The ClusterVersionOperatorConfiguration feature gate was promoted to the TechPreviewNoupgrade. However, the relevant functionality is not working in a TechPreviewNoupgrade cluster. 

    4.20, 4.21 development

  1/1  

    1.Launch a cluster; e.g. via a Cluster Bot by running `launch 4.20 aws`
    2. Patch the cluster FeatureGate to enable TechPreviewNoUpgrade

$ oc patch featuregates.config.openshift.io cluster --type=merge --patch '{"spec":{"featureSet":"TechPreviewNoUpgrade"}}'
     

    The ClusterVersionOperator CRD nor CR are applied. The CVO is logging errors such as:

W0917 16:20:08.436678 1 reflector.go:569] github.com/openshift/client-go/operator/informers/externalversions/factory.go:125: failed to list *v1alpha1.ClusterVersionOperator: the server could not find the requested resource (get clusterversionoperators.operator.openshift.io)E0917 16:20:08.436733 1 reflector.go:166] "Unhandled Error" err="github.com/openshift/client-go/operator/informers/externalversions/factory.go:125: Failed to watch *v1alpha1.ClusterVersionOperator: failed to list *v1alpha1.ClusterVersionOperator: the server could not find the requested resource (get clusterversionoperators.operator.openshift.io)" logger="UnhandledError"

The manifests are applied, no relevant errors are observable.    

The manifests are missing. The CRD must be imported by bumping up the openshift/api module in the CVO repository, and the CR must be created in the CVO repository.

The errors can be observed in the CI. For example:

https://gcsweb-ci.apps.ci.l2s4.p1.openshiftapps.com/gcs/test-platform-results/pr-logs/pull/openshift_cluster-version-operator/1206/pull-ci-openshift-cluster-version-operator-main-e2e-aws-ovn-techpreview/1963983229635530752/artifacts/e2e-aws-ovn-techpreview/gather-extra/artifacts/pods/openshift-cluster-version_cluster-version-operator-7f5fdd78fc-8m8xv_cluster-version-operator.log