argocd_cluster_events_total metric is high for ConfigMaps #21501
Replies: 6 comments
-
Yep!
Check your Kubernetes API audit logs. |
Beta Was this translation helpful? Give feedback.
-
Thanks! Looks like it is caused by |
Beta Was this translation helpful? Give feedback.
-
Something like this should work I suppose... |
Beta Was this translation helpful? Give feedback.
-
Well, it didn't. This metric is still high with that |
Beta Was this translation helpful? Give feedback.
-
Ok so I am reading this https://argo-cd.readthedocs.io/en/stable/operator-manual/reconcile/#ignoring-updates-for-untracked-resources - sounds like my customizations are not going to be picked up unless resources are directly tracked, otherwise resources needs to be annotated with |
Beta Was this translation helpful? Give feedback.
-
Anyway I think this actually is a bug, so I am creating #21506 and closing this question |
Beta Was this translation helpful? Give feedback.
-
See screenshot - this look suspicious to me. Can someone please explain what does that metric tracking, exactly? I don't see kubernetes events anywhere near that numbers, so it must be about resource updates then? How do I identify where and why it all comes from? ConfigMaps are far from the most frequently edited resource in my clusters...
Beta Was this translation helpful? Give feedback.
All reactions