Summary:
The Arena construction/destruction introduced significant overhead to read-heavy workload just by creating empty vectors for its blocks, so avoid it in RangeDelAggregator.
Closes https://github.com/facebook/rocksdb/pull/1547
Differential Revision: D4207781
Pulled By: ajkr
fbshipit-source-id: 9d1c130
Summary:
Since a RangeDelAggregator is created for each read request, these heap-allocating member variables were consuming significant CPU (~3% total) which slowed down request throughput. The map and pinning manager are only necessary when range deletions exist, so we can defer their initialization until the first range deletion is encountered. Currently lazy initialization is done for reads only since reads pass us a single snapshot, which is easier to store on the stack for later insertion into the map than the vector passed to us by flush or compaction.
Note the Arena member variable is still expensive, I will figure out what to do with it in a subsequent diff. It cannot be lazily initialized because we currently use this arena even to allocate empty iterators, which is necessary even when no range deletions exist.
Closes https://github.com/facebook/rocksdb/pull/1539
Differential Revision: D4203488
Pulled By: ajkr
fbshipit-source-id: 3b36279
Summary:
Remove the ticker count because:
* Having to reset the ticker count in WriteImpl is ineffiecent;
* It doesn't make sense to have it as a ticker count if multiple db
instance share a statistics object.
Closes https://github.com/facebook/rocksdb/pull/1531
Differential Revision: D4194442
Pulled By: yiwu-arbug
fbshipit-source-id: e2110a9
Summary:
Currently the compaction stats are printed to stdout. We want to export the compaction stats in a map format so that the upper layer apps (e.g., MySQL) could present
the stats in any format required by the them.
Closes https://github.com/facebook/rocksdb/pull/1477
Differential Revision: D4149836
Pulled By: maysamyabandeh
fbshipit-source-id: b3df19f
Summary:
Originally sequence ids were calculated, in recovery, based off of the first seqid found if the first log recovered. The working seqid was then incremented from that value based on every insertion that took place. This was faulty because of the potential for missing log files or inserts that skipped the WAL. The current recovery scheme grabs sequence from current recovering batch and increments using memtableinserter to track how many actual inserts take place. This works for 2PC batches as well scenarios where some logs are missing or inserts that skip the WAL.
Closes https://github.com/facebook/rocksdb/pull/1486
Differential Revision: D4156064
Pulled By: reidHoruff
fbshipit-source-id: a6da8d9
Summary:
copied from: 5ebfd2623a
Opening existing RocksDB attempts recovery from log files, which uses
wrong sequence number to create the memtable. This is a regression
introduced in change a400336.
This change includes a test demonstrating the problem, without the fix
the test fails with "Operation failed. Try again.: Transaction could not
check for conflicts for operation at SequenceNumber 1 as the MemTable
only contains changes newer than SequenceNumber 2. Increasing the value
of the max_write_buffer_number_to_maintain option could reduce the
frequency of this error"
This change is a joint effort by Peter 'Stig' Edwards thatsafunnyname
and me.
Closes https://github.com/facebook/rocksdb/pull/1458
Differential Revision: D4143791
Pulled By: reidHoruff
fbshipit-source-id: 5a25033
Summary:
A deadlock is possible if this happen
(1) Writer thread is stopped because it's waiting for compaction to finish
(2) Compaction is waiting for current IngestExternalFile() calls to finish
(3) IngestExternalFile() is waiting to be able to acquire the writer thread
(4) WriterThread is held by stopped writes that are waiting for compactions to finish
This patch fix the issue by not incrementing num_running_ingest_file_ except when we acquire the writer thread.
This patch include a unittest to reproduce the described scenario
Closes https://github.com/facebook/rocksdb/pull/1480
Differential Revision: D4151646
Pulled By: IslamAbdelRahman
fbshipit-source-id: 09b39db
Summary:
Note: reviewed in https://reviews.facebook.net/D65115
- DBIter maintains a range tombstone accumulator. We don't cleanup obsolete tombstones yet, so if the user seeks back and forth, the same tombstones would be added to the accumulator multiple times.
