Wednesday, 2023-02-22

gmanntc-members: I have created the etherpad for leaderless projects, we need to find leaders for 7 projects https://etherpad.opendev.org/p/2023.2-leaderless  03:11
opendevreviewTony Breeds proposed openstack/election master: Update package metadata and tox4 setup  https://review.opendev.org/c/openstack/election/+/87251003:37
opendevreviewTakashi Kajinami proposed openstack/governance master: Move os-(apply|collect|refresh)-config to Heat's governance  https://review.opendev.org/c/openstack/governance/+/87471905:13
opendevreviewTakashi Kajinami proposed openstack/governance master: Move os-(apply|collect|refresh)-config to Heat's governance  https://review.opendev.org/c/openstack/governance/+/87471905:14
opendevreviewDmitriy Rabotyagov proposed openstack/governance master: Exclusive management of projects by openstackci  https://review.opendev.org/c/openstack/governance/+/87478715:54
gmanntc-members: meeting in ~2 min from now15:58
gmann#startmeeting tc16:00
opendevmeetMeeting started Wed Feb 22 16:00:06 2023 UTC and is due to finish in 60 minutes.  The chair is gmann. Information about MeetBot at http://wiki.debian.org/MeetBot.16:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.16:00
opendevmeetThe meeting name has been set to 'tc'16:00
gmann#topic Roll call16:00
JayFo/16:00
gmanno/16:00
noonedeadpunko/16:00
TheJuliao/16:00
knikolla[m]o/16:00
dansmitho/16:00
bauzas\o /me sits again at the back :)16:00
gmannlet's wait for couple of min if other tc-members joining16:01
* mnaser sits at the side16:01
gmannmeanwhile this is today agenda #link https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee16:01
gouthamro/ 16:02
rosmaitao/16:03
gmannlet's start16:03
gmann#topic Follow up on past action items16:03
gmann#link https://meetings.opendev.org/meetings/tc/2023/tc.2023-02-15-16.00.html16:03
gmanntwo action item from previous meeting16:03
gmannnoonedeadpunk to add PyPi access policy in governance documentation16:03
slaweqo/16:03
noonedeadpunkI've pushed 2 patches as of today - https://review.opendev.org/q/topic:tc%252Fopenstackci_pypi16:03
JayFnoonedeadpunk: I think some of the effort I made got duplicated b/c I wasn't here last week to show my work -> https://review.opendev.org/c/opendev/infra-manual/+/87303316:04
noonedeadpunkI'm also looking at project-team-guide for place where to better mention this requirement16:04
JayFnoonedeadpunk: I've put some comments on your changes and I think a mix of what I wrote and what you put up will likely be ideal16:04
spotz[m]o/16:04
noonedeadpunkAha16:04
JayFas we need something in governance to say "openstack projects must do this" (what you're doing) and something in manuals saying how to (what I did)16:04
gmannsure, that will be helpful, we can make depends-on on JayF patch and modify the governance in parallel on policy16:05
gmannthanks noonedeadpunk for pushing those, let's review those16:05
gmannsecond action item is - gmann to prepare etherpad for grenade skip upgrade job data and send email asking required projects to add job16:06
gmannI did not get time to prepare this. I will try to finish this in this week so continuing it as action item16:06
gmann#action gmann to prepare etherpad for grenade skip upgrade job data and send email asking required projects to add job16:06
gmanntopic Gate health check16:07
gmann#topic Gate health check16:07
gmannany better news on gate?16:07
noonedeadpunktimouts still happening16:07
dansmithmy summary is "slightly better than last week"16:07
dansmithbut still quite poor16:07
slaweqlast week wasn't good for sure16:07
gmannfew timeout are gone but yes it still happening 16:07
dansmithnoonedeadpunk: I've seen fewer (I think) but agree, it's still happening way too often16:07
slaweqI also see that in number of rechecks needed to get something merged - gates weren't stable16:07
bauzasthanks for the hard efforts on tempest folks !16:08
noonedeadpunkWell, yes, I agree it's better, but far from perfect.16:08
dansmithslaweq: noonedeadpunk do you two see the volume test failures in your gates?16:08
dansmithI'm not sure what you run compared to us16:08
gmannyeah moving cirros image vesion to 0.6.1 alsao broke some job which is now reverted #link https://review.opendev.org/c/openstack/devstack/+/87462516:08
dansmithI've been trying to fix cinder tests that I find16:08
bauzasI wish I could have metrics to share about rechecks 16:08
bauzasand yeah, cirros-0.6.1 is creating a few issues that need to be addressed16:09
slaweqdansmith in Neutron we are running "integrated-networking" tests so no cinder related tests theere16:09
dansmithslaweq: okay16:09
JayFI now wonder if some of the seemingly-random ironic failures this week were potentially cirros related.16:09
bauzasslaweq: but we have sometimes some dhcp problems in the gate16:09
noonedeadpunkdansmith: no we run jsut couple of default integrational scenarios, so didn't see anything to off with cinder16:09
TheJuliaJayF: the failures over the last day after say 8am yesteday are all it16:09
fungiis the ovn fix in jammy-updates yet or still only in -proposed?16:09
bauzasthe client doesn't get the lease in time16:09
JayFTheJulia: ack; ty for the detail16:10
noonedeadpunkwe also don't run devstack, so...16:10
gmanndhcp problem was due to cirros, which should be fixed now16:10
bauzasafter 3 retries, each after 1 min16:10
slaweqbauzas can You point me to the failed job with such dhcp issue?16:10
gmannbauzas: feel free to recheck now16:10
bauzasgmann: because of a cirros upgrade ?16:10
dansmithyeah16:10
gmannbauzas: yes16:10
slaweqregarding cirros 0.6.1 I recently added support for dhcpcd in tempest16:10
slaweqit's merged16:10
TheJuliabauzas: if you find dhcp issue and have a fix, please let me know. dhcp failures are also a pain point for me16:11
slaweqso jobs which are using that needs to configure proper dhcp client to use, like we did in neutron https://review.opendev.org/c/openstack/neutron/+/87127216:11
dansmithwell, there are the usual dhcp issues, and the recent ones I think16:11
*** blarnath is now known as d34dh0r5316:11
bauzasslaweq: sec, I have a patch up for Nova16:11
gmannbut volume detach things are big blocker now a days16:11
