I'll try to bring more details here:
Actions:
1. alertmanager triggers pagerduty incident,
2. acknowledging the incident in PD console,
3. fixing the problem
4. manually resolve the incident - again in PD console
5. the real problem is still there - an alert in alertmanager is still in active state ( have not fixed it actually)
6. timeout_interval in alertmanager.yaml is set to 1h
7. what happens is that in a random interval ( it's less than 1h after I've resolved the incident) - I do receive a totally new PD incident on the initial issue.
According to this
link this is desired behavior or no? If it's desired - how come the incident in not triggered right after resolving?
If it's not desired - why do I receive the alert at all?
I'm running the latest version of Alertmanager.