PVC Consideration

[jrnt30]

Project looks interesting! One consideration on the "Future Roadmap" that would be worth considering is the Fault Domains and the PVCs that are associated with a pod.

"Evicting" the pod with a PVC due to LowNodeUtilization on another node would not result in actual re-placement of that pod, so it shouldn't be attempted.

[ravisantoshgudimetla]

@jrnt30 - Eviction of pod with a PVC seems unreasonable as this may cause data loss. So, descheduler should not do this. But I will add this to a later release.

[jrnt30]

Thanks for the response. To be a bit more clear about what I was hoping to resolve is:

• Descheduler sees a node is Under Utilized and wants to evict pods
• Pods are descheduled from nodes
• A specific pod (such as one in a stateful set or with a PVC) has a volume associated with it however there is no excess capacity on any nodes in the same fault domain (AWS AZ, Azure Zone, etc.) so the K8 Scheduler is unable at actually place the Pod again

At this point either a smart cluster autoscaler or operator would need to get involved to either uncordon the node or create a new node in the proper AZ to provide adequate headroom to place the pod again.

[ravisantoshgudimetla]

Thanks for providing the entire use-case.

Pods are descheduled from nodes

I believe this shouldn't happen in first place.

[fejta-bot]

Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle stale

[fejta-bot]

Stale issues rot after 30d of inactivity.
Mark the issue as fresh with /remove-lifecycle rotten.
Rotten issues close after an additional 30d of inactivity.

If this issue is safe to close now please do so with /close.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/lifecycle rotten

[fejta-bot]

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

[k8s-ci-robot]

@fejta-bot: Closing this issue.

In response to this:

Rotten issues close after 30d of inactivity.
Reopen the issue with /reopen.
Mark the issue as fresh with /remove-lifecycle rotten.

Send feedback to sig-testing, kubernetes/test-infra and/or fejta.
/close

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.