15:00:23 #startmeeting tc 15:00:23 Meeting started Thu Sep 16 15:00:23 2021 UTC and is due to finish in 60 minutes. The chair is gmann. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:23 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:23 The meeting name has been set to 'tc' 15:00:28 #topic Roll call 15:00:30 o/ 15:00:30 o/ 15:00:30 o/ 15:00:32 o/ 15:00:41 Howdy. 15:01:15 short agenda today, so we might finish early 15:01:31 #link https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting 15:02:08 o/ 15:03:11 let's start 15:03:14 #topic https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting 15:03:18 oh 15:03:18 o/ 15:03:25 #topic Follow up on past action items 15:03:32 gmann to remove projects pain points from next weekly meeting agenda 15:03:35 done 15:03:42 yoctozepto to send email to openstack-discuss about getting usage data of TC tags 15:03:51 Done. 15:04:03 that is also done and we will discuss it in detail in next topic 15:04:05 yeah 15:04:12 gmann to add folder or something to store the meeting recording/transcript in governance repo 15:04:41 I tried to do that but i am getting error from git on size of file I am checking in. meeting recording is of ~4 MB i think 15:04:57 gb? 15:05:22 4 GB? Yikes! 15:05:31 (4mb seems like really great compression for an hour-long video call anyway) 15:05:37 ah, ~400 MB 15:05:45 Ah, that makes more sense. 15:05:50 sorry 15:05:56 Because 4 GB would be terrible compression. 15:06:02 INDEED 15:06:03 or high fidelity 15:06:09 oops, caps lock bites again 15:06:21 No need to yell at me. ;-) 15:06:23 I think limit for git is ~100 MB or so 15:06:26 ;) 15:07:01 YouTube? 15:07:23 ^ we can survive a few advertisements 15:07:27 o/ 15:08:01 ok, I will check with foundation if they are ok to publish in openstack/openinfa channel. 15:08:08 ++ 15:08:13 or we want to maintain TC channel separately ? 15:08:39 like cinder did I think ? 15:08:49 Yes, that is what Cinder did. 15:09:25 we can do our own I guess 15:09:34 https://www.youtube.com/channel/UCJ8Koy4gsISMy0qW3CWZmaQ 15:10:41 ok. let me do that way 15:10:54 ++ 15:11:02 spotz to help in google meet setup for next meeting from RDO account. 15:11:16 spotz_ already gave me link. thanks again 15:11:32 #topic Gate health check (dansmith/yoctozepto) 15:12:09 literally start that today 15:12:15 any news on gate, dansmith yoctozepto or other members ? 15:12:16 apologies for the delay 15:12:23 #link https://etherpad.opendev.org/p/health_check 15:12:28 ricolin: hi 15:12:40 I was out of loop this last week 15:13:05 I haven't seen complaints despite it being one of the busiest times for the gate. 15:13:16 I have not seen much failure 15:13:54 yeah it's been good lately I think 15:14:02 \o/ 15:14:33 fwiw, 15:14:40 Hehe 15:14:43 I definitely think it's worth tracking this here 15:15:02 as we've been able to keep tabs on spikes pretty soon after they show up and work to get them resolved because of visibility 15:15:13 so even though it might seem boring, I think this is worth keeping on the agenda 15:15:15 Agreed. 15:15:41 yeah, this is permanent things in agenda even just saying 'all fine' 15:15:57 I know, 15:16:05 I'm just saying, I still think it's worth it :) 15:16:19 +1 15:16:21 now on ricolin etherpad which is on adding some framework to check project health #link https://etherpad.opendev.org/p/health_check 15:17:00 added few comments in etherpad 15:17:22 ricolin: so for contribution data or so you will collect from stackalytics or from gerrit ? 15:17:47 I think from gerrit 15:18:09 depends on what data stackalytics have and gerrit don't 15:18:14 if you need help with the gerrit api, let me know 15:18:21 single company contribution? 15:18:38 fungi, thx 15:18:46 gerrit has no information about contributor affiliation 15:19:22 however the openinfra foundation api (openstackid-resources.openstack.org) can be used to look up affiliation by contributor e-mail address 15:19:40 ok 15:19:42 we have examples of that in the openstack/elections repo 15:19:56 Okay, we can use that API 15:20:16 +1 15:20:17 on release, we need to check with release team about 'to stop doing release if no PTL/liaison ack' or keep doing like they do now and give us data 15:20:39 but release data is very important to check status 15:20:40 it's not perfect since a lot of contributors haven't created a foundation profile or haven't put in their affiliation info or haven't been terribly accurate when they did, but it's better than nothing at least 15:22:03 agree 15:22:26 anyways this is good start, thanks ricolin 15:22:28 so our origin plan is to have a period job for it? 15:22:38 or just a script? 