Tuesday, 2026-11-03

*** M00SE7 is now known as M00SE12:53
opendevreviewJeremy Stanley proposed openstack/election master: Update OpenInfra member API endpoint  https://review.opendev.org/c/openstack/election/+/101390914:58
fungiis https://opendev.org/openstack/goal-tools/src/branch/master/goal_tools/foundation.py used for anything these days?15:08
fungilooks like dhellmann added it 8 years ago and it hasn't been touched since then15:08
opendevreviewElod Illes proposed openstack/openstack-manuals master: [www] Set 2025.1 Epoxy state to Unmaintained  https://review.opendev.org/c/openstack/openstack-manuals/+/101391615:11
fungihttps://opendev.org/openstack/governance/src/branch/master/tools/validate_acs.py was also added by dhellmann around the same time, but has been updated by a few other people until about 3 years ago, and gets run as part of the non-voting `openstack-tox-validate_acs` job (as recently as a week ago)15:16
opendevreviewJeremy Stanley proposed openstack/governance master: Update OpenInfra member API endpoint  https://review.opendev.org/c/openstack/governance/+/101392015:19
gtemagouthamr: cberendt and myself are on the conference now and will not join the meeting today16:02
gouthamrhey gtema, ack - ty for letting me know16:02
gouthamrtc-members: a gentle reminder that our weekly IRC meeting will be hosted here in ~55 minutes16:05
gouthamragenda is here: https://etherpad.opendev.org/p/openstack-technical-committee-meeting16:05
dansmithgouthamr: I have a conflict, so I will be distracted16:06
gouthamrdansmith: ack16:15
* gouthamr counts16:24
* frickler is around, just afk for a moment now16:58
gouthamrack frickler 17:00
gouthamr#startmeeting tc17:00
opendevmeetMeeting started Tue Nov  3 17:00:13 2026 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://etherpad.opendev.org/p/openstack-technical-committee-meeting17:00
gouthamr#topic Roll Call17:00
noonedeadpunko/17:00
cardoeo/17:01
gouthamrnoted absence: m n a s i a d k a, d a n s m i t h, g t e m a, C h r i s t i a n B e r e n d t17:02
gouthamrfrickler will be around17:02
gouthamrcourtesy ping: spotz [m]   17:02
spotz[m]o/17:02
frickler\o17:02
gouthamrgreat, that's everyone! thank you for joining, let's get started17:03
spotz[m]Beat you!17:03
gouthamr:P17:03
gouthamr#topic Last week's action items17:03
gouthamrwe had a big one: the rust proposal needed our collective attention17:03
gouthamr#link https://review.opendev.org/c/openstack/governance/+/998652 (Add Rust use-case analysis for Keystone)17:04
gouthamrthe timebox worked i think, most of us have some thoughts expressed on the proposal17:04
gouthamrthanks for taking the time to do that.. artem's set the patch to WIP; lets give him some time to address comments, and this topic can reprise here when the gerrit patch needs attention17:05
gouthamrwould you like to note anything here wrt this change?17:06
gouthamr#topic A check on gate health17:07
fricklertox break it, tox make it again ;)17:08
gouthamrah! ty for https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/1013477 frickler 17:08
fungiclassic weekend release17:09
fricklerah, I need to revert that again, too, yes17:09
gouthamr#link https://github.com/tox-dev/tox/issues/4357 17:10
fricklerwell the developer apologized for the trouble after the issue was fixed, what more can you expect? :)17:10
frickleralso, not directly gate related, but 2025.1 is now eom. may affect upgrade jobs for 2025.2 and 2026.1, though17:11
gouthamrack :) CI wastage on our end.. but thanks for catching the issue, and for the upstream conversation17:11
fricklercredit to the upstream action goes to stephenfin 17:12
gouthamron 2025.1 going eol, projects will need to drop grenade jobs in 2025.2  and SLURP ones on 2026.1, correct?17:12
fungiand the weird llm agent auto-proposing pull requests for filed bug reports within minutes17:13
gouthamri don't remember if there's any other way out of this17:13
fricklerwell either that or base them on the unmaintained branches17:13
fungigouthamr: eom not eol, but yes17:13
gouthamrfungi: you're awarding "weird" to an LLM agent now. i thought you had to show proof of life to earn that tag17:13
gouthamrah yeah, eom17:14
fungifair enough, i've earned my own weirdness fair and square17:14
frickleralso regarding unmaintained branches in general, someone volunteering to do some housekeeping on the older eom branches would be great. not sure if elodilles planned to do anything in that regard17:14
fricklerlike https://review.opendev.org/q/topic:%22create-2024.1%22+status:open would be a good list of branches to shut down IMO17:14
frickleror some zuul config errors like https://zuul.opendev.org/t/openstack/config-errors?branch=unmaintained%2Fyoga&skip=017:15
gouthamrack... worth another effort to kill these branches and save our resources (and perceptions) - because of the automatic opt-in, this cleanup also needs someone's time and effort17:17
gouthamrfungi has a related topic for later17:19
gouthamrwas there anything further regarding the gate to discuss?17:19
