kubernetes: [upgrade test failure] Several network partition test failing
After we fixed upgrade test infra issue (https://github.com/kubernetes/kubernetes/issues/47379#issuecomment-309912047), now we get better signals on real failure: https://k8s-gubernator.appspot.com/build/kubernetes-jenkins/logs/ci-kubernetes-e2e-gce-1-6-1-7-upgrade-cluster/55 The following Network Partition tests are failed:
[k8s.io] Network Partition [Disruptive] [Slow] [k8s.io] [Job] should create new pods when node is partitioned
[k8s.io] Network Partition [Disruptive] [Slow] [k8s.io] [ReplicationController] should eagerly create replacement pod during network partition when termination grace is non-zero
[k8s.io] Network Partition [Disruptive] [Slow] [k8s.io] [ReplicationController] should recreate pods scheduled on the unreachable node AND allow scheduling of pods on a node after it rejoins the cluster
[k8s.io] Network Partition [Disruptive] [Slow] [k8s.io] [StatefulSet] should come back up if node goes down [Slow] [Disruptive]
[k8s.io] Network Partition [Disruptive] [Slow] [k8s.io] [StatefulSet] should not reschedule stateful pods if there is a network partition [Slow] [Disruptive]
About this issue
- Original URL
- State: closed
- Created 7 years ago
- Comments: 49 (46 by maintainers)
Triaging and looking into it, will have an update by tomorrow.
@kow3ns non-blocking, failing tests create a lot of overhead for the release team, particularly the test signal lead. If they’re non-blocking, can they be disabled on upgrade tests until someone has a chance to fix them?