Tuesday, 2025-12-02

opendevreviewMerged openstack/openstack-manuals master: Fix bug tracker link for heat  https://review.opendev.org/c/openstack/openstack-manuals/+/96902006:33
opendevreviewTakashi Kajinami proposed openstack/openstack-manuals master: Add bug tracker link for placement  https://review.opendev.org/c/openstack/openstack-manuals/+/96904012:02
opendevreviewTakashi Kajinami proposed openstack/openstack-manuals master: Add bug tracker link for placement  https://review.opendev.org/c/openstack/openstack-manuals/+/96904012:03
opendevreviewMerged openstack/openstack-manuals master: Add bug tracker link for aodh  https://review.opendev.org/c/openstack/openstack-manuals/+/96903912:06
mnasiadkaspotz[m]: Would appreciate some help - mod_auth_mellon in EPEL10 hasn’t budged even a bit - I’m happy to do the extra mile of making it work - but I’m clueless about being a Fedora maintainer - but it seems Neil Hanlon has applied for being a co-maintainer without any answer - https://bugzilla.redhat.com/show_bug.cgi?id=232653412:25
mnasiadkaspotz[m]: If you can help a bit - I’d be grateful12:26
spotz[m]Yeah let me ping14:25
spotz[m]Pinged but might not be around yet, one is PST14:31
mnasiadkaThanks, no rush, it’s been waiting since July :)14:41
spotz[m]Well we're going into the holidays which means it could sit until next year depending on PTO, shutdown, etc14:54
spotz[m]mnasiadka: They said to ping on epel-devel@lists.fedoraproject.org or in the bug to increase visibility but that it's definitely had enough time. I'd also put in your offer to help15:00
mnasiadkaLet me try both - thanks :)15:01
spotz[m]You're welcome15:17
gouthamrtc-members: a gentle reminder that our weekly IRC meeting will happen here in ~57 minutes16:03
gouthamr#startmeeting tc17:00
opendevmeetMeeting started Tue Dec  2 17:00:21 2025 UTC and is due to finish in 60 minutes.  The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot.17:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.17:00
opendevmeetThe meeting name has been set to 'tc'17:00
gouthamrWelcome to the weekly meeting of the OpenStack Technical Committee. A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct.17:00
gouthamrToday's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee17:00
gouthamr#topic Roll Call17:00
spotz[m]o/17:00
gtemao/17:01
bauzaso/ (but I'll barely pay attention)17:01
cardoeo/17:01
gouthamrcourtesy-ping: noonedeadpunk, frickler, mnasiadka17:02
mnasiadkao/17:02
gouthamrnoted absence: t o n y b17:02
noonedeadpunko/17:04
noonedeadpunksorry17:04
gouthamrno worries, lets get started.. 17:05
gouthamr#topic Last Week's AIs17:05
gouthamrwe chatted a bit about changing the FIPS goal 17:05
gouthamr#link https://review.opendev.org/c/openstack/governance/+/969145 (Move FIPS Compliance goal back to proposed)17:06
gouthamrthere's a ton of content on the commit message - i tried capturing the concerns during the PTG17:06
gouthamri can address the comments from frickler, but hoping this can get us started with whatever needs to be done regarding this17:06
gouthamrthe next AI was around documenting the PTL appointments, and preserving them through leadership style switches17:07
gouthamrstill assigned to to ny b; will follow up outside this meeting17:08
gouthamrgtema took an AI regarding passlib - i.e., to write to the list and chase some patches to the requirements17:08
gouthamr#link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/SGMP7SKKBOCKIH2FPFH53HYHNCHSRABS/ ([tc][all][security] Supporting Post-Quantum Cryptography in OpenStack code (all projects))17:09
gouthamr^ the details are here17:09
gtemapasslib change requires +w17:09
gtemaothers (chained) got it already17:09
gouthamrack, ty for working on this17:11
gtemawlcm17:11
gouthamrthe last AI i see is around the openstack-manuals patch regarding tracking metrics:17:12
gouthamr#link https://review.opendev.org/c/openstack/openstack-manuals/+/90561417:12
gouthamr^ we merged this with a comment, but, not against removing the tracking itself in teh future17:13
gouthamrthat's all the AIs i was tracking, was anyone working on anything else?17:13
cardoeNo. I've closed down the one about the AI tag in the commit.17:16
gouthamri missed it cardoe 17:16
gouthamrsorry, would need you to elaborate :) i didn't see anything on gerrit?17:18
gouthamrwe can catch up async.. lets move to the next topic17:22
gouthamr#topic Approaches to making big changes17:22
cardoeSorry. Drive by.17:22
gouthamri wanted to bring this up here because there were some interesting ideas17:22
cardoeSo I had asked if we should have a gerrit header like assisted-by to state what kind of assistance you did from AI.17:23
