database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-07-30 16:24:05 +03:00
Code Activity

MDEV-35465 Async replication stops working on Galera async replica node when parallel replication is enabled

Browse Source 
Parallel slave failed to retry in retry_event_group() with error

    WSREP: Parallel slave worker failed at wsrep_before_command() hook

Fix wsrep transaction cleanup/restart in retry_event_group() to properly
clean up previous transaction by calling wsrep_after_statement().
Also move call to reset error after call to wsrep_after_statement()
to make sure that it remains effective.

Add a MTR test galera_as_slave_parallel_retry to reproduce the error
when the fix is not present.

Other issues which were detected when testing with sysbench:

Check if parallel slave is killed for retry before waiting for prior
commits in THD::wsrep_parallel_slave_wait_for_prior_commit(). This
is required with slave-parallel-mode=optimistic to avoid deadlock
when a slave later in commit order manages to reach prepare phase
before a lock conflict is detected.

Suppress wsrep applier specific warning for slave threads.

Signed-off-by: Julius Goryavsky <julius.goryavsky@mariadb.com>
This commit is contained in:
Teemu Ollakka
2024-11-21 16:54:51 +02:00
committed by Julius Goryavsky
parent c772344510
commit a2575a0703
7 changed files with 113 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

29
sql/rpl_parallel.cc
View File

@ -134,7 +134,7 @@ handle_queued_pos_update(THD *thd, rpl_parallel_thread::queued_event *qev)
asynchronously, we need to be sure they will be completed before starting a
new transaction. Otherwise the new transaction might suffer a spurious kill.
*/
static void
void
wait_for_pending_deadlock_kill(THD *thd, rpl_group_info *rgi)
{
PSI_stage_info old_stage;
@ -831,8 +831,12 @@ do_retry:
err= 0;
errmsg= NULL;
#ifdef WITH_WSREP
thd->wsrep_cs().reset_error();
WSREP_DEBUG("retrying async replication event");
DBUG_EXECUTE_IF("sync.wsrep_retry_event_group", {
const char act[]= "now "
"SIGNAL sync.wsrep_retry_event_group_reached "
"WAIT_FOR signal.wsrep_retry_event_group";
debug_sync_set_action(thd, STRING_WITH_LEN(act));
};);
#endif /* WITH_WSREP */
/*
@ -981,15 +985,20 @@ do_retry:
*/
thd->reset_killed();
#ifdef WITH_WSREP
if (wsrep_before_command(thd))
if (WSREP(thd))
{
WSREP_WARN("Parallel slave worker failed at wsrep_before_command() hook");
err= 1;
goto err;
/* Exec after statement hook to make sure that the failed transaction
* gets cleared and reset error state. */
if (wsrep_after_statement(thd))
{
WSREP_WARN("Parallel slave worker failed at wsrep_after_statement() hook");
err= 1;
goto err;
}
thd->wsrep_cs().reset_error();
wsrep_start_trx_if_not_started(thd);
WSREP_DEBUG("parallel slave retry, after trx start");
}
wsrep_start_trx_if_not_started(thd);
WSREP_DEBUG("parallel slave retry, after trx start");
#endif /* WITH_WSREP */
strmake_buf(log_name, ir->name);
if ((fd= open_binlog(&rlog, log_name, &errmsg)) <0)