I have a interesting glitch going on with our xi setup. First off we have a core2quad system with 3gb of ram and 2x 1tb WD black drives. We are monitoring our vmware cluster with this using the vmware wizzard. We have a total of 5 host servers and our first host sometimes gives a False service check notification that something is failing. When I first added the vm hosts to XI our first VM host was showing that it was possibly flapping while i looked around for an issue I disabled flap detection. After this i started to get a lot of false notifications so i re-enabled flap detection and now im still getting false notifications. For example it will say -
2010-09-30 17:38:20HOST ALERT: kass-vm1.kasslaw.com;UP;SOFT;2;OK - 192.168.0.90: rta 0.256ms, lost 0%
2010-09-30 17:38:00HOST ALERT: kass-vm1.kasslaw.com;DOWN;SOFT;1;CRITICAL - 192.168.0.90: rta nan, lost 100%
2010-09-30 17:36:20Auto-save of retention data completed successfully.
2010-09-30 16:45:20HOST ALERT: kass-vm1.kasslaw.com;UP;SOFT;2;OK - 192.168.0.90: rta 0.363ms, lost 0%
2010-09-30 16:45:10HOST ALERT: kass-vm1.kasslaw.com;DOWN;SOFT;1;CRITICAL - 192.168.0.90: rta nan, lost 100%
But during this time I am actually remoted into one of the virtual machines on that specific vmware host and it is definitely not down. Also we use vmotion and our vmware guests don't get moved to other vmware hosts.
Hmm, we'll have to do some looking into this one, but it seems like this is a problem related to the actual check plugin itself. Try running your check manually from the command line and see if you get the same results. Outlined instructions below if you haven't done that before.
cd /usr/local/nagios/libexec/ (this is the directory for all of the check plugins)
./check_<plugin> (enter your command line arguments for that particular host here)
If it's coming back with the same info on the command line then it's probably related to the plugin itself. Can you send us the actual check command for a good host server, and then the check command for the one giving problems?
You may have to change things around a bit to monitor your environment. We don't have a vmotion environment and as such there are several options for the vmware check command that we were not able to add to the Wizard.
I believe you are looking to use the VMWare Data-center options instead of the VMWare Server options the Wizard uses.
Well not sure what happened but I was investigating it further and it appears to have cleared up on its own. Also just discovered now that our guests are automatically shifting between hosts via vmotion so the esx plugin is only valid for host machines now because if a guest moves hosts then It will loose it and will not find it again. Is there anything I can do about this or is adding all of my vmware guests manually by ip address pretty much my only option I really liked the data vmware was giving back to nagios.
The vmware wizard does not have these options. Use the wizard and then alter the resulting configuration to make use of the vmware plugin's other options.
