database/mariadb
1
0
Fork 0
mirror of https://github.com/MariaDB/server.git synced 2025-08-08 11:22:35 +03:00
Code Activity

MDEV-6718: Server crashed in Gtid_log_event::Gtid_log_event with parallel replication

Browse Source 
The bug occured in parallel replication when re-trying transactions that
failed due to deadlock. In this case, the relay log file is re-opened and the
events are read out again. This reading requires a format description event of
the appropriate version. But the code was using a description event stored in
rli, which is not thread-safe. This could lead to various rare races if the
format description event was replaced by the SQL driver thread at the exact
moment where a worker thread was trying to use it.

The fix is to instead make the retry code create and maintain its own format
description event. When the relay log file is opened, we first read the format
description event from the start of the file, before seeking to the current
position. This now uses the same code as when the SQL driver threads starts
from a given relay log position. This also makes sure that the correct format
description event version will be used in cases where the version of the
binlog could change during replication.
This commit is contained in:
Kristian Nielsen
2014-11-13 10:09:46 +01:00
parent a98a034c5e
commit 8a3e2f29bb
3 changed files with 112 additions and 64 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
sql/rpl_parallel.cc
View File

@@ -290,6 +290,7 @@ retry_event_group(rpl_group_info *rgi, rpl_parallel_thread *rpt,
THD *thd= rgi->thd;
rpl_parallel_entry *entry= rgi->parallel_entry;
ulong retries= 0;
Format_description_log_event *description_event= NULL;
do_retry:
event_count= 0;
@@ -355,6 +356,14 @@ do_retry:
goto err;
}
cur_offset= rgi->retry_start_offset;
delete description_event;
description_event=
read_relay_log_description_event(&rlog, cur_offset, &errmsg);
if (!description_event)
{
err= 1;
goto err;
}
my_b_seek(&rlog, cur_offset);
do
@@ -367,8 +376,7 @@ do_retry:
for (;;)
{
old_offset= cur_offset;
ev= Log_event::read_log_event(&rlog, 0,
rli->relay_log.description_event_for_exec /* ToDo: this needs fixing */,
ev= Log_event::read_log_event(&rlog, 0, description_event,
opt_slave_sql_verify_checksum);
cur_offset= my_b_tell(&rlog);
@@ -416,7 +424,12 @@ do_retry:
}
event_type= ev->get_type_code();
if (!Log_event::is_group_event(event_type))
if (event_type == FORMAT_DESCRIPTION_EVENT)
{
delete description_event;
description_event= (Format_description_log_event *)ev;
continue;
} else if (!Log_event::is_group_event(event_type))
{
delete ev;
continue;
@@ -472,6 +485,8 @@ do_retry:
err:
if (description_event)
delete description_event;
if (fd >= 0)
{
end_io_cache(&rlog);