Cross-posting from the General forum as suggested:
http://forums.mysql.com/read.php?20,625574,625574#msg-625574
I've got mysql 5.6.22 built from source on CentOS 6.3 x86_64. Using the standard (/etc/rc.d/init.d/mysql) script, I'm noticing that the server won't shut down cleanly. Both the mysqld_safe and mysqld processes hang around.. there is no disk activity (from what I can tell) and no activity from mysqld on `top`. The log shows:
2014-12-22 19:27:50 28608 [Note] /usr/libexec/mysqld: Normal shutdown
2014-12-22 19:27:50 28608 [Note] Giving 5 client threads a chance to die gracefully
2014-12-22 19:27:50 28608 [Note] Event Scheduler: Purging the queue. 0 events
2014-12-22 19:27:50 28608 [Note] Shutting down slave threads
2014-12-22 19:27:52 28608 [Note] Forcefully disconnecting 5 remaining clients
After which nothing else happens until I kill -9 the process(es).
When I restart the server, I get:
2014-12-22 19:29:05 29395 [Note] InnoDB: Database was not shutdown normally!
2014-12-22 19:29:05 29395 [Note] InnoDB: Starting crash recovery.
Is there any way to debug what's going on when mysqld is seemingly hung? I can't connect to it via a client..
This is a production VM, so there's only so much playing around I can do.
Thank you.