Commit Graph

3002 Commits

Author SHA1 Message Date
Yi Wu
06149429d9 WritePrepared Txn: Return NotSupported on iterator refresh
Summary:
A proper implementation of Iterator::Refresh() for WritePreparedTxnDB would require release and acquire another snapshot. Since MyRocks don't make use of Iterator::Refresh(), we just simply mark it as not supported.
Closes https://github.com/facebook/rocksdb/pull/3290

Differential Revision: D6599931

Pulled By: yiwu-arbug

fbshipit-source-id: 4e1632d967316431424f6e458254ecf9a97567cf
2017-12-18 22:29:30 -08:00
Maysam Yabandeh
78c2eedb4f fix release order in validateNumberOfEntries
Summary:
ScopedArenaIterator should be defined after range_del_agg so that it destructs the assigned iterator, which depends on range_del_agg, before it range_del_agg is already destructed.
Closes https://github.com/facebook/rocksdb/pull/3281

Differential Revision: D6592332

Pulled By: maysamyabandeh

fbshipit-source-id: 89a15d8ed13d0fc856b0c47dce3d91778738dbac
2017-12-18 14:27:28 -08:00
Anand Ananthabhotla
fccc12f386 Add a histogram stat for memtable flush
Summary:
Add a new histogram stat called rocksdb.db.flush.micros for memtable
flush
Closes https://github.com/facebook/rocksdb/pull/3269

Differential Revision: D6559496

Pulled By: anand1976

fbshipit-source-id: f5c771ba2568630458751795e8c37a493ff9b14d
2017-12-15 18:57:00 -08:00
Yi Wu
237b292515 BlobDB: Remove the need to get sequence number per write
Summary:
Previously we store sequence number range of each blob files, and use the sequence number range to check if the file can be possibly visible by a snapshot. But it adds complexity to the code, since the sequence number is only available after a write. (The current implementation get sequence number by calling GetLatestSequenceNumber(), which is wrong.) With the patch, we are not storing sequence number range, and check if snapshot_sequence < obsolete_sequence to decide if the file is visible by a snapshot (previously we check if first_sequence <= snapshot_sequence < obsolete_sequence).
Closes https://github.com/facebook/rocksdb/pull/3274

Differential Revision: D6571497

Pulled By: yiwu-arbug

fbshipit-source-id: ca06479dc1fcd8782f6525b62b7762cd47d61909
2017-12-15 13:27:30 -08:00
Andrew Kryczka
5a7e08468a fix ThreadStatus for bottom-pri compaction threads
Summary:
added `ThreadType::BOTTOM_PRIORITY` which is used in the `ThreadStatus` object to indicate the thread is used for bottom-pri compactions. Previously there was a bug where we mislabeled such threads as `ThreadType::LOW_PRIORITY`.
Closes https://github.com/facebook/rocksdb/pull/3270

Differential Revision: D6559428

Pulled By: ajkr

fbshipit-source-id: 96b1a50a9c19492b1a5fd1b77cf7061a6f9f1d1c
2017-12-14 14:57:49 -08:00
Siying Dong
def6a00740 Print out compression type of new SST files in logging
Summary: Closes https://github.com/facebook/rocksdb/pull/3264

Differential Revision: D6552768

Pulled By: siying

fbshipit-source-id: 6303110aff22f341d5cff41f8d2d4f138a53652d
2017-12-14 10:27:43 -08:00
Maysam Yabandeh
546a63272f disableWAL with WriteImplWALOnly
Summary:
Currently WriteImplWALOnly simply returns when disableWAL is set. This is an incorrect behavior since it does not allocated the sequence number, which is a side-effect of writing to the WAL. This patch fixes the issue.
Closes https://github.com/facebook/rocksdb/pull/3262

Differential Revision: D6550974

Pulled By: maysamyabandeh

fbshipit-source-id: 745a83ae8f04e7ca6c8ffb247d6ef16c287c52e7
2017-12-13 07:57:44 -08:00
Zhongyi Xie
51c2ea0feb Reduce heavy hitter for Get operation
Summary:
This PR addresses the following heavy hitters in `Get` operation by moving calls to `StatisticsImpl::recordTick` from `BlockBasedTable` to `Version::Get`

