Skip to content

OCPBUGS-84969: fix(e2e): wait for Karpenter node cleanup in parallel tests to prevent vCPU flake #8414

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
openshift-merge-bot merged 1 commit into from
May 5, 2026
Merged

OCPBUGS-84969: fix(e2e): wait for Karpenter node cleanup in parallel tests to prevent vCPU flake #8414

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 9 additions & 1 deletion test/e2e/karpenter_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -672,9 +672,11 @@ func testNodeClassVersionField(ctx context.Context, mgtClient, guestClient crcli
instanceID, instance.MetadataOptions.HttpTokens, instance.MetadataOptions.HttpEndpoint, *instance.MetadataOptions.HttpPutResponseHopLimit)
}

// Trigger cleanup; the t.Cleanup handles final deletion.
// Trigger cleanup and wait for nodes to fully terminate so stale
// NodeClaims don't leak vCPUs into subsequent sequential tests.
g.Expect(guestClient.Delete(ctx, testWorkLoads)).To(Succeed())
g.Expect(guestClient.Delete(ctx, testNodePool)).To(Succeed())
_ = e2eutil.WaitForReadyNodesByLabels(t, ctx, guestClient, hostedCluster.Spec.Platform.Type, 0, testNodeLabels)

// Verify that a version exceeding the allowed n-3 skew sets SupportedVersionSkew=False.
skewMajor, skewMinor, err := supportedversion.PreviousMinorVersion(cpVersion, 4)
Expand Down Expand Up @@ -887,6 +889,12 @@ func testCapacityReservation(ctx context.Context, mgtClient, guestClient crclien
g.Expect(aws.ToString(instance.CapacityReservationId)).To(Equal(crID),
"instance %s should have been launched into capacity reservation %s", instanceID, crID)
t.Logf("Instance %s correctly launched into capacity reservation %s", instanceID, crID)

// Delete workload and NodePool, then wait for nodes to fully terminate
// so stale NodeClaims don't leak vCPUs into subsequent sequential tests.
g.Expect(guestClient.Delete(ctx, crWorkload)).To(Succeed())
g.Expect(guestClient.Delete(ctx, crNodePool)).To(Succeed())
_ = e2eutil.WaitForReadyNodesByLabels(t, ctx, guestClient, hostedCluster.Spec.Platform.Type, 0, crNodeLabels)
}
}

Expand Down