- DBImpl::NewInternalIterator() (used to make DBIter's underlying iterator) adds memtable/L0 range tombstones, L1+ range tombstones are added on-demand during NewSecondaryIterator() (see D62205)
- DBIter uses ShouldDelete() when advancing to check whether keys are covered by range tombstones
Closes https://github.com/facebook/rocksdb/pull/1464
Differential Revision: D4131753
Pulled By: ajkr
fbshipit-source-id: be86559
Summary:
During Get()/MultiGet(), build up a RangeDelAggregator with range
tombstones as we search through live memtable, immutable memtables, and
SST files. This aggregator is then used by memtable.cc's SaveValue() and
GetContext::SaveValue() to check whether keys are covered.
added tests for Get on memtables/files; end-to-end tests mainly in https://reviews.facebook.net/D64761
Closes https://github.com/facebook/rocksdb/pull/1456
Differential Revision: D4111271
Pulled By: ajkr
fbshipit-source-id: 6e388d4
Summary:
Changed BuildTable() (used for flush) to (1) add range
tombstones to the aggregator, which is used by CompactionIterator to
determine which keys can be removed; and (2) add aggregator's range
tombstones to the table that is output for the flush.
Closes https://github.com/facebook/rocksdb/pull/1438
Differential Revision: D4100025
Pulled By: ajkr
fbshipit-source-id: cb01a70
Summary:
Currently the compression suppport and fast CRC support information is printed as info level. They should be in the same level as options, which is header level.
Also add ZSTD to this printing.
Closes https://github.com/facebook/rocksdb/pull/1448
Differential Revision: D4106608
Pulled By: yiwu-arbug
fbshipit-source-id: cb9a076
Summary:
DB Stats now are truncated if there are too many CFs. Extend the buffer size to allow more to be printed out. Also, separate out malloc to another log line.
Closes https://github.com/facebook/rocksdb/pull/1439
Differential Revision: D4100943
Pulled By: yiwu-arbug
fbshipit-source-id: 79f7218
Summary:
Currently, we don't use options.compression_per_level[0] as the compression style for L0 compression type, unless it is None. This behavior
doesn't look like on purpose. This diff will make sure L0 compress using the style of options.compression_per_level[0].
Reviewed and accepted in: https://reviews.facebook.net/D65607
Closes https://github.com/facebook/rocksdb/pull/1435
Differential Revision: D4099368
Pulled By: siying
fbshipit-source-id: cfbbdcd
Summary: Siying suggested to keep old code for normal mode prev() for safety
Test Plan: make check -j64
Reviewers: yiwu, andrewkr, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D65439
Summary:
change ioptions.comparator to user_comparator instread of internal_comparator.
Also change Comparator* to InternalKeyComparator* to make its type explicitly.
Test Plan: make all check -j64
Reviewers: andrewkr, sdong, yiwu
Reviewed By: yiwu
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D65121
Summary:
Changes in the diff
API changes:
- Introduce IngestExternalFile to replace AddFile (I think this make the API more clear)
- Introduce IngestExternalFileOptions (This struct will encapsulate the options for ingesting the external file)
- Deprecate AddFile() API
Logic changes:
- If our file overlap with the memtable we will flush the memtable
- We will find the first level in the LSM tree that our file key range overlap with the keys in it
- We will find the lowest level in the LSM tree above the the level we found in step 2 that our file can fit in and ingest our file in it
- We will assign a global sequence number to our new file
- Remove AddFile restrictions by using global sequence numbers
Other changes:
- Refactor all AddFile logic to be encapsulated in ExternalSstFileIngestionJob
Test Plan:
unit tests (still need to add more)
addfile_stress (https://reviews.facebook.net/D65037)
Reviewers: yiwu, andrewkr, lightmark, yhchiang, sdong
Reviewed By: sdong
Subscribers: jkedgar, hcz, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D65061
Summary:
Previously the WAL files that were avoided during recovery would never
be considered for deletion. That was because alive_log_files_ was only
populated when log files are created. This diff further populates
alive_log_files_ with existing log files that aren't flushed during recovery,
such that FindObsoleteFiles() can find them later.
