DRA extended resource quota by yliaog · Pull Request #134210 · kubernetes/kubernetes

yliaog · 2025-09-22T23:59:11Z

What type of PR is this?

/kind feature

What this PR does / why we need it:

adjusts DRA extended resource quota to include devices usages from regular resource claims

Which issue(s) this PR is related to:

#133671

Special notes for your reviewer:

Does this PR introduce a user-facing change?

ResourceQuota now counts device class requests within a ResourceClaim object as consuming two additional quotas when the DRAExtendedResource feature is enabled:
- `requests.deviceclass.resource.k8s.io/<deviceclass>` with a quantity equal to the worst case count of devices requested
- requests for device classes that map to an extended resource consume `requests.<extended resource name>`

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

k8s-triage-robot · 2025-09-23T00:41:50Z

This PR may require API review.

If so, when the changes are ready, complete the pre-review checklist and request an API review.

Status of requested reviews is tracked in the API Review project.

yliaog · 2025-11-04T06:23:21Z

@liggitt this PR is decoupled from #134882, it does not rely on the specifics of how device requests or <container, resource, request> mappings are created. It can be reviewed separately. Please take a look.

yliaog · 2025-11-05T15:16:55Z

/retest

liggitt · 2025-11-05T19:24:50Z

lgtm, go ahead and squash for merge

pohly

Some nits on the DRA part. Non-blocking, can also be a future cleanup PR.

pohly · 2025-11-05T20:25:46Z

+func (c *ExtendedResourceCache) updateClassMapping(deviceClass *resourceapi.DeviceClass) {
+	if deviceClass == nil {
+		return
+	}


Why the nil check?

As far as I can tell it's always called with a valid instance.

Might be worth removing also for the other methods, I missed that earlier.

removed the nil check in this and other methods

pohly · 2025-11-05T20:28:09Z

 	// mapping maps extended resource name to device class name
 	mapping map[v1.ResourceName]string
+	// classMapping maps device class name to extended resource name
+	classMapping map[string]string


Hmm, now the naming of these "mappings" is a bit unfortunate.

How about this?

mapping -> resourceName2class
classMapping -> class2ResourceName

The log message also seem a bit ambiguous.

renamed to resourceName2class and class2ResourceName

pohly · 2025-11-05T20:28:38Z

+func (c *ExtendedResourceCache) removeClassMapping(deviceClass *resourceapi.DeviceClass) {
+	if deviceClass == nil {
+		return
+	}


pohly · 2025-11-05T20:32:00Z

+			pod := b.Pod()
+			res := v1.ResourceList{}
+
+			// b.ExtendedResourceName(0) is added to the deivce class with name: b.ClassName()+"0"


Suggested change

// b.ExtendedResourceName(0) is added to the deivce class with name: b.ClassName()+"0"

// b.ExtendedResourceName(0) is added to the device class with name: b.ClassName()+"0"

…gular resource claims

yliaog · 2025-11-05T23:32:37Z

@liggitt squashed the commits, PTAL

yliaog · 2025-11-06T00:42:24Z

/retest

liggitt · 2025-11-06T01:25:34Z

/lgtm
/approve
/sig scheduling api-machinery node

k8s-ci-robot · 2025-11-06T01:25:41Z

LGTM label has been added.

Details

Git tree hash: 416626f4f143ad14d618f284299a1445545e2cbf

k8s-ci-robot · 2025-11-06T01:25:50Z

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: liggitt, yliaog

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

The pull request process is described here

Details

Needs approval from an approver in each of these files:

~~cmd/kube-controller-manager/OWNERS~~ [liggitt]
~~pkg/controller/OWNERS~~ [liggitt]
~~pkg/controlplane/OWNERS~~ [liggitt]
~~pkg/quota/v1/OWNERS~~ [liggitt]
~~plugin/pkg/admission/OWNERS~~ [liggitt]
~~staging/src/k8s.io/api/OWNERS~~ [liggitt]
~~staging/src/k8s.io/dynamic-resource-allocation/OWNERS~~ [liggitt]
~~test/OWNERS~~ [liggitt]

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

pacoxu · 2025-11-06T01:52:19Z

/retest-required

yliaog · 2025-11-06T02:28:53Z

/retest

yliaog · 2025-11-06T03:07:47Z

/retest

yliaog · 2025-11-06T03:44:48Z

/retest

yliaog · 2025-11-06T05:32:06Z

/retest

yliaog · 2025-11-06T05:53:47Z

/retest

yliaog · 2025-11-06T06:02:04Z

/retest

pohly · 2025-11-06T06:44:24Z

/lgtm
/retest

Known flakes.

pohly · 2025-11-06T06:46:02Z

squashed the commits, PTAL

Please avoid rebases during squashing. You can use git rebase -i --keep-base instead.

k8s-ci-robot · 2025-11-06T07:38:56Z

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

Test name	Commit	Details	Required	Rerun command
pull-kubernetes-e2e-gce	`870062d`	link	unknown	`/test pull-kubernetes-e2e-gce`

Full PR test history. Your PR dashboard. Please help us cut down on flakes by linking to an open issue when you hit one in your PR.

