-
Bug
-
Resolution: Done-Errata
-
Normal
-
4.13, 4.12, 4.14
-
None
Description of problem:
There are several testcases in conformance testsuite that are failing due to openshift-multus configuration.
We are running conformance testsuite as part of our Openshift on Openstack CI. We use that just to confirm correct functionality of the cluster. The command we are using to run the test suite is:
openshift-tests run --provider '{\"type\":\"openstack\"}' openshift/conformance/parallel
The name of the tests that failed are:
1. sig-arch] Managed cluster should ensure platform components have system-* priority class associated [Suite:openshift/conformance/parallel]
Reason is:
6 pods found with invalid priority class (should be openshift-user-critical or begin with system-): openshift-multus/whereabouts-reconciler-6q6h7 (currently "") openshift-multus/whereabouts-reconciler-87dwn (currently "") openshift-multus/whereabouts-reconciler-fvhwv (currently "") openshift-multus/whereabouts-reconciler-h68h5 (currently "") openshift-multus/whereabouts-reconciler-nlz59 (currently "") openshift-multus/whereabouts-reconciler-xsch6 (currently "")
2. [sig-arch] Managed cluster should only include cluster daemonsets that have maxUnavailable or maxSurge update of 10 percent or maxUnavailable of 33 percent [Suite:openshift/conformance/parallel]
Reason is:
fail [github.com/openshift/origin/test/extended/operators/daemon_set.go:105]: Sep 23 16:12:15.283: Daemonsets found that do not meet platform requirements for update strategy: expected daemonset openshift-multus/whereabouts-reconciler to have maxUnavailable 10% or 33% (see comment) instead of 1, or maxSurge 10% instead of 0 Ginkgo exit error 1: exit with code 1
3.[sig-arch] Managed cluster should set requests but not limits [Suite:openshift/conformance/parallel]
Reason is:
fail [github.com/openshift/origin/test/extended/operators/resources.go:196]: Sep 23 16:12:17.489: Pods in platform namespaces are not following resource request/limit rules or do not have an exception granted: apps/v1/DaemonSet/openshift-multus/whereabouts-reconciler/container/whereabouts defines a limit on cpu of 50m which is not allowed (rule: "apps/v1/DaemonSet/openshift-multus/whereabouts-reconciler/container/whereabouts/limit[cpu]") apps/v1/DaemonSet/openshift-multus/whereabouts-reconciler/container/whereabouts defines a limit on memory of 100Mi which is not allowed (rule: "apps/v1/DaemonSet/openshift-multus/whereabouts-reconciler/container/whereabouts/limit[memory]") Ginkgo exit error 1: exit with code 1
4. [sig-node][apigroup:config.openshift.io] CPU Partitioning cluster platform workloads should be annotated correctly for DaemonSets [Suite:openshift/conformance/parallel]
Reason is:
fail [github.com/openshift/origin/test/extended/cpu_partitioning/pods.go:159]: Expected
<[]error | len:1, cap:1>: [
<*errors.errorString | 0xc0010fa380>{
s: "daemonset (whereabouts-reconciler) in openshift namespace (openshift-multus) must have pod templates annotated with map[target.workload.openshift.io/management:{\"effect\": \"PreferredDuringScheduling\"}]",
},
]
to be empty
How reproducible: Always
Steps to Reproduce: Run conformance testsuite:
https://github.com/openshift/origin/blob/master/test/extended/README.md
Actual results: Testcases failing
Expected results: Testcases passing
- blocks
-
-
- Closed
-
- duplicates
-
-
- Closed
-
- is cloned by
-
-
- Closed
-
- links to
-
RHEA-2024:0041
OpenShift Container Platform 4.16.z bug fix update
-
RHSA-2023:7198
OpenShift Container Platform 4.15 security update
