| *** irclogbot_3 has quit IRC | 00:16 | |
| *** tetsuro has joined #openstack-tc | 00:19 | |
| *** irclogbot_2 has joined #openstack-tc | 00:20 | |
| *** tetsuro has quit IRC | 00:39 | |
| *** tetsuro has joined #openstack-tc | 00:41 | |
| *** markvoelker has joined #openstack-tc | 01:36 | |
| *** markvoelker has quit IRC | 01:41 | |
| *** ricolin has joined #openstack-tc | 01:43 | |
| *** tetsuro has quit IRC | 03:02 | |
| *** e0ne has joined #openstack-tc | 03:03 | |
| *** e0ne has quit IRC | 03:07 | |
| *** e0ne has joined #openstack-tc | 03:13 | |
| *** e0ne has quit IRC | 03:18 | |
| *** e0ne has joined #openstack-tc | 03:33 | |
| *** e0ne has quit IRC | 03:37 | |
| *** e0ne has joined #openstack-tc | 03:43 | |
| *** e0ne has quit IRC | 03:48 | |
| *** e0ne has joined #openstack-tc | 03:53 | |
| *** tetsuro has joined #openstack-tc | 03:57 | |
| *** e0ne has quit IRC | 03:57 | |
| *** e0ne has joined #openstack-tc | 04:03 | |
| *** e0ne has quit IRC | 04:07 | |
| *** e0ne has joined #openstack-tc | 06:13 | |
| *** e0ne has quit IRC | 06:17 | |
| *** e0ne has joined #openstack-tc | 06:32 | |
| *** tetsuro has quit IRC | 06:34 | |
| *** tetsuro has joined #openstack-tc | 06:36 | |
| *** e0ne has quit IRC | 06:37 | |
| *** e0ne has joined #openstack-tc | 06:43 | |
| *** e0ne has quit IRC | 06:47 | |
| *** Luzi has joined #openstack-tc | 07:02 | |
| *** e0ne has joined #openstack-tc | 07:03 | |
| *** e0ne has quit IRC | 07:07 | |
| *** tetsuro_ has joined #openstack-tc | 07:15 | |
| *** tetsuro has quit IRC | 07:19 | |
| *** e0ne has joined #openstack-tc | 07:26 | |
| *** tosky has joined #openstack-tc | 07:34 | |
| *** tetsuro_ has quit IRC | 07:41 | |
| *** tetsuro has joined #openstack-tc | 07:41 | |
| *** ralonsoh has joined #openstack-tc | 07:55 | |
| *** markvoelker has joined #openstack-tc | 08:04 | |
| *** markvoelker has quit IRC | 08:09 | |
| *** markvoelker has joined #openstack-tc | 08:14 | |
| *** markvoelker has quit IRC | 08:24 | |
| *** markvoelker has joined #openstack-tc | 08:27 | |
| ttx | smcginnis: I'm surprised that "the application of a trademark for Istio was denied"... Do you have a source for that info? | 08:32 |
|---|---|---|
| *** markvoelker has quit IRC | 08:32 | |
| ttx | The USPTO has ISTIO word mark owned by Google LLC since August 31, 2018 | 08:32 |
| *** ralonsoh_ has joined #openstack-tc | 08:53 | |
| *** ralonsoh has quit IRC | 08:55 | |
| *** tetsuro has quit IRC | 09:12 | |
| *** ricolin has quit IRC | 09:17 | |
| *** Ortet has joined #openstack-tc | 09:29 | |
| *** tosky has quit IRC | 10:08 | |
| *** Luzi has quit IRC | 10:08 | |
| *** irclogbot_2 has quit IRC | 10:08 | |
| *** johnthetubaguy has quit IRC | 10:08 | |
| *** tosky has joined #openstack-tc | 10:10 | |
| *** Luzi has joined #openstack-tc | 10:10 | |
| *** irclogbot_2 has joined #openstack-tc | 10:10 | |
| *** johnthetubaguy has joined #openstack-tc | 10:10 | |
| *** tkajinam has quit IRC | 10:29 | |
| *** Ortet has quit IRC | 11:37 | |
| smcginnis | ttx: https://tsdr.uspto.gov/documentviewer?caseId=sn88101573&docId=SUL20190623140201#docIndex=0&page=1 | 11:43 |
| smcginnis | Earlier rejection - https://tsdr.uspto.gov/documentviewer?caseId=sn88101573&docId=OOA20181217115244#docIndex=4&page=1 | 11:49 |
| smcginnis | "Registration of the applied-for mark is refused because of a likelihood of confusion with the mark in U.S. Registration No. 4635373." | 11:49 |
| openstackgerrit | Luigi Toscano proposed openstack/project-team-guide master: Zuulv3 migration guide: initial round of fixes https://review.opendev.org/741925 | 12:18 |
