2016-02-09 23:12:00 +00:00
|
|
|
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
2017-07-15 23:03:42 +00:00
|
|
|
// This source code is licensed under both the GPLv2 (found in the
|
|
|
|
// COPYING file in the root directory) and Apache 2.0 License
|
|
|
|
// (found in the LICENSE.Apache file in the root directory).
|
2014-04-15 20:39:26 +00:00
|
|
|
//
|
|
|
|
// Copyright (c) 2011 The LevelDB Authors. All rights reserved.
|
|
|
|
// Use of this source code is governed by a BSD-style license that can be
|
|
|
|
// found in the LICENSE file. See the AUTHORS file for names of contributors.
|
|
|
|
|
2015-10-13 00:35:43 +00:00
|
|
|
#ifndef NDEBUG
|
2014-04-15 20:39:26 +00:00
|
|
|
|
|
|
|
#include "db/db_impl.h"
|
2018-06-28 19:23:57 +00:00
|
|
|
#include "db/error_handler.h"
|
2017-04-06 02:02:00 +00:00
|
|
|
#include "monitoring/thread_status_updater.h"
|
2014-04-15 20:39:26 +00:00
|
|
|
|
2014-04-15 20:51:07 +00:00
|
|
|
namespace rocksdb {
|
|
|
|
|
2014-04-15 20:39:26 +00:00
|
|
|
uint64_t DBImpl::TEST_GetLevel0TotalSize() {
|
2015-02-05 05:39:45 +00:00
|
|
|
InstrumentedMutexLock l(&mutex_);
|
2014-10-31 15:48:19 +00:00
|
|
|
return default_cf_handle_->cfd()->current()->storage_info()->NumLevelBytes(0);
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
|
|
|
|
2017-09-28 00:37:08 +00:00
|
|
|
void DBImpl::TEST_SwitchWAL() {
|
2017-04-04 17:19:33 +00:00
|
|
|
WriteContext write_context;
|
2017-01-19 23:21:07 +00:00
|
|
|
InstrumentedMutexLock l(&mutex_);
|
2017-09-28 00:37:08 +00:00
|
|
|
SwitchWAL(&write_context);
|
2017-01-19 23:21:07 +00:00
|
|
|
}
|
|
|
|
|
2018-05-14 17:53:32 +00:00
|
|
|
bool DBImpl::TEST_WALBufferIsEmpty() {
|
|
|
|
InstrumentedMutexLock wl(&log_write_mutex_);
|
|
|
|
log::Writer* cur_log_writer = logs_.back().writer;
|
|
|
|
return cur_log_writer->TEST_BufferIsEmpty();
|
|
|
|
}
|
|
|
|
|
2014-04-15 20:39:26 +00:00
|
|
|
int64_t DBImpl::TEST_MaxNextLevelOverlappingBytes(
|
|
|
|
ColumnFamilyHandle* column_family) {
|
|
|
|
ColumnFamilyData* cfd;
|
|
|
|
if (column_family == nullptr) {
|
|
|
|
cfd = default_cf_handle_->cfd();
|
|
|
|
} else {
|
|
|
|
auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
|
|
|
|
cfd = cfh->cfd();
|
|
|
|
}
|
2015-02-05 05:39:45 +00:00
|
|
|
InstrumentedMutexLock l(&mutex_);
|
2014-10-31 15:48:19 +00:00
|
|
|
return cfd->current()->storage_info()->MaxNextLevelOverlappingBytes();
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
void DBImpl::TEST_GetFilesMetaData(
|
|
|
|
ColumnFamilyHandle* column_family,
|
|
|
|
std::vector<std::vector<FileMetaData>>* metadata) {
|
|
|
|
auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
|
|
|
|
auto cfd = cfh->cfd();
|
2015-02-05 05:39:45 +00:00
|
|
|
InstrumentedMutexLock l(&mutex_);
|
2014-04-15 20:39:26 +00:00
|
|
|
metadata->resize(NumberLevels());
|
|
|
|
for (int level = 0; level < NumberLevels(); level++) {
|
2014-10-27 22:49:46 +00:00
|
|
|
const std::vector<FileMetaData*>& files =
|
2014-10-31 15:48:19 +00:00
|
|
|
cfd->current()->storage_info()->LevelFiles(level);
|
2014-04-15 20:39:26 +00:00
|
|
|
|
|
|
|
(*metadata)[level].clear();
|
|
|
|
for (const auto& f : files) {
|
|
|
|
(*metadata)[level].push_back(*f);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
uint64_t DBImpl::TEST_Current_Manifest_FileNo() {
|
2014-11-04 01:45:55 +00:00
|
|
|
return versions_->manifest_file_number();
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
|
|
|
|
2017-10-10 20:07:00 +00:00
|
|
|
uint64_t DBImpl::TEST_Current_Next_FileNo() {
|
|
|
|
return versions_->current_next_file_number();
|
|
|
|
}
|
|
|
|
|
2014-04-15 20:39:26 +00:00
|
|
|
Status DBImpl::TEST_CompactRange(int level, const Slice* begin,
