Commit Graph

1313 Commits

Author SHA1 Message Date
mrambacher
13ae16c315 Cleanup includes in dbformat.h (#8930)
Summary:
This header file was including everything and the kitchen sink when it did not need to.  This resulted in many places including this header when they needed other pieces instead.

Cleaned up this header to only include what was needed and fixed up the remaining code to include what was now missing.

Hopefully, this sort of code hygiene cleanup will speed up the builds...

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8930

Reviewed By: pdillinger

Differential Revision: D31142788

Pulled By: mrambacher

fbshipit-source-id: 6b45de3f300750c79f751f6227dece9cfd44085d
2021-09-29 04:04:40 -07:00
mrambacher
7fd68b7c39 Make WalFilter, SstPartitionerFactory, FileChecksumGenFactory, and TableProperties Customizable (#8638)
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/8638

Reviewed By: zhichao-cao

Differential Revision: D31024729

Pulled By: mrambacher

fbshipit-source-id: 954c04ccab0b8dee64050a27aadf78ed119106c0
2021-09-28 05:32:02 -07:00
sdong
7c6a7e8fa8 FaultInjectionTestFS::InjectThreadSpecificReadError() should not corrupt mmaped bytes (#8952)
Summary:
Right now FaultInjectionTestFS::InjectThreadSpecificReadError() might try to corrupt return bytes, but these bytes might be from mmapped files, which would cause segfault. Instead FaultInjectionTestFS::InjectThreadSpecificReadError() should never corrupt data unless it is in caller's buffer.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8952

Test Plan: See db_stress still runs and make sure in a test run failurs are still injected in non-mmap cases.

Reviewed By: ajkr, ltamasi

Differential Revision: D31147318

fbshipit-source-id: 9484a64ff2aaa36685557203f449286e694e65f9
2021-09-23 12:00:47 -07:00
Levi Tamasi
be206db351 Deflake MySQLStyleTransactionTest.TransactionStressTest in "status checked" mode (#8947)
Summary:
There is a corner case when using WriteUnprepared transactions when
`WriteUnpreparedTxn::Get` returns `Status::TryAgain` instead of
propagating the result of `GetFromBatchAndDB`. The patch adds
`PermitUncheckedError` to make the `ASSERT_STATUS_CHECKED` build pass in
this case as well.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8947

Test Plan: `make check`

Reviewed By: riversand963

Differential Revision: D31125422

Pulled By: ltamasi

fbshipit-source-id: 42de51dcfa9384e032244c2b4d3f40e9a4111194
2021-09-22 16:40:25 -07:00
sdong
9320067703 Improve fault injection to MultiRead (#8937)
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
2021-09-21 14:48:15 -07:00
Peter Dillinger
5268cdc997 Finish BackupEngine migration to IOStatus (#8940)
Summary:
Updates a few remaining functions that should have been updated
from Status -> IOStatus, and adds to HISTORY for the overall change
including https://github.com/facebook/rocksdb/issues/8820.

This change is for inclusion in version 6.25.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8940

Test Plan: CI

Reviewed By: zhichao-cao

Differential Revision: D31085029

Pulled By: pdillinger

fbshipit-source-id: 91557c6a39ef1d90357d4f4dcd79af0645d87c7b
2021-09-21 11:13:17 -07:00
sdong
4f1dd05cec Implement TestFSRandomAccessFile::MultiRead() (#8925)
Summary:
Right now, the failure injection test for MultiGet() is not sufficient. Improve it with TestFSRandomAccessFile::MultiRead() injecting failures.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8925

Test Plan: Run crash test locally for a while.

Reviewed By: anand1976

Differential Revision: D31000529

fbshipit-source-id: 439c7e02cf7440ac5af82deb609e202abdca3e1f
2021-09-16 16:01:34 -07:00
Zhichao Cao
82e7631de6 Replace Status with IOStatus in the backupable_db (#8820)
Summary:
In order to populate the IOStatus up to the higher level, replace some of the Status to IOStatus.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8820

Test Plan: make check

Reviewed By: pdillinger

Differential Revision: D30967215

Pulled By: zhichao-cao

fbshipit-source-id: ccf9d5cfbd9d3de047c464aaa85f9fa43b474903
2021-09-15 15:09:48 -07:00
mrambacher
dafa584fd1 Change the File System File Wrappers to std::unique_ptr (#8618)
Summary:
This allows the wrapper classes to own the wrapped object and eliminates confusion as to ownership.  Previously, many classes implemented their own ownership solutions.  Fixes https://github.com/facebook/rocksdb/issues/8606

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8618

Reviewed By: pdillinger

Differential Revision: D30136064

Pulled By: mrambacher

fbshipit-source-id: d0bf471df8818dbb1770a86335fe98f761cca193
2021-09-13 08:46:19 -07:00
Peter Dillinger
bda8d93ba9 Fix and detect headers with missing dependencies (#8893)
Summary:
It's always annoying to find a header does not include its own
dependencies and only works when included after other includes. This
change adds `make check-headers` which validates that each header can
be included at the top of a file. Some headers are excluded e.g. because
of platform or external dependencies.

rocksdb_namespace.h had to be re-worked slightly to enable checking for
failure to include it. (ROCKSDB_NAMESPACE is a valid namespace name.)