- rocksdb.block.cache.bytes.write
- rocksdb.block.cache.add
- rocksdb.block.cache.data.miss
- rocksdb.block.cache.data.bytes.insert
- rocksdb.block.cache.data.add
- rocksdb.block.cache.hit
- rocksdb.block.cache.data.hit
- rocksdb.block.cache.bytes.read

The db_bench statistics before and after the change are:

|1GB block read|Children      |Self  |Command          |Shared Object        |Symbol|
|---|---|---|---|---|---|
|master:     |4.22%     |1.31%  |db_bench  |db_bench  |[.] rocksdb::StatisticsImpl::recordTick|
|updated:    |0.51%     |0.21%  |db_bench  |db_bench  |[.] rocksdb::StatisticsImpl::recordTick|
|     	     |0.14%     |0.14%  |db_bench  |db_bench  |[.] rocksdb::GetContext::record_counters|

|1MB block read|Children      |Self  |Command          |Shared Object        |Symbol|
|---|---|---|---|---|---|
|master:    |3.48%     |1.08%  |db_bench  |db_bench  |[.] rocksdb::StatisticsImpl::recordTick|
|updated:    |0.80%     |0.31%  |db_bench  |db_bench  |[.] rocksdb::StatisticsImpl::recordTick|
|    	     |0.35%     |0.35%  |db_bench  |db_bench  |[.] rocksdb::GetContext::record_counters|
Closes https://github.com/facebook/rocksdb/pull/3172

Differential Revision: D6330532

Pulled By: miasantreble

fbshipit-source-id: 2b492959e00a3db29e9437ecdcc5e48ca4ec5741
2017-12-12 21:11:33 -08:00
Islam AbdelRahman
9089373a01 Fix DeleteScheduler::MarkAsTrash() handling existing trash
Summary:
DeleteScheduler::MarkAsTrash() don't handle existing .trash files correctly
This cause rocksdb to not being able to delete existing .trash files on restart
Closes https://github.com/facebook/rocksdb/pull/3261

Differential Revision: D6548003

Pulled By: IslamAbdelRahman

fbshipit-source-id: c3800639412e587a690062c63076a5a08881e0e6
2017-12-12 18:17:13 -08:00
Yi Wu
e3a06f12d2 WritePrepared Txn: fix compaction filter snapshot checks
Summary:
Add snapshot_checker check whenever we need to check sequence against snapshots and decide what to do with an input key. The changes are related to one of:
* compaction filter
* single delete
* delete at bottom level
* merge
Closes https://github.com/facebook/rocksdb/pull/3251

Differential Revision: D6537850

Pulled By: yiwu-arbug

fbshipit-source-id: 3faba40ed5e37779f4a0cb7ae78af9546659c7f2
2017-12-12 11:12:24 -08:00
Zhongyi Xie
bb5ed4b1d1 exclude DynamicUniversalCompactionOptions from ROCKSDB_LITE
Summary:
since [SetOptions](https://github.com/facebook/rocksdb/blob/master/db/db_impl.cc#L494) is not supported in ROCKSDB_LITE
Right now unit test under lite is broken
Closes https://github.com/facebook/rocksdb/pull/3253

Differential Revision: D6539428

Pulled By: miasantreble

fbshipit-source-id: 13172b8ecbd75682330726498ea198969bc3e637
2017-12-11 16:28:20 -08:00
Yi Wu
9a27ac5d89 Fix drop column family data race
Summary:
A data race is caught by tsan_crash test between compaction and DropColumnFamily:
https://gist.github.com/yiwu-arbug/5a2b4baae05eeb99ae1719b650f30a44 Compaction checks if the column family has been dropped on each key input, while user can issue DropColumnFamily which updates cfd->dropped_, causing the data race. Fixing it by making cfd->dropped_ an atomic.
Closes https://github.com/facebook/rocksdb/pull/3250

Differential Revision: D6535991

Pulled By: yiwu-arbug

fbshipit-source-id: 5571df020beae7fa7db6fff5ad0d598f49962895
2017-12-11 13:57:48 -08:00
Zhongyi Xie
fcc8a6574d Make Universal compaction options dynamic
Summary:
Let me know if more test coverage is needed
Closes https://github.com/facebook/rocksdb/pull/3213

Differential Revision: D6457165

Pulled By: miasantreble

fbshipit-source-id: 3f944abff28aa7775237f1c4f61c64ccbad4eea9
2017-12-11 13:27:06 -08:00
Prashant D
6a183d1ae8 Fix coverity issues compaction_job, compaction_picker
Summary:
db/compaction_job.cc:
  ReportStartedCompaction(compaction);

CID 1419863 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
2. uninit_member: Non-static class member bottommost_level_ is not initialized in this constructor nor in any functions that it calls.

db/compaction_picker_universal.cc:
7struct InputFileInfo {
   	2. uninit_member: Non-static class member level is not initialized in this constructor nor in any functions that it calls.

