| noonedeadpunk | jrosser: fwiw I've reproduce the ceph failure in aio | 07:06 |
|---|---|---|
| noonedeadpunk | and that is fixed with `apt install python3-packaging` | 07:07 |
| jrosser | noonedeadpunk: thats really interesting, i was not able to reproduce it at all | 07:11 |
| noonedeadpunk | because apparently you have the package pre-installed in your image | 07:11 |
| jrosser | yes, that'll be it | 07:11 |
| jrosser | i have the upstream ubuntu cloud image | 07:12 |
| noonedeadpunk | and as I build mine alike to openinfra ones :D | 07:12 |
| jrosser | so there is something related to that actually | 07:12 |
| noonedeadpunk | but kinda question is - how better to fix that... | 07:12 |
| noonedeadpunk | in ceph-ansible? | 07:12 |
| noonedeadpunk | in our group_vars or smth? | 07:12 |
| jrosser | i had to change this https://review.opendev.org/c/openstack/openstack-ansible-ops/+/923586/6/mcapi_vexxhost/playbooks/mcapi_control_plane_k8s.yml#40 | 07:14 |
| jrosser | so in earlier versions of that patch i used the virtualenv module with name: [] | 07:15 |
| jrosser | to just create the virtualenv, but not to install any additional python modules into it | 07:15 |
| jrosser | and that was fine, until testing no noble | 07:16 |
| jrosser | when pip itself is seeming broken through lack of 'packaging' in the virtualenv | 07:16 |
| noonedeadpunk | well, but then we in all venvs ensure version of setuptools and pip | 07:16 |
| noonedeadpunk | which apparently you don't do when creating venv that way | 07:17 |
| noonedeadpunk | from https://opendev.org/openstack/openstack-ansible/src/branch/master/global-requirement-pins.txt#L9-L11 | 07:17 |
| jrosser | oh yes indeed | 07:17 |
| frickler | jrosser: noonedeadpunk: sorry to keep nagging you, but can we get https://review.opendev.org/c/openstack/openstack-ansible-tests/+/922408 in now, pretty please? :-) | 07:17 |
| noonedeadpunk | which would explain why we don't see same issues everywhere | 07:17 |
| noonedeadpunk | I will just +W https://review.opendev.org/c/openstack/openstack-ansible-tests/+/921434 | 07:18 |
| jrosser | noonedeadpunk: right but it has no explanation at all for the openinfra style images | 07:19 |
| jrosser | why would we see some change/regression in fundamental python behaviour when we should be using the distro python? | 07:19 |
| noonedeadpunk | setuptools is coming from uca? | 07:20 |
| noonedeadpunk | https://paste.openstack.org/show/bTIo3FuRRescJARkbBE1/ | 07:21 |
| jrosser | oh well then | 07:21 |
| jrosser | thats really not helpful | 07:21 |
| noonedeadpunk | depends on defenition :D | 07:21 |
| noonedeadpunk | kinda not some magical intermittent thing | 07:22 |
| noonedeadpunk | maybe worth adding packaging to the global-requirements and patch python_venv_build | 07:22 |
| noonedeadpunk | and openstack_hosts to ensure it's installed? | 07:22 |
| noonedeadpunk | (or system python) | 07:22 |
| noonedeadpunk | *for | 07:23 |
| jrosser | we should at least make the apt sources be the same | 07:27 |
| jrosser | as right now we can’t reproduce the CI results in the OS we state that is supported (and users will actually have) | 07:27 |
| noonedeadpunk | so I think apt sources are same. Though matter of pre-installed set of packages depending on the image... | 07:36 |
| noonedeadpunk | but not surue | 07:37 |
| noonedeadpunk | as python3-packaging goes from default repo and absent in UCA | 07:37 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-openstack_hosts master: Ensure python3-packaging is installed for distros https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/924994 | 07:46 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Delete glusterd data along with container data https://review.opendev.org/c/openstack/openstack-ansible/+/924910 | 07:47 |
| noonedeadpunk | let's see if that would be enough *for now* | 07:48 |
| opendevreview | Merged openstack/openstack-ansible-tests master: Bump Ansible version for tests repo https://review.opendev.org/c/openstack/openstack-ansible-tests/+/921434 | 07:51 |
| jrosser | noonedeadpunk: btw I am away next week, mostly offline for the whole week | 07:52 |
| opendevreview | OpenStack Proposal Bot proposed openstack/openstack-ansible-plugins master: Updated from OpenStack Ansible Tests https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/924995 | 08:01 |
| noonedeadpunk | jrosser: ok, gotcha | 08:20 |
| opendevreview | Merged openstack/openstack-ansible-tests master: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/922408 | 08:25 |
| noonedeadpunk | I think we need to land bumps for https://launchpad.net/bugs/2071734 then | 08:26 |
| opendevreview | Takashi Kajinami proposed openstack/ansible-config_template master: Remove compatibility with Python 2 https://review.opendev.org/c/openstack/ansible-config_template/+/917123 | 08:26 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump SHAs for master https://review.opendev.org/c/openstack/openstack-ansible/+/924997 | 08:29 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-config_template master: Bump version for bugfix release https://review.opendev.org/c/openstack/ansible-config_template/+/924998 | 08:31 |
| noonedeadpunk | let's try to land stable branches backports for the new point releases: https://review.opendev.org/q/parentproject:openstack/openstack-ansible+branch:%5Estable/.*+status:open+ | 08:35 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-role-uwsgi stable/2024.1: Ensure uWSGI is built with pcre support https://review.opendev.org/c/openstack/ansible-role-uwsgi/+/924876 | 08:37 |
