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:
Despite attempts to optimize `db_stress` setup phase (i.e.,
pre-`OperateDb()`) latency in https://github.com/facebook/rocksdb/issues/9470 and https://github.com/facebook/rocksdb/issues/9475, it still always took tens
of seconds. Since we still aren't able to setup a 100M key `db_stress`
quickly, we should reduce the number of keys. This PR reduces it 4x
while increasing `value_size_mult` 4x (from its default value of 8) so
that memtables and SST files fill at a similar rate compared to before this PR.
Also disabled bzip2 compression since we'll probably never use it and
I noticed many CI runs spending majority of CPU on bzip2 decompression.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9476
Reviewed By: siying
Differential Revision: D33898520
Pulled By: ajkr
fbshipit-source-id: 855021784ad9664f2be5bce21f0339a1cf93230d
Summary:
Even after https://github.com/facebook/rocksdb/issues/9461 could see
```
Error: please specify prefix_size for test_batches_snapshots test!
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9463
Test Plan:
run `make blackbox_crashtest` for a long time. (Unfortunately,
it's taking a long time to reproduce these failures)
Reviewed By: akankshamahajan15
Differential Revision: D33838152
Pulled By: pdillinger
fbshipit-source-id: b9a73c5bbb68df53f14c22b9b52f61d1f7ef38af
Summary:
Changes in https://github.com/facebook/rocksdb/issues/9453 could trigger
```
stderr:
Error: prefixpercent is non-zero while prefix_size is not positive!
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9461
Test Plan: run `make blackbox_crashtest` for a long time
Reviewed By: ajkr
Differential Revision: D33830751
Pulled By: pdillinger
fbshipit-source-id: be88377dcaa47e4bb7adb0347762639eff8f1476
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:
Recently we added the ability to verify some prefix of operations are recovered (AKA no "hole" in the recovered data) (https://github.com/facebook/rocksdb/issues/8966). Besides testing unsynced data loss scenarios, it is also useful to test WAL disabled use cases, where unflushed writes are expected to be lost. Note RocksDB only offers the prefix-recovery guarantee to WAL-disabled use cases that use atomic flush, so crash test always enables atomic flush when WAL is disabled.
To verify WAL-disabled crash-recovery correctness globally, i.e., also in whitebox and blackbox transaction tests, it is possible but requires further changes. I added TODOs in db_crashtest.py.
Depends on https://github.com/facebook/rocksdb/issues/9305.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9338
Test Plan: Running all crash tests and many instances of blackbox. Sandcastle links are in Phabricator diff test plan.
Reviewed By: riversand963
Differential Revision: D33345333
Pulled By: ajkr
fbshipit-source-id: f56dd7d2e5a78d59301bf4fc3fedb980eb31e0ce
Summary:
db_crashtest.py uses multiple CFs only when run without flag `--simple`.
The previous config set `-test_batches_snapshots=1` in that case for
blackbox mode. But `-test_batches_snapshots=1` cannot verify recovery
correctness, so it should not always be set for multi-CF blackbox tests.
We can instead randomly toggle it.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9303
Reviewed By: riversand963
Differential Revision: D33155229
Pulled By: ajkr
fbshipit-source-id: 4a6fdc4eddccc8ece664063baf6393ce1c5de6b7
Summary:
When table_options.prepopulate_block_cache is set to
BlockBasedTableOptions::PrepopulateBlockCache::kFlushOnly and
table_options.partition_filters is also set true, then there is
segmentation failure when top level filter is fetched because its
entered with wrong type in cache.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9263
Test Plan:
Updated unit tests;
Ran db_stress: make crash_test -j32
Reviewed By: pdillinger
Differential Revision: D32936566
Pulled By: akankshamahajan15
fbshipit-source-id: 8bd79e53830d3e3c1bb79787e1ffbc3cb46d4426
Summary:
The patch adds a new BlobDB configuration option `blob_compaction_readahead_size`
that can be used to enable prefetching data from blob files during compaction.
This is important when using storage with higher latencies like HDDs or remote filesystems.
If enabled, prefetching is used for all cases when blobs are read during compaction,
namely garbage collection, compaction filters (when the existing value has to be read from
a blob file), and `Merge` (when the value of the base `Put` is stored in a blob file).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9187
Test Plan: Ran `make check` and the stress/crash test.
Reviewed By: riversand963
Differential Revision: D32565512
Pulled By: ltamasi
fbshipit-source-id: 87be9cebc3aa01cc227bec6b5f64d827b8164f5d
Summary:
XXH3 - latest hash function that is extremely fast on large
data, easily faster than crc32c on most any x86_64 hardware. In
integrating this hash function, I have handled the compression type byte
in a non-standard way to avoid using the streaming API (extra data
movement and active code size because of hash function complexity). This
approach got a thumbs-up from Yann Collet.
Existing functionality change:
* reject bad ChecksumType in options with InvalidArgument
This change split off from https://github.com/facebook/rocksdb/issues/9058 because context-aware checksum is
likely to be handled through different configuration than ChecksumType.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/9069
Test Plan:
tests updated, and substantially expanded. Unit tests now check
that we don't accidentally change the values generated by the checksum
algorithms ("schema test") and that we properly handle
invalid/unrecognized checksum types in options or in file footer.
DBTestBase::ChangeOptions (etc.) updated from two to one configuration
changing from default CRC32c ChecksumType. The point of this test code
is to detect possible interactions among features, and the likelihood of
some bad interaction being detected by including configurations other
than XXH3 and CRC32c--and then not detected by stress/crash test--is
extremely low.
Stress/crash test also updated (manual run long enough to see it accepts
new checksum type). db_bench also updated for microbenchmarking
checksums.