dansmiththe cirros problem is just the recent increase I think, but in general, lack of networking to the guests is a common flaky failure point16:11
TheJuliadansmith: yeah, they need to be delienated if at all possible :)16:11
dansmithright16:11
TheJuliamy impression and the longer term class is "not configured in time"16:12
gmannslaweq: wait, so the new dchp client will work for cirror <0.6.2 also? because cirros version bump is reverted in devstack #link https://review.opendev.org/c/openstack/devstack/+/87462516:12
bauzaswe tried to test cirros-0.6 with dhcpcd https://review.opendev.org/c/openstack/nova/+/87393416:13
gmannthat should work16:13
dansmithTheJulia: I don't think it's that.. that's how it manifests, but I think it's not actually slowness as the guest is usually up and bored, but we still can't talk to it16:13
bauzasslaweq: this is the bug report we use for tracking the dhcp lease issues https://bugs.launchpad.net/nova/+bug/200646716:13
gmannanywyas let's take this to qa channel after meeting as we have many other topics to discuss today ?16:13
TheJuliadansmith: indeed16:13
gmannbauzas: thanks, that is helpful16:13
gmannany other gate things?16:13
fungiis the ovn fix in jammy-updates yet or still only in -proposed?16:14
fungijust wondering if nova was still running into those failures too16:14
dansmithwhich ovn fix?16:15
gmannwhich one? slaweq  do you know?16:15
slaweqnope16:15
gmannfungi: any link please?16:15
fungithe one last week which fnordahl pushed an sru for16:15
fungii'd have to dig the lp link out of the nova channel16:16
slaweqso I probably missed it, most likely ralonsoh will know16:16
ralonsohlet me read first16:16
bauzasfungi: hmm, sorry, I remember-ish that without context16:16
gmannok, let's check/talk on nova or qa channel about that16:16
bauzasthat does ring a bell, but a small one16:16
fungisorry, ovs16:17
fungi#link https://bugs.launchpad.net/ubuntu/+source/openvswitch/+bug/200306016:17
bauzasyup, openstack-qa seems a good place to discuss to me16:17
gmannyeah, let's move to next topic16:17
gmann#topic Discussion of "Add guidelines about naming versions of the OpenStack projects"16:17
gmann#link https://review.opendev.org/c/openstack/governance/+/87448416:17
gmannwe had discussion on version guidlines for document and release notes for consistency among openstack projects. which is ask by projects in PTG16:18
JayFI'd like to ask we keep the conversation to the content of that change -- I simply want it to be clear we can refer to individual packages, and their versions, without having to include information about the integrated release. There are no specific plans or details on when or where I want to do this; I just think it's valuable to retain that freedom.16:18
gmannslaweq: pushed the changes which was merged as per formal-vote criteira andwhat we discussed in PTG #link https://review.opendev.org/c/openstack/governance/+/87276916:18
JayFFor reasons that are partially my fault; it turned into a conversation about "standalone ironic" which was not my intention16:18
gmannJayF: has some updates suggestion on that ^^16:19
TheJuliaJayF: I think that is reasonable, although I'd honestly prefer a higher bandwidth medium so we can exercise listening as opposed to reading, as I think something gets lost in translation16:19
gmannsure, actually conversation on another things started from the comment on the proposal of not using 'openstack' but use cordinated release version16:19
gmannbut yes, let's discuss about the change and providing the two option in that guidlines one with OpenStack coordinated rlease and one with project version only 16:20
JayFTheJulia: frankly if there are other issues to discuss; we should discuss them in a proper venue -- I just don't want to be the nexus of those disucssions nor to have my (honestly, relatively minor) governance update tied up in them16:20
TheJuliaI think that is totally reasonable. My concern comes up when an explicit single way mandate comes into play.16:21
gmannI suggest that two option in that patch becuase using release version '2023.1' without 'openstack' word  seems odd16:21
bauzason the other hand, mentioning that the standalone product has been tested amongst a large list of other projects seems a quite good incentive for promoting OpenStack to others who just went to the standalone product without getting the whole stack16:21
noonedeadpunkI think we still should omit having cooredinated release but for projects with independant release model at very least16:21
TheJuliaJayF: absolutely agree, there are defintely related discussions occuring in the thread16:21
noonedeadpunkAs coordinated release is simply not applicable for it16:21
bauzasa coordinated release implies that we ensure all our projects are tested between themselves, even the standalone ones16:22
dansmithnoonedeadpunk: meaning make the doc say "independent release projects should omit the OpensStack 2023.1 part" ?16:22
JayFnoonedeadpunk: This sorta puts the crunch on Ironic; given we have users that use both our independent and coordinated releases. It's not as clean as one or the other.16:22
knikolla[m]that seemed to be the consensus. providing two ways to refers depending on whether it's being consumed as an integrated release (with openstack and release name), or by itself (without openstack release version).16:22
gmannnoonedeadpunk: but you mean standalone project to move to intendent release model too ?16:22
bauzasbut people can start consuming a project and after that integrating it with other projects 16:22
knikolla[m]There's also the fact that the integrated release name is on the url path of the docs 16:23
noonedeadpunkdansmith: well, you don't really know part of what they are?16:23
JayFMy preference, as demonstrated by the patch, is to trust the people closest to the project to know the best context to put things in rather than dictate it at a TC level.16:23
gmannyes, removing integrated release version completly is hard16:23
dansmithnoonedeadpunk: I'm asking if you're suggesting drawing the line at "independent release or not"16:23
noonedeadpunkFor exemple - they can have last release 2 years ago16:23
