| knikolla[m] | tc-members: meeting reminder, in 54 minutes | 15:06 |
|---|---|---|
| slaweq | thx knikolla | 15:21 |
| knikolla[m] | #startmeeting tc | 16:00 |
| opendevmeet | Meeting started Wed Mar 22 16:00:29 2023 UTC and is due to finish in 60 minutes. The chair is knikolla[m]. Information about MeetBot at http://wiki.debian.org/MeetBot. | 16:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 16:00 |
| opendevmeet | The meeting name has been set to 'tc' | 16:00 |
| knikolla[m] | #topic Roll call | 16:00 |
| dansmith | o/ | 16:00 |
| gmann | o/ | 16:00 |
| noonedeadpunk | o/ | 16:00 |
| JayF | o/ | 16:00 |
| jamespage | o/ | 16:00 |
| spotz[m] | o/ | 16:01 |
| rosmaita | o/ | 16:01 |
| knikolla[m] | o/ | 16:01 |
| knikolla[m] | Hi all, welcome to the weekly meeting of the OpenStack Technical Committee | 16:01 |
| slaweq | o/ | 16:01 |
| knikolla[m] | A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct | 16:01 |
| knikolla[m] | #topic Follow up on past action items | 16:02 |
| knikolla[m] | I see an item under JayF to notify PTLs about action-needed for PyPI maintainership cleanup | 16:02 |
| JayF | Yeah I have a quick update | 16:02 |
| JayF | I emailed the list, and asked PTLs to take action. Between that and any action already taken, we've seen movement. | 16:02 |
| JayF | When we started, there were 358 project extra maintainers, representing 156 humans with excess access. A week after asking PTLs to take action, we now have 49 fewer project extra maintainers, representing 25 fewer humans who have access. | 16:02 |
| gmann | #link https://lists.openstack.org/pipermail/openstack-discuss/2023-March/032780.html | 16:03 |
| JayF | So that's a nontrivial amount of success in a short period of time. | 16:03 |
| JayF | I've also had contributors reach out to me indicating they're starting the long process of recovering a PyPI account. | 16:03 |
| gmann | I just sent this on original thread also #link https://lists.openstack.org/pipermail/openstack-discuss/2023-March/032874.html | 16:03 |
| fungi | those of us who manage the inbox for the openstackci account in pypi are continuing to see notifications about the progress as well | 16:03 |
| JayF | So tl;dr: progress being made, it's slow | 16:03 |
| knikolla[m] | fantastic! | 16:03 |
| gmann | let's ping the projects PTLs we are involved in if they have not done this. | 16:04 |
| JayF | If you want, I can run the script and check progress in TC meetings, just as we have for other topics | 16:04 |
| gmann | so that we can close it during PTG | 16:04 |
| gouthamr | o/ | 16:04 |
| JayF | gmann: I don't think we've given enough time. It took me nearly a half-day to track down and email all Ironic-project-owners | 16:04 |
| knikolla[m] | Thank you JayF! | 16:04 |
| JayF | gmann: and that's as a long time Ironic'er who knows many people in the community. It'll take some time, especially for new PTLs, to track anyone down. I've already gotten emails to me personally asking for help with doing this. | 16:05 |
| gmann | let's see how many can do as this remaining week can be utilized for that | 16:05 |
| JayF | Actually, I'm going to email and sasy PTLs should utilize PTG to make those contacts | 16:05 |
| gmann | for QA, it was quick as most of those maintainers responded quickly | 16:05 |
| JayF | because at a PTG you have a bigger chance of someone knowing who `pypi_user_123` is or whatever :D | 16:05 |
| knikolla[m] | Any follow-up actions that we should note on the log? | 16:05 |
| JayF | If we want to track progress, I'm happy to keep an agenda item | 16:06 |
| JayF | but I don't think there's anything hashtag-action worthy :) | 16:06 |
| rosmaita | JayF: ++ on the follow up reporting to track, and also on the pre-PTG email | 16:06 |
| knikolla[m] | alright :) moving on then | 16:07 |
| knikolla[m] | #topic Deciding on meeting time | 16:07 |
| knikolla[m] | I saw that there was some confusion about doodle poll. It is in local time, and since I set the dates in April, they are past all DST changes. | 16:07 |
| gmann | JayF: please use this etehrpad for tracking where many projects has added info #link https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup | 16:07 |
| knikolla[m] | We're still missing one vote I think. | 16:07 |
| gmann | knikolla[m]: we have one more action item #link https://meetings.opendev.org/meetings/tc/2023/tc.2023-03-15-16.00.html | 16:07 |
| gmann | Remaining 5 TC members to respond to https://doodle.com/meeting/participate/id/er2LQQ2e/vote | 16:08 |
| gmann | but related to meeting things | 16:08 |
| gmann | this is where you can see all the recorded AI of meeting #link https://meetings.opendev.org/meetings/tc/2023/tc.2023-03-15-16.00.html | 16:08 |
| knikolla[m] | gmann: that action is related to the topic I just opened and I was going to bring it up now. | 16:08 |
| gmann | +1 | 16:08 |
| knikolla[m] | Unfortunately no time worked for all the TC. Each time has >= 2 absences. | 16:09 |
| knikolla[m] | I'm leaning towards making an alternating schedule for the technical committee meetings. | 16:09 |
| knikolla[m] | I know it's not ideal, but it seems the only choice we have to guarantee full participation. | 16:10 |
| dansmith | please no alternating meetings... | 16:10 |
| noonedeadpunk | it's even more imposible to read with more ppl that are voted... | 16:10 |