Depends on D64053.
Test Plan: new unit test, verifies it fails before this change and passes after
Reviewers: sdong, IslamAbdelRahman, yiwu
Reviewed By: yiwu
Subscribers: leveldb, dhruba, andrewkr
Differential Revision: https://reviews.facebook.net/D64059
Summary:
Issue scenario:
(1) We have 3 files in L1 and we issue a compaction that will compact them into 1 file in L2
(2) While compaction (1) is running, we flush a file into L0 and trigger another compaction that decide to move this file to L1 and then move it again to L2 (this file don't overlap with any other files)
(3) compaction (1) finishes and install the file it generated in L2, but this file overlap with the file we generated in (2) so we break the LSM consistency
Looks like this issue can be triggered by using non-exclusive manual compaction or AddFile()
Test Plan: unit tests
Reviewers: sdong
Reviewed By: sdong
Subscribers: hermanlee4, jkedgar, andrewkr, dhruba, yoshinorim
Differential Revision: https://reviews.facebook.net/D64947
Summary: I accidentally left out these changes from my commit of D64053 due to
messing up the merge conflict resolution.
Test Plan: ./db_wal_test
Reviewers:
Subscribers:
Tasks:
Blame Revision: D64053
Summary:
This reverts commit 9e4aa798c3,
which doesn't handle all cases (see inline comment).
I reimplemented the logic as suggested in the initial PR: https://github.com/facebook/rocksdb/pull/1313.
This approach has two benefits:
- All the parsing/filtering of full_scan_candidate_files is kept together in PurgeObsoleteFiles.
- We only need to check whether log file is recycled in one place where we've already determined it's a log file
Test Plan:
new unit test, verified fails before the original fix, still passes
now.
Reviewers: IslamAbdelRahman, yiwu, sdong
Reviewed By: yiwu, sdong
Subscribers: leveldb, dhruba, andrewkr
Differential Revision: https://reviews.facebook.net/D64053
Summary:
1) The previous solution for Prev() prefix support is not clean.
Since I add api SeekForPrev(), now the Prev() can be symmetric to Next().
and we do not need SeekToLast() to be called in Prev() any more.
Also, Next() will Seek(prefix_seek_key_) to solve the problem of possible inconsistency between db_iter and merge_iter when
there is merge_operator. And prefix_seek_key is only refreshed when change direction to forward.
2) This diff also solves the bug of Iterator::SeekToLast() with iterate_upper_bound_ with prefix extractor.
add test cases for the above two cases.
There are some tests for the SeekToLast() in Prev(), I will clean them later.
Test Plan: make all check
Reviewers: IslamAbdelRahman, andrewkr, yiwu, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D63933
Summary:
When constructing a write batch a client may now call MarkWalTerminationPoint() on that batch. No batch operations after this call will be added written to the WAL but will still be inserted into the Memtable. This facility is used to remove one of the three WriteImpl calls in 2PC transactions. This produces a ~1% perf improvement.
```
RocksDB - unoptimized 2pc, sync_binlog=1, disable_2pc=off
INFO 2016-08-31 14:30:38,814 [main]: REQUEST PHASE COMPLETED. 75000000 requests done in 2619 seconds. Requests/second = 28628
RocksDB - optimized 2pc , sync_binlog=1, disable_2pc=off
INFO 2016-08-31 16:26:59,442 [main]: REQUEST PHASE COMPLETED. 75000000 requests done in 2581 seconds. Requests/second = 29054
```
Test Plan: Two unit tests added.
