-
Bug
-
Resolution: Done
-
Major
-
None
Description of problem:
When creating a rosa-hcp with machine Pools distributed across more than 3 availability zones cannot delete the rosacontrolPlane CR and it shows error in the capa pod logs as below.
I1030 17:23:13.111099 1 rosamachinepool_controller.go:317] "Reconciling deletion of RosaMachinePool" controller="rosamachinepool" controllerGroup="infrastructure.cluster.x-k8s.io" controllerKind="ROSAMachinePool" ROSAMachinePool="ns-rosa-hcp/workers-2" namespace="ns-rosa-hcp" name="workers-2" reconcileID="a2723031-dbc5-4bd2-8b26-ec939650c71a" MachinePool="ns-rosa-hcp/workers-2" cluster="ns-rosa-hcp/rosa-hcp-1" I1030 17:23:13.229531 1 rosamachinepool_controller.go:317] "Reconciling deletion of RosaMachinePool" controller="rosamachinepool" controllerGroup="infrastructure.cluster.x-k8s.io" controllerKind="ROSAMachinePool" ROSAMachinePool="ns-rosa-hcp/workers-0" namespace="ns-rosa-hcp" name="workers-0" reconcileID="2d5f1538-bec2-4c4d-93d5-4ff3359d9897" MachinePool="ns-rosa-hcp/workers-0" cluster="ns-rosa-hcp/rosa-hcp-1" E1030 17:23:13.694458 1 controller.go:347] "Reconciler error" err="status is 400, identifier is '400', code is 'CLUSTERS-MGMT-400', at '2025-10-30T17:23:13Z' and operation identifier is '1fdb2fe0-1fdd-4741-8509-16f6c888ae26': A hosted cluster requires at least 2 replicas" controller="rosamachinepool" controllerGroup="infrastructure.cluster.x-k8s.io" controllerKind="ROSAMachinePool" ROSAMachinePool="ns-rosa-hcp/workers-2" namespace="ns-rosa-hcp" name="workers-2" reconcileID="a2723031-dbc5-4bd2-8b26-ec939650c71a"
Version-Release number of selected component (if applicable):
MCE v2.10 , CAPA v2.9.2
How reproducible:
100%
Steps to Reproduce:
- Create RosaControlplane with at least 3 RosamachinePool distributed into 3 availability zones.
- After the cluster is created try to delete the RosaControlplane.
- check the capa pod logs to see the error above
Actual results:
the RosaControlplane CR stuck in deleting.
Expected results:
the RosaControlplane CR to be deleted
Additional info:
upstream issue https://github.com/kubernetes-sigs/cluster-api-provider-aws/issues/5728