| openstackgerrit | Luigi Toscano proposed openstack/project-team-guide master: Zuul v3 migration guide: initial round of fixes https://review.opendev.org/741925 | 12:19 |
| tosky | I'm not sure who is the exact owner of project-team-guide; it seems that most of the people are past TC members, but not all currents | 12:20 |
| tosky | maybe the former documentation team? | 12:20 |
| tosky | anyway, I addressed most of the requests (and a few other fixes) for https://review.opendev.org/740727 (which tries to minimize the changes) through https://review.opendev.org/741925 | 12:21 |
| *** ricolin has joined #openstack-tc | 12:37 | |
| *** zbitter is now known as zaneb | 13:07 | |
| *** ricolin_ has joined #openstack-tc | 13:08 | |
| ttx | smcginnis: fun :) | 13:10 |
| *** ricolin has quit IRC | 13:10 | |
| *** ricolin has joined #openstack-tc | 13:16 | |
| *** ricolin_ has quit IRC | 13:17 | |
| mnaser | https://ironicbaremetal.org so i guess this is a tthing now | 13:22 |
| fungi | what a handsome bear | 13:28 |
| *** bnemec has joined #openstack-tc | 13:32 | |
| openstackgerrit | Merged openstack/project-team-guide master: Import the Zuul v3 porting guide https://review.opendev.org/740727 | 13:39 |
| knikolla | o/ | 14:04 |
| gmann | o/ | 14:04 |
| njohnston | o/ | 14:05 |
| *** dklyle has joined #openstack-tc | 14:59 | |
| *** Luzi has quit IRC | 15:09 | |
| *** ralonsoh_ is now known as ralonsoh | 15:10 | |
| *** jgwentworth is now known as melwitt | 15:19 | |
| gmann | tc-members, anyone would like to +W the dragonflow repo cleanup patch which is holding smcginnis to proceed - https://review.opendev.org/#/c/737564/2 | 15:59 |
| jungleboyj | gmann: Done. | 16:01 |
| gmann | jungleboyj: thanks | 16:01 |
| knikolla | beat me to it :) | 16:01 |
| jungleboyj | :-) | 16:02 |
| jungleboyj | The early bird ... | 16:02 |
| tosky | is it just me, or tox -e docs on the governance repository fails because I18n can't be found in projects.yaml? | 16:02 |
| knikolla | has had more coffee :P | 16:02 |
| gmann | tosky: is it? i think we merged many doc after i18 moved as sig | 16:03 |
| tosky | that's why I'm confused | 16:03 |
| gmann | this was latest doc run -https://review.opendev.org/#/c/739150/ | 16:04 |
| fungi | clean your doc/build dir | 16:04 |
| tosky | which should be the same as a local execution (almost) | 16:04 |
| fungi | you probably have stale files | 16:04 |
| *** yoctozepto has quit IRC | 16:12 | |
| tosky | I had already cleaned .tox and doc/build, but it ended up being a bit more tricky: I cleaned reference/projects/ (all but index.rst) | 16:12 |
| fungi | oh, yep | 16:13 |
| *** yoctozepto has joined #openstack-tc | 16:13 | |
| fungi | i forget it builds things outside doc/build | 16:13 |
| fungi | might be a good idea to have the tox target include a command to pre-clean those files | 16:13 |
| *** ricolin has quit IRC | 16:14 | |
| openstackgerrit | Luigi Toscano proposed openstack/governance master: V goals, Zuul v3 migration: update links and grenade https://review.opendev.org/741987 | 16:37 |
| *** dklyle has quit IRC | 17:21 | |
| *** dklyle has joined #openstack-tc | 17:21 | |
| openstackgerrit | Merged openstack/election master: Switch to newer openstackdocstheme version https://review.opendev.org/734854 | 17:35 |
| *** ralonsoh has quit IRC | 17:48 | |
| *** e0ne has quit IRC | 17:55 | |
| mnaser | tc-members: we really haven't had much going into https://etherpad.opendev.org/p/kubernetes-cross-community-topics | 18:03 |
| mnaser | (it also went to the ML and no response there either) | 18:06 |
| mnaser | tc-members: https://review.opendev.org/#/c/741859/2 do we think this is ok? | 18:50 |
