| noonedeadpunk | huh, I though we have sorted that out: https://zuul.opendev.org/t/openstack/build/fa3cfa7041c443b4814d9d84b19c01ba/log/job-output.txt#16218 | 09:10 |
|---|---|---|
| noonedeadpunk | OS minor versions wheels | 09:10 |
| noonedeadpunk | ah, not for rocky :( https://opendev.org/openstack/ansible-role-python_venv_build/commit/965cad09fa4d3d1d5f45b25039181fb51b453b85 | 09:11 |
| jrosser | good morning | 09:12 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/ansible-role-python_venv_build master: Use distribution_major_version for all distros except Ubuntu https://review.opendev.org/c/openstack/ansible-role-python_venv_build/+/901527 | 09:22 |
| noonedeadpunk | mornings o/ | 09:23 |
| damiandabrowski | hi! | 09:43 |
| jrosser | why do we specify this as a SHA? https://github.com/openstack/openstack-ansible/blob/master/ansible-collection-requirements.yml#L14 | 09:45 |
| noonedeadpunk | jrosser: you think worth using stable/$branch instead? | 09:55 |
| jrosser | not sure actually | 09:56 |
| noonedeadpunk | we don;'t have tags there as of today and we branch the repo as well | 09:56 |
| jrosser | i was just trying to understand why my cluster-api patch does not seem to use the depends-on changes in collections | 09:56 |
| noonedeadpunk | iirc, plugins depends-on should be respected today | 09:57 |
| jrosser | but i am not sure that specifying the hash in a-c-r affects that at all | 09:57 |
| jrosser | when we find a collection in an on-disk zuul repo the version is unspecified https://github.com/openstack/openstack-ansible/blob/master/scripts/get-ansible-collection-requirements.yml#L61-L64 | 09:58 |
| noonedeadpunk | IIRC we should be just overriding aything that is in zuul and a-cr | 09:58 |
| noonedeadpunk | and use whatever is in folder... | 09:58 |
| jrosser | yeah | 09:59 |
| noonedeadpunk | but galaxy.yml version is another thing actually | 09:59 |
| noonedeadpunk | as then it might try to find smth that would satisfy it or smth... | 10:00 |
| noonedeadpunk | but we have 1.0.0 there which is meh.... | 10:01 |
| noonedeadpunk | we indeed might want to switch release model of plugins to independent eventually... | 10:01 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-plugins master: Add role to set variables for a source installation https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/900527 | 10:06 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Add variable to control queues redundancy https://review.opendev.org/c/openstack/openstack-ansible/+/901534 | 10:07 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Test on ARM64 https://review.opendev.org/c/openstack/openstack-ansible/+/872423 | 10:12 |
| noonedeadpunk | we actually should land that one day ^ .... | 10:37 |
| noonedeadpunk | I guess rabbit is still biggest issue right now... | 10:38 |
| noonedeadpunk | (or better say erlang) | 10:38 |
| jrosser | i was hoping that the repos had got better now | 10:39 |
| jrosser | thought it was worth having a go again with it as we've sped up the metal jobs recently | 10:39 |
| noonedeadpunk | nah, they didn't... at least not for arm | 10:40 |
| noonedeadpunk | kolla does some workaround iirc for erlang in terms of repos | 10:40 |
| noonedeadpunk | or well, using `rabbitmq_install_method: distro` does work... | 10:41 |
| noonedeadpunk | and `rabbitmq_erlang_install_method: distro` | 10:41 |
| jrosser | noonedeadpunk: maybe this is now useable? https://binaries2.erlang-solutions.com/#ubuntu/pool/contrib/e/esl-erlang/ | 10:46 |
| jrosser | looks recent, arm+x86, and focal/jammy | 10:47 |
| jrosser | though they are missing rocky/centos9 | 10:51 |
| jrosser | NeilHanlon: something you might be interested in getting a gap filled ^ | 10:55 |
| noonedeadpunk | Well, they all have el8 but for el9 | 11:02 |
| noonedeadpunk | jrosser: I kind wonder for how long do they keep versions as well.... | 11:03 |
| noonedeadpunk | but yeah, might be for arm specifically it's good enough... | 11:03 |
| jrosser | i'll make a patch to see if i can unstick the arm stuff | 11:03 |
| jrosser | oh, no packages for bookworm either | 11:17 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-rabbitmq_server master: Use erlang-solutions repo for erlang packages. https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/901537 | 11:18 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible master: Test on ARM64 https://review.opendev.org/c/openstack/openstack-ansible/+/872423 | 11:19 |
| jrosser | argh https://ppa1.novemberain.com/rabbitmq/rabbitmq-server/deb/ubuntu/dists/jammy/main/binary-arm64/Packages 404 | 12:01 |
| jrosser | so even though the packages we want are architecture "all", this looks troublesome too | 12:02 |
