| opendevreview | sean mooney proposed openstack/governance master: clarify 2024.1 testing runtime requirements https://review.opendev.org/c/openstack/governance/+/895160 | 10:28 |
|---|---|---|
| opendevreview | sean mooney proposed openstack/governance master: clarify 2024.1 testing runtime requirements https://review.opendev.org/c/openstack/governance/+/895160 | 10:41 |
| knikolla | Proposed https://review.opendev.org/c/openstack/releases/+/897848 to update release status for the new Unmaintained policy | 15:25 |
| knikolla | And https://review.opendev.org/c/openstack/project-team-guide/+/897505 updates the Project Team Guide as well. | 15:26 |
| JayF | thanks for that | 15:50 |
| gmann | knikolla: thanks, reviewed the release change but it seems you have WIP in p-t-g one, is that more content you are adding there or is it ready? https://review.opendev.org/c/openstack/project-team-guide/+/897505 | 16:07 |
| knikolla | gmann: I'll remove the WIP tag. | 16:10 |
| gmann | perfect, thanks | 16:10 |
| opendevreview | Kristi Nikolla proposed openstack/project-team-guide master: Update docs for Unmaintained https://review.opendev.org/c/openstack/project-team-guide/+/897505 | 16:13 |
| JayF | Just going to post a few governance things here which are close to landing for folks to have a look; I'll be looking to land eligible items after the meeting | 16:43 |
| JayF | https://review.opendev.org/897520 mark monasca inactive has several yes votes but is only 4 days old; I'll let that one sit at least 7 before landing but could use your review if you're not looked yet | 16:43 |
| JayF | https://review.opendev.org/896585 appoint james page PTL of sunbeam; 13 days old; this is majority-yes right now, no votes against, and I will merge after meeting if no objection or negative vote before then | 16:44 |
| JayF | There are two patches to update sunbeam project with new charms; I'll wait on those until after the ^ ptl patch lands but they should be trivial once we get there | 16:45 |
| JayF | https://review.opendev.org/896487 and https://review.opendev.org/896378 | 16:45 |
| JayF | Last but far from least; PTI is up, has lots of positive feedback at this point after many revisions (thanks Sean!); https://review.opendev.org/895160 | 16:46 |
| JayF | tc-members: Please take note; meeting in 75 minutes. I've also posted some governance reviews, with context, above that you may want to look at before the meeting. | 16:46 |
| TheJulia | fungi: quick question if your around, https://etherpad.opendev.org/p/2024.1-leaderless#L103 seems weird to me since I thought you had to acquire membership to sign agreement to commit code. Is that the case, different case? | 17:38 |
| clarkb | TheJulia: that requirement went away a long long time ago. In part to simplify the process and in part because gerrit stopped supporting the phone home process during CLA signing | 17:58 |
| spotz[m] | I know you can be just a community member and not a foundation member | 17:58 |
| JayF | #startmeeting tc | 18:01 |
| opendevmeet | Meeting started Tue Oct 10 18:01:10 2023 UTC and is due to finish in 60 minutes. The chair is JayF. Information about MeetBot at http://wiki.debian.org/MeetBot. | 18:01 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 18:01 |
| opendevmeet | The meeting name has been set to 'tc' | 18:01 |
| JayF | Welcome 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. | 18:01 |
| JayF | Today's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee. | 18:01 |
| dansmith | o/ | 18:01 |
| JayF | #topic Roll Call | 18:01 |
| rosmaita | o/ | 18:01 |
| jamespage | o/ | 18:01 |
| slaweq | o/ | 18:01 |
| knikolla | o/ | 18:01 |
| spotz[m] | o/ | 18:01 |
| gmann | o/ | 18:01 |
| frickler | \o | 18:01 |
| JayF | 9 outta 9 in twenty seconds; awesome :) | 18:02 |
| JayF | #topic Follow up on tracked action items | 18:02 |
| JayF | Starting with mine; I have three items being tracked | 18:02 |
