Why do (external) incident tasks have a different format of ActivityInstanceId?

We have noticed that normal ‘running’ external tasks have an ActivityInstanceId that follows the format of [ActivityId + some GUID] but when a failure is reported for them, a new ‘Incident’ activity instance is created which has an ActivityInstanceId of [ a new GUID].

Is this intended or some kind of a bug in the system? If it is intended, I would like to know the reasoning behind this logic. Thanks!

Hi @vigb,

what do you mean with


Incidents don’t have an activity instance id. Can you provide an example?

Best regards,