CID 1405355 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
4. uninit_member: Non-static class member index is not initialized in this constructor nor in any functions that it calls.
 38  InputFileInfo() : f(nullptr) {}

db/dbformat.h:
 ParsedInternalKey()
 84      : sequence(kMaxSequenceNumber)  // Make code analyzer happy

CID 1168095 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
2. uninit_member: Non-static class member type is not initialized in this constructor nor in any functions that it calls.
 85  {}  // Intentionally left uninitialized (for speed)
Closes https://github.com/facebook/rocksdb/pull/3091

Differential Revision: D6534558

Pulled By: yiwu-arbug

fbshipit-source-id: 5ada975956196d267b3f149386842af71eda7553
2017-12-11 11:57:15 -08:00
Prashant D
34aa245dd8 Fix coverity issues version, write_batch
Summary:
db/version_builder.cc:
117        base_vstorage_->InternalComparator();

CID 1351713 (#1 of 1): Uninitialized pointer field (UNINIT_CTOR)
2. uninit_member: Non-static class member field level_zero_cmp_.internal_comparator is not initialized in this constructor nor in any functions that it calls.

db/version_edit.h:
145  FdWithKeyRange()
146      : fd(),
147        smallest_key(),
148        largest_key() {

CID 1418254 (#1 of 1): Uninitialized pointer field (UNINIT_CTOR)
2. uninit_member: Non-static class member file_metadata is not initialized in this constructor nor in any functions that it calls.
149  }

db/version_set.cc:
120    }

CID 1322789 (#1 of 1): Uninitialized pointer field (UNINIT_CTOR)
4. uninit_member: Non-static class member curr_file_level_ is not initialized in this constructor nor in any functions that it calls.
121  }

db/write_batch.cc:
 939    assert(cf_mems_);

CID 1419862 (#1 of 1): Uninitialized scalar field (UNINIT_CTOR)
3. uninit_member: Non-static class member rebuilding_trx_seq_ is not initialized in this constructor nor in any functions that it calls.
 940  }
Closes https://github.com/facebook/rocksdb/pull/3092

Differential Revision: D6505666

Pulled By: yiwu-arbug

fbshipit-source-id: fd2c68948a0280772691a419d72ac7e190951d86
2017-12-07 11:57:36 -08:00
Andrew Kryczka
2e3a00987e fix ASAN for DeleteFilesInRange test case
Summary:
error message was

```
==3095==ERROR: AddressSanitizer: stack-use-after-scope on address 0x7ffd18216c40 at pc 0x0000005edda1 bp 0x7ffd18215550 sp 0x7ffd18214d00
...
Address 0x7ffd18216c40 is located in stack of thread T0 at offset 1952 in frame
     #0 internal_repo_rocksdb/db_compaction_test.cc:1520 rocksdb::DBCompactionTest_DeleteFileRangeFileEndpointsOverlapBug_Test::TestBody()
```

It was unsafe to have slices referring to the temporary string objects' buffers, as those strings were destroyed before the slices were used. Fixed it by assigning the strings returned by `Key()` to local variables.
Closes https://github.com/facebook/rocksdb/pull/3238

Differential Revision: D6507864

Pulled By: ajkr

fbshipit-source-id: dd07de1a0070c6748c1ab4f3d7bd31f9a81889d0
2017-12-07 11:12:43 -08:00
Sagar Vemuri
bbef8c3884 Log GetCurrentTime failures during Flush and Compaction
Summary:
`GetCurrentTime()` is used to populate `creation_time` table property during flushes and compactions. It is safe to ignore `GetCurrentTime()` failures here but they should be logged.

(Note that `creation_time` property was introduced as part of TTL-based FIFO compaction in #2480.)

Tes Plan:
`make check`
Closes https://github.com/facebook/rocksdb/pull/3231

Differential Revision: D6501935

Pulled By: sagar0

fbshipit-source-id: 376adcf4ab801d3a43ec4453894b9a10909c8eb6
2017-12-06 20:56:53 -08:00
Andrew Kryczka
78d1a5ec72 Preserve overlapping file endpoint invariant
Summary:
Fix for #2833.

