Wednesday, 2020-02-26

ricolino/00:58
gmanno/01:00
zanebo/01:00
*** gagehugo has quit IRC01:54
zanebspotz: I think I'm saying it doesn't matter if it's a separate or unified group, because if anybody were going to volunteer to do any work they'd already have done so02:20
*** tetsuro has quit IRC05:20
*** tetsuro has joined #openstack-tc05:22
*** gagehugo has joined #openstack-tc05:23
*** evrardjp has quit IRC05:34
*** evrardjp has joined #openstack-tc05:35
*** tetsuro has quit IRC05:39
*** e0ne has joined #openstack-tc05:59
*** e0ne has quit IRC06:15
*** e0ne has joined #openstack-tc06:16
*** e0ne has quit IRC06:30
*** tetsuro has joined #openstack-tc06:31
*** slaweq_ has joined #openstack-tc07:42
*** slaweq_ is now known as slaweq07:45
*** tosky has joined #openstack-tc08:20
*** iurygregory has joined #openstack-tc08:21
*** witek has joined #openstack-tc09:46
*** witek has quit IRC10:00
*** witek has joined #openstack-tc10:08
*** witek has quit IRC10:15
*** e0ne has joined #openstack-tc10:37
*** e0ne has quit IRC10:38
*** e0ne has joined #openstack-tc10:54
*** e0ne has quit IRC11:27
ttxfungi: I tried dropping the outer link, and it did not result in making the individual badge links in the SVG clickable (at least on GitHub)11:31
ttxSee https://github.com/ttx/glance/blob/master/README.rst11:31
*** ianychoi_ is now known as ianychoi11:32
*** jaosorior has joined #openstack-tc11:32
openstackgerritMerged openstack/governance master: Collect Automation SIG repos  https://review.opendev.org/70838211:35
openstackgerritMerged openstack/governance master: Split OpenDev out of OpenStack Infra  https://review.opendev.org/70313411:36
mnaseruhm11:37
mnaseris that really just a project-update? cause to me it's a little bit more than that.11:37
mnaseri really dont think 4 votes is all it takes to split opendev out of openstack-infra (cc: tc-members)11:39
ttxmnaser: I agree it would have been nice to have more votes on this one, even if it technically *is* a deliverable list change, which only requires PTL signoff and lazy TC approval11:41
mnaseri agree on the technicality, but i don't think we should just treat things technically in that way.11:42
mnaserwhats opendev exactly now? does it live under the foundation? is it outside the foundation?11:42
ttxI guess it's easy to revert if you'd file an objection11:42
mnaserthere no governance proposed/existing afaik inside a repo or anything, as a infra donor i dont know where/who manages those resources11:43
ttxthere is a governance sort-of proposal https://review.opendev.org/#/c/703488/11:43
ttxthe depends-on could have been the other way around11:44
mnaserwhere's this advisory board, do we have members listed already..11:44
ttxbut without standing objections on either reviews... moving forward was probably the right way to go11:44
openstackgerritMohammed Naser proposed openstack/governance master: Revert "Split OpenDev out of OpenStack Infra"  https://review.opendev.org/71001811:46
ttxwe said we'd ask for forgiveness and revert liberally on those lazy-consensus ones, to avoid holding them forever11:46
ttxmnaser: Directly approved, per our house rules and charter.11:50
ttx(see my comment on https://review.opendev.org/#/c/710018/1 if you need more context)11:51
evrardjpmnaser: why didn't you vote negatively then?11:53
evrardjpttx: thanks for doing so.11:53
ttxIt's just hard to commit all of this transition atomically, so personally I was fine with merging this one first.11:55
ttxi.e. merging the chicken before nailing the details of the egg11:57
evrardjpyup exactly12:03
evrardjpand we had already an agreement that our voice will be heard12:04
mnaseri did vote negatively initially12:10
openstackgerritMerged openstack/governance master: Revert "Split OpenDev out of OpenStack Infra"  https://review.opendev.org/71001812:11
mnaserand usually when i chairing at least, once we got a revision, id wait for those who asked for things to be clarified/previously had a -1 if they're satisfied with the current state of the change before merging personally12:11
mnaserbut lets not get bogged down with this and that12:12
evrardjpmnaser: you didn't.12:12
evrardjpWe let a few days too12:13
