Description of problem (please be detailed as possible and provide log snippests): Disconnected deployment from the latest version is failing, because rook-ceph-operator-* pod is stuck in ImagePullBackOff state. The reason seems to be the fact, that rook-ceph-operator image is referenced via version tag instead of checksum ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ $ oc get deployment -n openshift-storage rook-ceph-operator -o yaml | grep image: image: docker.io/rook/ceph:v1.15.0 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Version of all relevant components (if applicable): OCP Version: Client Version: 4.17.0-0.nightly-2024-08-28-171119 Kustomize Version: v5.0.4-0.20230601165947-6ce0bf390ce3 Server Version: 4.17.0-0.nightly-2024-08-28-171119 Kubernetes Version: v1.30.3 ODF Version: ocs-registry:4.17.0-87 Does this issue impact your ability to continue to work with the product (please explain in detail what is the user impact)? ODF Deployment on disconnected cluster is not possible. Is there any workaround available to the best of your knowledge? Rate from 1 - 5 the complexity of the scenario you performed that caused this bug (1 - very simple, 5 - very complex)? 3 Can this issue reproducible? Yes Can this issue reproduce from the UI? N/A If this is a regression, please provide more details to justify this: Yes, disconnected deployment was working in previous versions. Steps to Reproduce: 1. Prepare environment for disconnected deployment (mirror required images, etc.) 2. Start ODF deployment on the disconnected environment. 3. Check status of rook-ceph-operator CSV and rook-ceph-operator pod. 4. Check the image definition for rook-ceph-operator. Actual results: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ $ oc get csv -n openshift-storage rook-ceph-operator.v4.17.0-87.stable NAME DISPLAY VERSION REPLACES PHASE rook-ceph-operator.v4.17.0-87.stable Rook-Ceph 4.17.0-87.stable Failed $ oc get pod -n openshift-storage --selector app=rook-ceph-operator NAME READY STATUS RESTARTS AGE rook-ceph-operator-7b67976c46-d4j78 0/1 ImagePullBackOff 0 59m $ oc get deployment -n openshift-storage rook-ceph-operator -o yaml | grep image: image: docker.io/rook/ceph:v1.15.0 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Expected results: Rook-ceph-operator CSV is in Succeeded, all pods are running (rook-ceph-operator image is referenced via checksum). Additional info:
Hi, This started to happen recently due to the change in the upstream rook code to include registry names in the images[1], as a result it surfaced a hidden bug in the csv-gen script[2] of the rook-ceph-operator bundle that is responsible for updating the images correctly in the csv for downstream codebase. I will be sending a PR soon to fix this issue, and once we have a new build with the fix please use that. 1. https://github.com/rook/rook/pull/14550 2. https://github.com/red-hat-storage/rook/blob/master/build/csv/csv-gen.sh Thanks Nikhil
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Important: Red Hat OpenShift Data Foundation 4.17.0 Security, Enhancement, & Bug Fix Update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2024:8676