descheduler Add new strategy ‘RemoveContainersNotReadyPods’

When the node network or CRI is error ， the init container is always initialized when deploy app .

It may be caused by image download failure or insufficient IP , in this case, pods and containers will always be in this state

But the pod is not in the failed state, if you try to reschedule to another node in this case, the image may be downloaded successfully and the IP may be sufficient，the container will be in a ready state or succeed.

I think it is necessary to add a non-ready policy for containers, because in this case rescheduling may make the container run successfully

for example：

apiVersion: "descheduler/v1alpha1"
kind: "DeschedulerPolicy"
strategies:
  "RemoveContainersNotReadyPods":
     enabled: true
     params:
       containersNotReadyPods:
         reasons:
         - "ContainersNotReady"
         includingInitContainers: true
         minPodLifetimeSeconds: 600

May 16 '22 13:05 wxuedong

Would this need to be an entirely new strategy? Or is it just a status phase we can parse and check in PodLifetime or RemoveFailedPods?

May 16 '22 14:05 damemi

Would this need to be an entirely new strategy? Or is it just a status phase we can parse and check in PodLifetime or RemoveFailedPods?

RemoveFailedPods currently only support pods phase is failed ,if support other phase(Running and Pending) or PodLifetime support pods phase is failed and parameter support filter reasons .

In this case, there is no need to add new strategies to solve the above problems

May 17 '22 00:05 wxuedong

/assign

I can create a PR once #834 is reviewed/approved/merged

Jun 05 '22 03:06 a7i

On second though, PodLifeTime is likely not a good candidate for this given that it is included in conditions at some point in time.

For example, take the following example:

  conditions:
    - type: Initialized
      status: 'True'
      lastProbeTime: null
      lastTransitionTime: '2019-02-13T20:31:37Z'
    - type: Ready
      status: 'False'
      lastProbeTime: null
      lastTransitionTime: '2019-02-13T20:31:37Z'
      reason: ContainersNotReady
      message: 'containers with unready status: [...]'
    - type: ContainersReady
      status: 'False'
      lastProbeTime: null
      lastTransitionTime: '2019-02-13T20:31:37Z'
      reason: ContainersNotReady
      message: 'containers with unready status: [...]'
    - type: PodScheduled
      status: 'True'
      lastProbeTime: null
      lastTransitionTime: '2019-02-13T20:31:37Z'

The container was not ready at some point in time, but eventually, it was scheduled.

Jun 25 '22 19:06 a7i

@a7i so this is only reported through the conditions? We could probably do some logic to check that it's the most recent condition. But I think that could be enough for a new strategy (RemovePodsWithConditions?)

Jul 01 '22 17:07 damemi

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

Sep 29 '22 18:09 k8s-triage-robot

/remove-lifecycle stale /assign

@damemi this would be a good opportunity for me to create a new plugin and see how that goes.

Sep 29 '22 18:09 a7i

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

Dec 28 '22 18:12 k8s-triage-robot

/remove-lifecycle stale

Dec 28 '22 19:12 a7i

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

Mar 28 '23 19:03 k8s-triage-robot

/remove-lifecycle stale

Mar 28 '23 19:03 a7i

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

Jun 26 '23 20:06 k8s-triage-robot

/remove-lifecycle stale

Jun 27 '23 02:06 a7i

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

Jan 23 '24 09:01 k8s-triage-robot

The Kubernetes project currently lacks enough active 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 rotten
Close this issue with /close
Offer to help out with Issue Triage

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

/lifecycle rotten

Feb 22 '24 09:02 k8s-triage-robot

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

This bot triages 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:

Reopen this issue with /reopen
Mark this issue as fresh with /remove-lifecycle rotten
Offer to help out with Issue Triage

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

/close not-planned

Mar 23 '24 10:03 k8s-triage-robot

@k8s-triage-robot: Closing this issue, marking it as "Not Planned".

In response to this:

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

This bot triages 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:

Reopen this issue with /reopen

Mark this issue as fresh with /remove-lifecycle rotten

Offer to help out with Issue Triage

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

/close not-planned

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.

Mar 23 '24 10:03 k8s-ci-robot

descheduler descheduler copied to clipboard

Add new strategy ‘RemoveContainersNotReadyPods’

descheduler
descheduler copied to clipboard