Update Skew Docs to Fix Compatible Issues between Controller Clients and the API-Server

[network-charles]

This pull request addresses an issue reported by a user regarding the compatibility of the aggregated apiserver library with Kubernetes v1.28 clusters. After updating to version v1.29 of the aggregated apiserver library, the user encountered failures on the Kubernetes v1.28 clusters, accompanied by the following errors:

W0425 14:25:27.110980       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
E0425 14:25:27.111035       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.PriorityLevelConfiguration: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
W0425 14:25:34.571189       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.FlowSchema: the server could not find the requested resource
E0425 14:25:34.571265       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.FlowSchema: failed to list *v1.FlowSchema: the server could not find the requested resource
W0425 14:25:39.776616       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
E0425 14:25:39.776683       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.PriorityLevelConfiguration: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
W0425 14:25:58.606055       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.FlowSchema: the server could not find the requested resource
E0425 14:25:58.606080       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.FlowSchema: failed to list *v1.FlowSchema: the server could not find the requested resource
W0425 14:26:01.846754       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
E0425 14:26:01.846786       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.PriorityLevelConfiguration: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
W0425 14:25:27.211376       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
E0425 14:25:27.211394       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.PriorityLevelConfiguration: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
W0425 14:25:34.498183       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.FlowSchema: the server could not find the requested resource
E0425 14:25:34.498228       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.FlowSchema: failed to list *v1.FlowSchema: the server could not find the requested resource
W0425 14:25:35.545134       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
E0425 14:25:35.545287       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.PriorityLevelConfiguration: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
W0425 14:25:49.677632       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
E0425 14:25:49.677657       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.PriorityLevelConfiguration: failed to list *v1.PriorityLevelConfiguration: the server could not find the requested resource
W0425 14:25:51.325292       1 reflector.go:539] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: failed to list *v1.FlowSchema: the server could not find the requested resource
E0425 14:25:51.325325       1 reflector.go:147] pkg/mod/k8s.io/[email protected]/tools/cache/reflector.go:229: Failed to watch *v1.FlowSchema: failed to list *v1.FlowSchema: the server could not find the requested resource

Changes Made

• Updated the skew docs to instruct users not to have controller clients such as kubelet, kube-controller-manager, and kube-scheduler, as well as aggregated apiservers of a higher version than the kube-apiserver they interact with.

References

Kubernetes/kubernetes#124533 Kubernetes/kubernetes#124655

[netlify[bot]]

✅ Pull request preview available for checking

Built without sensitive environment variables

Name	Link
🔨 Latest commit	906570d987e830c63d7ccfa9d5aa165dc398a83d
🔍 Latest deploy log	https://app.netlify.com/sites/kubernetes-io-main-staging/deploys/663950132f482c00083c4738
😎 Deploy Preview	https://deploy-preview-46109--kubernetes-io-main-staging.netlify.app
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.

---

To edit notification comments on pull requests, go to your Netlify site configuration.

[dims]

/assign @liggitt

[liggitt]

I agree with @liggitt, this document isn't the right place to document this. I will be closing the PR.

I think we can mention in this doc that other components have their own skew requirements and their documentation should be consulted to determine what kube-apiserver versions they support and whether they should be upgraded before / after kube-apiserver.

Separately, the aggregated API servers in various sig projects (metrics-server? custom-metrics-server?) should document their supported kube-apiserver versions. It looks like https://github.com/kubernetes-sigs/metrics-server?tab=readme-ov-file#compatibility-matrix and https://github.com/kubernetes-sigs/custom-metrics-apiserver?tab=readme-ov-file#compatibility need updates.

[network-charles]

Separately, the aggregated API servers in various sig projects (metrics-server? custom-metrics-server?) should document their supported kube-apiserver versions. It looks like https://github.com/kubernetes-sigs/metrics-server?tab=readme-ov-file#compatibility-matrix and https://github.com/kubernetes-sigs/custom-metrics-apiserver?tab=readme-ov-file#compatibility need updates.

Okay, Jordan.

I think we can mention in this doc that other components have their own skew requirements and their documentation should be consulted to determine what kube-apiserver versions they support and whether they should be upgraded before / after kube-apiserver.

The skew documentation contains kube-apiserver version requirements for each component. If there's another documentation, kindly let me know.

[liggitt]

The skew documentation contains kube-apiserver version requirements for each component

This page documents supported skew of core kubernetes components. Mentioning in this doc that supported skew of other components should be considered might prompt admins to consider a category of compatibility that they are not considering today.

Separately, the places where kubernetes sigs are documenting compatibility of aggregated API servers they release should be updated to be explicit about the supported kube-apiserver skew. Issues tracking this request in each repo and referencing this discussion would be a good start. Examples:

• https://github.com/kubernetes-sigs/metrics-server?tab=readme-ov-file#compatibility-matrix
• https://github.com/kubernetes-sigs/custom-metrics-apiserver?tab=readme-ov-file#compatibility

[network-charles]

Mentioning in this doc that supported skew of other components should be considered might prompt admins to consider a category of compatibility that they are not considering today.

That's true, I will add it.

Separately, the places where kubernetes sigs are documenting compatibility of aggregated API servers they release should be updated to be explicit about the supported kube-apiserver skew. Issues tracking this request in each repo and referencing this discussion would be a good start. Examples:

https://github.com/kubernetes-sigs/metrics-server?tab=readme-ov-file#compatibility-matrix https://github.com/kubernetes-sigs/custom-metrics-apiserver?tab=readme-ov-file#compatibility

Great.

[tengqm]

Thanks. /approve

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: tengqm

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• ~~content/en/releases/OWNERS~~ [tengqm]

Approvers can indicate their approval by writing /approve in a comment Approvers can cancel approval by writing /approve cancel in a comment

[sftim]

I recommend cancelling approval until the hold from https://github.com/kubernetes/website/pull/46109#discussion_r1586655268 is addressed. Then we can LGTM and re-approve.

[network-charles]

Thanks for the review, @sftim, the outcome was addressed here and here.

[k8s-triage-robot]

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

This bot triages 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 PR is closed

You can:

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

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

/lifecycle stale

[BenTheElder]

/remove-lifecycle stale