Why Is This So Hard?How do I Monitor a UPS? (check_apcupsd)

[gormank]

Anyone can buy a used book for $5 shipped off the internet for Nagios core that will show the stuff we all need to know.
Chances are with a little searching the same book can be got for free as a pdf if $5 is too much. I dropped $10 and bought 2. I point coworkers to the pdfs.
Learning core, or at least reading about using it, before XI is a good thing. XI makes it too easy to do bad things, while core requires some thinking of how to do things.

As far as monitoring a UPS, it begs for SNMP from square one. 1st traps, then grab data actively w/ snmpget. Both require minimal config on the UPS. Both can be set up initially w/ a wizard, then the resulting service generalized to allow easy monitoring of a hundred similar devices. I do a lot of active SNMP monitoring to tell me about hardware.

Look at it this way, you took a wrong turn at the beginning, but think how much you learned by doing so. The next thing you start to monitor will be much easier. I've been doing it wrong for a while now, and I think I'm starting to find my way back. :)

Can you take a screen shot of the relevent active service in CCM and post it? Crop as needed.

Hmm, I think you mentioned things that had the same name. If so, try renaming things w/ unique names.
You might also look at the files in /usr/local/nagios/etc/services and ../hosts. This is where the rubber meets the road, since XI is layered on top of core. That means Nagios runs off the config files. XI stores the info in mysql and writes the config files when you tell it to (apply configuration), and then restarts core to reread the files. If you watch you can see a short outage, that if Nagios is monitoring itself occasionally catches.

[rkennedy]

Thanks @gormank!

Can you send over a profile for us to look at? (Admin -> System Profile -> Download Profile)

We can look at your core config files from there, and see what's going on exactly.

[Moneer81]

Thank you for all your help on this.

Somehow, magically, the service appeared a while later. I am not sure why or what caused it to happen. I am wondering if the fact that this is a virtual image has anything to do with it. At one point, the machine hosting that image of Nagios did reboot, so maybe that had anything to do with it.

In any case, the service is now there.

The other service that I defined (SNMP traps) keeps showing "Trap Reset" but I might have to make sure that I defined that correctly.

If anyone has any quick fixes for the Trap Reset issue, then let me know, otherwise we can assume this issue has been resolved. I might do some more research on the SNMP trap issue.

Also, if anyone has a good suggestion for what is the best way to monitor a UPS or what plugin they use, I would be interested in that too.

Thanks a lot for all the help guys!

[Box293]

Your SNMP Trap service has "freshness" checks enabled.

Freshness means, if you don't receive an update within x (x=freshness threhsold) then execute the check command assigned to this service.

xiwizard_snmptrap_service is the template being used by your service and it's the one generating the TRAP RESET message.

It's an easy way of resetting things to normal.

To turn this off,edit the service in CCM.
On the Check Settings tab set "Check freshness" to Off

Save an apply config.

Now this service will always display the last trap it received.