15:00:36 <noonedeadpunk> #startmeeting openstack_ansible_meeting 15:00:36 <opendevmeet> Meeting started Tue Jun 24 15:00:36 2025 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:36 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:36 <opendevmeet> The meeting name has been set to 'openstack_ansible_meeting' 15:00:41 <noonedeadpunk> #topic rollcall 15:00:42 <noonedeadpunk> o/ 15:00:45 <NeilHanlon> o/ 15:00:48 <NeilHanlon> welcome back noonedeadpunk :) 15:00:53 <noonedeadpunk> thanks :) 15:00:56 <DavidGomez> o/ 15:01:03 <NeilHanlon> hope you had a good time off 15:01:05 <noonedeadpunk> sorry - I clean forgot to send email last week :( 15:01:08 <NeilHanlon> we relaxed like you told us to :D 15:01:25 <noonedeadpunk> but yeah, it was quite good. Though I wish I took even slightly more time off... 15:01:43 <NeilHanlon> same, i need a real vacation soon heh 15:01:56 <jrosser> o/ hello 15:02:15 <noonedeadpunk> for sure, for sure 15:03:41 <damiandabrowski> hi! 15:03:57 <noonedeadpunk> #topic office hours 15:05:58 <noonedeadpunk> today I've updated SHA bumps, as they were not reviewed/merged for 2 weeks, so did not make sense to leave them as-is 15:06:59 <opendevreview> Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/2023.2: Bump SHAs for EOL-ing 2023.2 https://review.opendev.org/c/openstack/openstack-ansible/+/950893 15:07:19 <noonedeadpunk> #link https://review.opendev.org/q/topic:%22bump_osa%22+status:open 15:07:29 <noonedeadpunk> but some are failing now :( 15:07:45 <noonedeadpunk> another thing about releases - that 2023.2 went to EOL last week IIRC. 15:08:15 <noonedeadpunk> So 2023.2-eol tag was created for all our roles, only integrated repo left to finalize this. 15:08:42 <noonedeadpunk> NeilHanlon: are you aware of any progress on adding Rocky 10 as CI image? 15:09:02 <NeilHanlon> not yet, but I will start to look at that 15:09:24 <noonedeadpunk> or C10S... 15:10:04 <noonedeadpunk> ok, gotcha 15:10:52 <noonedeadpunk> also, as wrote previously today, our hosts scenario is broken now due to merging https://review.opendev.org/c/openstack/openstack-ansible-plugins/+/951869 but not https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/951852 15:11:34 <jrosser> i wonder if we need a pip/setuptools etc bump on master for the sha bump failures there 15:11:39 <noonedeadpunk> so I'd ask to review https://review.opendev.org/c/openstack/openstack-ansible-openstack_hosts/+/951852 with some prio 15:12:06 <noonedeadpunk> you mean this? https://review.opendev.org/c/openstack/openstack-ansible/+/951288/1/global-requirement-pins.txt 15:12:20 <jrosser> oh /o\ 15:13:16 <noonedeadpunk> the error with this bump is `ValueError: check_hostname requires server_hostname` 15:13:31 <noonedeadpunk> which apparently goes from pip and is quite confusing... 15:14:43 <noonedeadpunk> NeilHanlon: btw right now mnasiadka has proposed Rocky 10 image: https://review.opendev.org/c/openstack/diskimage-builder/+/952548 :) 15:14:56 <NeilHanlon> oh, excellent :) 15:15:34 <mnasiadka> yeah, it's a long chain of patches before we can merge this one though (switch CI approach in DIB and some others...) 15:16:10 <opendevreview> Dmitriy Chubinidze proposed openstack/openstack-ansible master: Ensure the remote_tmp directory exists with correct permissions https://review.opendev.org/c/openstack/openstack-ansible/+/953218 15:16:13 <noonedeadpunk> it;'s not _that_ long, but yeah 15:16:44 <noonedeadpunk> I can hopefully test this one internally, as we rely on DIB for building images locally as well. 15:18:58 <noonedeadpunk> mnasiadka: I kinda wonder why we're going path of having different yum.repos.d there 15:21:15 <noonedeadpunk> as this probably could/should be `metalink=https://mirrors.centos.org/metalink?repo=centos-baseos-$releasever-stream&arch=$basearch` ? 15:23:56 <noonedeadpunk> anyway 15:25:15 <noonedeadpunk> other then that my plan is to proceed with reviews on pki/vault topic this week 15:28:59 <mnasiadka> noonedeadpunk: I ,,inherited'' that patch after somebody else, thanks for spotting that ;-) 15:32:29 <noonedeadpunk> I guess that's prety much it from my side then 15:32:59 <noonedeadpunk> ah. we also got a reported issue with skyline here: 15:33:10 <noonedeadpunk> #link https://answers.launchpad.net/openstack-ansible/+question/821993 15:33:31 <noonedeadpunk> I've tried to reproduce the issue, and have a really weird behavior on metal installation... 15:34:33 <noonedeadpunk> as I can't even auth in Skyline, getting 403 while trying to auth. 15:35:01 <noonedeadpunk> I think it might be related to HSTS actually, as permissions were made for Horizon, while Skyline is kinda different 15:40:42 <jrosser> hmm yes i can see this being incorrect 15:41:04 <jrosser> i think it shows up in the browser console? i forgot a lot about this tbh 15:41:15 <noonedeadpunk> buyt still I see 401 even in /api/v1/login for gunicorn 15:41:30 <noonedeadpunk> so it's smth different it seems 15:44:46 <opendevreview> Dmitriy Chubinidze proposed openstack/openstack-ansible master: Define /tmp directory as default https://review.opendev.org/c/openstack/openstack-ansible/+/953218 15:46:12 <noonedeadpunk> I'm really not sure why I have issues with Skyline though, as I'm pretty sure it was working normally on Caracal for me before... 15:46:43 <noonedeadpunk> so potentially another thing to look at... 15:49:25 <noonedeadpunk> as there are apperntly no more topics to discuss, let's conclude the meeting 15:49:28 <noonedeadpunk> #endmeeting