| opendevreview | Merged openstack/openstack-ansible stable/zed: Bump SHAs for Zed https://review.opendev.org/c/openstack/openstack-ansible/+/901235 | 12:11 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_magnum master: Adopt for usage openstack_resources role https://review.opendev.org/c/openstack/openstack-ansible-os_magnum/+/901185 | 12:20 |
| noonedeadpunk | hopefully, https://review.opendev.org/q/topic:%22osa/openstack_resources%22 is ready .... | 12:25 |
| noonedeadpunk | I just recalled that we're missing cinder volume types there, but I dunno if that has landed to collection itself or not... | 12:26 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.1: Bump SHAs for 2023.1 https://review.opendev.org/c/openstack/openstack-ansible/+/901234 | 12:31 |
| noonedeadpunk | also likely we need to have this https://review.opendev.org/c/openstack/ansible-role-python_venv_build/+/901527 to fix gates for rocky... | 12:32 |
| opendevreview | Merged openstack/openstack-ansible stable/yoga: Bump SHAs for Yoga https://review.opendev.org/c/openstack/openstack-ansible/+/901236 | 13:07 |
| opendevreview | Merged openstack/openstack-ansible stable/2023.1: Fix incorrect release name https://review.opendev.org/c/openstack/openstack-ansible/+/901461 | 13:07 |
| opendevreview | Merged openstack/openstack-ansible stable/yoga: Fix incorrect release name https://review.opendev.org/c/openstack/openstack-ansible/+/901462 | 13:07 |
| opendevreview | Merged openstack/openstack-ansible stable/zed: Fix incorrect release name https://review.opendev.org/c/openstack/openstack-ansible/+/901397 | 13:07 |
| opendevreview | Merged openstack/openstack-ansible-os_octavia master: Adjust condition for availability_zone definition https://review.opendev.org/c/openstack/openstack-ansible-os_octavia/+/899127 | 13:21 |
| opendevreview | Merged openstack/openstack-ansible-os_glance master: Add glance_bin to rootwrap defenition https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/900930 | 13:35 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_glance stable/2023.1: Add glance_bin to rootwrap defenition https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/901562 | 13:36 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_glance stable/zed: Add glance_bin to rootwrap defenition https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/901563 | 13:37 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_glance stable/yoga: Add glance_bin to rootwrap defenition https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/901564 | 13:37 |
| NeilHanlon | jrosser thanks, i'll check that out | 13:38 |
| opendevreview | Merged openstack/openstack-ansible-os_murano master: Add example playbook and linters job to the role https://review.opendev.org/c/openstack/openstack-ansible-os_murano/+/899247 | 13:54 |
| jrosser | ah so removing ssh means synchronize module doesnt work any more | 14:06 |
| noonedeadpunk | hmmmmm | 14:08 |
| noonedeadpunk | yeah, synchronize indeed needs rsync I guess.... Somehow I thought it's still through connection plugin... But yeah, might be you're right.... | 14:10 |
| noonedeadpunk | you're running it with smth internal? | 14:10 |
| jrosser | the collection that deploys k8s uses it | 14:10 |
| noonedeadpunk | yeah, then we probably need to do like in keystone... | 14:11 |
| jrosser | i wonder if that can be refactored, or there is something smart rsync is doing | 14:11 |
| noonedeadpunk | Well. We can turn that other way around I guess and return SSH and rsync to containers. And then drop our connection plugin... | 14:12 |
| noonedeadpunk | as why to do things through lxc hosts when we have ssh available... | 14:12 |
| opendevreview | Christian Rohmann proposed openstack/openstack-ansible-rabbitmq_server stable/yoga: Add ability to add custom configuration for RabbitMQ https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/901556 | 14:48 |
| noonedeadpunk | #startmeeting openstack_ansible_meeting | 15:02 |
| opendevmeet | Meeting started Tue Nov 21 15:02:20 2023 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:02 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:02 |
| opendevmeet | The meeting name has been set to 'openstack_ansible_meeting' | 15:02 |
| noonedeadpunk | #topic rollcall | 15:02 |
| noonedeadpunk | o/ | 15:02 |
| NeilHanlon | o/ | 15:03 |
| NeilHanlon | apologies in advance if I'm flakey today, got a handful of things going on | 15:03 |
| NeilHanlon | mostly: rocky 9.3 was released yesterday | 15:03 |
| noonedeadpunk | Yeah, also a bit double-scheduled today | 15:07 |
| noonedeadpunk | #topic office hours | 15:07 |
| noonedeadpunk | So I think that openstack_resources topic should be ready for review | 15:08 |
| noonedeadpunk | #link https://review.opendev.org/q/topic:%22osa/openstack_resources%22 | 15:08 |
