Goal
- As an OpenShift cluster admin, I want to upgrade my cluster from an el9-based release to an el10-based one without harming my running CNV virtual machines.
- Consider an OCP-4.18 (el9-based) cluster running CNV VMs. Each VM is run in a Pod with el9 libvirt+qemu on el9 kernel.
- OCP is upgraded to an el10-based release, say OCP-4.19. One of the nodes is evicted and reboots with el10-based rhcos.
- Until OCP upgrade is finished, CNV remains on 4.18, so the new host would run el9-libvirt with el10 kernel.
- During OCP upgrade, VMs are migrated to the new node.
- VMs may migrate away from the node, too.
- After OCP upgrade, CNV is upgraded to 4.19 and its libvirt+qemu would be taken from el10.
- The CNV team would like to ensure that these userspace and kernel combination are supported and tested by RHEL VIRT:
- (u=9, k=9) <-> (u=9, k=10)
- (u=9, k=10) -> (u=10, k=10)
Acceptance Criteria
A list of verification conditions, successful functional tests, or expected outcomes in order to declare this story/task successfully completed.
- Verify X
- Verify Y
- Verify Z
- is cloned by
-
RHEL-56817 [Dual stream for OpenShift Virtualization] Support two major versions of RHEL CoreOS in a single OCP release & cluster
- New
- is related to
-
CNV-45755 [Dual stream] Support two major versions of RHEL CoreOS in a single CNV release & cluster
- Backlog