Server silently stopped sending alerts, then died at restart attempt

[walterheck]

Hello,

we run zabbix server 1.8.2 and we have suffered a weird problem the past few days. Zabbix server silently stopped sending alerts around 36 hours before it completely failed.

has anyone else experienced this before? Any ideas of where to look for a solution?

The error log shows this every once in a while, but it doesn't seem to bother the server too much:

/var/log/zabbix-server/zabbix_server.log 14007:20110611:081014.549 [Z3005] Query failed: [2006] MySQL server has gone away [select hostid,status from hosts where host='yomamma' and status in (0,1) and proxy_hostid=0 and hostid between 000000000000000 and 099999999999999]
/var/log/zabbix-server/zabbix_server.log: 14007:20110611:081014.549 Database is down. Retrying in 10 seconds

any suggestions are more then welcome

Walter

[Alexei]

It is difficult to troubleshoot such cases without any log files. It looks very much that your database was down, at least Zabbix was not able to connect to it.

[walterheck]

the problem is that the above logs output is appearing both before and after the silent stopping of alerts, as well as after succesfully fixing the situation.
The above log stuff shouldn't interfere with the server's behaviour either, it should just retry the same query (more info on mysql errors and how to handle them here: http://openquery.com/blog/error-hand...l-applications)

For reference, I attached the terminal output from the two terminals used to fixed this. Any ideas are still more then welcome