Summary:
Before this patch, any function call to ThreadStatusUtil might automatically initialize and register the thread status data. However, if it is the user-thread making this call, the allocated thread-status-data will never be released as such threads are not managed by rocksdb.
In this patch, I remove the automatic-initialization part. Thread-status data is only initialized and uninitialized in Env during the thread creation and destruction.
Test Plan:
db_test
thread_list_test
listener_test
Reviewers: igor, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D40017
Summary:
Add Env::GetThreadID(), which returns the ID of the current thread.
In addition, make GetThreadList() and InfoLog use same unique ID for the same thread.
Test Plan:
db_test
listener_test
Reviewers: igor, rven, IslamAbdelRahman, kradhakrishnan, sdong
Reviewed By: sdong
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D39735
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:
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:
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:
Right now if they system we are compiling on is not Linux and not Mac we will get a compilation error
this diff use chrono as a fallback when we are compiling on something other than Linux/FreeBSD/Mac
Test Plan:
compile on CentOS/FreeBSD
./db_test (still running)
Reviewers: igor
Reviewed By: igor
Subscribers: dhruba
Differential Revision: https://reviews.facebook.net/D35277
Summary:
On RocksDB, when there are multiple instances doing
flushes/compactions in the background, the close call takes a long time
because the flushes/compactions need to complete before the database can
shut down. If another instance is using the background threads and the compaction for this instance is in the queue since it has been scheduled, we still cannot shutdown. We now remove the scheduled background tasks which have not yet started running, so that shutdown is speeded up.
Test Plan: DB Test added.
Reviewers: yhchiang, igor, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D33741
Summary:
For some reason, libstdc++ implements steady_clock::now() using syscall instead of VDSO optimized clock_gettime() when using glibc 2.16 and earlier. This leads to significant performance degradation for users with older glibcs. See bug reported here: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=59177
We observed this behavior when testing mongo on AWS hosts. Facebook hosts are unaffected since we use glibc2.17 and 2.20.
Revert "Fix timing"
This reverts commit 965d9d50b8.
Revert "Use chrono for timing"
This reverts commit 001ce64dc7.
Test Plan: make check
Reviewers: MarkCallaghan, yhchiang, rven, meyering, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D34371
Summary:
When using latest clang (3.6 or 3.7/trunck) rocksdb is failing with many errors. Almost all of them are missing override errors. This diff adds missing override keyword. No manual changes.
Prerequisites: bear and clang 3.5 build with extra tools
```lang=bash
% USE_CLANG=1 bear make all # generate a compilation database http://clang.llvm.org/docs/JSONCompilationDatabase.html
% clang-modernize -p . -include . -add-override
% make format
```
Test Plan:
Make sure all tests are passing.
```lang=bash
% #Use default fb code clang.
% make check
```
Verify less error and no missing override errors.
```lang=bash
% # Have trunk clang present in path.
% ROCKSDB_NO_FBCODE=1 CC=clang CXX=clang++ make
```
Reviewers: igor, kradhakrishnan, rven, meyering, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D34077
Summary:
scan-build complains with division by zero warning in a test. Added an assertion to prevent this.
scan-build report: http://home.fburl.com/~sugak/latest6/report-c61be9.html#EndPath
Test Plan:
Make sure scan-build does not report 'Division by zero' and all tests are passing.
```lang=bash
% make analyze
% make check
```
Reviewers: igor, meyering
Reviewed By: meyering
Subscribers: sdong, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D33495
Summary:
Add structures for exposing events and operations. Event describes
high-level action about a thread such as doing compaciton or
doing flush, while an operation describes lower-level action
of a thread such as reading / writing a SST table, waiting for
mutex. Events and operations are designed to be independent.
One thread would typically involve in one event and one operation.
Code instrument will be in a separate diff.
Test Plan:
Add unit-tests in thread_list_test
make dbg -j32
./thread_list_test
export ROCKSDB_TESTS=ThreadList
./db_test
Reviewers: ljin, igor, sdong
Reviewed By: sdong
Subscribers: rven, jonahcohen, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D29781
Summary:
GetThreadList() feature depends on the thread creation and destruction, which is currently handled under Env.
This patch moves GetThreadList() feature under Env to better manage the dependency of GetThreadList() feature
on thread creation and destruction.
Renamed ThreadStatusImpl to ThreadStatusUpdater. Add ThreadStatusUtil, which is a static class contains
utility functions for ThreadStatusUpdater.
Test Plan: run db_test, thread_list_test and db_bench and verify the life cycle of Env and ThreadStatusUpdater is properly managed.