- In `DeleteFilesInRange`, use `GetCleanInputsWithinInterval` instead of `GetOverlappingInputs` to make sure we get a clean cut set of files to delete.
- In `GetCleanInputsWithinInterval`, support nullptr as `begin_key` or `end_key`.
- In `GetOverlappingInputsRangeBinarySearch`, move the assertion for non-empty range away from `ExtendFileRangeWithinInterval`, which should be allowed to return an empty range (via `end_index < begin_index`).
Closes https://github.com/facebook/rocksdb/pull/2843

Differential Revision: D5772387

Pulled By: ajkr

fbshipit-source-id: e554e8461823c6be82b21a9262a2da02b3957881
2017-12-06 18:56:54 -08:00
Yi Wu
a7d32776f0 Fix write_callback_test compile error
Summary:
Rename shadow variable name db_impl.

Fixing #3227
Closes https://github.com/facebook/rocksdb/pull/3235

Differential Revision: D6504051

Pulled By: yiwu-arbug

fbshipit-source-id: 186c9378dabb11f8d6db56f45c95cc3b029fcb88
2017-12-06 17:12:27 -08:00
Yi Wu
20995c5729 Make iterator invalid on Merge error
Summary:
Since #1665, on merge error, iterator will be set to corrupted status, but it doesn't invalidate the iterator. Fixing it.
Closes https://github.com/facebook/rocksdb/pull/3226

Differential Revision: D6499094

Pulled By: yiwu-arbug

fbshipit-source-id: 80222930f949e31f90a6feaa37ddc3529b510d2c
2017-12-06 11:56:39 -08:00
Andrew Kryczka
63f1c0a57d fix gflags namespace
Summary:
I started adding gflags support for cmake on linux and got frustrated that I'd need to duplicate the build_detect_platform logic, which determines namespace based on attempting compilation. We can do it differently -- use the GFLAGS_NAMESPACE macro if available, and if not, that indicates it's an old gflags version without configurable namespace so we can simply hardcode "google".
Closes https://github.com/facebook/rocksdb/pull/3212

Differential Revision: D6456973

Pulled By: ajkr

fbshipit-source-id: 3e6d5bde3ca00d4496a120a7caf4687399f5d656
2017-12-01 10:42:05 -08:00
Maysam Yabandeh
18dcf7f98d WritePrepared Txn: PreReleaseCallback
Summary:
Add PreReleaseCallback to be called at the end of WriteImpl but before publishing the sequence number. The callback is used in WritePrepareTxn to i) update the commit map, ii) update the last published sequence number in the 2nd write queue. It also ensures that all the commits will go to the 2nd queue.
These changes will ensure that the commit map is updated before the sequence number is published and used by reading snapshots. If we use two write queues, the snapshots will use the seq number published by the 2nd queue. If we use one write queue (the default, the snapshots will use the last seq number in the memtable, which also indicates the last published seq number.
Closes https://github.com/facebook/rocksdb/pull/3205

Differential Revision: D6438959

Pulled By: maysamyabandeh

fbshipit-source-id: f8b6c434e94bc5f5ab9cb696879d4c23e2577ab9
2017-11-30 23:50:45 -08:00
zhangjinpeng1987
ffacaaa3ea fix Seek with lower_bound
Summary:
When Seek a key less than `lower_bound`, should return `lower_bound`.
ajkr PTAL
Closes https://github.com/facebook/rocksdb/pull/3199

Differential Revision: D6421126

Pulled By: ajkr

fbshipit-source-id: a06c825830573e0040630704f6bcb3f7f48626f7
2017-11-29 22:56:29 -08:00
kapitan-k
75d57a5d53 C API: Add some block based table options
Summary: Closes https://github.com/facebook/rocksdb/pull/3159

Differential Revision: D6428220

Pulled By: sagar0

fbshipit-source-id: 60508d09b5281f54b907a1c40e9631fc08343131
2017-11-28 14:12:44 -08:00
Yi Wu
3cf562be31 Fix IOError on WAL write doesn't propagate to write group follower
Summary:
This is a simpler version of #3097 by removing all unrelated changes.

