Nagios Support Forum

I'd like to get SLA info on some checks which are redundantly running on multiple (4 for example) hosts. If 3 of 4 of the checks succeed, the app is still working and there's no customer SLA impact. Of course there are lots of hosts and many checks.
Using hostgroups would include hosts so if one or more is down for example, it would impact the regular SLA report. Servicegroups are very tedious and don't seem like a good solution.
Any ideas?

What I would probably do is create a BPI group and use the BPI Wizard to monitor that BPI group and use that for the SLA (in a servicegroup).

Or you could technically use the check_cluster plugin to do what you want:

https://assets.nagios.com/downloads/nag ... sters.html

Both would create a new service that would show the status of what you want (the real SLA).

I can't really think of any other options.

That doesn't really fix the SLA report though, only using the servicegroups would limit the services.

I'd never seen the cluster check before. Thanks for the pointer to it. Looks interesting.