Hi,
We've setup an alertmanager cluster multi site and use it to forward our alerts from many prom instances out to opsgenie /
alerta.io dashboard.
alerta dashboard has the ability to generate dynamic heartbeats based on the watchdog rules using a specific label to create a new heartbeat when received.
Opsgenie has the provision via webhook to send a heartbeat but
is heartbeat has to match up the URL with the internal heartbeat name.
I would like to make this dynamic so that I do not need to configure an individual route for every heartbeat that fails. (we have >10 from all out prom instances)
what I thought would/should be possible would be something like
however, this doesn't work. currently with a single HB config I have to wait until all heartbeats have failed before triggering the HB alert in opsgenie (not ideal) however it will tell me that the alertmanager is down.
Does this make sense? or am i mis-understanding the role of the dead-man switch? Is there any provision to get access to label values in the URL or another way to construct the URL?
Thanks