Summary:
We introduced smallest_prep optimization in this commit b225de7e10, which enables storing the smallest uncommitted sequence number along with the snapshot. This enables the readers that read from the snapshot to skip further checks and safely assumed the data is committed if its sequence number is less than smallest uncommitted when the snapshot was taken. The problem was that smallest uncommitted and the snapshot must be taken atomically, and the lack of atomicity had led to readers using a smallest uncommitted after the snapshot was taken and hence mistakenly skipping some data.
This patch fixes the problem by i) separating the process of removing of prepare entries from the AddCommitted function, ii) removing the prepare entires AFTER the committed sequence number is published, iii) getting smallest uncommitted (from the prepare list) BEFORE taking a snapshot. This guarantees that the smallest uncommitted that is accompanied with a snapshot is less than or equal of such number if it was obtained atomically.
Tested by running MySQLStyleTransactionTest/MySQLStyleTransactionTest.TransactionStressTest that was failing sporadically.
Closes https://github.com/facebook/rocksdb/pull/3703
Differential Revision: D7581934
Pulled By: maysamyabandeh
fbshipit-source-id: dc9d6f4fb477eba75d4d5927326905b548a96a32
Summary:
Adding some stats that would be helpful to monitor if the DB has gone to unlikely stats that would hurt the performance. These are mostly when we end up needing to acquire a mutex.
Closes https://github.com/facebook/rocksdb/pull/3683
Differential Revision: D7529393
Pulled By: maysamyabandeh
fbshipit-source-id: f7d36279a8f39bd84d8ddbf64b5c97f670c5d6d9
Summary:
When `max_valid_backups_to_open` is set, the `BackupEngine` doesn't know about the files referenced by existing backups. This PR prevents us from deleting valid files when that option is set, in cases where we are unable to accurately determine refcount. There are warnings logged when we may miss deleting unreferenced files, and a recommendation in the header for users to periodically unset this option and run a full `GarbageCollect`.
Closes https://github.com/facebook/rocksdb/pull/3518
Differential Revision: D7008331
Pulled By: ajkr
fbshipit-source-id: 87907f964dc9716e229d08636a895d2fc7b72305
Summary:
Make blob_dump tool able to show uncompressed values if the blob file is compressed. Also show total compressed vs. raw size at the end if --show_summary is provided.
Closes https://github.com/facebook/rocksdb/pull/3633
Differential Revision: D7348926
Pulled By: yiwu-arbug
fbshipit-source-id: ca709cb4ed5cf6a550ff2987df8033df81516f8e
Summary:
This change models Optimistic Tx db after Pessimistic TX db. The motivation for this change is to make the ptr polymorphic so it can be held by the same raw or smart ptr.
Currently, due to the inheritance of the Opt Tx db not being rooted in the manner of Pess Tx from a single DB root it is more difficult to write clean code and have clear ownership of the database in cases when options dictate instantiate of plan DB, Pess Tx DB or Opt tx db.
Closes https://github.com/facebook/rocksdb/pull/3566
Differential Revision: D7184502
Pulled By: yiwu-arbug
fbshipit-source-id: 31d06efafd79497bb0c230e971857dba3bd962c3
Summary:
The is an optimization to reduce lookup in the CommitCache when querying IsInSnapshot. The optimization takes the smallest uncommitted data at the time that the snapshot was taken and if the sequence number of the read data is lower than that number it assumes the data as committed.
To implement this optimization two changes are required: i) The AddPrepared function must be called sequentially to avoid out of order insertion in the PrepareHeap (otherwise the top of the heap does not indicate the smallest prepare in future too), ii) non-2PC transactions also call AddPrepared if they do not commit in one step.
Closes https://github.com/facebook/rocksdb/pull/3649
Differential Revision: D7388630
Pulled By: maysamyabandeh
fbshipit-source-id: b79506238c17467d590763582960d4d90181c600
Summary:
Currently if the CommitTimeWriteBatch is set to be used only as a state that is required only for recovery , the user cannot see that in DB until it is restarted. This while the state is already inserted into the DB after the memtable flush. It would be useful for debugging if make this state visible to the user after the flush by committing it. The patch does it by a invoking a callback that does the commit on the recoverable state.