Fixing the bug where concurrent writes may get Status::OK while it actually gets IOError on WAL write. This happens when multiple writes form a write batch group, and the leader get an IOError while writing to WAL. The leader failed to pass the error to followers in the group, and the followers end up returning Status::OK() while actually writing nothing. The bug only affect writes in a batch group. Future writes after the batch group will correctly return immediately with the IOError.
Closes https://github.com/facebook/rocksdb/pull/3201

Differential Revision: D6421644

Pulled By: yiwu-arbug

fbshipit-source-id: 1c2a455c5b73f6842423785eb8a9dbfbb191dc0e
2017-11-28 11:42:48 -08:00
Andrew Kryczka
1bdb44de95 optimize file ingestion checks for range deletion overlap
Summary:
Before we were checking every file in the level which was unnecessary. We can piggyback onto the code for checking point-key overlap, which already opens all the files that could possibly contain overlapping range deletions. This PR makes us check just the range deletions from those files, so no extra ones will be opened.
Closes https://github.com/facebook/rocksdb/pull/3179

Differential Revision: D6358125

Pulled By: ajkr

fbshipit-source-id: 00e200770fdb8f3cc6b1b2da232b755e4ba36279
2017-11-28 11:27:02 -08:00
Griffin Smith
2f09524762 Expose all remaining read and write options via the C API
Summary:
Expose read and write options via the C API
Closes https://github.com/facebook/rocksdb/pull/3185

Differential Revision: D6389658

Pulled By: sagar0

fbshipit-source-id: 1848912750329a476805b3cb2f315e7b71f61472
2017-11-28 10:28:46 -08:00
Maysam Yabandeh
e59cb2a19b Add seq_per_batch to WriteWithCallbackTest
Summary:
Augment WriteWithCallbackTest to also test when seq_per_batch is true.
Closes https://github.com/facebook/rocksdb/pull/3195

Differential Revision: D6398143

Pulled By: maysamyabandeh

fbshipit-source-id: 7bc4218609355ec20fed25df426a8455ec2390d3
2017-11-22 13:56:44 -08:00
Zhongyi Xie
5fac4729cc make compaction_readahead_size_ thread safe
Summary:
this should fix the failing tsan_check
Closes https://github.com/facebook/rocksdb/pull/3192

Differential Revision: D6390004

Pulled By: miasantreble

fbshipit-source-id: 6cadfc6f68febb1a77b0abcdb5416570dad926a5
2017-11-21 20:11:38 -08:00
anand1976
d394a6bb48 Add a ticker stat for number of keys skipped during iteration
Summary:
This diff adds a new ticker stat, NUMBER_ITER_SKIP, to count the
number of internal keys skipped during iteration. Keys can be skipped
due to deletes, or lower sequence number, or higher sequence number
than the one requested.

Also, fix the issue when StatisticsData is naturally aligned on cacheline boundary,
padding becomes a zero size array, which the Windows compiler doesn't
like. So add a cacheline worth of padding in that case to keep it happy.
We cannot conditionally add padding as gcc doesn't allow using sizeof
in preprocessor directives.
Closes https://github.com/facebook/rocksdb/pull/3177

Differential Revision: D6353897

Pulled By: anand1976

fbshipit-source-id: 441d5a09af9c4e22e7355242dfc0c7b27aa0a6c2
2017-11-20 21:26:37 -08:00
Gustav Davidsson
2d04ed65e4 Make trash-to-DB size ratio limit configurable
Summary:
Allow users to configure the trash-to-DB size ratio limit, so
that ratelimits for deletes can be enforced even when larger portions of
the database are being deleted.
Closes https://github.com/facebook/rocksdb/pull/3158

Differential Revision: D6304897

Pulled By: gdavidsson

fbshipit-source-id: a28dd13059ebab7d4171b953ed91ce383a84d6b3
2017-11-17 11:58:17 -08:00
Zhongyi Xie
32e31d49d1 Make DBOption compaction_readahead_size dynamic
Summary: Closes https://github.com/facebook/rocksdb/pull/3004

Differential Revision: D6056141

Pulled By: miasantreble

fbshipit-source-id: 56df1630f464fd56b07d25d38161f699e0528b7f
2017-11-16 17:57:25 -08:00
Maysam Yabandeh
54b43563be WritePrepared Txn: Refactoring WriteCallback
Summary:
Refactor the logic around WriteCallback in the write path to clarify when and how exactly we advance the sequence number and making sure it is consistent across the code.
Closes https://github.com/facebook/rocksdb/pull/3168