| mnaser | because i'm not ok with running glance-api under eventlet | 18:50 |
| mnaser | and not ok with having glance as the unicorn project that needs to still run it's own application server | 18:50 |
| mnaser | sure there's no interest to fix it but if there's no interest to fix it but interest in inserting $some_other_fancy_feature | 18:51 |
| mnaser | we're failing to do "what's best to openstack first" | 18:51 |
| mnaser | it's beyond frustrating. it's very much a kick to folks working on openstack-ansible (because that's how it deployed it) and the overall community effort of going towards uwsgi | 18:52 |
| mnaser | mostly to support some mechanism that is not even fully mature or seen much use yet, image import. | 18:52 |
| fungi | on a related note, it may be worth tc members thinking about the implications of https://review.opendev.org/741846 | 18:55 |
| mnaser | and in another fun note, not only did that change get merged to master, it was backported to ussuri -- https://review.opendev.org/#/c/741992/ | 18:56 |
| mnaser | we've literally went from not recommended to unsupported in a backport | 18:56 |
| dansmith | in like 20 minutes | 18:56 |
| *** tosky has quit IRC | 18:57 | |
| smcginnis | It can be considered a bug that it said it was "not recommended" rather than unsupported though. "You can run this way if you want (even though we know it's horribly broken)". Personally, I'd rather have my software work than have it be broken for the sake of consistency with other projects. None of this is ideal though. | 19:02 |
| dansmith | smcginnis: yeah, unless there's a reason, no need to just remove pure wsgi as an option, right? | 19:03 |
| dansmith | smcginnis: this came about because I wanted to make the pure wsgi server abort if you've asked for import to be enabled, just to avoid advertising to the API clients that it's supported | 19:03 |
| dansmith | my preference would be to keep pure wsgi, but not let it advertise things we know are not usable | 19:04 |
| mnaser | smcginnis: and how about this? https://review.opendev.org/#/c/741497/4 | 19:04 |
| smcginnis | dansmith: Yeah, that makes sense to me. | 19:04 |
| mnaser | instead of warning the user 'you will get degraded experience' or refuse to start if you are in an unsupported config | 19:04 |
| dansmith | smcginnis: https://review.opendev.org/#/c/741497/ | 19:04 |
| mnaser | we instead mark stuff as unsupported and -2 a change like this | 19:04 |
| mnaser | and also now talking about ripping out the wsgi file entirely from glance too. | 19:05 |
| mnaser | so like in case you had a chance to even do it, you're being 'forbidden'. | 19:05 |
| mnaser | to support a feature that openstack client literally added support for, like, a couple weeks ago, because mordred did the work on it. so i cant imagine it being ~so widespread~ | 19:05 |
| dansmith | mnaser: also the failure path is the API takes the import request and says "202: I will do this" and then nothing ever happens.. no indication of failure, etc because a thread just doesn't get scheduled | 19:06 |
| gmann | i agree, not sure why API (with default) should not tell the reality. | 19:07 |
| dansmith | hence my wanting to make sure the operator can't even start in that kind of mode | 19:07 |
| mnaser | to me dansmith is trying to push a trivial fix to improve quality of life for operators and we're just dropping a -2 on it | 19:07 |
| mnaser | and it bothers me because i know i can just work around it but it's a huge disservice to all our users | 19:07 |
| mnaser | and this is why "openstack is hard" | 19:07 |
