Description of problem:

After enabling cluster-api-provider-azure-preview in the MCE, the azureserviceoperator-controller-manager pod fails to start and keep crashing. This prevents the MCE from moving beyond Progressing.

The capz-controller-manager pod also seems to randomly crash and restart with this error.

E0120 18:37:37.977498       1 controller.go:258] "Could not wait for Cache to sync" err="failed to wait for azuremachine caches to sync kind source: *v1beta2.Machine: timed out waiting for cache to be synced for Kind *v1beta2.Machine" controller="azuremachine" controllerGroup="infrastructure.cluster.x-k8s.io" controllerKind="AzureMachine" source="kind source: *v1beta2.Machine"

Version-Release number of selected component (if applicable):

"acm_image": "quay.io:443/acm-d/acm-dev-catalog:2.16.0-DOWNSTREAM-2026-01-16-11-15-08",
"mce_image": "quay.io:443/acm-d/mce-dev-catalog:2.11.0-DOWNSTREAM-2026-01-20-05-18-12",

How reproducible:

Always

Steps to Reproduce:

1. Install build above
2. Enable cluster-api-provider-azure-preview

Actual results:

Pod azureserviceoperator-controller-manager fails to start, but in some instances, after about 9 attempts it did finally start. The pod should start on the first attempt and this will cause automation failures for the installer team.

Expected results:

Pod should start in a reasonable time