| gmann | that can make it more complex i think | 16:10 |
| gmann | yeah in term of voting, agreement etc | 16:10 |
| rosmaita | i've been thinking about this, though ... i think an alternating meeting would be OK | 16:11 |
| spotz[m] | Ok weird I voted but not seeing myself listed | 16:11 |
| JayF | I don't see any fair option without alternating meetings. It's not OK for us to completely block out 2 of our members due to time conflicts. | 16:11 |
| spotz[m] | hang on | 16:11 |
| JayF | I would say, if it's possible at all for TC members to adjust their downstream meeting schedule to help, we should pull those levers if we haven't already. | 16:11 |
| gmann | last meeting we were discussion about current time, I have noted for it I have other schedule but I am ok to adjust my schedule | 16:11 |
| gmann | so 1 absence of mine can be counted as presence in the same time slot we have now | 16:12 |
| gmann | JayF: +1, I can do that for current time. | 16:12 |
| noonedeadpunk | knikolla[m]: I wonder of your workday given any time from 12pm to 3am UTC works for you :D | 16:13 |
| slaweq | I can eventually attend e.g. on Wednesday 8-9 PM CEST time (I think it will be 6-7 UTC) | 16:13 |
| spotz[m] | Ok voted, not sue what happened the first time | 16:13 |
| gmann | I selected the current time as available if that help | 16:13 |
| dansmith | is there some way to better visualize all the votes? I can only see three lines at a time | 16:14 |
| rosmaita | i gotta say, that doodle interface is the worst | 16:14 |
| knikolla[m] | noonedeadpunk: if European folks have to stay until late, I figured I should show I can make the same effort as TC chair. | 16:14 |
| dansmith | it is the *worst* | 16:14 |
| spotz[m] | dansmith: I generally just look at the number listed under the date | 16:14 |
| dansmith | spotz[m]: yeah I just want to see the matrix | 16:14 |
| JayF | knikolla[m]: I'll note, you can assume I have the same willinness even if I didn't vote that way, although I obviously don't prefer to be meeting at late/early my time, I assume European/APAC folks don't wanna either. | 16:15 |
| gmann | so then let's vote here if current time works for everyone ? | 16:15 |
| dansmith | +1 for current time from me | 16:15 |
| spotz[m] | Doesn't work for me, but I said that last time | 16:15 |
| spotz[m] | -1 | 16:16 |
| slaweq | for me it's fine now but it will not be that good after DST (next week) :/ | 16:16 |
| knikolla[m] | Current time also has 2 absences. | 16:16 |
| JayF | I am -1 to a solution which leads to any TC member unable to attend most regular meetings. | 16:16 |
| knikolla[m] | I can see the matrix as poll organizer. Didn't know it shower you a different view. | 16:17 |
| knikolla[m] | showed* | 16:17 |
| JayF | It's OK if we have rotating misses (like we would for an alternating meeting) but why should we be a 9 person body if we're willing to effectively whittle down to 7 for purposes of meetings? | 16:17 |
| spotz[m] | Tuesday at 1(sorry might be CST) has 7 and 2 maybes(noonedeadpunk and myself) | 16:17 |
| gmann | no matrix there, we need to go to slots one by one | 16:17 |
| slaweq | spotz (@_oftc_spotz:matrix.org) this one may be okish for me | 16:18 |
| spotz[m] | No slawek or jamespage though so NM | 16:18 |
| slaweq | it's a bit late but I can attend usually | 16:18 |
| knikolla[m] | gmann: understood. Noted that it shows the matrix view only to the organizer and to use a different tool next time. | 16:18 |
| slaweq | it's 8pm to 9 pm my time but I can handle that | 16:18 |
| noonedeadpunk | Thursday as well | 16:18 |
| gmann | it is hard to find all members available at same time unless we want to adjust other things for TC meeting | 16:18 |
| spotz[m] | I was hoping it was 7 + the two maybes | 16:18 |
| jamespage | just checking on that Tuesday timeslot | 16:19 |
| jamespage | I may be able to make it | 16:20 |
| knikolla[m] | So it seems like we might be able to get full attendance on that Tue time slot | 16:20 |
| slaweq | yeap | 16:21 |
| rosmaita | which slot is that? | 16:21 |
| JayF | Can we restate the specific time, in UTC, we're considering? | 16:21 |
| JayF | I'm having trouble locating it on the doodle | 16:21 |
| gmann | which one? 6 UTC ? | 16:21 |
| rosmaita | JayF: ++ | 16:21 |
| knikolla[m] | 1800 UTC | 16:21 |
| knikolla[m] | Tuesday | 16:21 |
| slaweq | that's the only one with 7 votes on doodle | 16:22 |
| gmann | slaweq: jamespage: not present and spotz[m]tentative ? | 16:22 |
| JayF | gmann: that reflects what I see, too | 16:22 |
| noonedeadpunk | I think it's 1900UTC | 16:22 |
| dansmith | gmann: jamespage is checking | 16:22 |
| gmann | ok | 16:22 |
| spotz[m] | Yeah though it does count the 2 maybes as full counts. I can make it work, noonedeadpunk you were the other maybe can you do it? | 16:22 |
| slaweq | gmann yes, as it's 8pm for me I said "no" but if there's no other option I can change my vote to "tentative" and attend at that time :) | 16:22 |
| slaweq | so please count me in for that slot | 16:22 |
| fungi | 19:00 utc tuesdays is the time of the opendev sysadmins meeting, but conflicts are always going to be a risk | 16:23 |
| gmann | slaweq: appreciate that. | 16:23 |
| JayF | slaweq: thank you for doing that; seriously I appreciate it | 16:23 |
