Configuring Nagios to monitor a Windows service

JeremyInNC · 07-29-2008, 09:49 AM

I am running Nagios 3.x on a RHEL box. It's able to do the default ntcheck stuff like CPU utilization and see the NSClient++ version, but I can't verify that a specific service is running (shows up as unknown). Here's the entry in the host's cfg file for the service:

Code:

define service{
        use                     generic-service
        host_name               gsovmacs01
        service_description     Cisco Secure WWW Server
        check_command           check_nt!SERVICESTATE!-d SHOWALL -l CSAdmin
        }

The weird thing is I *can* telent into the Windows box on the Nagios client port and when I open the NSCLient++ command injector and do the CheckServiceState for csadmin it comes back happy. What am I missing?

Quote:

Cisco Secure WWW Server

WARNING 07-29-2008 10:42:07 0d 0h 52m 12s 4/4 CSAdmin: Unknown

JeremyInNC · 07-29-2008, 10:43 AM

For anyone interested it's because I'm stupid

I've been cp the template file and then doing a %s/oldname/newname/g on it...but never changed the IP address. So suprisingly enough the "error" I was getting about the service not being found was actually accurate. All is now well, except I need to figure out how to stop my default group from showing up in the CGI.

mab1376 · 01-13-2009, 09:10 AM

What does the -d parameter do? and can i have it check a single service such as Oracle instead of using SHOWALL?

dguitar · 01-13-2009, 09:19 AM

Running the check_nt with --help (via the CLI) will give you a good explanation of what to use it for.

You get a lot more flexibility using NRPE rather than check_nt. Check out the NSClient++ website for more info.