| JayF | #action JayF Before next video meeting, write up a short document on pros/cons of moving TC video meetings to jitsi-meet. | 18:02 |
| JayF | no action on that one, carrying over | 18:02 |
| JayF | rosmaita was appointed as vice chair; thank you and congrats | 18:02 |
| rosmaita | \o/ | 18:03 |
| JayF | And for the last one; I was to schedule a cross-performance session at the PTG centered around apparent heavy DB usage | 18:03 |
| JayF | I have an email out to Stephenfin and to Mike, the SQLA maintainer; Mike has responded but not stephenfin yet. I will put something on the calendar when I hear from him. | 18:03 |
| JayF | I would expect this to be scheduled sometime Monday early, if that works for Stephen. | 18:04 |
| JayF | I had one other action item which wasn't properly documented that was carried over from last time as well; | 18:04 |
| clarkb | zzzeek once told me to ping him in openstack irc channels for oepnstack related stuff becuse then he knows it is related to openstack stuff and can be prioritized appropriately | 18:04 |
| JayF | Yeah; he responded to my email quickly and was incredibly nice :) | 18:04 |
| clarkb | so maybe try IRC | 18:04 |
| JayF | I'm waiting on Stephen, and I neglected to send the email until yesterday. | 18:04 |
| clarkb | oh I misread zzzeek did resposne already. great | 18:04 |
| JayF | so this is on track except I've been busy :D | 18:05 |
| JayF | the one additional item; I emailed the tripleo team about their maintenance intentions | 18:05 |
| JayF | let me get the exact quote | 18:05 |
| JayF | According to James Slagle: | 18:05 |
| JayF | > We don't plan to maintain any wallaby branches in TripleO after around the end of October. All these branches can be transitioned to EOL at that time. For the CI repos (tripleo-ci, tripleo-quickstart, tripleo-quickstart-extras), those can also EOL'd unless the TC wants those around to support Zed for some reason. | 18:05 |
| JayF | This should simplify our work around unmaintained branches; as we should be able to retire the tripleo related branches and repos without impacting that team. | 18:06 |
| dansmith | wow, that's surprising to me, but yeah, easier | 18:06 |
| knikolla | yay | 18:06 |
| JayF | We don't have a dedicated item for it; but I imagine the review for unmaintained branch implementation would be a good topic for open discussion; so we can use that information as a tool there. | 18:06 |
| clarkb | any idea if that has been communicated to users? | 18:06 |
| JayF | clarkb: this was a private email from me to TripleO DPLs. If it's been communicated, it wasn't in a location I've seen anytime recently. | 18:07 |
| JayF | clarkb: that being said; Wallaby is an EM branch and we do not make promises about support lifetimes for EM branches. | 18:07 |
| JayF | clarkb: your comment is taken under advisement; I'll reply to James and suggest they communicate that to the list. | 18:08 |
| clarkb | yes, but useres may not undersatnd that if it is the last release of tripleo. | 18:08 |
| gmann | as per the new stable policy, stable/wallaby is going to 'unmaintained' status and if no maintainer for any project like tripleO then it can be EOL. | 18:08 |
| clarkb | we had people discussing use of tripleo as recently as end of august on the mailing list | 18:08 |
| clarkb | (I'm not saying don't eol, I'm saying make sure people know tripleo is really EOL at this point) | 18:08 |
| gmann | Em did not get release anyways. so stable/wallaby releaed long back | 18:08 |
| JayF | Yeah, it's a good point. I'd rather a DPL be the one to start that thread so I will nudge them but will send an email myself if they will not. | 18:08 |
| JayF | Going to ask we table further discussion on this until open discussion | 18:09 |
| JayF | Moving on to other action items; knikolla had two | 18:09 |
| JayF | > Complete documentation for unmaintained branch policy in releases.openstack.org | 18:09 |
