Maybe you also have a scraping/data collection problem. I saw all those outages in your "up == 0" graph - that's not good. Fix that first.
The value of a metric is whatever the last value that was scraped, up to a maximum look-back time which defaults to 5 minutes. You will be able to determine this by entering your alerting expression directly into the PromQL web browser and then switching to "graph" mode. That is, enter rabbitmq_queue_messages > 10000 into the query box. The graph will show lines wherever alerts would be generated. If the lines are horizontal, then either the same metric value was being returned in each scrape (perhaps it was cached on the rabbitmq side?) or else the scraping failed (look at the corresponding "up" metric for the job which collects that data) This is stuff that you'll have to sort out on your own system, I'm afraid. -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to prometheus-users+unsubscr...@googlegroups.com. To view this discussion visit https://groups.google.com/d/msgid/prometheus-users/e9d21390-e973-409e-a1b6-8e16b971c79an%40googlegroups.com.