| dansmith | it's 1800UTC though right? | 16:23 |
| noonedeadpunk | spotz: not perfect, but I can | 16:23 |
| knikolla[m] | it is 1800 UTC | 16:23 |
| jamespage | likewise - I will from time-to-time have a conflict but I should be able to make it | 16:23 |
| fungi | cool, noonedeadpunk seemed to think it was 19 | 16:23 |
| gmann | dansmith: yes, 18 UTC tuesday we are taking. if I am not wrong :) | 16:23 |
| knikolla[m] | because the local times there are after DST changes. | 16:23 |
| dansmith | yeah, doing this in the DST change envelope is very confusing | 16:24 |
| gmann | yeah | 16:24 |
| noonedeadpunk | Hm... I'm just UTC+1 now I assume, and Tusday slot with 2 tenatnive and 7 overall is showing as 8PM-9PM... | 16:24 |
| knikolla[m] | next time I'll send out polls in UTC. | 16:24 |
| spotz[m] | Doing anything until everyone catches up is hard | 16:24 |
| slaweq | noonedeadpunk but it's for April 4th so after DST for us | 16:24 |
| gmann | let's do ti in PTG then | 16:25 |
| slaweq | that's at least my assumption :) | 16:25 |
| knikolla[m] | slaweq: ++, that's why I set the date to that. to save you all the confusion. | 16:25 |
| spotz[m] | I will be on PTO 4/4:) | 16:25 |
| gmann | because I see it is confusing for EU DST things | 16:25 |
| rosmaita | everyone, go here and add your city: https://www.timeanddate.com/worldclock/converter.html?iso=20230404T180000&p1=1440 | 16:25 |
| knikolla[m] | so you could just look at your calendars. | 16:25 |
| slaweq | so then we will be UTC+2 | 16:25 |
| gmann | I feel current time has more majority (8 members) but if spotz can adjust the things but not sure if that is something cannot be. just checking | 16:26 |
| noonedeadpunk | slaweq: Ok, I'm totally confuused about DST then, as I was thinking it in like, time that's written, not converting to utc after dst... | 16:26 |
| noonedeadpunk | what a mess | 16:26 |
| slaweq | yeah, it's a mess | 16:26 |
| dansmith | right, the problem is it's not clear if the times in the doodle are pre- or post-adjustment | 16:26 |
| gmann | noonedeadpunk: :) its not just you. I selected time wrongly due to that | 16:26 |
| dansmith | me too | 16:27 |
| noonedeadpunk | Thursday at 4 also have 7 votes just in case | 16:27 |
| slaweq | but if we are talking about Tuesday 1800 UTC now, this will be for sure 2000 for us after DST | 16:27 |
| gmann | I feel https://framadate.org/ is more simple in poll than doodle | 16:27 |
| JayF | Is the best approach at this point for knikolla[m] to take an action to arrange a menu of times based on the doodle, and us discuss at PTG? | 16:27 |
| dansmith | I thought we're settled | 16:27 |
| dansmith | on tue at 1800UTC, no? | 16:27 |
| knikolla[m] | Ok, with that in mind, and regardless of all the assumptions about DST | 16:28 |
| knikolla[m] | I'm going to start a vote | 16:28 |
| knikolla[m] | about 1800UTC on Tuesdays | 16:28 |
| knikolla[m] | Yes but need a vote to formalize. | 16:28 |
| dansmith | ack, JayF was talking about delaying, but I think we're good to vote | 16:28 |
| JayF | I was only because I thought we still had an off-by-one err | 16:28 |
| JayF | I'm happy to not table the issue if we have a happy place | 16:29 |
| gmann | I think so. not sure if jamespage has checked but good to vote for 16 UTC Tuesday | 16:29 |
| dansmith | jamespage: said he could usually make it | 16:29 |
| jamespage | I did | 16:29 |
| gmann | ohk. perfect | 16:29 |
| rosmaita | gmann: i thought it was 18UTC? | 16:29 |
| gmann | sorry, I missed | 16:29 |
| gmann | so sorry 18 UTC | 16:29 |
| gmann | let's do vote | 16:29 |
| rosmaita | i'm ok with voting now | 16:30 |
| knikolla[m] | #startvote Starting after the vPTG, hold the weekly Technical Committee meeting on Tuesdays 1800UTC? Yes, No | 16:30 |
| opendevmeet | Begin voting on: Starting after the vPTG, hold the weekly Technical Committee meeting on Tuesdays 1800UTC? Valid vote options are Yes, No. | 16:30 |
| opendevmeet | Vote using '#vote OPTION'. Only your last vote counts. | 16:30 |
| dansmith | #vote Yes | 16:30 |
| gmann | #vote yes | 16:30 |
| knikolla[m] | (Did I mess up the syntax? ) | 16:30 |
| spotz[m] | #vote yes | 16:30 |
| jamespage | #vote yes | 16:30 |
| noonedeadpunk | #vote Yes | 16:30 |
| slaweq | #vote yes | 16:30 |
| rosmaita | #vote Yes | 16:30 |
| knikolla[m] | #vote Yes | 16:30 |
| JayF | #vote yes | 16:31 |
| knikolla[m] | #endvote | 16:31 |
| opendevmeet | Voted on "Starting after the vPTG, hold the weekly Technical Committee meeting on Tuesdays 1800UTC?" Results are | 16:31 |
| opendevmeet | Yes (9): JayF, dansmith, spotz[m], noonedeadpunk, knikolla[m], slaweq, jamespage, rosmaita, gmann | 16:31 |
| rosmaita | \o/ | 16:31 |
| knikolla[m] | Thank you all! | 16:31 |
| JayF | I know we have an official meeting invite, but I usually maintain one on my calendar as well. Please DM me your email if you want me to add you on the invite for the new meeting time and you can get a curated email invite from me :D | 16:31 |
| gmann | we have one on ircmeeting also | 16:32 |
| dansmith | I subscribe to the ics, yeah, so that needs updating | 16:32 |
| fungi | i've given aprice a heads up since this will impact the zoom bridge | 16:32 |
| JayF | Yeah, I've had trouble importing those ICS files in the past, so I have my own invite. We should make sure that gets updated though. | 16:32 |
