Summary:
We occasionally get write stalls (>1s Write() calls) on HDD under read load. The following timers explain almost all of the stalls:
- perf_context.db_mutex_lock_nanos
- perf_context.db_condition_wait_nanos
- iostats_context.open_time
- iostats_context.allocate_time
- iostats_context.write_time
- iostats_context.range_sync_time
- iostats_context.logger_time
In my experiments each of these occasionally takes >1s on write path under some workload. There are rare cases when Write() takes long but none of these takes long.
Test Plan: Added code to our application to write the listed timings to log for slow writes. They usually add up to almost exactly the time Write() call took.
Reviewers: rven, yhchiang, sdong
Reviewed By: sdong
Subscribers: march, dhruba, tnovak
Differential Revision: https://reviews.facebook.net/D39177
Summary: It used to be no good (known to me) non-intrusive way to wrap WritableFile - you can't call protected virtual methods of the wrapped pointer to WritableFile. This diff adds a convenience class WritableFileWrapper that makes wrapping WritableFile both possible and easy.
Test Plan: `make clean; make -j release`, `make clean; OPT=-DROCKSDB_LITE make release`, `make clean; USE_CLANG=1 make -j all`.
Reviewers: sdong, yhchiang, rven
Reviewed By: rven
Subscribers: dhruba, tnovak, march
Differential Revision: https://reviews.facebook.net/D39147
Summary: Optimistic transactions supporting begin/commit/rollback semantics. Currently relies on checking the memtable to determine if there are any collisions at commit time. Not yet implemented would be a way of enuring the memtable has some minimum amount of history so that we won't fail to commit when the memtable is empty. You should probably start with transaction.h to get an overview of what is currently supported.
Test Plan: Added a new test, but still need to look into stress testing.
Reviewers: yhchiang, igor, rven, sdong
Reviewed By: sdong
Subscribers: adamretter, MarkCallaghan, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D33435
Summary:
For transactions, we are using the memtables to validate that there are no write conflicts. But after flushing, we don't have any memtables, and transactions could fail to commit. So we want to someone keep around some extra history to use for conflict checking. In addition, we want to provide a way to increase the size of this history if too many transactions fail to commit.
After chatting with people, it seems like everyone prefers just using Memtables to store this history (instead of a separate history structure). It seems like the best place for this is abstracted inside the memtable_list. I decide to create a separate list in MemtableListVersion as using the same list complicated the flush/installalflushresults logic too much.
This diff adds a new parameter to control how much memtable history to keep around after flushing. However, it sounds like people aren't too fond of adding new parameters. So I am making the default size of flushed+not-flushed memtables be set to max_write_buffers. This should not change the maximum amount of memory used, but make it more likely we're using closer the the limit. (We are now postponing deleting flushed memtables until the max_write_buffer limit is reached). So while we might use more memory on average, we are still obeying the limit set (and you could argue it's better to go ahead and use up memory now instead of waiting for a write stall to happen to test this limit).
However, if people are opposed to this default behavior, we can easily set it to 0 and require this parameter be set in order to use transactions.
Test Plan: Added a xfunc test to play around with setting different values of this parameter in all tests. Added testing in memtablelist_test and planning on adding more testing here.
Reviewers: sdong, rven, igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37443
Summary:
Allow EventLogger to directly log from a JSONWriter. This allows
the JSONWriter to be shared by EventLogger and potentially EventListener,
which is an important step to integrate EventLogger and EventListener.
This patch also rewrites EventLoggerHelpers::LogTableFileCreation(),
which uses the new API to generate identical log.
Test Plan:
Run db_bench in debug mode and make sure the log is correct and no
assertions fail.
Reviewers: sdong, anthony, igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38709
Summary: Having stats in our LOG more often will help a lot with perf debugging.
Test Plan: none
Reviewers: sdong, MarkCallaghan
Reviewed By: MarkCallaghan
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38781
Summary: Rename JSONWritter to JSONWriter
Test Plan: make
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38733
Summary:
sync_file_range is not always asyncronous and thus can block writes if we do this for WAL in the foreground thread. See more here: http://yoshinorimatsunobu.blogspot.com/2014/03/how-syncfilerange-really-works.html
Some users don't want us to call sync_file_range on WALs. Some other do.