| JayF | > Investigate or delegate research on DB usage patterns in Keystone in devstack. Due before PTG. | 18:09 |
| JayF | I assume you want the "before PTG" one kicked ahead a week? | 18:09 |
| knikolla | For the docs side of things, I have two patches up | 18:09 |
| knikolla | https://review.opendev.org/c/openstack/project-team-guide/+/897505 updates the project team guide | 18:10 |
| knikolla | And https://review.opendev.org/c/openstack/releases/+/897848 updates the release series status (it does not EOL the branches by itself, that's for a future series of patches) | 18:10 |
| knikolla | Please take a look and comment | 18:10 |
| JayF | Thank you for those; I already left some comments. It's much easier to think about this with a concrete suggestion. | 18:11 |
| JayF | #action knikolla Investigate or delegate research on DB usage patterns in Keystone in devstack. Due before PTG. | 18:11 |
| knikolla | For the keystone and devstack side of things. I did some progress, collected all the database queries for a stack.sh run and am analyzing them. so far, most seems to be related to users and roles. | 18:11 |
| gmann | I also left comment on the release patch, I think we should merge the p-t-g change first so that anyone looking at release page knows what is 'unmaintained' means | 18:11 |
| JayF | That's neat to hear. I am looking forward to seeing the full results of those. | 18:12 |
| JayF | The last TC member with tracked action item was slaweq | 18:12 |
| knikolla | It does feel like a lot of those queries should have been cached, and I'm gonna dig further to see if there are any caching issues | 18:12 |
| dansmith | knikolla: sweet | 18:12 |
| JayF | > To propose a patch to openstack/governance for TC consideration to mark monasca inactive | 18:12 |
| JayF | knikolla: \o/ thank you very much for this, it'll be huge | 18:12 |
| slaweq | I sent a patch and email to ML about Monasca | 18:12 |
| slaweq | https://lists.openstack.org/pipermail/openstack-discuss/2023-October/035281.html | 18:12 |
| slaweq | no response for that email so far | 18:12 |
| JayF | #link https://lists.openstack.org/pipermail/openstack-discuss/2023-October/035281.html | 18:13 |
| JayF | #link https://review.opendev.org/897520 | 18:13 |
| gmann | previous PTL added late candidacy, maybe they are not seeing ML? | 18:13 |
| gmann | or at least gerrit | 18:13 |
| slaweq | I can try to contact with previous PTL this week | 18:14 |
| JayF | Lets get through gate health check and we can hit that in detail during leaderless projects? | 18:14 |
| JayF | #topic Gate Health Check | 18:14 |
| JayF | Anything going on with the gate? | 18:14 |
| dansmith | mostly still pretty good I think | 18:15 |
| dansmith | things are merging at rates I've not seen in a while | 18:15 |
| gmann | yeah, seems good | 18:15 |
| dansmith | just based on my butt dyno | 18:15 |
| gmann | devstack setup for stable/2023.2 is not yet finished. we had few jobs failing there which are marked non voting for now. hopefully those will merge today/tomorrow. | 18:16 |
| JayF | We're in the eye of the hurricane so to speak :D (or maybe the storm is over?) | 18:16 |
| JayF | thank you for the work to keep things stable | 18:16 |
| clarkb | that is a pretty normal cycle for openstack | 18:16 |
| JayF | Aight, if that's all for gate moving on | 18:17 |
| clarkb | the release rush makes things worse in two ways a) we are our own noisy neighbors and b) landing new functionality and tests that may not be super stable | 18:17 |
| JayF | #topic Leaderless projects | 18:18 |
| JayF | gmann: slaweq: do we wanna un-table monasca conversation? | 18:18 |
| gmann | we have volunteer for Skyline project, seems they missed the election things | 18:18 |
| slaweq | for Monasca I think I will try to reach out to the previous PTL and see what answer I will get, if any | 18:18 |
| slaweq | and then we can hopefully discuss about it next week | 18:19 |