evrardjpif you did, I would have been aware of. I thought it was just a convo and everything was clear now.12:13
evrardjpbut yeah, let's ignore that.12:13
openstackgerritMohammed Naser proposed openstack/governance master: Revert "Revert "Split OpenDev out of OpenStack Infra""  https://review.opendev.org/71002012:14
mnaseryeah overall the fact that "who is opendev" is still not answered and "we'll figure it out" seems a bit odd personally12:14
mnasernot to be a PITA but i think we should have that nailed down before we split things off12:14
evrardjpare you afraid of something specific, that should be raised?12:17
mnaserwhat is opendev, is it an openstack project, is it going to use all four opens like we have historically done, who's stepping up to run it, there's so many questions that if we don't answer, we'll just end up in a worse off situation12:19
mnaseri'm not against opendev, i think it is a good idea, but we need to make sure we're doing it for the right reasons12:19
*** witek_ has joined #openstack-tc12:36
*** witek_ has quit IRC12:47
*** ijolliffe has quit IRC12:49
spotzzaneb TheJulia fungi - candidates do have to be AUC as it stands. We had 2 candidates who wanted to run but came back to us after the deadline. I kow I submitteed at lleast 1 person for AUC status who was not on the list I was given as an election official12:52
mordredmnaser: are you thinking we should also get opendev to be noted as an OSF pilot project (or something similar) so that it's got an official relationship with the OSF umbrella? (I think that's what we did with the zuul split out?) If so, I believe the OSF staff has the authority to just make pilot projects without board approval, so maybe we should poke staff?13:00
mordredmnaser: (also, I agree, everyone should feel solid on these answers - it's a big and important piece of our community)13:00
*** e0ne has joined #openstack-tc13:04
smcginnisI suppose pilot project status (and presumably not long after, fully approved project) would at least make it clear that OpenDev is and will remain something under the OpenStack Foundation. And then it clearly places it as something that the other OSF projects could expect to work with.13:06
fungispotz: osf bylaws state "1. Membership of User Committee. The User Committee will consist of five members. Beginning with the first election, all members selected to the User Committee shall be Individual Members. A member of the User Committee may cease to be an Individual Member during her or his term, but must be an Individual Member at the time of nomination."13:10
fungihttps://www.openstack.org/legal/user-committee-member-policy/13:10
fungiso nothing in there about requirements for who can hold a uc seat other than being an osf individual member (that's the same as for the tc, in fact)13:11
spotzfungi I'll have to look for it because part of our election official dutiees aree to make sure thee person has AUC status13:12
fungispotz: if so, i can't find anywhere that's backed up in the osf bylaws13:13
spotzBut yeeah I think in this round wee just needed a bit more time to fill the seats but merging had already been brought to us13:14
fungii'm not sure the uc can decide to limit who is eligible to run for a seat beyond the requirements specified in the bylaws, as that would allow seated uc members to directly exclude rivals13:14
fungisame reason the tc can't13:14
fungiit's something we'll likely want input on13:15
fungialso i see a copy-paste error in 2(c) where it refers to "newly elected tc members" ;)13:15
*** ijolliffe has joined #openstack-tc13:17
spotzI joined in the first election wheen it went to 5 from the 3 appointed members mmebers. So don't know aa whole lot of the original formation history. We could track down Melvin or Edgar though13:18
smcginnisIt was likley the policy that was decided on, just not something that was absolutely required in the bylaws.13:20
fungispotz: well, that user committee member policy was added the same time the uc started being an elected body, if memory serves, so you've been on the uc roughly as long as it's been written13:22
fungithough i suppose not for the discussions which led up to it13:23