cardoeBut after getting feedback from folks I think it's not a great idea. So putting that down.17:23
gouthamroh, i get it now17:23
gouthamrty cardoe, will catch up after this topic if we have room or after the meeting17:24
cardoe+117:24
gouthamralright, around this $topic17:25
gouthamrfully agree with many sentiments that clarkb shared over the initial message, i felt like we needed to put it down somewhere, akin to this document here:17:26
gouthamr#link https://governance.openstack.org/tc/reference/technical-vision.html?_ga=2.42535660.850854672.1745410202-224944396.1732558908 17:26
* gouthamr GAH, google analytics17:26
gouthamr#undo17:26
opendevmeetRemoving item from minutes: #link https://governance.openstack.org/tc/reference/technical-vision.html?_ga=2.42535660.850854672.1745410202-224944396.173255890817:26
gouthamr#link https://governance.openstack.org/tc/reference/technical-vision.html17:26
gouthamrthe core argument was derived from several ongoing threads that seemed to indicate that our community's existing policies and procedures are merely "bad bureaucracy"17:27
gouthamrwe definitely need a better process to update them, and i was thinking that we headed there with the "ideas" work17:28
gouthamr#link https://governance.openstack.org/ideas/17:28
clarkbyes, I was hoping to point out that the rules we've got exists for reasons, but those reasons may not apply any longer and our goals may have shifted. We need to be able to update policy to accomodate. And ideally we'd do so in a way that continues to preserve the four opens rather than leading to people just secretly doing a bunch of work and saying take it or leave it17:29
spotz[m]So use ideas repo vs someone submitting a change to the vision?17:29
gouthamrclarkb: +117:29
fungiideas was the replacement for the more formal community specs we had prior, and then later evolved into the current goals framework17:30
gouthamrspotz[m]: nope, not suggesting that we use it, i wanted to know what the intent of it was, and if we should explore that route again17:30
gouthamrfungi: wow; but, a goal is an accepted idea with a definitive timeline? its not a place for brainstorming is it? 17:31
fungiwhat was attractive, and also ultimately regrettable, about the ideas process is that it allowed people to suggest work for others to do and not take responsibility for making happen themselves17:32
gouthamrwe'll have a lot of that, even with goals.. but, its valuable, isn't it17:33
fungii guess it's a matter of opinion, at least in most open source communities any suggestion that doesn't come with a commitment to invest one's time on typically fails to get implemented17:35
fungiwhich is why goals at least require a champion to project-manage them before they can be accepted17:36
mnasiadkaWell, we have active goals that are not so active - so even having a champion doesn’t warrant that17:37
fungiright, so if even the accepted goals aren't a priority to implement...17:37
gouthamrmnasiadka: i think actively pushing a goal back to "proposed" or "abandoned" would help17:38
fungias happened with the fips testing goal, and yes that was a good move17:38
gouthamrspitballing - we need to let innovation happen with the mindset that something will always challenge norms/guidelines.. but, it's okay, because we can change these norms/guidelines by providing sound reason..17:40
gouthamr 17:40
mnasiadkagouthamr: sure, but I also understand frickler’s comments about what happens if nobody ever picks that up - and we have ,,artefacts’’ of previous implementation attempt in the ecosystem17:40
gouthamrmnasiadka: yes, its a bother cleaning them up, but, it taught us some valuable questions to ask when accepting a community goal17:41
mnasiadkaActually FIPS is something I never understood, given the ,,target audience’’ of that feature - but let’s not open that discussion :)17:42
clarkbfwiw the goals behind policies are at a higher level than the project goals implementation imop17:42
clarkb*imo17:42
gouthamryes17:42
clarkbthings like the use cases we consider fundamental, the platforms we support, our approach to testing (gating, etc), and so on17:43
fungii guess what i don't understand is why adjustments to policies and processes can't start out as a discussion somewhere that leads to one or more changes in the governance repo17:44
mnasiadkaI wanted to ask what’s the problem in discussing needed changes in policies and doing the change in governance repo17:45
clarkbI don't think there is one. I think the problem is more that there is either a perception or reality that if one does so you will always be shot down17:45
gouthamrthey surely can, we're not forcing one pattern of making change.. but the governance repo doesn't capture _everything_ and surely not "half baked" things17:45