Fixes mostly involve adding and cleaning up #includes, but for
FileTraceWriter, a constructor was out-of-lined to make a forward
declaration sufficient.

This check is not currently run with `make check` but is added to
CircleCI build-linux-unity since that one is already relatively fast.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8893

Test Plan: existing tests and resolving issues detected by new check

Reviewed By: mrambacher

Differential Revision: D30823300

Pulled By: pdillinger

fbshipit-source-id: 9fff223944994c83c105e2e6496d24845dc8e572
2021-09-10 10:00:26 -07:00
mrambacher
0fb938c448 Add support to the ObjectRegistry for ManagedObjects (#8658)
Summary:
ManagedObjects are  shared pointer objects where RocksDB wants to share a single object between multiple configurations.  For example, the Cache may be shared between multiple column families/tables or the Statistics may be shared between multiple databases.

ManagedObjects are stored in the ObjectRegistry by Type (e.g. Cache) and ID.  For a given type/ID name, a single object is stored.

APIs were added to get/set/create these objects.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8658

Reviewed By: pdillinger

Differential Revision: D30806273

Pulled By: mrambacher

fbshipit-source-id: 832ac4423b210c4c4b4a456b35897334775d3160
2021-09-10 05:21:04 -07:00
hx235
45175ca2e1 Charge read to rate limiter in BackupEngine (#8722)
Summary:
Context:
While all the non-trivial write operations in BackupEngine go through the RateLimiter, reads currently do not. In general, this is not a huge issue because (especially since some I/O efficiency fixes) reads in BackupEngine are mostly limited by corresponding writes, for both backup and restore. But in principle we should charge the RateLimiter for reads as well.
- Charged read operations in `BackupEngineImpl::CopyOrCreateFile`, `BackupEngineImpl::ReadFileAndComputeChecksum`, `BackupEngineImpl::BackupMeta::LoadFromFile` and `BackupEngineImpl::GetFileDbIdentities`

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8722

Test Plan:
- Passed existing tests
- Passed added unit tests

Reviewed By: pdillinger

Differential Revision: D30610464

Pulled By: hx235

fbshipit-source-id: 9b08c9387159a5385c8d390d6666377a0d0117e5
2021-09-08 16:24:40 -07:00
Andrew Kryczka
dd092c2d11 prevent stranded LATEST_BACKUP in BackupEngineTest.NoDeleteWithReadOnly (#8887)
Summary:
A "LATEST_BACKUP" file was left in the backup directory by
"BackupEngineTest.NoDeleteWithReadOnly" test, affecting future test
runs. In particular, it caused "BackupEngineTest.IOStats" to fail since
it relies on backup directory containing only data written by its
`BackupEngine`.

The fix is to promote "LATEST_BACKUP" to an explicitly managed file so
it is deleted in `BackupEngineTest` constructor if it exists.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8887

Test Plan:
below command used to fail. Now it passes:

```
$ TEST_TMPDIR=/dev/shm ./backupable_db_test --gtest_filter='BackupEngineTest.NoDeleteWithReadOnly:BackupEngineTest.IOStats'
```

Reviewed By: pdillinger

Differential Revision: D30812336

Pulled By: ajkr

fbshipit-source-id: 32dfbe1368ebdab872e610764bfea5daf9a2af09
2021-09-08 13:39:01 -07:00
Andrew Kryczka
9308ff366c Bytes read/written stats for CreateNewBackup*() (#8819)
Summary:
Gets `Statistics` from the options associated with the `DB` undergoing backup, and populates new ticker stats with the thread-local `IOContext` read/write counters for the threads doing backup work.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8819

Reviewed By: pdillinger

Differential Revision: D30779238

Pulled By: ajkr

fbshipit-source-id: 75ccafc355f90906df5cf80367f7245b985772d8
2021-09-07 18:25:16 -07:00
Peter Dillinger
0ef88538c6 Improve support for using regexes (#8740)
Summary:
* Consolidate use of std::regex for testing to testharness.cc, to
minimize Facebook linters constantly flagging uses in non-production
code.
* Improve syntax and error messages for asserting some string matches a
regex in tests.
* Add a public Regex wrapper class to encapsulate existing usage in
ObjectRegistry.
* Remove unnecessary include <regex>
* Put warnings that use of Regex in production code could cause bad
performance or stack overflow.

Intended follow-up work:
* Replace std::regex with another underlying implementation like RE2
* Improve ObjectRegistry interface in terms of possibly confusing literal
string matching vs. regex and in terms of reporting invalid regex.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8740

Test Plan:
tests updated, basic unit test for public Regex, and some manual
testing of temporary changes to see example error messages:

utilities/backupable/backupable_db_test.cc:917: Failure
000010_1162373755_138626.blob (child.name)
does not match regex
[0-9]+_[0-9]+_[0-9]+[.]blobHAHAHA (pattern)

db/db_basic_test.cc:74: Failure
R3SHSBA8C4U0CIMV2ZB0 (sid3)
does not match regex [0-9A-Z]{20}HAHAHA

Reviewed By: mrambacher

Differential Revision: D30706246

Pulled By: pdillinger

fbshipit-source-id: ba845e8f563ccad39bdb58f44f04e9da8f78c3fd
2021-09-07 13:05:23 -07:00
Peter Dillinger
4750421ece Replace most typedef with using= (#8751)
Summary:
Old typedef syntax is confusing