| gmann | yeah but we really really need commitment from them otherwise it will be same story next cycle | 18:19 |
| JayF | For skyline; it's especially concerning their electorate is so small. Should we consider modifying requirements around foundation membership to be more inclusive of those contributors? | 18:19 |
| gmann | so slaweq proposal going with Inactive project way is good idea | 18:19 |
| JayF | I second the concern about Monasca: no PTL volunteer is only a small reason it appears inactive; the bigger reason is no independent contributor has been doing maintenance on it, and instead it's been being done by openstack-wide teams | 18:19 |
| gmann | I mentioned the DPL option for them if election is overhead | 18:20 |
| gmann | I mean DPL model for Skyline | 18:20 |
| JayF | Yeah; I realize, but they also are impacted by TC decisions | 18:20 |
| JayF | and if they are not represented in the overall OpenStack electorate, it's not really an equitable situation | 18:21 |
| gmann | also, sent email to previous PTL + ML for call of PTL/DPL model for Rally and Mistral | 18:21 |
| clarkb | JayF: that may be encoded in the bylaws? I don't remember | 18:21 |
| TheJulia | foundation bylaws can be changed | 18:21 |
| TheJulia | and project governance is largely up to the project itself deciding how it wants to run it's project | 18:22 |
| slaweq | regarding Rally I think many projects have some CI jobs based on this tool so any action there may imact other projects | 18:22 |
| JayF | I suspect, in general, the OIF board would be aligned with us trying to ensure projects get a reasonable say in their governance | 18:22 |
| gmann | slaweq: good point | 18:22 |
| slaweq | just saying :) | 18:22 |
| JayF | I'll take an informal-action to dig into that; I don't wanna promise I'll have time but I'll make a note on my todo list to see what'd be neccessary for such a change | 18:22 |
| fungi | right, not all openinfra projects require electors to be foundation members, however it's a good way to reduce the risk of ballot stuffing since someone would have to apply for multiple foundation memberships in order to do so | 18:23 |
| fungi | which has its own legal implications | 18:23 |
| JayF | So we've had three separate conversation threads here; I want to summarize for minutes: | 18:23 |
| JayF | #info Skyline project has leadership candidate who missed nomination period and appears active. We hope they will have DPL or PTL soon. | 18:24 |
| clarkb | looks like it is encoded in the bylaws for the TC but not for the projects themselves | 18:24 |
| JayF | #info Rally and Mistral, email sent to ML and previous PTL asking if there is interest in continuing to maintain. | 18:25 |
| fungi | and that's part of the bylaws we're working with the board and legal counsel to get dropped anyway | 18:25 |
| JayF | #info Monasca is proposed for inactivity; but despite no response on ML and proposal to mark inactive; a PTL candidate has emerged. | 18:25 |
| gmann | yeah, those TC specific things from bylaw will go away soon | 18:25 |
| JayF | #action slaweq to sync with Monasca potential PTL candidate and report back to TC next meeting | 18:25 |
| JayF | do those infos and actions appear to represent the discussion around leaderless projects faithfully? | 18:26 |
| gmann | ++ | 18:26 |
| JayF | alright, going to move on, thank you for the work on this gmann | 18:26 |
| gmann | and that is all from my side on leaderless projects, all notes are in etherpad too #link https://etherpad.opendev.org/p/2024.1-leaderless#L55 | 18:26 |
| frickler | where was the skyline candidate seen? I think I missed that | 18:26 |
| JayF | #link https://etherpad.opendev.org/p/2024.1-leaderless | 18:26 |
| JayF | frickler: https://lists.openstack.org/pipermail/openstack-discuss/2023-October/035303.html not in the proper place, but they did volunteer | 18:26 |
