-
Story
-
Resolution: Unresolved
-
Critical
-
None
-
None
-
False
-
-
False
-
None
-
None
-
None
On the hypershift go modules, update all CAPI providers to a CAPI 1.11 compatible version and kubernetes depencencies to 1.33.
The following needs to be performed:
- Change CAPI and providers module imports to use downstream instead of upstream.
- Bump all providers to the latest version supporting k8s 1.33+
- Keep all the K8s dependencies to 1.34 after: OCPBUGS-66211
- Revert the temporary fix added for the CAPI bump:
CNTRLPLANE-2152 - Revert temporary CAPI fix used in OCPBUGS-66211
Acceptance critaria
- A management cluster with some running HCs
- Upgrade HO to install both v1beta1/v1beta2 served with storage:v1beta1
- Conversion via webhook enabled
- Some of our clients updated to v1beta2
- Extensive unit coverage and see our e2e green in that state
- is blocked by
-
OCPBUGS-77514 CAPI v1.11: MachineDeployment marked as completed with old machines still not deleted
-
- New
-
-
OCPBUGS-76337 Scale from zero fails because nodepool IAM policies lack ec2:DescribeInstanceTypes permission
-
- POST
-
- links to