PS-3345: LP #1527463: Waiting for binlog lock (5.7) #3426
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
percona-ysorokin
requested changes
Sep 13, 2019
| # When it finds the deadlock, it throws assert. | ||
| ################################################################################ | ||
|
|
||
| --source include/have_debug.inc |
There was a problem hiding this comment.
Also
--source include/have_debug_sync.inc
--source include/have_innodb.inc
| --source include/rpl_connection_slave.inc | ||
| --source include/only_mts_slave_parallel_workers.inc | ||
| --source include/only_mts_slave_parallel_type_logical_clock.inc | ||
| --source include/stop_slave_sql.inc |
There was a problem hiding this comment.
Consider --let $rpl_skip_start_slave= 1 before including master-slave.inc instead.
There was a problem hiding this comment.
This test needs stop_slave_sql.inc as a sync point.
| --echo # | ||
| --source include/rpl_connection_master.inc | ||
| CREATE TABLE t1(c1 INT PRIMARY KEY, c2 INT, INDEX(c2)) ENGINE = InnoDB; | ||
| SET debug = "+d,set_commit_parent_100"; |
There was a problem hiding this comment.
Why do we need this?
There was a problem hiding this comment.
Without this 2 INSERTs are processed with a single worker thread.
gl-sergei
suggested changes
Sep 15, 2019
sql/binlog.cc
Outdated
| { | ||
| Slave_worker *worker= dynamic_cast<Slave_worker *>(thd->rli_slave); | ||
|
|
||
| static bool skip_first_query= true; |
There was a problem hiding this comment.
I am not sue if it really works when test is run second time without server restart.
There was a problem hiding this comment.
I improved this a little bit and verified that it works fine with --repeat with and without deadlock.
gl-sergei
approved these changes
Sep 17, 2019
Fix 3-way deadlock that can be achieved with 2 slave threads working and parallel and with 1 slave client that executes LOCK BINLOG FOR BACKUP. And the deadlock is: worker0: applying INSERT INTO t1 VALUES(11, NULL); worker1: applying INSERT INTO t1 VALUES(12, NULL); worker1: calls backup_binlog_lock.acquire_protection() worker1: waits for worker0 in wait_for_its_turn() client: executes LOCK BINLOG FOR BACKUP client: waits in backup_binlog_lock.acquire(), but protection is acquired by worker1 worker0: calls backup_binlog_lock.acquire_protection(), but it's blocked by client
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix 3-way deadlock that can be achieved with 2 slave threads working and parallel and with 1 slave client that executes LOCK BINLOG FOR BACKUP.
And the deadlock is:
worker0: applying INSERT INTO t1 VALUES(11, NULL);
worker1: applying INSERT INTO t1 VALUES(12, NULL);
worker1: calls backup_binlog_lock.acquire_protection()
worker1: waits for worker0 in wait_for_its_turn()
client: executes LOCK BINLOG FOR BACKUP
client: waits in backup_binlog_lock.acquire(), but protection is acquired by worker1
worker0: calls backup_binlog_lock.acquire_protection(), but it's blocked by client