Nagios Support Forum

Good time of day,

Our currently design for server monitoring is as follows:

A Service config template has several NRPE-based services with FreeVariables passed as a Warning/Critical values
This config template is a assigned to appropriate HostGroup
HostGroup contains several servers of similar type

the values for Free Variables are declared on each host (thusly alerts can be tuned uniquely based on volume/size of each server without re-configuring the whole hostgroup)

We are looking for a way to add time schedules to alerting variables.
For example,
if we monitor the concurrent call usage on asterisk. Currently, a free variable associated with Warning value set to value of 200 statically(24x7 schedule), so if over 200 calls found, Warning is given. This value is good fit for alert during business hours, however during the off-peak, we would like to trigger the Warning on value of 5 concurrent calls instead of a 200 on the same host..

Our pre-sales team recommended looking into Adaptive Monitoring:
https://assets.nagios.com/downloads/nag ... ptive.html

CHANGE_CUSTOM_HOST_VAR external command seems to be a good fit for the above functionality;
however the documentation listed does not really describe on steps required to implement adaptive monitoring in Nagios XI.

Would you be able to advise to what exactly we need to do to active this external command?
How do we then attach this external command to pre-configured schedule etc...

Thank you in advance.

Given your use case, I'm not sure adaptive monitoring is what you need. Why not just set up two different services with two different timeperiods? One for after hours and one for the business day?

It's certainly not the only way to do what you want, but it seems way more straight forward and less prone to error than the other methods. If you do become a customer, you might want to look at https://support.nagios.com/forum/viewto ... 16&t=43193 for some ideas (it's customer-only, and while I could copy/paste the text, I think the context might get lost).

If you do want to go with adapter monitoring, I think https://old.nagios.org/developerinfo/ex ... and_id=108 probably makes more sense than CUSTOM_HOST_VAR. You aren't using customer vars as far as I can see. You want to change the warning and critical values. Also, I suspect you want to do this on a service rather than a host, but you know your infrastructure better than I do.