-
Bug
-
Resolution: Done
-
Major
-
None
-
Future, ACM 2.13.0
-
False
-
None
-
False
-
-
-
-
None
Description of problem:
Attempt to install 4.19 nightly build on spoke using latest ACM 2.13 build via gitops workflow, installation failed with following error repeated in spoke journal log.
Seems to be related to this PR: https://github.com/openshift/assisted-installer/pull/899
Feb 27 19:15:14 localhost.localdomain installer[5965]: time="2025-02-27T19:15:14Z" level=info msg="node-image-pull.service status: activating" Feb 27 19:15:14 localhost.localdomain assisted-installer[5963]: time="2025-02-27T19:15:14Z" level=info msg="node-image-pull.service status: activating" Feb 27 19:15:19 localhost.localdomain installer[5965]: time="2025-02-27T19:15:19Z" level=info msg="node-image-pull.service status: activating" Feb 27 19:15:19 localhost.localdomain assisted-installer[5963]: time="2025-02-27T19:15:19Z" level=info msg="node-image-pull.service status: activating" Feb 27 19:15:20 localhost.localdomain ironic-agent[3813]: 2025-02-27 19:15:20.438 1 DEBUG ironic_coreos_install [-] Assisted Agent is in state active (result success) _is_assisted_running /usr/lib/python3.9/site-packages/ironic_coreos_install.py:136 Feb 27 19:15:20 localhost.localdomain ironic-agent[3813]: 2025-02-27 19:15:20.438 1 DEBUG ironic_coreos_install [-] Still waiting for the assisted agent to finish start_assisted_install /usr/lib/python3.9/site-packages/ironic_coreos_install.py:157 Feb 27 19:15:20 localhost.localdomain podman[3794]: 2025-02-27 19:15:20.438 1 DEBUG ironic_coreos_install [-] Assisted Agent is in state active (result success) _is_assisted_running /usr/lib/python3.9/site-packages/ironic_coreos_install.py:136 Feb 27 19:15:20 localhost.localdomain podman[3794]: 2025-02-27 19:15:20.438 1 DEBUG ironic_coreos_install [-] Still waiting for the assisted agent to finish start_assisted_install /usr/lib/python3.9/site-packages/ironic_coreos_install.py:157 Feb 27 19:15:22 localhost.localdomain ostree-containe[12742]: Fetching ostree-unverified-registry:quay.io/openshift-release-dev/ocp-v4.0-art-dev@sha256:60b021d870d9243875776b5314c61a2dc7efd5f0b8f31633ba5a5755ed224096 Feb 27 19:15:22 localhost.localdomain node-image-pull.sh[12742]: layers already present: 8; layers needed: 43 (1.2 GB) Feb 27 19:15:22 localhost.localdomain ostree-containe[12742]: layers already present: 8; layers needed: 43 (1.2 GB) Feb 27 19:15:22 localhost.localdomain node-image-pull.sh[12742]: error: Importing: Unencapsulating base: Layer sha256:d06832d17d6e4203201a2aa56f0cf41e0d3db2f9b17fc241f77e65e0925342fc: mkdirat: Read-only file system Feb 27 19:15:22 localhost.localdomain node-image-pull.sh[6447]: Failed to fetch release image; retrying...
Version-Release number of selected component (if applicable):
hub OCP: 4.19.0-ec.2
ACM/MCE: 2.13.0-85/2.8.0-201
Spoke OCP: 4.19.0-0.nightly-2025-02-26-050226
How reproducible:
100%
Steps to Reproduce:
- Attempt to install 4.19 nightly on spoke using ACM 2.13/MCE 2.8
- ...
Actual results:
- agent stuck at installing state.
- repeated failed to fetch release image error from node-image-pull.sh on spoke journal log
Expected results:
installation succeeds
Additional info:
- duplicates
-
OCPBUGS-51204 node-image-pull.sh in bootstrap fails to pull images in PXE boot
-
- Verified
-
- is related to
-
OCPBUGS-50585 OCP 4.19 failing to deploy due to bootkube.sh error, missing oc command
-
- Closed
-