fungiwrapping that in yet another process seems like propagating the existing complaint that we have too many processes17:46
clarkband this may be a case where explicitly working through what we've got that we know isn't working anymore or is no longer reality and leading by example may help break down that perception/reality17:46
clarkbI think the problem is more that we don't practice this as a thing we do17:47
clarkbso everyone assumes it isn't something that can be done17:47
clarkbwhcih is why in my email I suggested maybe we just start doing it every 6 months or something17:47
clarkbthe golang platform is a perfect example for where reality doesn't match policy imo17:48
clarkbthere is no golang code in openstack as far as I know. Why do we keep that as a valid option?17:48
clarkbif we start from there we can reevaluate the origianl goals behind adding go and consider if those are still valid. If so then we know why we keep it is an otpion. Otherwise we can pivot17:49
spotz[m]Ok so is there a problem with policy? Or a problem with perception and potentially documentation?17:50
clarkbspotz[m]: I think both, but one leads to the other17:50
clarkbfrom my perspective our policy is bit rotting17:50
clarkband there isn't an effort to address it because the perception is that doing so is futile17:50
mnasiadkaWell, the vision document doesn’t really look like a policy - or we’re talking about some different document than the one gouthamr linked?17:51
clarkbmnasiadka: I think you start from that vision document to determine what is important for policy to address17:51
clarkbmnasiadka: for example that document doesn't metnion python17:51
clarkbto me that implies that python is an implementation detail. There are many good reasons to use python, but maybe we don't only have to use python etc17:52
clarkb(or go or ...)17:52
mnasiadkaWell, the question if we use python, or something else - is mostly dictated by the testing infrastructure - currently only Python is in PTI IIRC17:53
gouthamryes; at some point, we defined "PTI" as complementing that vision.. and we had a python pti, golang pti etc 17:53
spotz[m]Ok so it sounds like a review and a re-write of the vision is the first place to start, then increase communication, etc17:53
gouthamrmnasiadka: https://governance.openstack.org/tc/reference/pti/golang.html 17:53
fungithe vision is due to be revised anyway if the tc wants to maintain one. the entire point of vision statements (based on the workshop we attended when we drafted the first one) is that they're supposed to be time-bounded17:53
gouthamr(and this: https://governance.openstack.org/tc/reference/pti/javascript.html)17:53
mnasiadkagouthamr: thanks, I was blind :)17:53
clarkbspotz[m]: yes I think that is a reasonable first step17:54
gouthamri agree, i would do it, but poorly, i wonder if, clarkb you'd like to take a stab or work with me on it? 17:56
spotz[m]So my suggestion is to focus there if everyone is in agreement on that being the first step. Start small and then grow from there or we won't make progress on the whole issue17:56
fungithe gist of the workshop was that we were supposed to imagine the state of the tc five years out, and then describe that in a way that all the tc members could agree on17:56
clarkbgouthamr: I'm happy to help. I don't know that I can drive it. But happy to review and draft as I'm able17:57
gouthamrfungi: i need notes of this workshop :D 17:57
fungii'll go spelunking for them17:57
gouthamri haven't said it enough, we need to index your brain into an OpenStack History spanning 500+ volumes17:57
spotz[m]Yeah I'm not sure what workshop17:57
clarkbthe openstack tc went to a deli in michigan years and years ago and did a leadership workshop17:58
clarkba major output of that workshop was the vision document iirc (I wasn't there)17:58
fungi#link https://review.opendev.org/c/openstack/governance/+/45326217:58
fungiit was driven by the stewardship working group (which no longer exists)17:59
fungi#link https://wiki.openstack.org/wiki/Stewardship_Working_Group#Build_a_vision_for_the_TC17:59
fungithe thing at the deli in michigan was put on by the same group, but was on servant leadership18:00
clarkbah separate activities then18:00
fungithe vision drafting mini-workshop happened at an in-person meeting in boston18:00
spotz[m]names that are definitely a blast from the past18:00
gouthamrhaha, good reading.. ty for digging these up18:01
fungitons of review comments on the vision draft change18:01
gouthamri suggest we reflect on this in light of the original question and continue this discussion in the next meeting 18:02
