Join GitHub today
GitHub is home to over 31 million developers working together to host and review code, manage projects, and build software together.Sign up
[Flaky test] [k8s.io] Pods should support pod readiness gates [NodeFeature:PodReadinessGate] in ci-kubernetes-e2e-gce-cos-k8sbeta-default #75355
Which jobs are failing:
Which test(s) are failing:
Since when has it been failing:
Reason for failure:
It looks like updating the status after the patch to the readiness gate takes longer than expected.
Looking at this failure: https://prow.k8s.io/view/gcs/kubernetes-jenkins/logs/ci-kubernetes-e2e-gce-cos-k8sbeta-default/9858
From the test logs:
From the kubelet logs:
The pod goes to Ready at 1:53:12, and we stopped checking at 1:53:05
This looks expected, based on the description of the maxReadyStatusUpdateTolerance parameter in the test.
I don't think this should be release-blocking.
When this tests was added (2 or 3 release back), I remember the 10 second waiting timeout maxReadyStatusUpdateTolerance is enough for kubelet to react and update the pod ready condition.
IIRC, kubelet reconciles pod status every 15s. But if pod readiness gate condition is updated, kubelet should reconcile ASAP skipping the 15s period.
There might be a bug here.