| mnaser | we're +2W something that marks an entire deployment model (used by one of our own projects and deployers) unsupported, backporting it in 20 minutes | 19:09 |
| mnaser | but we're -2 on a change that stops the user from shooting themselves in the foot. | 19:09 |
| mnaser | and after mordred tirelessly worked on improving glance support in openstacksdk and client | 19:11 |
| mnaser | jokke> mnaser: we just recently dropped the osc promotion from our docs and went back to support only python-glanceclient | 19:11 |
| mnaser | \o/ | 19:11 |
| gmann | hummm | 19:11 |
| knikolla | it's disheartening, to say the least. | 19:11 |
| dansmith | yeah that is like ... o.O | 19:11 |
| mnaser | tbh, i am concerned about the health of the project | 19:12 |
| gmann | we need to sort out these things first before we start keep adding more and more things in openstack | 19:12 |
| mnaser | https://github.com/openstack/openstacksdk/commit/75ae5bf4aadedf47e26264f278c19adb64871735#diff-c2c5af58d554c3e61b4fd2f33ef64bb0 | 19:13 |
| mnaser | https://github.com/openstack/openstacksdk/commit/a66639f806bee1858bee8c929d848b49ae917fb6#diff-c2c5af58d554c3e61b4fd2f33ef64bb0 | 19:13 |
| mnaser | mordred literally went ahead and wrote the code for them | 19:13 |
| mnaser | and they still drop it | 19:13 |
| smcginnis | You think a team of two running a "core" project is watching other projects like this to see if anything related to them have changed? | 19:14 |
| gmann | i think foundation has intern worked in glance in past(if i am not wrong) and we can have more intern to solve these issue if manpower is the issue. | 19:15 |
| mnaser | smcginnis: yes, they are aware of that effort, let me drag out _why_ that surfaced the first place | 19:15 |
| smcginnis | Don't bother. | 19:15 |
| smcginnis | My point is, this is a small team that has been on the "stuff is about to completely blow up over here" list for about 3+ years now. | 19:16 |
| mnaser | ok, but when someone is helping that team and they are disregarding that help and throwing it out | 19:16 |
| mnaser | what do we do then | 19:16 |
| smcginnis | Complaining that they are trying to make sure users of their code actually have an operational system isn't productive. | 19:16 |
| smcginnis | The work dansmith has been doing has definitely been helping. | 19:17 |
| * dansmith slips smcginnis the promised $20 under the table for saying that | 19:17 | |
| gmann | :) | 19:17 |
| smcginnis | I think part of the issue is there hasn't been consistent support from anyone over the last few years. | 19:17 |
| smcginnis | :) | 19:17 |
| * dansmith yanks his $20 back | 19:17 | |
| mnaser | but now we have support, yet it's being greeted with "sorry, we got our own agenda here" | 19:18 |
| dansmith | smcginnis: fwiw I'm well outside of my original feature blast radius, trying to improve everything around this, including devstack and testing, but I definitely understand the drive-by aspect of your point | 19:18 |
| smcginnis | dansmith: Yeah, I was happy to see the work you're doing over there. | 19:18 |
| dansmith | okay you can just keep the $20 | 19:19 |
| mnaser | it hurts nobody to disallow the service from starting and disallow opt-in for image import | 19:19 |
| mnaser | s/disallow/allow/ | 19:19 |
| smcginnis | dansmith: Haha. I can owe you change. | 19:19 |
| dansmith | heh | 19:19 |
| mnaser | y'all can settle this, in the next in-person open infra summit being held in 2026 ;'( | 19:20 |
| dansmith | mnaser: if they're trying to deprecate and remove a thing, then it kindaish makes sense, but I think the much larger issue is that root issue of wanting that in the first place, | 19:20 |
