Created attachment 1850091 [details] Create NMState button Description of problem: When installing nmstate-operator from OperatorHub, a button "Create NMState" appears on the web console. However, clicking on the button doesn't create NMState and shows a 404 page. Attached are the picture of the button and the 404 error message. When creating NMState from the Installed Operator NMState tab, the NMState created successfully. Version-Release number of selected component (if applicable): 4.10.0-0.nightly-s390x-2022-01-10-013938 How reproducible: Install nmstate-operator from OperatorHub and click create NMState button. Actual results: 404 Not Found Expected results: NMState should be created.
Created attachment 1850092 [details] 404 Not Found error
i don't see the operator present in 4.10.0-0.nightly-s390x-2022-01-11-033836. my guess is that till 4.10 GAs, we won't have it.
Is there any team(s) to confirm that operators may not be present until 4.10 GA?
I asked Justin in #aos-art for a new build of the 4.10 operators in case that is the issue since the kube rebase has landed. Hi Muhammad, a couple additional pieces of information would be helpful: which operator index was the nmstate operator deployed from? can you add the must gather logs to the bz from the failed state when the error occurs in the web console?
Issue is seen on Power as well using the art index image: quay.io/openshift-release-dev/ocp-release-nightly:iib-int-index-art-operators-4.10 # oc get csv -A | grep nmstate openshift-nmstate kubernetes-nmstate-operator.4.10.0-202112172036 Kubernetes NMState Operator 4.10.0-202112172036 Succeeded
As part of Multi-Arch bug triage, we are re-assigning this bug to the x86 nmstate team for further evaluation. Hi team - this bug has been discovered on s390x and ppc64le, therefore, we believe that it may occur on all arches. If this bug is not discovered on x86, feel free to re-assign back to us.
Hi Doug, I am using the same index and operator version as used by P. I will check how can I provide must-gather logs.
I can confirm that this also happens on x86. It appears to be related to the initialization resource[0] that we added to the manifests in 4.10. If you use the "Create instance" button under "Provided APIs" it should work. Clearly we do need to fix this though. The immediate fix will probably be to remove the initialization resource and then we can add it back in later if we get it working correctly. I'm marking this as blocker- since this is new functionality for 4.10, not a regression. There's also a simple way to get past the problem if we for some reason can't get it fixed before release. 0: https://github.com/openshift/kubernetes-nmstate/blob/0fd00ac4371e5c753a07198b7c254693ec1bb9fc/manifests/4.10/kubernetes-nmstate-operator.v4.10.0.clusterserviceversion.yaml#L5
Moving back to assigned because it turns out the fix that merged was not complete. Another PR is on the way.
Oops, I forgot to link the second PR to this bug. https://github.com/openshift/kubernetes-nmstate/pull/248 has merged so this bug is ready for verification.
Verified with 4.10.0-202201310820 Steps: Install nmstate-operator from OperatorHub Click create NMState button. Nmstate instance created, all the pods are running
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.10.3 security update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2022:0056