DX Application Performance Management

Hello,

I have created an alert for swap space used, I have my alert set for Caution at 40.

Comparison Operator is "Greater Than", Trigger Alert Notification is "When severity increases"

I set the Periods Over Threshold to 1 and the Observed periods to 240. I also hit the "apply" button

at the top of the hour.

My question now is, I did get an alert / email that I expected, except when I view the graph, the alert hits when the

swap first jumped over the 40% in use, and I want it to trip when it is over 40% for an hour. Is the alert checking or resetting itself on the top of each hour?

Feedback welcome,

Pat O'Connor

Hello Pat,

In this case the alert re-evaluates itself every 15 seconds using data from the last hour. (This slides forward so it is 2:30, data from 1:30 to 2:30 is considered).

In your case, as the alert is defined, if one 15 second interval was over 40% in the last hour, the alert would trigger. You may want to set Periods Over Threshold to 240 out of 240 Observed Periods. In this case every data point for the last hour would need to be over 40% for the alert to trigger.

Hi Thomas and Matt,

Thank you guys for the clarification on this one, I read the admin guide and still had not quite figured it out,

I will give that a try.

I Appreciate your Help on this one!

Best Regards,

Pat O'Connor

Hi Pat,

The alert does not reset itself in the sense that it waits a full hour.  After each resolution is met, it rechecks.  For example, starting at 2:30:00 PM, the alert will check from 1:30:00 PM to 2:30:00 PM and evaluate the points in between.  When the alert advances to 2:30:15, it evaluates between 1:30:15 PM to 2:30:15 PM and so on.  Below are the definitions of the periods.  Since you set Periods Over Threshold to 1 and Observed Periods, that means the Alert monitors the last 240 periods (noted by what you set "Resolution" to) and trigger the alert if any 1 of those 240 observed periods goes over the Caution line.  

• Periods Over Threshold
  Specifies the maximum number of periods that the threshold can exceed before an alert is triggered.
• Observed Periods
  Specifies the total number of periods being monitored for each alert iteration.

Thanks,
Matt