| spotz[m] | I've got that repo locally I can do it | 16:33 |
| gmann | knikolla[m]: please update the chair info also here #link https://meetings.opendev.org/#Technical_Committee_Meeting | 16:33 |
| knikolla[m] | gmann: ah, thanks for pointing that out. | 16:33 |
| gmann | fungi: ah, right. we need to check for video call things also knikolla[m] | 16:33 |
| gmann | knikolla[m]: I was waiting for new time so we can do name and time change togehter | 16:33 |
| knikolla[m] | spotz: you said you can propose the changes to meetings.opendev.org? | 16:34 |
| fungi | she checked and the zoom bridge you've been using is free at that time so no changes are needed there | 16:34 |
| spotz[m] | changing it now | 16:34 |
| knikolla[m] | thank you :) | 16:34 |
| * dansmith puts the pin back in the grenade | 16:34 | |
| gmann | fungi: aprice : perfect thanks | 16:34 |
| knikolla[m] | #topic Gate health check | 16:35 |
| dansmith | About the same as last week for me, which is good news | 16:35 |
| dansmith | gmann has been moving grenade and devstack targets forward, which will unblock nova from using the skip-level-always job | 16:36 |
| gmann | only one thing for grenade jobs | 16:36 |
| fungi | node capacity has been good recently since we tweaked some parameters to squeeze out a bit better quota utilization | 16:36 |
| gmann | +1, all patches merged for that | 16:36 |
| dansmith | (which would be advisable for other projects as well if they're willing) | 16:36 |
| gmann | but octavia-greande job failing and need DIB fix to be released #link https://review.opendev.org/c/openstack/diskimage-builder/+/878089 | 16:37 |
| dansmith | also, there are backports for the mysql memory mitigation to help the stability there, not sure if are merg{ing,ed} yet or not | 16:37 |
| johnsom | gmann I am going to propose that release now | 16:37 |
| gmann | dansmith: yeah, I will prepare the etherpad for grenade skip upgrade testing and greande-skip-level-always usage also | 16:37 |
| gmann | johnsom: perfect, thanks | 16:37 |
| dansmith | cool | 16:37 |
| dansmith | this for zed for the mysql thing: https://review.opendev.org/c/openstack/devstack/+/878135 | 16:38 |
| dansmith | not yet merged, but we should probably do that | 16:38 |
| gmann | dansmith: +2 | 16:38 |
| dansmith | stable branch stability has been sucking similarly to how master was a bit ago, so.. | 16:38 |
| dansmith | yeah | 16:38 |
| gmann | dansmith: do you want to switch the setting on master ? | 16:39 |
| slaweq | gmann that skip-level-always job is opt-in job which we discussed few weeks ago, right? | 16:39 |
| dansmith | gmann: yeah, we should probably do that too, I'll propose later | 16:39 |
| dansmith | slaweq: yeah | 16:39 |
| gmann | it will be good timing if we see anything and need to be discussed in PTG | 16:39 |
| slaweq | I remember I have to add something to neutron but I forgot what it was :) | 16:39 |
| slaweq | ok, thx | 16:39 |
| dansmith | that's all from me | 16:39 |
| gmann | slaweq: I can add it in integrated template and make use of that in neutron, let's discuss it tomorrow or so | 16:39 |
| slaweq | ok | 16:40 |
| knikolla[m] | great :) due to time constraints i want to move on unless there's something that requires consensus | 16:40 |
| gmann | no other failure i observed or info from my side too | 16:40 |
| knikolla[m] | #topic 2023.2 cycle Leaderless projects | 16:40 |
| knikolla[m] | There's still 6 projects without candidacies. | 16:40 |
| knikolla[m] | We have a rally candidacy. Which is great. | 16:41 |
| knikolla[m] | We're still missing for Monasca, Sahara, Swift, TripleO, Vitrage and Winstackers. | 16:41 |
| * bauzas waves super late | 16:41 | |
| noonedeadpunk | TripleO is deprecated, so sorted out as well | 16:41 |
| gmann | noonedeadpunk: no, we need PTL there | 16:41 |
| slaweq | for Monasca I have contacted previous ptl and he told me that he's not able to be ptl again | 16:41 |
| gmann | even deprecation patch need PTL +1 #link https://review.opendev.org/c/openstack/governance/+/877132/2 | 16:41 |
| noonedeadpunk | Do we need a PTL for maintenance of EM branches? | 16:41 |
| slaweq | and I don't have any other info from him about who could we ask for help with it | 16:42 |
| knikolla[m] | I think as TC we can replace the PTL in that instance. | 16:42 |
| JayF | I think we might be waiting a very long time if we wait for someone to volunteer to PTL TripleO :/ | 16:42 |
| slaweq | so IMO it's candidate for marking as inactive project really | 16:42 |
| gmann | noonedeadpunk: as it is not retired only deprecated, we still need PTL for stable/wallaby maintenance point of contact or so | 16:42 |
| knikolla[m] | We already voted as a governing body to deprecate the project. | 16:42 |
| noonedeadpunk | ^ this, yeah | 16:42 |
| knikolla[m] | Ah, thanks for pointing out the difference between retired and deprecated. | 16:42 |
| JayF | It doesn't matter if we need/want a PTL, the operative question is if we'll get anyone volunteering to be a PTL. | 16:42 |
| gmann | knikolla[m]: we need PTL ack also. | 16:42 |
| knikolla[m] | But I think we can just hold another vote as a governing body to explicitly retire, no? | 16:43 |
| gmann | and I think james on email said they are ok to volunteer if needed | 16:43 |
| knikolla[m] | As we have the power to assign PTLs, this is just a formality. | 16:43 |