| opendevreview | Jens Harbott proposed openstack/openstack-ansible-tests stable/2024.1: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925001 | 09:19 |
| opendevreview | Merged openstack/openstack-ansible-tests stable/2024.1: Bump Ansible version for tests repo https://review.opendev.org/c/openstack/openstack-ansible-tests/+/921435 | 09:20 |
| opendevreview | Jens Harbott proposed openstack/openstack-ansible-tests stable/2023.2: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925002 | 09:20 |
| opendevreview | Jens Harbott proposed openstack/openstack-ansible-tests stable/2023.1: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925003 | 09:21 |
| opendevreview | OpenStack Proposal Bot proposed openstack/openstack-ansible-os_skyline stable/2024.1: Updated from OpenStack Ansible Tests https://review.opendev.org/c/openstack/openstack-ansible-os_skyline/+/925004 | 09:29 |
| opendevreview | OpenStack Proposal Bot proposed openstack/openstack-ansible-plugins stable/2024.1: Updated from OpenStack Ansible Tests https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/925005 | 09:29 |
| opendevreview | Jens Harbott proposed openstack/openstack-ansible-tests stable/2024.1: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925001 | 10:09 |
| noonedeadpunk | so yes, https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/924994 does indeed fix the ceph job | 10:10 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump SHAs for master https://review.opendev.org/c/openstack/openstack-ansible/+/924997 | 10:10 |
| noonedeadpunk | ie https://zuul.opendev.org/t/openstack/build/b2d6ff349d4a489b971db0c2b93a6d7e | 10:10 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Use the 'generated' apparmor profile for all containers https://review.opendev.org/c/openstack/openstack-ansible/+/924661 | 10:12 |
| opendevreview | Merged openstack/openstack-ansible-os_nova stable/2023.2: Allow to supply multiline overrides to vendor_data https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/924281 | 10:15 |
| opendevreview | Merged openstack/openstack-ansible-os_nova stable/2023.1: Allow to supply multiline overrides to vendor_data https://review.opendev.org/c/openstack/openstack-ansible-os_nova/+/924282 | 10:29 |
| opendevreview | Merged openstack/openstack-ansible-plugins stable/2023.2: Update plugins collection version https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/922721 | 11:15 |
| frickler | noonedeadpunk: jrosser: I'm in cleanup mood today, do you want me to eol the three osa branches here (and same for pike+queens)? https://zuul.opendev.org/t/openstack/config-errors?branch=stable%2Focata&skip=0 | 11:21 |
| frickler | also, are there any other manual cleanups that are still pending? | 11:22 |
| jrosser | i think that those ancient branches are the ones we can't easily address ourselves, so subject to confirmation from noonedeadpunk then yes i think those should be eol | 11:26 |
| jrosser | i see there is cleanp we need to do of centos-7 nodeset particularly in the opestack-ansible-ops repo | 11:26 |
| jrosser | there might also be a few more very old branches in other repos too, looking at the bottom of https://zuul.opendev.org/t/openstack/config-errors?skip=550 | 11:28 |
| frickler | oh, right, so https://zuul.opendev.org/t/openstack/config-errors?branch=stable%2Fqueens&skip=0 is a better list to start from. sadly zuul doesn't allow filtering for multiple branches at a time | 11:31 |
| jrosser | it would be great if it could filter by parent project too | 11:32 |
| jrosser | as we have tons of repos | 11:32 |
| opendevreview | Merged openstack/openstack-ansible-tests stable/2024.1: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925001 | 11:56 |
| noonedeadpunk | frickler: so we'd really want to drop branches, but once I've reached releases time it was advised to give myself a write permissions to these repos/branches to manually delete them | 12:34 |
| noonedeadpunk | So if you can do that with your infra-root hat on - that will be pretty much appreciated | 12:34 |
| noonedeadpunk | as I never get to a point to propose patch escalating privileges to project-config... | 12:35 |
| fungi | zuul doesn't have a concept of parent projects in its data model (nor do the systems it's relying on for triggering data), but maybe a regex filter on project could suffice for projects that share a common prefix or substring | 12:38 |
| noonedeadpunk | having regex support would help a lot as well | 12:39 |
| fungi | or maybe a filter on queue name, but that seems like a boundary violation of sorts and i haven't thought it through very far so there might be reasons that wouldn't be a suitable lookup key anyway | 12:40 |
| noonedeadpunk | jrosser: so the folk who wrote a ML couple of days ago wit hkeystone just filled in a bug report: https://bugs.launchpad.net/openstack-ansible/+bug/2074196 | 12:45 |
| opendevreview | Merged openstack/openstack-ansible stable/2023.2: Use UCA mirror in CI for ubuntu https://review.opendev.org/c/openstack/openstack-ansible/+/924603 | 13:54 |
| frickler | noonedeadpunk: actually it is my release hat in this case I think. let me create an etherpad listing branches and SHAs, then, just to make sure we're on the same page^H^Hd | 14:38 |
| opendevreview | Merged openstack/openstack-ansible-tests stable/2023.2: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925002 | 20:53 |
| opendevreview | Merged openstack/openstack-ansible-tests stable/2023.1: Remove legacy centos-8-stream job definition https://review.opendev.org/c/openstack/openstack-ansible-tests/+/925003 | 20:53 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!