It is possible to implement auto-resolution with Pagerduty API integration. When a monitoring system detects an issue based on some metric and a threshold, it would create a Pagerduty incident through the API integration. Later, if the metric falls below the threshold, it could automatically resolve the incident, in case it is still open.
This feature could be very useful for several reasons:
Cases where a large number of incidents get triggered due to a common root cause
In this case, the on call person is spared from having to ack/resolve so many incidents and instead focus on troubleshooting and fixing the problem.
Auto-resolution would capture the incident duration more accurately
Humans are not good in doing things on a timely basis. If we are computing MTTR based on manual resolution of incidents, it may lead to inaccurate measurements.
Has anyone implemented auto-resolution with Pagerduty? If so, would you be able to talk about the pros and cons?