spotzsmcginnis: Yeah I'm not sure, but the needing to be AUC was I told when I was an official before I was elected and it's part of the communicaation we send out so if not in the by-laaws it has to be somewhere else13:23
fungithe uc added that limitation in their charter: "All members selected to the UC shall be Individual Members and Active User Contributors (AUC) based on the definition in the next section." https://governance.openstack.org/uc/reference/charter.html#user-committee-members13:25
spotzYeah fungi and his search skills:)13:25
fungiwhich means it's also possible for the uc (whatever that winds up being defined as) to change that charter13:25
spotzWe haad considered reducing down to 3 candidaates but tabled that with the mergeer tlaks13:26
spotzerr 3 memberrs13:28
fungireducing it to 3 members would necessitate a bylaws change though13:28
spotzWhich were aalreaady being discussed for aa merrger13:29
fungias one of the options, yep13:30
ttxmnaser, evrardjp: a change of that magnitude could have benefited from a "resolution", in addition to the repo list adjustment.13:34
fungispotz: i think it's only appendix 10 which would have to be adjusted to reduce the number of uc seats, but article ix doesn't appear to have a carve-out for simple votes of the board in modifying that, unlike for the technical committee member policy (appendix 4), so unless the last paragraph of 9.2(d) applies (where it talks about insubstantial changes of member policies) reducing the size of the uc13:36
fungimay require a full vote of the foundation membership13:36
mordredfungi, ttx: you're both smart ... I need to publish artifacts from the ansible-openstack-sig to galaxy.ansible.com into the openstack namespace on galaxy. publishing to galaxy requires github credentials. (publishing to that namespace is currently squatted on with my personal account) ... what do we think the appropriate ownership and management of such credentials should be?13:37
fungimordred: does the namespace have to match the name of a github org used for credentialing? or are they independent namespaces?13:39
mordredfungi: uh. that's a _really_ good question. let me go find out13:40
*** e0ne has quit IRC13:43
fungimordred: though one way to look at it is that the ansible-openstack-sig is not part of openstack, so its output published on galaxy shouldn't masquerade as a product of the openstack community13:43
*** witek_ has joined #openstack-tc13:44
clarkbmnaser: it would be extremely helpful to have that feedback in review13:48
clarkbI'm out today, but we do review specifically to avoid these issues13:49
clarkbbut as far as I can see there was no followup on my response?13:49
clarkbI also think if we try to define everything on day -1 we'll never get this done13:51
mordredfungi: so - I hope that's not how we look at it, because it's only a sig because the openstack community decided that it was inappropriate for openstack to publish GPL content - which is what this content is due to its lineage having come from the ansible repo originally but has now been split out and control over the content has been moved to the openstack community13:53
*** jamesmcarthur has joined #openstack-tc13:53
mordredfor us to decide that collectively we cannot publish content to the openstack namespace on galaxy would be a very unfortunate reading of the situation13:53
fungimordred: oh, this is a sig in openstack, not a sig in ansible? i misunderstood13:54
mordredyes13:54
mordredsorry :)13:54
fungii thought it was like the kubernetes openstack sig13:54
mordrednope, this is an openstack sig we're using to manage the ansible modules used for talking to openstack post divestiture of such modules by the ansible community13:55
mordred(although to be clear, the primary humans involved have always also been openstack community members :) )13:56
evrardjpI didn't realise you were considering yourself human13:56
fungiso you have an openstack sig producing some output which you want to publish in an openstack namespace on galaxy. that seems reasonable, though knowing the credential requirements better would help work out what the access ought to be13:56