noonedeadpunkdansmith: yes, I'd say that for project with independant release cadence it doesn't make much sense to mention coordinated release in docs or release notes16:24
slaweqnoonedeadpunk like Tripleo for example :)16:24
noonedeadpunkdansmith: but they still can/should be mentioned on releases page16:24
gmannbut ironic is independent release model or coordinated ?16:24
TheJuliacycle with intermediacy16:24
noonedeadpunkslaweq: I was thinking more about tempest or oslo, but yeah... :D16:24
dansmithnoonedeadpunk: agree, and by the same token coordinated or independent-and-coordinated (or whatever it's called) would continue to refer to which coordinated release the doc goes with right?16:25
dansmithyeah, cycle-with-intermediate is what I meant ^16:25
gmannthen mentioning only about independent release does not solve ironic case right?16:25
noonedeadpunkdansmith: yeah, sure16:25
dansmithgmann: correct, but I think that's what noonedeadpunk is suggesting16:25
JayFI do not intend on updating my governance change to draw the line based on specific release models. There is little value in micromanaging to that level.16:25
JayFAnd I would not be in favor of a different change that drew that line.16:25
gmannstandalone vs non-standalone is right thing to suggest?16:25
dansmithJayF: that's your opinion, I think it's clear that for a bunch of people here, we see a value there16:26
dansmithvalue in consistency16:26
gmannconsistency is the key and from where these guidelines requirement came from16:26
TheJuliaI think this whole release model discussion is contrary to the heart of the issue, which is how do projects refer to other projects16:27
gmannas mnaser also mentioned in patch and I also commented, having a top level line of saying  'this project can be used as integrated openstck and standalone and ignore the openstack version if you are a standalone user and do not know openstack' can clear the thigns16:27
dansmithI'm not sure it's that so much as "how do projects refer to the group of versions of other projects that they were tested with"16:27
TheJuliaAnd if there are going be changes there, that has a wide reaching impact which may or may not be desirable, so I would encourage lots of thought, discusison, and consideration of feedback16:27
noonedeadpunkBut I think my biggest question to the change about motives behind it? As I'm not really sure that intention was to distantiate project from openstack, but was about being more clear for end-users that are standalone?16:27
spotz[m]Could we refer to other projects by release or tag?16:27
mnaserconsistency is super important16:28
gmannspotz[m]: tag? tag are release version for projects16:28
bauzas+116:28
TheJuliaconsistency is important, but to be consistent one has to be mindful of their audience16:29
bauzasif I start using project A with 2023.1, I somehow want to make sure that if I use project B later, 2023.1 release will work with project B16:29
bauzasproject A* my bad16:29
JayFspotz[m]: My change, as written, just explicitly permits contributors when sensible in context to refer only to the component name and version. Without my change, it appears any reference to an integrated project needs to be prefaced with the integrated version.16:29
mnaserare you saying the audience of ironic is people who will lose their hope at using the software when they see openstack16:29
TheJuliatheir audience has a starting context, they need to manage that context, and then any author needs to go onward from there16:29
TheJuliamnaser: that is one of the feedback items we tend to get16:30
mnaser"ugh, openstack, forget ironic, it has the o word, i will ignore it as a very capable toolkit to manage by baremetal"16:30
mnaserwell maybe educating those people might be the way to go rather than dodging the whole thing16:30
TheJuliaand that is a whole perception issue which is... differnet16:30
knikolla[m]I just want to call out that the conversation is devolving. 16:30
TheJuliaknikolla[m]: indeed!16:30
dansmithI really do not understand the concern that ironic users may see the word openstack in the doc (ignoring that it's already in the URL) and be concerned that they're going to have to deploy more than they expect16:30
knikolla[m]And we're not discussing the item at hand.16:30
TheJuliawe should stick to the topic at hand, how projects should refer to other projects.16:30
knikolla[m]Which is, in the case of standalone projects, does it make sense to allow them in some use cases to omit integrated release name. Consensus was that it does. 16:31
JayFPlease, keep the focus on the difference as proposed in the patch: should contributors be permitted to reference their projects' actual-version without prefacing it with the integrated release when sensible16:31
dansmithbut if that's the concern, how about a note or a footnote that clearly indicates that ironic can be deployed standalone?16:31
noonedeadpunkJayF: but would suffixing openstack version isntead of prefixing it solve your issue?16:31
mnaserall projects should refer to other projects by the openstack release number so that people who are trying to run this arent stuck looking for this all the time16:31
gmannok so we are missing thing here again. I think both are somwhere related where ironic as standalone want version without openstack 16:31
knikolla[m]Consistency when using those projects together with other projects would be preserved, because the integrated release version in that case would still be included. 16:31
gmannit is hard to separate the things16:31
JayFI do not want to hypothetical or hypothesize about other things; I'm not trying to split any hair other than maintaining my and other Ironic contributors' editorial independence when writing docs about these.16:31
dansmithknikolla[m]: consensus where16:31
mnaserits frustrating enoguh opening releases.openstack.org all the time because i have no idea what release nova 23 is.16:31
dansmithmnaser: I feel the same :)16:32
noonedeadpunkmnaser: well, I think that's why we have https://releases.openstack.org/antelope/index.html16:32
bauzas++16:32
dansmithmnaser: I work on nova and I don't even know what nova version numbers match up to anything :D16:32
