| @adam:sheeplauncher.net | ah I didn't realize we had started switching to nox! interesting | 02:15 |
|---|---|---|
| @adam:sheeplauncher.net | so rather than try to work around the tox issue, we should just... switch the projects to use nox? that seems like a very good task for an LLM I'd guess | 02:15 |
| @adam:sheeplauncher.net | unless it's more nuanced than I'd expect? but nox is designed to be drop-in for tox right? | 02:16 |
| @adam:sheeplauncher.net | or is https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/1013446 the answer? | 02:19 |
| @adam:sheeplauncher.net | agent seems to think we need additional shared template work in openstack-zuul-jobs for nox, i'm not sure if that's true | 03:07 |
| @adam:sheeplauncher.net | I guess this doesn't look too crazy: https://paste.opendev.org/show/b5CuVQPWSz3z3NJYp8eJ/ | 03:13 |
| @adam:sheeplauncher.net | and I guess https://paste.opendev.org/show/bIGHI9qCPoOUQdBnRdHD/ | 03:13 |
| @adam:sheeplauncher.net | if that's the right approach I can post those | 03:13 |
| @adam:sheeplauncher.net | eh i'll just post them as WIP, can abandon easily | 03:19 |
| @adam:sheeplauncher.net | will prove if it works or not | 03:20 |
| -@gerrit:opendev.org- Adam Harwell proposed wip: [openstack/project-config] 1013452: ci: trial nox publishing jobs https://review.opendev.org/c/openstack/project-config/+/1013452 | 03:30 | |
| @adam:sheeplauncher.net | huh, gerrit bot looks nicer here IMO when posting about changes that opendevreview bot does in other channels, interesting | 03:36 |
| @tafkamax:matrix.org | This is a Matrix channel and other channels are IRC, that's why I presume. | 07:16 |
| @adam:sheeplauncher.net | ah | 07:36 |
| @adam:sheeplauncher.net | I think some others are matrix too like neutron, but just they use a different bot | 07:37 |
| @tafkamax:matrix.org | Could be, I am just an observer-bot 😁 | 07:37 |
| @adam:sheeplauncher.net | anyway, I posted two changes that may be workable, but I'm not overly attached to any one method, I assume ya'll know the best path forward: | 07:38 |
| https://review.opendev.org/c/openstack/project-config/+/1013452 | ||
| https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/1013453 | ||
| and it looks like I may need a third here too... | ||
| @adam:sheeplauncher.net | (for the tox to nox thing) | 07:39 |
| -@gerrit:opendev.org- Adam Harwell proposed wip: [zuul/zuul-jobs] 1013467: fix(nox): ignore session descriptions https://review.opendev.org/c/zuul/zuul-jobs/+/1013467 | 07:54 | |
| -@gerrit:opendev.org- Michal Arbet proposed: [zuul/zuul-jobs] 1013473: tox: make package staging path private https://review.opendev.org/c/zuul/zuul-jobs/+/1013473 | 08:23 | |
| @adam:sheeplauncher.net | Jens Harbott: saw your -2, that’s fine, just looking to discuss / brainstorm directions to go since folks here were saying we might want to just use this as an opportunity to go to nox, what do you actually think overall? Since we need to do SOMETHING now, would you prefer Michal’s approach instead? I have no dog in this race really lol | 11:39 |
| @adam:sheeplauncher.net | Just did a minimal PoC and it seems like it works, obviously there’s more to do | 11:40 |
| -@gerrit:opendev.org- yatin proposed: [zuul/zuul-jobs] 1013504: [tox-cover] Block tox 4.67.5 https://review.opendev.org/c/zuul/zuul-jobs/+/1013504 | 12:39 | |
| @harbott.osism.tech:regio.chat | afaict the idea to migrate to tox was meant for opendev \setminus openstack. we have denylisted the broken tox release for most openstack jobs, with some gaps still to be filled, I hope stephenfins upstream issue will be dealt with soon | 13:32 |
| @fungicide:matrix.org | yes, when i mentioned switching remaining uses from tox to nox i meant for *opendev's* projects (those in the `opendev/` git namespace, mainly), not other projects hosted in opendev | 14:01 |
| @fungicide:matrix.org | though the zuul community has already switched off tox in favor of nox as well | 14:02 |
| @fungicide:matrix.org | rm_you: the gerritbot you're seeing in openstack irc channels is using irc protocol because openstack is relying on irc, if you're joining those channels with a matrix client, you're seeing them through oftc's matrix bridge | 14:03 |