Differential Revision: D6324312

Pulled By: maysamyabandeh

fbshipit-source-id: 9a34f479561fdb2a5d01ef6d37a28908d03bbe33
2017-11-15 08:27:06 -08:00
Maysam Yabandeh
53863b76f9 WritePrepared Txn: fix bug with Rollback seq
Summary:
The sequence number was not properly advanced after a rollback marker. The patch extends the existing unit tests to detect the bug and also fixes it.
Closes https://github.com/facebook/rocksdb/pull/3157

Differential Revision: D6304291

Pulled By: maysamyabandeh

fbshipit-source-id: 1b519c44a5371b802da49c9e32bd00087a8da401
2017-11-15 08:27:06 -08:00
Maysam Yabandeh
175d5d6a9e Properly destruct rebuilding_trx_
Summary:
When testing rebuilding_trx_ in MemTableInserter might still be set before the tests finishes which would cause ASAN alarms for leaks. This patch deletes the pointers in MemTableInserter destructor.
Closes https://github.com/facebook/rocksdb/pull/3162

Differential Revision: D6317113

Pulled By: maysamyabandeh

fbshipit-source-id: a68be70709a4fff7ac2b768660119311968f9c21
2017-11-14 08:56:50 -08:00
Maysam Yabandeh
2515266725 WritePrepared Txn: Refactoring TrackKeys
Summary:
This patch clarifies and refactors the logic around tracked keys in transactions.
Closes https://github.com/facebook/rocksdb/pull/3140

Differential Revision: D6290258

Pulled By: maysamyabandeh

fbshipit-source-id: 03b50646264cbcc550813c060b180fc7451a55c1
2017-11-11 13:14:20 -08:00
Maysam Yabandeh
2edc92bc28 WritePrepared Txn: cross-compatibility test
Summary:
Add tests to ensure that WritePrepared and WriteCommitted policies are cross compatible when the db WAL is empty. This is important when the admin want to switch between the policies. In such case, before the switch the admin needs to empty the WAL by i) committing/rollbacking all the pending transactions, ii) FlushMemTables
Closes https://github.com/facebook/rocksdb/pull/3118

Differential Revision: D6227247

Pulled By: maysamyabandeh

fbshipit-source-id: bcde3d92c1e89cda3b9cfa69f6a20af5d8993db7
2017-11-11 11:28:37 -08:00
Maysam Yabandeh
857adf388f WritePrepared Txn: Refactor conf params
Summary:
Summary of changes:
- Move seq_per_batch out of Options
- Rename concurrent_prepare to two_write_queues
- Add allocate_seq_only_for_data_
Closes https://github.com/facebook/rocksdb/pull/3136

Differential Revision: D6304458

Pulled By: maysamyabandeh

fbshipit-source-id: 08e685bfa82bbc41b5b1c5eb7040a8ca6e05e58c
2017-11-10 17:28:12 -08:00
Dmitri Smirnov
f8e2db0717 Fix crashes, address test issues and adjust windows test script
Summary:
Add per-exe execution capability
  Add fix parsing of groups/tests
  Add timer test exclusion

 Fix unit tests
  Ifdef threadpool specific tests that do not pass on Vista threadpool.
  Remove spurious outout from prefix_test so test case listing works
  properly.
  Fix not using standard test directories results in file creation errors
  in sst_dump_test.

  BlobDb fixes:
    In C++ end() iterators can not be dereferenced. They are not valid.
	When deleting blob_db_ set it to nullptr before any other code executes.
	Not fixed:. On Windows you can not delete a file while it is open.
	[ RUN      ] BlobDBTest.ReadWhileGC
	d:\dev\rocksdb\rocksdb\utilities\blob_db\blob_db_test.cc(75): error: DestroyBlobDB(dbname_, options, bdb_options)
	IO error: Failed to delete: d:/mnt/db\testrocksdb-17444/blob_db_test/blob_dir/000001.blob: Permission denied
	d:\dev\rocksdb\rocksdb\utilities\blob_db\blob_db_test.cc(75): error: DestroyBlobDB(dbname_, options, bdb_options)
	IO error: Failed to delete: d:/mnt/db\testrocksdb-17444/blob_db_test/blob_dir/000001.blob: Permission denied

  write_batch
    Should not call front() if there is a chance the container is empty
Closes https://github.com/facebook/rocksdb/pull/3152

