22:00:59 #startmeeting EPEL (2020-12-04) 22:00:59 Meeting started Fri Dec 4 22:00:59 2020 UTC. 22:00:59 This meeting is logged and archived in a public location. 22:00:59 The chair is tdawson. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:00:59 Useful Commands: #action #agreed #halp #info #idea #link #topic. 22:00:59 The meeting name has been set to 'epel_(2020-12-04)' 22:01:00 #meetingname epel 22:01:00 The meeting name has been set to 'epel' 22:01:02 #chair nirik tdawson bstinson pgreco carlwgeorge 22:01:02 Current chairs: bstinson carlwgeorge nirik pgreco tdawson 22:01:03 #topic aloha 22:01:15 michel_slm: Hi 22:01:46 hi, I'm here, sort of 22:01:47 .hello salimma 22:01:48 michel_slm: salimma 'Michel Alexandre Salim' 22:01:53 I guess at 20% attention 22:01:55 morning 22:02:13 20% of a Hi to pgreco 22:02:17 Hi nirik 22:03:28 howdy yall 22:03:47 Hi carlwgeorge 22:04:59 Has anyone seen smooge today? He had a topic he wanted to talk about. 22:05:27 tdawson: you might have to give a 500% hi to make sure pgreco notices :) 22:05:38 :) 22:05:59 I'm sorta surprised pgreco is here, he told me earlier he probrbuly wouldn't make it. 22:06:16 #topic Old Business 22:06:19 yeah, I'm here, fighting fires in the other monitor 22:06:26 epel8-playground / fedpkg 22:06:40 #info fedpkg 1.40 is in Bodhi 22:06:41 It looks like fedpkg has been updated 22:06:42 I think the fedpkg updates finally went out 22:06:48 Ya!! 22:06:52 ah, I can't do info 22:07:06 #info fedpkg 1.40 is in Bodhi 22:07:09 you can, it doesn't output anything 22:07:20 except to the report/summary it sends/creates 22:07:28 my bugzilla entry didn't get linked to the update though - https://bugzilla.redhat.com/show_bug.cgi?id=1900134 22:07:49 not sure if we should just close it as Rawhide unless Ondrej edits the bodhi update 22:08:34 I can go add it 22:09:14 nirik has special power 22:09:46 any provenpackager can I think 22:10:22 odd, I can't 22:10:31 my bad, I can 22:11:18 ok, going to find a package I want branched for epel8 and give that a spin after this meeting 22:12:01 michel_slm: Thank you very much for getting fedpkg changed. 22:12:23 no problem. Ondrej ended up fixing my botched attempt, hope he got his beer 22:12:45 * michel_slm wondering why there's a mystery delivery from RH coming on Monday 22:13:08 It's your fed pkg :) 22:13:15 ha 22:13:35 Anything else for playground before we move on? 22:14:10 Next on agenda ... epel8-next 22:14:36 carlwgeorge: What's the progress? And what help do you need? 22:15:07 smooge was able to get cs8 mirrored into fedora infra (releng#9850). now we're waiting on grobisplitting the modules (releng#9851). 22:15:28 it's blocked on that 22:15:37 Cool ... sounds like good progress. 22:15:49 I know it's blocked, but it's still one step closeer. 22:16:23 * carlwgeorge queues up johnny cash one piece at a time on spotify 22:16:35 :) 22:17:06 Outside of grobisplitting is there anything else you need help with? 22:17:50 i don't believe so, onces that's done we can complete the draft implementation in staging, make sure it works as expected, then repeat it in prod 22:18:27 That's still going to be some work, but it's getting closer. 22:18:42 any known package failing right now in 8.x as a good candidate to start testing? 22:21:38 Since it sounds like things are moving along, I'm going to go to the next topic. 22:21:50 EPEL-Packaging-SIG 22:22:01 What are the next steps? 22:22:32 #info I've updated https://fedoraproject.org/wiki/EPEL/Packagers with some tentative joining instructions 22:22:54 TL;DR simplified from the provenpackagers one, if you want the commit bit you file a ticket but only need one sponsor unless anyone object 22:23:47 and we need to figure out a way to evaluate stalled branch requests. I added an initial query that seems to work for finding requests over a week old. two questions though 22:23:59 1. would it make sense to have a template and make it easy to find (where?) 22:24:21 2. would it make sense to have a tracking bug, and if so how do we get an alias for it 22:24:45 Tracking bug is easy 22:25:25 I don't think I have any special privilidges, but to make an alias, you just fill out the alias part of the bugzilla 22:26:05 but what do you mean about a template? A template that people fill out when they want a package in epel8? 22:26:20 the infra we need to get collaborators branching ability is not available yet, https://pagure.io/epel/issue/106. once that's done we can move on to figuring out a way of expediting packages we care about without invoking the non-responsive maintainer policy 22:26:58 yeah, a template to standardize what such requests look like and automatically block on the tracking bug. in practice it seems that searching Fedora and Fedora EPEL for bugs with summaries that mention epel work well enough 22:27:50 or even just a standard whiteboard word would be nice... 22:27:57 Here's an example EPEL tracking bug, it's what I use for EPEL7 "won't install" bugs - https://bugzilla.redhat.com/show_bug.cgi?id=1647564 22:27:57 but you will never get 100% of people using it. ;( 22:28:37 so re: process -- two things we can decide now (or in the next week to give people more time). is the joining instruction reasonable, and also, should we triage the EPEL branch requests and decide which ones the packagers SIG should take on? (unless we assume we automatically claim any package nobody wants to support, but that seems unrealistic) 22:30:20 might be good to look over the list before commiting to anything? 