Most but not all changes with

    perl -pi -e 's/typedef (.*) ([a-zA-Z0-9_]+);/using $2 = $1;/g' list_of_files
    make format

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8751

Test Plan: existing

Reviewed By: zhichao-cao

Differential Revision: D30745277

Pulled By: pdillinger

fbshipit-source-id: 6f65f0631c3563382d43347896020413cc2366d9
2021-09-07 11:31:59 -07:00
Peter Dillinger
32752551b9 Fix a buffer size race condition in BackupEngine (#8732)
Summary:
If RateLimiter burst bytes changes during concurrent Restore
operations

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8732

Test Plan: updated unit test fails with TSAN before change, passes after

Reviewed By: ajkr

Differential Revision: D30683879

Pulled By: pdillinger

fbshipit-source-id: d0ddb3587ade91ee2a4d926b475acf7781b03086
2021-09-01 14:28:58 -07:00
Akanksha Mahajan
f9ffeaed3f Update branch name to main in env_librados.md (#8738)
Summary: Pull Request resolved: https://github.com/facebook/rocksdb/pull/8738

Reviewed By: ltamasi

Differential Revision: D30705691

Pulled By: akankshamahajan15

fbshipit-source-id: 44ac8e1c906b1d1d31e9017a700aab5eefe94253
2021-09-01 14:28:58 -07:00
Peter Dillinger
13ded69484 Built-in support for generating unique IDs, bug fix (#8708)
Summary:
Env::GenerateUniqueId() works fine on Windows and on POSIX
where /proc/sys/kernel/random/uuid exists. Our other implementation is
flawed and easily produces collision in a new multi-threaded test.
As we rely more heavily on DB session ID uniqueness, this becomes a
serious issue.

This change combines several individually suitable entropy sources
for reliable generation of random unique IDs, with goal of uniqueness
and portability, not cryptographic strength nor maximum speed.

Specifically:
* Moves code for getting UUIDs from the OS to port::GenerateRfcUuid
rather than in Env implementation details. Callers are now told whether
the operation fails or succeeds.
* Adds an internal API GenerateRawUniqueId for generating high-quality
128-bit unique identifiers, by combining entropy from three "tracks":
  * Lots of info from default Env like time, process id, and hostname.
  * std::random_device
  * port::GenerateRfcUuid (when working)
* Built-in implementations of Env::GenerateUniqueId() will now always
produce an RFC 4122 UUID string, either from platform-specific API or
by converting the output of GenerateRawUniqueId.

DB session IDs now use GenerateRawUniqueId while DB IDs (not as
critical) try to use port::GenerateRfcUuid but fall back on
GenerateRawUniqueId with conversion to an RFC 4122 UUID.

GenerateRawUniqueId is declared and defined under env/ rather than util/
or even port/ because of the Env dependency.

Likely follow-up: enhance GenerateRawUniqueId to be faster after the
first call and to guarantee uniqueness within the lifetime of a single
process (imparting the same property onto DB session IDs).

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8708

Test Plan:
A new mini-stress test in env_test checks the various public
and internal APIs for uniqueness, including each track of
GenerateRawUniqueId individually. We can't hope to verify anywhere close
to 128 bits of entropy, but it can at least detect flaws as bad as the
old code. Serial execution of the new tests takes about 350 ms on
my machine.

Reviewed By: zhichao-cao, mrambacher

Differential Revision: D30563780

Pulled By: pdillinger

fbshipit-source-id: de4c9ff4b2f581cf784fcedb5f39f16e5185c364
2021-08-30 15:20:41 -07:00
Merlin Mao
6c2bd28a61 Update comments, fix typos. (#8721)
Summary:
- Removed the default empty constructors of `TraceWriter` and `TraceReader`.
- Removed unused `ReadFooter()` from `ReplayerImpl`.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8721

Test Plan: None

Reviewed By: zhichao-cao

Differential Revision: D30609743

Pulled By: autopear

fbshipit-source-id: 7e2626b015bd57ebb408a2836b4b4217cea10002
2021-08-27 13:16:32 -07:00
Merlin Mao
f6437ea4d7 Refactor TraceAnalyzer to use TraceRecord::Handler to avoid casting. (#8678)
Summary:
`TraceAnalyzer` privately inherits `TraceRecord::Handler` and `WriteBatch::Handler`.

`trace_analyzer_test` can pass with this change.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8678

Reviewed By: zhichao-cao

Differential Revision: D30459814

Pulled By: autopear

fbshipit-source-id: a27f59ac4600f7c3682830c9b1d9dc79e53425be
2021-08-23 17:18:27 -07:00
Merlin Mao
d10801e983 Allow Replayer to report the results of TraceRecords. (#8657)
Summary:
`Replayer::Execute()` can directly returns the result (e.g, request latency, DB::Get() return code, returned value, etc.)
`Replayer::Replay()` reports the results via a callback function.

New interface:
`TraceRecordResult` in "rocksdb/trace_record_result.h".