gmannmnaser: right, I do check for tempest also many time even I do releaase for it :)16:32
TheJuliamnaser: that is a good data point, can you articulate how you get to that starting point of looking for nova 23 ?16:32
noonedeadpunkWhich must be source of truth kind of for integration between components16:32
mnaserpip freeze in an env?16:32
bauzasit even takes me 0.5secs to remember the nova release number for every cycle when I write the relnotes16:32
TheJuliawell... that is a *whole* different issue then16:32
knikolla[m]dansmith: the majority in the patch discussion had no objection, you included if i'm not wrong, to allow an option to use case for omitting the coordinated release version when considering a project separately from the rest. 16:32
bauzasthis is taking too much brain cycles to calculate this in my mind16:32
mnaserno because now i will end up on the ironic 19.0.0 release page docs16:33
mnaserand i will have no idea what version of openstack it is16:33
dansmithTheJulia: I'd like to point out that you commented on my mention of that concern and maybe even implied malice on my part :)16:33
gmanndo we have any other stanadalone project case here and what they thing?16:33
dansmithknikolla[m]: I don't think you're reading that right ;)16:33
JayFThe way the conversation has devolved into a referendum on the decisions Ironic has made with regard to what else it's integrated with does feel malicious in some ways. I understand my feelings may not align with reality but it's hard not to connect them when scope keeps getting balooned.16:33
gmannare they also feel having openstack version in doc confuse users?16:33
knikolla[m]dansmith: then apologies for that. i will go and read it again. 16:34
noonedeadpunkTo be frank - for me some middle ground would be making main focus on project version and reference openstack version afterwards16:34
gmannnoonedeadpunk: that is ok if that solve the issue16:34
noonedeadpunkSo instead saying `OpenStack 2023.1 (Nova 27.0.0)` we can do `Nova 27.0.0 (OpenStack 2023.1)`16:34
gmannwe had that option in pTG discussion16:35
JayFnoonedeadpunk: that was essentially my original proposal16:35
JayFnoonedeadpunk: if you look at Patchset 1, per your original suggestion16:35
JayFat this point; I think I prefer my current patchset to that one though16:35
noonedeadpunkAnd I'm not sure where all this fuss about ironic and distantiation from openstack has come from...16:35
bauzasnoonedeadpunk: I strongly dislike mentioning the nova release number before the openstack release16:35
noonedeadpunkoh?16:35
gmannJayF: no. proposal; in PS1 was "Nova 2023.1 Series (27.X.Y)"16:36
bauzasand I'll encourage my projects contributors to always refer to the openstack release16:36
dansmithgmann: right, which is incorrect, because the nova version is not 2023.116:36
TheJuliadansmith: not implying malice on your part, but there is a potential risk there, which is an entirely separate discussion. If that is a driver to be dsicussed, then so be it but we shouldn't merge the issues together. Hearing mnaser express frustration from pip freeze to what version is a very valid issue, but we're going to impact identity at that point and then we are re-framing what is openstack16:36
bauzasyeah it's a meaningless 27.0.016:36
JayFgmann: ack; that's fair16:36
TheJuliaat which point, a higher level discussion is truly needed to validate perceptions16:36
bauzaswhich needs you need to calculate how many nova releases you got since Austin16:36
bauzasand for Neutron this is weirder16:37
noonedeadpunkbauzas: but we also do have usecases of project that are deployable and used quite heavily outside of openstack. where openstack version is not that important for some users16:37
gmannso we need to solve standalone case and how standalone projects can consume both version for both cases16:37
bauzasyou have to take in account this was incepted later as a Quantum project, somewhere in Essex release16:37
gmannand I think talking to few more standalone projct will give us more clarity now16:37
JayFI have proposed that solution: trust the contributors of the standalone project to reference them reasonably. If we publish documents, as an Ironic team, that people have an issue with, we can deal with them specifically16:37
noonedeadpunkbauzas: so basically you're saying that "Nova 2023.1 Series (27.X.Y)" would be good for you?16:38
TheJuliaFolks, there are several different issues at play here, each one is a separate discussion.16:38
JayFfor now; I just want to not feel handcuffed when writing up official things about Ironic that might be targetted more towards standalone users16:38
bauzasnoonedeadpunk: this is a message I tend to disagree : even standalone users can take benefits of a release number that guarantees you a compatibility with other products from the same foundation16:38
spotz[m]It seems to me this topic is a larger issue needing further discussion and maybe a wider audience16:38
JayFThis is the focus of my patch; none of this larger picture stuff16:38
dansmithnoonedeadpunk: not fine with me16:38
gmannnoonedeadpunk: that is too confusing 16:38
noonedeadpunkgmann: dansmith I'm just trying to understand16:38
gmannk16:38
dansmithJayF: I get that you don't want to discuss the larger issue, but for what seems like a bunch of us, we can't understand the "smaller" change without the larger context16:38
bauzasnoonedeadpunk: pesonnally, I think 50% of the users probably misunderstand the different projects we have16:39
dansmithJayF: if I'm to ignore the larger context on the "smaller" change then I'm -1 because it makes no sense to me16:39
noonedeadpunkbauzas: I think we're jsut talking now where to place coordinated release and where project version16:39
bauzasnoonedeadpunk: mentioning the project first seems always weird to me16:39
JayFdansmith: bluntly; I'd rather have a straight up-and-down vote lost than continue down these unfruitful discussions of the last two or three days16:39
mnaseri think this small change is contextual out of a larger context16:39
gmannok so let's do one things. I think it is hard to conclude in meeting or in review. let's dodiscussion or re-discuss in PTG on that ?16:39
