| *** ysandeep|PTO is now known as ysandeep | 05:07 | |
| noonedeadpunk | mornings | 07:43 |
|---|---|---|
| noonedeadpunk | I'd say to address that we can try add some condition to ca's as well as we add to certs | 07:44 |
| noonedeadpunk | ah, we already do that | 07:45 |
| noonedeadpunk | https://opendev.org/openstack/ansible-role-pki/src/branch/master/vars/main.yml#L32 | 07:45 |
| jrosser_ | morning | 07:54 |
| noonedeadpunk | tbh I think run_once is correct there. It would be really too much overhead without it imo | 07:57 |
| noonedeadpunk | like all computes/containers/net nodes will fetch exact same file for quite small reason | 07:57 |
| jrosser_ | right - thats what i was thinking here that the collection of the CA from the pki setup node was consistent across all the hosts | 08:02 |
| jrosser_ | but i can also see how normal ansible expectations would want vars in group_vars to do the right thing | 08:02 |
| noonedeadpunk | well... yes hard to disagree. But I don't know how to tick both boxes here | 08:23 |
| *** ysandeep is now known as ysandeep|lunch | 09:08 | |
| *** ysandeep|lunch is now known as ysandeep | 10:46 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 11:12 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 11:13 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/train: Disable upgrade jobs on EM branch https://review.opendev.org/c/openstack/openstack-ansible/+/861858 | 11:14 |
| anskiy | I have a question about updates. As far as I've understood, I would not be able to update to Zed if I'm running Focal, and I would have to wait until Antiloper release, when I'll need to upgrade to Jammy along with OpenStack upgrade, right? | 11:17 |
| jrosser_ | anskiy: https://docs.openstack.org/openstack-ansible/latest/admin/upgrades/compatibility-matrix.html | 11:18 |
| anskiy | jrosser_: I see, thank you :) | 11:19 |
| jrosser_ | Zed will be a release with support for focal and Jammy so thats a pretty good place to do your OS upgrade | 11:20 |
| noonedeadpunk | I think even on antelope both should be still supported... | 11:21 |
| anskiy | jrosser_: got it | 11:22 |
| anskiy | noonedeadpunk: isn't there some problem with minimal python version for antelope? | 11:22 |
| anskiy | I mean, on focal, you have 3.9 | 11:22 |
| noonedeadpunk | anskiy: so there should be Y->A direct updates. While they're not "officially supported" all project should have CI jobs to test this path | 11:23 |
| noonedeadpunk | And A->A+2 will be first officially supported upgrade path | 11:24 |
| noonedeadpunk | so Y->A is kind of "training" but this also means that projects will still need to support that somehow | 11:24 |
| *** dviroel|out is now known as dviroel | 11:25 | |
| anskiy | and 3.9 is supported, but OSA uses 3.8 on focal right now | 11:27 |
| noonedeadpunk | you actually can't use 3.9 for nova at least iirc? | 11:28 |
| noonedeadpunk | as there're no python3-libvirt bindings for 3.9? | 11:28 |
| noonedeadpunk | also it will be problematic for ceph as well | 11:29 |
| jrosser_ | imho this needs to be better documented in the PTI (looking at the tc :) ) | 11:30 |
| jrosser_ | the current approach considers each release in isolation, but with the adding of A->A+2 type upgrades then the lifecycle of the distros and pythons (becasue of the bindings) needs to be understood | 11:31 |
| anskiy | noonedeadpunk: yeah, true. | 11:32 |
| anskiy | so it looks like I would have to run mixed Ubuntu versions on Z, as I would still prefer to do direct upgrades, without skipping anything in between | 11:35 |
| anskiy | thanks for the info! | 11:35 |
| noonedeadpunk | jrosser_: oh yes, I was going to raise that today on PTG | 11:36 |
| noonedeadpunk | anskiy: let us know how mixed versions working as of today | 11:36 |
| jrosser_ | noonedeadpunk: it would be great if there was some graphical view | 11:37 |
| noonedeadpunk | as we aimed to fix that support in Y but never ran production env with that | 11:37 |
| jrosser_ | like this sort of thing https://en.wikipedia.org/wiki/Ubuntu_version_history#Version_timeline | 11:38 |
| anskiy | noonedeadpunk: I think, I would be able to start this only at the very end of a year, but sure! | 11:38 |
