Triggering incidents after multiple threshold checks
Hi I am monitoring some Activemq queues through DT and I'm trying to figure out how to throw an alert if the queuesize doesn't change over time. I want to check queuesize every minute, and trigger an incident only if the queuesize does not change for 5 consecutive checks (indicating the queue is stalled). I know that dynatrace gets all the data it needs to be able to do this, but I don't see a native way within the application to measure in this way. Has anyone else attempted this, or does anyone have any tips on how to achieve this withing DT?
The built-in Incidents that are based on measures work in a way that the Incidents trigger in case a measure exceeds a certain threshold. In your case you want to alert if nothing changes. Thats not possible with the built-in capabilities
What you can do is write your own Monitor Plugin that monitors your Queue Sizes. This plugin returns a simply value saying "Alert Yes/No". The logic of that plugin checks whether these values change. In case they dont change 5 times in row this plugin returns a value 1 (=Alert). Otherwise it returns 0. You can then define an incident that triggers in case that return value goes to 1
Andi
