Description of problem: We should be using the same client library version as the core of the openshift cluster and in particular the openshift and kube API servers The following components will need to be bumped: - Cluster API - Cluster CAPI Operator (Downstream) - Cluster API Operator (Upstream) - Cluster API Provider AWS - Cluster API Provider Azure - Cluster API Provider vSphere - Cluster API Provider GCP - Cluster API Provider OpenStack We should also make sure that the Go version is updated to Go 1.18 and that the controller runtime version is updated to v0.12.x
Currently all other CAPI components (upstream Cluster API operator and all the providers) are blocked as they rely on https://github.com/kubernetes-sigs/cluster-api library, and its 1.24 hasn't been released yet. When upstream components are updated, we will backport those changes to OpenShift repos.
Moving to assigned we can move it to ON_QA once upstream dependency is resolved .
This will now be worked on for a 4.11.z stream, this isn't a blocker for GA
Hi Mike , could you add others PR links as well , I would make sure , we had CI run atleast once successfully after the merge to those repos ? thanks for adding azure , vsphere provider already
Please add PR for openstack as well to keep it documented at one Bug .
Went through the latest regression results , LGTM will move to verify . Let me know if openstack updates are not needed as of now. AWS - https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/298160 Vsphere - https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/298110 Azure - https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/297969 gcp-https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/297872 techpreview - GCP - https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/297833 AWS-https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/295392 VSPHERE-https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/294105 Azure - https://reportportal-openshift.apps.ocp-c1.prod.psi.redhat.com/ui/#prow/launches/619/293324 Additional info - Checking few failed cases in Azure, looks like automation flake , will move to verified once it is confirmed .
No, OpenStack cluster API provider update is not required for now. We don't use it and don't plan to do so in some near future. Wrt Azure, currently it doesn't work with OpenShift, and we have a related epic to enable it https://issues.redhat.com/browse/OCPCLOUD-1577. I updated its dependencies because we ship the image anyway.
Based on above comments and regression results moving to VERIFIED
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 (Important: OpenShift Container Platform 4.13.0 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-2023:1326