-
Bug
-
Resolution: Unresolved
-
Major
-
None
-
4.21
-
Quality / Stability / Reliability
-
False
-
-
None
-
None
-
None
-
None
-
None
-
Approved
-
CORENET Sprint 280
-
1
-
None
-
None
-
None
-
None
-
None
-
None
-
None
Sippy AI-assisted description; please review details for accuracy.
Filed from: Test Regression Details
Test Name
[sig-network] Conntrack should be able to preserve UDP traffic when server pod cycles for a ClusterIP service and client is hostNetwork [Suite:openshift/conformance/parallel ] [Suite:k8s]
Brief Overview
Extreme regression detected. Fishers Exact probability of a regression: 100.00%. Test pass rate dropped from 100.00% to 85.00%.
Statistics Section
Release: 4.21
Time Period: 2025-11-03T00:00:00Z to 2025-11-10T20:00:00Z
Success Rate: 85.0%
Successes: 17
Failures: 3
Flakes: 0
Release: 4.20
Time Period: 2025-10-11T00:00:00Z to 2025-11-10T20:00:00Z
Success Rate: 100.0%
Successes: 105
Failures: 0
Flakes: 0
Sample Failure Outputs
No specific error messages or stack traces were found in the e2e-test logs for the failed job runs using the current analysis.
Links to Relevant Jobs
- periodic-ci-openshift-release-master-nightly-4.21-e2e-vsphere-ovn-techpreview
- periodic-ci-openshift-release-master-nightly-4.21-e2e-vsphere-ovn-techpreview
- periodic-ci-openshift-release-master-nightly-4.21-e2e-vsphere-ovn-techpreview
Patterns and Insights
This is an existing test that has regressed significantly. The test had a 100% success rate in the base release (4.20) but has dropped to 85% in the sample release (4.21), indicating a clear regression. The failures appear to be consistent within the `periodic-ci-openshift-release-master-nightly-4.21-e2e-vsphere-ovn-techpreview` job. There were no flakes observed in either the base or sample periods, suggesting a direct failure rather than intermittent flakiness.
Filed by: weliang@redhat.com
- links to