`DBTest2.TraceAndReplay` and `DBTest2.TraceAndManualReplay` are updated accordingly.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8657

Reviewed By: ajkr

Differential Revision: D30290216

Pulled By: autopear

fbshipit-source-id: 3c8d4e6b180ec743de1a9d9dcaee86064c74f0d6
2021-08-18 17:06:14 -07:00
Yanqin Jin
2b367fa8cc Fix bug caused by releasing snapshot(s) during compaction (#8608)
Summary:
In debug mode, we are seeing assertion failure as follows

```
db/compaction/compaction_iterator.cc:980: void rocksdb::CompactionIterator::PrepareOutput(): \
Assertion `ikey_.type != kTypeDeletion && ikey_.type != kTypeSingleDeletion' failed.
```

It is caused by releasing earliest snapshot during compaction between the execution of
`NextFromInput()` and `PrepareOutput()`.

In one case, as demonstrated in unit test `WritePreparedTransaction.ReleaseEarliestSnapshotDuringCompaction_WithSD2`,
incorrect result may be returned by a following range scan if we disable assertion, as in opt compilation
level: the SingleDelete marker's sequence number is zeroed out, but the preceding PUT is also
outputted to the SST file after compaction. Due to the logic of DBIter, the PUT will not be
skipped and will be returned by iterator in range scan. https://github.com/facebook/rocksdb/issues/8661 illustrates what happened.

Fix by taking a more conservative approach: make compaction zero out sequence number only
if key is in the earliest snapshot when the compaction starts.

Another assertion failure is
```
Assertion `current_user_key_snapshot_ == last_snapshot' failed.
```

It's caused by releasing the snapshot between the PUT and SingleDelete during compaction.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8608

Test Plan: make check

Reviewed By: jay-zhuang

Differential Revision: D30145645

Pulled By: riversand963

fbshipit-source-id: 699f58e66faf70732ad53810ccef43935d3bbe81
2021-08-17 22:14:20 -07:00
Merlin Mao
74a652a45f Code cleanup for trace replayer (#8652)
Summary:
- Remove extra `;` in trace_record.h
- Remove some unnecessary `assert` in trace_record_handler.cc
- Initialize `env_` after` exec_handler_` in `ReplayerImpl` to let db be asserted in creating the handler before getting `db->GetEnv()`.
- Update history to include the new `TraceReader::Reset()`

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8652

Reviewed By: ajkr

Differential Revision: D30276872

Pulled By: autopear

fbshipit-source-id: 476ee162e0f241490c6209307448343a5b326b37
2021-08-12 09:22:43 -07:00
Merlin Mao
f58d276764 Make TraceRecord and Replayer public (#8611)
Summary:
New public interfaces:
`TraceRecord` and `TraceRecord::Handler`, available in "rocksdb/trace_record.h".
`Replayer`, available in `rocksdb/utilities/replayer.h`.

User can use `DB::NewDefaultReplayer()` to create a Replayer to auto/manual replay a trace file.

Unit tests:
- `./db_test2 --gtest_filter="DBTest2.TraceAndReplay"`: Updated with the internal API changes.
- `./db_test2 --gtest_filter="DBTest2.TraceAndManualReplay"`: New for manual replay.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8611

Reviewed By: ajkr

Differential Revision: D30266329

Pulled By: autopear

fbshipit-source-id: 1ecb3cbbedae0f6a67c18f0cc82e002b4d81b6f8
2021-08-11 19:32:46 -07:00
Peter Dillinger
a7fd1d0881 Make backup restore atomic, with sync option (#8568)
Summary:
Guarantees that if a restore is interrupted, DB::Open will fail. This works by
restoring CURRENT first to CURRENT.tmp then as a final step renaming to CURRENT.

Also makes restore respect BackupEngineOptions::sync (default true). When set,
the restore is guaranteed persisted by the time it returns OK. Also makes the above
atomicity guarantee work in case the interruption is power loss or OS crash (not just
process interruption or crash).

Fixes https://github.com/facebook/rocksdb/issues/8500

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8568

Test Plan:
added to backup mini-stress unit test. Passes with
gtest_repeat=100 (whereas fails 7 times without the CURRENT.tmp)

Reviewed By: akankshamahajan15

Differential Revision: D29812605

Pulled By: pdillinger

fbshipit-source-id: 24e9a993b305b1835ca95558fa7a7152e54cda8e
2021-08-06 09:50:21 -07:00
mrambacher
d057e8326d Make MergeOperator+CompactionFilter/Factory into Customizable Classes (#8481)
Summary:
- Changed MergeOperator, CompactionFilter, and CompactionFilterFactory into Customizable classes.
 - Added Options/Configurable/Object Registration for TTL and Cassandra variants
 - Changed the StringAppend MergeOperators to accept a string delimiter rather than a simple char.  Made the delimiter into a configurable option
 - Added tests for new functionality