mordredfungi: anywho - the namespace does not have to match the name of a github org used for credentialing13:56
mordredHOWEVER - there is a tie between the galaxy namespace and the openstack namespace13:56
evrardjpwhat's the hold up on putting all of this inside openstack/ ?13:57
mordredbetween the galaxy openstack namespace and the github openstack organization that is13:57
evrardjp(and make it part of openstack)13:57
mordredevrardjp: the GPL13:57
evrardjplet me rephrase this13:58
evrardjpis it thinkable of having an exception, to make this happen ?13:58
mordredoh - I don't think there's an issue with the publication ...13:58
mordredthe main quesiton is - what account should we use to publish to the openstack/ namespace on galaxy13:59
mordredright now the answer is "emonty" - which I'm 100% certain is incorrect13:59
evrardjp:)13:59
mordredthen - once we decide what acount, assuming it's not an account that is already owned and managed by openstack-infra, how do we manage the credentials13:59
*** e0ne has joined #openstack-tc14:01
evrardjpthat sounds like a generic question for any artifact we publish outside our control14:01
mordredwell ...14:02
fungimordred: i'm not on the tc, but my suggestion would be for the tc to recognize the openstack-ansible-sig's authority over the openstack namespace in ansible galaxy and let them work out/share the necessary publication credentials with one another (with a primary goal of setting them as zuul secrets)14:02
mordredfungi: nod14:02
fungithem being the sig14:02
mordredyeah. I may also have not been clear that this is about having Zuul publish the content14:02
mordrednot about having humans do so14:03
evrardjpI am puzzled, what fungi said sounded very ... obvious. so I must be missing something.14:03
mordredevrardjp: traditionally openstack-infra takes care of managing credentials used to publish things to openstack namespaces in places - and has a long establish process for managing access to such credentials. but it's not clear to me that this falls into the bucket of that management, and without that, I'm not sure what the answer is14:04
mordredlike - the infra team manages the pypi credentials that are used to upload things to pypi so that invidual teams do not have to worry about that14:05
*** slaweq has quit IRC14:05
mordredfungi: as I say that, I'm not 100% sure why this isn't an infra-managed secret14:06
fungipypi does not have namespaces though14:06
mordred(and I mean that specifically as infra as opposed to opendev assuming such a split exists)14:07
fungiand the infra team is not managing, for example, the credentials used to mirror git refs into the airship namespace on github, or the kolla namespace on dockerhub, or what have you14:07
mordredfungi: right. becuase those aren't openstack projects, which drives the opendev split ... but this is an _openstack_ thing, not an non-openstack opendev thing14:08
mordredoh - you said kllla14:08
mordredkolla14:08
evrardjpmordred: fungi: maybe a change can be introduced in governance-sigs/sigs.yaml to add in Ansible SIG's scope that it handles the publication on galaxy.ansible.com, and we have a resolution in governance explaining this ?14:09
*** slaweq has joined #openstack-tc14:09
mordredyeah - we can do that14:09
fungisome of that kolla namespace situation is also driven by the penultimate bullet (#6) at https://governance.openstack.org/tc/resolutions/20170530-binary-artifacts.html#guidelines14:09
evrardjpI can tackle it, but it sounds like the best place, and the easiest to track14:09
evrardjpI will propose the patches real quick14:10
fungier, and more so bullet #5: We do not want the artifacts to be seen as owned by the community at large, if only a few people can actually address issues. For this reason, the metadata for binary artifacts must not say or imply that they are produced by “the OpenStack community.” Specific team names should be used instead.14:10
mordredevrardjp: cool, thanks14:10
fungithough i suppose ansible roles don't have compiled binary representations independent from their source, so that resolution is probably less applicable14:11
*** witek_ has quit IRC14:12
openstackgerritJean-Philippe Evrard proposed openstack/governance-sigs master: Clarify scope to explain ownership of openstack ns in galaxy  https://review.opendev.org/71004514:12