| dansmith | especially with the problems import brings | 19:20 |
| mnaser | import, as an operator, is a huge operational headache | 19:20 |
| dansmith | like requiring a common nfs mount for the staging area across all api workers :( | 19:20 |
| mnaser | it means i need to have enough disk space on my controllers | 19:20 |
| mnaser | ^ that too, because it has no 'service architecture' so theres' no locality | 19:21 |
| mnaser | its not like 'ok im uploaded to controller-5 so glance-importer on controller-5 do your thing' | 19:21 |
| dansmith | there's no way "mount staging on NFS" is going to be workable for anyone real, IMHO | 19:21 |
| mnaser | (i actually thought it would indicate which host it was uploaded on and process it there) | 19:21 |
| mnaser | but this is .. really impossible :\ | 19:21 |
| dansmith | yeah, no | 19:22 |
| dansmith | I'm not sure what the path forward is here, but I think that we probably need to communicate in some way that dropping everything in favor of import isn't okay, and maybe decide if uwsgi (even if handicapped) is something we need them to keep | 19:25 |
| mnaser | i agree with that | 19:26 |
| gmann | +1 | 19:26 |
| knikolla | ++ | 19:26 |
| gmann | what step we can take here from tc ? discussion with glance team (not irc :) but audio or video)? | 19:27 |
| mnaser | i think so, i'd also like the rest of the tc to be involved. i know cloudnull struggled back when they were around OSA with the whole uwsgi thing | 19:28 |
| dansmith | presumably we also need a set of acceptance criteria for how import can be improved to make it acceptable as a replacement | 19:28 |
| mnaser | dansmith: that's very reasonable too | 19:28 |
| dansmith | so, is there a plan forward? | 20:12 |
| mnaser | knikolla: do you think you can help setup an adhoc meeting (i know, its a pita, because we still don't have meeting times) to get to the bottom of this? | 20:14 |
| mnaser | knikolla: i've seen you seem partially involved and bring in operator feedback to this too | 20:14 |
| knikolla | mnaser: sure. it seems more of a how do we get there and who does it rather than lack of good will thing :) | 20:15 |
| knikolla | my involvement is mostly from a "want to understand and help improve" position than from real stakes in the matter | 20:16 |
| *** e0ne has joined #openstack-tc | 20:20 | |
| dansmith | if we're going to communicate that uwsgi needs to be supported even if handicapped, I guess I will wait to abandon my patch | 20:21 |
| dansmith | in case it ends up palatable | 20:21 |
| dansmith | I should say ya'll're instead of we're I guess :) | 20:22 |
| mnaser | tc-members: ^ thoughts on above? | 20:23 |
| knikolla | i'd rather us push to have uwsgi fully supported and find a way to get there | 20:25 |
| knikolla | same for osc | 20:25 |
| dansmith | knikolla: well, landing fixes to the uwsgi mode, even if they're just to call out things that aren't supported now, seems like a step in that direction | 20:26 |
| knikolla | if it's a technical challenge on how to implement a API/worker model, it can be adopted from/to other services as well who may have a need tomorrow. | 20:26 |
| knikolla | dansmith: yeah, totally, just don't want to see that as the endgame. | 20:26 |
| dansmith | knikolla: in case you weren't around earlier, my patch makes sure the api doesn't expose that import works via wsgi, because right now we tell the api clients it works but doesn't | 20:26 |
| fungi | with my not-so-tc-any-more hat on, i favor consistency across our services... attempts at consistency are a big part of what makes them openstack and not just some disconnected pile of projects | 20:26 |
