Description of problem: The overcloud fails to deploy when using a satellite server for rpms/containers. The error is: 2022-06-30 16:55:16.400957 | 5254009d-c1eb-8140-be5f-000000006265 | FATAL | SATELLITE 6 | Execute katello-rhsm-consumer | controller-1 | error={"changed": true, "cmd": "katello-rhsm-consumer", "delta": "0:00:00.550173", "end": "2022-06-30 16:55:16.369568", "msg": "non-zero return code", "rc": 78, "start": "2022-06-30 16:55:15.819395", "stderr": "subscription-manager is disabled when running inside a container. Please refer to your host system for subscription management.", "stderr_lines": ["subscription-manager is disabled when running inside a container. Please refer to your host system for subscription management."], "stdout": "", "stdout_lines": []} Version-Release number of selected component (if applicable): RHOS-17.0-RHEL-9-20220628.n.1 How reproducible: Every time Steps to Reproduce: 1. Execute overcloud deploy from a satellite server. e.g. Jenkins job: https://rhos-ci-jenkins.lab.eng.tlv2.redhat.com/view/DFG/view/df/view/deployment/job/DFG-df-deployment-17.0-virthost-3cont_2comp_3ceph-ceph-ipv4-geneve-satellite-local-registry/ 2. 3. Actual results: Overcloud fails to deploy with error shown above Expected results: Overcloud deploys successfully Additional info:
Hey David, Care to check on the OC node if the symlink /etc/local-rhsm -> /etc/rhsm exists? If not, that would mean the patch is probably not in the OC image.. Cheers, C.
Hm, in fact, that's not related to OC image content - it's in t-h-t, so on the UC. Unfortunately, the job you linked doesn't show anything relevant, they apparently crashed before reaching the OC step... And a job I just checked[1] failed due to the very same thing, but at the UC level.... Not really execpted, since, at least at the time of the patch creation and the different tests, it was actually working as expected :(. IIRC, you even could work it through your job. Now, on that same job, I see an old version of the t-h-t package is installed: openstack-tripleo-heat-templates.noarch 14.3.1-0.20220617150351.66bbda8.el9ost @Default_Organization_OSP17_0_RPM_passed_phase1 It should be, at least, 20220625 in order to have the needed patch. Maybe it's just that? Cheers, C. [1] http://rhos-ci-logs.lab.eng.tlv2.redhat.com/logs/rcj/DFG-df-deployment-17.0-virthost-3cont_1comp_3ceph_3db_2net_3msg-ceph-ipv4-geneve-satellite-composable-advanced-subscription/19/undercloud1701-0/home/stack/undercloud_install.log.gz
We may need https://github.com/containers/podman/pull/14582 in podman-4.0 (and 4.1 if it is to come to el9).
Merged on upstream master, it should therefore be taken downstream over the weekend.
Using RHOS-17.0-RHEL-9-20220711.n.1 when doing the overcloud deploy from a satellite the subscription-manager is disabled when running inside a container error did not occur.
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 (Release of components for Red Hat OpenStack Platform 17.0 (Wallaby)), 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/RHEA-2022:6543