Friday, 2025-10-24

opendevreviewTony Breeds proposed zuul/zuul-jobs master: [tox-siblings] pre-create the logfile  https://review.opendev.org/c/zuul/zuul-jobs/+/96473501:23
*** mrunge_ is now known as mrunge06:48
stephenfinclarkb: I don't know if we can drop anything. While new pip does things efficiently, I don't know if older pip (e.g. pip 9.x on Ubuntu 18.04) does10:27
opendevreviewMatthieu Huin proposed zuul/zuul-jobs master: merge-output-to-logs: add option to also copy dotfiles  https://review.opendev.org/c/zuul/zuul-jobs/+/96476110:38
clarkbstephenfin: ack that makes sense13:20
opendevreviewClark Boylan proposed zuul/zuul-jobs master: [tox-siblings] pre-create the logfile  https://review.opendev.org/c/zuul/zuul-jobs/+/96473513:25
opendevreviewClark Boylan proposed opendev/system-config master: Explictly install docker.io package  https://review.opendev.org/c/opendev/system-config/+/96364213:36
opendevreviewClark Boylan proposed opendev/system-config master: Force gitea http(s) connectivity through the load balancer  https://review.opendev.org/c/opendev/system-config/+/96472813:36
clarkbstephenfin: I'm slowly working my way through the stack of PBR changes again and notice the second to the last is failing tests so something may have happened in the later portions of the refactor13:45
clarkbstephenfin: and the diffs look a lot cleaner thank you!13:51
clarkbstephenfin: ya everything looks good up until https://review.opendev.org/c/openstack/pbr/+/964715 where testing gets unhappy14:00
clarkbzm05 got Error response from daemon: {"message":"unauthorized: authentication required"} when pulling zuul container images. That would be against quay.io so I guess be on the lookout for any issues with aws again?14:11
clarkbnone of the other zuul servers hit that though so I think it must be rare if persistent14:11
fungiodd, i guess that's the response you get when the url goes to a nonexistent s3 bucket or something?14:13
clarkbor maybe its a cdn issue?14:15
clarkbI should note that error occurred in the 1400 hourly infra-prod job14:17
clarkbfungi: did that email about pushing video files to gerrit need to be moderated or are they subscribed now?14:22
clarkbnevermind headers says nonmember-moderation was hit so I don't think they are subscribed :/14:23
fungiyeah, i approved it through moderation and it indicated the reason for holding is that the poster is not subscribed14:23
clarkbI'm just worried we're going to end up playing a bad game of telephone if they say respond to sean directly14:24
clarkbso was hoping they had taken the time to subscribe finally. But this is the same repo I've tried to reach out to them about in the past too14:24
clarkbthoguh I guess I didn't send email and asked horace to reach out since email has been a struggle repviously14:25
clarkbthe prior emails were about the git history in a different repo being "corrupted" and I asked them to revert rather than have us force push to correct14:25
clarkbbut this repo is the one that is already larger than nova because they keep stashing binary artifacts in it14:26
clarkbI wonder if we should force this thread to horace and see if he can broker some sort of discussion that is a bit more productive one way or another14:26
clarkbthoughts?14:26
fungiwell, i do agree, but also if git-review is failing to push a commit then that likely indicates a bug somewhere14:33
fungialso sean has helpfully made your case14:35
stephenfinclarkb: fixed14:49
stephenfinConflict between an import and a function defined in that file14:49
stephenfinI've also spotted a few more pkg_resources calls in the functional tests, but I've deferred fixing those yet. I'll get to them next week14:50
clarkbstephenfin: ack thanks I'll look at that change in a bit14:51
clarkbfungi: right I'm not discounting a bug more pointing out that these users have had potentially problematic behaviors and are difficult to communicate with so I'm hoping we can figure out some way to communicate better and open to idea on that front14:51
clarkbfungi: it could be something as simple as the great firewall blocking the connection after so many bytes14:51
clarkbor it could be gerrit exploding on something it doesn't like (I haven't looked at the server logs)14:52
clarkb`computing-native` does not show up in the gerrit error log for today or yesterday14:54
fungiyeah, it could be any number of things of course14:58
opendevreviewClark Boylan proposed opendev/system-config master: Force gitea http(s) connectivity through the load balancer  https://review.opendev.org/c/opendev/system-config/+/96472816:01
opendevreviewClark Boylan proposed opendev/system-config master: Force gitea http(s) connectivity through the load balancer  https://review.opendev.org/c/opendev/system-config/+/96472816:04
fungigonna go grab lunch with some friends, won't be too long16:04
opendevreviewClark Boylan proposed zuul/zuul-jobs master: [tox-siblings] pre-create the logfile  https://review.opendev.org/c/zuul/zuul-jobs/+/96473516:24
opendevreviewMerged zuul/zuul-jobs master: Find tox/nox siblings with normalized package names  https://review.opendev.org/c/zuul/zuul-jobs/+/96472417:19
clarkbrosmaita: ^ fyi I think your stable branch job should be happy now17:20
rosmaita\o/17:35
opendevreviewMerged zuul/zuul-jobs master: Install the docker-compose-plugin when installing upstream docker  https://review.opendev.org/c/zuul/zuul-jobs/+/92308317:42
opendevreviewClark Boylan proposed opendev/system-config master: Force gitea http(s) connectivity through the load balancer  https://review.opendev.org/c/opendev/system-config/+/96472818:26
opendevreviewClark Boylan proposed opendev/system-config master: Force gitea http(s) connectivity through the load balancer  https://review.opendev.org/c/opendev/system-config/+/96472819:42
clarkboh cool ^ passes now21:48
clarkbso I think that is an option we can discuss next week during the team meeting (whcih I think we'll have since we cancelled this week)21:48
fungisounds good, thanks for working through that21:48
clarkbfungi: if you find time reviewing that stack of pbr updates would probably be helpful. I think only the first half or so are what we expect to need for setuptools deprecation updates21:54
clarkbthen subsequent changes are further cleanups working towards an eventual state where setup.py won't be needed21:55
fungihopefully i'll have time on monday, all the ptg sessions i was planning to hit are during the other 4 days of the week anyway22:12
*** dmellado9 is now known as dmellado22:37

Generated by irclog2html.py 4.0.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!