| @adam:sheeplauncher.net | yeah ok, so then we'll just patch up tox support for now and leave it alone? I can abandon those WIPs | 14:15 |
| @adam:sheeplauncher.net | it WOULD work though, but I can see how we might want that to be a larger discussion / community goal type thing | 14:15 |
| @adam:sheeplauncher.net | granted it... seems almost trivially mechanical with LLMs | 14:15 |
| @fungicide:matrix.org | also trivially mechanical anyway, an llm only complicates that work and gets in the way of you understanding the basic concepts (as it does with most things) | 14:22 |
| @fungicide:matrix.org | it's not about leaving us alone, opendev simply doesn't make those decisions for other communities. the openstack community has a technical committee, quality assurance team, and a testing and collaboration tools sig (they discuss such things in their #openstack-tc, #openstack-qa, and #openstack-infra channels respectively) | 14:25 |
| @adam:sheeplauncher.net | yeah could POSSIBLY just be pure-mechanical, but the LLM caught some stuff I didn't see at first so I feel like it had value | 14:27 |
| @adam:sheeplauncher.net | granted that's as you say, likely just because I am not super familiar with parts of how zuul works 😛 | 14:28 |
| @fungicide:matrix.org | though it's also worth being aware that the openstack community already considered switching from tox to nox back when tox v4 broke tons of projects, and their decision at the time was to stick with tox and update configuration to work with the new versions. i doubt they would come to a different conclusion over this recent behavior change | 14:28 |
| @adam:sheeplauncher.net | yep, makes sense | 14:29 |
| @adam:sheeplauncher.net | will just abandon these for now | 14:30 |
| @adam:sheeplauncher.net | a pin will cover it for the time being until they can resolve it one way or another | 14:30 |
| @jim:acmegating.com | nox is great and they love their users and it shows. if openstack wants to consider switching, that would be a swell idea. obviously, something for the TC to consider, it should be a concerted and deliberate effort. but we did all the work in the zuul-jobs repo to share, so it's ready whenever openstack is. | 14:48 |
| @jim:acmegating.com | it probably makes sense to just fix the current thing with tox -- but once that's done, why not consider switching to nox between now and the next time everything breaks? :) | 14:49 |
| @adam:sheeplauncher.net | yeah unfortunately when there's no urgency, nothing happens T_T | 15:02 |
| @adam:sheeplauncher.net | that's why I jump on stuff like this as opportunities to ram progress through 😛 | 15:02 |
| @adam:sheeplauncher.net | I wonder if we could land some of the stuff that'd be required as a precursor (like the two changes I had linked earlier) which would ALLOW projects to opt into switching? | 15:03 |
| @adam:sheeplauncher.net | and leave it up to the cores to choose? or if it would really be something that MUST be decided at a SIG leve | 15:03 |
| @adam:sheeplauncher.net | *level | 15:04 |
| @adam:sheeplauncher.net | I'm not sure if there was like an official "we use tox" directive, I guess it's likely there was | 15:04 |
| @jim:acmegating.com | well, i think the main thing is to try to keep things standardized so it's easier to manage. at least, that was the rationale when i worked on this more. | 15:04 |
| @adam:sheeplauncher.net | yeah I am so used to ya'll being "the openstack infra folks" that I forget that isn't your only focus now | 15:04 |
| @jim:acmegating.com | weirdly, the project testing interface only specifies 'tox' in the docs and releasenotes sections: https://governance.openstack.org/tc/reference/project-testing-interface.html | 15:06 |
| @jim:acmegating.com | i think it used to be more prescriptive for unit test jobs | 15:07 |
| @jim:acmegating.com | it kind of reads like the tc decided to give the individual projects more freedom to chose their own test runners | 15:10 |
| @jim:acmegating.com | oh wait here it is: | 15:10 |
| @jim:acmegating.com | https://governance.openstack.org/tc/reference/pti/python.html | 15:11 |
| @jim:acmegating.com | rm_work: so that's the document all the projects should be working from ^ | 15:11 |
| @jim:acmegating.com | and that's what the tc would need to agree to change in order for projects to make the switch | 15:12 |
