Hi!
I was wondering how to group a sum over time for a value which is coming
from multiple instances?
For example consider a distributed environment. The same metric is reported
by instance 'a' and instance 'b'. So I have the same metric for server
restarts for example reported by the same metric reporter but twice.
I would like to have a sum over time for server restarts to see which
server restarted how many times over the last 2 days.
I tried various combinations, but how do I ignore the duplicate entries? I
tried something like...
sum_over_time(server_statistics_total{restart_reason="failed"}[7d])>0
But this results in insane large numbers. The metric is a Gauge. So that
doesn't help. :) Because I'm guessing when the gauge decreases because the
server normalises, that will also be reported as 1 and will be counted
towards server restarts... Uh. :/
Any advice would be much appreciated.
Cheers,
Gergely.
--
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 on the web visit
https://groups.google.com/d/msgid/prometheus-users/d0e46728-1c52-4d10-89de-1f286c57b93bn%40googlegroups.com.
