I would like to create an alert based on particular exception within request like below capture. Example, if this exception occur 10 times in 5 minutes, Dynatrace will send the alert.
But I don't know how to create alert based on particular exception like this in Dynatrace Managed, as I know in AppMon is very simple and easy to create this case.
Anybody can give me idea to create this case ?
Solved! Go to Solution.
I can think of two different approaches here, depending on your use case:
First would be to define a calculated service metric that will be a count of requests having that particular exception and then define a custom event for alerting on that metric. You then will have a precise counter of requests with that particular exception.
Second would be to modify the error detection rules for your particular service. You can add an exception to the "Custom handled exceptions". This will then mark requests with the exception as failed. Anomaly detection for service will then see failed requests and you can alert on that. With this approach, you won't distinguish between exception and other request failures.
In Settings -> Server side monitoring -> Calculated service metrics, I'd define something like this:
You might want to define additional conditions such as tags or process groups. This will create a new metric representing number of requests having hit particular exception with the exception message and you can alert based on this metric.
However, since your example is just about IOException, I'd recommend maybe reconsider monitoring that particular endpoint (either with OneAgent or if it's unmonitored host, then by creating a custom device) and not alert on such generic exception.