### Performance microbenchmark (PORTABLE=0 DEBUG_LEVEL=0, Broadwell processor)
./db_bench -benchmarks=crc32c,xxhash,xxhash64,xxh3,crc32c,xxhash,xxhash64,xxh3,crc32c,xxhash,xxhash64,xxh3
crc32c : 0.200 micros/op 5005220 ops/sec; 19551.6 MB/s (4096 per op)
xxhash : 0.807 micros/op 1238408 ops/sec; 4837.5 MB/s (4096 per op)
xxhash64 : 0.421 micros/op 2376514 ops/sec; 9283.3 MB/s (4096 per op)
xxh3 : 0.171 micros/op 5858391 ops/sec; 22884.3 MB/s (4096 per op)
crc32c : 0.206 micros/op 4859566 ops/sec; 18982.7 MB/s (4096 per op)
xxhash : 0.793 micros/op 1260850 ops/sec; 4925.2 MB/s (4096 per op)
xxhash64 : 0.410 micros/op 2439182 ops/sec; 9528.1 MB/s (4096 per op)
xxh3 : 0.161 micros/op 6202872 ops/sec; 24230.0 MB/s (4096 per op)
crc32c : 0.203 micros/op 4924686 ops/sec; 19237.1 MB/s (4096 per op)
xxhash : 0.839 micros/op 1192388 ops/sec; 4657.8 MB/s (4096 per op)
xxhash64 : 0.424 micros/op 2357391 ops/sec; 9208.6 MB/s (4096 per op)
xxh3 : 0.162 micros/op 6182678 ops/sec; 24151.1 MB/s (4096 per op)
As you can see, especially once warmed up, xxh3 is fastest.
### Performance macrobenchmark (PORTABLE=0 DEBUG_LEVEL=0, Broadwell processor)
Test
for I in `seq 1 50`; do for CHK in 0 1 2 3 4; do TEST_TMPDIR=/dev/shm/rocksdb$CHK ./db_bench -benchmarks=fillseq -memtablerep=vector -allow_concurrent_memtable_write=false -num=30000000 -checksum_type=$CHK 2>&1 | grep 'micros/op' | tee -a results-$CHK & done; wait; done
Results (ops/sec)
for FILE in results*; do echo -n "$FILE "; awk '{ s += $5; c++; } END { print 1.0 * s / c; }' < $FILE; done
results-0 252118 # kNoChecksum
results-1 251588 # kCRC32c
results-2 251863 # kxxHash
results-3 252016 # kxxHash64
results-4 252038 # kXXH3
Reviewed By: mrambacher
Differential Revision: D31905249
Pulled By: pdillinger
fbshipit-source-id: cb9b998ebe2523fc7c400eedf62124a78bf4b4d1
Summary:
The current BlobDB garbage collection logic works by relocating the valid
blobs from the oldest blob files as they are encountered during compaction,
and cleaning up blob files once they contain nothing but garbage. However,
with sufficiently skewed workloads, it is theoretically possible to end up in a
situation when few or no compactions get scheduled for the SST files that contain
references to the oldest blob files, which can lead to increased space amp due
to the lack of GC.
In order to efficiently handle such workloads, the patch adds a new BlobDB
configuration option called `blob_garbage_collection_force_threshold`,
which signals to BlobDB to schedule targeted compactions for the SST files
that keep alive the oldest batch of blob files if the overall ratio of garbage in
the given blob files meets the threshold *and* all the given blob files are
eligible for GC based on `blob_garbage_collection_age_cutoff`. (For example,
if the new option is set to 0.9, targeted compactions will get scheduled if the
sum of garbage bytes meets or exceeds 90% of the sum of total bytes in the
oldest blob files, assuming all affected blob files are below the age-based cutoff.)
The net result of these targeted compactions is that the valid blobs in the oldest
blob files are relocated and the oldest blob files themselves cleaned up (since
*all* SST files that rely on them get compacted away).
These targeted compactions are similar to periodic compactions in the sense
that they force certain SST files that otherwise would not get picked up to undergo
compaction and also in the sense that instead of merging files from multiple levels,
they target a single file. (Note: such compactions might still include neighboring files
from the same level due to the need of having a "clean cut" boundary but they never
include any files from any other level.)
This functionality is currently only supported with the leveled compaction style
and is inactive by default (since the default value is set to 1.0, i.e. 100%).
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8994
Test Plan: Ran `make check` and tested using `db_bench` and the stress/crash tests.
Reviewed By: riversand963
Differential Revision: D31489850
Pulled By: ltamasi
fbshipit-source-id: 44057d511726a0e2a03c5d9313d7511b3f0c4eab
Summary:
`FaultInjectionTest{Env,FS}::ReopenWritableFile()` functions were accidentally deleting WALs from previous `db_stress` runs causing verification to fail. They were operating under the assumption that `ReopenWritableFile()` would delete any existing file. It was a reasonable assumption considering the `{Env,FileSystem}::ReopenWritableFile()` documentation stated that would happen. The only problem was neither the implementations we offer nor the "real" clients in RocksDB code followed that contract. So, this PR updates the contract as well as fixing the fault injection client usage.
The fault injection change exposed that `ExternalSSTFileBasicTest.SyncFailure` was relying on a fault injection `Env` dropping unsynced data written by a regular `Env`. I changed that test to make its `SstFileWriter` use fault injection `Env`, and also implemented `LinkFile()` in fault injection so the unsynced data is tracked under the new name.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8995
Test Plan:
- Verified it fixes the following failure:
```
$ ./db_stress --clear_column_family_one_in=0 --column_families=1 --db=/dev/shm/rocksdb_crashtest_whitebox --delpercent=5 --expected_values_dir=/dev/shm/rocksdb_crashtest_expected --iterpercent=0 --key_len_percent_dist=1,30,69 --max_key=100000 --max_key_len=3 --nooverwritepercent=1 --ops_per_thread=1000 --prefixpercent=0 --readpercent=60 --reopen=0 --target_file_size_base=1048576 --test_batches_snapshots=0 --write_buffer_size=1048576 --writepercent=35 --value_size_mult=33 -threads=1
...
$ ./db_stress --avoid_flush_during_recovery=1 --clear_column_family_one_in=0 --column_families=1 --db=/dev/shm/rocksdb_crashtest_whitebox --delpercent=5 --destroy_db_initially=0 --expected_values_dir=/dev/shm/rocksdb_crashtest_expected --iterpercent=10 --key_len_percent_dist=1,30,69 --max_bytes_for_level_base=4194304 --max_key=100000 --max_key_len=3 --nooverwritepercent=1 --open_files=-1 --open_metadata_write_fault_one_in=8 --open_write_fault_one_in=16 --ops_per_thread=1000 --prefix_size=-1 --prefixpercent=0 --readpercent=50 --sync=1 --target_file_size_base=1048576 --test_batches_snapshots=0 --write_buffer_size=1048576 --writepercent=35 --value_size_mult=33 -threads=1
...
Verification failed for column family 0 key 000000000000001300000000000000857878787878 (1143): Value not found: NotFound:
Crash-recovery verification failed :(
...
