gibi | fungi: did we updated zuul to the recent fix of cross provider worker allocation during the weekend? I still see hits where a nova multinode job fails as the compute1 worker has no access to rabbitmq running on the controller. I added the recent hits to the bug https://bugs.launchpad.net/nova/+bug/2115980 | 10:33 |
---|---|---|
frickler | gibi: there was a second issue found yesterday, which should be fixed since tonight. if you didn't find failures from today, things might be better now | 11:09 |
gibi | yeah I will check later today to see if this is gone now. Thanks | 11:29 |
gibi | I have couple jobs running but indeed so far the last such failure was 2025-07-07T23:20:42 | 11:36 |
gibi | so it looks good so far | 11:36 |
elodilles | frickler: hi, since i've approved "Puppet OpenStack's Bobcat EOL" patch i'm planning to run the eol-cleanup script. is there any infra activity for that the script could cause interference, or is it OK if I run now the script? | 12:27 |
frickler | elodilles: should be fine afaict | 12:42 |
elodilles | ACK, will run it then soon | 12:52 |
fungi | gibi: we think it's fixed as of the manual restart of the launchers at 22:24 utc yesterday, so any multinode job builds for changes enqueued after that should get their nodes from the same provider. if you see any since then that didn't, we might still be missing another bug or need to do some provider-specific tuning | 13:14 |
elodilles | result for the clean up: https://paste.opendev.org/show/bIxOD2WGtpOvFQdlAuFx/ | 13:26 |
fungi | thanks! | 13:26 |
elodilles | there were 2 missing 2023.2-eol tag. those have to be checked :S | 13:26 |
elodilles | i'm trying to search now the tagging job logs | 13:27 |
fungi | let the zuul job errors commence | 13:27 |
fungi | s/job/config/ | 13:27 |
fungi | which two are missing? | 13:27 |
fungi | oh, i see them in the paste | 13:27 |
fungi | openstack/puppet-openstack-integration contains eol stale branch (2023.2) No 2023.2-eol tag found! Branch cannot be deleted. Skipping. | 13:28 |
fungi | openstack/puppet-openstack_spec_helper contains eol stale branch (2023.2) No 2023.2-eol tag found! Branch cannot be deleted. Skipping. | 13:28 |
elodilles | yepp ^^^ those | 13:28 |
fungi | i think those may both be retired | 13:28 |
elodilles | hmmm | 13:29 |
fungi | nope, at least https://opendev.org/openstack/puppet-openstack-integration looks live | 13:29 |
fungi | https://opendev.org/openstack/puppet-openstack_spec_helper too | 13:30 |
fungi | so neither is retired | 13:30 |
elodilles | tagging was not initiated, as i don't see them in the tag job list: https://zuul.opendev.org/t/openstack/builds?pipeline=tag&skip=0 | 13:31 |
fungi | sounds like a good explanation then, yes | 13:32 |
sean-k-mooney | so i was just reviewing the "OpenDev Team Meeting Agenda for July 8, 2025" email | 13:34 |
sean-k-mooney | fungi: you use weechat as your irc client right? | 13:35 |
sean-k-mooney | it does not have matrix support does it? | 13:35 |
sean-k-mooney | or maybe it does and i have forgoten | 13:35 |
sean-k-mooney | clarkb: i guess that more your topic since yoru going to write a spec on the proposal to move coms to matrix | 13:36 |
elodilles | (so this is the job log and the two relevant lines, that unfortunately don't tell much :/ https://zuul.opendev.org/t/openstack/build/57a74bbfb33d4519afac04b890147b7e/log/job-output.txt#7193 and https://zuul.opendev.org/t/openstack/build/57a74bbfb33d4519afac04b890147b7e/log/job-output.txt#7198 ) | 13:37 |
elodilles | (there are no 'Tagging 2023.2-eol in [..]' lines after these two lines) | 13:37 |
elodilles | (the 'skipping closed series' lines are a bit misleading as those are only logs for branching operations) | 13:38 |
fungi | sean-k-mooney: yes, i do | 13:40 |
fungi | sean-k-mooney: and it doesn't have integrated matrix support, no, though there are two plugins that add it, an outdated one in python that mostly doesn't work any more, and an incomplete but still in-progress one in rust which i'm using | 13:41 |
sean-k-mooney | ya so looking i think the python one is the one i used before | 13:41 |
sean-k-mooney | but there is a rust rewrite becuase rust | 13:41 |
sean-k-mooney | https://github.com/poljar/weechat-matrix-rs | 13:41 |
fungi | right, it's the same author, they abandoned the python one (which then fell behind on more recent protocol changes) and started a rewrite in rust | 13:42 |
fungi | but per usual they underestimated the amount of work rewriting your software in a different language can be | 13:42 |
fungi | so it's still not reached feature parity with the old one | 13:43 |
sean-k-mooney | ya... the question is is it functional enouch to experiment with | 13:43 |
sean-k-mooney | or shoudl i just stick with irc for now. i have element and thruderbird set up as matix clients too | 13:44 |
sean-k-mooney | fungi: are you plannig to try it out with the experiemtn or wait a bit | 13:44 |
fungi | i mean, i'm using it. the main shortcomings are 1. it doesn't yet have the ability to complete session verification so your session in it shows up as unverified by your other clients, and 2. it lacks the ability to join and leave rooms so you need to use e.g. the element browser client if you want to adjust the rooms you're in (but it will automatically link up buffers to | 13:45 |
