Received SIGTERM, exiting gracefully

[jerome2710]

A former employee of us setup Prometheus through Puppet for our servers. Unfortunately, it seems that Alertmanager keeps stopping, preventing us from receiving alerts. This creates a problem, where we are not automatically notified of issues anymore.

I am not super familiar with Puppet, but know a thing or two about our servers. Does anyone know what could be causing our dead alertmanager instance? I tried restarting it several times, but it somehow stops every time.

[TheMeier]

The probability is very slim that this has anything to do with the puppet module. Received SIGTERM, exiting gracefully... means the process received the TERM signal for some reason. Maybe something else is running on that node, that intentionally stops the service.

Maybe running alertmanager with --log.level=debug gives you some more info. Also please post module version, operating system and the relevant puppet code you are using.

[TheMeier]

Looking at it again, it even says systemd[1]: Stopping Prometheus Aletmanager ~8 seconds after its was started.