15:01:20 <noonedeadpunk> #startmeeting openstack_ansible_meeting 15:01:20 <opendevmeet> Meeting started Tue Jul 22 15:01:20 2025 UTC and is due to finish in 60 minutes. The chair is noonedeadpunk. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:20 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:20 <opendevmeet> The meeting name has been set to 'openstack_ansible_meeting' 15:01:25 <noonedeadpunk> #topic rollcall 15:02:15 <noonedeadpunk> o/ 15:02:32 <damiandabrowski> hi! 15:06:26 <noonedeadpunk> #topic office hours 15:07:07 <noonedeadpunk> I saw there was quite a good progress on PKI role, though I failed to look into details of changes so far 15:07:20 <noonedeadpunk> damiandabrowski: do we have anything to discuss on this matter? 15:08:37 <damiandabrowski> yeah, I pushed some fixes recently and I'm waiting for reviews :D 15:08:49 <damiandabrowski> most of the unresolved comments are here: https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/949420 15:10:12 <noonedeadpunk> it's quite a long read :p 15:11:07 <damiandabrowski> yeah... :D bu in general, when we agree on changes in os_neutron role(which can be used as an example), I can start pushing changes to all other roles 15:11:25 <noonedeadpunk> ok, yes, that sounds very reasonable 15:11:57 <damiandabrowski> and when I do that for at least all roles that are included in base aio scenario, we can run openbao scenario to ensure that everything works fine 15:13:15 <jrosser> o/ hello 15:15:28 <noonedeadpunk> ok, good then 15:15:49 <noonedeadpunk> Yesterday an LXC package seems to be built for EL10 15:15:59 <noonedeadpunk> (in EPEL) 15:16:20 <noonedeadpunk> so it's about time to check on the CI 15:16:41 <noonedeadpunk> While there's still no systemd-networkd for rocky, we have good chance of C10S 15:17:18 <noonedeadpunk> MariaDB should have new release this week, so potentially is is gonna be sovled as well 15:17:42 <noonedeadpunk> though my test patch for adding CI fails super early actually on adding EPEL 15:18:11 <noonedeadpunk> Did not check it yet in details though, but it's smth I wanna do this week. 15:20:18 <noonedeadpunk> another obviously important thing, is to ensure https://review.opendev.org/c/openstack/openstack-ansible/+/955226 works 15:20:25 <noonedeadpunk> as there's no neutron-server at the moment 15:20:35 <noonedeadpunk> which also blocks switching CAPI to Noble 15:21:06 <noonedeadpunk> I think we need to enable uwsgi one more time 15:21:26 <noonedeadpunk> though worth double-checking if we have all obligatory supporting services present as well 15:26:32 <noonedeadpunk> I guess we have no choice at this point regarding uwsgi usage there 15:26:56 <noonedeadpunk> but also I don't see what we were missing for neutron previous time 15:27:04 <noonedeadpunk> seems like all required services were in place 15:28:02 <opendevreview> Dmitriy Rabotyagov proposed openstack/openstack-ansible-os_neutron master: Switch Neutron to uWSGI one more time https://review.opendev.org/c/openstack/openstack-ansible-os_neutron/+/955610 15:28:21 <opendevreview> Dmitriy Rabotyagov proposed openstack/openstack-ansible master: Bump SHAs for master https://review.opendev.org/c/openstack/openstack-ansible/+/955226 15:29:01 <jrosser> oh yes i did revisit the uwsgi patches a bit 15:29:11 <jrosser> there are still a few leftovers to merge 15:29:28 <noonedeadpunk> #link https://review.opendev.org/q/topic:%22osa-wsgi-module%22+status:open 15:29:46 <noonedeadpunk> Though I think we might wanna sha bump first for some of them 15:29:50 <jrosser> magnum CI is still broken 15:30:40 <noonedeadpunk> I wonder if heat driver is gone completely at this point 15:30:48 <jrosser> i think the sha bump might depend on https://review.opendev.org/c/openstack/openstack-ansible-os_keystone/+/955268 15:31:00 <noonedeadpunk> yup, correct 15:31:24 <noonedeadpunk> nope, heat is still there 15:31:31 <jrosser> thats an easy review if anyone has a moment 15:32:10 <noonedeadpunk> I think once SHA bump lands we can look into magnum issue more closely as well... 15:32:41 <noonedeadpunk> or well, at least there seems to be dependency for CAPI test 15:33:55 <noonedeadpunk> I think these are pretty much all ongoing things at the moment? 15:40:45 <noonedeadpunk> fwiw, about failing migrations - switching back to tunneled migrations (nova_libvirtd_listen_tls: 0 and nova_libvirtd_listen_tcp: 1) in fact seems to perform better then native libvirt migrations with TLS. 15:41:12 <noonedeadpunk> and this kinda solves the problem, unless you wanna leverage local storage migrations... 16:32:10 <opendevreview> Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Allow to override ceph_components from defaults https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/955615 16:33:24 <opendevreview> Dmitriy Rabotyagov proposed openstack/openstack-ansible-ceph_client master: Allow to override ceph_components from defaults https://review.opendev.org/c/openstack/openstack-ansible-ceph_client/+/955615 17:02:03 <noonedeadpunk> #endmeeting