```
- `make check -j48`
Reviewed By: ltamasi
Differential Revision: D31495388
Pulled By: ajkr
fbshipit-source-id: 7886ccb6a07cb8b78ad7b6c1c341ccf40bb68385
Summary:
Enable SingleDelete with user defined timestamp in db_bench,
db_stress and crash test
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8971
Test Plan:
1. For db_stress, ran the command for full duration: i) python3 -u tools/db_crashtest.py
--enable_ts whitebox --nooverwritepercent=100
ii) make crash_test_with_ts
2. For db_bench, ran: ./db_bench -benchmarks=randomreplacekeys
-user_timestamp_size=8 -use_single_deletes=true
Reviewed By: riversand963
Differential Revision: D31246558
Pulled By: akankshamahajan15
fbshipit-source-id: 29cd8740c9921341e52f09242fca3c44d75a12b7
Summary:
This is a precursor refactoring to enable an upcoming feature: persistence failure correctness testing.
- Changed `--expected_values_path` to `--expected_values_dir` and migrated "db_crashtest.py" to use the new flag. For persistence failure correctness testing there are multiple possible correct states since unsynced data is allowed to be dropped. Making it possible to restore all these possible correct states will eventually involve files containing snapshots of expected values and DB trace files.
- The expected values directory is managed by an `ExpectedStateManager` instance. Managing expected state files is separated out of `SharedState` to prevent `SharedState` from becoming too complex when the new files and features (snapshotting, tracing, and restoring) are introduced.
- Migrated expected values file access/management out of `SharedState` into a separate class called `ExpectedState`. This is not exposed directly to the test but rather the `ExpectedState` for the latest values file is accessed via a pass-through API on `ExpectedStateManager`. This forces the test to always access the single latest `ExpectedState`.
- Changed the initialization of the latest expected values file to use a tempfile followed by rename, and also add cleanup logic for possible stranded tempfiles.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8913
Test Plan:
run in several ways; try to make sure it's not obviously broken.
- crashtest blackbox without TEST_TMPDIR
```
$ python3 tools/db_crashtest.py blackbox --simple --write_buffer_size=1048576 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --max_key=100000 --value_size_mult=33 --compression_type=none --duration=120 --interval=10 --compression_type=none --blob_compression_type=none
```
- crashtest blackbox with TEST_TMPDIR
```
$ TEST_TMPDIR=/dev/shm python3 tools/db_crashtest.py blackbox --simple --write_buffer_size=1048576 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --max_key=100000 --value_size_mult=33 --compression_type=none --duration=120 --interval=10 --compression_type=none --blob_compression_type=none
```
- crashtest whitebox with TEST_TMPDIR
```
$ TEST_TMPDIR=/dev/shm python3 tools/db_crashtest.py whitebox --simple --write_buffer_size=1048576 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --max_key=100000 --value_size_mult=33 --compression_type=none --duration=120 --interval=10 --compression_type=none --blob_compression_type=none --random_kill_odd=88887
```
- db_stress without expected_values_dir
```
$ ./db_stress --write_buffer_size=1048576 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --max_key=100000 --value_size_mult=33 --compression_type=none --ops_per_thread=10000 --clear_column_family_one_in=0 --destroy_db_initially=true
```
- db_stress with expected_values_dir and manual corruption
```
$ ./db_stress --write_buffer_size=1048576 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --max_key=100000 --value_size_mult=33 --compression_type=none --ops_per_thread=10000 --clear_column_family_one_in=0 --destroy_db_initially=true --expected_values_dir=./
// modify one byte in "./LATEST.state"
$ ./db_stress --write_buffer_size=1048576 --target_file_size_base=1048576 --max_bytes_for_level_base=4194304 --max_key=100000 --value_size_mult=33 --compression_type=none --ops_per_thread=10000 --clear_column_family_one_in=0 --destroy_db_initially=false --expected_values_dir=./
...
Verification failed for column family 0 key 0000000000000000 (0): Value not found: NotFound:
...
```
Reviewed By: riversand963
Differential Revision: D30921951
Pulled By: ajkr
fbshipit-source-id: babfe218062e55d018c9b046536c0289fb78f41c
Summary:
For now, disable it since the below command indicates it can cause a
failure. Running that command with `-experimental_mempurge_threshold=0`
has been running successfully for several minutes, whereas before it
failed in seconds.
```
$ while rm -rf /dev/shm/single_stress && ./db_stress --clear_column_family_one_in=0 --column_families=1 --db=/dev/shm/single_stress --experimental_mempurge_threshold=5.493146827397074 --flush_one_in=10000 --reopen=0 --write_buffer_size=262144 --value_size_mult=33 --max_write_buffer_number=3 -ops_per_thread=10000; do : ; done
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8958
Reviewed By: ltamasi
Differential Revision: D31187059
Pulled By: ajkr
fbshipit-source-id: 04d5bfb4fcc4f5b66233e691427dfd940c67037f
Summary:
Several improvements to MultiRead:
1. Fix a bug in stress test which causes false positive when both MultiRead() return and individual read request have failure injected.
2. Add two more types of fault that should be handled: empty read results and checksum mismatch
3. Add a message indicating which type of fault is injected
4. Increase the failure rate
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8937
Reviewed By: anand1976
Differential Revision: D31085930
fbshipit-source-id: 3a04994a3cadebf9a64d25e1fe12b14b7a272fba
Summary:
Current implementation does not support user-defined timestamp when
block-based filter is used. Will implement the support in the future, or
wait to see if block-based filter can be deprecated and removed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8703
Test Plan: make whitebox_crash_test_with_ts
Reviewed By: pdillinger
Differential Revision: D30528931
Pulled By: riversand963
fbshipit-source-id: 60dd74ee0a6194e69072069d8c4bd876f249f38d
Summary:
This is essentially resurrection and fixing of the part of
https://github.com/facebook/rocksdb/issues/8198 that was reverted in https://github.com/facebook/rocksdb/issues/8212, using data added in https://github.com/facebook/rocksdb/issues/8246. Basically,
when configuring Ribbon filter, you can specify an LSM level before which
Bloom will be used instead of Ribbon. But Bloom is only considered for
Leveled and Universal compaction styles and file going into a known LSM
level. This way, SST file writer, FIFO compaction, etc. use Ribbon filter as
you would expect with NewRibbonFilterPolicy.
So that this can be controlled with a single int value and so that flushes
can be distinguished from intra-L0, we consider flush to go to level -1 for
the purposes of this option. (Explained in API comment.)