| gmann | frickler: #link https://lists.openstack.org/pipermail/openstack-discuss/2023-October/035303.html | 18:26 |
| gmann | yeah that one :) | 18:26 |
| JayF | frickler: there is clearly a disconnect between our procedures and the contributors to the project; I suspect it may be around language. | 18:27 |
| frickler | yes, language was something I was also thinking about, maybe we can check with the i18n whether they could also translate governance docs? | 18:27 |
| frickler | *the i18n team | 18:28 |
| TheJulia | And maybe the filter for who can vote at the project level is also a bit self limiting in some of the more organic origin cases | 18:28 |
| gmann | ++, at least translated election doc/notification can help | 18:28 |
| JayF | I haven't interacted much with i18n team; is there someone in the TC willing to reach out and inquire with them about this? | 18:28 |
| spotz[m] | How are we doing on the new platform? | 18:28 |
| fungi | it's an i18n sig now | 18:29 |
| fungi | (no longer a team) | 18:29 |
| rosmaita | i can follow up with the i18n sig | 18:29 |
| JayF | #action rosmaita to engage i18n sig about possibilty of translating governance docs | 18:29 |
| frickler | refreshing the translation platform is a related open issue | 18:29 |
| JayF | Thank you, I appreciate it. | 18:29 |
| gmann | ianychoi[m]: is chair of i18n SIG but not sure about how much bandwidth they have as we already asked for volunteer to help in translation | 18:29 |
| gmann | rosmaita: ++, thanks | 18:30 |
| JayF | I may also reach out to the PTL who emailed the list directly and try to help them get their paperwork in order in the short term. | 18:30 |
| JayF | If they are doing what's needed technically; I wanna help them succeed in the governance/social bits. | 18:30 |
| JayF | s/PTL/PTL candidate/ | 18:30 |
| spotz[m] | I think Seongsoo is working with him | 18:30 |
| gmann | I think this is good topic for PTG, how to improve election notification or governance process communication | 18:30 |
| JayF | oh, wonderful, that's nice to hear, I was hoping something like that would be the case | 18:31 |
| JayF | but if you assume someone else it helping that's a good path to nobody helping :D | 18:31 |
| JayF | gmann: I'd encourage you to add it to the topic list. | 18:31 |
| gmann | we are facing it in every cycle and many things we tried did not actually helped much | 18:31 |
| gmann | sure, will add | 18:31 |
| JayF | If I can figure out the details around the electorate question, that'd be a good topic for ti too, but I don't want to add it until I feel informed enough to participate in the discussion. | 18:31 |
| JayF | Moving on to the next topic now. | 18:31 |
| JayF | oh, good transition | 18:32 |
| JayF | #topic PTG Scheduling and Agenda | 18:32 |
| JayF | I have reserved Monday 1600-1800 UTC in Folsom room for TC/Community interaction | 18:32 |
| JayF | And Thursday 1500-1900 UTC, Friday 1500-1900 UTC for other TC topics | 18:32 |
| fungi | JayF: i think you interpreted spotz[m]'s response as stating that seongsoo is helping the skyline ptl, but i believe she meant he's helping ianychoi with the translation platform move | 18:33 |
| JayF | I have not performed any mapping of specific topics to timeslots yet for the TC due to the small number of agenda items we currently have listed. | 18:33 |
| JayF | fungi: ack; understood thank you for the clarification | 18:33 |
| JayF | This is the time to say if those times create a hardship for you. | 18:34 |
| frickler | according to the ptg schedule, the sessions end at 18 UTC | 18:34 |
| rosmaita | not a hardship, but i may have to leave early on Friday | 18:34 |
| gmann | I think we are 4 hrs on Thurs and friday? | 18:35 |
| JayF | frickler: you're right, Friday ends at 1800, I fat fingered it in the etherpad | 18:35 |
| gmann | frickler: I requested many times to open slot for 18-19 UTC but it was rejected every times. | 18:35 |
| gmann | I find having that in ptgbot easy in communication | 18:35 |