Reviewers: igor, sdong
Reviewed By: sdong
Subscribers: ljin, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D30057
Summary: Why do we assert here? This doesn't seem like user friendly thing to do :)
Test Plan: none
Reviewers: sdong, yhchiang, rven
Reviewed By: rven
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D30027
Summary:
Add GetThreadList API, which allows developer to track the
status of each process. Currently, calling GetThreadList will
only get the list of background threads in RocksDB with their
thread-id and thread-type (priority) set. Will add more support
on this in the later diffs.
ThreadStatus currently has the following properties:
// An unique ID for the thread.
const uint64_t thread_id;
// The type of the thread, it could be ROCKSDB_HIGH_PRIORITY,
// ROCKSDB_LOW_PRIORITY, and USER_THREAD
const ThreadType thread_type;
// The name of the DB instance where the thread is currently
// involved with. It would be set to empty string if the thread
// does not involve in any DB operation.
const std::string db_name;
// The name of the column family where the thread is currently
// It would be set to empty string if the thread does not involve
// in any column family.
const std::string cf_name;
// The event that the current thread is involved.
// It would be set to empty string if the information about event
// is not currently available.
Test Plan:
./thread_list_test
export ROCKSDB_TESTS=GetThreadList
./db_test
Reviewers: rven, igor, sdong, ljin
Reviewed By: ljin
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D25047
Summary: Store links to live files in directory on same disk
Test Plan:
Take snapshot and open it. Added a test GetSnapshotLink in
db_test.
Reviewers: sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D28713
Summary: So iOS size_t is 32-bit, so we need to static_cast<size_t> any uint64_t :(
Test Plan: TARGET_OS=IOS make static_lib
Reviewers: dhruba, ljin, yhchiang, rven, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D28743
Summary:
We need to turn on -Wshorten-64-to-32 for mobile. See D1671432 (internal phabricator) for details.
This diff turns on the warning flag and fixes all the errors. There were also some interesting errors that I might call bugs, especially in plain table. Going forward, I think it makes sense to have this flag turned on and be very very careful when converting 64-bit to 32-bit variables.
Test Plan: compiles
Reviewers: ljin, rven, yhchiang, sdong
Reviewed By: yhchiang
Subscribers: bobbaldwin, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D28689
Summary:
With the patch, thread pool size will be automatically increased if DB's options ask for more parallelism of compactions or flushes.
Too many users have been confused by the API. Change it to make it harder for users to make mistakes
Test Plan: Add two unit tests to cover the function.
Reviewers: yhchiang, rven, igor, MarkCallaghan, ljin
Reviewed By: ljin
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D27555
Summary: Revmoed this in D25641, causing compiler complain. put it back
Test Plan: make release
Reviewers: igor, yhchiang, rven, sdong
Reviewed By: sdong
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D27891
Summary:
ftruncate does not always free preallocated unused space at the end of file.
In some cases, we pin too much disk space than it should
Test Plan: env_test
Reviewers: sdong, rven, yhchiang, igor
Reviewed By: igor
Subscribers: nkg-, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D25641
Summary: Since we depend on C++11, we might as well use it for timing, instead of this platform-depended code.
Test Plan: Ran autovector_test, which reports time and confirmed that output is similar to master
Reviewers: ljin, sdong, yhchiang, rven, dhruba
Reviewed By: dhruba
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D25587
Summary:
BlockBasedTable sst file size can grow to a large size when universal
compaction is used. When index block exceeds 2G, pread seems to fail and
return truncated data and causes "trucated block" error. I tried to use
```
#define _FILE_OFFSET_BITS 64
```
But the problem still persists. Splitting a big write/read into smaller
batches seems to solve the problem.
Test Plan:
successfully compacted a case with resulting sst file at ~90G (2.1G
index block size)
Reviewers: yhchiang, igor, sdong
Reviewed By: sdong
Subscribers: leveldb
Differential Revision: https://reviews.facebook.net/D22569
Summary: This is a linux-specific system call.
Test Plan: ran db_bench
Reviewers: igor, yhchiang, sdong
Reviewed By: sdong
Subscribers: haobo, leveldb
Differential Revision: https://reviews.facebook.net/D21183
Summary:
Add option and plugin rate limiter for PosixWritableFile. The rate
limiter only applies to flush and compaction. WAL and MANIFEST are
excluded from this enforcement.
Test Plan: db_test
Reviewers: igor, yhchiang, sdong
Reviewed By: sdong
Subscribers: leveldb
Differential Revision: https://reviews.facebook.net/D19425
Summary:
This diff allows the I/O stats about Flush and Compaction to be reported
in a more accurate way. Instead of measuring the size of a file, it
measure I/O cost in per read / write basis.