Closes https://github.com/facebook/rocksdb/pull/3661
Differential Revision: D7424577
Pulled By: maysamyabandeh
fbshipit-source-id: 137f9408662f0853938b33fa440f27f04c1bbf5c
Summary:
Current commit cache size is 2^21. This was due to a type. With 2^23 commit entries we can have transactions as long as 64s without incurring the cost of having them evicted from the commit cache before their commit. Here is the math:
2^23 / 2 (one out of two seq numbers are for commit) / 2^16 TPS = 2^6 = 64s
Closes https://github.com/facebook/rocksdb/pull/3657
Differential Revision: D7411211
Pulled By: maysamyabandeh
fbshipit-source-id: e7cacf40579f3acf940643d8a1cfe5dd201caa35
Summary:
Currently log_writer->AddRecord in WriteImpl is protected from concurrent calls via FlushWAL only if two_write_queues_ option is set. The patch fixes the problem by i) skip log_writer->AddRecord in FlushWAL if manual_wal_flush is not set, ii) protects log_writer->AddRecord in WriteImpl via log_write_mutex_ if manual_wal_flush_ is set but two_write_queues_ is not.
Fixes#3599
Closes https://github.com/facebook/rocksdb/pull/3656
Differential Revision: D7405608
Pulled By: maysamyabandeh
fbshipit-source-id: d6cc265051c77ae49c7c6df4f427350baaf46934
Summary:
Currently AddPrepared is performed only on the first sub-batch if there are duplicate keys in the write batch. This could cause a problem if the transaction takes too long to commit and the seq number of the first sub-patch moved to old_prepared_ but not the seq of the later ones. The patch fixes this by calling AddPrepared for all sub-patches.
Closes https://github.com/facebook/rocksdb/pull/3651
Differential Revision: D7388635
Pulled By: maysamyabandeh
fbshipit-source-id: 0ccd80c150d9bc42fe955e49ddb9d7ca353067b4
Summary:
This commit fixes a race condition on calling SetLastPublishedSequence. The function must be called only from the 2nd write queue when two_write_queues is enabled. However there was a bug that would also call it from the main write queue if CommitTimeWriteBatch is provided to the commit request and yet use_only_the_last_commit_time_batch_for_recovery optimization is not enabled. To fix that we penalize the commit request in such cases by doing an additional write solely to publish the seq number from the 2nd queue.
Closes https://github.com/facebook/rocksdb/pull/3641
Differential Revision: D7361508
Pulled By: maysamyabandeh
fbshipit-source-id: bf8f7a27e5cccf5425dccbce25eb0032e8e5a4d7
Summary:
The 10MB buffer in BackupEngineImpl::BackupMeta::StoreToFile can be corrupted with a large number of files. Added a check to determine current buffer length and append data to file if buffer becomes full.
Resolves https://github.com/facebook/rocksdb/issues/3228
Closes https://github.com/facebook/rocksdb/pull/3636
Differential Revision: D7354160
Pulled By: ajkr
fbshipit-source-id: eec12d38095a0d17551a4aaee52b99d30a555722
Summary:
* Fix BlobDBImpl::GCFileAndUpdateLSM doesn't close the new file, and the new file will not be able to be garbage collected later.
* Fix BlobDBImpl::GCFileAndUpdateLSM doesn't copy over metadata from old file to new file.
Closes https://github.com/facebook/rocksdb/pull/3639
Differential Revision: D7355092
Pulled By: yiwu-arbug
fbshipit-source-id: 4fa3594ac5ce376bed1af04a545c532cfc0088c4
Summary:
`Writer::WriteBuffer` was always called at the beginning of checkpoint/backup. But that log writer has no internal synchronization, which meant the same buffer could be flushed twice in a race condition case, causing a WAL entry to be duplicated. Then subsequent WAL entries would be at unexpected offsets, causing the 32KB block boundaries to be overlapped and manifesting as a corruption.
This PR fixes the behavior to only use `WriteBuffer` (via `FlushWAL`) in checkpoint/backup when manual WAL flush is enabled. In that case, users are responsible for providing synchronization between WAL flushes. We can also consider removing the call entirely.
Closes https://github.com/facebook/rocksdb/pull/3603
Differential Revision: D7277447
Pulled By: ajkr
fbshipit-source-id: 1b15bd7fd930511222b075418c10de0aaa70a35a
Summary:
This diff handles cases where compaction causes an ENOSPC error.
This does not handle corner cases where another background job is started while compaction is running, and the other background job triggers ENOSPC, although we do allow the user to provision for these background jobs with SstFileManager::SetCompactionBufferSize.
It also does not handle the case where compaction has finished and some other background job independently triggers ENOSPC.
Usage: Functionality is inside SstFileManager. In particular, users should set SstFileManager::SetMaxAllowedSpaceUsage, which is the reference highwatermark for determining whether to cancel compactions.
Closes https://github.com/facebook/rocksdb/pull/3449
Differential Revision: D7016941
Pulled By: amytai
fbshipit-source-id: 8965ab8dd8b00972e771637a41b4e6c645450445
Summary:
This patch addressed several issues.
Portability including db_test std::thread -> port::Thread Cc: @
and %z to ROCKSDB portable macro. Cc: maysamyabandeh
Implement Env::AreFilesSame
Make the implementation of file unique number more robust
Get rid of C-runtime and go directly to Windows API when dealing
with file primitives.
Implement GetSectorSize() and aling unbuffered read on the value if
available.
Adjust Windows Logger for the new interface, implement CloseImpl() Cc: anand1976
Fix test running script issue where $status var was of incorrect scope
so the failures were swallowed and not reported.
DestroyDB() creates a logger and opens a LOG file in the directory
being cleaned up. This holds a lock on the folder and the cleanup is
prevented. This fails one of the checkpoin tests. We observe the same in production.
We close the log file in this change.
Fix DBTest2.ReadAmpBitmapLiveInCacheAfterDBClose failure where the test
attempts to open a directory with NewRandomAccessFile which does not
work on Windows.
Fix DBTest.SoftLimit as it is dependent on thread timing. CC: yiwu-arbug
Closes https://github.com/facebook/rocksdb/pull/3552
Differential Revision: D7156304
Pulled By: siying
fbshipit-source-id: 43db0a757f1dfceffeb2b7988043156639173f5b
Summary:
Improving blob db FIFO eviction with the following changes,
* Change blob_dir_size to max_db_size. Take into account SST file size when computing DB size.
* FIFO now only take into account live sst files and live blob files. It is normal for disk usage to go over max_db_size because there are obsolete sst files and blob files pending deletion.
* FIFO eviction now also evict TTL blob files that's still open. It doesn't evict non-TTL blob files.
* If FIFO is triggered, it will pass an expiration and the current sequence number to compaction filter. Compaction filter will then filter inlined keys to evict those with an earlier expiration and smaller sequence number. So call LSM FIFO.
* Compaction filter also filter those blob indexes where corresponding blob file is gone.
* Add an event listener to listen compaction/flush event and update sst file size.
* Implement DB::Close() to make sure base db, as well as event listener and compaction filter, destruct before blob db.
* More blob db statistics around FIFO.
* Fix some locking issue when accessing a blob file.
Closes https://github.com/facebook/rocksdb/pull/3556
Differential Revision: D7139328
Pulled By: yiwu-arbug
fbshipit-source-id: ea5edb07b33dfceacb2682f4789bea61de28bbfa
Summary:
Move DuplicateDetector and SetComparator to its own header file in util. It would also address a complaint in the unity test.
Closes https://github.com/facebook/rocksdb/pull/3567
Differential Revision: D7163268
Pulled By: maysamyabandeh
fbshipit-source-id: 6ddf82773473646dbbc1284ae601a78c4907c778
Summary:
Fix the following bugs:
- During recovery a duplicate key was inserted twice into the write batch of the recovery transaction,
once when the memtable returns false (because it was duplicates) and once for the 2nd attempt. This would result into different SubBatch count measured when the recovered transactions is committing.
- If a cf is flushed during recovery the memtable is not available to assist in detecting the duplicate key. This could result into not advancing the sequence number when iterating over duplicate keys of a flushed cf and hence inserting the next key with the wrong sequence number.
- SubBacthCounter would reset the comparator to default comparator after the first duplicate key. The 2nd duplicate key hence would have gone through a wrong comparator and not being detected.
Closes https://github.com/facebook/rocksdb/pull/3562
Differential Revision: D7149440
Pulled By: maysamyabandeh
fbshipit-source-id: 91ec317b165f363f5d11ff8b8c47c81cebb8ed77
Summary:
Red diff to remove existing implementation of garbage collection. The current approach is reference counting kind of approach and require a lot of effort to get the size counter right on compaction and deletion. I'm going to go with a simple mark-sweep kind of approach and will send another PR for that.
CompactionEventListener was added solely for blob db and it adds complexity and overhead to compaction iterator. Removing it as well.
Closes https://github.com/facebook/rocksdb/pull/3551
Differential Revision: D7130190
Pulled By: yiwu-arbug
fbshipit-source-id: c3a375ad2639a3f6ed179df6eda602372cc5b8df
Summary:
The zeroed entries were not removed from prepared_section_completed_ map. This patch adds a unit test to show the problem and fixes that by refactoring the code. The new code is more efficient since i) it uses two separate mutex to avoid contention between commit and prepare threads, ii) it uses a sorted vector for maintaining uniq log entires with prepare which avoids a very large heap with many duplicate entries.