mordredevrardjp: I think that makes perfect sense14:22
mordredevrardjp: thank you14:23
evrardjpit's just the first part, I will make sure this is acknowledged also in governance using a resolution14:23
mordredevrardjp: cool. then I think what the sig will do as a followup is request that the openstack-infra team be the caretakers of the credentials used to do the uploading, since it's an unambiguous team/upload location and the infra team is good at that14:26
*** jamesmcarthur has quit IRC14:27
fungi(openstack-infra independently from opendev, once that comes to pass)14:28
mordredyah.14:28
*** jamesmcarthur has joined #openstack-tc14:30
evrardjpyeah I think who will hold the credentials don't matter, as we explain who is in charge/who owns those :)14:30
evrardjpwell it does matter, but you got my idea :D14:31
*** jamesmcarthur has quit IRC14:32
mordredevrardjp: yeah. the TC doesn't specifically care about that part :)14:38
fungimaintenance of the credentials can be delegated by whoever is granted authority for them, sure14:39
evrardjpfungi: damn I wrote a text that was three sentences, and now you write a single sentence that is better in this chat.14:47
fungii've come to learn that writing is my primary job these days14:47
evrardjpYou'll be glad to know that I wasn't too far off !14:49
evrardjphaha14:49
*** iurygregory has quit IRC14:49
openstackgerritJean-Philippe Evrard proposed openstack/governance master: Handling the OpenStack name in external services  https://review.opendev.org/71004814:49
mnaserId like to chime in eventually btw14:53
mnaserI’m just in the midst of flying and transferring through airport14:54
mnaserSo I will try and reply soon sorry.14:54
* mnaser is in a gate14:54
*** mugsie has quit IRC15:20
*** mugsie has joined #openstack-tc15:22
*** iurygregory has joined #openstack-tc15:34
njohnstonmnaser: Do you have to write +2+W on your boarding pass before they will let you on?15:40
zanebspotz: if we have volunteers for the UC (albeit after the deadline) then I think we should let them join. the discussion AIUI is premised on the fact that there were no volunteers15:44
spotzzaneb: I agree and offered to stay if needed as well. I didn't re-run to free up aa leadership spot with me on the board. It really cmes down in my mind at least making sure everyone is represented15:48
ttxzaneb: I would not say the lack of volunteers is the premise. The real premise is simplifying our community management and further removing barriers between users and code contributors15:52
zanebthere's something to be said for changing the bylaws to make OpenStack less special, and more like other open infra projects with a single leadership body. that would be simpler.15:54
zanebbut changing the bylaws is not simple :(15:54
*** e0ne has quit IRC15:59
fungithe other open infrastructure projects don't all have single leadership bodies either15:59
*** e0ne has joined #openstack-tc15:59
zanebtrue. at least they get to decide for themselves though :)16:00
fungiairship has a "technical committee" and a "working committee" https://github.com/airshipit/governance#committees16:00
fungiwell, they get to write up what they want and then ask the osf board of directors to approve it, whereas openstack's tc and uc actually get to write their charters and then tell the osf board what those are16:02
fungimaking openstack more like the other projects would mean giving that up, and giving the osf board of directors approval control over all changes to the tc and uc charters16:03
ttxyeah, it's really because openstack was "in" before the Foundation actually was. While other projects were "in" after the Foundation existed16:05
*** iurygregory has quit IRC16:05
ttxso the order of "adoption" was reversed in a way16:06
*** e0ne has quit IRC16:10
*** jamesmcarthur has joined #openstack-tc16:36
*** jaosorior has quit IRC16:57
mnasernjohnston: yep!17:01
gmannevrardjp: you mean  March meeting in Subject... - http://lists.openstack.org/pipermail/openstack-discuss/2020-February/012844.html17:08
evrardjpgmann: oh shoot!17:11