spotz[m]Or in a deli in Michigan:)18:02
fungi(the boston meeting was in the dla piper offices, if memory serves)18:03
gouthamrhaha, /me thinks of pastrami sandwiches or bagels and lox  18:03
gouthamrwe're over by a few mins.. but, this was an insightful discussion that was worth doing 18:03
gouthamranything to note for the minutes today?18:04
mnasiadkaIt’s 7pm here, don’t make me hungry18:04
gouthamrmnasiadka: something i learned in the US, breakfast is meant to be eaten all day - ignore anyone that says otherwise :D 18:04
spotz[m]I'm a pro at scheduling meetings in the Boston RH office18:04
gouthamralright, thank you for joining/participating everyone! please continue the chatter here.. 18:05
fungii've been eating breakfast for 7 hours nonstop18:05
gouthamr:D18:05
gouthamr#endmeeting18:05
opendevmeetMeeting ended Tue Dec  2 18:05:56 2025 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)18:05
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tc/2025/tc.2025-12-02-17.00.html18:05
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tc/2025/tc.2025-12-02-17.00.txt18:05
opendevmeetLog:            https://meetings.opendev.org/meetings/tc/2025/tc.2025-12-02-17.00.log.html18:05
gouthamr^ that was important to get into the minutes18:06
fungiabsolutely18:06
spotz[m]hehe18:07
fungiand yeah, for a little more background, in early 2018 there was an in-person openstack foundation (before the rename) board meeting which included time for a joint meeting with the board and openstack project leadership. the board also had a day of private executive session during which time the tc did the vision workshop with the zingerman's deli trainers18:09
fungier, early 2017 i mean18:09
spotz[m]Yeah the in-person board meetings ended with COVID at least as a separate meeting, there have been a few at events18:10
fungifor two days we had a couple of large, beautiful, conference rooms in the highrise office building for the foundation's legal counsel (dla piper)18:10
fungihttps://www.zingtrain.com/article/why-and-how-visioning-works/18:12
fungithat's the folks that put on the mini-workshop for the tc, and then a number of openstack community leaders (not just tc members) also separately attended a multi-day workshop at the zingerman's headquarters in ann arbor18:14
fungithough that later workshop was around https://www.zingtrain.com/article/servant-leadership/18:15
fungioh, and the lunches were (perhaps unsurprisingly) amazing, of course18:50
gouthamrgood read on a situation we're all too familar with :( 22:05
gouthamrhttps://www.theregister.com/2025/12/02/ingress_nginx_opinion/22:05
clarkbI think part of the problem there is that nginx has their own direct competitor.22:16
fungiwhich is only available under a proprietary license right?22:17
fungior am i confusing it with something else maybe...22:18
mnasernginx https://github.com/nginx/kubernetes-ingress apache2.0 license22:18
mnaserkubernetes (cncf?) https://github.com/kubernetes/ingress-nginx also apache 2.0 license22:18
mnaserbut afaik the nginx (corp one) has some features that work only with nginx "pro" which is the non-community version of nginx22:19
clarkbya its open core aiui https://docs.nginx.com/nginx-ingress-controller/install/license-secret/ you need secrets for the plus version22:19
fungiah okay. so i guess the concern is that it's not a drop-in replacement22:19
mnaserwell to be more clear, nginx itself is the part htat need to be licenesed22:19
clarkbfungi: well and that it canablizes the community that is most likely to be able to maintain the software22:20
mnaserthe funny thing is22:20
mnasermost people used the upstream one, not the nginx one22:20
mnaserand i guess nginx folks never pushed things into the upstream one because they wnat to be able to add features to it that support their nginx pro22:21
fungisounds like that community didn't maintain the software though, so i can understand the reluctance to pretend they suddenly will22:21
fungiand if that community really does want to maintain it after all, it's open source, they can just fork it under a new name and move forward22:22
mnaserthe other part is there is a new "way" that replacing ingress by gateway api in k8s22:22
mnaserand so even if itw as maintained, it'll eventually disappear, so i can understand maintainers being more interested by building something that supports gateway api rather than supporting outgoing software22:23
clarkbI suspect the biggest pain here is simply the short timeline. There are alternatives and converting for a lot of stuff is likely to be straightforward. Then you've got the long tail of complicated configs22:30

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