| @jim:acmegating.com | (minor usability note: i didn't find the correct sub-page because it was under the "Linux Distributions" section heading) | 15:13 |
| @adam:sheeplauncher.net | Heh k | 15:15 |
| @adam:sheeplauncher.net | Well that’s fine for now | 15:15 |
| -@gerrit:opendev.org- Takashi Kajinami proposed: [openstack/diskimage-builder] 1013557: Migrate setup configuration to pyproject.toml https://review.opendev.org/c/openstack/diskimage-builder/+/1013557 | 15:25 | |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [openstack/project-config] 1013558: Add gitiles repo to zuul config https://review.opendev.org/c/openstack/project-config/+/1013558 | 15:29 | |
| @mordred:waterwanders.com | Clark: that ^^ isn't important for opendev until 3.15, but it would be handy for me now | 15:33 |
| @mordred:waterwanders.com | but we'll also need to add it to required projects and the repo playbook once we make 3.15 jobs | 15:33 |
| -@gerrit:opendev.org- Zuul merged on behalf of Monty Taylor https://matrix.to/#/@mordred:inaugust.com: [openstack/project-config] 1013558: Add gitiles repo to zuul config https://review.opendev.org/c/openstack/project-config/+/1013558 | 15:50 | |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [opendev/system-config] 1013582: Simplify gerrit container image https://review.opendev.org/c/opendev/system-config/+/1013582 | 17:31 | |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1013589: Disable IPv4 options to mitigate CVE-2026-98432 https://review.opendev.org/c/opendev/system-config/+/1013589 | 18:19 | |
| -@gerrit:opendev.org- Clark Boylan proposed: | 18:24 | |
| - [opendev/system-config] 1013590: Set go options in node-exporter container https://review.opendev.org/c/opendev/system-config/+/1013590 | ||
| - [opendev/system-config] 1013591: Set go options in codesearch hound container https://review.opendev.org/c/opendev/system-config/+/1013591 | ||
| - [opendev/system-config] 1013592: Set go options in prometheus containers https://review.opendev.org/c/opendev/system-config/+/1013592 | ||
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1013593: Set go options in jaeger container https://review.opendev.org/c/opendev/system-config/+/1013593 | 18:25 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1013594: Set go options in grafana container https://review.opendev.org/c/opendev/system-config/+/1013594 | 18:25 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1013595: Set go options in mailman anubis container https://review.opendev.org/c/opendev/system-config/+/1013595 | 18:25 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1013596: Set go options in gitea containers including anubis https://review.opendev.org/c/opendev/system-config/+/1013596 | 18:25 | |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1013589: Disable IPv4 options to mitigate CVE-2026-98432 https://review.opendev.org/c/opendev/system-config/+/1013589 | 18:36 | |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: | 18:37 | |
| - [opendev/system-config] 1013582: Simplify gerrit container image https://review.opendev.org/c/opendev/system-config/+/1013582 | ||
| - [opendev/system-config] 1013598: Add gitiles to gerrit master build https://review.opendev.org/c/opendev/system-config/+/1013598 | ||
| @mordred:waterwanders.com | fungi: wow - I'm glad that iptables line isn't completely cryptic or anything | 18:38 |
| @fungicide:matrix.org | for the record, i'm a big ipfilter/packetfilter syntax fan, hate ipchains/iptables | 19:01 |
| @fungicide:matrix.org | nftables is a big improvement, at least | 19:01 |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1013589: Disable IPv4 options to mitigate CVE-2026-98432 https://review.opendev.org/c/opendev/system-config/+/1013589 | 19:13 | |
| -@gerrit:opendev.org- Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org proposed: [opendev/system-config] 1013589: Disable IPv4 options to mitigate CVE-2026-98432 https://review.opendev.org/c/opendev/system-config/+/1013589 | 19:34 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1013605: Update etherpad to 3.3.13 https://review.opendev.org/c/opendev/system-config/+/1013605 | 21:05 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 1013605: Update etherpad to 3.3.13 https://review.opendev.org/c/opendev/system-config/+/1013605 | 21:06 | |