gouthamroh, fungi mentioned that m nasia dka has taken on the wiki.openstack.org revival 17:20
fungino current timeline on restoration still17:20
gouthamr++17:20
gouthamri think project teams are using this as an opportunity to clean up their docs pointing to the wiki - in some cases, they never meant to 17:21
gouthamrmeeting agendas can totally use etherpad i think, i plan to do that permanently with the TC's meeting, any objections there?17:21
fricklerdo you plan to create a new entry each week? otherwise we'll lose a bit of the history17:22
gouthamrwe have the meeting logs and the summaries on the discuss list17:22
fungii guess the timeslider in etherpad doesn't count17:22
clarkbwith opendev I hadn't planned to make a new list every week and instead rely on the timeslider and emailed agendas to serve as the historical record17:23
gouthamri could hit the star and create a time-stamped revision too17:23
fricklermaybe then include the agenda in the weekly mail instead of linking it? that would suffice for me17:23
fricklers/instead/in addition to/17:23
fungiyeah, opendev's meeting announcements just carry a verbatim copy of the agenda, and don't even mention the url to an editable version17:24
gouthamri can do that.. helps me enforce the discipline to not squeeze something in that 24-hour window :D 17:24
fungithere's always "any other business/open discussion" for anything that doesn't make it in time17:25
gouthamr++17:25
gouthamralright, let's move to the next topic17:26
gouthamr#topic TC Tracker, PTG Follow up17:26
gouthamr#link https://etherpad.opendev.org/p/tc-2027.1-tracker (Technical Committee activity tracker - 2027.1)17:26
gouthamrgoals need cleanup.. 17:29
gouthamr if you're haven't looked at these proposals, please do:17:29
gouthamr#link https://review.opendev.org/c/openstack/governance/+/1009912 (SRBAC goal completion)17:29
gouthamr#link https://review.opendev.org/c/openstack/governance/+/1009913 (Disband pop up team)17:29
gouthamri intend to workflow them this week, but can wait a couple of days if anyone needs time to17:29
gouthamrhberaud told me that he'd like to bring in damani as the goal champion for the remainder of the eventlet-removal goal; and i've poked ralonsoh for an update on the rootwrap->privsep migration goal.. I'll try and capture the conversation here regarding memory management strategies for privsep implementations17:30
gouthamrthat currently looks to me like bugs and fixes to each service that uses privsep, and does something wrong.. 17:31
gouthamrbut like mikal documented (thanks for doing it!), i think profiling and identifying the import tax would point to "bad" implementations to fix17:32
gouthamr#link https://www.madebymikal.com/chasing-openstack-memory-usage-the-memory-your-python-services-spend-on-code-they-never-run/ (privsep memory consumption notes)17:32
gouthamrbesides the other issue of using very broad capabilities, like the initial goal also called out:17:34
gouthamr#link https://governance.openstack.org/tc/goals/selected/migrate-to-privsep.html17:34
gouthamrbesides these, there are new goal proposals. I'd like to get them past the first "proposal" phase through reviews:17:36
gouthamr#link https://review.opendev.org/c/openstack/governance/+/1001777 (Add PQC Readiness as a proposed community goal)17:36
gouthamr#link https://review.opendev.org/c/openstack/governance/+/1009958 (document core reviewer processes)17:36
gouthamr^ both of these got some airtime at the PTG17:37
gouthamrmharley and cardoe both had some comments from there to address perhaps? 17:37
gouthamrnevertheless, please do add this to your review backlog.. goal proposal != goal selected17:38
gouthamrso this first phase is an evaluation, and mostly documentation:17:39
gouthamr"Goal definitions should use the provided template so they are all formatted consistently. The goal definition does not take the place of any blueprints, spec documents, or other planning tools used within a project to track its work, but can be referenced from those documents."17:39
gouthamr#link https://governance.openstack.org/tc/goals/#defining-goals 17:39
gouthamrso in that regard, these two seem fine to me; i need one/two more pair/s of eyes to confirm17:41
gouthamrthat's a good rollcall on the tracker topics for today.. please do move any task you're working on to the tracker if it doesn't exist17:42
cardoeI'll work on updates17:43
gouthamrty cardoe 17:43
gouthamralright anything else to chat about the tracker/ptg?17:43
gouthamr#topic Stable and unmaintained branch transitions for retired deliverables (fungi)17:44
fungiwhile processing the 2025.1 eom transition recently, the release team hit a snag with venus because openstack's "retired" acl in gerrit prevents our automation from proposing .gitreview/tox.ini updates17:44
fungithis led to the observation that we probably shouldn't be transitioning stable branches of retired projects to unmaintained, but also raised related questions like should existing unmaintained (and even stable) branches transition directly to eol at deliverable retirement?17:44