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. I understand the commands that are listed here.

pacoxu · 2025-11-06T07:49:03Z

/test pull-kubernetes-e2e-gce
https://prow.k8s.io/view/gs/kubernetes-ci-logs/pr-logs/pull/134210/pull-kubernetes-e2e-gce/1986325424531050496
Kubernetes e2e suite: [It] [sig-node] FileKeyRef [FeatureGate:EnvFiles] [Beta] should allow ephemeralContainer to consume fileKeyRef, @HirazawaUi could you check if this flake is a new flake after #134414.

pohly · 2025-11-06T11:04:35Z

 		b := drautils.NewBuilder(f, driver)
 		b.UseExtendedResourceName = true

+		ginkgo.It("must run a pod with extended resource with resource quota", func(ctx context.Context) {


This is flaking: #135177

github-project-automation Bot added this to SIG Apps Sep 22, 2025

k8s-ci-robot added the sig/scheduling Categorizes an issue or PR as relevant to SIG Scheduling. label Sep 22, 2025

github-project-automation Bot added this to SIG Scheduling Sep 22, 2025

github-project-automation Bot moved this to Needs Triage in SIG Apps Sep 22, 2025

github-project-automation Bot added this to SIG Node: code and documentation PRs and SIG Node CI/Test Board Sep 22, 2025

k8s-ci-robot added sig/testing Categorizes an issue or PR as relevant to SIG Testing. wg/device-management Categorizes an issue or PR as relevant to WG Device Management. labels Sep 22, 2025

github-project-automation Bot moved this to Triage in SIG Node CI/Test Board Sep 22, 2025

github-project-automation Bot moved this to Triage in SIG Node: code and documentation PRs Sep 22, 2025

github-project-automation Bot added this to Dynamic Resource Allocation Sep 22, 2025

github-project-automation Bot moved this to 🆕 New in Dynamic Resource Allocation Sep 22, 2025

k8s-ci-robot requested review from bart0sh and cici37 September 22, 2025 23:59

yliaog force-pushed the admit_quota branch from e48e9cc to 8790f59 Compare September 23, 2025 16:17

k8s-ci-robot added size/XL Denotes a PR that changes 500-999 lines, ignoring generated files. and removed size/XXL Denotes a PR that changes 1000+ lines, ignoring generated files. labels Nov 3, 2025

liggitt reviewed Nov 4, 2025

View reviewed changes

Comment thread pkg/quota/v1/evaluator/core/registry.go Outdated

Comment thread pkg/quota/v1/evaluator/core/resource_claims.go Outdated

liggitt reviewed Nov 5, 2025

View reviewed changes

Comment thread pkg/quota/v1/evaluator/core/resource_claims.go Outdated

Comment thread pkg/quota/v1/evaluator/core/resource_claims.go Outdated

pohly reviewed Nov 5, 2025

View reviewed changes

adjusts DRA extended resource quota to include devices usages from re…

870062d

…gular resource claims

yliaog force-pushed the admit_quota branch from 6eb9809 to 870062d Compare November 5, 2025 23:30

k8s-ci-robot added the release-note Denotes a PR that will be considered when it comes time to generate release notes. label Nov 6, 2025

pohly mentioned this pull request Nov 6, 2025

Implement scoring for extended resources backed up by DRA #134058

Merged

pohly reviewed Nov 6, 2025

View reviewed changes

yliaog mentioned this pull request Feb 2, 2026

DRA Extended Resource: promote to Beta in 1.36 #135048

Merged

	// b.ExtendedResourceName(0) is added to the deivce class with name: b.ClassName()+"0"
	// b.ExtendedResourceName(0) is added to the device class with name: b.ClassName()+"0"

Conversation

yliaog commented Sep 22, 2025 • edited by liggitt 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 is related to:

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!

k8s-triage-robot commented Sep 23, 2025

Uh oh!

yliaog commented Nov 4, 2025

Uh oh!

Uh oh!

Uh oh!

yliaog commented Nov 5, 2025

Uh oh!

Uh oh!

Uh oh!

liggitt commented Nov 5, 2025

Uh oh!

pohly left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

yliaog commented Nov 5, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

liggitt commented Nov 6, 2025

Uh oh!

k8s-ci-robot commented Nov 6, 2025

Uh oh!

k8s-ci-robot commented Nov 6, 2025

Uh oh!

pacoxu commented Nov 6, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

yliaog commented Nov 6, 2025

Uh oh!

pohly commented Nov 6, 2025

Uh oh!

pohly commented Nov 6, 2025

Uh oh!

k8s-ci-robot commented Nov 6, 2025

Uh oh!

pacoxu commented Nov 6, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

pohly Nov 6, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

yliaog commented Sep 22, 2025 •

edited by liggitt

Loading

pacoxu commented Nov 6, 2025 •

edited

Loading

pohly Nov 6, 2025 •

edited

Loading