[rabbitmq-discuss] Odd behavior where server stops responding

[Jason McIntosh]

SO we finally figured out the issue.  Apparently there's a rather nasty bug
in the 2.6.32 kernels:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/805341
http://www.novell.com/support/kb/doc.php?id=7009834
https://www.ibm.com/developerworks/community/blogs/anthonyv/entry/208_day_reboot_bug3?lang=en

These are just a few of the sites listing the problem and upgrade notes.
 The behavior on this for us was really kinda random.  We've had some
systems just completely freeze, the OS being totally hung.  Other systems
seem to manage to make it past the limit by a day or two then die.  In the
case of the servers I mentioned, only the Rabbit process itself hung.  But
the one consistent thing we discovered was that all the servers having an
issue had an uptime right around 208 days - we missed this on our initial
investigation.  SO If anyone is running CentOS 6.2, upgrade!

Jason


-- 
Jason McIntosh
https://github.com/jasonmcintosh/
573-424-7612
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20140317/d16adf2b/attachment.html>