|
|
|
|
const Slice* end,
|
Allowing L0 -> L1 trivial move on sorted data
Summary:
This diff updates the logic of how we do trivial move, now trivial move can run on any number of files in input level as long as they are not overlapping
The conditions for trivial move have been updated
Introduced conditions:
- Trivial move cannot happen if we have a compaction filter (except if the compaction is not manual)
- Input level files cannot be overlapping
Removed conditions:
- Trivial move only run when the compaction is not manual
- Input level should can contain only 1 file
More context on what tests failed because of Trivial move
```
DBTest.CompactionsGenerateMultipleFiles
This test is expecting compaction on a file in L0 to generate multiple files in L1, this test will fail with trivial move because we end up with one file in L1
```
```
DBTest.NoSpaceCompactRange
This test expect compaction to fail when we force environment to report running out of space, of course this is not valid in trivial move situation
because trivial move does not need any extra space, and did not check for that
```
```
DBTest.DropWrites
Similar to DBTest.NoSpaceCompactRange
```
```
DBTest.DeleteObsoleteFilesPendingOutputs
This test expect that a file in L2 is deleted after it's moved to L3, this is not valid with trivial move because although the file was moved it is now used by L3
```
```
CuckooTableDBTest.CompactionIntoMultipleFiles
Same as DBTest.CompactionsGenerateMultipleFiles
```
This diff is based on a work by @sdong https://reviews.facebook.net/D34149
Test Plan: make -j64 check
Reviewers: rven, sdong, igor
Reviewed By: igor
Subscribers: yhchiang, ott, march, dhruba, sdong
Differential Revision: https://reviews.facebook.net/D34797
2015-06-04 23:51:25 +00:00
|
|
|
ColumnFamilyHandle* column_family,
|
|
|
|
bool disallow_trivial_move) {
|
2014-04-15 20:39:26 +00:00
|
|
|
ColumnFamilyData* cfd;
|
|
|
|
if (column_family == nullptr) {
|
|
|
|
cfd = default_cf_handle_->cfd();
|
|
|
|
} else {
|
|
|
|
auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
|
|
|
|
cfd = cfh->cfd();
|
|
|
|
}
|
|
|
|
int output_level =
|
2014-11-18 18:20:10 +00:00
|
|
|
(cfd->ioptions()->compaction_style == kCompactionStyleUniversal ||
|
|
|
|
cfd->ioptions()->compaction_style == kCompactionStyleFIFO)
|
2014-04-15 20:39:26 +00:00
|
|
|
? level
|
|
|
|
: level + 1;
|
2018-04-27 18:48:21 +00:00
|
|
|
return RunManualCompaction(cfd, level, output_level, 0, 0, begin, end, true,
|
Allowing L0 -> L1 trivial move on sorted data
Summary:
This diff updates the logic of how we do trivial move, now trivial move can run on any number of files in input level as long as they are not overlapping
The conditions for trivial move have been updated
Introduced conditions:
- Trivial move cannot happen if we have a compaction filter (except if the compaction is not manual)
- Input level files cannot be overlapping
Removed conditions:
- Trivial move only run when the compaction is not manual
- Input level should can contain only 1 file
More context on what tests failed because of Trivial move
```
DBTest.CompactionsGenerateMultipleFiles
This test is expecting compaction on a file in L0 to generate multiple files in L1, this test will fail with trivial move because we end up with one file in L1
```
```
DBTest.NoSpaceCompactRange
This test expect compaction to fail when we force environment to report running out of space, of course this is not valid in trivial move situation
because trivial move does not need any extra space, and did not check for that
```
```
DBTest.DropWrites
Similar to DBTest.NoSpaceCompactRange
```
```