I also expect the most common and recommended Ribbon configuration to
use Bloom during flush, to minimize slowing down writes and because according
to my estimates, Ribbon only pays off if the structure lives in memory for
more than an hour. Thus, I have changed the default for NewRibbonFilterPolicy
to be this mild hybrid configuration. I don't really want to add something like
NewHybridFilterPolicy because at least the mild hybrid configuration (Bloom for
flush, Ribbon otherwise) should be considered a natural choice.
C APIs also updated, but because they don't support overloading,
rocksdb_filterpolicy_create_ribbon is kept pure ribbon for clarity and
rocksdb_filterpolicy_create_ribbon_hybrid must be called for a hybrid
configuration. While touching C API, I changed bits per key options from
int to double.
BuiltinFilterPolicy is needed so that LevelThresholdFilterPolicy doesn't inherit
unused fields from BloomFilterPolicy.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8679
Test Plan: new + updated tests, including crash test
Reviewed By: jay-zhuang
Differential Revision: D30445797
Pulled By: pdillinger
fbshipit-source-id: 6f5aeddfd6d79f7e55493b563c2d1d2d568892e1
Summary:
Changes the API of the MemPurge process: the `bool experimental_allow_mempurge` and `experimental_mempurge_policy` flags have been replaced by a `double experimental_mempurge_threshold` option.
This change of API reflects another major change introduced in this PR: the MemPurgeDecider() function now works by sampling the memtables being flushed to estimate the overall amount of useful payload (payload minus the garbage), and then compare this useful payload estimate with the `double experimental_mempurge_threshold` value.
Therefore, when the value of this flag is `0.0` (default value), mempurge is simply deactivated. On the other hand, a value of `DBL_MAX` would be equivalent to always going through a mempurge regardless of the garbage ratio estimate.
At the moment, a `double experimental_mempurge_threshold` value else than 0.0 or `DBL_MAX` is opnly supported`with the `SkipList` memtable representation.
Regarding the sampling, this PR includes the introduction of a `MemTable::UniqueRandomSample` function that collects (approximately) random entries from the memtable by using the new `SkipList::Iterator::RandomSeek()` under the hood, or by iterating through each memtable entry, depending on the target sample size and the total number of entries.
The unit tests have been readapted to support this new API.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8628
Reviewed By: pdillinger
Differential Revision: D30149315
Pulled By: bjlemaire
fbshipit-source-id: 1feef5390c95db6f4480ab4434716533d3947f27
Summary:
Add `experimental_mempurge_policy` flag to `db_stress` and `db_crashtest.py`.
This flag is only read if the `experimental_allow_mempurge` flag is set to `true`. This flag can take the following values: `kAlways`, and `kAlternate` (default).
- `kAlways`: a flush is always redirected to a mempurge. If the mempurge aborts, the a regular flush proceeds.
- `kAlternate`: if one or more of the flush input memtables is an mempurge output memtable, then a flush is performed, else a mempurge is carried out. Similar to kAlways, if a mempurge aborts, the FlushJob proceeds to a regular flush to storage.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8588
Reviewed By: pdillinger
Differential Revision: D29934251
Pulled By: bjlemaire
fbshipit-source-id: 90c1debed2029b9915d066914556547507c33dae
Summary:
Add `experiemental_allow_mempurge` flag support for `db_stress` and `db_crashtest.py`, with a `false` default value.
I succesfully tested locally both `whitebox` and `blackbox` crash tests with `experiemental_allow_mempurge` flag set as true.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8545
Reviewed By: akankshamahajan15
Differential Revision: D29734513
Pulled By: bjlemaire
fbshipit-source-id: 24316c0eccf6caf409e95c035f31d822c66714ae
Summary:
Inject read failures in DB reopen, just as what we do for metadata writes and writes.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8476
Test Plan: Some manual tests and make sure failures are triggered.
Reviewed By: anand1976
Differential Revision: D29507283
fbshipit-source-id: d04da0163973447041038bd87701686a417c4e0c
Summary:
Previously Stress can inject metadata write failures when reopening a DB. We extend it to file append too, in the same way.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8474
Test Plan: manually run crash test with various setting and make sure the failures are triggered as expected.
Reviewed By: zhichao-cao
Differential Revision: D29503116
fbshipit-source-id: e73a446e80ccbd09301a579280e56ff949381fab
Summary:
Add a ```-secondary_cache_uri``` to db_stress to allow the user to specify a custom ```SecondaryCache``` object from the object registry. Also allow db_crashtest.py to be run with an alternate db_stress location. Together, these changes will allow us to run db_stress using FB internal components.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8455
Reviewed By: zhichao-cao
Differential Revision: D29371972
Pulled By: anand1976
fbshipit-source-id: dd1b1fd80ebbedc11aa63d9246ea6ae49edb77c4
Summary:
Whitebox crash test can run significantly over the time limit for test slowness or no kiling points. This indefinite job can create problem when this test is periodically scheduled as a job. Instead, kill the job if it is 15 minutes over the limit.
Refactor the code slightly to consolidate the code for executing commands for white and black box tests.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8341
Test Plan: Run both of black and white box tests with both of natual and explicit kill condition.
Reviewed By: jay-zhuang
Differential Revision: D28756170
fbshipit-source-id: f253149890e62ace78f871be927e093e9b12f49b
Summary:
In https://github.com/facebook/rocksdb/issues/8268, the `db_stress` stdout began containing both the strings
"fail" and "error" (case-insensitive). The whitebox crash test
failed upon seeing either of those strings.
I checked that all other occurrences of "fail" and "error"
(case-insensitive) that `db_stress` produces are printed to `stderr`. So
this PR separates the handling of `db_stress`'s stdout and stderr, and
only fails when one those bad strings are found in stderr.
The downside of this PR is `db_stress`'s original interleaving of stdout/stderr is not preserved in `db_crashtest.py`'s output.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8272
Test Plan:
run it; see it succeeds for several runs until encountering a real error
```
$ python3 tools/db_crashtest.py whitebox --simple --random_kill_odd=8887 --max_key=1000000 --value_size_mult=33
...
db_stress: cache/clock_cache.cc:483: bool rocksdb::{anonymous}::ClockCacheShard::Unref(rocksdb::{anonymous}::CacheHandle*, bool, rocksdb::{anonymous}::CleanupContext*): Assertion `CountRefs(flags) > 0' failed.
TEST FAILED. Output has 'fail'!!!