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8481

Reviewed By: zhichao-cao

Differential Revision: D30136050

Pulled By: mrambacher

fbshipit-source-id: 271d1772835935b6773abaf018ee71e42f9491af
2021-08-06 08:27:25 -07:00
anand76
c268859aaa Remove corruption error injection in FaultInjectionTestFS (#8616)
Summary:
```FaultInjectionTestFS``` injects various types of read errors in ```FileSystem``` APIs. One type of error is corruption errors, where data is intentionally corrupted or truncated. There is corresponding validation in db_stress to verify that an injected error results in a user visible Get/MultiGet error. However, for corruption errors, its hard to know when a corruption is supposed to be detected by the user request, due to prefetching and, in case of direct IO, padding. This results in false positives. So remove that functionality.

Block checksum validation for Get/MultiGet is confined to ```BlockFetcher```, so we don't lose a lot by disabling this since its a small surface area to test.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8616

Reviewed By: zhichao-cao

Differential Revision: D30074422

Pulled By: anand1976

fbshipit-source-id: 6a61fac18f95514c15364b75013799ddf83294df
2021-08-04 15:48:54 -07:00
Merlin Mao
4811115b3e Revert checkpoint fix (#8607)
Summary:
PR https://github.com/facebook/rocksdb/pull/8572 looses custom types in the options file. Need more API changes to fix this issue. Revert this PR.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8607

Reviewed By: ajkr

Differential Revision: D30058289

Pulled By: autopear

fbshipit-source-id: 78f5a154c0bf193e8441bae4a36fa79b95277fd4
2021-08-02 18:29:35 -07:00
Mikhail Golubev
8f52972cf9 Allow to use a string as a delimiter in StringAppendOperator (#8536)
Summary:
An arbitrary string can be used as a delimiter in StringAppend merge operator
flavor. In particular, it allows using an empty string, combining binary values for
the same key byte-to-byte one next to another.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8536

Reviewed By: mrambacher

Differential Revision: D29962120

Pulled By: zhichao-cao

fbshipit-source-id: 4ef5d846a47835cf428a11200409e30e2dbffc4f
2021-08-02 16:50:41 -07:00
mrambacher
ab7f7c9e49 Allow WAL dir to change with db dir (#8582)
Summary:
Prior to this change, the "wal_dir"  DBOption would always be set (defaults to dbname) when the DBOptions were sanitized.  Because of this setitng in the options file, it was not possible to rename/relocate a database directory after it had been created and use the existing options file.

After this change, the "wal_dir" option is only set under specific circumstances.  Methods were added to the ImmutableDBOptions class to see if it is set and if it is set to something other than the dbname.  Additionally, a method was added to retrieve the effective value of the WAL dir (either the option or the dbname/path).

Tests were added to the core and ldb to test that a database could be created and renamed without issue.  Additional tests for various permutations of wal_dir were also added.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8582

Reviewed By: pdillinger, autopear

Differential Revision: D29881122

Pulled By: mrambacher

fbshipit-source-id: 67d3d033dc8813d59917b0a3fba2550c0efd6dfb
2021-07-30 12:16:44 -07:00
Yanqin Jin
066b51126d Several simple local code clean-ups (#8565)
Summary:
This PR tries to remove some unnecessary checks as well as unreachable code blocks to
improve readability. An obvious non-public API method naming typo is also corrected.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8565

Test Plan: make check

Reviewed By: lth

Differential Revision: D29963984

Pulled By: riversand963

fbshipit-source-id: cc96e8f09890e5cfe9b20eadb63bdca5484c150a
2021-07-30 12:07:49 -07:00
mrambacher
3aee4fbd41 Make EventListener into a Customizable Class (#8473)
Summary:
- Added Type/CreateFromString
- Added ability to load EventListeners to DBOptions
- Since EventListeners did not previously have a Name(), defaulted to "".  If there is no name, the listener cannot be loaded from the ObjectRegistry.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8473

Reviewed By: zhichao-cao

Differential Revision: D29901488

Pulled By: mrambacher

fbshipit-source-id: 2d3a4aa6db1562ac03e7ad41b360e3521d486254
2021-07-27 07:47:02 -07:00
Merlin Mao
55f7ded80d Checkpoint dir options fix (#8572)
Summary:
Originally the 2 options `db_log_dir` and `wal_dir` will be reused in a snapshot db since the options files are just copied. By default, if `wal_dir` was not set when a db was created, it is set to the db's dir. Therefore, the snapshot db will use the same WAL dir. If both the original db and the snapshot db write to or delete from the WAL dir, one may modify or delete files which belong to the other. The same applies to `db_log_dir` as well, but as info log files are not copied or linked, it is simpler for this option.

2 arguments are added to `Checkpoint::CreateCheckpoint()`, allowing to override these 2 options.

`wal_dir`:  If the function argument `wal_dir` is empty, or set to the original db location, or the checkpoint location, the snapshot's `wal_dir` option will be updated to the checkpoint location. Otherwise, the absolute path specified in the argument will be used. During checkpointing, live WAL files will be copied or linked the new location, instead of the current WAL dir specified in the original db.

`db_log_dir`: Same as `wal_dir`, but no files will be copied or linked.

A new unit test was added: `CheckpointTest.CheckpointWithOptionsDirsTest`.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8572

Test Plan:
New unit test
```
checkpoint_test --gtest_filter="CheckpointTest.CheckpointWithOptionsDirsTest"
```

Output
```
Note: Google Test filter = CheckpointTest.CheckpointWithOptionsDirsTest
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from CheckpointTest
[ RUN      ] CheckpointTest.CheckpointWithOptionsDirsTest
[       OK ] CheckpointTest.CheckpointWithOptionsDirsTest (11712 ms)
[----------] 1 test from CheckpointTest (11712 ms total)

