20:00:18 #startmeeting EPEL (2023-06-28) 20:00:18 Meeting started Wed Jun 28 20:00:18 2023 UTC. 20:00:18 This meeting is logged and archived in a public location. 20:00:18 The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 20:00:18 Useful Commands: #action #agreed #halp #info #idea #link #topic. 20:00:18 The meeting name has been set to 'epel_(2023-06-28)' 20:00:19 #meetingname epel 20:00:19 The meeting name has been set to 'epel' 20:00:21 #chair nirik tdawson pgreco carlwgeorge salimma dcavalca dherrera gotmax23 smooge 20:00:21 Current chairs: carlwgeorge dcavalca dherrera gotmax23 nirik pgreco salimma smooge tdawson 20:00:22 #topic aloha 20:00:33 .hi 20:00:35 rcallicotte: rcallicotte 'Robby Callicotte' 20:00:40 .hi 20:00:42 jonathanspw: jonathanspw 'Jonathan Wright' 20:00:47 Hi rcallicotte and jonathanspw 20:00:48 on time this time :D 20:00:54 :) 20:01:00 .hi 20:01:03 carlwgeorge: carlwgeorge 'Carl George' 20:01:13 morning 20:01:19 .hello salimma 20:01:19 .hi 20:01:21 michel-slm: salimma 'Michel Alexandre Salim' 20:01:24 dherrera: dherrera 'Diego Herrera' 20:01:25 * rcallicotte waves at nirik 20:01:30 .hello robert 20:01:31 let's all please be extra nice today, i'm kinda spent 20:01:33 rsc: robert 'Robert Scheck' 20:01:44 Hi carlwgeorge and dherrera 20:01:47 Morning nirik 20:01:50 carlwgeorge: dangit i was gonna flail and yell 20:01:56 Hello rsc 20:02:23 jonathanspw: LOUD NOISES 20:02:37 the squeaky wheel gets the oil, riiiiight? 20:03:06 the squeaky wheel gets my computer turned off and i go outside to touch grass 20:03:31 * nirik waves back at rcallicotte 20:03:54 nirik: is "nirik77" you or someone else? 20:04:32 nirik77 isn't me... nirik99 is... 20:04:49 i saw nirik99 the other day and thought it was an imposter 20:04:53 ah 99, that's it. i was confusing another name's "77" 20:04:58 #topic End Of Life (EOL) 20:04:59 carlwgeorge: yep that's basically what i wasn't sure of 20:05:00 RHEL 7 / epel-7 will go EOL on 2024-06-30 20:05:01 CentOS Stream 8 / epel-8-next goes EOL in 2024-05-31 20:05:03 CentOS Stream 9 / epel-9-next goes EOL in 2027-05-31 20:05:25 nirik99 is my irc client. nirik is matrix (which I use much more) 20:05:41 makes sense. i thought matrix would always have [m] at the end. 20:05:46 ah 20:05:50 or start, or wherever it puts that 20:05:55 #topic EPEL Issues https://pagure.io/epel/issues 20:05:56 https://pagure.io/epel/issues?tags=meeting&status=Open 20:06:47 We currently have two. Let's start with the old packages in testing. 20:07:08 .epel 230 20:07:09 tdawson: Issue #230: old epel7/8 updates in testing - epel - Pagure.io - https://pagure.io/epel/issue/230 20:07:24 dherrera: Do you have an update for us? 20:07:48 you can set your irc side nick to anything you like. just talk to the bridge and tell it. 20:08:20 I pushed a PR for the docs as we discussed last week, haven't created the ticket for releng yet, but my plan is to get that done today 20:09:41 jonathanspw: 77 is Fortran, 99 is C ;) 20:09:50 TIL 20:09:51 That would be this pull request - https://pagure.io/epel/pull-request/237#request_diff 20:10:05 yes :) 20:11:10 dherrera: thank you for that. 20:12:28 I don't really have any other questions ... does anything else have anything before we move on? 20:12:36 anyone else? 20:13:14 Sorry ... that was supposed to be /anything else/anyone else/ 20:14:28 Again, thank you for your work on this. And especially for the pull request for the documentation, so next time this comes up, we can easily remove packages. 20:14:47 no problem :) 20:15:04 .epel 236 20:15:07 tdawson: Issue #236: update rocksdb to newer version in el8 and el9 - epel - Pagure.io - https://pagure.io/epel/issue/236 20:15:24 This is another update request. 20:15:40 so, I didn't have time to look at this too closely, but... is this really an incompatible update? 