```
Reviewed By: zhichao-cao
Differential Revision: D28239233
Pulled By: ajkr
fbshipit-source-id: 3b8602a0d570466a7e2c81bb9c49468f7716091e
Summary:
In testing for https://github.com/facebook/rocksdb/issues/8225 I found cache_bench would crash with
-use_clock_cache, as well as db_bench -use_clock_cache, but not
single-threaded. Smaller cache size hits failure much faster. ASAN
reported the failuer as calling malloc_usable_size on the `key` pointer
of a ClockCache handle after it was reportedly freed. On detailed
inspection I found this bad sequence of operations for a cache entry:
state=InCache=1,refs=1
[thread 1] Start ClockCacheShard::Unref (from Release, no mutex)
[thread 1] Decrement ref count
state=InCache=1,refs=0
[thread 1] Suspend before CalcTotalCharge (no mutex)
[thread 2] Start UnsetInCache (from Insert, mutex held)
[thread 2] clear InCache bit
state=InCache=0,refs=0
[thread 2] Calls RecycleHandle (based on pre-updated state)
[thread 2] Returns to Insert which calls Cleanup which deletes `key`
[thread 1] Resume ClockCacheShard::Unref
[thread 1] Read `key` in CalcTotalCharge
To fix this, I've added a field to the handle to store the metadata
charge so that we can efficiently remember everything we need from
the handle in Unref. We must not read from the handle again if we
decrement the count to zero with InCache=1, which means we don't own
the entry and someone else could eject/overwrite it immediately.
Note before this change, on amd64 sizeof(Handle) == 56 even though there
are only 48 bytes of data. Grouping together the uint32_t fields would
cut it down to 48, but I've added another uint32_t, which takes it
back up to 56. Not a big deal.
Also fixed DisownData to cooperate with ASAN as in LRUCache.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8261
Test Plan:
Manual + adding use_clock_cache to db_crashtest.py
Base performance
./cache_bench -use_clock_cache
Complete in 17.060 s; QPS = 2458513
New performance
./cache_bench -use_clock_cache
Complete in 17.052 s; QPS = 2459695
Any difference is easily buried in small noise.
Crash test shows still more bug(s) in ClockCache, so I'm expecting to
disable ClockCache from production code in a follow-up PR (if we
can't find and fix the bug(s))
Reviewed By: mrambacher
Differential Revision: D28207358
Pulled By: pdillinger
fbshipit-source-id: aa7a9322afc6f18f30e462c75dbbe4a1206eb294
Summary:
DB Stress to add --open_metadata_write_fault_one_in which would randomly fail in some file metadata modification operations during DB Open, including file creation, close, renaming and directory sync. Some operations can fail before and after the operations take place.
If DB open fails, db_stress would retry without the failure ingestion, and DB is expected to open successfully.
This option is enabled in crash test in half of the time.
Some follow up changes would allow write failures in open time, and ingesting those failures in non-DB open cases.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8235
Test Plan: Run stress tests for a while and see failures got triggered. This can reproduce the bug fixed by https://github.com/facebook/rocksdb/pull/8192 and a similar one that fails when fsyncing parent directory.
Reviewed By: anand1976
Differential Revision: D28010944
fbshipit-source-id: 36a96da4dc3633e5f7680cef3ea0a900fcdb5558
Summary:
This partially reverts commit 10196d7edc.
The problem with this change is because of important filter use cases:
FIFO compaction and SST writer. FIFO "compaction" always uses level 0 so
would only use Ribbon filters if specifically including level 0 for the
Ribbon filter policy. SST writer sets level_at_creation=-1 to indicate
unknown level, and this would be treated the same as level 0 unless
fixed.
We are keeping the part about committing to permanent schema, which is
only changes to API comments and HISTORY.md.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8212
Test Plan: CI
Reviewed By: jay-zhuang
Differential Revision: D27896468
Pulled By: pdillinger
fbshipit-source-id: 50a775f7cba5d64fb729d9b982e355864020596e
Summary:
Since the Ribbon filter schema seems good (compatible back to
6.15.0), this change commits to long term support of the SST schema,
even though we expect the API for enabling Ribbon to change (still
called NewExperimentalRibbonFilterPolicy).
This also adds support for "hybrid" configuration in which some levels
use Bloom (higher levels, lower numbered) for speed and the rest use
Ribbon (lower levels, higher numbered) for memory space efficiency.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8198
Test Plan: unit test added, crash test support
Reviewed By: jay-zhuang
Differential Revision: D27831232
Pulled By: pdillinger
fbshipit-source-id: 90e528677689474d293ed6710b42ba89fbd5b5ab
Summary:
Enable backup/restore functionality with Integrated BlobDB in
db_stress and crash test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8165
Test Plan:
Ran python3 -u tools/db_crashtest.py --simple whitebox along
with :
1. decreased "backup_in_one" value for backups to be more frequent and
2. manually changed code for "enable_blob_file" to be always true and
apply blobdb params 100% for testing purpose.
Reviewed By: ltamasi
Differential Revision: D27636025
Pulled By: akankshamahajan15
fbshipit-source-id: 0d0e0d1479ced163f992872dc998e79c581bfc99
Summary:
According to https://github.com/facebook/rocksdb/issues/5907, each filter partition "should include the bloom of the prefix of the last
key in the previous partition" so that SeekForPrev() in prefix mode can return correct result.
The prefix of the last key in the previous partition does not necessarily have the same prefix
as the first key in the current partition. Regardless of the first key in current partition, the
prefix of the last key in the previous partition should be added. The existing code, however,
does not follow this. Furthermore, there is another issue: when finishing current filter partition,
`FullFilterBlockBuilder::AddPrefix()` is called for the first key in next filter partition, which effectively
overwrites `last_prefix_str_` prematurely. Consequently, when the filter block builder proceeds
to the next partition, `last_prefix_str_` will be the prefix of its first key, leaving no way of adding
the bloom of the prefix of the last key of the previous partition.
Prefix extractor is FixedLength.2.
```
[ filter part 1 ] [ filter part 2 ]
abc d
```
When SeekForPrev("abcd"), checking the filter partition will land on filter part 2 because "abcd" > "abc"
but smaller than "d".
If the filter in filter part 2 happens to return false for the test for "ab", then SeekForPrev("abcd") will build
incorrect iterator tree in non-total-order mode.
Also fix a unit test which starts to fail following this PR. `InDomain` should not fail due to assertion
error when checking on an arbitrary key.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8137
Test Plan:
```
make check
```
Without this fix, the following command will fail pretty soon.