| dansmith | retire is what we didnt' want to do right? | 16:43 |
| JayF | I have seen zero interest on any of the TripleO posts to the mailing list about deprecation and such, yeah? Does anyone know of anyone willing to work to maintain TripleO, or stable branches? | 16:43 |
| gmann | i think we should ask them to stepup as PTL | 16:43 |
| dansmith | me too | 16:43 |
| dansmith | it should be very minimal work, so I think it's a reasonable ask | 16:43 |
| gmann | I can check with james about PTL things | 16:43 |
| noonedeadpunk | dansmith: no, retire means nothing can't land anywhere | 16:43 |
| gmann | dansmith: yeah | 16:43 |
| dansmith | noonedeadpunk: exactly? | 16:44 |
| knikolla[m] | ah, understood. | 16:44 |
| gmann | until it is retired we need PTL and other formalities | 16:44 |
| knikolla[m] | That cleared it out for me. Since it's not retired, it still needs a PTL despite deprecation. | 16:44 |
| gmann | yeah | 16:44 |
| knikolla[m] | It took me a seconds to click. | 16:44 |
| fungi | you'll have the same problem with projects like monasca too though, right? | 16:44 |
| gmann | let me reachout to james and ask if they are ok to serve as PTL | 16:44 |
| gmann | fungi: if anyone want to maintain stable branch for monasca then yes otherwise we just retire them like we did for congress etc | 16:45 |
| noonedeadpunk | Well, we have EM described as `community members maintaining it` | 16:45 |
| gmann | searchlight also same examle | 16:45 |
| noonedeadpunk | So I'm not sure I agree this inclines requirement of PTL | 16:45 |
| dansmith | especially since they said they're keeping up maintenance for their last repo (wallaby?) I think that there's someone around that can ack PTL required things | 16:45 |
| fungi | wallaby is no longer maintained. it's under extended maintenance now | 16:45 |
| gmann | yeah but extended maintenance is still under governance right | 16:46 |
| dansmith | fungi: I'm talking about their declaration email | 16:46 |
| fungi | right, just pointing out that with no remaining branches officially maintained, it's effectively retired | 16:46 |
| gmann | let's me ask james about it and if we do not have any then we will see how it goes but I think they said ok for PTL in email | 16:47 |
| JayF | gmann: I looked back thru that thread and didn't see him say that; but it wouldn't hurt to ask in either event. | 16:47 |
| noonedeadpunk | it's under governance, but does it require PTL is smth I'm not fully agree with. It requires release liason at least... | 16:47 |
| noonedeadpunk | but yeah, dunno | 16:47 |
| knikolla[m] | Can we say DPL under the TC? | 16:47 |
| noonedeadpunk | I like that ^ | 16:47 |
| knikolla[m] | And point to whoever is current TC chair or a liaison of the TC? | 16:47 |
| JayF | We still need humans willing to fulfill those distributed roles, right? | 16:47 |
| fungi | ptl or a dpl. but also anyone could just volunteer to be the ptl or dpl in name only, and not actually have any responsibilities since the project is retired | 16:48 |
| gmann | JayF: its there in first email only #link https://lists.openstack.org/pipermail/openstack-discuss/2023-February/032083.html | 16:48 |
| knikolla[m] | fungi: ++, i just want them to have a group of people to hide behind | 16:48 |
| fungi | sort of like how the tc is also the uc in an official capacity, but only because bylaws say we have a uc | 16:48 |
| JayF | gmann: aha, he said they'd find someone moreso than volunteering himself, same thing though, thank you for pointing it out to me | 16:49 |
| gmann | knikolla[m]: any project leadership is under governance/TC if they are official openstack project | 16:49 |
| noonedeadpunk | So basically why I'm kind of regarding PTL for deperecated projects, because retiring project that jsut doesn;t happen to have PTL now is kind of too aggressive.... | 16:50 |
| knikolla[m] | gmann: I understand that. But we require a formal PTL role or DPL liaisons even under that official leadership. | 16:50 |
| gmann | DPL, PTL or any new in future. that is just a project level leadership but everything in OpenStack official projects under TC | 16:50 |
| gmann | even SIG etc | 16:50 |
| fungi | noonedeadpunk: how about retiring a project which no longer has any maintained stable branches? | 16:50 |
| noonedeadpunk | Ie, ppl have running deployments of monasca, there could be still vulnarabilities found and patched by community | 16:50 |
| noonedeadpunk | But since we require PTL we retire project instead of deprecate it | 16:50 |
| JayF | fungi: So you're saying: 1) tripleo only says they'll support wallaby, 2) wallaby is EM, not maintained, leads to 3) we should just retire it now? | 16:51 |
| gmann | I will say if tripleO is not able to give PTL then we must retire it instead of deprecation | 16:51 |
| JayF | fungi: if that's what you're saying, I think I agree | 16:51 |
| fungi | JayF: no, you're right that we're still treating the newer stable branches as maintained until they reach em | 16:51 |
| knikolla[m] | gmann: that doesn't match my understanding of what we voted on. As we would still keep accepting patches, and retiring means not accepting patches. | 16:51 |
| dansmith | right, we decided to keep zed alive | 16:51 |