DBTest.DeleteObsoleteFilesPendingOutputs
This test expect that a file in L2 is deleted after it's moved to L3, this is not valid with trivial move because although the file was moved it is now used by L3
```
```
CuckooTableDBTest.CompactionIntoMultipleFiles
Same as DBTest.CompactionsGenerateMultipleFiles
```
This diff is based on a work by @sdong https://reviews.facebook.net/D34149
Test Plan: make -j64 check
Reviewers: rven, sdong, igor
Reviewed By: igor
Subscribers: yhchiang, ott, march, dhruba, sdong
Differential Revision: https://reviews.facebook.net/D34797
2015-06-04 23:51:25 +00:00
|
|
|
disallow_trivial_move);
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
|
|
|
|
2017-10-03 16:08:07 +00:00
|
|
|
Status DBImpl::TEST_SwitchMemtable(ColumnFamilyData* cfd) {
|
|
|
|
WriteContext write_context;
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
if (cfd == nullptr) {
|
|
|
|
cfd = default_cf_handle_->cfd();
|
|
|
|
}
|
|
|
|
return SwitchMemtable(cfd, &write_context);
|
|
|
|
}
|
|
|
|
|
2018-08-29 18:58:13 +00:00
|
|
|
Status DBImpl::TEST_FlushMemTable(bool wait, bool allow_write_stall,
|
2019-03-27 23:13:08 +00:00
|
|
|
ColumnFamilyHandle* cfh) {
|
2014-04-15 20:39:26 +00:00
|
|
|
FlushOptions fo;
|
|
|
|
fo.wait = wait;
|
2018-08-29 18:58:13 +00:00
|
|
|
fo.allow_write_stall = allow_write_stall;
|
2016-04-18 18:11:51 +00:00
|
|
|
ColumnFamilyData* cfd;
|
|
|
|
if (cfh == nullptr) {
|
|
|
|
cfd = default_cf_handle_->cfd();
|
|
|
|
} else {
|
|
|
|
auto cfhi = reinterpret_cast<ColumnFamilyHandleImpl*>(cfh);
|
|
|
|
cfd = cfhi->cfd();
|
|
|
|
}
|
2018-02-09 20:09:55 +00:00
|
|
|
return FlushMemTable(cfd, fo, FlushReason::kTest);
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
Status DBImpl::TEST_WaitForFlushMemTable(ColumnFamilyHandle* column_family) {
|
|
|
|
ColumnFamilyData* cfd;
|
|
|
|
if (column_family == nullptr) {
|
|
|
|
cfd = default_cf_handle_->cfd();
|
|
|
|
} else {
|
|
|
|
auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
|
|
|
|
cfd = cfh->cfd();
|
|
|
|
}
|
2018-10-26 22:06:44 +00:00
|
|
|
return WaitForFlushMemTable(cfd, nullptr, false);
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
|
|
|
|
2018-03-07 00:13:05 +00:00
|
|
|
Status DBImpl::TEST_WaitForCompact(bool wait_unscheduled) {
|
2014-04-15 20:39:26 +00:00
|
|
|
// Wait until the compaction completes
|
|
|
|
|
|
|
|
// TODO: a bug here. This function actually does not necessarily
|
|
|
|
// wait for compact. It actually waits for scheduled compaction
|
|
|
|
// OR flush to finish.
|
|
|
|
|
2015-02-05 05:39:45 +00:00
|
|
|
InstrumentedMutexLock l(&mutex_);
|
2017-08-03 22:36:28 +00:00
|
|
|
while ((bg_bottom_compaction_scheduled_ || bg_compaction_scheduled_ ||
|
2018-03-07 00:13:05 +00:00
|
|
|
bg_flush_scheduled_ ||
|
|
|
|
(wait_unscheduled && unscheduled_compactions_)) &&
|
Auto recovery from out of space errors (#4164)
Summary:
This commit implements automatic recovery from a Status::NoSpace() error
during background operations such as write callback, flush and
compaction. The broad design is as follows -
1. Compaction errors are treated as soft errors and don't put the
database in read-only mode. A compaction is delayed until enough free
disk space is available to accomodate the compaction outputs, which is
estimated based on the input size. This means that users can continue to
write, and we rely on the WriteController to delay or stop writes if the
compaction debt becomes too high due to persistent low disk space
condition
2. Errors during write callback and flush are treated as hard errors,
i.e the database is put in read-only mode and goes back to read-write
only fater certain recovery actions are taken.