```
./db_stress --acquire_snapshot_one_in=10000 --avoid_flush_during_recovery=0 \
--avoid_unnecessary_blocking_io=0 --backup_max_size=104857600 --backup_one_in=0 \
--batch_protection_bytes_per_key=0 --block_size=16384 --bloom_bits=17 \
--bottommost_compression_type=disable --cache_index_and_filter_blocks=1 --cache_size=1048576 \
--checkpoint_one_in=0 --checksum_type=kxxHash64 --clear_column_family_one_in=0 \
--compact_files_one_in=1000000 --compact_range_one_in=1000000 --compaction_ttl=0 \
--compression_max_dict_buffer_bytes=0 --compression_max_dict_bytes=0 \
--compression_parallel_threads=1 --compression_type=zstd --compression_zstd_max_train_bytes=0 \
--continuous_verification_interval=0 --db=/dev/shm/rocksdb/rocksdb_crashtest_whitebox \
--db_write_buffer_size=8388608 --delpercent=5 --delrangepercent=0 --destroy_db_initially=0 --enable_blob_files=0 \
--enable_compaction_filter=0 --enable_pipelined_write=1 --file_checksum_impl=big --flush_one_in=1000000 \
--format_version=5 --get_current_wal_file_one_in=0 --get_live_files_one_in=1000000 --get_property_one_in=1000000 \
--get_sorted_wal_files_one_in=0 --index_block_restart_interval=4 --index_type=2 --ingest_external_file_one_in=0 \
--iterpercent=10 --key_len_percent_dist=1,30,69 --level_compaction_dynamic_level_bytes=True \
--log2_keys_per_lock=10 --long_running_snapshots=1 --mark_for_compaction_one_file_in=0 \
--max_background_compactions=20 --max_bytes_for_level_base=10485760 --max_key=100000000 --max_key_len=3 \
--max_manifest_file_size=1073741824 --max_write_batch_group_size_bytes=16777216 --max_write_buffer_number=3 \
--max_write_buffer_size_to_maintain=8388608 --memtablerep=skip_list --mmap_read=1 --mock_direct_io=False \
--nooverwritepercent=0 --open_files=500000 --ops_per_thread=20000000 --optimize_filters_for_memory=0 --paranoid_file_checks=1 --partition_filters=1 --partition_pinning=0 --pause_background_one_in=1000000 \
--periodic_compaction_seconds=0 --prefixpercent=5 --progress_reports=0 --read_fault_one_in=0 --read_only=0 \
--readpercent=45 --recycle_log_file_num=0 --reopen=20 --secondary_catch_up_one_in=0 \
--snapshot_hold_ops=100000 --sst_file_manager_bytes_per_sec=104857600 \
--sst_file_manager_bytes_per_truncate=0 --subcompactions=2 --sync=0 --sync_fault_injection=False \
--target_file_size_base=2097152 --target_file_size_multiplier=2 --test_batches_snapshots=0 --test_cf_consistency=0 \
--top_level_index_pinning=0 --unpartitioned_pinning=1 --use_blob_db=0 --use_block_based_filter=0 \
--use_direct_io_for_flush_and_compaction=0 --use_direct_reads=0 --use_full_merge_v1=0 --use_merge=0 \
--use_multiget=0 --use_ribbon_filter=0 --use_txn=0 --user_timestamp_size=8 --verify_checksum=1 \
--verify_checksum_one_in=1000000 --verify_db_one_in=100000 --write_buffer_size=4194304 \
--write_dbid_to_manifest=1 --writepercent=35
```
Reviewed By: pdillinger
Differential Revision: D27553054
Pulled By: riversand963
fbshipit-source-id: 60e391e4a2d8d98a9a3172ec5d6176b90ec3de98
Summary:
Currently, partitioned filter does not support user-defined timestamp. Disable it for now in ts stress test so that
the contrun jobs can proceed.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8127
Test Plan: make crash_test_with_ts
Reviewed By: ajkr
Differential Revision: D27388488
Pulled By: riversand963
fbshipit-source-id: 5ccff18121cb537bd82f2ac072cd25efb625c666
Summary:
Fix the following error while running `make crash_test`
```
Traceback (most recent call last):
File "tools/db_crashtest.py", line 705, in <module>
main()
File "tools/db_crashtest.py", line 696, in main
blackbox_crash_main(args, unknown_args)
File "tools/db_crashtest.py", line 479, in blackbox_crash_main
+ list({'db': dbname}.items())), unknown_args)
File "tools/db_crashtest.py", line 414, in gen_cmd
finalzied_params = finalize_and_sanitize(params)
File "tools/db_crashtest.py", line 331, in finalize_and_sanitize
dest_params.get("user_timestamp_size") > 0):
TypeError: '>' not supported between instances of 'NoneType' and 'int'
```
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8091
Test Plan: make crash_test
Reviewed By: ltamasi
Differential Revision: D27268276
Pulled By: riversand963
fbshipit-source-id: ed2873b9587ecc51e24abc35ef2bd3d91fb1ed1b
Summary:
Add some basic test for user-defined timestamp to db_stress. Currently,
read with timestamp always tries to read using the current timestamp.
Due to the per-key timestamp-sequence ordering constraint, we only add timestamp-
related tests to the `NonBatchedOpsStressTest` since this test serializes accesses
to the same key and uses a file to cross-check data correctness.
The timestamp feature is not supported in a number of components, e.g. Merge, SingleDelete,
DeleteRange, CompactionFilter, Readonly instance, secondary instance, SST file ingestion, transaction,
etc. Therefore, db_stress should exit if user enables both timestamp and these features at the same
time. The (currently) incompatible features can be found in
`CheckAndSetOptionsForUserTimestamp`.
This PR also fixes a bug triggered when timestamp is enabled together with
`index_type=kBinarySearchWithFirstKey`. This bug fix will also be in another separate PR
with more unit tests coverage. Fixing it here because I do not want to exclude the index type
from crash test.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8061
Test Plan: make crash_test_with_ts
Reviewed By: jay-zhuang
Differential Revision: D27056282
Pulled By: riversand963
fbshipit-source-id: c3e00ad1023fdb9ebbdf9601ec18270c5e2925a9
Summary:
Since our stress/crash tests by default generate values of size 8, 16, or 24,
it does not make much sense to set `min_blob_size` to 256. The patch
updates the set of potential `min_blob_size` values in the crash test
script and in `db_stress` where it might be set dynamically using
`SetOptions`.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/8085
Test Plan: Ran `make check` and tried the crash test script.
