You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
With the default addon if the pod can't be scheaduled for whichever reason and the autoscaler tries to update the pod it fails due to lack of rbac permission.
E0805 11:34:43.722542 1 event.go:199] Server rejected event ..................... 'User "system:serviceaccount:kube-system:cluster-autoscaler" cannot patch events in the namespace "hystrix". (patch events grafana-1238051556-9h6h8.14d7effda08f51bb)' (will not retry!)
E0805 11:24:02.840091 1 utils.go:72] Error during reseting pod condition for hystrix/grafana-1238051556-9h6h8: User "system:serviceaccount:kube-system:cluster-autoscaler" cannot update pods/status in the namespace "hystrix". (put pods grafana-1238051556-9h6h8)
The text was updated successfully, but these errors were encountered:
Automatic merge from submit-queue
Fixing clusterautoscaler rbac
Fixes#3144
I have tested on my cluster and it can now update the status of the pods. Im not sure if it also needs the permission to `patch` on the `endpoint`
Maybe someone with better RBAC knowlage can pick up on this? @chrislovecnm
I also gave the file some pretty treatment.. sorry habbit
With the default addon if the pod can't be scheaduled for whichever reason and the autoscaler tries to update the pod it fails due to lack of rbac permission.
The text was updated successfully, but these errors were encountered: