Comments on: Debugging metadata locking in MySQL 5.5 http://www.xaprb.com/blog/2012/08/28/debugging-metadata-locking-in-mysql-5-5/ Stay curious! Fri, 10 May 2013 18:25:19 +0000 hourly 1 http://wordpress.org/?v=3.5.1 By: Sheeri K. Cabral http://www.xaprb.com/blog/2012/08/28/debugging-metadata-locking-in-mysql-5-5/#comment-20208 Sheeri K. Cabral Thu, 06 Sep 2012 20:11:07 +0000 http://www.xaprb.com/blog/?p=1582#comment-20208 Would it be possible for pt-deadlock-logger to store this information too? (or is that where it gets it?) I’ve enjoyed using it but it only explains one of the 2 locks that cause the deadlock :(

]]>
By: Daniël van Eeden http://www.xaprb.com/blog/2012/08/28/debugging-metadata-locking-in-mysql-5-5/#comment-20202 Daniël van Eeden Mon, 03 Sep 2012 05:34:12 +0000 http://www.xaprb.com/blog/?p=1582#comment-20202 Maybe watching open files with lsof could give some hints about what’s going on?
It could also be hardware/OS related. (full filesystem, out of memory, SSD firmware bug, etc)

]]>
By: Normann http://www.xaprb.com/blog/2012/08/28/debugging-metadata-locking-in-mysql-5-5/#comment-20191 Normann Tue, 28 Aug 2012 18:43:47 +0000 http://www.xaprb.com/blog/?p=1582#comment-20191 if i kill a huge insert qry, I get a “Waiting for table metadata lock” too. but the connection status is just “killed”

]]>
By: Xaprb http://www.xaprb.com/blog/2012/08/28/debugging-metadata-locking-in-mysql-5-5/#comment-20190 Xaprb Tue, 28 Aug 2012 18:02:09 +0000 http://www.xaprb.com/blog/?p=1582#comment-20190 I have a feeling something like this is the problem in my case too.

]]>
By: sbester http://www.xaprb.com/blog/2012/08/28/debugging-metadata-locking-in-mysql-5-5/#comment-20189 sbester Tue, 28 Aug 2012 17:34:49 +0000 http://www.xaprb.com/blog/?p=1582#comment-20189 one that is particularly tricky to diagnose when innodb table is not involved. e.g.

conn1: begin;
conn1: select * from myisam_table;
conn2: drop table myisam_table;

conn2 will simply hang until lock_wait_timeout expires and there’s no easy way to know that conn1 is responsible.

]]>