| gmann | +1, that is my take too. if cannot support fully then just document or fail the non-supported features like 741497 | 20:27 |
| knikolla | dansmith: yeah, i agree on documenting what doesn't work | 20:28 |
| dansmith | so, I'm not sure if this fits that or not: https://review.opendev.org/#/c/741992/ | 20:28 |
| gmann | consistency is much needed from usage perspective | 20:28 |
| dansmith | but we just landed docs in *ussuri* that says "don't ever deploy with uwsgi" | 20:28 |
| knikolla | fungi: not only that but it reduces knowledge/tech debt. | 20:28 |
| dansmith | so we might want to consider asking them to revert that? or change it to a list of things that won't work for the time being | 20:29 |
| fungi | knikolla: sure, i see that as a reason for openstack being openstack | 20:29 |
| gmann | +1, that is not backport candidate. | 20:29 |
| knikolla | i'm in favor of asking them to revert the docs patch, if the rest of tc agrees. | 20:30 |
| dansmith | gmann: it's merged already | 20:30 |
| dansmith | er, it's in the gate | 20:30 |
| gmann | knikolla: it will be good we list the things to solve as first step and long term. like revert of backport and keep support wsgi . | 20:30 |
| gmann | dansmith: yeah. i mean that should not have been approved. | 20:30 |
| dansmith | ack | 20:31 |
| mnaser | should we start gathering these in an etherpad? | 20:36 |
| knikolla | mnaser: already on it | 20:36 |
| gmann | yeah | 20:36 |
| knikolla | https://etherpad.opendev.org/p/tc-glance-wsgi | 20:39 |
| dansmith | knikolla: this is the patch to make uwsgi mode not lie to users, which is how this all started: https://review.opendev.org/#/c/741497/ | 20:41 |
| dansmith | fir your references section | 20:41 |
| dansmith | *for | 20:41 |
| knikolla | dansmith: thanks, i was just hunting down for that now | 20:41 |
| dansmith | cool | 20:41 |
| knikolla | please feel free to reword rewrite | 20:43 |
| gmann | added few things we need agreement on to avoid same discussion again. | 20:46 |
| dansmith | mnaser: did you have a pointer to the patch that removed osc language from their docs? | 20:47 |
| dansmith | that's probably relevant too | 20:48 |
| knikolla | https://review.opendev.org/#/c/685247/ i think | 20:49 |
| mnaser | there's the doc change but afaik it goes beyond that | 20:49 |
| dansmith | ah yeah | 20:50 |
| dansmith | mnaser: yep, just context I think | 20:50 |
| mnaser | yeah its just docs | 20:53 |
| knikolla | alright, i'm off for the day. take care everyone. | 20:55 |
| *** bnemec has quit IRC | 20:55 | |
| gmann | knikolla: thanks, GN | 20:55 |
| *** markvoelker has joined #openstack-tc | 20:58 | |
| *** markvoelker has quit IRC | 21:02 | |
| *** bnemec has joined #openstack-tc | 21:10 | |
| *** markvoelker has joined #openstack-tc | 21:19 | |
| *** markvoelker has quit IRC | 21:23 | |
| *** e0ne has quit IRC | 21:45 | |
| *** e0ne has joined #openstack-tc | 22:15 | |
| *** e0ne has quit IRC | 22:20 | |
| *** e0ne has joined #openstack-tc | 22:25 | |
| *** e0ne has quit IRC | 22:30 | |
| *** e0ne has joined #openstack-tc | 22:36 | |
| *** e0ne has quit IRC | 22:40 | |
| *** e0ne has joined #openstack-tc | 22:46 | |
| *** e0ne has quit IRC | 22:50 | |
| *** tkajinam has joined #openstack-tc | 22:53 | |
| *** markvoelker has joined #openstack-tc | 22:53 | |
| *** markvoelker has quit IRC | 23:03 | |
| *** bnemec has quit IRC | 23:20 | |
| *** e0ne has joined #openstack-tc | 23:22 | |
| *** e0ne has quit IRC | 23:26 | |
| *** e0ne has joined #openstack-tc | 23:32 | |
| *** e0ne has quit IRC | 23:37 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!