8000 Replace ceph monitors in Pending state on non-existing hosts more quickly · Issue #15953 · rook/rook · GitHub
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

Replace ceph monitors in Pending state on non-existing hosts more quickly #15953

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
telmich opened this issue Jun 5, 2025 · 0 comments
Open
Labels

Comments

@telmich
Copy link
telmich commented Jun 5, 2025

Is this a bug report or feature request?

  • Feature Request

What should the feature do:

When the rook-operator find a ceph monitor deployment that is

  • a) in state PENDING
  • b) having a nodeSelector for a node that does not exist (anymore)

immediately create a new deployment using a new, existing node from the kubernetes cluster.

What is use case behind this feature:

The background is mostly explained in #6663 (comment), but as a summary:

  • Nodes can vanish in a kubernetes environment (being deleted, hardware failure, etc.)
  • The ceph monitors on those will not be deploy-able anymore, as the containe a nodeSelector field referencing a non-existing node
  • Thus the pods will get stuck in PENDING state
  • Rook has a standard 2x 10 minute timeout for updating stuck monitors, but in case a&b are given, a deployment can

Environment:

Kubernetes

@telmich telmich added the feature label Jun 5, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

1 participant
0