Closes https://github.com/facebook/rocksdb/pull/3545
Differential Revision: D7106071
Pulled By: maysamyabandeh
fbshipit-source-id: b3ae17cb6cd37ef10b6b35e0086c15c758768a48
Summary:
Make use of the index in WriteBatchWithIndex to also count the number of sub-batches. This eliminates the need to separately scan the batch to count the number of sub-batches once a duplicate key is detected.
Closes https://github.com/facebook/rocksdb/pull/3529
Differential Revision: D7049947
Pulled By: maysamyabandeh
fbshipit-source-id: 81cbf12c4e662541c772c7265a8f91631e25c7cd
Summary:
Use the rsync tempfile naming convention in our `BackupEngine`. The temp file follows the format, `.<filename>.<suffix>`, which is later renamed to `<filename>`. We fix `tmp` as the `<suffix>` as we don't need to use random bytes for now. The benefit is gluster treats this tempfile naming convention specially and applies hashing only to `<filename>`, so the file won't need to be linked or moved when it's renamed. Our gluster team suggested this will make things operationally easier.
Closes https://github.com/facebook/rocksdb/pull/3463
Differential Revision: D6893333
Pulled By: ajkr
fbshipit-source-id: fd7622978f4b2487fce33cde40dd3124f16bcaa8
Summary:
Under a certain sequence of accessing PreparedHeap, there was a bug that would not successfully empty the heap. This would result in performance issues when the heap content is moved to old_prepared_ after max_evicted_seq_ advances the orphan prepared sequence numbers. The patch fixed the bug and add more unit tests. It also does more logging when the unlikely scenarios are faced
Closes https://github.com/facebook/rocksdb/pull/3526
Differential Revision: D7038486
Pulled By: maysamyabandeh
fbshipit-source-id: f1e40bea558f67b03d2a29131fcb8734c65fce97
Summary:
In case it is found that a key is already marked as locked in a
stripe's map of locked keys, it is not necessary to look it up
again using `std::unordered_map<std::string, ...>::at(size_t)`.
Instead, we can use the already found position using the iterator
produced by the previous `find` operation. Reusing the iterator
will avoid having to hash the key again and do additional "random"
memory lookups in the map of keys (though the data will very
likely sit available in caches here already due to the previous
find operation)
Closes https://github.com/facebook/rocksdb/pull/3505
Differential Revision: D7036446
Pulled By: sagar0
fbshipit-source-id: cced51547b2bd2d49394f6bc8c5896f09fa80f68
Summary:
- made `CreateCheckpoint` properly return `InvalidArgument` when called with an empty directory. Previously it triggered an assertion failure due to a bug in the logic.
- made `ldb` set empty `checkpoint_dir` if that's what the user specifies, so that we can use it to properly test `CreateCheckpoint` in the future.
Differential Revision: D6874562
fbshipit-source-id: dcc1bd41768261d9338987fa7711444289707ed7
Summary:
Add a static cast to perform the left shift as with an unsigned type.
make ubsan_check
Closes https://github.com/facebook/rocksdb/pull/3517
Reviewed By: sagar0
Differential Revision: D7016044
Pulled By: igorsugak
fbshipit-source-id: baf72f6197edd8f7220d010b15a23d6de6a72c49
Summary:
These are optimization that we applied to improve sysbech's update_noindex performance.
1. Make use of LIKELY compiler hint
2. Move std::atomic so the subclass
3. Make use of skip_prepared in non-2pc transactions.
Closes https://github.com/facebook/rocksdb/pull/3512
Differential Revision: D7000075
Pulled By: maysamyabandeh
fbshipit-source-id: 1ab8292584df1f6305a4992973fb1b7933632181
Summary:
- removed a few unneeded variables
- fused some variable declarations and their assignments
- fixed right-trimming code in string_util.cc to not underflow
- simplifed an assertion
- move non-nullptr check assertion before dereferencing of that pointer
- pass an std::string function parameter by const reference instead of by value (avoiding potential copy)
Closes https://github.com/facebook/rocksdb/pull/3507
Differential Revision: D7004679
Pulled By: sagar0
fbshipit-source-id: 52944952d9b56dfcac3bea3cd7878e315bb563c4
Summary:
Somehow the indentation was incorrect in this file.
The only change in this PR is to get it right again in order to make the code more readable.
Please reject if you think it's not worth it.