| slaweq | noonedeadpunk for monasca it's not just lack of ptl, there is almost no activity in the project, some gates are broken as I checked | 16:51 |
| noonedeadpunk | I don't like retiring projects due to reasons like not having ptl... | 16:51 |
| slaweq | so there's more things to consider there | 16:52 |
| JayF | noonedeadpunk: slaweq: last commit to Monasca was 11 months ago from gmann | 16:52 |
| gmann | knikolla[m]: yes but we need PTL for all deprecated projects | 16:52 |
| bauzas | do we have contributions or activity on such projects ? | 16:52 |
| slaweq | JayF exactly | 16:52 |
| knikolla[m] | I want to timebox this discussion here for TripleO specifically and for the PTL role. | 16:52 |
| gmann | I am saying if no one step up as PTL then option 5 is best way to go #link https://etherpad.opendev.org/p/2023.2-leaderless | 16:52 |
| slaweq | and gmann is also most active contributor to Monasca in last cycle or two | 16:52 |
| fungi | right, monasca came up because they basically had no 2023.1 release but took too long to decide that was the case so they still got listed as part of 2023.1 | 16:52 |
| knikolla[m] | We can talk about that async as it's mostly implementation details, and worst case we can talk during the PTG. | 16:52 |
| gmann | slaweq: heh :). i should stop that helping them on gate | 16:52 |
| JayF | What about other PTL-less projects? Are we taking action to find PTLs? I see Swift, for instance, still has activity but not a PTL. | 16:52 |
| slaweq | so lack of PTL is just a red flag for me to check other things | 16:53 |
| knikolla[m] | Thanks JayF, was just going to bring up Swift. | 16:53 |
| knikolla[m] | And want to dedicate the last few minutes to it. | 16:53 |
| gmann | if we cannot find leaders (PTL or DPL) we need to think about retirement | 16:53 |
| slaweq | gmann++ | 16:53 |
| knikolla[m] | Does anyone have a communication channel with the team? | 16:54 |
| JayF | looking at https://opendev.org/openstack/swift/commits/branch/master I see at least 3, maybe 4 contributors in the last month, have we reached out to them to voluntell one of them to be PTL>? | 16:54 |
| dansmith | does retirement mean deleting the repo that they say they still want to contribute to? | 16:54 |
| noonedeadpunk | Dunno, that still kind of... extreme to me... Ok, monasca, but like retiring swift that really huge deployments and other projects rely on.... | 16:54 |
| dansmith | repo/branch I mean | 16:54 |
| gmann | dansmith: yes but 2nd option is to move to x namespace (out of openstack) and contriute/maintain | 16:55 |
| dansmith | gmann: right, so that doesn't match with what I think is going on for tripleo | 16:55 |
| noonedeadpunk | Aren't we causing regressions with that for other projects? Given severe vulnarability is going to be found in Zed for example | 16:55 |
| dansmith | which is why I think deprecate and ask for a PTL, which I think is reasonable if they're still around | 16:55 |
| fungi | fork not move, the copy in the openstack namespace gets closed down but it can be forked elsewhere | 16:55 |
| gmann | dansmith: yes. agree. I will ask on ML if there is any PTL until retirement | 16:55 |
| bauzas | I'm quite sad with moving a project to the x namespace | 16:55 |
| bauzas | as an example, blazar is one of the oldest projects from openstack | 16:55 |
| bauzas | who just resurrected from the deads | 16:56 |
| gmann | bauzas: yeah, that is unfortunate that we have contributor but need to move out of openstack | 16:56 |
| knikolla[m] | tc-members: can we take an action item to reach out to the mentioned teams next week during the PTG? | 16:56 |
| fungi | the way it works is that the repository in the openstack namespace remains but gets closed down and code in the master branch replaced by a readme. if anyone wants to continue development on it they can start a new project in another namespace (or elsewhere outside opendev even) and fork the original | 16:56 |
| bauzas | if we put projects into sidelines, I'm not sure those could resurrect the same way | 16:56 |
| slaweq | shouldn't we first mark such project as inactive https://governance.openstack.org/tc/reference/emerging-technology-and-inactive-projects.html and ask for PTL/Liaisons ? | 16:57 |
| JayF | knikolla[m]: I'll reach out to those swift contributors. | 16:57 |
| JayF | knikolla[m]: Ironic has a big dep on swift so we have investment in it. | 16:57 |
| slaweq | and then we can think of retirement in next cycle https://governance.openstack.org/tc/reference/emerging-technology-and-inactive-projects.html#timeline | 16:57 |
| JayF | well, big optional dep :) | 16:57 |
| gmann | slaweq: we are passed to that stage i think, I mean if no PTL/leader to lead that project then we should talk to contributor about it if no then retirement etc. inactive is more of we still ahve PTL etc and no activity | 16:57 |
| gmann | slaweq: we need to retire in same cycle as no PTL/leaders or may be no release they can produce | 16:58 |
| slaweq | gmann in this document it's written differently | 16:58 |
| gmann | keeping project without PTL/release liaison is difficult situation for release team as well as for us | 16:58 |
| JayF | We have two meeting minutes left, we should get the actions down for people to contact the remaining leaderless teams before it runs out | 16:59 |
