Recently we upgraded from 5.0.8NE to 5.0.15NE on SLES10.

We are now starting to get alert emails indicating the logger service has stopped.

Both the Admin Console and a grep of /var/log/zimbra.log for logger status shows the logger service is running just fine. There are no gaps in the Server Statistics graphs in the Admin Console either.

The alert emails seem to come at fairly random times, typically twice per day. We got one this morning around 8:30am and yesterday at 6:30pm. Specifically for example:
  1. 8:18:06 zimbra.log has Logger Status as RUNNING.
  2. 8:18:43 admin alert email is generated and successfully sent.
  3. 8:20:10 zimbra.log has Logger Status as RUNNING.

Plus, we never get an admin alert email that the logger service is once again running.

Initially we were thinking this could be a (benign?) log rotation error, but the randomness of the alerts made us think otherwise.

We've gone through the wiki document to troubleshoot logger issues and all seems well. We've searched the forums for someone with a similar problem here, to no avail. We've grepped a number of the log files in ~/log as well as /var/log/messages, /var/log/warn, etc. looking for clues--to no avail.

If anyone has seen this or has any ideas where we should be looking, that would be great, thank you!

All the best,