Restrict scaling out MPIJob

[tenzen-y]

What would you like to be added:
I would like to restrict scaling out MPIJob in admission webhooks when clusterQueue lacks resources.

Why is this needed:
MPIJob supports scaling out workers by updating replicas.
Currently, if we scale out MPIJob that uses under-resourced clusterQueue, kueue sets suspend=true to MPIJob, and then the mpi-operator kills all MPIJob replicas.

However, I don't think that users want to kill all MPI replicas in that case.

Completion requirements:

This enhancement requires the following artifacts:

• Design doc
• API change
• Docs update

The artifacts should be linked in subsequent comments.

[tenzen-y]

/cc @mimowo @alculquicondor

[alculquicondor]

We discussed this before in the context of Job #467

I think it makes sense to just restrict this via webhooks while we don't have the support.

@mwielgus do you think this is still not worth since we will eventually remove this piece of code?

[tenzen-y]

We discussed this before in the context of Job #467

I think it makes sense to just restrict this via webhooks while we don't have the support.

@mwielgus do you think this is still not worth since we will eventually remove this piece of code?

It's reasonable. For the first step, we can introduce webhooks for both batch/job and mpijob.

[k8s-triage-robot]

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

This bot triages un-triaged issues 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 as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

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

/lifecycle stale

[tenzen-y]

/remove-lifecycle stale

[k8s-triage-robot]

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

This bot triages un-triaged issues 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 as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

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

/lifecycle stale

[tenzen-y]

I think this still could be possible although we can close this once #77 is realized.

/remove-lifecycle stale

[k8s-triage-robot]

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

This bot triages un-triaged issues 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 as fresh with /remove-lifecycle stale
• Close this issue with /close
• Offer to help out with Issue Triage

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

/lifecycle stale

[tenzen-y]

This would be resolved by #77.
Considering the current effort for #77, we can close this issue for now.
/close

[k8s-ci-robot]

@tenzen-y: Closing this issue.

In response to this:

This would be resolved by #77.
Considering the current effort for #77, we can close this issue for now.
/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.