| JayF | So limiting ourself to what's reserved in the bot, that gives us 3 hours Thursday and Friday. | 18:36 |
| JayF | Honestly it's probably enough. | 18:36 |
| frickler | so that brings up a good question: who decides which slots can be used at the ptg. because the kolla team wants to move into the other direction | 18:36 |
| gmann | we can book 4 hrs like we did in last two cycle and mention it in ML. | 18:36 |
| slaweq | ++ | 18:36 |
| gmann | slots in ptgbot should not be a limitation | 18:37 |
| fungi | the amer+emea block is already 5 hours long (13:00-18:00 utc). extending that to 19:00 utc would either mean the possibility of being in conference calls for 6 hours straight or removing some of the earlier part of that block to shoft it forward | 18:37 |
| JayF | If the majority of the TC sees value in holding our slots open through 1900 UTC; I personally have no objection to that, but I'm also skeptical we have 8 hours of content and that timeboxing to 6 hours won't be an improvement. | 18:37 |
| gmann | I am in favor of 4 hrs slot on both days. till 19 UTC | 18:37 |
| spotz[m] | +1 to fungi | 18:38 |
| gmann | fungi: that is not actually true not every team book all open slot in ptbot so team can decide what 4 hrs they want to book or if they are ok with 5-6 hrs then also it is fine | 18:38 |
| fungi | if we pretend the ptg is not a cross-project event, then that would be true | 18:39 |
| gmann | it is about opening more slots and team can book as per their need | 18:39 |
| JayF | #startvote When should TC sessions end? 1800, 1900 | 18:39 |
| opendevmeet | Begin voting on: When should TC sessions end? Valid vote options are 1800, 1900. | 18:39 |
| opendevmeet | Vote using '#vote OPTION'. Only your last vote counts. | 18:39 |
| fungi | the event organizers are trying to promote personal health of our contributors by not recommending overly long blocks of time for back-to-back ptg sessions | 18:39 |
| gmann | fungi: who gurantee what members are not attending more than 4 hrs, we have !13 slots open per day :) | 18:40 |
| gmann | s/what/that | 18:40 |
| JayF | I suggest we take this as an informal poll to see what tc members think. | 18:40 |
| fungi | gmann: no more than 4-5 hours without a significant break | 18:40 |
| slaweq | #vote 1900 | 18:41 |
| gmann | break is different thing :) which can be done in between of 4 hrs too. | 18:41 |
| JayF | If we go until 1900 on Thursday, for instance, I will be on Ironic + TC PTG from 1300 UTC through 1900 UTC with minimal breaks. | 18:41 |
| gmann | #vote 1900 | 18:41 |
| JayF | #vote 1800 | 18:41 |
| spotz[m] | #vote 1800 | 18:41 |
| frickler | to me even 1800 is very late, especially on friday | 18:41 |
| dansmith | I'll abstain, I don't really care | 18:41 |
| knikolla | #vote 1800 | 18:41 |
| rosmaita | i was a proponent of the "extra" block in the past, but the last ptg kicked my butt, the day was too long | 18:41 |
| jamespage | #vote 1800 | 18:41 |
| rosmaita | #vote 1800 | 18:41 |
| dansmith | I have a bucket under my desk | 18:41 |
| JayF | that is 5 votes for 1800, 1 abstain, 2 votes for 1900 and 1 not voting | 18:42 |
| JayF | #stopvote | 18:42 |
| rosmaita | i would not want to have to empty that bucket | 18:42 |
| knikolla | If we need extra time for a specific topic, we can schedule ad hoc meetings or discuss asynchronously. | 18:42 |
| JayF | #endvote | 18:42 |
| opendevmeet | Voted on "When should TC sessions end?" Results are | 18:42 |
| opendevmeet | 1900 (2): slaweq, gmann | 18:42 |
| opendevmeet | 1800 (5): knikolla, JayF, spotz[m], jamespage, rosmaita | 18:42 |
| spotz[m] | When you add all the other sessions on top earlier in the week people will be burntout on things by Thursday and Friday | 18:42 |
| JayF | We will end at 1800. I will offer to be available for targetted topic discussions afterwards if a TC member needs it. | 18:42 |