Differential Revision: D6293274

Pulled By: sagar0

fbshipit-source-id: 318c3717c22087fae13b18715dffb24565dbd956
2017-11-10 10:41:57 -08:00
Shaohua Li
eefd75a228 Stream
Summary:
Add a simple policy for NVMe write time life hint
Closes https://github.com/facebook/rocksdb/pull/3095

Differential Revision: D6298030

Pulled By: shligit

fbshipit-source-id: 9a72a42e32e92193af11599eb71f0cf77448e24d
2017-11-10 09:26:24 -08:00
kapitan-k
f1c5eaba56 updated c ingestexternalfileoptions for ingest behind
Summary: Closes https://github.com/facebook/rocksdb/pull/3151

Differential Revision: D6293861

Pulled By: ajkr

fbshipit-source-id: f8db0a71509d1cd8237f2d377bf9e1bb0464bdbf
2017-11-09 18:15:09 -08:00
Andrew Kryczka
93f69cb93a use bottommost compression when base level is bottommost
Summary:
The previous compression type selection caused unexpected behavior when the base level was also the bottommost level. The following sequence of events could happen:

- full compaction generates files with `bottommost_compression` type
- now base level is bottommost level since all files are in the same level
- any compaction causes files to be rewritten `compression_per_level` type since bottommost compression didn't apply to base level

I changed the code to make bottommost compression apply to base level.
Closes https://github.com/facebook/rocksdb/pull/3141

Differential Revision: D6264614

Pulled By: ajkr

fbshipit-source-id: d7aaa8675126896684154a1f2c9034d6214fde82
2017-11-09 17:42:00 -08:00
Sagar Vemuri
a6d8e30c05 Remove unnecessary status check in TableCache::NewIterator
Summary:
While investigating the usage of `new_table_iterator_nanos` perf counter, I saw some code was wrapper around with unnecessary status check ... so removed it.
Closes https://github.com/facebook/rocksdb/pull/3120

Differential Revision: D6229181

Pulled By: sagar0

fbshipit-source-id: f8a44fe67f5a05df94553fdb233b21e54e88cc34
2017-11-03 14:42:08 -07:00
Andrew Kryczka
24ad430600 pass key/value samples through zstd compression dictionary generator
Summary:
Instead of using samples directly, we now support passing the samples through zstd's dictionary generator when `CompressionOptions::zstd_max_train_bytes` is set to nonzero. If set to zero, we will use the samples directly as the dictionary -- same as before.

Note this is the first step of #2987, extracted into a separate PR per reviewer request.
Closes https://github.com/facebook/rocksdb/pull/3057

Differential Revision: D6116891

Pulled By: ajkr

fbshipit-source-id: 70ab13cc4c734fa02e554180eed0618b75255497
2017-11-02 22:56:36 -07:00
Andrew Kryczka
c4c1f961e7 dynamically change current memtable size
Summary:
Previously setting `write_buffer_size` with `SetOptions` would only apply to new memtables. An internal user wanted it to take effect immediately, instead of at an arbitrary future point, to prevent OOM.

This PR makes the memtable's size mutable, and makes `SetOptions()` mutate it. There is one case when we preserve the old behavior, which is when memtable prefix bloom filter is enabled and the user is increasing the memtable's capacity. That's because the prefix bloom filter's size is fixed and wouldn't work as well on a larger memtable.
Closes https://github.com/facebook/rocksdb/pull/3119

Differential Revision: D6228304

Pulled By: ajkr

fbshipit-source-id: e44bd9d10a5f8c9d8c464bf7436070bb3eafdfc9
2017-11-02 22:28:10 -07:00
Yi Wu
62578d80c1 Blob DB: Add compaction filter to remove expired blob index entries
Summary:
After adding expiration to blob index in #3066, we are now able to add a compaction filter to cleanup expired blob index entries.
Closes https://github.com/facebook/rocksdb/pull/3090

Differential Revision: D6183812

Pulled By: yiwu-arbug

fbshipit-source-id: 9cb03267a9702975290e758c9c176a2c03530b83
2017-11-02 17:27:38 -07:00
Yi Wu
7bfa88037e Blob DB: fix snapshot handling
Summary:
Blob db will keep blob file if data in the file is visible to an active snapshot. Before this patch it checks whether there is an active snapshot has sequence number greater than the earliest sequence in the file. This is problematic since we take snapshot on every read, if it keep having reads, old blob files will not be cleanup. Change to check if there is an active snapshot falls in the range of [earliest_sequence, obsolete_sequence) where obsolete sequence is
1. if data is relocated to another file by garbage collection, it is the latest sequence at the time garbage collection finish
2. otherwise, it is the latest sequence of the file
Closes https://github.com/facebook/rocksdb/pull/3087