fungi | each room you've joined with another client) | 13:45 |
sean-k-mooney | ah ok | 13:46 |
sean-k-mooney | the verificaiton is a littel annoying but the room contol i can live with | 13:46 |
sean-k-mooney | i dont really change that often | 13:46 |
fungi | if you can accept those problems while it's still being worked on, then it's definitely in a state that you can use it for conversations | 13:46 |
sean-k-mooney | ack one followup | 13:46 |
sean-k-mooney | how does it handel matix usere doing long multi line messages | 13:47 |
sean-k-mooney | does it actully render properly or is it like irc where you just get a link to the rest of the message | 13:47 |
sean-k-mooney | thats one of the most annoying things on the irc side today | 13:47 |
fungi | it doesn't do anything special with them, you just see the whole message. it's not like irc where there's a smallish protocol limit to per-message size | 13:47 |
sean-k-mooney | well thats basically what i want just show the full thing inline as normal | 13:48 |
fungi | if you're talking about in channels bridged to irc, then of course you'd need to join the matrix bridge room for that channel rather than the irc channel itself | 13:48 |
fungi | i haven't bothered to do that for any irc channels personally, i just use the plugin for matrix-only rooms | 13:49 |
sean-k-mooney | ya so i have openstakc-nova as default room for several years | 13:49 |
fungi | so i can't speak to the experience with that specifically, but i assume it would work like any other matrix room | 13:49 |
fungi | since on the matrix side that's all it is | 13:49 |
sean-k-mooney | ya | 13:50 |
sean-k-mooney | ok well good to know. i might look into it if i have time. im not agianst matirx but on the other hand i dont really want to have to be in more rooms. the reaosn i stead with irc in the past was latency in messages showing up | 13:51 |
sean-k-mooney | it weas fine 90% and then ocationall it used to lag for a a bit and then go back to being fine | 13:51 |
sean-k-mooney | but its been a year or two since i tired useign it day to day | 13:51 |
opendevreview | Elod Illes proposed openstack/project-config master: [release-tools] Improve tagging logs https://review.opendev.org/c/openstack/project-config/+/954340 | 13:54 |
sean-k-mooney | looks like only hte python version is aviable in nix currenly so while i shoudl jsut need to drop libmatrix.so in .weechat/plugins/ compiling that and testing it is mroe then i want to do right now :) | 13:56 |
fungi | yeah, i just `cargo build && mv ...` | 14:03 |
elodilles | just out of curiosity: project-config looks like not require 'signed-off-by' tag. is that intentional? :-o | 14:03 |
fungi | elodilles: yes, it didn't previously require the cla either | 14:03 |
fungi | basically none of the openstack tact sig repos require it | 14:03 |
elodilles | oh, i see | 14:04 |
fungi | it's strictly necessary for all official openstack deliverable repositories | 14:04 |
fungi | but not e.g. sigs repos or other things that aren't deliverables | 14:04 |
elodilles | (oh, btw, the root cause of the missing 2023.2-eol tags is trivial: neither puppet-openstack-integration nor puppet-openstack-spec-helper is cycle-trailing project /o\) | 15:23 |
fungi | hah, makes sense | 15:24 |
sean-k-mooney | did ye update gitea? it looks prettier then i remember | 16:26 |
clarkb | yes the latest version added fancier icons for files/dirs and also the file browser has a navigation bar on the left now | 16:27 |
clarkb | we do our best to keep up with upstream releases | 16:27 |
sean-k-mooney | ya thats what i was noticiceign i twas like no this is my browser not my ide/editor | 16:28 |
fungi | we upgraded on 2025-06-23, a little over 2 weeks ago | 16:29 |
sean-k-mooney | ack i still typicly use the github mirror more so i just noticed it now | 16:32 |
opendevreview | Dmitriy Rabotyagov proposed openstack/project-config master: Switch translate proposal jobs to noble https://review.opendev.org/c/openstack/project-config/+/954370 | 19:41 |
opendevreview | Jeremy Stanley proposed openstack/project-config master: Remove CLA enforcement from all projects and lock https://review.opendev.org/c/openstack/project-config/+/954374 | 21:04 |
opendevreview | Ian Y. Choi proposed openstack/project-config master: Move to use prepare-weblate-client job https://review.opendev.org/c/openstack/project-config/+/954377 | 21:17 |
opendevreview | Ian Y. Choi proposed openstack/openstack-zuul-jobs master: [WIP] Add ansible play for weblate client configuration https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/921878 | 21:35 |
opendevreview | Ian Y. Choi proposed openstack/openstack-zuul-jobs master: [WIP] Add ansible play for weblate client configuration https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/921878 | 22:04 |
opendevreview | Ian Y. Choi proposed openstack/openstack-zuul-jobs master: [WIP] Add ansible play for weblate client configuration https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/921878 | 22:32 |
opendevreview | Merged openstack/project-config master: Switch translate proposal jobs to noble https://review.opendev.org/c/openstack/project-config/+/954370 | 22:45 |
Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!