Nagios Support Forum

We have several customers that hide more than one (Windows/NSClient++) server behind one single IP-address. What I usually do in these cases is forward increasing port numbers to those hosts (12489-> HostA, 12490->HostB, 12491->HostC, etc.). On the Nagios side I need to create a service definition for every combination of portnumber and service ( one for CPULOAD at 12489, CPULOAD at 12490, etc.) which tends to fill the service-definitions-file with a bunch of definition entries.

Can anybody stir me towards a better (i.e. more readable and manageable) setup for this problem? I'd love a way to pass the port number to the host definition and use a generic definition for each service. I feel like I could add the port number to the host address (12.34.56.78:1290) but am at a loss on how to parse this in the command definition.

Any suggestion?

The other way you can deal with this is with passive checks. The remote systems can just be identified with a host name, and the check results sent to the Nagios server via NSCA.