| jrosser | noonedeadpunk: can i sanity check with you want we expect to happen with this pin https://github.com/openstack/openstack-ansible-rabbitmq_server/blob/master/tasks/install_apt.yml#L32 | 10:30 |
|---|---|---|
| jrosser | this is the output of `apt-policy` and shows two pins the same but with different priorities https://paste.opendev.org/show/bnqbyrgGLyJQ1QpxuJwE/ | 10:32 |
| jrosser | oh doh - the 999 is for rabbitmq-erlang isnt it | 10:33 |
| noonedeadpunk | yeah | 10:33 |
| noonedeadpunk | but frankly - I was also a bit o_O at why we do that way | 10:34 |
| noonedeadpunk | like ping erlang and not pin rabbitmq | 10:34 |
| noonedeadpunk | *pin | 10:34 |
| jrosser | yeah indeed | 10:34 |
| noonedeadpunk | and now with defining versions explicitly, we probably just don't need that anymore | 10:34 |
| jrosser | we get the situation where a specific version is removed from the ppa and then we have $random thing take its place? | 10:35 |
| noonedeadpunk | we got that, yes | 10:35 |
| noonedeadpunk | but pinning won't really help here | 10:35 |
| noonedeadpunk | as if what we've pinned is not there - it will go down by prio as fallback | 10:35 |
| noonedeadpunk | so that's what reason behind https://opendev.org/openstack/openstack-ansible-rabbitmq_server/commit/2f686f4bfd67e78f884dad8267195af8b78bb576 | 10:36 |
| jrosser | ahha right so for erlang we try to install the very specific package | 10:37 |
| noonedeadpunk | and for rabbitmq now as well | 10:37 |
| jrosser | yep ok | 10:38 |
| noonedeadpunk | that line https://opendev.org/openstack/openstack-ansible-rabbitmq_server/blame/branch/master/vars/debian.yml#L50 | 10:38 |
| jrosser | ^ andrewbonney i think this answers our question, that the pins cannot enforce the version, but you kind of can at package install time | 10:38 |
| jrosser | well, at least make it fail in a "safe" way | 10:38 |
| noonedeadpunk | so I guess we don't need that pinning you've mentioned, but probably it doesn't hurt, soI just left it for now | 10:39 |
| jrosser | yep sure - i had missed that we improved the specificness of the package to install | 10:40 |
| noonedeadpunk | but I'd say we should adjust that to be consistent for erlang/rabbitmq | 10:40 |
| jrosser | right, because if we do get some odd behaviour in the future it will be asymmetric currently | 10:40 |
| noonedeadpunk | rabbitmq roleneeded some love for sure... | 10:41 |
| jrosser | i double checked the ppa1.rabbitmq Releases file, and it still references cloudsmith in the Origin: field | 10:41 |
| noonedeadpunk | I still haven't looked into force re-bootstrap of the cluster | 10:41 |
| noonedeadpunk | yup | 10:41 |
| jrosser | hmm ok so those ppa1 resolve to cloudflare IP | 10:44 |
| jrosser | so this is quite likley CDN on top of cloudsmith or something | 10:44 |
| opendevreview | Jonathan Rosser proposed openstack/openstack-ansible-rabbitmq_server master: Ensure that rabbitmq-erlang and rabbitmq-server repos use the same pins https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/933724 | 10:51 |
| noonedeadpunk | that was some mirror of cloudsmith that was jsut storing more versions for cheaper | 11:01 |
| noonedeadpunk | or smth like that | 11:01 |
| mgariepy | noonedeadpunk, https://bugs.launchpad.net/nova/+bug/2055245 <--- this seens to be my issue... | 15:30 |
| mgariepy | not 100% sure why it's working on some node and not other tho.. but well that's another question | 15:30 |
| mgariepy | hmm i'm already running this patch ;( | 15:40 |
| jrosser | Bah nothing provides python3 = 3.9.18-3.el9_4.5 needed by python3-devel-3.9.18-3.el9_4.5.x86_64 from appstream | 15:44 |
| noonedeadpunk | mgariepy: ah, so it bnever affected 2024.1 | 16:58 |
| noonedeadpunk | which is the only version I'm using with ovn | 16:59 |
| mgariepy | not sure i do have the patch running on 2023.1 .. so i'll see. | 17:04 |
| mgariepy | maybe something is missing ... somewhere :D | 17:05 |
| *** amorin_ is now known as amorin | 17:35 | |
| opendevreview | Dmitriy Rabotyagov proposed openstack/openstack-ansible-repo_server master: Increase timeouts and add retries for UC fetching over HTTP https://review.opendev.org/c/openstack/openstack-ansible-repo_server/+/933769 | 17:57 |
| mgariepy | ho. at the end i don't run it everywhere. | 17:58 |
| noonedeadpunk | how you ended up with different shas then? | 18:00 |
| mgariepy | yep. | 18:01 |
| mgariepy | my control plane isn't on the same sha than the last compute node. | 18:02 |
| mgariepy | it's weird that it works on the older node tho but. anyway | 18:02 |
| -opendevstatus- NOTICE: The Gerrit service on review.opendev.org will be offline momentarily at 20:30 utc (half an hour from now) to apply a configuration change | 20:01 | |
| -opendevstatus- NOTICE: The Gerrit service on review.opendev.org will be offline momentarily to apply a configuration change | 20:30 | |
| opendevreview | Merged openstack/openstack-ansible-rabbitmq_server master: Use json formatter for feature_flags fetcher https://review.opendev.org/c/openstack/openstack-ansible-rabbitmq_server/+/931903 | 21:16 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!