Bug #66485 Replication of Events (Event Scheduler)
Submitted: 21 Aug 2012 17:14 Modified: 3 Oct 2012 17:19
Reporter: Adam Newman Email Updates:
Status: No Feedback Impact on me:
None 
Category:MySQL Server: Replication Severity:S3 (Non-critical)
Version:5.5 OS:Windows
Assigned to: CPU Architecture:Any
Tags: Drop, event scheduler, replication

[21 Aug 2012 17:14] Adam Newman
Description:
Hello!

Please see forum post for problem description:

http://forums.mysql.com/read.php?119,565767,565767#msg-565767

I've tried all kinds of ways to work around the problem but nothing i've tried seems to help. I've tried 

- Revoking privaliges for EVENTS from the slave, master and root user (root is the user creating the event on the master). Replication slave doesn't appear to care about privaleges when reading the binlog
- Adding: "replicate-ignore-table=cmdtedb.events" to my.ini on both master and slave side. This settings gets ignored and the events statements are still replicated.

How to repeat:
Establish replication and execute a long-running event on master. Stop/start sql service (which will purge events) on slave. When even completes on master, slave tries to execute "DROP EVENT" but the event is no longer present (events table is empty), so error will occur in replication on the slave.

Suggested fix:
- Change "DROP EVENT" statement to "DROP EVENT IF EXISTS" on event scheduler purge so that replication remains stable.
- Implement means to filter EVENTS through replication.
[23 Aug 2012 18:32] Sveta Smirnova
Thank you for the report.

Please send us full slave error log file.
[30 Aug 2012 10:01] Adam Newman
Sorry, I am on leave until Monday 3rd August so I can't get into our system until then. Is it the binlog that you need?

Have you been able to reproduce this yet?
[3 Sep 2012 17:19] Sveta Smirnova
Thank you for the feedback.

Not, I need slave error log file. You can find its location if run query SELECT @@log_error; on slave.
[4 Oct 2012 1:00] Bugs System
No feedback was provided for this bug for over a month, so it is
being suspended automatically. If you are able to provide the
information that was originally requested, please do so and change
the status of the bug back to "Open".