kubernetes: Updating kube-proxy configmap doesn't reload/restart kube-proxy

/kind bug

What happened: I’ve updated the kube-proxy configmap after the master IPs changed but had to delete kube-proxy pods manually for the changes to take effect.

What you expected to happen: That kube-proxy reloads or get restarted on configmap changes.

How to reproduce it (as minimally and precisely as possible):

  • Create cluster with kube-proxy and kube-proxy configmap pointing to the apiserver
  • Change the kube-proxy config map, pointing to a new apiserver
  • Deleted the old apiserver

After that, the cluster should become unresponsive due to kube-proxy still talking to the old apiservers.

Environment:

  • Kubernetes version (use kubectl version): 1.8.4

About this issue

  • Original URL
  • State: closed
  • Created 7 years ago
  • Comments: 22 (14 by maintainers)

Commits related to this issue

Most upvoted comments

We could make kube-proxy inotify it’s config file or just poll it and restart. That seems reasonable.