Reviewers: sdong, yiwu, IslamAbdelRahman
Reviewed By: yiwu
Subscribers: hermanlee4, dhruba, andrewkr
Differential Revision: https://reviews.facebook.net/D64599
Summary:
Fix the conflict bug between AddFile() and CompactRange() by
- Make sure that no AddFile calls are running when asking CompactionPicker to pick compaction for manual compaction
- If AddFile() run after we pick the compaction for the manual compaction it will be aware of it since we will add the manual compaction to running_compactions_ after picking it
This will solve these 2 scenarios
- If AddFile() is running, we will wait for it to finish before we pick a compaction for the manual compaction
- If we already picked a manual compaction and then AddFile() started ... we ensure that it never ingest a file in a level that will overlap with the manual compaction
Test Plan: unit tests
Reviewers: sdong
Reviewed By: sdong
Subscribers: andrewkr, yoshinorim, jkedgar, dhruba
Differential Revision: https://reviews.facebook.net/D64449
Summary: We didn't recompute compaction score on SetOptions, and end up not having compaction if no flush happens afterward. The PR fixing it.
Test Plan: See unit test.
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D64167
Summary:
Since AddFile unlock/lock the mutex inside LogAndApply() we need to ensure that during this period other compactions cannot run since such compactions are not aware of the file we are ingesting and could create a compaction that overlap wit this file
this diff add
- WaitForAddFile() call that will ensure that no AddFile() calls are being processed right now
- Call `WaitForAddFile()` in 3 locations
-- When doing manual Compaction
-- When starting automatic Compaction
-- When doing CompactFiles()
Test Plan: unit test
Reviewers: lightmark, yiwu, andrewkr, sdong
Reviewed By: sdong
Subscribers: andrewkr, yoshinorim, jkedgar, dhruba
Differential Revision: https://reviews.facebook.net/D64383
Summary: Use ImmutableDBOptions/MutableDBOptions internally and DBOptions only for user-facing APIs. MutableDBOptions is barely a placeholder for now. I'll start to move options to MutableDBOptions in following diffs.
Test Plan:
make all check
Reviewers: yhchiang, IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D64065
Summary:
Revert the behavior where we don't read sequence id from WAL, but increase it as we replay the log. We still keep the behave for 2PC for now but will fix later.
This change fixes github issue 1339, where some writes come with WAL disabled and we may recover records with wrong sequence id.
Test Plan: Added unit test.
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D64275
Summary:
Mitigate regression bug of options.max_successive_merges hit during DB Recovery
For https://reviews.facebook.net/D62625
Test Plan: make all check
Reviewers: horuff, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D62655
Summary: WritableFile::SetPreallocationBlockSize() requires parameter as size_t, and options used in DBImpl::GetWalPreallocateBlockSize() are all size_t. WritableFile::SetPreallocationBlockSize() should return size_t to avoid build break if size_t is not uint64_t.
Test Plan: Run existing tests.
Reviewers: andrewkr, IslamAbdelRahman, yiwu
Reviewed By: yiwu
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D64137
Summary: Currently the WAL file preallocation size is 1.1 * write_buffer_size. This, however, will be over-estimated if options.db_write_buffer_size or options.max_total_wal_size is set and is much smaller.
Test Plan: Add a unit test.
Reviewers: andrewkr, yiwu
Reviewed By: yiwu
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D63957
Summary: One more small refactor before I split DBOptions into mutable and immutable parts.
Test Plan: existing unit tests.
Reviewers: yhchiang, IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D64047
Summary:
if one or more CFs had no data in the WAL, the log number that's used
by FindObsoleteFiles() wasn't updated. We need to treat this case the same as
if the data for that WAL had been flushed.
Test Plan: new unit test
Reviewers: IslamAbdelRahman, yiwu, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D63963
Summary:
* Change constructor of MutableCFOptions to depends only on ColumnFamilyOptions.
* Move `max_subcompactions`, `compaction_options_fifo` and `compaction_pri` to ImmutableCFOptions to make it clear that they are immutable.
Test Plan: existing unit tests.
Reviewers: yhchiang, IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D63945
If log recycling is enabled with the rocksdb (recycle_log_file_num=16)
db->Writebatch is erroring out with keynotfound after ~5-6 hours of run
(1M seq but can happen to any workload I guess).See my detailed bug
report here (https://github.com/facebook/rocksdb/issues/1303).
This commit is the fix for this, a check is been added not to delete
the log file if it is already there in the recycle list.
Test Plan:
Unit tested it and ran the similar profile. Not reproducing anymore.
Summary: As title, make sure Prev() works as expected with Next() when the current iter->key() in the range of the same prefix in prefix seek mode
Test Plan: make all check -j64 (add prefix_test with PrefixSeekModePrev test case)
Reviewers: andrewkr, sdong, IslamAbdelRahman
Reviewed By: IslamAbdelRahman
Subscribers: yoshinorim, andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D61419
Summary:
After 1b8a2e8fdd, DB Pointer is passed to WriteBatchInternal::InsertInto() while DB recovery. This can cause deadlock if options.max_successive_merges hits. In that case DB::Get() will be called. Get() will try to acquire the DB mutex, which is already held by the DB::Open(), causing a deadlock condition.
This commit mitigates the problem by not passing the DB pointer unless 2PC is allowed.
Test Plan: Add a new test and run it.
Reviewers: IslamAbdelRahman, andrewkr, kradhakrishnan, horuff
Reviewed By: kradhakrishnan
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D62625
Summary:
Move the manual memtable flush for databases containing data that has
bypassed the WAL from DBImpl's destructor to CancleAllBackgroundWork().
CancelAllBackgroundWork() is a publicly exposed API which allows
async operations performed by background threads to be disabled on a
database. In effect, this places the database into a "shutdown" state
in advance of calling the database object's destructor. No compactions
or flushing of SST files can occur once a call to this API completes.
When writes are issued to a database with WriteOptions::disableWAL
set to true, DBImpl::has_unpersisted_data_ is set so that
memtables can be flushed when the database object is destroyed. If
CancelAllBackgroundWork() has been called prior to DBImpl's destructor,
this flush operation is not possible and is skipped, causing unnecessary
loss of data.
Since CancelAllBackgroundWork() is already invoked by DBImpl's destructor
in order to perform the thread join portion of its cleanup processing,
moving the manual memtable flush to CancelAllBackgroundWork() ensures
data is persisted regardless of client behavior.
Test Plan:
Write an amount of data that will not cause a memtable flush to a rocksdb
database with all writes marked with WriteOptions::disableWAL. Properly
"close" the database. Reopen database and verify that the data was
persisted.
Reviewers: IslamAbdelRahman, yiwu, yoshinorim, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D62277
Summary: With read_options.background_purge_on_iterator_cleanup=true, File deletion and closing can still happen in forward iterator, or WAL file closing. Cover those cases too.
Test Plan: I am adding unit tests.
Reviewers: andrewkr, IslamAbdelRahman, yiwu
Reviewed By: yiwu
Subscribers: leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D61503
Summary:
My understanding is that the purpose of write stall triggers are to wait for auto-compaction to catch up. Without auto-compaction, we don't need to stall writes.
Also with this diff, flush/compaction conditions are recalculated on dynamic option change. Previously the conditions are recalculate only when write stall options are changed.
Test Plan: See the new test. Removed two tests that are no longer valid.
Reviewers: IslamAbdelRahman, sdong
Reviewed By: sdong
Subscribers: andrewkr, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D61437
Summary:
Stale log files can be deleted out of order. This can happen for various reasons. One of the reason is that no data is ever inserted to a column family and we have an optimization to update its log number, but not all the old log files are cleaned up (the case shown in the unit tests added). It can also happen when we simply delete multiple log files out of order.
This causes data corruption because we simply increase seqID after processing the next row and we may end up with writing data with smaller seqID than what is already flushed to memtables.
In DB recovery, for the oldest files we are replaying, if there it contains no data for any column family, we ignore the sequence IDs in the file.
Test Plan: Add two unit tests that fail without the fix.
Reviewers: IslamAbdelRahman, igor, yiwu
Reviewed By: yiwu
Subscribers: hermanlee4, yoshinorim, leveldb, andrewkr, dhruba
Differential Revision: https://reviews.facebook.net/D60891
Andrew Kryczka