Zimbra and SNMP--monitoring port status?

[nsmarler]

Is there a way to monitor ports on our ZCS box (5.01, running on Ubuntu 6.06.2 Dapper Drake) so that we can tell if it stops responding on, say, port 465 or other common ports? Maybe through a configuration of the SNMP service? The documentation I found wasn't very clear on how to leverage this to monitor the Zimbra server's health and status.

The reason we want to do this is that we occasionally find that the server stops listening on those ports--the web portal works, but mail clients can't send or receive mail, and restarting services resolves the issue. We'd like a way to proactively monitor this, though, so that we can be warned if the server stops listening, rather than letting frustrated users do the monitoring for us

[p24t]

I use Nagios for monitoring all sorts of services and ports on my various servers.

Have you investigated the reason for the service stopping? You may want to look at your Log Files to see if you can find any errors. (if you haven't already, or if you have time : )

[mmorse]

Originally Posted by nsmarler

5.01, running on Ubuntu 6.06.2 Dapper Drake

Might be time to upgrade as well, current is 5.0.6 & 5.0.7 is just around the corner.

If you're a Hyperic fan: Zimbra Gallery - Hyperic HQ Plugin for Zimbra

[nsmarler]

Hi mmorse,

Thanks for the good info. I have a stupid question, though--if we want to install Hyperic, and our Zimbra installation is running in an Ubuntu virtualization on an Xserve running Leopard, then can we install Hyperic on the Xserve in OS X and still do the monitoring of Zimbra, or does it need to get installed on the Ubuntu virtualization? Then, if it does need to be installed in the Ubuntu virtualization, will Hyperic's web-page monitoring interfere with the Zimbra web portal or other Zimbra-critical operations?

Thanks in advance for all your help.

[LMStone]

Originally Posted by nsmarler

Is there a way to monitor ports on our ZCS box (5.01, running on Ubuntu 6.06.2 Dapper Drake) so that we can tell if it stops responding on, say, port 465 or other common ports? Maybe through a configuration of the SNMP service? The documentation I found wasn't very clear on how to leverage this to monitor the Zimbra server's health and status.

The reason we want to do this is that we occasionally find that the server stops listening on those ports--the web portal works, but mail clients can't send or receive mail, and restarting services resolves the issue. We'd like a way to proactively monitor this, though, so that we can be warned if the server stops listening, rather than letting frustrated users do the monitoring for us

We use Nagios as well, not only to monitor the Zimbra ports for proper responses, but also for NIC and Disk issues. Screen shot attached from one of our ZCS systems running on HP hardware where the HP Insight Agents are doing the work of monitoring the hardware (e.g. disk health within a RAID array).

Hope that helps,
Mark

[rlingampally]

Nagios look's cool , is it possible to post the configurations for both nagios and zimbra .

[gnyce]

Don't think you would need it, as it is not that hard. For example, nagios comes with a "check_http" plugin that will do exactly that, make an http call. You can look for a specific return string, or a 200 OK... whatever. You can also use it to check https, or on a specific port (7100). There is also a check_tcp plugin, which you can point to any port, to get a response (e.g. smtp, pop3, imap, etc).