OCPBUGS-62270: UPSTREAM: <carry>: Skip CPU resource status for workload-pinned pods #2490

harche · 2025-10-16T16:56:16Z

What type of PR is this?

/kind bug

What this PR does / why we need it:

This PR fixes CPU double-counting for workload-partitioned (managed) pods when
InPlacePodVerticalScaling is enabled (default in Kubernetes 1.33+).

When workload partitioning is enabled, pods have CPU resources specified as
management.workload.openshift.io/cores in spec.requests. However, the kubelet also populates cpu
in status.containerStatuses[].resources.requests by reading from cgroups. This causes the scheduler
to count CPU twice:

From management.workload.openshift.io/cores in pod spec
From cpu in pod status

This leads to scheduler seeing artificially inflated CPU usage, resulting in false "Insufficient
cpu" errors when scheduling new pods.

Which issue(s) this PR fixes:

Fixes # https://issues.redhat.com/browse/OCPBUGS-62270

Special notes for your reviewer:

Does this PR introduce a user-facing change?

Fixed CPU double-counting issue for workload-partitioned pods that caused false "Insufficient cpu"
  scheduling errors when InPlacePodVerticalScaling is enabled.

Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

openshift-ci-robot · 2025-10-16T16:56:26Z

@harche: the contents of this pull request could not be automatically validated.

The following commits could not be validated and must be approved by a top-level approver:

edb8bd5|Skip CPU resource status for workload-pinned pods: does not specify an upstream backport in the commit message

Comment /validate-backports to re-evaluate validity of the upstream PRs, for example when they are merged upstream.

openshift-ci · 2025-10-16T16:57:16Z

[APPROVALNOTIFIER] This PR is NOT APPROVED

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

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

Needs approval from an approver in each of these files:

pkg/kubelet/DOWNSTREAM_OWNERS

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

haircommander · 2025-10-16T17:01:35Z

pkg/kubelet/kuberuntime/kuberuntime_container.go

 		return nil, nil, err
 	}

+	isManagedPod := utilfeature.DefaultFeatureGate.Enabled(features.InPlacePodVerticalScaling) && managed.IsManagedPodFromRuntimeService(ctx, m.runtimeService, activePodSandboxID)


I don't think we need to check for feature gate here. cStatus.Resources != nil will only be true if the feature is enabled

Addressed, thanks.

haircommander · 2025-10-16T17:03:05Z

pkg/kubelet/managed/managed.go


+// IsPodSandboxManagedPod checks if a pod sandbox belongs to a managed pod
+// by looking for workload pinning annotations.
+func IsPodSandboxManagedPod(sandboxAnnotations map[string]string) bool {


nit: can this be defined below where it's called in IsManagedPodFromRuntimeService

Addressed, thanks.

openshift-ci-robot · 2025-10-16T17:34:43Z

@harche: the contents of this pull request could not be automatically validated.

The following commits could not be validated and must be approved by a top-level approver:

3fc7c95|Skip CPU resource status for workload-pinned pods: does not specify an upstream backport in the commit message

Comment /validate-backports to re-evaluate validity of the upstream PRs, for example when they are merged upstream.

openshift-ci-robot · 2025-10-20T17:46:14Z

@harche: the contents of this pull request could not be automatically validated.

The following commits could not be validated and must be approved by a top-level approver:

ff0c855|Skip CPU resource status for workload-pinned pods: does not specify an upstream backport in the commit message

Comment /validate-backports to re-evaluate validity of the upstream PRs, for example when they are merged upstream.

openshift-ci-robot · 2025-10-20T17:47:52Z

@harche: This pull request references Jira Issue OCPBUGS-62270, which is invalid:

expected the bug to target the "4.21.0" version, but no target version was set

Comment /jira refresh to re-evaluate validity if changes to the Jira bug are made, or edit the title of this pull request to link to a different bug.

The bug has been updated to refer to the pull request using the external bug tracker.

In response to this:

What type of PR is this?

What this PR does / why we need it:

Which issue(s) this PR fixes:

Fixes #

Special notes for your reviewer:

Does this PR introduce a user-facing change?
Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

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 openshift-eng/jira-lifecycle-plugin repository.

Signed-off-by: Harshal Patil <[email protected]>

openshift-ci-robot · 2025-10-20T17:49:51Z

@harche: the contents of this pull request could not be automatically validated.

