Bug #61562 | Replication slows down after network glitch, needs stop/start slave to recover | ||
---|---|---|---|
Submitted: | 19 Jun 2011 8:46 | Modified: | 4 Jul 2011 8:38 |
Reporter: | Kees Jan Koster | Email Updates: | |
Status: | Closed | Impact on me: | |
Category: | MySQL Server: Replication | Severity: | S3 (Non-critical) |
Version: | 5.5.8 | OS: | FreeBSD |
Assigned to: | CPU Architecture: | Any |
[19 Jun 2011 8:46]
Kees Jan Koster
[19 Jun 2011 8:47]
Kees Jan Koster
Munin graph for mysql qps on slave
Attachment: mysql_queries-day.png (image/png, text), 19.27 KiB.
[19 Jun 2011 8:48]
Kees Jan Koster
cron job and cron job output
Attachment: mysql-slave-restart.txt (text/plain), 10.28 KiB.
[19 Jun 2011 8:50]
Kees Jan Koster
The graph and the cron job are in files, attached to this issue.
[28 Jun 2011 13:39]
Kees Jan Koster
Interesting. I updated the slave database to MySQL 5.5.13 and it has been running without these strange slowdowns for an entire week (previously unheard of). I have reasonably good hopes that this problem has been resolved with the version upgrade. I'll keep you posted.
[29 Jun 2011 17:48]
Valeriy Kravchuk
Thank you for status update. Please, inform about any results of your further testing with 5.5.13.
[4 Jul 2011 8:38]
Kees Jan Koster
I am closing this bug report. Upgrading the slave to version 5.5.13 solved the problem for me. Thanks.