Comments on: Avoiding statement-based replication warnings http://www.xaprb.com/blog/2012/08/23/avoiding-statement-based-replication-warnings/ Stay curious! Thu, 02 May 2013 12:36:53 +0000 hourly 1 http://wordpress.org/?v=3.5.1 By: Simon Mudd http://www.xaprb.com/blog/2012/08/23/avoiding-statement-based-replication-warnings/#comment-20188 Simon Mudd Fri, 24 Aug 2012 19:14:35 +0000 http://www.xaprb.com/blog/?p=1775#comment-20188 I’ve also reported this issue to Oracle and I think that some work has been done on this (in 5.6 iirc) but it’s still rather infuriating to see mysqld.log file files growing to tens of GB in size on some of the systems I manage. Completely unnecessary.

My real objection to some of these “warning messages” is that they are not actually correct (as you say) and for some of the cases where the limit statement makes the SQL deterministic then no warning should be displayed. That’s just a bug. If there’s a real issue: fine report it, and if necessary “throttle the warnings” if they are likely to be large in number.

]]> By: Xaprb http://www.xaprb.com/blog/2012/08/23/avoiding-statement-based-replication-warnings/#comment-20187 Xaprb Fri, 24 Aug 2012 15:26:23 +0000 http://www.xaprb.com/blog/?p=1775#comment-20187 Daniël, that requires a lot of care.

For anyone who’s interested in following this issue, there is currently a note in the MySQL manual about it: “Currently, when using STATEMENT mode, warnings are issued for DML statements containing LIMIT even when they also have an ORDER BY clause (and so are made deterministic). This is a known issue. (Bug #42851)”

]]> By: Daniël van Eeden http://www.xaprb.com/blog/2012/08/23/avoiding-statement-based-replication-warnings/#comment-20186 Daniël van Eeden Fri, 24 Aug 2012 13:31:55 +0000 http://www.xaprb.com/blog/?p=1775#comment-20186 What about binlog_format = MIXED?

]]>