Thus, I'm adding a separate option wal_bytes_per_sync to control calling
sync_file_range on WAL files. bytes_per_sync will apply only to table
files now.
Test Plan: no more sync_file_range for WAL as evidenced by strace
Reviewers: yhchiang, rven, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38253
Summary:
Fixed the following compile errors due to some gcc does not have std::map::emplace
util/thread_status_impl.cc: In static member function ‘static std::map<std::basic_string<char>, long unsigned int> rocksdb::ThreadStatus::InterpretOperationProperties(rocksdb::ThreadStatus::OperationType, const uint64_t*)’:
util/thread_status_impl.cc:88:20: error: ‘class std::map<std::basic_string<char>, long unsigned int>’ has no member named ‘emplace’
util/thread_status_impl.cc:90:20: error: ‘class std::map<std::basic_string<char>, long unsigned int>’ has no member named ‘emplace’
util/thread_status_impl.cc:94:20: error: ‘class std::map<std::basic_string<char>, long unsigned int>’ has no member named ‘emplace’
util/thread_status_impl.cc:96:20: error: ‘class std::map<std::basic_string<char>, long unsigned int>’ has no member named ‘emplace’
util/thread_status_impl.cc:98:20: error: ‘class std::map<std::basic_string<char>, long unsigned int>’ has no member named ‘emplace’
util/thread_status_impl.cc:101:20: error: ‘class std::map<std::basic_string<char>, long unsigned int>’ has no member named ‘emplace’
make: *** [util/thread_status_impl.o] Error 1
Test Plan: make db_bench
Reviewers: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38643
Summary: Call Flush() function instead
Test Plan: make all check
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D38583
Summary:
Allow GetThreadList to report Flush properties, which includes:
* job id
* number of bytes that has been written since flush started.
* total size of input mem-tables
Test Plan:
./db_bench --threads=30 --num=1000000 --benchmarks=fillrandom --thread_status_per_interval=100 --value_size=1000
Sample output from db_bench which tracks same flush job
ThreadID ThreadType cfName Operation ElapsedTime Stage State OperationProperties
140213879898240 High Pri default Flush 5789 us FlushJob::WriteLevel0Table BytesMemtables 4112835 | BytesWritten 577104 | JobID 8 |
ThreadID ThreadType cfName Operation ElapsedTime Stage State OperationProperties
140213879898240 High Pri default Flush 30.634 ms FlushJob::WriteLevel0Table BytesMemtables 4112835 | BytesWritten 1734865 | JobID 8 |
Reviewers: rven, igor, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38505
Summary: Make ThreadStatus::InterpretOperationProperties take const uint64_t*
Test Plan:
make
make OPT=-DROCKSDB_LITE shared_lib
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D38445
Summary:
Added these events:
* Recovery start, finish and also when recovery creates a file
* Trivial move
* Compaction start, finish and when compaction creates a file
* Flush start, finish
Also includes small fix to EventLogger
Also added option ROCKSDB_PRINT_EVENTS_TO_STDOUT which is useful when we debug things. I've spent far too much time chasing LOG files.
Still didn't get sst table properties in JSON. They are written very deeply into the stack. I'll address in separate diff.
TODO:
* Write specification. Let's first use this for a while and figure out what's good data to put here, too. After that we'll write spec
* Write tools that parse and analyze LOGs. This can be in python or go. Good intern task.
Test Plan: Ran db_bench with ROCKSDB_PRINT_EVENTS_TO_STDOUT. Here's the output: https://phabricator.fb.com/P19811976
Reviewers: sdong, yhchiang, rven, MarkCallaghan, kradhakrishnan, anthony
Reviewed By: anthony
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37521
Test Plan: Verified that valgrind build passes for cache_test
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37665
Summary: Added keyword override for SetCapacity()
Test Plan: Fixes build
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37647
Summary:
When new capacity is larger than existing capacity, simply update the capacity to the new valie
When new capacity is less than existing capacity, but more than the usage, simply update the capacity to new value
When new capacity is less than the existing capacity and existing usage both, try to purge entries in LRU if feasible to make usage < capacity
Test Plan: Created unit tests in cache_test.cc
Reviewers: sdong, rven, yhchiang, igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37527
Summary:
Make it build for CYGWIN.
