Zabbix upgrade to 2.0.6 causes agent to stop sending updates

[nilie]

Hello everybody,

I am currently evaluating Zabbix in a lab environment and the experience I had so far is overwhelmingly positive. The documentation is excellent and the installation on CentOS v6 using Zabbix repositories was flawless for me. I started with version 2.0.5 and when 2.0.6 came out, I did an upgrade from the same repository, which also went just fine. However, following the upgrade, the Zabbix agent installed on the Zabbix server itself stopped sending updates. In Administration -> Queue screen, the Zabbix agent items were red and everything else was green, including Zabbix agent active. All other checks were working and updates were coming in (hardly surprising since they were coming from the localhost). I looked in the documentation and I've searched through the menus for a way of forcing some sort of update but without any luck. The last action I took before things came back to normal was to enable and disable the discovery but I don't know this is the trick that solved my problem since I haven't configured the discovery at all.
So here are my questions :
1 - Has anyone else experienced this kind of behavior ?
2 - What would be the best way to force the Zenoss server to contact the agent and ask for updates

[Heilig]

1 - Has anyone else experienced this kind of behavior ?

In such situations, helps to check the log files for error.

2 - What would be the best way to force the Zenoss server to contact the agent and ask for updates

This is not possible.

[nilie]

In such situations, helps to check the log files for error.

This is not possible.

Thank you for your reply.

I did check the log files at the time but for some unknown reason the tail command displayed all kind of ASCII characters and I (mistakenly) thought the log file must have some sort of binary format instead of plain text. Now after reading your reply I did check again and indeed they are in plain text format. Here's the relevant part :

Server log
14715:20130430:155730.432 Zabbix agent item [system.swap.size[,free]] on host [Zabbix server] failed: first network error, wait for 15 seconds
14718:20130430:155730.432 Zabbix agent item [system.swap.size[,pfree]] on host [Zabbix server] failed: first network error, wait for 15 seconds
14720:20130430:155745.737 resuming Zabbix agent checks on host [Zabbix server]: connection restored

Agent log
14704:20130430:152832.635 active check configuration update from [127.0.0.1:10051] started to fail (cannot connect to [[127.0.0.1]:10051]: [111] Connection refused)
14704:20130430:152932.645 active check configuration update from [127.0.0.1:10051] is working again

Other than this there are only shutdown/startup and housekeeper related messages. Although the log files don't show that, it took about 29 minutes until the updates were back in.

I'm runnig Zabbix on a VMware virtual machine with CentOS 6.4 64-bit.

Any idea on why the connection to 127.0.0.1 was being refused after upgrading the server and the agent ?