Details
-
Bug
-
Status: Closed (View Workflow)
-
Critical
-
Resolution: Duplicate
-
10.1.12
-
None
-
CentOS 7
MariaDB 10.1.12
Parallel Replication with 16 threads
Description
I'm using a MariaDB with replication parallel.
Whenever I need to stop the MySQL service, it does not stop effectively.
It seems that a zombie is still up process, but I can not aCesar MySQL.
But if I stop replication before stopping the MySQL service, the MySQL service is stoped successfully.
How are you stopping the service?
Can you use gdb to grab a backtrace while it is in its zombie status?
gdb --batch --eval-command="thread apply all bt" program $(pgrep -xn mysqld)
What also are the last messages written to the error log?
If you do a STOP SLAVE SQL_THREAD; does this complete ok?
What are your replication settings? What version is the master? Are you replicating in gtid mode? Are you using any aggressive mode replication?