[----------] Global test environment tear-down
[==========] 1 test from 1 test case ran. (11713 ms total)
[  PASSED  ] 1 test.
```
This test will fail without this patch. Just modify the code to remove the 2 arguments introduced in this patch in `CreateCheckpoint()`.

Reviewed By: zhichao-cao

Differential Revision: D29832761

Pulled By: autopear

fbshipit-source-id: e6a639b4d674380df82998c0839e79cab695fe29
2021-07-23 11:13:01 -07:00
Drewryz
3b27725245 Fix a minor issue with initializing the test path (#8555)
Summary:
The PerThreadDBPath has already specified a slash. It does not need to be specified when initializing the test path.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8555

Reviewed By: ajkr

Differential Revision: D29758399

Pulled By: jay-zhuang

fbshipit-source-id: 6d2b878523e3e8580536e2829cb25489844d9011
2021-07-23 08:38:45 -07:00
Jay Zhuang
c4a503f3df Fix an race condition during multiple DB opening (#8574)
Summary:
ObjectLibrary is shared between multiple DB instances, the
Register() could have race condition.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8574

Test Plan: pass the failed test

Reviewed By: ajkr

Differential Revision: D29855096

Pulled By: jay-zhuang

fbshipit-source-id: 541eed0bd495d2c963d858d81e7eabf1ba16153c
2021-07-22 13:43:06 -07:00
Zhichao Cao
87e82a41a9 Fix incorrect Status::NoSpace() status check (#8504)
Summary:
If we want to check whether a Status s is NoSpace() or not, we should check the subcode instread of using s==Status::NoSpace(). Fix some of the incorrect check in the ErrorHandler.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8504

Test Plan: make check

Reviewed By: anand1976

Differential Revision: D29601764

Pulled By: zhichao-cao

fbshipit-source-id: cdab56a827891c23746bba9cbb53f169fe35f086
2021-07-20 18:09:51 -07:00
sdong
39a07c9651 DB Stress Reopen write failure to skip WAL (#8548)
Summary:
When DB Stress enables write failure in reopen, WAL files are also created with a wrapper writalbe file which buffers write until fsync. However, crash test currently expects all writes to WAL is persistent. This is at odd with the unsynced bytes dropped. To work it around temporarily, we disable WAL write failure for now.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8548

Test Plan: Run db_stress. Manual printf to make sure only WAL files are skipped.

Reviewed By: jay-zhuang

Differential Revision: D29745095

fbshipit-source-id: 1879dd2c01abad7879ca243ee94570ec47c347f3
2021-07-16 16:09:33 -07:00
Peter Dillinger
df5dc73bec Don't hold DB mutex for block cache entry stat scans (#8538)
Summary:
I previously didn't notice the DB mutex was being held during
block cache entry stat scans, probably because I primarily checked for
read performance regressions, because they require the block cache and
are traditionally latency-sensitive.

This change does some refactoring to avoid holding DB mutex and to
avoid triggering and waiting for a scan in GetProperty("rocksdb.cfstats").
Some tests have to be updated because now the stats collector is
populated in the Cache aggressively on DB startup rather than lazily.
(I hope to clean up some of this added complexity in the future.)

This change also ensures proper treatment of need_out_of_mutex for
non-int DB properties.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8538

Test Plan:
Added unit test logic that uses sync points to fail if the DB mutex
is held during a scan, covering the various ways that a scan might be
triggered.

Performance test - the known impact to holding the DB mutex is on
TransactionDB, and the easiest way to see the impact is to hack the
scan code to almost always miss and take an artificially long time
scanning. Here I've injected an unconditional 5s sleep at the call to
ApplyToAllEntries.

Before (hacked):

    $ TEST_TMPDIR=/dev/shm ./db_bench.base_xxx -benchmarks=randomtransaction,stats -cache_index_and_filter_blocks=1 -bloom_bits=10 -partition_index_and_filters=1 -duration=30 -stats_dump_period_sec=12 -cache_size=100000000 -statistics -transaction_db 2>&1 | egrep 'db.db.write.micros|micros/op'
    randomtransaction :     433.219 micros/op 2308 ops/sec;    0.1 MB/s ( transactions:78999 aborts:0)
    rocksdb.db.write.micros P50 : 16.135883 P95 : 36.622503 P99 : 66.036115 P100 : 5000614.000000 COUNT : 149677 SUM : 8364856
    $ TEST_TMPDIR=/dev/shm ./db_bench.base_xxx -benchmarks=randomtransaction,stats -cache_index_and_filter_blocks=1 -bloom_bits=10 -partition_index_and_filters=1 -duration=30 -stats_dump_period_sec=12 -cache_size=100000000 -statistics -transaction_db 2>&1 | egrep 'db.db.write.micros|micros/op'
    randomtransaction :     448.802 micros/op 2228 ops/sec;    0.1 MB/s ( transactions:75999 aborts:0)
    rocksdb.db.write.micros P50 : 16.629221 P95 : 37.320607 P99 : 72.144341 P100 : 5000871.000000 COUNT : 143995 SUM : 13472323

Notice the 5s P100 write time.

After (hacked):

    $ TEST_TMPDIR=/dev/shm ./db_bench.new_xxx -benchmarks=randomtransaction,stats -cache_index_and_filter_blocks=1 -bloom_bits=10 -partition_index_and_filters=1 -duration=30 -stats_dump_period_sec=12 -cache_size=100000000 -statistics -transaction_db 2>&1 | egrep 'db.db.write.micros|micros/op'
    randomtransaction :     303.645 micros/op 3293 ops/sec;    0.1 MB/s ( transactions:98999 aborts:0)
