Capi-provider image can't be pulled off

[youssefazrak]

Hi team,

I'm going through the documentation and installing an HCP cluster.
I'm following the documentation (see this link, prerequisite section), and building the binaries as it should be. Hypershift is installed but when creating an HCP cluster (whether I use hcp or hypershift cli), the capi-provider image is in ImagePullBackOff.

  Warning  Failed          12m (x6 over 14m)    kubelet            Error: ImagePullBackOff
  Normal   Pulling         12m (x4 over 14m)    kubelet            Pulling image "registry.ci.openshift.org/ocp/4.18:cluster-api-provider-kubevirt"
  Warning  Failed          12m (x4 over 14m)    kubelet            Failed to pull image "registry.ci.openshift.org/ocp/4.18:cluster-api-provider-kubevirt": reading manifest cluster-api-provider-kubevirt in registry.ci.openshift.org/ocp/4.18: authentication required
  Warning  Failed          12m (x4 over 14m)    kubelet            Error: ErrImagePull
  Normal   BackOff         4m1s (x44 over 14m)  kubelet            Back-off pulling image "registry.ci.openshift.org/ocp/4.18:cluster-api-provider-kubevirt"

What surprises me is the authentication requirement. I should be authenticated as the cluster has been created with this command:
./hypershift create cluster kubevirt --name hostedtest1 --network-type Other --node-pool-replicas=2 --base-domain test.local --pull-secret ~/pullsecret --generate-ssh --release-image quay.io/openshift-release-dev/ocp-release:4.16.26-multi --arch amd64 --vm-node-selector node.kubernetes.io/instance-type=m5.metal

My pull secret is the same one I use to build my OCP clusters (or ROSA clusters).

I've tried with multiple different versions, with no success.

Any ideas?

[openshift-bot]

Issues go stale after 90d of inactivity.

Mark the issue as fresh by commenting /remove-lifecycle stale.
Stale issues rot after an additional 30d of inactivity and eventually close.
Exclude this issue from closing by commenting /lifecycle frozen.

If this issue is safe to close now please do so with /close.

/lifecycle stale