Reviewed By: riversand963
Differential Revision: D27238620
Pulled By: ltamasi
fbshipit-source-id: 4a96f9944b1ed9220d3045c5ab0b34c49009aeee
Summary:
For dictionary compression, we need to collect some representative samples of the data to be compressed, which we use to either generate or train (when `CompressionOptions::zstd_max_train_bytes > 0`) a dictionary. Previously, the strategy was to buffer all the data blocks during flush, and up to the target file size during compaction. That strategy allowed us to randomly pick samples from as wide a range as possible that'd be guaranteed to land in a single output file.
However, some users try to make huge files in memory-constrained environments, where this strategy can cause OOM. This PR introduces an option, `CompressionOptions::max_dict_buffer_bytes`, that limits how much data blocks are buffered before we switch to unbuffered mode (which means creating the per-SST dictionary, writing out the buffered data, and compressing/writing new blocks as soon as they are built). It is not strict as we currently buffer more than just data blocks -- also keys are buffered. But it does make a step towards giving users predictable memory usage.
Related changes include:
- Changed sampling for dictionary compression to select unique data blocks when there is limited availability of data blocks
- Made use of `BlockBuilder::SwapAndReset()` to save an allocation+memcpy when buffering data blocks for building a dictionary
- Changed `ParseBoolean()` to accept an input containing characters after the boolean. This is necessary since, with this PR, a value for `CompressionOptions::enabled` is no longer necessarily the final component in the `CompressionOptions` string.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7970
Test Plan:
- updated `CompressionOptions` unit tests to verify limit is respected (to the extent expected in the current implementation) in various scenarios of flush/compaction to bottommost/non-bottommost level
- looked at jemalloc heap profiles right before and after switching to unbuffered mode during flush/compaction. Verified memory usage in buffering is proportional to the limit set.
Reviewed By: pdillinger
Differential Revision: D26467994
Pulled By: ajkr
fbshipit-source-id: 3da4ef9fba59974e4ef40e40c01611002c861465
Summary:
The patch adds checkpoint support to BlobDB. Blob files are hard linked or
copied, depending on whether the checkpoint directory is on the same filesystem
or not, similarly to table files.
TODO: Add support for blob files to `ExportColumnFamily` and to the checksum
verification logic used by backup/restore.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7959
Test Plan: Ran `make check` and the crash test for a while.
Reviewed By: riversand963
Differential Revision: D26434768
Pulled By: ltamasi
fbshipit-source-id: 994be55a8dc08133028250760fca440d2c7c4dc5
Summary:
The patch adds support for the options related to the new BlobDB implementation
to `db_stress`, including support for dynamically adjusting them using `SetOptions`
when `set_options_one_in` and a new flag `allow_setting_blob_options_dynamically`
are specified. (The latter is used to prevent the options from being enabled when
incompatible features are in use.)
The patch also updates the `db_stress` help messages of the existing stacked BlobDB
related options to clarify that they pertain to the old implementation. In addition, it
adds the new BlobDB to the crash test script. In order to prevent a combinatorial explosion
of jobs and still perform whitebox/blackbox testing (including under ASAN/TSAN/UBSAN),
and to also test BlobDB in conjunction with atomic flush and transactions, the script sets
the BlobDB options in 10% of normal/`cf_consistency`/`txn` crash test runs.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7900
Test Plan: Ran `make check` and `db_stress`/`db_crashtest.py` with various options.
Reviewed By: jay-zhuang
Differential Revision: D26094913
Pulled By: ltamasi
fbshipit-source-id: c2ef3391a05e43a9687f24e297df05f4a5584814
Summary:
This PR adds the foundation classes for key-value integrity protection and the first use case: protecting live updates from the source buffers added to `WriteBatch` through the destination buffer in `MemTable`. The width of the protection info is not yet configurable -- only eight bytes per key is supported. This PR allows users to enable protection by constructing `WriteBatch` with `protection_bytes_per_key == 8`. It does not yet expose a way for users to get integrity protection via other write APIs (e.g., `Put()`, `Merge()`, `Delete()`, etc.).
The foundation classes (`ProtectionInfo.*`) embed the coverage info in their type, and provide `Protect.*()` and `Strip.*()` functions to navigate between types with different coverage. For making bytes per key configurable (for powers of two up to eight) in the future, these classes are templated on the unsigned integer type used to store the protection info. That integer contains the XOR'd result of hashes with independent seeds for all covered fields. For integer fields, the hash is computed on the raw unadjusted bytes, so the result is endian-dependent. The most significant bytes are truncated when the hash value (8 bytes) is wider than the protection integer.
When `WriteBatch` is constructed with `protection_bytes_per_key == 8`, we hold a `ProtectionInfoKVOTC` (i.e., one that covers key, value, optype aka `ValueType`, timestamp, and CF ID) for each entry added to the batch. The protection info is generated from the original buffers passed by the user, as well as the original metadata generated internally. When writing to memtable, each entry is transformed to a `ProtectionInfoKVOTS` (i.e., dropping coverage of CF ID and adding coverage of sequence number), since at that point we know the sequence number, and have already selected a memtable corresponding to a particular CF. This protection info is verified once the entry is encoded in the `MemTable` buffer.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7748
Test Plan:
- an integration test to verify a wide variety of single-byte changes to the encoded `MemTable` buffer are caught
- add to stress/crash test to verify it works in variety of configs/operations without intentional corruption
- [deferred] unit tests for `ProtectionInfo.*` classes for edge cases like KV swap, `SliceParts` and `Slice` APIs are interchangeable, etc.
Reviewed By: pdillinger
Differential Revision: D25754492
Pulled By: ajkr
fbshipit-source-id: e481bac6c03c2ab268be41359730f1ceb9964866
Summary:
Inject the random write error to stress test, it requires set reopen=0 and disable_wal=true.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7653
Test Plan: pass db_stress and python3 db_crashtest.py blackbox
Reviewed By: ajkr
Differential Revision: D25354132
Pulled By: zhichao-cao
fbshipit-source-id: 44721104eecb416e27f65f854912c40e301dd669
Summary:
Added experimental public API for Ribbon filter:
NewExperimentalRibbonFilterPolicy(). This experimental API will
take a "Bloom equivalent" bits per key, and configure the Ribbon
filter for the same FP rate as Bloom would have but ~30% space
savings. (Note: optimize_filters_for_memory is not yet implemented
for Ribbon filter. That can be added with no effect on schema.)