    rocksdb.db.write.micros P50 : 16.061871 P95 : 33.978834 P99 : 60.018017 P100 : 616315.000000 COUNT : 187619 SUM : 4097407
    $ TEST_TMPDIR=/dev/shm ./db_bench.new_xxx -benchmarks=randomtransaction,stats -cache_index_and_filter_blocks=1 -bloom_bits=10 -partition_index_and_filters=1 -duration=30 -stats_dump_period_sec=12 -cache_size=100000000 -statistics -transaction_db 2>&1 | egrep 'db.db.write.micros|micros/op'
    randomtransaction :     310.383 micros/op 3221 ops/sec;    0.1 MB/s ( transactions:96999 aborts:0)
    rocksdb.db.write.micros P50 : 16.270026 P95 : 35.786844 P99 : 64.302878 P100 : 603088.000000 COUNT : 183819 SUM : 4095918

P100 write is now ~0.6s. Not good, but it's the same even if I completely bypass all the scanning code:

    $ TEST_TMPDIR=/dev/shm ./db_bench.new_skip -benchmarks=randomtransaction,stats -cache_index_and_filter_blocks=1 -bloom_bits=10 -partition_index_and_filters=1 -duration=30 -stats_dump_period_sec=12 -cache_size=100000000 -statistics -transaction_db 2>&1 | egrep 'db.db.write.micros|micros/op'
    randomtransaction :     311.365 micros/op 3211 ops/sec;    0.1 MB/s ( transactions:96999 aborts:0)
    rocksdb.db.write.micros P50 : 16.274362 P95 : 36.221184 P99 : 68.809783 P100 : 649808.000000 COUNT : 183819 SUM : 4156767
    $ TEST_TMPDIR=/dev/shm ./db_bench.new_skip -benchmarks=randomtransaction,stats -cache_index_and_filter_blocks=1 -bloom_bits=10 -partition_index_and_filters=1 -duration=30 -stats_dump_period_sec=12 -cache_size=100000000 -statistics -transaction_db 2>&1 | egrep 'db.db.write.micros|micros/op'
    randomtransaction :     308.395 micros/op 3242 ops/sec;    0.1 MB/s ( transactions:97999 aborts:0)
    rocksdb.db.write.micros P50 : 16.106222 P95 : 37.202403 P99 : 67.081875 P100 : 598091.000000 COUNT : 185714 SUM : 4098832

No substantial difference.

Reviewed By: siying

Differential Revision: D29738847

Pulled By: pdillinger

fbshipit-source-id: 1c5c155f5a1b62e4fea0fd4eeb515a8b7474027b
2021-07-16 14:13:08 -07:00
longlijian
4e4ec16957 Replace the namespace "rocksdb" to "ROCKSDB_NAMESPACE" (#8531)
Summary:
For more detail can reference the https://github.com/facebook/rocksdb/issues/6433
(https://github.com/facebook/rocksdb/pull/6433)

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8531

Reviewed By: siying

Differential Revision: D29717057

Pulled By: ajkr

fbshipit-source-id: 3ccad9501e5612590e54a7cf8c447118f323c7f4
2021-07-15 17:23:39 -07:00
Myth
bbdc4f2e9a Fix a minor issue in checkpoint test case (#8483)
Summary:
A very simple change :)

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8483

Reviewed By: ajkr

Differential Revision: D29558904

Pulled By: mrambacher

fbshipit-source-id: bbe68c20c861103726cb6231ca3fb8fbe1e5a546
2021-07-12 09:09:09 -07:00
sdong
b1a53db327 FaultInjectionTestFS::DeleteFilesCreatedAfterLastDirSync() to recover… (#8501)
Summary:
… small overwritten files.
If a file is overwritten with renamed and the parent path is not synced, FaultInjectionTestFS::DeleteFilesCreatedAfterLastDirSync() will delete the file. However, RocksDB relies on file renaming to be atomic no matter whether the parent directory is synced or not, and the current behavior breaks the assumption and caused some false positive: https://github.com/facebook/rocksdb/pull/8489

Since the atomic renaming is used in CURRENT files, to fix the problem, in FaultInjectionTestFS::DeleteFilesCreatedAfterLastDirSync(), we recover the state of overwritten file if the file is small.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8501

Test Plan: Run stress test for a while and see it doesn't break.

Reviewed By: anand1976

Differential Revision: D29594384

fbshipit-source-id: 589b5c2f0a9d2aca53752d7bdb0231efa5b3ae92
2021-07-07 16:23:23 -07:00
Andrew Kryczka
ed8eb436db Move slow valgrind tests behind -DROCKSDB_FULL_VALGRIND_RUN (#8475)
Summary:
Various tests had disabled valgrind due to it slowing down and timing
out (as is the case right now) the CI runs. Where a test was disabled with no comment,
I assumed slowness was the cause. For these tests that were slow under
valgrind, as well as the ones identified in https://github.com/facebook/rocksdb/issues/8352, this PR moves them
behind the compiler flag `-DROCKSDB_FULL_VALGRIND_RUN`.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8475

Test Plan: running `make full_valgrind_test`, `make valgrind_test`, `make check`; will verify they appear working correctly

Reviewed By: jay-zhuang

Differential Revision: D29504843

Pulled By: ajkr

fbshipit-source-id: 2aac90749cfbd30d5ce11cb29a07a1b9314eeea7
2021-07-07 11:14:05 -07:00
anand76
df4197ca6e Bypass buffer in TestFSWritableFile if direct IO is enabled (#8490)
Summary:
```TestFSWritableFile``` buffers data in ```Append``` in order to simulate unsynced data loss on crash. This is only required for buffered IO and should be disabled for direct IO. Otherwise, it causes crash tests to assert on the buffer address alignment - ```db_stress: env/io_posix.cc:1194: virtual rocksdb::IOStatus rocksdb::PosixWritableFile::Append(const rocksdb::Slice&, const rocksdb::IOOptions&, rocksdb::IODebugContext*): Assertion `IsSectorAligned(data.data(), GetRequiredBufferAlignment())' failed.```.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8490