15:22:45 to collect from APIs? 15:23:05 ohk, i thought periodic jobs for gate jobs status 15:23:25 I know last job I was an independent 15:23:36 I think periodic monthly jobs can be fine 15:23:45 +1 15:23:49 or once in two month 15:24:11 like checking 2-3 time in a release is more than enough or even 2 times only 15:24:36 sounds like a good plan 15:24:46 everyone please comment on etherpad 15:25:02 if it's going to be that infrequent, it can just be a script someone runs on their workstation. the gerrit and foundation apis are public and anonymous, you don't even need credentials to query them 15:25:38 having a periodic zuul pipeline which only triggers once every 60 days would be a little odd 15:25:56 that also work and based on when we decide the time to check in TC meeting or so we can run that time 15:26:07 i find that scripts run that infrequently often need troubleshooting because of changes to apis 15:26:25 Makes sense 15:26:33 so it's better to have a human running them who can debug them when they inevitably break 15:27:25 :-) Makes sense. 15:27:27 and that script/job can also help in our one of the future things of 'different project level' checks which we are going to discuss in PTG 15:27:28 I think we can decide after I got the script done 15:27:32 I'm find either way 15:27:40 sure 15:28:17 ricolin: we will check it as part of Xena tracker topic next week. or you want to add it as separate topic ? 15:28:59 let's do it with Xena tracker 15:29:12 I mean do it `in` 15:29:17 ok, +1 15:29:37 moving next 15:29:38 #topic TC tags analysis 15:29:44 #link http://lists.openstack.org/pipermail/openstack-discuss/2021-September/024804.html 15:29:53 yoctozepto: no reply on that seems 15:30:20 who could have expected that? (-: 15:30:38 :-( Yeah, didn't see any responses. 15:30:58 :) 15:31:20 ok, what is the next step then? pausing discussion till ptg? 15:31:24 anyways I am adding that in next newsletter also 15:31:32 I have not received more feedback from you either 15:31:37 newsletter++ 15:31:40 yoctozepto: yeah, we can decide in PTG and wait until than if anyone reply 15:31:41 gmann: That is a good idea. 15:31:42 isn't that the answer we're looking for? not useful? 15:31:52 dansmith: to me it is 15:31:56 me too :P 15:31:57 Seems to be. 15:32:08 but we can maximise the time window as there is no rush 15:32:09 yeah, seems like 15:32:20 and then just drop most of this stuff 15:32:30 in PTG we can decide and start cleanup yup or remove the framework itself 15:32:45 yoctozepto: the once with no project are safer to drop anyways 15:32:50 the one 15:33:06 yeah 15:33:26 thanks for sending email and starting this work 15:33:34 :-) 15:33:48 #topic Open Reviews 15:33:50 #link https://review.opendev.org/q/projects:openstack/governance+is:open 15:34:11 I think PTL assignments are ready to merge, let me check after meeting 15:34:47 this is failing gate #link https://review.opendev.org/c/openstack/governance/+/807837 15:34:59 yoctozepto: fungi I have not read log but you seems discussed it?> 15:35:13 before meeting or so 15:35:37 the error message yoctozepto quoted sounded like something i've run into with setuptools deprecation, so could be a new setuptools release 15:36:13 yeah, it seems we must be using some legacy stuff in governance gate 15:36:56 error in pydot2 setup command: use_2to3 is invalid. 15:36:57 yeah, py38 running as n-v I just realized 15:36:59 i merged some changes in pbr a while back to avoid hitting that codepath in setuptools 15:37:18 maybe it's a combination with old/pinned pbr? i'd have to look 15:37:29 I think the pydot2 package doesn't support python3 properly 15:37:30 yeah, need to look; I haven't had time 15:37:38 I know that e.g. docs repo merged today fine 15:37:43 clarkb: oh, that would also explain it 15:37:53 oh, then we need to drop pydot2 15:38:00 wonder where we use it exactly 15:38:20 tools/universe_dot.py 15:38:29 if the pydot2 package is too old and still tries to set use_2to3 metadata through setuptools that could start breaking 15:38:47 "A tool that will translate a projects.yaml file into a visualized graph." 15:39:04 anyways let's check after meeting 15:39:11 https://codesearch.opendev.org/?q=%5Cbuniverse_dot%5Cb&i=nope&literal=nope&files=&excludeFiles=&repos= 15:39:13 not used 15:39:17 yeah 15:39:20 let's drop with pydot2 15:39:37 yoctozepto: may be on governance site side 15:39:50 that package was last updated in like 2014 15:39:55 pydot seems to be maintained 15:40:29 gmann: we would see that still, no? 15:40:49 yoctozepto: c.o.o should give but not 100% sure 15:41:06 anyways let's check after meeting 15:41:09 that's all for today meeting, anything else to discuss as we have ~20 min time 15:41:52 if nothing else, let's close it. 15:41:53 thanks everyone for joining 15:41:58 #endmeeting