autoscaler: Missing events on status config map

Even though Scale-up: group events (https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/core/scale_up.go#L715) should have been triggered as well for the status configMap as shows the log:

Event(v1.ObjectReference{Kind:"ConfigMap", Namespace:"test", Name:"cluster-autoscaler-status", UID:"6f9ba031-9377-4984-8a91-a8f6ef363f63", APIVersion:"v1", ResourceVersion:"225776", FieldPath:""}): type: 'Normal' reason: 'ScaledUpGroup' Scale-up: group test/e2e-d6645-w-2 size set to 2

I can always only see/receive one single Scale-up: setting (https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/core/scale_up.go#L701) event:

kubectl describe configmap cluster-autoscaler-status

Events:
  Type    Reason         Age   From                Message
  ----    ------         ----  ----                -------
  Normal  ScaledUpGroup  50s   cluster-autoscaler  Scale-up: setting group test/e2e-d6645-w-1 size to 2

Not sure what I’m missing.

About this issue

  • Original URL
  • State: closed
  • Created 5 years ago
  • Comments: 19 (12 by maintainers)

Commits related to this issue

Most upvoted comments

Hi @enxebre. The PR you mentioned aggregates events so that CA for scalability reason where we don’t want to send a lot of similar events which might happen if we have too many unschedulable pods.

We’ll fix this by having a different config for the logrecoder. Expect the fix some time next week

+1
vivekbagade on Oct 30, 2019
Read more comments on GitHub
← autoscaler: Memory leak?
autoscaler: My pod is in the CrashLoopBackOff state after configuring cluster-autoscaler  →