Reviewed By: zhichao-cao

Differential Revision: D29565080

Pulled By: anand1976

fbshipit-source-id: 682831fd66ed3b9597caa74fc453e22dfaf9b973
2021-07-06 16:46:16 -07:00
sdong
f33611d5e9 Stress test to inject read failures in DB reopen (#8476)
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
2021-07-06 11:05:27 -07:00
sdong
ba224b75c7 Stress Test to inject write failures in reopen (#8474)
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
2021-06-30 16:46:41 -07:00
Zhichao Cao
a904c62d28 Using existing crc32c checksum in checksum handoff for Manifest and WAL (#8412)
Summary:
In PR https://github.com/facebook/rocksdb/issues/7523 , checksum handoff is introduced in RocksDB for WAL, Manifest, and SST files. When user enable checksum handoff for a certain type of file, before the data is written to the lower layer storage system, we calculate the checksum (crc32c) of each piece of data and pass the checksum down with the data, such that data verification can be down by the lower layer storage system if it has the capability. However, it cannot cover the whole lifetime of the data in the memory and also it potentially introduces extra checksum calculation overhead.

In this PR, we introduce a new interface in WritableFileWriter::Append, which allows the caller be able to pass the data and the checksum (crc32c) together. In this way, WritableFileWriter can directly use the pass-in checksum (crc32c) to generate the checksum of data being passed down to the storage system. It saves the calculation overhead and achieves higher protection coverage. When a new checksum is added with the data, we use Crc32cCombine https://github.com/facebook/rocksdb/issues/8305 to combine the existing checksum and the new checksum. To avoid the segmenting of data by rate-limiter before it is stored, rate-limiter is called enough times to accumulate enough credits for a certain write. This design only support Manifest and WAL which use log_writer in the current stage.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8412

Test Plan: make check, add new testing cases.

Reviewed By: anand1976

Differential Revision: D29151545

Pulled By: zhichao-cao

fbshipit-source-id: 75e2278c5126cfd58393c67b1efd18dcc7a30772
2021-06-25 00:47:17 -07:00
Levi Tamasi
3ab0eae860 Deflake BlobDBTest.SnapshotAndGarbageCollection (#8444)
Summary:
This test case has been failing occasionally due to automatic
compactions kicking in, resulting in GC generating additional
blob files that the test did not expect. Disabling automatic
compactions to get rid of this flakiness.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8444

Test Plan: `gtest-parallel --repeat=1000 ./blob_db_test --gtest_filter="BlobDBTest.SnapshotAndGarbageCollection"`

Reviewed By: jay-zhuang

Differential Revision: D29316987

Pulled By: ltamasi

fbshipit-source-id: 9815d189ed7d63890622768675a01990e3680221
2021-06-22 17:34:03 -07:00
Jay Zhuang
f89423a57a Revert "Revert "Snapshot release triggered compaction without multiple tombstones (#8357)" (#8410)" (#8438)
Summary:
This reverts commit 25be1ed66a.

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8438

Test Plan: Run the impacted mysql test 40 times

Reviewed By: ajkr

Differential Revision: D29286247

Pulled By: jay-zhuang

fbshipit-source-id: d3bd056971a19a8b012d5d0295fa045c012b3c04
2021-06-22 11:10:03 -07:00
聂佩轩
d53f7ff69a Add DeteleRange support for DBWithTTL (#8384)
Summary:
This commit is for enabling `DBWithTTL` to use `DeteleRange` which it cannot before.
As (int32_t)Timestamp is suffixed to values in `DBWithTTL`, there is no reason that it
cannot use the common used api. I added `DeleteRangeCF` in `DBWithTTLImpl::Write`
so that we can use `DeteleRange` normally. When we run code like
`dbWithTtl->DeleteRange(start, end)`, it executes`WriteBatchInternal::DeleteRange`
internally. Intended to fix https://github.com/facebook/rocksdb/issues/7218

Pull Request resolved: https://github.com/facebook/rocksdb/pull/8384

Test Plan: added corresponded testing logic to existing unit test

Reviewed By: jay-zhuang

Differential Revision: D29176734

fbshipit-source-id: 6874ed979fc08e1d138149d03653e43a75f0e0e6
2021-06-17 16:00:50 -07:00