Status of ESX host services

[spasq01]

I recently installed the VMware wizard and added two ESX hosts to monitoring. I checked off all of the options for what I wanted to monitor on each host. I used the ESX host's root credentials when I went through the wizard. The hosts have been successfully added to Nagios. In the Host Status page both hosts show as Up but when viewing Service Status the services have a status of Critical and the Status Information column is showing a value of 'null'. I went through the documentation for monitoring VMware and I did install the Perl SDK as well as the Perl plugins. Everything appeared to install properly. Is there anything to needs to be installed on the ESX hosts themselves? I didn't see anything about an agent. It appears that communication between Nagios and the ESX servers is not working or there is a rights issue with monitoring. Has anyone else run into this issue? Thanks!

[mmestnik]

We did discuss this at length and I almost think I understand, at least well enough to pass this on. The VMWare monitoring software monitors the VMWare software provided by VMWare, it does not monitor the software provided by Microsoft or RedHat for example.

It's true that when you add monitors for the VMWare server you can choose to monitor the values related to a guest instance, but it's much more like monitoring the number of hits to a web page and has nothing to do with the browsers making the request.

[spasq01]

I'm not really sure I follow what you are saying. Is what I am seeing normal behavior? Does the VMware wizard use snmp? Would I have to edit the snmpd.conf file on my ESX server to follow the Nagios server to poll it?

[mmestnik]

We use the VMWare SDK. I happen to know that this software uses http and perhaps soap or something similar.

The fact is that this software does not monitor the guest's. It monitors the host's and specifically the parts of the host that pertain to a particular guest.

[spasq01]

I'm not trying to monitor the guests, just the ESX hosts. Is there any documentation out there on how to set this up?

[tonyyarusso]

If you're using the wizard made by us (as opposed the the community-contributed one), yes - http://support.nagios.com/forum/viewtopic.php?f=8&t=554

[spasq01]

I've already read this document. It helped to get the wizard and appropriate components installed but it doesn't give any details on configuring it so it works.

[mmestnik]

Perhaps we should take a step or two backwards, describe your setup and what you'd like to accomplish.

As for terminology, we are using the Host/Guest ideology. Though previously I got all confused, I'm in favor of using the Target/Source ideology to avoid further confusion... Though it's lesser used in the industry.

[spasq01]

I was using Host/Guest in terms of vmware. I'm basically trying to evaluate this product. I'm running through all the wizards. I used the documentation to install the vmware wizard. I went through the wizard twice adding two ESX servers to be monitored. I checked off all of the options to monitor (cpu, memory, networking, etc...) After successfully finishing the wizard I see the ESX server in my Host Detail with a status of Up. When I check Services Detail all of the services I have selected in the wizard show a status of (null) in the Status Information column. I explained all of this before. It appears that Nagios cannot poll these services on the ESX servers. I am trying to figure out if there is additional configuration that needs to be done, or after running the wizard, in theory, everything should work?

[mmestnik]

Please explain your environment. I'm unsure of what to ask about but if you start thinking in terms of what I don't know, like the brand of servers switches ect, then you should be able to complete a view of your situation that I could verify.