Fix flush not being commit while writing manifest
Summary:
Fix flush not being commit while writing manifest, which is a recent bug introduced by D60075.
The issue:
# Options.max_background_flushes > 1
# Background thread A pick up a flush job, flush, then commit to manifest. (Note that mutex is released before writing manifest.)
# Background thread B pick up another flush job, flush. When it gets to `MemTableList::InstallMemtableFlushResults`, it notices another thread is commiting, so it quit.
# After the first commit, thread A doesn't double check if there are more flush result need to commit, leaving the second flush uncommitted.
Test Plan: run the test. Also verify the new test hit deadlock without the fix.
Reviewers: sdong, igor, lightmark
Reviewed By: lightmark
Subscribers: andrewkr, omegaga, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D60969
2016-07-21 19:10:41 +02:00
|
|
|
// Copyright (c) 2011-present, Facebook, Inc. All rights reserved.
|
|
|
|
// This source code is licensed under the BSD-style license found in the
|
|
|
|
// LICENSE file in the root directory of this source tree. An additional grant
|
|
|
|
// of patent rights can be found in the PATENTS file in the same directory.
|
|
|
|
//
|
|
|
|
// 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.
|
|
|
|
|
|
|
|
#include "db/db_test_util.h"
|
|
|
|
#include "port/stack_trace.h"
|
|
|
|
#include "util/sync_point.h"
|
|
|
|
|
|
|
|
namespace rocksdb {
|
|
|
|
|
|
|
|
class DBFlushTest : public DBTestBase {
|
|
|
|
public:
|
|
|
|
DBFlushTest() : DBTestBase("/db_flush_test") {}
|
|
|
|
};
|
|
|
|
|
|
|
|
// We had issue when two background threads trying to flush at the same time,
|
|
|
|
// only one of them get committed. The test verifies the issue is fixed.
|
|
|
|
TEST_F(DBFlushTest, FlushWhileWritingManifest) {
|
|
|
|
Options options;
|
|
|
|
options.disable_auto_compactions = true;
|
|
|
|
options.max_background_flushes = 2;
|
|
|
|
Reopen(options);
|
|
|
|
FlushOptions no_wait;
|
|
|
|
no_wait.wait = false;
|
|
|
|
|
|
|
|
SyncPoint::GetInstance()->LoadDependency(
|
|
|
|
{{"VersionSet::LogAndApply:WriteManifest",
|
|
|
|
"DBFlushTest::FlushWhileWritingManifest:1"},
|
|
|
|
{"MemTableList::InstallMemtableFlushResults:InProgress",
|
|
|
|
"VersionSet::LogAndApply:WriteManifestDone"}});
|
|
|
|
SyncPoint::GetInstance()->EnableProcessing();
|
|
|
|
|
|
|
|
ASSERT_OK(Put("foo", "v"));
|
|
|
|
ASSERT_OK(dbfull()->Flush(no_wait));
|
|
|
|
TEST_SYNC_POINT("DBFlushTest::FlushWhileWritingManifest:1");
|
|
|
|
ASSERT_OK(Put("bar", "v"));
|
|
|
|
ASSERT_OK(dbfull()->Flush(no_wait));
|
|
|
|
// If the issue is hit we will wait here forever.
|
|
|
|
dbfull()->TEST_WaitForFlushMemTable();
|
2016-07-21 23:32:12 +02:00
|
|
|
#ifndef ROCKSDB_LITE
|
Fix flush not being commit while writing manifest
Summary:
Fix flush not being commit while writing manifest, which is a recent bug introduced by D60075.
The issue:
# Options.max_background_flushes > 1
# Background thread A pick up a flush job, flush, then commit to manifest. (Note that mutex is released before writing manifest.)
# Background thread B pick up another flush job, flush. When it gets to `MemTableList::InstallMemtableFlushResults`, it notices another thread is commiting, so it quit.
# After the first commit, thread A doesn't double check if there are more flush result need to commit, leaving the second flush uncommitted.
Test Plan: run the test. Also verify the new test hit deadlock without the fix.
Reviewers: sdong, igor, lightmark
Reviewed By: lightmark
Subscribers: andrewkr, omegaga, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D60969
2016-07-21 19:10:41 +02:00
|
|
|
ASSERT_EQ(2, TotalTableFiles());
|
2016-07-21 23:32:12 +02:00
|
|
|
#endif // ROCKSDB_LITE
|
Fix flush not being commit while writing manifest
Summary:
Fix flush not being commit while writing manifest, which is a recent bug introduced by D60075.
The issue:
# Options.max_background_flushes > 1
# Background thread A pick up a flush job, flush, then commit to manifest. (Note that mutex is released before writing manifest.)
# Background thread B pick up another flush job, flush. When it gets to `MemTableList::InstallMemtableFlushResults`, it notices another thread is commiting, so it quit.
# After the first commit, thread A doesn't double check if there are more flush result need to commit, leaving the second flush uncommitted.
Test Plan: run the test. Also verify the new test hit deadlock without the fix.
Reviewers: sdong, igor, lightmark
Reviewed By: lightmark
Subscribers: andrewkr, omegaga, dhruba, leveldb
Differential Revision: https://reviews.facebook.net/D60969
2016-07-21 19:10:41 +02:00
|
|
|
}
|
|
|
|
|
|
|
|
} // namespace rocksdb
|
|
|
|
|
|
|
|
int main(int argc, char** argv) {
|
|
|
|
rocksdb::port::InstallStackTraceHandler();
|
|
|
|
::testing::InitGoogleTest(&argc, argv);
|
|
|
|
return RUN_ALL_TESTS();
|
|
|
|
}
|