Nagios Support Forum

Hi,

Is there a way to exclude some services from the Service Status Summary (at least the Unhandled list) while keeping the checks active?

For example, we would like Nagios Xi to keep monitoring default MSSQL checks (from the MSSQL configuration wizards) on a dozen hosts (about 50 service definitions per host) to collect metrics and help analyzing these at a later time while keeping an eye on what is considered an issue with default threshold values (I know default values don't fit all situations and some of these probably just need to have warn/crit threshold adjusted for our uses), but right now they are flooding the unhandled/problem lists.

It also could be useful for other low-priority/informational service checks (e.g. pending reboots, software versions, etc.).

For now we can ACK these service problems to make them disappear (while still checking them from time to time if needed, using a Service group summary list), but some of them keep coming back frequently (especially the MSSQL ones) whenever the service change status again.

Of course we could adjust the warn/crit triggers to make sure they always remain green so no more Unhandled flooding, but we would prefer to leave the warn/crit triggers as is for now and just exclude them from Unhandled list if there is a way.

Thanks

Unfortunately, currently only acknowledgements/downtime/ok states would remove them from the unhandled list.

I'm wondering if you can just submit a long scheduled (or recurring) downtime for them as an alternative.

Another alternative would be to use the negate plugin to negate all the WARN/CRIT/UNKNOWN and make them OK states, but that would require you defining a new command (that is a copy of your existing commands that each check uses) like this below example:

Current command definition in Configure > Core Config Manager > Commands:
Your new one:
- This will turn warning, critical, and unknown states into OK states for whatever service uses this check command
Then edit the services and change the check_command to the new one, and apply configuration. While not the easiest to manage it's an alternative.

Will one of those work for you? If not, I can submit a feature request to development requesting they add some sort of exclude functionality? Please keep in mind that the decision to implement the enhancement is at the discretion of our development team.

ssax wrote:I'm wondering if you can just submit a long scheduled (or recurring) downtime for them as an alternative.

Maybe we could try that. We're never bothered too much with using downtimes until recently.

ssax wrote:Another alternative would be to use the negate plugin to negate all the WARN/CRIT/UNKNOWN and make them OK states, but that would require you defining a new command (that is a copy of your existing commands that each check uses) like this below example:

The idea was to keep an eye on the warn/crit statuses on the Status Window or Service group, just take these out of the unhandled/problems list. That workaround would remove the warn/crit statuses. I'll keep that workaround in mind in case we need it later for something else, but don't think it's worth the trouble editing the 50ish services for this particular case.

ssax wrote:Will one of those work for you? If not, I can submit a feature request to development requesting they add some sort of exclude functionality? Please keep in mind that the decision to implement the enhancement is at the discretion of our development team.

Thanks. I guess it doesn't hurt to submit a feature request. I'm not sure there is a huge demand for this, and by the time they implement it (if at all) there's good chances we have had time to adjust settings for these MSSQL checks by then, but there's definitely other cases it could be useful.

I've created the request: