Alert when Runner crashes

It happens that Runner does panic. Can we somehow scrape logs to detect this situation and make it obligatory to take a look and fix?

Runner crashes are catastrophic to currently running builds, so this should be immediately fixed by CI/CD team, but currently we can only see that by looking at uptime of the runner process in Grafana. Maybe this is something that could generate an alert?

You can see what to look for by looking comments of this issue: https://gitlab.com/gitlab-com/infrastructure/issues/1073

Assignee Loading
Time tracking Loading