dansmithJayF: roger that16:39
JayFNo; I am not OK with that gmann 16:39
JayFbecause that means the policy as written/landed while I was out will be in effect for all docs I write for EOC Antelope16:40
JayFand starting of Bobcat ycle16:40
JayFSo I do not want this punted; I want it voted on. If it loses; it loses; but I will not kick the can further down the road.16:40
mnaserwell16:40
mnaserthen the tc votes16:40
mnaserand then we can discuss in the ptg again16:40
gmannsure, that work fine16:41
mnaserand then we revert all the release notes if we decide agaisnt it16:41
noonedeadpunkbauzas: well, I also have usecase where porject version matters. As depending on project version you can tell exactly SHA of other dependencies that were installed16:41
mnaseroops, im not part of the tc, but thats what i was thinking16:41
spotz[m]Hehe16:41
noonedeadpunkI don't care much what would be first though as it looks quite insufficient for me16:41
gmannno it is ok to vote on proposal and if still objection on results we can discus in PTG16:41
rosmaitamnaser: you are tc-emeritus, i think16:41
spotz[m]He is16:41
mnaseri am but i'm running off with it as a tc-member :)16:42
bauzasnoonedeadpunk: fwiw we pin our RPC versions with the openstack release name, so...16:42
mnaserbut my opinion is that it should get voted on, and if it goes through, great, if it doesn't, great.  it can be rediscussed in ptg16:42
mnaserand then anything can be easily reverted16:42
noonedeadpunkbauzas: but error out with ID in case of mismath :p16:42
bauzas27.0.0 is only an openstack/releases number that very insightful people know :p16:43
gmanntc-members: all ok with that and vote means vote on gerrit right?   16:43
dansmithwell, what I'm hearing is that JayF wants a vote on the patch, and very specifically without any larger context about why the change might be needed16:43
gmannif any tc members want to ready the patch/conversation16:43
dansmithgmann: yep, sounds like it16:43
gmannyeah. let's vote on patch and in next video call we take decision if still not concluded on gerrit?16:43
noonedeadpunkBut yeah, as I said I personally don't care much about what will be first, but I can understand why some may preffer X over Y16:43
slaweqgmann +116:44
gmannI will say to get more opinion of other standalone projects but I will try to get those people if I can16:44
JayFI'd just say when voting: have some trust in your contributors? My career, as is most of the others in here, is tied to *OpenStack*. Show some trust that if you give the freedom it will not be abused.16:44
gmannJayF: it works for you?16:44
mnaserJayF: you're not the only one that has been doing openstack and only openstack :)16:44
JayFgmann: sure16:44
gmannok16:45
mnaserbut some of us here do more than just one project and _have_ to know the big context16:45
JayFmnaser: I know, I just think some folks are reading bad motives into things here, and there are none16:45
gmannmoving next, as we are running out of time16:45
mnaserthe less context we have, the harder it is to see your pov, but i will let gmann run the rest of the meeting )16:45
mnaser:)16:45
TheJuliaBut context here may be entirely different from outside this channel16:45
gmannsure, let's vote on patch and other things we can discuss as separte, feel free to add it in PTG etherpad16:46
gmann#topic TC 2023.1 tracker status checks 16:46
gmann#link https://etherpad.opendev.org/p/tc-2023.1-tracker16:46
gmannI will skip this one but please do check your assigned itme as we are approaching to cycle end16:46
gmann#topic Deprecation process for TripleO16:47
noonedeadpunkwe're full of hot topics todya, huh16:48
gmannas discussed in last meeting, I asked about maintaining zed branch too #link https://lists.openstack.org/pipermail/openstack-discuss/2023-February/032239.html16:48
knikolla[m]starting the new tc term with fireworks16:48
gmannjames mentioned he will check it in tripleO team and get back to TC16:48
gmannbut no response yet so I will say let's wait and I will ping them again16:49
gmannknikolla[m]: noonedeadpunk  :)16:49
gmannanything else anyone want to discuss on this ?16:49
dansmithyeah, that's disappointing.. I suspect maintaining zed would mostly be a ceremonial thing, not requiring a lot of work on their part16:49
gmannhope so but not sure what tripleo plan is16:50
JayFHave we reached out to PTLs of other deployment projects to see if they have an interest in developing a tripleo migration plan?16:50
JayFIt'd be a good way to get tripleo users on OSA/KA/et16:50
JayF*etc16:50
noonedeadpunkI think at very least we should provide them with guidance on deperecation process of master branch?16:50
gmannits there on ML but i did not see anyone shown interest16:50
dansmithJayF: meaning a plan to migrate tripleo users to something else like OSA?16:50
JayFYeah; some kind of off-ramp so they can keep using openstack-blessed deployment methods16:51
JayFeven with tripleo being retired16:51
noonedeadpunkI'm kind of interested, but lacking time right now16:51
gmannnoonedeadpunk: sure, let me ping them and we can give path forward for master at least16:51
noonedeadpunkI think this is really good idea to be frank16:51
JayFit seems like a way for us to prevent loss of face in the light of the users who might be marooned16:51
knikolla[m]aren't we forced to find a way to maintain TripleO back to Xena, as it was part of an official OpenStack release? 16:52
noonedeadpunkAnd OSA has path wit husage of RDO packages 16:52
gmannlet me check with Tripleo about zed maintenance and way to proceed on master if that is blocking them on something16:52
gmannknikolla[m]: there is no xena. its wallaby and then zed tjhat is all confusion16:52
gmannk moving to next topic?16:52
dansmithJayF: it'd be nice for sure, but knowing what I do (and don't) know about tripleo, that would likely be a massive amount of work16:53
noonedeadpunkgmann: wait a sec16:53
JayFknikolla[m]: there is no function to force the people who know about a project to maintain it if they don't want to; that's part of why I get nervous about projects primarily with one corporate sponsor16:53
gmannk16:53