| fungi | i'm already burned out and the ptg isn't even starting for two weeks | 18:43 |
| JayF | So thanks for the lively discussion about timing :D | 18:43 |
| JayF | #link https://etherpad.opendev.org/p/tc-ptg-october-2023 | 18:43 |
| JayF | #link https://etherpad.opendev.org/p/tc-leaders-interaction-2024-1 | 18:43 |
| JayF | these are up, ready for topics if you have any to add. | 18:43 |
| JayF | I will wait until the week before PTG to map topics to times; if you have preferences (e.g. for rosmaita who may not be here for the whole PTG) please indicate and I'll schedule so you're available. | 18:44 |
| JayF | #topic Open Discussion | 18:44 |
| JayF | I'll note we have multiple open goverance reviews; I intend on merging eligible changes after this meeting. Please review them. | 18:44 |
| JayF | One item we tabled earlier was discussion of unmaintained/, specifically in relation to tripleo, we can untable that discussion | 18:45 |
| rosmaita | i have a question about the python 3.8 situation | 18:45 |
| JayF | or discuss anything | 18:45 |
| JayF | sure rosmaita you have the floor | 18:45 |
| rosmaita | iirc, we reinstated it for 2023.2 because a library or two dropped support and broke everyone | 18:45 |
| rosmaita | do we have a transition plan for getting off 3.8, or are we just kicking the can down the road? | 18:45 |
| rosmaita | it's just not obvious to me that we need to support python 3.8 in 2024.1 | 18:46 |
| fungi | when it came up, we floated a few possible transition plans | 18:46 |
| dansmith | when we discussed it we said we'd keep language compatibility (not necessarily testing) on supported pythons | 18:46 |
| dansmith | so at least "new code doesn't use some new feature that makes it a parse error in 3.8" | 18:47 |
| gmann | yeah at least unit tests make sure that | 18:47 |
| dansmith | it's an incredibly low bar | 18:47 |
| fungi | keeping one unit test job is a safe way of checking that continues to be the case | 18:47 |
| gmann | and we also agree that as much as we can keep the min version it is good for everyone | 18:47 |
| frickler | except when libs drop py38 support | 18:47 |
| JayF | Well, sorta, it does have a pretty big side effect of limiting us to versions of libraries that also run under python 3.8, which frickler indicated was already keeping us from upgrading some library versions. | 18:47 |
| fungi | frickler: i meant in the libs | 18:48 |
| gmann | python 3.8 EOLing on Oct 2024 so keeping it in 2024.1 which suppose to release in Mar 2024 seems ok to me | 18:48 |
| fungi | but yes, for third-party libraries it's a shackle | 18:48 |
| frickler | yes, but you cannot run unit tests without dependencies | 18:48 |
| gmann | yeah, external deps lib can be good point too along with EOLing of python version itself | 18:48 |
| JayF | gmann: I'm concerned that only looks at the top layer; there may be things we depend on which will not have supported versions for python 3.8 when we release | 18:49 |
| rosmaita | JayF: ++ | 18:49 |
| gmann | JayF: yes, if we encounter such thing we can think of dropping or how to test | 18:49 |
| JayF | gmann: but it's tough to make that case as well as the opposite: it's easy to see when python 3.8 breaks someone running 3.8; it's not easy to see when some random library is missing a two-line fix because we're running 1.x instead of 2.x | 18:49 |
| frickler | see https://review.opendev.org/c/openstack/requirements/+/896904 for an example for how this is treated in reqs | 18:49 |
| dansmith | yeah I'd rather deal with that if it happens | 18:49 |
| rosmaita | and who will want to run openstack in an EOL python version, which will happen with 2024.1 like two months after release | 18:49 |
| gmann | yeah, if we have that situation then we can discuss about how to handle it. and if no way to test then drop | 18:50 |
| fungi | rosmaita: two months after 2024.2 actually | 18:50 |
