Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
10.2(EOL), 10.3(EOL), 10.4(EOL), 10.5
Description
A race described in MDEV-24526 is not fixed over there for a case of the server started
with a non-default --binlog-optimize-thread-scheduling=0.
Specifically a group of committing transaction threads need to lock a sequence
of mutexes LOCK_log, LOCK_after_binlog_sync, LOCK_commit_ordered in a crabbing manner, that is with releasing the preceding mutex upon acquisition of the current one.
MYSQL_BIN_LOG::trx_group_commit_leader() shows the pattern for the default
--binlog-optimize-thread-scheduling=1.
Attachments
Issue Links
- relates to
-
MDEV-24526 binlog rotate via FLUSH LOGS may obsolate binlog file for recovery too early
- Closed