Differential Revision: D6182519

Pulled By: yiwu-arbug

fbshipit-source-id: cdf4c35281f782eb2a9ad6a87b6727bbdff27a45
2017-11-02 15:58:27 -07:00
Andrew Kryczka
6778690b51 fix duplicate definition of GetEntryType()
Summary:
It's also defined in db/dbformat.cc per 7fe3b32896
Closes https://github.com/facebook/rocksdb/pull/3111

Differential Revision: D6219140

Pulled By: ajkr

fbshipit-source-id: 0f2b14e41457334a4665c6b7e3f42f1a060a0f35
2017-11-01 22:56:17 -07:00
Maysam Yabandeh
02693f64fc WritePrepared Txn: ValidateSnapshot
Summary:
Implements ValidateSnapshot for WritePrepared txns and also adds a unit test to clarify the contract of this function.
Closes https://github.com/facebook/rocksdb/pull/3101

Differential Revision: D6199405

Pulled By: maysamyabandeh

fbshipit-source-id: ace509934c307ea5d26f4bbac5f836d7c80fd240
2017-11-01 19:11:09 -07:00
Mikhail Antonov
7fe3b32896 Added support for differential snapshots
Summary:
The motivation for this PR is to add to RocksDB support for differential (incremental) snapshots, as snapshot of the DB changes between two points in time (one can think of it as diff between to sequence numbers, or the diff D which can be thought of as an SST file or just set of KVs that can be applied to sequence number S1 to get the database to the state at sequence number S2).

This feature would be useful for various distributed storages layers built on top of RocksDB, as it should help reduce resources (time and network bandwidth) needed to recover and rebuilt DB instances as replicas in the context of distributed storages.

From the API standpoint that would like client app requesting iterator between (start seqnum) and current DB state, and reading the "diff".

This is a very draft PR for initial review in the discussion on the approach, i'm going to rework some parts and keep updating the PR.

For now, what's done here according to initial discussions:

Preserving deletes:
 - We want to be able to optionally preserve recent deletes for some defined period of time, so that if a delete came in recently and might need to be included in the next incremental snapshot it would't get dropped by a compaction. This is done by adding new param to Options (preserve deletes flag) and new variable to DB Impl where we keep track of the sequence number after which we don't want to drop tombstones, even if they are otherwise eligible for deletion.
 - I also added a new API call for clients to be able to advance this cutoff seqnum after which we drop deletes; i assume it's more flexible to let clients control this, since otherwise we'd need to keep some kind of timestamp < -- > seqnum mapping inside the DB, which sounds messy and painful to support. Clients could make use of it by periodically calling GetLatestSequenceNumber(), noting the timestamp, doing some calculation and figuring out by how much we need to advance the cutoff seqnum.
 - Compaction codepath in compaction_iterator.cc has been modified to avoid dropping tombstones with seqnum > cutoff seqnum.

Iterator changes:
 - couple params added to ReadOptions, to optionally allow client to request internal keys instead of user keys (so that client can get the latest value of a key, be it delete marker or a put), as well as min timestamp and min seqnum.

TableCache changes:
 - I modified table_cache code to be able to quickly exclude SST files from iterators heep if creation_time on the file is less then iter_start_ts as passed in ReadOptions. That would help a lot in some DB settings (like reading very recent data only or using FIFO compactions), but not so much for universal compaction with more or less long iterator time span.

What's left:

 - Still looking at how to best plug that inside DBIter codepath. So far it seems that FindNextUserKeyInternal only parses values as UserKeys, and iter->key() call generally returns user key. Can we add new API to DBIter as internal_key(), and modify this internal method to optionally set saved_key_ to point to the full internal key? I don't need to store actual seqnum there, but I do need to store type.
Closes https://github.com/facebook/rocksdb/pull/2999

Differential Revision: D6175602

Pulled By: mikhail-antonov

fbshipit-source-id: c779a6696ee2d574d86c69cec866a3ae095aa900
2017-11-01 18:56:43 -07:00