Follow-up from "Use AlertManagement::Payload for processing Prometheus notification"
The following discussion from !39271 (merged) should be addressed:
-
@syasonik started a discussion: (+1 comment) This is a scenario where we might swallow an alert. If we have effectively a self-resolving alert, we'd never persist it to the DB.
Maybe the firing alert request timed out or errored on the prometheus side or something, but I don't think that's a superb reason to just ignore this alert, then.
Again, possibly something for a follow-up if others feel similarly.