3. Both types of recovery rely on the SstFileManagerImpl to poll for
sufficient disk space. We assume that there is a 1-1 mapping between an
SFM and the underlying OS storage container. For cases where multiple
DBs are hosted on a single storage container, the user is expected to
allocate a single SFM instance and use the same one for all the DBs. If
no SFM is specified by the user, DBImpl::Open() will allocate one, but
this will be one per DB and each DB will recover independently. The
recovery implemented by SFM is as follows -
a) On the first occurance of an out of space error during compaction,
subsequent
compactions will be delayed until the disk free space check indicates
enough available space. The required space is computed as the sum of
input sizes.
b) The free space check requirement will be removed once the amount of
free space is greater than the size reserved by in progress
compactions when the first error occured
c) If the out of space error is a hard error, a background thread in
SFM will poll for sufficient headroom before triggering the recovery
of the database and putting it in write-only mode. The headroom is
calculated as the sum of the write_buffer_size of all the DB instances
associated with the SFM
4. EventListener callbacks will be called at the start and completion of
automatic recovery. Users can disable the auto recov ery in the start
callback, and later initiate it manually by calling DB::Resume()
Todo:
1. More extensive testing
2. Add disk full condition to db_stress (follow-on PR)
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4164
Differential Revision: D9846378
Pulled By: anand1976
fbshipit-source-id: 80ea875dbd7f00205e19c82215ff6e37da10da4a
2018-09-15 20:36:19 +00:00
|
|
|
(error_handler_.GetBGError() == Status::OK())) {
|
2014-04-15 20:39:26 +00:00
|
|
|
bg_cv_.Wait();
|
|
|
|
}
|
2018-06-28 19:23:57 +00:00
|
|
|
return error_handler_.GetBGError();
|
2014-04-15 20:39:26 +00:00
|
|
|
}
|
2014-04-29 17:27:58 +00:00
|
|
|
|
2019-03-27 23:13:08 +00:00
|
|
|
void DBImpl::TEST_LockMutex() { mutex_.Lock(); }
|
2014-09-05 22:20:05 +00:00
|
|
|
|
2019-03-27 23:13:08 +00:00
|
|
|
void DBImpl::TEST_UnlockMutex() { mutex_.Unlock(); }
|
2014-09-05 22:20:05 +00:00
|
|
|
|
|
|
|
void* DBImpl::TEST_BeginWrite() {
|
2015-08-05 23:56:28 +00:00
|
|
|
auto w = new WriteThread::Writer();
|
|
|
|
write_thread_.EnterUnbatched(w, &mutex_);
|
2014-09-05 22:20:05 +00:00
|
|
|
return reinterpret_cast<void*>(w);
|
|
|
|
}
|
|
|
|
|
|
|
|
void DBImpl::TEST_EndWrite(void* w) {
|
2014-09-12 23:23:58 +00:00
|
|
|
auto writer = reinterpret_cast<WriteThread::Writer*>(w);
|
2015-08-05 23:56:28 +00:00
|
|
|
write_thread_.ExitUnbatched(writer);
|
2014-09-08 15:01:25 +00:00
|
|
|
delete writer;
|
2014-09-05 22:20:05 +00:00
|
|
|
}
|
|
|
|
|
2015-03-30 19:04:10 +00:00
|
|
|
size_t DBImpl::TEST_LogsToFreeSize() {
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
return logs_to_free_.size();
|
|
|
|
}
|
|
|
|
|
2015-07-02 21:27:00 +00:00
|
|
|
uint64_t DBImpl::TEST_LogfileNumber() {
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
return logfile_number_;
|
|
|
|
}
|
|
|
|
|
2015-11-04 01:52:17 +00:00
|
|
|
Status DBImpl::TEST_GetAllImmutableCFOptions(
|
|
|
|
std::unordered_map<std::string, const ImmutableCFOptions*>* iopts_map) {
|
|
|
|
std::vector<std::string> cf_names;
|
|
|
|
std::vector<const ImmutableCFOptions*> iopts;
|
|
|
|
{
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
for (auto cfd : *versions_->GetColumnFamilySet()) {
|
|
|
|
cf_names.push_back(cfd->GetName());
|
|
|
|
iopts.push_back(cfd->ioptions());
|
|
|
|
}
|
|
|
|
}
|
|
|
|
iopts_map->clear();
|
|
|
|
for (size_t i = 0; i < cf_names.size(); ++i) {
|
|
|
|
iopts_map->insert({cf_names[i], iopts[i]});
|
|
|
|
}
|
|
|
|
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
2016-04-18 18:11:51 +00:00
|
|
|
uint64_t DBImpl::TEST_FindMinLogContainingOutstandingPrep() {
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-03 22:35:11 +00:00
|
|
|
return logs_with_prep_tracker_.FindMinLogContainingOutstandingPrep();
|
2016-04-18 18:11:51 +00:00
|
|
|
}
|
|
|
|
|
2018-03-02 04:33:41 +00:00
|
|
|
size_t DBImpl::TEST_PreparedSectionCompletedSize() {
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-03 22:35:11 +00:00
|
|
|
return logs_with_prep_tracker_.TEST_PreparedSectionCompletedSize();
|
2018-03-02 04:33:41 +00:00
|
|
|
}
|
|
|
|
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-03 22:35:11 +00:00
|
|
|
size_t DBImpl::TEST_LogsWithPrepSize() {