Test Plan: make all check
Reviewers: sdong, igor, ljin
Reviewed By: ljin
Subscribers: leveldb
Differential Revision: https://reviews.facebook.net/D19383
Summary: 220132b65e correctly fixed the issue of thread ID printing when terminating a thread. Nothing wrong with it. This diff prints the ID in the same way as in PosixLogger::logv() so that users can be more easily to correlates them.
Test Plan: run env_test and make sure it prints correctly.
Reviewers: igor, haobo, ljin, yhchiang
Reviewed By: yhchiang
Subscribers: dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D18819
Summary: Per request from @nkg-, temporarily print thread ID when a thread terminates. It is a temp solution as we try to minimized stderr messages.
Test Plan: env_test
Reviewers: haobo, igor, dhruba
Reviewed By: igor
CC: nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D18753
Summary:
Add a feature to decrease the number of threads in thread pool.
Also instantly schedule more threads if number of threads is increased.
Here is the way it is implemented: each background thread needs its thread ID. After decreasing number of threads, all threads are woken up. The thread with the largest thread ID will terminate. If there are more threads to terminate, the thread will wake up all threads again.
Another change is made so that when number of threads is increased, more threads are created and all previous excessive threads are woken up to do the work.
Test Plan: Add a unit test.
Reviewers: haobo, dhruba
Reviewed By: haobo
CC: yhchiang, igor, nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D18675
Summary:
When TransactionLogIterator comes to EOF, it calls UnmarkEOF and continues reading. However, if glibc cached the EOF status of the file, it will get EOF again, even though the new data might have been written to it.
This has been causing errors in Mac OS.
Test Plan: test passes, was failing before
Reviewers: dhruba, haobo, sdong
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D18381
Summary: Calling Fsync()/Sync() on a file should give the guarantee that whatever you written to the file is now persisted. This is currently not the case, since we might have some data left in application cache as we do Fsync()/Sync(). For example, BuildTable() calls Fsync() without the flush, assuming all sst data is now persisted, but it's actually not. This may result in big inconsistencies.
Test Plan: no test
Reviewers: sdong, dhruba, haobo, ljin, yhchiang
Reviewed By: sdong
CC: leveldb
Differential Revision: https://reviews.facebook.net/D18159
Summary:
Fixed a compile error which tries to check whether a size_t < 0 in env_posix.cc
util/env_posix.cc:180:16: error: comparison of unsigned expression < 0 is always false [-Werror,-Wtautological-compare]
} while (r < 0 && errno == EINTR);
~ ^ ~
1 error generated.
Test Plan: make check all
Reviewers: igor, haobo
Reviewed By: igor
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17379
Summary: EINTR means 'please retry'. We don't do that currenty. We should.
Test Plan: make check, although it doesn't really test the new code. we'll just have to believe in the code!
Reviewers: haobo, ljin
Reviewed By: haobo
CC: leveldb
Differential Revision: https://reviews.facebook.net/D17349
Summary:
Based on my recent findings (posted in our internal group), if we use fallocate without KEEP_SIZE flag, we get superior performance of fdatasync() in append-only workloads.
This diff provides an option for user to not use KEEP_SIZE flag, thus optimizing his sync performance by up to 2x-3x.
At one point we also just called posix_fallocate instead of fallocate, which isn't very fast: http://code.woboq.org/userspace/glibc/sysdeps/posix/posix_fallocate.c.html (tl;dr it manually writes out zero bytes to allocate storage). This diff also fixes that, by first calling fallocate and then posix_fallocate if fallocate is not supported.
Test Plan: make check
Reviewers: dhruba, sdong, haobo, ljin
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16761
Summary: Add a function to Env so that users can query the waiting queue length of each thread pool
Test Plan: add a test in env_test
Reviewers: haobo
Reviewed By: haobo
CC: dhruba, igor, yhchiang, ljin, nkg-, leveldb
Differential Revision: https://reviews.facebook.net/D16755
Summary:
Change to store the return value from ftruncate().
The reason is that ftruncate() has "warn_unused_result" attribute in some environment.
Signed-off-by: Yumikiyo Osanai <yumios.art@gmail.com>
Summary:
Blocks allocated with fallocate will take extra space on disk even if they are unused and the file is close.
Now we remove the extra blocks at the end of the file by calling `ftruncate`.
Test Plan: added a test to env_test
Reviewers: dhruba
Reviewed By: dhruba
CC: leveldb
Differential Revision: https://reviews.facebook.net/D16647
Summary:
This is not a generic thread local implementation in the sense that it
only takes pointer. But it does support multiple instances per thread
and lets user plugin function to perform cleanup when thread exits or an
instance gets destroyed.
Test Plan: unit test for now
Reviewers: haobo, igor, sdong, dhruba
Reviewed By: igor
CC: leveldb, kailiu
Differential Revision: https://reviews.facebook.net/D16131
Yueh-Hsuan Chiang