tripleo-ci-centos-8-scenario001-standalone broken with cephadm-16.1.0-4

[Rabi Mishra]

Looks like cephadm-16.1.0-4 needs podman version 2.0.2 or later. Either we've to pin cephadm or make this job non-voting I guess.

Noticed at:

https://5924d6108c185fae6617-478b3307a49eb1f7fda658d719f5a641.ssl.cf5.rackcdn.com/782978/2/check/tripleo-ci-centos-8-scenario001-standalone/8f8a107/logs/undercloud/home/zuul/standalone-ansible-3d8t58js/cephadm/cephadm_command.log

Running /home/zuul/standalone-ansible-3d8t58js/cephadm/cephadm_command.sh
2021-03-25 16:07:20,765 p=73140 u=root n=ansible | [WARNING]: Skipping key (deprecated) in group (overcloud) as it is not a
mapping, it is a <class 'ansible.parsing.yaml.objects.AnsibleUnicode'>

2021-03-25 16:07:20,776 p=73140 u=root n=ansible | [WARNING]: Found both group and host with same name: standalone

2021-03-25 16:07:21,705 p=73140 u=root n=ansible | PLAY [Deploy Ceph with cephadm] ************************************************
2021-03-25 16:07:21,842 p=73140 u=root n=ansible | 2021-03-25 16:07:21.841079 | bc764e01-95e9-6aa2-0ba8-00000000000e | TASK | Install cephadm package
2021-03-25 16:07:27,506 p=73140 u=root n=ansible | 2021-03-25 16:07:27.505039 | bc764e01-95e9-6aa2-0ba8-00000000000e | CHANGED | Install cephadm package | standalone
2021-03-25 16:07:27,540 p=73140 u=root n=ansible | 2021-03-25 16:07:27.539481 | bc764e01-95e9-6aa2-0ba8-00000000000f | TASK | Stat cephadm file
2021-03-25 16:07:28,282 p=73140 u=root n=ansible | 2021-03-25 16:07:28.281082 | bc764e01-95e9-6aa2-0ba8-00000000000f | OK | Stat cephadm file | standalone
2021-03-25 16:07:28,319 p=73140 u=root n=ansible | 2021-03-25 16:07:28.318804 | bc764e01-95e9-6aa2-0ba8-000000000010 | TASK | Fail if cephadm is not available
2021-03-25 16:07:28,422 p=73140 u=root n=ansible | 2021-03-25 16:07:28.421262 | bc764e01-95e9-6aa2-0ba8-000000000010 | SKIPPED | Fail if cephadm is not available | standalone
2021-03-25 16:07:28,456 p=73140 u=root n=ansible | 2021-03-25 16:07:28.455478 | bc764e01-95e9-6aa2-0ba8-000000000011 | TASK | List Ceph daemon instances on this host
2021-03-25 16:07:29,680 p=73140 u=root n=ansible | 2021-03-25 16:07:29.679366 | bc764e01-95e9-6aa2-0ba8-000000000011 | FATAL | List Ceph daemon instances on this host | standalone | error={"changed": true, "cmd": "/usr/sbin/cephadm ls --no-detail", "delta": "0:00:00.550744", "end": "2021-03-25 16:07:29.641497", "msg": "non-zero return code", "rc": 1, "start": "2021-03-25 16:07:29.090753", "stderr": "ERROR: podman version 2.0.2 or later is required", "stderr_lines": ["ERROR: podman version 2.0.2 or later is
required"], "stdout": "", "stdout_lines": []}

[Francesco Pantano (fmount)]

The issue [1] is reported on the Centos-infra system, waiting for that package to be removed.

[1] https://pagure.io/centos-infra/issue/265

[Francesco Pantano (fmount)]

Now ceph-octopus repos should contain the right cephadm versions [1].
Rechecking [2] and double check scenario001 is working again :/

[1] http://mirror.centos.org/centos/8/storage/x86_64/ceph-octopus/Packages/c/
[2] https://review.opendev.org/c/openstack/tripleo-heat-templates/+/781069

[Rabi Mishra (rabi)]

Looks like the issue has been resolved. Recent jobs are green.

https://storage.bhs.cloud.ovh.net/v1/AUTH_dcaab5e32b234d56b626f72581e3644c/zuul_opendev_logs_8ca/782978/2/check/tripleo-ci-centos-8-scenario001-standalone/8ca48ee/