evrardjpmeans I will have trouble finding it after the meeting :p17:11
evrardjpsorry for that.17:12
gmann:)17:14
gmannon UC merge,  now i think doing Bylaw change to close the UC will be good long term solution than having same situation of no UC volunteer either in merged with TC or separate group.17:15
gmannhow costly changing bylaw is now should be less costly than solving the current situation again and again.17:16
fungibut a big part of the question is what to do in the years between now and when that's done17:23
spotzAnd saaying the UC does nothing on the ML by the TC is NOT helping an OPS/Dev divide. While the TC is more in your face the UC often works with the OSF to help bring on new users, work with sharing user feedback and so on17:29
gmannspotz: its not ref to  "current UC not doing anything", it was with new team with 'no duties' idea.17:31
spotzgmann: Not how it came across:(17:32
fungithe point is more that the uc only has responsibilities it has itself declared, so any reformulation of the uc can declare as few responsibilities as it wants17:34
*** evrardjp has quit IRC17:34
*** evrardjp has joined #openstack-tc17:35
mordredfungi: wow. I just noticed that one can be removed from the TC for conviction of a felony17:37
mordredwait - that's just removing chair from the TC chair17:37
evrardjpyeah, sorry I didn't do that yet :p17:38
evrardjpI shouldn't joke about this.17:39
evrardjpsorry17:39
smcginnisHah17:39
fungimordred: i was about to say, i guess you don't know why i'm no longer on the tc ;)17:39
mordredfungi: hahaha17:40
*** e0ne has joined #openstack-tc17:48
ttxspotz: yeah, i can easily see how that can be misinterpreted, especially if someone did not follow all that thread17:49
*** jamesmcarthur has quit IRC17:58
*** jamesmcarthur has joined #openstack-tc18:00
*** jamesmcarthur has quit IRC18:05
openstackgerritHervé Beraud proposed openstack/ideas master: [WIP] voting procedure and decision making  https://review.opendev.org/71010718:06
openstackgerritHervé Beraud proposed openstack/ideas master: [WIP] voting procedure and decision making  https://review.opendev.org/71010718:09
openstackgerritHervé Beraud proposed openstack/ideas master: [WIP] voting procedure and decision making  https://review.opendev.org/71010718:09
openstackgerritHervé Beraud proposed openstack/ideas master: [WIP] voting procedure and decision making  https://review.opendev.org/71010718:12
openstackgerritHervé Beraud proposed openstack/ideas master: [WIP] voting procedure and decision making  https://review.opendev.org/71010718:14
openstackgerritHervé Beraud proposed openstack/ideas master: [WIP] voting procedure and decision making  https://review.opendev.org/71010718:16
openstackgerritKendall Nelson proposed openstack/election master: Set Victoria  configuration  https://review.opendev.org/70847018:18
*** jamesmcarthur has joined #openstack-tc18:21
*** tosky has quit IRC18:56
*** jamesmcarthur has quit IRC19:31
*** jamesmcarthur has joined #openstack-tc19:40
*** jamesmcarthur has quit IRC19:58
*** e0ne has quit IRC20:06
*** jamesmcarthur has joined #openstack-tc20:21
clarkbmordred: my fishing license prominently states my non felon status21:20
mordredclarkb: wouldn't want them felons fishin21:21
*** bnemec has quit IRC21:23
*** slaweq has quit IRC21:32
*** jamesmcarthur has quit IRC21:36
*** jamesmcarthur has joined #openstack-tc21:36
fungiindeed, make them buy fish from law-fearing fishermen and fisherwomen (fishpeople?)21:43
*** jamesmcarthur has quit IRC21:44
*** jamesmcarthur has joined #openstack-tc21:45
*** jamesmcarthur has quit IRC21:59
*** jamesmcarthur has joined #openstack-tc22:00
*** jamesmcarthur has quit IRC22:01
*** jamesmcarthur has joined #openstack-tc22:05
*** tosky has joined #openstack-tc22:12
*** jamesmcarthur has quit IRC22:29
*** jamesmcarthur has joined #openstack-tc22:30
*** ijolliffe has quit IRC22:34
*** jamesmcarthur has quit IRC22:35
*** jamesmcarthur has joined #openstack-tc22:35
*** gagehugo has quit IRC22:41
*** jamesmcarthur has quit IRC22:49
*** jamesmcarthur has joined #openstack-tc22:54
*** jamesmcarthur has quit IRC23:15
*** bnemec has joined #openstack-tc23:18

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!