knikolla[m]ah :/ right. 16:53
dansmithknikolla[m]: yeah, it's really messy :(16:53
bauzasas a reminder, tripleo is a collection of repos16:53
noonedeadpunkThere's also a ML and PR regarding picking up some delivareables by Heat16:53
bauzasmaybe some have more interests for consumers than others16:53
gmannnoonedeadpunk: that config one? I did not see that yet but that will be great16:53
knikolla[m]JayF: I understand. I mean that the openstack releases page makes a promise about level of support and we are missing the mark on it. 16:54
noonedeadpunkI see no reason to refuse this request, but I wonder if we should block it or not until master will be deprecated16:54
gmannnoonedeadpunk: agree, any other project taking maaintainance of few thigns is good16:54
noonedeadpunk#link https://review.opendev.org/c/openstack/governance/+/87471916:54
dansmithknikolla[m]: that's my primary concern btw16:54
knikolla[m]errr... maintainance... not support16:54
dansmithknikolla[m]: what it means for our "reputation"16:54
gmannnoonedeadpunk: well we can move that deliverables form TripleO to heat and no deprecation thngs for that deliverbales 16:54
dansmithknikolla[m]: and tripleo being very niche doesn't help, because the easy thing is to say "oh well, nobody cares anyway" but it sets the precedent for the next thing16:55
noonedeadpunk++16:55
gmannwe can do that before release things happen. I will check that right after the meeting16:55
knikolla[m]dansmith: ++16:55
gmannlet's move next now16:55
gmann#topic Cleanup of PyPI maintainer list for OpenStack Projects16:55
gmannEtherpad for audit and cleanup of additional PyPi maintainers16:55
gmann#link https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup16:56
gmannML discussion: #link https://lists.openstack.org/pipermail/openstack-discuss/2023-January/031848.html16:56
gmannJayF: any update on email template things ?16:56
gmannI think its good to go now unless no objection on etherpad16:56
JayFit exists; I think it's linked in there?16:56
JayF#link https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup-email-template16:57
gmannFYI, QA team reachout out to all additionla maintainers and they removed themself and given the ownership to openstackci16:57
JayFit's not linked, but I'll add it now16:57
JayFalthoguh it's not really good to go until after my infra-manual change lands16:57
gmannJayF: ok, I am saying we can go ahead on the email template. or you are waiting for infra patch to merge?16:57
JayFand it'll be able to be more strongly worded if we land noonedeadpunk's proposed policy16:57
gmannohk, will check after meeting16:57
JayFI would suggest waiting; fungi has a +1 on my infra-manuals patch now16:57
JayFso I suspect it's extremely close to landing16:57
gmannack16:57
JayFwe have little stannding to ask without hte policy being written16:58
noonedeadpunkI will update my patch right after the meeting16:58
JayFlike, we can ask; but there's no power of TC-force to apply to it yet16:58
gmanncool16:58
noonedeadpunkI hope being able to predict URL :D16:58
gmannJayF: sure16:58
gmannthanks noonedeadpunk JayF 16:58
gmannmoving to next topic16:58
gmann#toic Election updates & post-election work16:58
gmann#topic Election updates & post-election work16:58
* gmann fingers are hurting by typing so much :)16:59
gmannElection will be closed by today 23:45 UTC16:59
dansmithgmann: suck it up chief!16:59
gmannhttps://governance.openstack.org/election/16:59
gmannLeaderless projects16:59
gmann#linkhttps://etherpad.opendev.org/p/2023.2-leaderless16:59
gmann#link https://etherpad.opendev.org/p/2023.2-leaderless16:59
gmannprepared the etherpad for leaderless projects please add your opinion there or any candidate you now for those projects17:00
gmannwe are on time but I would like to extend it for 5-10 min to cover next topics. hoe it is ok for eveyrone17:00
noonedeadpunkWe can drop tripleo right away17:00
gmannTC Chair election17:00
gmannas election will be closed today, we need to start the TC chair election process17:01
gmannnomiantion process is listed here #link https://governance.openstack.org/tc/reference/tc-chair-elections.html#tc-chair-nomination17:01
gmannI am not planning to run for chair this cycle17:01
gmannanyone want to run for Chair please add your nomination17:01
dansmithboo :(17:01
rosmaitagmann: thanks for all your service as chair!17:01
knikolla[m]As mentioned in my TC candidacy, I'm planning to run.17:02
rosmaitabut i understand that your typing fingers may need a rest17:02
dansmithgmann: you're the only one I know of who has a 28 hour day, it's not fair to ask us with only 24 hours to do it17:02
spotz[m]Thanks for your work gmann17:02
JayFgmann: congratulations on the free time that I'm sure the QA project won't fill up immediately ;) 17:02
gmannwhich  will be open for 3 days after election are closed. I will calculate the exact time and let you all know17:02
noonedeadpunkI love that we have a process for that now :)17:02
gmannknikolla[m]: perfect17:02
knikolla[m]gmann: thanks for your great work!17:02
gmannrosmaita: :)17:02
gmannJayF: true17:02
slaweqthx gmann for all Your work as TC chair17:02
noonedeadpunkyeah. that's quite a bummer17:03
gmannthanks everyone and I will continue Chair till we select the new CHair and also help in transition 17:03
noonedeadpunkbut thatnks indeed, you did really awesome job17:03
gmannthanks again17:03
gmannone last topic17:03
gmann#topic Select time to discuss the 'Less Diversity' discussion with foundation staff (Jimmy)17:03
gmannthis one we discussed in last meeting and agreed to have discussion in PTG17:04
gmannbut Jimmy is not available in PTG, do we want to do this discussion before PTG or after PTG ?17:04
gmannthis is not urgent things to sovle but very important to continue the discussion 17:04
fungithe foundation staff are all in a week of in-person meetings17:04
spotz[m]Need to get to next thing on my calendar17:04
rosmaitai think we were also going to start referring to it as "Need More Diversity" ?17:05
JayFShould we lump in the discussion that TheJulia was requesting above, too?17:05
gmannI think we all will be busy pre-PTG 17:05
clarkbjimmy mentioned that allison should be able to be ther eat the ptg17:05
