Summary
Installing Vault Secrets Operator 1.3.0 from OperatorHub on OpenShift running on IBM Z (linux/s390x) fails because the kube-rbac-proxy sidecar image digest referenced by the operator deployment/CSV does not include an s390x manifest.
The main manager container pulls successfully, but the kube-rbac-proxy sidecar enters ImagePullBackOff, leaving the pod stuck at 1/2 containers running.
Environment
- OpenShift on IBM Z /
s390x
- Vault Secrets Operator
1.3.0
- Install source: OperatorHub
Expected behavior
The operator should install successfully on linux/s390x, and all containers in the controller-manager pod should start.
Actual behavior
The sidecar image pull fails because the referenced digest does not have an s390x image in its manifest list.
Pod status observed:
vault-secrets-operator-controller-manager-57f6d9b79f-vkjzf 1/2 ImagePullBackOff
Relevant event:
Failed to pull image "registry.redhat.io/openshift4/ose-kube-rbac-proxy@sha256:64f6654c142df021651ae8c1ae9a290293040edf46f6bbf50b78e5e3db1ad0cc": choosing an image from manifest list docker://registry.redhat.io/openshift4/ose-kube-rbac-proxy@sha256:64f6654c142df021651ae8c1ae9a290293040edf46f6bbf50b78e5e3db1ad0cc: no image found in manifest list for architecture "s390x", variant "", OS "linux"
Meanwhile, the main manager container does pull successfully:
registry.connect.redhat.com/hashicorp/vault-secrets-operator@sha256:eed45038ea64171a643e9afb63a11300ea71b3f06660c4e9ad4fc56b1d567643
Workaround
Manually edit the deployment/operator bundle reference and replace the kube-rbac-proxy image digest with an s390x-compatible digest.
Request
Can you update the OperatorHub/operator bundle artifacts for VSO 1.3.0 so the referenced kube-rbac-proxy image supports linux/s390x?
Summary
Installing Vault Secrets Operator 1.3.0 from OperatorHub on OpenShift running on IBM Z (
linux/s390x) fails because thekube-rbac-proxysidecar image digest referenced by the operator deployment/CSV does not include ans390xmanifest.The main
managercontainer pulls successfully, but thekube-rbac-proxysidecar entersImagePullBackOff, leaving the pod stuck at1/2containers running.Environment
s390x1.3.0Expected behavior
The operator should install successfully on
linux/s390x, and all containers in the controller-manager pod should start.Actual behavior
The sidecar image pull fails because the referenced digest does not have an
s390ximage in its manifest list.Pod status observed:
Relevant event:
Meanwhile, the main manager container does pull successfully:
Workaround
Manually edit the deployment/operator bundle reference and replace the
kube-rbac-proxyimage digest with ans390x-compatible digest.Request
Can you update the OperatorHub/operator bundle artifacts for VSO 1.3.0 so the referenced
kube-rbac-proxyimage supportslinux/s390x?