| frickler | well py38 is EOL only upstream, I assume redhat and ubuntu will continue to support what is in their LTS distros? | 18:50 |
| fungi | 'Provided irregularly on an “as-needed” basis until October 2024.' https://peps.python.org/pep-0569/#source-only-security-fix-releases | 18:50 |
| gmann | frickler: hope so. | 18:51 |
| fungi | not eol upstream | 18:51 |
| dansmith | frickler: correct | 18:51 |
| frickler | s/is/will be soon/ | 18:51 |
| gmann | and we are using focal to test the python3.8 so it should be there at least | 18:51 |
| fungi | cpython is continuing to release versions of 3.8 for the next ~year | 18:51 |
| fungi | i'm not advocating for keeping support for 3.8, just making sure the timeline around it is clear | 18:52 |
| JayF | Both sides have incredibly compelling arguments. It's not an easy choice to make, and I don't think there's a clear right choice. | 18:52 |
| JayF | What is clear is we have to decide one way or another very, very soon. | 18:52 |
| dansmith | well, the right choice is to be conservative and deal with something practically if it happens | 18:53 |
| gmann | yeah | 18:53 |
| dansmith | if it doesn't, then we're fine and if it does, we'll have a better idea of what the impact is to inform a move | 18:53 |
| JayF | I default more to trying to avoid disruptions in the middle of the busy part of the development cycle; just different trade offs. | 18:53 |
| JayF | Either way; TC members need to put a vote on this soon. | 18:54 |
| gmann | I thought we are getting agreement about keeping it on gerrit? https://review.opendev.org/c/openstack/governance/+/895160 | 18:54 |
| dansmith | yeah I thought we were settled | 18:54 |
| JayF | Given there's contention, I will not merge this after meeting like originally intending to, but I will give at least 48 more hours for folks to update votes and set a reminder to land it on Thursday. | 18:54 |
| gmann | I did not see any -1 vote on the change | 18:54 |
| JayF | I'll note this started with a question from a tc member who had not voted on the change yet :) | 18:55 |
| JayF | so there is some consensus but working to grow it doesn't hurt, and all of us being able to explain why we decided a thing is good :D | 18:55 |
| JayF | Is there anything else for open discussion? 5 minute warning. | 18:55 |
| frickler | just noting that I found two two year old project retirements that are not completed | 18:56 |
| frickler | the governance patches have merged, but project repo cleanup and zuul config cleanup are pending | 18:56 |
| gmann | that js lib ? | 18:56 |
| JayF | Is there any action we (TC?) can take to complete that? | 18:56 |
| frickler | I wonder if there is a gap in the process there | 18:56 |
| frickler | js lib is one | 18:56 |
| frickler | the tc could track the status and ping people if things stop moving | 18:57 |
| gmann | frickler: I think it is just missed, usually we complete all the steps on retirement but yes a few can be missed | 18:57 |
| gmann | 2-3 years back we did a lot of cleanup on retired repos and most of them were in good state | 18:58 |
| gmann | I can take js lib forward if ricolin does not have time and busy in other thing | 18:58 |
| gmann | I will check with him and propose remaining changes | 18:58 |
| frickler | ok then | 19:00 |
| JayF | Thanks for that, and we're outta time. | 19:00 |
| JayF | have a good Tuesday o/ | 19:00 |
| JayF | #endmeeting | 19:00 |
| opendevmeet | Meeting ended Tue Oct 10 19:00:25 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 19:00 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/tc/2023/tc.2023-10-10-18.01.html | 19:00 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/tc/2023/tc.2023-10-10-18.01.txt | 19:00 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/tc/2023/tc.2023-10-10-18.01.log.html | 19:00 |
| slaweq | o/ | 19:00 |
| spotz[m] | thanks all | 19:00 |
| jamespage | thanks! | 19:00 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!