|
|
|
|
return logs_with_prep_tracker_.TEST_LogsWithPrepSize();
|
|
|
|
}
|
2018-03-02 04:33:41 +00:00
|
|
|
|
2016-04-18 18:11:51 +00:00
|
|
|
uint64_t DBImpl::TEST_FindMinPrepLogReferencedByMemTable() {
|
Skip deleted WALs during recovery
Summary:
This patch record min log number to keep to the manifest while flushing SST files to ignore them and any WAL older than them during recovery. This is to avoid scenarios when we have a gap between the WAL files are fed to the recovery procedure. The gap could happen by for example out-of-order WAL deletion. Such gap could cause problems in 2PC recovery where the prepared and commit entry are placed into two separate WAL and gap in the WALs could result into not processing the WAL with the commit entry and hence breaking the 2PC recovery logic.
Before the commit, for 2PC case, we determined which log number to keep in FindObsoleteFiles(). We looked at the earliest logs with outstanding prepare entries, or prepare entries whose respective commit or abort are in memtable. With the commit, the same calculation is done while we apply the SST flush. Just before installing the flush file, we precompute the earliest log file to keep after the flush finishes using the same logic (but skipping the memtables just flushed), record this information to the manifest entry for this new flushed SST file. This pre-computed value is also remembered in memory, and will later be used to determine whether a log file can be deleted. This value is unlikely to change until next flush because the commit entry will stay in memtable. (In WritePrepared, we could have removed the older log files as soon as all prepared entries are committed. It's not yet done anyway. Even if we do it, the only thing we loss with this new approach is earlier log deletion between two flushes, which does not guarantee to happen anyway because the obsolete file clean-up function is only executed after flush or compaction)
This min log number to keep is stored in the manifest using the safely-ignore customized field of AddFile entry, in order to guarantee that the DB generated using newer release can be opened by previous releases no older than 4.2.
Closes https://github.com/facebook/rocksdb/pull/3765
Differential Revision: D7747618
Pulled By: siying
fbshipit-source-id: d00c92105b4f83852e9754a1b70d6b64cb590729
2018-05-03 22:35:11 +00:00
|
|
|
autovector<MemTable*> empty_list;
|
|
|
|
return FindMinPrepLogReferencedByMemTable(versions_.get(), nullptr,
|
|
|
|
empty_list);
|
2016-04-18 18:11:51 +00:00
|
|
|
}
|
2016-05-17 20:11:56 +00:00
|
|
|
|
|
|
|
Status DBImpl::TEST_GetLatestMutableCFOptions(
|
|
|
|
ColumnFamilyHandle* column_family, MutableCFOptions* mutable_cf_options) {
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
|
|
|
|
auto cfh = reinterpret_cast<ColumnFamilyHandleImpl*>(column_family);
|
|
|
|
*mutable_cf_options = *cfh->cfd()->GetLatestMutableCFOptions();
|
|
|
|
return Status::OK();
|
|
|
|
}
|
|
|
|
|
2016-10-14 19:25:39 +00:00
|
|
|
int DBImpl::TEST_BGCompactionsAllowed() const {
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
2017-05-24 18:25:38 +00:00
|
|
|
return GetBGJobLimits().max_compactions;
|
|
|
|
}
|
|
|
|
|
|
|
|
int DBImpl::TEST_BGFlushesAllowed() const {
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
return GetBGJobLimits().max_flushes;
|
2016-10-14 19:25:39 +00:00
|
|
|
}
|
|
|
|
|
2017-11-11 01:18:01 +00:00
|
|
|
SequenceNumber DBImpl::TEST_GetLastVisibleSequence() const {
|
2017-12-01 07:39:56 +00:00
|
|
|
if (last_seq_same_as_publish_seq_) {
|
2017-10-23 21:20:22 +00:00
|
|
|
return versions_->LastSequence();
|
2017-11-11 01:18:01 +00:00
|
|
|
} else {
|
|
|
|
return versions_->LastAllocatedSequence();
|
2017-10-23 21:20:22 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2018-09-26 17:34:56 +00:00
|
|
|
size_t DBImpl::TEST_GetWalPreallocateBlockSize(
|
|
|
|
uint64_t write_buffer_size) const {
|
|
|
|
InstrumentedMutexLock l(&mutex_);
|
|
|
|
return GetWalPreallocateBlockSize(write_buffer_size);
|
|
|
|
}
|
|
|
|
|
2019-02-20 23:46:59 +00:00
|
|
|
void DBImpl::TEST_WaitForDumpStatsRun(std::function<void()> callback) const {
|
move dump stats to a separate thread (#4382)
Summary:
Currently statistics are supposed to be dumped to info log at intervals of `options.stats_dump_period_sec`. However the implementation choice was to bind it with compaction thread, meaning if the database has been serving very light traffic, the stats may not get dumped at all.
