backport "daemonset-eviction-for-empty-nodes" and "daemonset-eviction-for-occupied-nodes" from 1.22 to 1.21

[jmenan]

Which component are you using?:

cluster-autoscaler

Is your feature request designed to solve a problem? If so describe the problem this feature should solve.:

With version 1.22 cluster-autoscaler introduces the possibility to manage the eviction of daemonsets differently :

• Add a flag to control DaemonSet eviction on non-empty nodes (Add a flag to control DaemonSet eviction on non-empty nodes #4162).
• Allow DaemonSet pods to opt in/out from eviction (Allow DaemonSet pods to opt in/out from eviction #4172).

But in previous versions this is not possible and may cause bugs.
One of these bugs is perfectly described in this issue: #4337

Describe the solution you'd like.:

backport the options daemonset-eviction-for-empty-nodes and daemonset-eviction-for-occupied-nodes to previous version of cluster-autoscaler ( ≤1.21)

Describe any alternative solutions you've considered.:

Additional context.:

In our case, this causes a bug during a downscale if there is a pod on the node that uses a PVC.
because we use an "aws-ebs-csi-driver" daemonset to manage our PVCs and in this case the "aws-ebs-csi-driver" pod is killed before our application pod and so it loses communication with its PVC

[Shubham82]

If it is valid to do then I would like to work on it.

Hi, @gjtempleton @MaciekPytel @jbartosik your thoughts on this.

[Shubham82]

/assign

[Shubham82]

Fixed it here: #4916

PTAL!

[k8s-triage-robot]

The Kubernetes project currently lacks enough contributors to adequately respond to all issues and PRs.

This bot triages issues and PRs according to the following rules:

• After 90d of inactivity, lifecycle/stale is applied
• After 30d of inactivity since lifecycle/stale was applied, lifecycle/rotten is applied
• After 30d of inactivity since lifecycle/rotten was applied, the issue is closed

You can:

• Mark this issue or PR as fresh with /remove-lifecycle stale
• Mark this issue or PR as rotten with /lifecycle rotten
• Close this issue or PR with /close
• Offer to help out with Issue Triage

Please send feedback to sig-contributor-experience at kubernetes/community.

/lifecycle stale

[Shubham82]

/close
with reference to #4916

[k8s-ci-robot]

@Shubham82: Closing this issue.

In response to this:

/close
with reference to #4916

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.