Need to define "-std=gnu++11" instead of "-std=c++11" and use some replacement functions.
Test Plan: Build it and run some unit tests in CYGWIN
Reviewers: yhchiang, rven, anthony, kradhakrishnan, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37605
Summary:
Based on feedback from D37083.
Are all of these correct? In some spaces it seems like we're doing SetMaxPossibleForUserKey() although we want the smallest possible internal key for user key.
Test Plan: make check
Reviewers: sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37341
Summary: To further distinguish the corruption cases were caused by storage media or in memory states when writing it, add a paranoid check after writing the file to iterate all the rows.
Test Plan: Add a new unit test for it
Reviewers: rven, igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D37335
Summary:
Without this change, someone on the machine on which
I run "make check" could cause me to overwrite arbitrary
files owned by me, via a symlink attack.
Instead of using a predictable temporary directory and
accepting to use a preexisting one, always create a new
one using mkdtemp. If $TEST_IOCTL_FRIENDLY_TMPDIR is
set and usable, attempt first to find a usable
temporary directory therein. If not, or if unusable,
then try /var/tmp and /tmp. If none of those is usable
abort with a diagnostic.
To do that, I added a new class.
Its constructor finds a suitable directory or aborts,
the sole member prints that directory's name, and the
destructor unlinks what should be an empty directory.
Note that while the code before this did not remove
its temporary directory, there was only one per $UID.
Now, there would be at least one per run or one per
test, depending on implementation, so it is important
to remove them.
Test Plan:
Run this on a fedora rawhide system, where /tmp
is a tmpfs file system, and /var/tmp is ext4.
# This gives a diagnostic that /dev/shm is not suitable
# and ends up using /var/tmp.
TEST_IOCTL_FRIENDLY_TMPDIR=/dev/shm ./env_test
# Uses /var/tmp; same as when envvar not set.
TEST_IOCTL_FRIENDLY_TMPDIR=/var/tmp ./env_test
# Uses /tmp unless it's tmpfs, in which case it gives
# a diagnostic and uses /var/tmp.
TEST_IOCTL_FRIENDLY_TMPDIR=/tmp ./env_test
Reviewers: ljin, rven, igor.sugak, yhchiang, sdong, igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37287
Summary:
[NB: this is a prerequisite for the /tmp-abuse-fixing patch]
This avoids spurious test failure on Linux systems
like Fedora for which /tmp is a tmpfs file system.
On a devtmpfs file
system, ioctl(fd, FS_IOC_GETVERSION, &version) returns -1 with
errno == ENOTTTY, indicating that that ioctl is not supported
on such a file system. Do not let this cause test failures, e.g.,
where env_test would assert that file->GetUniqueId(...) > 0.
Before this change, ./env_test would fail these three tests
on a fedora rawhide system:
[ FAILED ] 3 tests, listed below:
[ FAILED ] EnvPosixTest.RandomAccessUniqueID
[ FAILED ] EnvPosixTest.RandomAccessUniqueIDConcurrent
[ FAILED ] EnvPosixTest.RandomAccessUniqueIDDeletes
3 FAILED TESTS
The fix:
When support for that ioctl is lacking, skip each affected test.
Could be improved by noting which sub-tests are being skipped.
Test Plan:
run these on F21 and note that they now pass.
TEST_TMPDIR=/dev/shm/rdb ./env_test
./env_test
Reviewers: ljin, rven, igor.sugak, yhchiang, sdong, igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D37323
Summary: For some reason reduce_levels is opening the databse with 65.000 levels. This makes ComputeCompactionScore() function terribly slow and the tests is also very slow (20seconds).
Test Plan: mr reduce_levels_test now takes 20ms
Reviewers: sdong, rven, kradhakrishnan, yhchiang
Reviewed By: yhchiang
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D37059
Summary:
Allow users to give a callback function with parameter using sync point, so more complicated verification can be done in tests.
Use it in DBTest.DynamicLevelCompressionPerLevel2 so that failures will be more easy to debug.
Test Plan: Run all tests. Run DBTest.DynamicLevelCompressionPerLevel2 with valgrind check.
Reviewers: rven, yhchiang, anthony, kradhakrishnan, igor
Reviewed By: igor
Subscribers: leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D36999
Summary: We should use mocked-out env for these tests to make it more realiable. Added benefit is that instead of actually sleeping for 3 seconds, we can instead pretend to sleep and just increase time counters.
Test Plan: for i in `seq 100`; do ./wal_manager_test --gtest_filter=WalManagerTest.WALArchivalTtl ;done
Reviewers: rven, meyering
Reviewed By: meyering
Subscribers: meyering, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36951
Summary: As title. For every operation we're asserting Valid(), which sorts the data. That's pretty terrible. We have to be careful to have decent performance even with DEBUG builds.
Test Plan: make check
Reviewers: sdong, rven, yhchiang, MarkCallaghan
Reviewed By: MarkCallaghan
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36969
Summary:
1. it doesn't work
2. we're not using it
In the future, if we need general benchmark framework, we should probably use https://github.com/google/benchmark
Test Plan: make all
Reviewers: yhchiang, rven, anthony, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36777
Summary:
Fixed xfunc related compile errors in ROCKSDB_LITE
Now make OPT=-DROCKSDB_LITE shared_lib -j32 would work
Test Plan:
make clean
make OPT=-DROCKSDB_LITE shared_lib -j32
make clean
make OPT=-DROCKSDB_LITE static_lib -j32
Reviewers: sdong, igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36825
Summary: Now we add warnings when user configures compression and the compression is not supported.
Test Plan:
Configured compression to non-supported values. Observed messages in my log:
2015/03/26-12:17:57.586341 7ffb8a496840 [WARN] Compression type chosen for level 2 is not supported: LZ4. RocksDB will not compress data on level 2.
2015/03/26-12:19:10.768045 7f36f15c5840 [WARN] Compression type chosen is not supported: LZ4. RocksDB will not compress data.
Reviewers: rven, sdong, yhchiang
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D35979
Summary:
Currently users have no idea a key is add, delete or merge from TablePropertiesCollector call back. Add a new function to add it.
Also refactor the codes so that
(1) make table property collector and internal table property collector two separate data structures with the later one now exposed
(2) table builders only receive internal table properties
Test Plan: Add cases in table_properties_collector_test to cover both of old and new ways of using TablePropertiesCollector.
Reviewers: yhchiang, igor.sugak, rven, igor
Reviewed By: rven, igor
Subscribers: meyering, yoshinorim, maykov, leveldb, dhruba
Differential Revision: https://reviews.facebook.net/D35373
Summary:
With this change, we use L1 and up to store compaction outputs in universal compaction.
The compaction pick logic stays the same. Outputs are stored in the largest "level" as possible.
If options.num_levels=1, it behaves all the same as now.
Test Plan:
1) convert most of existing unit tests for universal comapaction to include the option of one level and multiple levels.
2) add a unit test to cover parallel compaction in universal compaction and run it in one level and multiple levels
3) add unit test to migrate from multiple level setting back to one level setting
4) add a unit test to insert keys to trigger multiple rounds of compactions and verify results.
Reviewers: rven, kradhakrishnan, yhchiang, igor
Reviewed By: igor
Subscribers: meyering, leveldb, MarkCallaghan, dhruba
Differential Revision: https://reviews.facebook.net/D34539
Summary:
Just couple of small changes:
1. removed signal_test, since it doesn't seem useful and we don't even run it as part of `make check`
2. moved perf_context_test to TESTS instead of PROGRAMS
3. `make release` probably shouldn't compile benchmarks. We currently rely on `make release` building db_bench (via Jenkins), so I left db_bench there.
This is just a minor cleanup. We need to rethink our targets since they are a bit messy right now. We can do this during our tech debt week.
Test Plan: make release
Reviewers: anthony, rven, yhchiang, sdong, meyering
Reviewed By: meyering
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D36171
Summary:
This makes run_flash_bench.sh configurable. Previously it was hardwired for 1B keys and tests
ran for 12 hours each. That kept me from using it. This makes it configuable, adds more tests,
makes the duration per-test configurable and refactors the test scripts.
Adds the seekrandomwhilemerging test to db_bench which is the same as seekrandomwhilewriting except
the writer thread does Merge rather than Put.
Forces the stall-time column in compaction IO stats to use a fixed format (H:M:S) which makes
it easier to scrape and parse. Also adds an option to AppendHumanMicros to force a fixed format.
Sometimes automation and humans want different format.
Calls thread->stats.AddBytes(bytes); in db_bench for more tests to get the MB/sec summary
stats in the output at test end.
Adds the average ingest rate to compaction IO stats. Output now looks like:
https://gist.github.com/mdcallag/2bd64d18be1b93adc494
More information on the benchmark output is at https://gist.github.com/mdcallag/db43a58bd5ac624f01e1
For benchmark.sh changes default RocksDB configuration to reduce stalls:
* min_level_to_compress from 2 to 3
* hard_rate_limit from 2 to 3
* max_grandparent_overlap_factor and max_bytes_for_level_multiplier from 10 to 8
* L0 file count triggers from 4,8,12 to 4,12,20 for (start,stall,stop)
Task ID: #6596829
Blame Rev:
Test Plan:
run tools/run_flash_bench.sh
Revert Plan:
Database Impact:
Memcache Impact:
Other Notes:
EImportant:
- begin *PUBLIC* platform impact section -
Bugzilla: #
- end platform impact -
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D36075
The error was:
util/logging.cc: In function 'int rocksdb::AppendHumanMicros(uint64_t, char*, int)':
error: util/logging.cc:41:39: integer overflow in expression [-Werror=overflow]
} else if (micros < 1000000l * 60 * 60) {
^
error: util/logging.cc:41:39: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
Summary: Fix compile error when NROCKSDB_THREAD_STATUS is not used.
Test Plan: make dbg OPT=-DNROCKSDB_THREAD_STATUS -j32
Reviewers: sdong, igor, rven
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D35847
Summary:
We have addded new stats and perf_context for measuring the merge and filter operation time consumption.
We have bounded all the merge operations within the GUARD statment and collected the total time for these operations in the DB.
Test Plan: WIP
Reviewers: rven, yhchiang, kradhakrishnan, igor, sdong
Reviewed By: sdong
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D34377
Summary:
Report elapsed time of a thread operation in micros in ThreadStatus
instead of start time of a thread operation in seconds since the
Epoch, 1970-01-01 00:00:00 (UTC).
Test Plan:
./db_bench --benchmarks=fillrandom --num=100000 --threads=40 \
--max_background_compactions=10 --max_background_flushes=3 \
--thread_status_per_interval=1000 --key_size=16 --value_size=1000 \
--num_column_families=10
Sample Output:
ThreadID ThreadType cfName Operation ElapsedTime Stage State
140667724562496 High Pri column_family_name_000002 Flush 772.419 ms FlushJob::WriteLevel0Table
140667728756800 High Pri default Flush 617.845 ms FlushJob::WriteLevel0Table
140667732951104 High Pri column_family_name_000005 Flush 772.078 ms FlushJob::WriteLevel0Table
140667875557440 Low Pri column_family_name_000008 Compaction 1409.216 ms CompactionJob::Install
140667737145408 Low Pri
140667749728320 Low Pri
140667816837184 Low Pri column_family_name_000007 Compaction 1071.815 ms CompactionJob::ProcessKeyValueCompaction
140667787477056 Low Pri column_family_name_000009 Compaction 772.516 ms CompactionJob::ProcessKeyValueCompaction
140667741339712 Low Pri
140667758116928 Low Pri column_family_name_000004 Compaction 620.739 ms CompactionJob::ProcessKeyValueCompaction
140667753922624 Low Pri
140667842003008 Low Pri column_family_name_000006 Compaction 1260.079 ms CompactionJob::ProcessKeyValueCompaction
140667745534016 Low Pri
Reviewers: sdong, igor, rven
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D35769
Summary:
Limiting verbose printing to "command=scan"
Test Plan:
Run make check and manual testing of sst_dump_test
Reviewers: sdong
CC: leveldb
Task ID: #6575982
Blame Rev:
Summary: Modified rocksdb status assertions ASSERT_OK and EXPECT_OK to print error message from Status::ToString() when failed.
Test Plan: Modify a test to fail status assertions ASSERT_OK and EXPECT_OK and notice an error message that came from Status::ToString()
Reviewers: meyering, sdong, yhchiang, igor
Reviewed By: igor
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D35469
Summary: Cleaning util/testharness.h
Test Plan:
Make completes with no errors.
```
% make all
```
Reviewers: meyering, igor, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D35487
Mike Kolupaev