JayFIf we're setting aside time for foundation-adjacent discussions17:05
fungiwhat diversity and inclusion issues did you want to discuss?17:05
gmannrosmaita: oh, I think I changed that but i thnk in my weekly sumamry ? I will do17:05
rosmaita:)17:05
gmannit is for diversity thing 17:05
dansmithfungi: diversity of contributing companies17:06
gmannyeah that one17:06
TheJuliarosmaita: and maybe "creating a diverse environment where diversity flourishes"17:06
aprice[m]yes, i will be there17:06
gmannwe can do post-PTG also or in PTG with allison from staff ?17:06
aprice[m]please just let me know once that specific time slot is scheduled so I can make sure I am there. 17:06
* TheJulia blinks at aprice[m] joining suddenly17:06
gmannaprice[m]: cool17:06
aprice[m]:) 17:06
gmannlet's do it PTG and I will inform Jimmy, aprice[m] about exact time once we finalized that 17:07
fungioh, a ptg session on increasing diversity of contributing organizations in openstack projects?17:07
aprice[m]ack 17:07
dansmithfungi: a discussion about feedback the foundation supposedly has for us17:07
fungithanks, i misunderstood, thought you had concerns about diversity of ptg attendance17:07
dansmithfungi: remember, you suggested jimmy?17:07
dansmithno17:08
gmannno, it was what from we discussed in baord sync up call17:08
gmannfungi: seems you need more coffee like me :)17:08
* TheJulia could also use more coffee17:08
gmannanyways. skipping 'Recurring tasks check' and 'Open Reviews' topic and let's close the meeting17:08
gmannthanks everyone for joining17:09
slaweqthx :)17:09
gmann#endmeeting17:09
opendevmeetMeeting ended Wed Feb 22 17:09:21 2023 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)17:09
opendevmeetMinutes:        https://meetings.opendev.org/meetings/tc/2023/tc.2023-02-22-16.00.html17:09
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/tc/2023/tc.2023-02-22-16.00.txt17:09
opendevmeetLog:            https://meetings.opendev.org/meetings/tc/2023/tc.2023-02-22-16.00.log.html17:09
slaweqo/17:10
knikolla[m]Thanks all! 17:10
fungisorry, conference room wifi picked a terrible time to die on us. but yes, i remember, you wanted to talk with foundation account management and share input both for things they can bring up with current or prospective member companies17:11
gmannfungi: yeah that one. we will discuss it in PTG with foundation staff17:12
gmannfungi: added it in PTG etherpad and will infoirm Jimmy, aprice[m] about exact schedule once ready https://etherpad.opendev.org/p/tc-2023-2-ptg#L6117:13
fungiand yeah, definitely time for more coffee17:13
fungithanks!17:13
bauzasfwiw, I'm gonna promote a couple of things in the meet&greet sessions at the forum to address the lack of reviewers17:18
bauzaswe also gonna discuss how to attract more contributors in the nova PTG17:19
bauzasno magic17:19
bauzasbut maybe a few rules and a solid communication17:19
fungioh, on a related note, very few people used contributor codes for early bird summit registration, we need to make sure people remember to register if they're planning to attend in vancouver17:20
JayFIt's explicitly part of my job to do this kind of developer outreach and openstack advocacy; and I'm happy to hear ideas as to how to do that more effectively17:20
JayFincluding if you have new contributors who might be struggling, send them to my office hours and I'll help them if I can 17:21
fungithe ptg and forum there aren't going to be super productive if contributors don't come17:21
gmannfungi: oh it is and I think price first increase deadline passed right ? 17:21
fungiyes17:21
gmannI did register I remember. but yeah for forum and PTG we need more contributors 17:22
JayFfungi: FWIW; I think Ironic is mostly planning for it to be virtual due to feedback from our contributors and inability to get travel budget17:22
bauzasfungi: I personnally feel that my outreach is way more beneficial at Summits17:23
gmannbauzas: ++, that will be good for other projects to try too. feel free to advertise it on ML in case any other project want to do same if not yet planned17:23
fungiyep, just want to make sure people remind their colleagues about it. the event organizers have been sending out reminders to mailing lists, but i'm not sure people pay much attention to those17:23
bauzasgmann: it will be first discussed as a PTG topic17:23
bauzasI need my coworkers's support :)17:23
gmanncool, sounds good17:24
clarkbJayF: (and everyone else) There is travel support for the summit if there is interest. I don't know that this will change your plans, but may still be helpful https://openinfra.dev/summit/vancouver-2023/summit-faq/17:31
JayFclarkb: in many of these cases, I don't think it's always 'travel budget', in some cases it's difficulties getting visas or travel due to political constraints. I appreciate the pointer but I doubt that'd have an appreciable impact for Ironic contributors.17:32
JayFPlus I think it'd be weird for RH employees to be tapping into that; it doesn't seem like what it's meant for but IMBW17:32
clarkbsure I don't know all the individual circumstances. I just want o make sure people are aware of hte program for where it makes sense17:33
bauzasin general, we (the RH associates) are asked to not use the travel program17:39
bauzasand leave the program for SME contributors that deserve it17:40
JayFyeah, that's what I would've expected17:40
bauzaspersonally, I plan to do some midcycle thingies at the physical PTG17:42
bauzasno design decisions17:42
bauzassince it's depending on a quorum17:42
bauzasbut we'll still do some work, which will be more or less productive depending on the number of people who can trabel17:43
bauzastravel*17:43
clarkbI also sympathize with the pain of requesting visas particularly to Canada. Did that for my mom a few years back. I had to boot windows for adobe acrobat17:43
bauzasanyway, for outreach, Forum it is17:43
bauzasPTGs are difficult for attracting new contributors17:43
mnaserclarkb: hey us canadians aint that much of a pain!! :p17:44
clarkbya I'm sure a lot of the pain is common across the world since a lot of this type of thing is recipricol17:45