| -@gerrit:opendev.org- Clark Boylan proposed: [opendev/system-config] 840972: DNM force etherpad failure to hold node https://review.opendev.org/c/opendev/system-config/+/840972 | 21:07 | |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [opendev/system-config] 1013582: Simplify gerrit container image https://review.opendev.org/c/opendev/system-config/+/1013582 | 21:13 | |
| @mordred:waterwanders.com | fwiw - 1013598 turned the non-voting master build green again - its most recent failure is here: https://zuul.opendev.org/t/openstack/build/b27fcdd0992d495590dfb03f51275d84 - so I think that one is worth going ahead and landing, since it should help in the future | 21:19 |
| @mordred:waterwanders.com | my other container image patch is not quite ready yet | 21:19 |
| @clarkb:matrix.org | mordred: you only add the repo to the master build. Does 3.13 and 3.14 need that too or does the if -d {{item}} check in the playbook that is shared by all versions skip it safeyl on the older versions? | 21:22 |
| @mordred:waterwanders.com | Clark: they don't - it was added as part of the 3.15 release | 21:25 |
| @mordred:waterwanders.com | and the if -d check skips it safely on the other versions | 21:25 |
| @clarkb:matrix.org | ack so the playbook is backward compatible then | 21:25 |
| @mordred:waterwanders.com | yah | 21:25 |
| @clarkb:matrix.org | and ya ls gerrit/modules doesn't show gitiles on stable-3.13 | 21:26 |
| @mordred:waterwanders.com | they're using it to power blame calculations - pre 3.15 they were using a release jar, but shifted to the submodule approach | 21:27 |
| @clarkb:matrix.org | I'll add that change to the meeting agenda under the gerrit upgrade planning topic | 21:38 |
| @clarkb:matrix.org | ok I've got the meeting agenda updated. Was there anything else we'd like to see included in that before I send it out later today? | 21:59 |
| @fungicide:matrix.org | none i can think of, just everything's on fire so make sure to bring your fireproof boots | 22:01 |
| @clarkb:matrix.org | I think mine have already burned up | 22:02 |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [opendev/system-config] 1013582: Simplify gerrit container image https://review.opendev.org/c/opendev/system-config/+/1013582 | 22:08 | |
| @fungicide:matrix.org | https://pypi.org/project/pip/26.3/ now out, keep an eye for more broken i guess | 22:29 |
| -@gerrit:opendev.org- Monty Taylor https://matrix.to/#/@mordred:inaugust.com proposed: [opendev/system-config] 1013582: Simplify gerrit container image https://review.opendev.org/c/opendev/system-config/+/1013582 | 22:39 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Jeremy Stanley https://matrix.to/#/@fungicide:matrix.org: [opendev/system-config] 1013589: Disable IPv4 options to mitigate CVE-2026-98432 https://review.opendev.org/c/opendev/system-config/+/1013589 | 22:50 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [opendev/system-config] 1013596: Set go options in gitea containers including anubis https://review.opendev.org/c/opendev/system-config/+/1013596 | 22:50 | |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: | 22:50 | |
| - [opendev/system-config] 1013590: Set go options in node-exporter container https://review.opendev.org/c/opendev/system-config/+/1013590 | ||
| - [opendev/system-config] 1013591: Set go options in codesearch hound container https://review.opendev.org/c/opendev/system-config/+/1013591 | ||
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: | 22:50 | |
| - [opendev/system-config] 1013592: Set go options in prometheus containers https://review.opendev.org/c/opendev/system-config/+/1013592 | ||
| - [opendev/system-config] 1013593: Set go options in jaeger container https://review.opendev.org/c/opendev/system-config/+/1013593 | ||
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: | 22:57 | |
| - [opendev/system-config] 1013594: Set go options in grafana container https://review.opendev.org/c/opendev/system-config/+/1013594 | ||
| - [opendev/system-config] 1013595: Set go options in mailman anubis container https://review.opendev.org/c/opendev/system-config/+/1013595 | ||
Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!