Application Polling Adjustments

Orion Platform 2018.2 HF6, SRM 6.7.0, IPAM 4.7.0, SAM 6.7.0, DPAIM 11.1.0, NCM 7.8, NetPath 1.1.3, NPM 12.3, NTA 4.2.3, WPM 2.2.2

I am working to build alerting for services but need to account for the recovery actions set on the service's properties on the server. I want to only alert if the service would still remain down after the recovery actions were attempted.

It looks like there are options to wait x number of polls for performance metrics but not for the "not running" status.

Current config:

adjusted polling to wait , but I seem to get an alert no matter if i am at 300,600 or 900

Polling Frequency:	1200 seconds
Polling Timeout:	1200 seconds

Does this polling adjustment work the same as if I had the option to adjust to "has to occur for more than x polls"?

This is on a Windows Server 2016 Node polling via wmi w/ wmi service monitor applied.

Any advice or personal successes would be welcome.

Find more posts tagged with

application management

Application

thresholds & polling

sam 6.7

advanced alerting

Accepted answers

stevenstadel

The polling frequency does not determine threshold status, it is just the amount of time between polls, and the timeout allotted before giving up. The threshold values in the application template will only set the levels at which the component will turn into a Warning or Critical Status.

Alerting is a different thing altogether. Alerts check the database at an interval to see if a set of conditions have been met. In the alert details is where you have the option of establishing a wait condition.

If you want to get creative you can also use Escalation levels in your Trigger Actions to try an action and if the condition is not resolved in a specific time-frame (like the 10 minute example below) then execute an escalation action like sending an email.

All comments

stevenstadel

bobby_dd

Thanks for the response, that would make sense why only adjusting the polling frequency would still trigger the alert if it occurred at some point in the window. I was trying to avoid creating custom alerts or solution for each app monitor given the amount of apps in our environment. It is sounding like that may be the only option tho if it cant be addressed on the templates. Does my assessment of the situation sound accurate stevenstadel ?

stevenstadel

Yes, it is best to think of the thresholds and the alerting as two separate entities. You tweak the template to suite how often you want to check for the condition and then setup an alert on how often you want to send a notification or perform an action.