clarkbbut I had to boot windows. I think that was the last time I did that too17:45
gmanncanada visa is taking time now a dyas it is 3-4 month compare to 1 month time before17:47
mnaserclarkb: i wonder if its still the case17:48
gmannmnaser: it is just 20 min drive from me this time :)17:48
JayFgmann: Oh really? Where are you? 17:48
mnasergmann: oh you've moved then!17:48
JayFI'm driving up to Vancouver summit from Tacoma :D 17:48
mnaseryou actually live not in the middle of nowhere anymore gmann =)17:48
gmannJayF: mnaser yeah moved in OCt to BC17:48
JayFBC, CA is one of my favorite places I've visited17:49
gmannmnaser: heh no, in better place now17:49
bauzasgmann: so, next time when the Summit will be in Europe, you gonna move again ?17:50
bauzasman, that's quite an expensive decision17:50
mnaserbc is nice but the cost of living is insane17:50
gmannnice weather here. I have visited Stanley park many times even during winter 17:50
bauzasI knew you were one of the best chairs but I didn't thought of that much dedication :p17:50
mnaserbut then again i remember when i visited CERN to talk about vgpus17:50
mnasermcdonalds was like $40 for one person or smething17:50
gmannbauzas: haha, no more moving now. 17:50
bauzasmnaser: you're talking of a place that's despite being physically in Europe is actully more like Singapour17:51
gmannmnaser: is it? It is costly but not much different except housing price17:51
mnaserbauzas: yeah that sure feels like it17:51
bauzasbut nah, bigmacs aren't that expensive even with a weak dollar17:52
mnasergmann: yeah i guess housing is the wild one, but these days housing is weird in other weird ways :)17:52
gmanncostly but worth with no snow place17:52
mnaseryeah that's true17:52
JayFcost of real estate anywhere in the pacific northwest is pretty insane17:52
mnaserwhile i sit covered in snow outside17:52
bauzasone advice tho : driving fines have to be avoided at most.17:52
mnaserno actual visible streets17:52
bauzasone friend got fined in Geneva for 3 km/h above the limit, he had to pay nearly 700€17:53
mnaserouch17:53
mnaserso everything is expensive, even the fines =)17:53
gmannmnaser: you might be used to it but we had hard time in winnipeg and want to move to no/less snow place at any cost17:53
JayFIs it like in Finland, where fines are indexed to income/net worth?17:53
bauzasnah17:54
mnasergmann: i've just accepted it by now 17:54
mnaser:P17:54
gmann:)17:54
mnaseri dont go out much anyways so17:54
mnasercompany cleans the driveway and the door and if you have to leave car is parked in garage so its kinda easy to live with17:54
gmannyeah17:55
bauzasmnaser: any way to pack and ship a few of those companies and send them to my place ?17:56
bauzas10 centimeters and we're lost for 3 days.17:56
mnaseryeah i guess if its not common its not that big of adeal17:56
mnaserits a super normal thing here to pay for snow removal17:56
mnaserwell, for driveway it is common... for front door i'm a little lazy :)17:57
bauzas:)17:58
knikolla[m]Canada definitely had one of the worst visa processes i've had to go through so far, but not even close to how terrible Ireland and UK were. 18:16
gmannknikolla[m]: is it? it was super easy for me in all different type of visa for me and my family. but the not sure current situation. or you may be you do not need to go through USA visa, 2 years waiting to get a visa application appointment only :) 18:18
JayFUS citizens don't have to get a visa to Canada ahead of time; they are issued on entry. Somewhat lucky for us to not have to do as much of the visa dance. 18:20
JayFbut even as a citizen I always am a little nervous crossing back home18:20
knikolla[m]gmann: Oh yeah, the US is in a league of its own. My parents had to wait 2 years for an appointment and they finally got one a few weeks ago.18:21
knikolla[m]At least both US and Canada issue 10 year tourist visas. Ireland was single entry 6 months, and the paperwork was insane. 18:22
gmannknikolla[m]: yeah, glad they got in finally. For my Mom, Canada long-term visa to stay with me was approved within a week.18:22
gmannyeah18:22
gmannI think most of Europian visa is 6 month single entry most of time18:22
knikolla[m]Interesting. I haven't needed VISAs to Europe, with the exception of Ireland and UK, so I can't comment on that.18:23
gmanntc-members: knikolla[m]: As per the election close date/time, TC chair nomination will be open until "Monday 27th Feb 23:45 UTC". please plan before that. 18:32
knikolla[m]gmann: thanks, I will have a patch up by tomorrow EOD at the latest. 18:41
gmanncool18:45
TheJuliaOh no, visa discussions19:49
TheJulia:)19:49
TheJuliaI kind of appreciated how China does it without appointments. But the paperwork felt a little confusing19:51
*** JasonF is now known as JayF19:51
gmannnoonedeadpunk: did you get chance to look into this? It is ready to merge but waiting as you mentioned to check last week https://review.opendev.org/c/openstack/governance/+/87223320:02
gmanntc-members: this is ready to review too https://review.opendev.org/c/openstack/governance/+/87471920:03
JayFlooking20:04
gmannthanks20:04
JayFheat riding in on horseback to save some tripleo repos \o/ RC+120:04
gmannJayF: you did Code-Review +1 but i think you want to RC+1 on 874719 ?20:06
JayFyes, yes I do20:06
gmannI am sure that is because of gerrit view of voting button :)20:06
JayFI'll let my brain take at least half of the blame LOL 20:06
opendevreviewMerged openstack/governance master: Make DPL/PTL model change deadline as election nomination start  https://review.opendev.org/c/openstack/governance/+/87344020:07
spotzjayf can you post the link to your patch? It's not in my backscroll20:14
JayFhttps://review.opendev.org/c/openstack/governance/+/87448420:15
spotzThanks20:16
JayFthanks for considering it o/20:16
spotzI get it, the version of a service isn't the same across all projects and as CERN has shown you can run multiple versions from different releases20:20
opendevreviewMerged openstack/governance master: Move os-(apply|collect|refresh)-config to Heat's governance  https://review.opendev.org/c/openstack/governance/+/87471920:41

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!