-
Bug
-
Resolution: Done
-
Major
-
premerge
-
Quality / Stability / Reliability
-
False
-
-
None
-
None
-
None
-
None
-
None
-
None
-
OCP Node Sprint 273 (Green)
-
1
-
In Progress
-
Release Note Not Required
-
None
-
None
-
None
-
None
-
None
Description of problem:
Kueue will manage deployments that are labeled with Kueue. But if you try to submit a deployment with no kueue name, I expect admission will work without any issue. In this case, replica sets fails to create the pods. status: conditions: - lastTransitionTime: "2025-06-23T15:32:43Z" message: 'Internal error occurred: failed calling webhook "mpod.kb.io": failed to call webhook: Post "https://kueue-webhook-service.openshift-kueue-operator.svc:443/mutate--v1-pod?timeout=10s": context deadline exceeded'
Version-Release number of selected component (if applicable):
Kueue 0.2.0
How reproducible:
Everytime
Steps to Reproduce:
Create a labelled namespace: apiVersion: v1 kind: Namespace metadata: labels: kubernetes.io/metadata.name: kueue-test kueue.openshift.io/managed: "true" kueue.io/managed: "true" name: kueue-test Submit a deployment that does not use kueue into this namespace: apiVersion: apps/v1 kind: Deployment metadata: namespace: kueue-test name: deployment-no-kueue spec: selector: matchLabels: app: name replicas: 3 template: metadata: labels: app: name spec: containers: - name: container image: busybox command: ["/bin/sh", "-c"] # Define the command to execute args: ["sleep 100000"] ports: - containerPort: 8080 protocol: TCP env: [] imagePullSecrets: [] strategy: type: RollingUpdate rollingUpdate: maxSurge: 25% maxUnavailable: 25% paused: false Deployment replica set does not create pods. If you describe the replicaset that the deployment creates you can see it fails due to some webhook failure. Conditions: Type Status Reason ---- ------ ------ ReplicaFailure True FailedCreate Events: Type Reason Age From Message ---- ------ ---- ---- ------- Warning FailedCreate 41s (x20 over 19m) replicaset-controller Error creating: Internal error occurred: failed calling webhook "mpod.kb.io": failed to call webhook: Post "https://kueue-webhook-service.openshift-kueue-operator.svc:443/mutate--v1-pod?timeout=10s": context deadline exceeded
Additional info:
Kueue seems to work with deployments that are labelled.
- is related to
-
OCPBUGS-56429 Can't schedule plain pods through kueue
-
- Closed
-