| slaweq | if inactive project will not have leaders before milestone-2, there will be no release of it in the cycle | 16:59 |
| knikolla[m] | We're out of time now | 16:59 |
| slaweq | and next cycle TC will discuss retirement | 16:59 |
| gmann | slaweq: even for PTL-less projects? | 16:59 |
| knikolla[m] | #action JayF to reach to Swift about lack of PTL candidacy | 16:59 |
| gmann | slaweq: we can check/updaet that if that is written for leaderless projects | 16:59 |
| gmann | I will check for tripleo | 17:00 |
| knikolla[m] | The rest please fill out on the etherpad if you plan to reach out to one of the project teams | 17:00 |
| knikolla[m] | Next week is the Virtual PTG, so there will be no weekly meeting | 17:00 |
| slaweq | gmann++ let's check it later | 17:00 |
| knikolla[m] | On Monday there will be the TC+Community Leaders interaction | 17:00 |
| knikolla[m] | And Thu and Fri the TC Slots. | 17:00 |
| knikolla[m] | I will send out calendar invites for all of the above and prepare the agenda. | 17:00 |
| knikolla[m] | Reach out to me for items and scheduling. | 17:00 |
| knikolla[m] | Thanks all | 17:00 |
| knikolla[m] | #endmeeting | 17:01 |
| opendevmeet | Meeting ended Wed Mar 22 17:01:00 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 17:01 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/tc/2023/tc.2023-03-22-16.00.html | 17:01 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/tc/2023/tc.2023-03-22-16.00.txt | 17:01 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/tc/2023/tc.2023-03-22-16.00.log.html | 17:01 |
| slaweq | o/ | 17:01 |
| JayF | o/ | 17:01 |
| JayF | thanks knikolla[m] | 17:01 |
| bauzas | fwiw, I won't be able to attend next TC meetings after the time change | 17:01 |
| bauzas | 8pm is nearly impossible for me as a work hour | 17:01 |
| JayF | That's unfortunate, but I'm glad I think since the first time I've been on TC we have a time which we might get 9 attendees :) | 17:02 |
| bauzas | but I'm happy other european people can make it, despite it's probably an hard ask for them | 17:02 |
| spotz[m] | Thanks all | 17:02 |
| bauzas | anyway, this is a NP-problem (finding a meeting slot that suits a distributed team) and being not part of the TC, I just provide comments | 17:02 |
| noonedeadpunk | dansmith: do you accidentally happen to know if s3 driver is back to glance? as I can't find it there but saw some reverts and actions wrt | 17:02 |
| dansmith | noonedeadpunk: nope | 17:03 |
| dansmith | I mean, I don't know | 17:03 |
| dansmith | rosmaita might | 17:03 |
| noonedeadpunk | As I'm kind of thinking about consequences for all projects with swift retirement... | 17:03 |
| bauzas | JayF: I'm actually positively surprised that 8pm suits people without putting burden on their personal times | 17:03 |
| fungi | maybe 18:00z will attract some apac participants at least | 17:04 |
| JayF | bauzas: This is part of why I said I'd be willing to do alternating times; it's not awesome for our European friends to always shoulder that burden. | 17:04 |
| bauzas | fungi: that's very early for them | 17:04 |
| fungi | it is, but not impossibly early at least | 17:05 |
| fungi | 18z is currently 5am in sydney, i think | 17:05 |
| bauzas | 1800 UTC being 2am in China ? well, I am pretty unsure | 17:05 |
| noonedeadpunk | It kind of sounds to be cheaper to maintain swift then replace swiftclient with s3 client everywhere... | 17:05 |
| JayF | noonedeadpunk: swiftclient is used by many openstack deployers to access ceph as well, fwiw | 17:06 |
| JayF | noonedeadpunk: I think there's enough activity we'll keep it going. It's too important for it not to keep going. | 17:06 |
| noonedeadpunk | JayF: do you really think RGW will keep swiftclient compatability? | 17:06 |
| clarkb | swift is active as far as I know? | 17:06 |
| clarkb | I know they didn't sign up a ptl but people are working on it | 17:06 |
| noonedeadpunk | They're kind of dreaming to cut it off it | 17:06 |
| JayF | clarkb: yeah, but no PTL... I'm not so sure why we're eager to shovel dirt on it :) | 17:06 |
| fungi | swift is active but nobody has had luck reaching someone to be the ptl | 17:07 |
| JayF | noonedeadpunk: I know many upset downstreams if they did... including mine and many others running it as a swift-emulation | 17:07 |
| clarkb | timburke is active in the swift irc channel right now | 17:07 |
| noonedeadpunk | Yeah, but us burrying swift will give perfect excuse to do so | 17:07 |
| timburke | i can continue to be ptl | 17:07 |
| noonedeadpunk | awesome | 17:07 |
| noonedeadpunk | issue solved :D | 17:08 |
| fungi | noonedeadpunk: yes, radosgw swift emulation is apparently still problematic/incomplete and not really a viable substitute for a multi-tenant object store, so its claims of compatibility serve more as an attractive nuisance when operators think they can just reuse their block storage backend as a pseudo-swift without realizing the drawbacks | 17:09 |
| timburke | confluence of issues led to me not getting my candidacy in, then failed to circle back on it. sorry for the confusion/disarray | 17:09 |
| JayF | timburke: thank you! If there's ever any actual-help you need keeping things running, feel free to reach out. | 17:09 |
| gmann | knikolla[m]: sent email to contact TripleO for PTL https://lists.openstack.org/pipermail/openstack-discuss/2023-March/032884.html | 17:09 |
| JayF | Hey knikolla[m], I finished my action item :P LOL | 17:09 |
| knikolla[m] | what swift progress. | 17:10 |
| * TheJulia notices the pun and smiles | 17:10 | |
| JayF | I'm going to put that in my pun container for later. | 17:11 |
| TheJulia | I'm sure this is taking place swiftly | 17:12 |
| gmann | bauzas: just to remind it for nova, we need an extra step from PTLs for pypi maintainers cleanup, if you can do for nova. thanks https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup#L23 | 17:18 |
| bauzas | gmann: I thought the fact we said "yes to remove the extra maintainers" was sufficient | 17:19 |
| JayF | gmann: I'll see if more progress gets made this week. If not I'll start doing things like, organizing it by person and project and contacting PTLs with directive lists instead of higher level instructions :) | 17:19 |
| bauzas | do I need to reach out all of thm | 17:19 |
| bauzas | ? | 17:19 |
| bauzas | ah I see | 17:20 |
| JayF | bauzas: https://etherpad.opendev.org/p/pypi-maintainer-cleanup-details this is a detailed set of steps I laid out fro the Magnum PTL who emailed me in response | 17:20 |
| gmann | JayF: +1, great | 17:20 |
| bauzas | ok cool | 17:20 |
| gmann | bauzas: yeah, that is better way to communicate. thanks | 17:20 |
| bauzas | so basically, those old maintainers need to self-remove from the list and add openstackci ? | 17:21 |
| gmann | yeah | 17:21 |
| bauzas | I see | 17:21 |
| bauzas | cool, then I'll do my duty, noted. | 17:21 |
| JayF | Thanks bauzas, I appreciate it :) Hopefully nova doesn't have the same ... deliverable sprawl Ironic does :) It took me a while | 17:22 |
| fungi | openstackci will already be there, they just may need to switch it from maintainer role to owner role | 17:22 |
| gmann | bauzas: thanks | 17:22 |
| bauzas | JayF: I'll see as a blast from the past | 17:22 |
| bauzas | I'll see it* | 17:22 |
| bauzas | I'll just reconnect with old friends and coworkers, not the hardest thing to do | 17:23 |
| JayF | Just perfect your "come back home" paragraph for old contributors you really miss in the reply back ;) | 17:23 |
| bauzas | yeah I should propose tea with biscuits | 17:24 |
| fungi | speaking of inactive and tech preview... does anyone happen to know what the remaining gap is for skyline to become a full-fledged openstack project? | 17:38 |
| gmann | sent email for Winstackers and Vitrage PTL | 17:45 |
| gmann | knikolla[m]: sahara is the only project left to contact | 17:45 |
| gmann | done for sahara also | 18:37 |
| gmann | timburke: thanks for volunteer for swift PTL, can you please propose the governance change like I did last time https://review.opendev.org/c/openstack/governance/+/858978 | 18:40 |
| rosmaita | noonedeadpunk: s3 driver was added back to glance in ussuri | 19:41 |
| tobias-urdin | damn, i feel kind of sad, imagine if canonical, redhat, vexxhost atmosphere, yaook, openstack-helm, kolla, and all collaborated and built the openstack on k8s operators and container images together | 20:31 |
| tobias-urdin | i'm happy to see all of it moving in the k8s direction but redhat building openstack k8s operators with go on github, canonical doing their own operators in python in opendev etc | 20:33 |
| tobias-urdin | just had to get that off my chest | 20:33 |
| fungi | i guess it does reiterate that there's significant continuing interest in openstack though | 21:02 |
| fungi | even if deployment orchestrators have become the commercial differentiator these days | 21:02 |
| JayF | Arguably deployment automation has been the commercial differentiator for many OpenStack-based companies for almost the whole time it's existed :) | 21:31 |
| JayF | I agree it's disappointing that everyone doesn't agree on everything and work together, but there are lots of places to highlight where that cooperation is happening good. My go-to example is always the partnership between metal3.io and Ironic -- k8s consuming openstack projects without any fanfare or notice. Things like kolla and kolla-ansible are partial-counterexamples to | 21:32 |
| JayF | your specific case, too. | 21:32 |
| opendevreview | Tim Burke proposed openstack/governance master: Appoint Tim Burke as Swift PTL https://review.opendev.org/c/openstack/governance/+/878286 | 21:58 |
| rosmaita | \o/ | 21:58 |
| rosmaita | way to step up, Tim! | 21:58 |
| JayF | thanks timburke :) feel free to cash in some karma to get some of the nvidia driver patches in IPA reviewed :D | 22:05 |
| fungi | JayF: yeah, i was thinking back to when api extensions were the commercial differentiator, before services started ripping out extension points and we added trademark requirements that companies serve unadulterated apis | 22:36 |
| fungi | but you're right, that's been... a while | 22:37 |
| JayF | fungi: when all the contributors are public clouds and system integrators, there is little incentive to write good docs or deployment tooling downstream. Some of that problem still exists now, but not nearly to the degree it did many years ago. | 22:41 |
| JayF | s/downstream/upstream/ | 22:42 |
| JayF | In fact, I'd point at kolla and kolla-ansible as maybe a thing that would've been difficult to exist before it did exist :D | 22:43 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!