| noonedeadpunk | yeah, and eventually confusing part is that for A it's currently obviously stated only 22.04 and python310 which is very confusing | 11:38 |
| jrosser_ | i don't believe there is much (any?) consideration for overlap and upgrades when the PTI is defined, it's too simplistic | 11:39 |
| jrosser_ | also the cadence of python releases is getting much quicker | 11:40 |
| jrosser_ | it's no longer like the past with a decade of python2.7 so some of the previous approaches to setting target distros/pythons are maybe not as valid as in the far far past | 11:41 |
| noonedeadpunk | well, some subset of python versions still need to be tested supported. And because of bindings it's still tighten to distros. IMO nowadays I'd say it's more distros that should adjust their midset in terms of building python bindings, as indeed lifetime of ubuntu lts is way longer then python it provides | 11:43 |
| jrosser_ | perhaps things can be added to UCA and the like there | 11:45 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-rabbitmq_server stable/stein: Use cloudsmith repo for rabbit and erlang https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/862104 | 12:07 |
| *** ysandeep is now known as ysandeep|mtg | 12:15 | |
| mgariepy | good morning | 12:15 |
| noonedeadpunk | \o/ | 12:15 |
| mgariepy | how are you guys doing this morning ? | 12:16 |
| noonedeadpunk | meh.... continue digging in old crap... | 12:20 |
| noonedeadpunk | also had weird thing wrt cinder on beginning of xena (like 24.0.1 tag of osa) | 12:21 |
| noonedeadpunk | there's obviously some bug with clustering or dunno | 12:21 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/ussuri: Do not user urljoin for deprecated script https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/862143 | 13:25 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible stable/ussuri: Switch to tracking stable/ussuri for EM release https://review.opendev.org/c/openstack/openstack-ansible/+/853029 | 13:25 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 13:27 |
| *** ysandeep|mtg is now known as ysandeep | 13:44 | |
| *** dviroel is now known as dviroel|ptg | 14:08 | |
| *** ysandeep is now known as ysandeep|dinner | 14:29 | |
| *** ysandeep|dinner is now known as ysandeep | 14:55 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/ussuri: Return CentOS 7 jobs to voting https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861789 | 14:57 |
| *** dviroel|ptg is now known as dviroel|lunch\ | 16:06 | |
| *** ysandeep is now known as ysandeep|out | 16:20 | |
| *** dviroel|lunch\ is now known as dviroel | 16:53 | |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible master: Do not enable glance_show_multiple_locations https://review.opendev.org/c/openstack/openstack-ansible/+/862167 | 17:01 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible-os_glance master: Add external_lb_vip_address to glance_cors_allowed_origin by default https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/862171 | 17:06 |
| opendevreview | Damian Dąbrowski proposed openstack/openstack-ansible-os_glance master: Add external_lb_vip_address to glance_cors_allowed_origin by default https://review.opendev.org/c/openstack/openstack-ansible-os_glance/+/862171 | 17:08 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-tests stable/train: Bump pluggy back to support py27 https://review.opendev.org/c/openstack/openstack-ansible-tests/+/862173 | 17:10 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 17:11 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 17:21 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-tests stable/train: Bump pluggy back to support py27 https://review.opendev.org/c/openstack/openstack-ansible-tests/+/862173 | 17:48 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 18:03 |
| opendevreview | Merged openstack/openstack-ansible master: Imported Translations from Zanata https://review.opendev.org/c/openstack/openstack-ansible/+/861697 | 18:20 |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-lxc_hosts stable/train: Use legacy image retrieval for CentOS 7 https://review.opendev.org/c/openstack/openstack-ansible-lxc_hosts/+/861849 | 18:34 |
| *** dviroel is now known as dviroel|out | 22:47 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!