22:30:23 tdawson: thanks. I'll create a tracking bug once we figure out how to decide what package we care about. maybe some rough guideline e.g. 1) the package is useful on its own, in which case we vote to decide if it is, and 2) the package is a dependency, in which case if the Fedora maintainer doesn't care we have no choice 22:30:26 I'd like to read the instructions a few times to make sure I'm udnerstanding them ... so how about we give that part another week. 22:30:39 * nirik nods. 22:30:58 yup. I'll add the criteria for 'adopting' a package today so it can be evaluated next week too 22:31:40 one last thing: since we agreed to use the epel Pagure tracker, would it make sense to add a tag for epel-packagers-sig? 22:31:51 so we can easily find the issues we're tracking 22:31:57 sure. 22:32:12 Sounds like a good idea 22:32:35 if one of the 5 members want to do that, that'd be great 22:32:48 ok, I think that's all I have for now 22:33:10 we could add more folks to that too... 22:33:19 * michel_slm puts his hand up 22:33:36 I agree with nirik ... let's add a few more people there. 22:34:27 What do you want the tag to be? 22:34:59 do we have existing tags? if not, off the top of my head epel-packagers or epel-packagers-sig 22:36:00 OK, tag added 22:36:02 I see at least one existing tag (qa) but can't see the settings 22:36:13 michel_slm: whats your fas account again? 22:36:52 I'll let nirik setup the user account ... I'm going to move onto the next subject 22:36:58 nirik: salimma 22:37:05 #info EPEL-6 is EOL and archived 22:37:17 hooray 22:37:48 any major complaints out there? 22:38:00 a few on the centos-devel list... 22:38:24 Ya ... I wanna comment on those ... and yet ... I won't 22:39:11 probibly best :) 22:39:41 There was a couple comments on the Scientific Linux list as well ... but it's from the same guy(s) who complain at each EOL ... as well as each new release. 22:40:00 my only complaint is that the mirrorlist for CentOS 6 was broken on purpose... you must hard code the archive URL for yum to work 22:40:41 but same happened for 5. EPEL doesn't do that breakage, thankfully, IIRC 22:41:44 is the archive mirrored? if not, probably doesn't make sense to have a mirrorlist (but that's not an EPEL issue I suppose) 22:41:57 it is not mirrored 22:42:19 I mean, there are a couple of servers, but not distributed or anything like the real mirrors 22:42:28 and having the mirrorlist return vault urls would overload it with traffic from what i'm told 22:42:30 ah. so.. having some friction is probably a good thing anyway. if you admin a server that's running an EOL release you should have to do something 22:42:57 Ya, through all of those, I didn't hear any complaints about EPEL6 ... just CentOS6 and the general Scientific Linux 22:42:57 up until not long ago, vault was only 1 server 22:43:02 right like "make a local mirror" 22:44:11 guaranteed folks on RHEL 6 will continue using EPEL for 3.5 more years, even in its archived state 22:44:31 Thanks to smooge who did the archiving. Everything seemed to go smooth. 22:45:06 OK, I'm assuming it was smooge ... if it was someone else, let me know so my thanks go to them. 22:45:27 And ... moving on before we run out of time. 22:45:35 #topic EPEL-7 22:46:28 I'm basically done with my RHEL7.9 "does it install" campain. Most everything was removed ore rebuilt. 22:47:13 But ... just found out today that we have over 150 old bodhi packages ... packages that have been in epel-testing for over a year. 22:48:13 ah, are these before packages got auto-promoted unless they have negative karma? 22:48:22 Most of them, yes 22:48:32 I was just surprised at the number of them. 22:48:55 I haven't done any checks to see if any of them have negative karma or something like that. 22:49:05 tdawson: what state were tose in? they weren't in the mails 22:49:13 *those 22:49:41 Well ... the first one I looked at is just in testing ... nothing negative or anything - https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2019-abb5b76382 22:50:13 But, I think that one was before we did the auto karma 22:51:18 I'm going to go through them this week, probrubly send out an email with numbers and stats, and have a plan by next week. 22:52:36 At the very least, I'm going to do a comment on them all telling the owners to merge or they will be dropped. 22:53:01 yeah, that one I looked at... it just had no autopush set... 22:53:08 so it needs the maintainer to do something. 22:53:41 Yep 22:53:50 OK, moving on 22:53:52 #topic EPEL-8 22:54:01 Anything for epel8 specifically? 22:54:20 * nirik can't think of anything 22:54:33 #topic General Issues / Open Floor 22:55:08 smooge wanted to bring up the "Retire python34 proposal" for EPEL7 22:55:13 .epel 70 22:55:14 tdawson: Issue #70: Retire python34 proposal - epel - Pagure.io - https://pagure.io/epel/issue/70 22:55:28 Oh ... I guess that would have been for EPEL7 section. :) 22:55:55 I don't know if he was planning on helping with that, or if he was looking for volunteers 22:56:23 queuing up one sort of topic but maybe for when epel9 is branched 22:56:59 * nirik can help, but doesn't want to drive it. 22:57:09 michel_slm: Which topic is that? 22:57:34 nirik: Ya, I really wish the python group had the cycles to do more. 22:58:08 once we have epel9, would it make sense to have fedora-review available on it? (right now it's impossible because it has many dependencies that depend on newer RPM versions or Python versions) 22:58:39 michel_slm: definatly 22:59:07 Get it in early before the dependencies change, so we have at least some version of it in EPEL9 22:59:22 I cross-posted to epel-devel and python-devel so I'll see if Miro has any objections 23:00:33 Looks like our time is up. I'll put the retire python34 proposal on next weeks agenda. 23:00:51 Thanks to everyone for coming, contributing, and helping. 23:00:56 thanks tdawson 23:00:58 Talk to you next week. 23:01:05 #endmeeting