Monitoring VMWare DataCenter & Hosts with Nagios

[buckshot44]

Hello All,

I'm wondering if someone out there can help point me in the right direction. We use Nagios to monitor our network and servers and what not. Up until now, I've been moniotring individual VMWare hosts. Recently we've turned on the power saving feature which will put VMWare hosts into "Standby" mode (powered off) until the resources are needed. This obviously saves the company money on power comsumption, but Nagios freaks out everytime thinking the host is "down". I'm wondering if there is a slick method of monitoring just the datacenter to get the proper status of the host. (Ie. "Standby" and not "Down")

I've been playing around with the check_esx3 perl script, but haven't been able to get it right.

Any help is much much appreciated.

Thanks,

Buckshot44

[mcowger]

I would recommend setting up an event handler in nagios for the host. Have the event handler do a check of vCenter to determine if the host has been placed into the standby mode, and if it has, to ACK the alert automatically.

OR

create a new host check command/script that rather than the default (which just does a ping) does a ping and THEN a check of vCenter viathe APIs as well).

--Matt

VCP, vExpert, Unix Geek

[buckshot44]

Hi Mcowger,

Thanks for your quick reply. I'm not familiar with the VCenter API's, but I think that is the route I'd like to go. I want something that is smarter than just a ping, something that will return the "State" of the host.

Any suggestions on some reading material or how I can go about it?

Thanks again!

[mcowger]

Check out lamw's host check script - its long, but will show you the perl API call required to check the 'state' of the host (which is probably just a few lines).

--Matt

VCP, vExpert, Unix Geek