Internally, the Ribbon filter is configured using a "one_in_fp_rate"
value, which is 1 over desired FP rate. For example, use 100 for 1%
FP rate. I'm expecting this will be used in the future for configuring
Bloom-like filters, as I expect people to more commonly hold constant
the filter accuracy and change the space vs. time trade-off, rather than
hold constant the space (per key) and change the accuracy vs. time
trade-off, though we might make that available.
### Benchmarking
```
$ ./filter_bench -impl=2 -quick -m_keys_total_max=200 -average_keys_per_filter=100000 -net_includes_hashing
Building...
Build avg ns/key: 34.1341
Number of filters: 1993
Total size (MB): 238.488
Reported total allocated memory (MB): 262.875
Reported internal fragmentation: 10.2255%
Bits/key stored: 10.0029
----------------------------
Mixed inside/outside queries...
Single filter net ns/op: 18.7508
Random filter net ns/op: 258.246
Average FP rate %: 0.968672
----------------------------
Done. (For more info, run with -legend or -help.)
$ ./filter_bench -impl=3 -quick -m_keys_total_max=200 -average_keys_per_filter=100000 -net_includes_hashing
Building...
Build avg ns/key: 130.851
Number of filters: 1993
Total size (MB): 168.166
Reported total allocated memory (MB): 183.211
Reported internal fragmentation: 8.94626%
Bits/key stored: 7.05341
----------------------------
Mixed inside/outside queries...
Single filter net ns/op: 58.4523
Random filter net ns/op: 363.717
Average FP rate %: 0.952978
----------------------------
Done. (For more info, run with -legend or -help.)
```
168.166 / 238.488 = 0.705 -> 29.5% space reduction
130.851 / 34.1341 = 3.83x construction time for this Ribbon filter vs. lastest Bloom filter (could make that as little as about 2.5x for less space reduction)
### Working around a hashing "flaw"
bloom_test discovered a flaw in the simple hashing applied in
StandardHasher when num_starts == 1 (num_slots == 128), showing an
excessively high FP rate. The problem is that when many entries, on the
order of number of hash bits or kCoeffBits, are associated with the same
start location, the correlation between the CoeffRow and ResultRow (for
efficiency) can lead to a solution that is "universal," or nearly so, for
entries mapping to that start location. (Normally, variance in start
location breaks the effective association between CoeffRow and
ResultRow; the same value for CoeffRow is effectively different if start
locations are different.) Without kUseSmash and with num_starts > 1 (thus
num_starts ~= num_slots), this flaw should be completely irrelevant. Even
with 10M slots, the chances of a single slot having just 16 (or more)
entries map to it--not enough to cause an FP problem, which would be local
to that slot if it happened--is 1 in millions. This spreadsheet formula
shows that: =1/(10000000*(1 - POISSON(15, 1, TRUE)))
As kUseSmash==false (the setting for Standard128RibbonBitsBuilder) is
intended for CPU efficiency of filters with many more entries/slots than
kCoeffBits, a very reasonable work-around is to disallow num_starts==1
when !kUseSmash, by making the minimum non-zero number of slots
2*kCoeffBits. This is the work-around I've applied. This also means that
the new Ribbon filter schema (Standard128RibbonBitsBuilder) is not
space-efficient for less than a few hundred entries. Because of this, I
have made it fall back on constructing a Bloom filter, under existing
schema, when that is more space efficient for small filters. (We can
change this in the future if we want.)
TODO: better unit tests for this case in ribbon_test, and probably
update StandardHasher for kUseSmash case so that it can scale nicely to
small filters.
### Other related changes
* Add Ribbon filter to stress/crash test
* Add Ribbon filter to filter_bench as -impl=3
* Add option string support, as in "filter_policy=experimental_ribbon:5.678;"
where 5.678 is the Bloom equivalent bits per key.
* Rename internal mode BloomFilterPolicy::kAuto to kAutoBloom
* Add a general BuiltinFilterBitsBuilder::CalculateNumEntry based on
binary searching CalculateSpace (inefficient), so that subclasses
(especially experimental ones) don't have to provide an efficient
implementation inverting CalculateSpace.
* Minor refactor FastLocalBloomBitsBuilder for new base class
XXH3pFilterBitsBuilder shared with new Standard128RibbonBitsBuilder,
which allows the latter to fall back on Bloom construction in some
extreme cases.
* Mostly updated bloom_test for Ribbon filter, though a test like
FullBloomTest::Schema is a next TODO to ensure schema stability
(in case this becomes production-ready schema as it is).
* Add some APIs to ribbon_impl.h for configuring Ribbon filters.
Although these are reasonably covered by bloom_test, TODO more unit
tests in ribbon_test
* Added a "tool" FindOccupancyForSuccessRate to ribbon_test to get data
for constructing the linear approximations in GetNumSlotsFor95PctSuccess.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7658
Test Plan:
Some unit tests updated but other testing is left TODO. This
is considered experimental but laying down schema compatibility as early
as possible in case it proves production-quality. Also tested in
stress/crash test.
Reviewed By: jay-zhuang
Differential Revision: D24899349
Pulled By: pdillinger
fbshipit-source-id: 9715f3e6371c959d923aea8077c9423c7a9f82b8
Summary:
crash tests donot run in DEBUG_MODE=0 on tmpfs when
use_direct_reads/use_direct_io_for_flush_and_compaction is set randomly because
direct I/O is not supported on tmpfs and tests exit.
Fix: Sanitize direct I/O read options in DEBUG_LEVEL=0 so that crash
tests can run in tmpfs. When mmap_reads is set, direct I/O reads options are
unset so we can sanitize direct I/O reads options in case of tmpfs as well.
Pull Request resolved: https://github.com/facebook/rocksdb/pull/7643
Test Plan:
1. export DEBUG_LEVEL=0; export TEST_TMPDIR="/dev/shm";
export CRASH_TEST_EXT_ARGS="--use_direct_reads=1 --mmap_read=0";
make crash_test -j64
2. In DEBUG_LEVEL=1 mode: make crash_test -j64
Reviewed By: jay-zhuang
Differential Revision: D24766550
Pulled By: akankshamahajan15
fbshipit-source-id: 021720b2343c12c72004f84b26147625d3991d9e
Hui Xiao