Scalix Forums

I've been using Nagios here for some time, but it was originally set up by another person. Only recently have I been digging into the guts of the thing--especially as it pertains to Scalix. I was wondering how many other Scalix users are monitoring with Nagios? Is it enough to warrant a sub-forum of its own?

Mike

I use Nagios for a few clients, and it works quite well.

I see very few questions about it though, so I'd suggest that if you have questions or issues, you could use the Third Party Integration forum.

Thanks
Kev

I use nagios, but at the moment not to monitor Scalix. Need/want to setup Scalix monitoring in the future

If you're already using nagios, adding Scalix into the mix isn't too hard. I'd be glad to help you out. I take it you are running nagios on a machine other than the Scalix box? Are you familiar with using nrpe fro nagios?

There is a Scalix document which outlines how to set up Nagios for monitoring Scalix. I haven't looked with 11, but with 10, it was in the admin resource kit. I would guess that it has now been moved to the wiki.

Kev.

no it is not in the wiki, still the old version in the ressource kit.


can someone share his modified config to have it cetrally available?

We used Nagios for a while. It's a very nice product but I always found the initial configuration very cumbersome, and often times too much irrelevant information being presented, which was confusing. Then we switched to openNMS in combination with Cacti, and it's much better now.

openNMS always gives me a snapshot view of what is going on with all systems on the network. Cacti gives me a graph history of several performance parameters on the most critical systems. Both are very simple and quick to setup and modify.

how does openNMS talk to Scalix? Are there "plugins" or does it use SNMP or something else? I'll check out their website, but want to know about how you have it working with Scalix.

Yikes! It runs on java? I'll bet it's a giant resource hog!

OpenNMS is SNMP driven. It's not resource based at all. In fact we use one of our oldest and retired servers just for that purpose. We configured it with a number of monitoring fixtures such as Cacti, Ganglia, MRTG, OpenNMS and SmokePing. Best of all, I can access all these applications from my BlackBerry's tiny web browser, graphics included. That's actually one of the criteria for my choosing of some of these applications.

OpenNMS provides information on all services running not only on the Scalix server, but on all other critical servers we have in the company. The monitoring server itself is setup to email me OpenNMS warnings and alarm information directly to my BB, independently and by-passing both the scalix server and the firewall.

Unlike Nagios (and Zenoss as another example), OpenNMS information is presented on a extremely simplified and straight to the point format, only 3 colors: green, yellow, red.
If you have a bit of time to do some research give some if these applications a try. You won't regret.

I just saw that it is written in java--which was surprising. Do you have to install anything on the remote hosts to monitor anything other than services (HTTP, FTP, etc.)? I'm monitoring a lot of systems now with nagios, but will look into this a bit further.

For most cases, within the same vlan, you don't have to do much on the clients. The ICMP Auto Discovery is pretty good. You can find more info on how to configure here:
http://www.howtoforge.com/opennms_network_management