The following commits could not be validated and must be approved by a top-level approver:

9382bbe|UPSTREAM: : Skip CPU resource status for workload-pinned pods: does not specify an upstream backport in the commit message

Comment /validate-backports to re-evaluate validity of the upstream PRs, for example when they are merged upstream.

openshift-ci-robot · 2025-10-20T17:53:10Z

@harche: This pull request references Jira Issue OCPBUGS-62270, which is invalid:

expected the bug to target the "4.21.0" version, but no target version was set

Comment /jira refresh to re-evaluate validity if changes to the Jira bug are made, or edit the title of this pull request to link to a different bug.

The bug has been updated to refer to the pull request using the external bug tracker.

In response to this:

What type of PR is this?

/kind bug

What this PR does / why we need it:

This PR fixes CPU double-counting for workload-partitioned (managed) pods when
InPlacePodVerticalScaling is enabled (default in Kubernetes 1.33+).

When workload partitioning is enabled, pods have CPU resources specified as
management.workload.openshift.io/cores in spec.requests. However, the kubelet also populates cpu
in status.containerStatuses[].resources.requests by reading from cgroups. This causes the scheduler
to count CPU twice:

From management.workload.openshift.io/cores in pod spec

From cpu in pod status

This leads to scheduler seeing artificially inflated CPU usage, resulting in false "Insufficient
cpu" errors when scheduling new pods.

Which issue(s) this PR fixes:

Fixes # https://issues.redhat.com/browse/OCPBUGS-62270

Special notes for your reviewer:

Does this PR introduce a user-facing change?
Fixed CPU double-counting issue for workload-partitioned pods that caused false "Insufficient cpu"
 scheduling errors when InPlacePodVerticalScaling is enabled.
Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

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 openshift-eng/jira-lifecycle-plugin repository.

harche · 2025-10-20T17:54:45Z

/jira refresh

openshift-ci-robot · 2025-10-20T17:54:54Z

@harche: This pull request references Jira Issue OCPBUGS-62270, which is valid. The bug has been moved to the POST state.

3 validation(s) were run on this bug

bug is open, matching expected state (open)
bug target version (4.21.0) matches configured target version for branch (4.21.0)
bug is in the state ASSIGNED, which is one of the valid states (NEW, ASSIGNED, POST)

In response to this:

/jira refresh

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 openshift-eng/jira-lifecycle-plugin repository.

harche · 2025-10-20T17:59:06Z

/hold for another round of testing with real cluster after resolving merge conflicts during rebasing.

harche · 2025-10-20T19:01:13Z

/retest-required

harche · 2025-10-21T02:35:37Z

/retest-required

openshift-ci · 2025-10-21T06:59:55Z

@harche: The following tests failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/unit	`9382bbe`	link	true	`/test unit`
ci/prow/okd-scos-images	`9382bbe`	link	true	`/test okd-scos-images`
ci/prow/e2e-aws-ovn-fips	`9382bbe`	link	true	`/test e2e-aws-ovn-fips`
ci/prow/okd-scos-e2e-aws-ovn	`9382bbe`	link	false	`/test okd-scos-e2e-aws-ovn`
ci/prow/e2e-aws-crun-wasm	`9382bbe`	link	true	`/test e2e-aws-crun-wasm`
ci/prow/verify-commits	`9382bbe`	link	true	`/test verify-commits`
ci/prow/k8s-e2e-conformance-aws	`9382bbe`	link	true	`/test k8s-e2e-conformance-aws`
ci/prow/e2e-aws-ovn-crun	`9382bbe`	link	true	`/test e2e-aws-ovn-crun`
ci/prow/verify	`9382bbe`	link	true	`/test verify`
ci/prow/e2e-aws-ovn-hypershift	`9382bbe`	link	true	`/test e2e-aws-ovn-hypershift`
ci/prow/e2e-aws-ovn-cgroupsv2	`9382bbe`	link	true	`/test e2e-aws-ovn-cgroupsv2`
ci/prow/images	`9382bbe`	link	true	`/test images`
ci/prow/e2e-aws-ovn-serial	`9382bbe`	link	true	`/test e2e-aws-ovn-serial`
ci/prow/integration	`9382bbe`	link	true	`/test integration`
ci/prow/e2e-aws-ovn-runc	`9382bbe`	link	true	`/test e2e-aws-ovn-runc`

Full PR test history. Your PR dashboard.

