Background: I would like to count (and eventually alert on) the number of database calls that exceed 30 sec. Currently, I am using the Count (Threshold Violation) with a source measure DB Total Time measure with an upper severe of 30 sec.
Question: With this configuration, is the count measure determined by the individual calls that are over the 30 sec or is it counting the number of times the DB Total Time aggregated measure exceeds the 30 sec threshold? If the latter is true, which aggregation (average, min, max etc.) does it use and what resolution is it?
Solved! Go to Solution.
I have had the understanding that it will be looking at the count of PurePaths that contain a violation of the threshold of the source measure that you configure. So if you set the DB threshold to 10 seconds it will be a count of PurePaths that contain a DB call longer than 10 seconds. I'm not sure if it is counting the number of PurePaths or number of calls (I'm leaning towards the former) but I'm quite certain it is not the case of using some aggregation as you can't set any aggregation for the source measure and what is used in charting wouldn't matter.
Someone please correct me if I'm mistaken,