We decided to separate stats dumping into a new timed thread using `TimerQueue`, which is already used in blob_db. This will allow us schedule new timed tasks with more deterministic behavior.
Tested with db_bench using `--stats_dump_period_sec=20` in command line:
> LOG:2018/09/17-14:07:45.575025 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG:2018/09/17-14:08:05.643286 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG:2018/09/17-14:08:25.691325 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG:2018/09/17-14:08:45.740989 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
LOG content:
> 2018/09/17-14:07:45.575025 7fe99fbfe700 [WARN] [db/db_impl.cc:605] ------- DUMPING STATS -------
2018/09/17-14:07:45.575080 7fe99fbfe700 [WARN] [db/db_impl.cc:606]
** DB Stats **
Uptime(secs): 20.0 total, 20.0 interval
Cumulative writes: 4447K writes, 4447K keys, 4447K commit groups, 1.0 writes per commit group, ingest: 5.57 GB, 285.01 MB/s
Cumulative WAL: 4447K writes, 0 syncs, 4447638.00 writes per sync, written: 5.57 GB, 285.01 MB/s
Cumulative stall: 00:00:0.012 H:M:S, 0.1 percent
Interval writes: 4447K writes, 4447K keys, 4447K commit groups, 1.0 writes per commit group, ingest: 5700.71 MB, 285.01 MB/s
Interval WAL: 4447K writes, 0 syncs, 4447638.00 writes per sync, written: 5.57 MB, 285.01 MB/s
Interval stall: 00:00:0.012 H:M:S, 0.1 percent
** Compaction Stats [default] **
Level Files Size Score Read(GB) Rn(GB) Rnp1(GB) Write(GB) Wnew(GB) Moved(GB) W-Amp Rd(MB/s) Wr(MB/s) Comp(sec) Comp(cnt) Avg(sec) KeyIn KeyDrop
Pull Request resolved: https://github.com/facebook/rocksdb/pull/4382
Differential Revision: D9933051
Pulled By: miasantreble
fbshipit-source-id: 6d12bb1e4977674eea4bf2d2ac6d486b814bb2fa
2018-10-09 05:52:58 +00:00
|
|
|
if (thread_dump_stats_ != nullptr) {
|
|
|
|
thread_dump_stats_->TEST_WaitForRun(callback);
|
|
|
|
}
|
|
|
|
}
|
2019-02-20 23:46:59 +00:00
|
|
|
|
|
|
|
void DBImpl::TEST_WaitForPersistStatsRun(std::function<void()> callback) const {
|
|
|
|
if (thread_persist_stats_ != nullptr) {
|
|
|
|
thread_persist_stats_->TEST_WaitForRun(callback);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
bool DBImpl::TEST_IsPersistentStatsEnabled() const {
|
|
|
|
return thread_persist_stats_ && thread_persist_stats_->IsRunning();
|
|
|
|
}
|
|
|
|
|
|
|
|
size_t DBImpl::TEST_EstiamteStatsHistorySize() const {
|
|
|
|
return EstiamteStatsHistorySize();
|
|
|
|
}
|
2014-04-15 20:51:07 +00:00
|
|
|
} // namespace rocksdb
|
2015-10-13 00:35:43 +00:00
|
|
|
#endif // NDEBUG
|