise with `-sck_keep_bits=41` and `42`, which takes more running time to get enough data:
```
(keep 41 bits) 16 collisions after 4 x 90 days, est 22.5 days between (1.03763e+20 corrected)
(keep 42 bits) 19 collisions after 10 x 90 days, est 47.3684 days between (1.09224e+20 corrected)
```
The generalized prediction still holds. With the `-sck_randomize` option, we can see that we are beating "random" cache keys (except offsets still non-randomized) by a modest amount (roughly 20x less collision prone than random), which should make us reasonably comfortable even in "degenerate" cases:
```
197 collisions after 1 x 90 days, est 0.456853 days between (4.21372e+18 corrected)
```
I've run other tests to validate other conditions behave as expected, never behaving "worse than random" unless we start chopping off structured data.
Reviewed By: zhichao-cao
Differential Revision: D33171746
Pulled By: pdillinger
fbshipit-source-id: f16a57e369ed37be5e7e33525ace848d0537c88f
Summary:
`db_stress` is a user of `FaultInjectionTestFS`. After injecting a write error, `db_stress` probabilistically determins
data drop (https://github.com/facebook/rocksdb/blob/6.27.fb/db_stress_tool/db_stress_test_base.cc#L2615:L2619).
In some of our recent runs of `db_stress`, we found duplicate trailing entries corresponding to file trivial move in
the MANIFEST, causing the recovery to fail, because the file move operation is not idempotent: you cannot delete a
file from a given level twice.
Investigation suggests that data buffering in both `WritableFileWriter` and `FaultInjectionTestFS` may be the root cause.
WritableFileWriter buffers data to write in a memory buffer, `WritableFileWriter::buf_`. After each
`WriteBuffered()`/`WriteBufferedWithChecksum()` succeeds, the `buf_` is cleared.
If the underlying file `WritableFileWriter::writable_file_` is opened in buffered IO mode, then `FaultInjectionTestFS`
buffers data written for each file until next file sync. After an injected error, user of `FaultInjectionFS` can
choose to drop some or none of previously buffered data. If `db_stress` does not drop any unsynced data, then
such data will still exist in the `FaultInjectionTestFS`'s buffer.
Existing implementation of `WritableileWriter::WriteBuffered()` does not clear `buf_` if there is an error. This may lead
to the data being buffered two copies: one in `WritableFileWriter`, and another in `FaultInjectionTestFS`.
We also know that the `WritableFileWriter` of MANIFEST file will close upon an error. During `Close()`, it will flush the
content in `buf_`. If no write error is injected to `FaultInjectionTestFS` this time, then we end up with two copies of the
data appended to the file.
To fix, we clear the `WritableFileWriter::buf_` upon failure as well. We focus this PR on files opened in non-direct mode.
This PR includes a unit test to reproduce a case when write error injection
to `WritableFile` can cause duplicate trailing entries.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9236
Test Plan: make check
Reviewed By: zhichao-cao
Differential Revision: D33033984
Pulled By: riversand963
fbshipit-source-id: ebfa5a0db8cbf1ed73100528b34fcba543c5db31
Peter Dillinger