| noonedeadpunk | This is one of the things we scoped for 2023.2 during PTG | 15:08 |
| jrosser | o/ | 15:09 |
| * jrosser in another meeting also | 15:09 | |
| noonedeadpunk | One more thing for 2023.2 was skyline role. | 15:11 |
| noonedeadpunk | I've tried to spawn it up but fail to make it working properly for quite weird reasons. Like API docs for skyline-apiserver was saying no params should be passed for some methods, but they were expecting some | 15:12 |
| noonedeadpunk | I've seen some potentially related patches landed recently I about to abandon the idea to scope it for 2023.2 | 15:12 |
| NeilHanlon | i am +0 on it. i think it'd be cool to have, but as there's no one really both asking for + working on it, it seems something we can't really afford to support | 15:23 |
| noonedeadpunk | well, I can recall couple of times we were asked for that, and some already use some version of the role in a way... | 15:25 |
| noonedeadpunk | But yeah, as of today I failed a bit on spawning it... Maybe I should run devstack and see how things are done there... | 15:25 |
| noonedeadpunk | But I don't feel it will fit timeline for the releasing | 15:26 |
| noonedeadpunk | So I wanna wait for decision on openstack_resources and propose branching of roles | 15:26 |
| noonedeadpunk | as that is by far the last thing, except docs for pretty endpoint names, that should be done for 2023.2 | 15:35 |
| noonedeadpunk | #endmeeting | 15:55 |
| opendevmeet | Meeting ended Tue Nov 21 15:55:05 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:55 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/openstack_ansible_meeting/2023/openstack_ansible_meeting.2023-11-21-15.02.html | 15:55 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/openstack_ansible_meeting/2023/openstack_ansible_meeting.2023-11-21-15.02.txt | 15:55 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/openstack_ansible_meeting/2023/openstack_ansible_meeting.2023-11-21-15.02.log.html | 15:55 |
| opendevreview | Merged openstack/openstack-ansible-rabbitmq_server stable/2023.1: Add ability to add custom configuration for RabbitMQ https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/900734 | 17:24 |
| noonedeadpunk | hm, have anybody seen very weird behaviour for our ssh connection plugin on Antelope? | 18:03 |
| noonedeadpunk | As from what I see - it ignores passed user to the command | 18:03 |
| noonedeadpunk | https://paste.openstack.org/show/bH5vjLPnDbqe7G0MRauE/ | 18:07 |
| noonedeadpunk | very weirdly, but I see that only in 1 environment while others on same version are kinda fine.... | 18:09 |
| noonedeadpunk | which makes me pretty much o_O | 18:10 |
| mgariepy | hmm | 18:13 |
| noonedeadpunk | what is waaaaaaay more scary though | 18:22 |
| noonedeadpunk | https://paste.openstack.org/show/buNXGr0A3lXo06FlKbQm/ | 18:23 |
| noonedeadpunk | I wonder that is possible because of SSH certificates... Though it would be a bit weird I guess... | 18:28 |
| mgariepy | ansible using the cert while ssh would not ? | 18:29 |
| noonedeadpunk | I though it's not.... | 18:30 |
| noonedeadpunk | I don't think this was ever intended to be frank. | 18:30 |
| noonedeadpunk | or configured on deploy hosts in our case | 18:31 |
| noonedeadpunk | and on compute it should work only for nova I assume, if that;s what pricipal stand for... | 18:31 |
| noonedeadpunk | fweh, that was just persistent connection.... | 18:33 |
| noonedeadpunk | but passed user to SSH connection is oviously not respected.... | 18:34 |
| mgariepy | indeed. it looks like so. | 18:35 |
| mgariepy | i currently only use root for now. | 18:35 |
| mgariepy | so i don't have this particular issue i guess. | 18:36 |
| noonedeadpunk | I just was o_O when trying to provision new host, that has only `ubuntu` user | 18:36 |
| noonedeadpunk | and that I can't run anything against it... | 18:37 |
| mgariepy | i guess you will need to find the bug in openstack.osa.ssh transport. | 18:40 |
| mgariepy | maybe this ? https://opendev.org/openstack/openstack-ansible-plugins/commit/faf4b76ea5f906a08a87ed6388e4b0df60b5e7fb | 18:43 |
| mgariepy | but yeah.. not sure | 18:44 |
| mgariepy | i wonder if it has ever worked.. | 18:46 |
| noonedeadpunk | and `-e ansible_user=ubuntu` works.... | 18:49 |
| noonedeadpunk | /o\ | 18:49 |
| noonedeadpunk | I went back to Xena and it didn't work there either | 18:49 |
| noonedeadpunk | so it likely never worked and `-e ansible_user` was always used... | 18:51 |
| mgariepy | https://opendev.org/openstack/openstack-ansible-plugins/src/branch/master/plugins/connection/ssh.py#L130-L141 | 18:52 |
| mgariepy | juste like the documentation say.. | 18:52 |
| mgariepy | ;p | 18:52 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!