gouthamrat retirement? 17:45
fungidid irc cut off the end of the sentence?17:45
gouthamrno, i was thinking out aloud17:46
gouthamrhmmm, that's a fair thought given these retirements were motivated by the project team disbanding17:46
fungianyway, if the tc has a preference on these topics, maybe we can get some guidance encoded in the unmaintained policy and/or retirement workflow documentation17:46
gouthamrbut, will that always be the case?17:46
fungiit's been policy that to land changes on older branches you have to backport from newer branches. if the newer branches are skipped we've considered that a blocker (e.g. tripleo's deprecation)17:47
fricklerso currently for retired repos we leave stable branches in place untouched forever, right? I meant to check that, but didn't get to that yet17:47
fungithat seems to be what's been happening, right17:48
gouthamryeah: https://opendev.org/openstack/tripleo-ansible/src/branch/stable/zed/17:48
fungiwe empty the master branch, but don't do anything to the stable branches while they rot indefinitely, but for non-retired projects we do close those branches down over time17:48
fricklerI think before the unmaintained policy those may also have gotten closed down when the stable branch goes eol? but I'd need to check that, too17:49
fungii guess part of the question is whether you foresee a team resurrecting a retired project as likely to revive its old stable branches from prior to retirement17:50
fungior to work with a clean slate going forward, starting only from the restored master branch17:50
frickleranother part is whether we need to keep the stable branches supported for their usual lifetime if they weren't deprecated before already17:51
fungiin most cases, we didn't retire deliverables until they had been inactive long enough that all existing stable branches reached end of maintenance, instead we'd hold them in a "deprecated" state waiting for those to age out first17:51
fricklerand that would also be the policy change - if that is needed - that I'd suggest: eol the stable branches when their time has come (like for 2025.1 right now), don't transition to unmaintained17:52
gouthamr^ i think that's simple enough17:52
gouthamrbut, i don't really like this automatic transition either :( 17:52
gouthamri really think "unmaintained" needs to be opt-in17:52
gouthamrper branch, per project17:53
fungii guess that's what's different about the venus case wrt existing stable branches, retiring the deliverables with branches under maintenance (rather than deprecating) is not a typical workflow17:53
fungiwhich is why this hasn't caused problems in recent history17:54
gouthamri see17:54
fungiso maybe that's the other hole in policy and/or process17:55
gouthamrmonasca was in the same position as venus - i.e., the project team disappeared17:56
gouthamr#link https://opendev.org/openstack/monasca-api 17:56
gouthamr^ we've deleted all the branches17:56
gouthamrand my memory may be mushy here, frickler helped with this by actually cleaning up the older branches during the retirement process 17:57
gouthamr(not after)17:57
fungiwas it inactive for long enough before retirement that the branches had aged out without creating new ones?17:57
gouthamrcould be; venus spent some time being "inactive" too.. but maybe we limped along differently17:59
fricklerhttps://review.opendev.org/c/openstack/releases/+/95807617:59
fungiyes, venus skipped two releases17:59
frickleryes, needed to do that by "hand" ... well script ;)17:59
gouthamrfungi: would you be able to propose your thought as a resolution? or would you want one of us to?18:00
fungii can propose a resolution amending the unmaintained policy18:00
gouthamrwe're at the hour, and i don't expect resistance to your idea.. you're right that this is a gap in our policy18:00
fungilike the prior amendment18:00
gouthamrack18:01
gouthamrty for bringing the topic here, and for the cleanup!18:01
gouthamralright, we're past the hour. Is there anything else you'd like to add to the minutes today18:02
fungimy pleasure18:02
gouthamrif not, i hope, after all the DST changes, that the time for this meeting still works in your calendars. If you have an issue, please let me know18:03
gouthamrthank you all for participating today18:03
gouthamr#endmeeting18:03
opendevmeetMeeting ended Tue Nov  3 18:03:40 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)18:03
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tc/2026/tc.2026-11-03-17.00.html18:03
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tc/2026/tc.2026-11-03-17.00.txt18:03
opendevmeetLog:            https://meetings.opendev.org/meetings/tc/2026/tc.2026-11-03-17.00.log.html18:03
dansmithgouthamr: sorry, I watched most of that delayed, but I'm _still_ on my conflict18:10
gouthamrdansmith: no problem at all :) 18:34
*** elodilles is now known as elodilles_pto20:24

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!