KEP-5776: Configurable cAdvisor Metrics Collection

[NahumLitvin]

Summary

This KEP proposes adding a cadvisor configuration section to KubeletConfiguration that allows operators to selectively disable expensive cAdvisor metric collectors. The primary use case is disabling ProcessMetrics collection, which scans /proc for every thread in every container and causes extreme CPU overhead on high-density nodes.

Production Evidence

Testing on EKS 1.31 clusters with ~200 pods/node showed:

• Kubelet CPU: 1332% → 8% (99.4% reduction)
• Node Total CPU: 44% → 13% (70% reduction)
• ~10 cores freed per node

Proposal

• New alpha feature gate: ConfigurableCAdvisorMetrics (target 1.33)
• New KubeletConfiguration.cadvisor.includedMetrics.processMetrics (default true)
• Maintains backward compatibility (all metrics enabled by default)

Related

• CAdvisor stats collection should be configurable. kubernetes#123340 (original issue)
• Implementation PR (pending): kubernetes/kubernetes

/sig node
/kind kep

[k8s-ci-robot]

Welcome @NahumLitvin!

It looks like this is your first PR to kubernetes/enhancements 🎉. Please refer to our pull request process documentation to help your PR have a smooth ride to approval.

You will be prompted by a bot to use commands during the review process. Do not be afraid to follow the prompts! It is okay to experiment. Here is the bot commands documentation.

You can also check if kubernetes/enhancements has its own contribution guidelines.

You may want to refer to our testing guide if you run into trouble with your tests not passing.

If you are having difficulty getting your pull request seen, please follow the recommended escalation practices. Also, for tips and tricks in the contribution process you may want to read the Kubernetes contributor cheat sheet. We want to make sure your contribution gets all the attention it needs!

Thank you, and welcome to Kubernetes. 😃

[k8s-ci-robot]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: NahumLitvin
Once this PR has been reviewed and has the lgtm label, please assign derekwaynecarr for approval. For more information see the Code Review Process.

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

Details

Needs approval from an approver in each of these files:

• keps/sig-node/OWNERS

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

[k8s-ci-robot]

Hi @NahumLitvin. Thanks for your PR.

I'm waiting for a kubernetes member to verify that this patch is reasonable to test. If it is, they should reply with /ok-to-test on its own line. Until that is done, I will not automatically test new commits in this PR, but the usual testing commands by org members will still work. Regular contributors should join the org to skip this step.

Once the patch is verified, the new status will be reflected by the ok-to-test label.

I understand the commands that are listed here.

Details

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-sigs/prow repository.

[SergeyKanzhelev]

you need to have a KEP issue first

/sig instrumentation

[rexagod]

Per-component metric manipulations should ideally happen through a common interface: kubernetes/kubernetes#131572, for multiple reasons (common underlying DX expectations, ability to ship faster as most changes can be made in component-base and trickle down to components, etc.).

For component-specific behaviors, it'd be good to have those gates and their configurations exposed through the same interface, albeit the implementation currently does not account for that, and will need to be updated to accomodate for per-component behaviors.

This is also something that should be discussed first because this effectively goes in the other direction, as currently all options exposed in the Metrics API as well as their behaviors all live in component-base.