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. I understand the commands that are listed here.

openshift-ci-robot added the backports/unvalidated-commits Indicates that not all commits come to merged upstream PRs. label Oct 16, 2025

openshift-ci bot added the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label Oct 16, 2025

openshift-ci bot requested review from mrunalp and rphillips October 16, 2025 16:57

haircommander reviewed Oct 16, 2025

View reviewed changes

harche force-pushed the OCPBUGS-61220 branch from edb8bd5 to 3fc7c95 Compare October 16, 2025 17:34

openshift-merge-robot added the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Oct 19, 2025

harche force-pushed the OCPBUGS-61220 branch from 3fc7c95 to ff0c855 Compare October 20, 2025 17:46

openshift-merge-robot removed the needs-rebase Indicates a PR cannot be merged because it has merge conflicts with HEAD. label Oct 20, 2025

harche changed the title ~~[WIP] Skip CPU resource status for workload-pinned pods~~ OCPBUGS-62270: Skip CPU resource status for workload-pinned pods Oct 20, 2025

openshift-ci bot removed the do-not-merge/work-in-progress Indicates that a PR should not merge because it is a work in progress. label Oct 20, 2025

openshift-ci-robot added jira/severity-critical Referenced Jira bug's severity is critical for the branch this PR is targeting. jira/valid-reference Indicates that this PR references a valid Jira ticket of any type. labels Oct 20, 2025

openshift-ci-robot added the jira/invalid-bug Indicates that a referenced Jira bug is invalid for the branch this PR is targeting. label Oct 20, 2025

harche changed the title ~~OCPBUGS-62270: Skip CPU resource status for workload-pinned pods~~ OCPBUGS-62270: UPSTREAM: <carry>: Skip CPU resource status for workload-pinned pods Oct 20, 2025

UPSTREAM: <carry>: Skip CPU resource status for workload-pinned pods

9382bbe

Signed-off-by: Harshal Patil <[email protected]>

harche force-pushed the OCPBUGS-61220 branch from ff0c855 to 9382bbe Compare October 20, 2025 17:49

openshift-ci-robot added jira/valid-bug Indicates that a referenced Jira bug is valid for the branch this PR is targeting. and removed jira/invalid-bug Indicates that a referenced Jira bug is invalid for the branch this PR is targeting. labels Oct 20, 2025

openshift-ci bot added the do-not-merge/hold Indicates that a PR should not merge because someone has issued a /hold command. label Oct 20, 2025

OCPBUGS-62270: UPSTREAM: <carry>: Skip CPU resource status for workload-pinned pods #2490

Are you sure you want to change the base?

OCPBUGS-62270: UPSTREAM: <carry>: Skip CPU resource status for workload-pinned pods #2490

Conversation

harche commented Oct 16, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What type of PR is this?

What this PR does / why we need it:

Which issue(s) this PR fixes:

Special notes for your reviewer:

Does this PR introduce a user-facing change?

Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

Uh oh!

openshift-ci-robot commented Oct 16, 2025

Uh oh!

openshift-ci bot commented Oct 16, 2025

Uh oh!

haircommander Oct 16, 2025

Choose a reason for hiding this comment

Uh oh!

harche Oct 16, 2025

Choose a reason for hiding this comment

Uh oh!

haircommander Oct 16, 2025

Choose a reason for hiding this comment

Uh oh!

harche Oct 20, 2025

Choose a reason for hiding this comment

Uh oh!

openshift-ci-robot commented Oct 16, 2025

Uh oh!

openshift-ci-robot commented Oct 20, 2025

Uh oh!

openshift-ci-robot commented Oct 20, 2025

What type of PR is this?

What this PR does / why we need it:

Which issue(s) this PR fixes:

Special notes for your reviewer:

Does this PR introduce a user-facing change?

Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

Uh oh!

openshift-ci-robot commented Oct 20, 2025

Uh oh!

openshift-ci-robot commented Oct 20, 2025

What type of PR is this?

What this PR does / why we need it:

Which issue(s) this PR fixes:

Special notes for your reviewer:

Does this PR introduce a user-facing change?

Additional documentation e.g., KEPs (Kubernetes Enhancement Proposals), usage docs, etc.:

Uh oh!

harche commented Oct 20, 2025

Uh oh!

openshift-ci-robot commented Oct 20, 2025

Uh oh!

harche commented Oct 20, 2025

Uh oh!

harche commented Oct 20, 2025

Uh oh!

harche commented Oct 21, 2025

Uh oh!

openshift-ci bot commented Oct 21, 2025

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

harche commented Oct 16, 2025 •

edited

Loading