Closes https://github.com/facebook/rocksdb/pull/3504
Differential Revision: D6996011
Pulled By: miasantreble
fbshipit-source-id: 060514a3a8c910d34bad795b36eb4d278512b154
Summary:
TransactionDB::Write can receive some optimization hints from the user. One is to skip the concurrency control mechanism. WritePreparedTxnDB is currently ignoring such hints. This patch optimizes WritePreparedTxnDB::Write for skip_concurrency_control and skip_duplicate_key_check hints.
Closes https://github.com/facebook/rocksdb/pull/3496
Differential Revision: D6971784
Pulled By: maysamyabandeh
fbshipit-source-id: cbab10ad538fa2b8bcb47e37c77724afe6e30f03
Summary:
WritePreparedTransactionTest has the UBSAN error because the wrong order of its parent class construction. Fix it.
Closes https://github.com/facebook/rocksdb/pull/3478
Differential Revision: D6928975
Pulled By: siying
fbshipit-source-id: 13edfd5cb9cf73f1ac5ae3b6f53061d32783733d
Summary:
Compared to DB::Write, TransactionDB::Write has the additional overhead of creating and initializing an internal transaction object, as well as the overhead of locking/unlocking the keys. This patch extends the TransactionDB::Write with an skip_cc option to allow the users to indicate that the write batch do not conflict with others and the concurrency control and its overhead can be skipped. TransactionDB::Write by default calls DB::Write when skip_cc is set, which works for WriteCommitted WritePolicy. Any other flavor of TransactionDB that is not compatible with this default behavior (such as WritePreparedTxnDB) can extend ::Write and implement their own approach for taking into account the skip_cc optimization.
Closes https://github.com/facebook/rocksdb/pull/3457
Differential Revision: D6877318
Pulled By: maysamyabandeh
fbshipit-source-id: 56f4e21db87ff71492db4e376fb7c2b03dfeab6b
Summary:
Deletes the transaction object at the end of the test.
Verified by:
- COMPILE_WITH_ASAN=1 make -j32 transaction_test
- ./transaction_test --gtest_filter="DBA**Duplicate*"
Closes https://github.com/facebook/rocksdb/pull/3470
Differential Revision: D6916473
Pulled By: maysamyabandeh
fbshipit-source-id: 8303df25408635d5d3ac2b25f309a3d15957c937
Summary:
This patch takes advantage of memtable being able to detect duplicate <key,seq> and returning TryAgain to handle duplicate keys in WritePrepared Txns. Through WriteBatchWithIndex's index it detects existence of at least a duplicate key in the write batch. If duplicate key was reported, it then pays the cost of counting the number of sub-patches by iterating over the write batch and pass it to DBImpl::Write. DB will make use of the provided batch_count to assign proper sequence numbers before sending them to the WAL. When later inserting the batch to the memtable, it increases the seq each time memtbale reports a duplicate (a sub-patch in our counting) and tries again.
Closes https://github.com/facebook/rocksdb/pull/3455
Differential Revision: D6873699
Pulled By: maysamyabandeh
fbshipit-source-id: db8487526c3a5dc1ddda0ea49f0f979b26ae648d
Summary:
Without this patch, ubsan_check is currently failing with this error:
```
utilities/transactions/write_prepared_transaction_test.cc:369:63: runtime error: member call on address 0x0000051649f8 which does not point to an object of type 'WithParamInterface'
0x0000051649f8: note: object has invalid vptr
```
Tested by `COMPILE_WITH_UBSAN=1 make -j32 transaction_test` and running `./write_prepared_transaction_test --gtest_filter=TwoWriteQueues/SnapshotConcurrentAccessTest.SnapshotConcurrentAccessTest1/0`
Closes https://github.com/facebook/rocksdb/pull/3444
Differential Revision: D6850087
Pulled By: maysamyabandeh
fbshipit-source-id: 5b254da8504b8757f7aec8a820ad464154da1a1d
Summary:
previously if `checkpoint_dir` contained a trailing slash, we'd attempt to create the `.tmp` directory under `checkpoint_dir` due to simply concatenating `checkpoint_dir + ".tmp"`. This failed because `checkpoint_dir` hadn't been created yet and our directory creation is non-recursive. This PR fixes the issue by always creating the `.tmp` directory in the same parent as `checkpoint_dir` by stripping trailing slashes before concatenating.
Closes https://github.com/facebook/rocksdb/pull/3275
Differential Revision: D6574952
Pulled By: ajkr
fbshipit-source-id: a6daa6777a901eac2460cd0140c9515f7241aefc
Maysam Yabandeh