20:15:50 We don't have to vote on it this week, but I wanted people to see it. 20:17:25 i personally would like more reassurance than "AFAIK nothing will break" 20:18:30 I did verify that nothing requires it ... at least not in RHEL and/or EPEL. 20:18:38 it may be the case that the maintainer looked into it sufficiently and things are highly compatible, but a little more reassuring statement of that would be nice. or examples of the incompatibilities, however small, that are known about. 20:18:53 tdawson: that certainly helps 20:19:01 "nothing breaks on EPEL" does not always mean "nothing breaks for people that use EPEL" 20:19:08 yeah, not clear the changes? abi? api? users need to migrate things manually? 20:19:17 dherrera: You are correct. 20:19:40 .hello ngompa 20:19:44 Eighth_Doctor: ngompa 'Neal Gompa' 20:19:52 Hello Eighth_Doctor 20:20:22 👋 20:22:04 Sounds like we need to ask more questions. Anyone up for putting them on? 20:23:45 Well, we have a week ... let's put the questions on and get answers before next week. 20:24:10 That's it for issues. 20:24:22 #topic Old Business 20:24:51 Does anyone have any old business they want to bring up? 20:26:01 Is everyone being quiet, or is the matrix bridge down again? 20:26:42 i'm here, just super distracted 20:26:53 but i'm also on the irc side 20:27:02 i'm up on matrix 20:27:03 OK, not a problem. 20:27:12 everyone is tired... well at least I am 20:27:12 I'm here 20:27:19 Then moving on ... 20:27:20 haven't checked the Matrix side but yeah dealing with some outage 20:27:27 yeah, it's been an exhausting week 20:27:28 #topic General Issues / Open Floor 20:27:36 I have something quick 20:27:42 michel-slm: Go for it. 20:28:19 still working with gotmax and miro on getting that flit-core fix so we can update flit-core/flit on EPEL9 and get awscli2 in - getting close, I think, it's just getting a clean enough patch 20:28:37 I'll need to ping the bugs for the other missing dependencies, since I just realized it's been two weeks. 20:28:50 What else is missing? 20:28:58 so... ETA another two weeks if we need to ask releng to get access 20:29:05 one sec, let me check 20:29:42 https://bugzilla.redhat.com/show_bug.cgi?id=2215098 20:30:42 Is ruamel-yaml-clib in RHEL? 20:30:53 ruamel-yaml-clib, awscrt 20:31:05 yeah apparently it's supposed to be in CRB, I need to double check 20:31:34 $ fedrq pkgs python3-ruamel-yaml-clib -b c9s -Fnevrr 20:31:34 python3-ruamel-yaml-clib-0.1.2-8.el9.1.x86_64 fedrq-centos-stream-crb 20:31:36 oh crap 20:31:41 the version is too old 20:31:49 we need >= 0.2 -- ok, this will be fun 20:32:31 once we get flit-core in I'll see about seeing if I can just lower that version dependency, otherwise it'll need a request that this library be updated 20:32:31 It's in CRB, so it should be able to update if the maintainers will work with you 20:32:40 yup 20:32:52 michel-slm: if you happen to know, whats the mailman3 for epel9 status these days? 20:32:52 anyway, that's it for me for now, the other thing is not really EPEL specific 20:33:07 i have one depressing thing 20:33:23 nirik: need to find some spare time to look at any missing Django dependency - I'll try and squeeze it in before next week's meeting 20:33:26 thanks for reminding! 20:33:57 let me know if I can help (although I am busy as always, I can try and find cycles) 20:34:42 michel-slm: Not wanting to rush, but are we done so we can get on to carlwgeorge's depressing thing. 20:35:16 me and dherrera are gathering data on the impact to epel based on last week's news from red hat. of course i hope all epel contributors feel welcome to continue participating, but i understand if anyone can no longer justify as much or any time on it, or just has lost motivation. 20:36:08 tdawson: yeah, sorry, let's get to Carl'sthing 20:36:19 so my ask is this: if you are considering stopping or reducing your epel maintenance activities, up to and including orphaning packages, please let us know so we can count up the affected or potentially affected packages. 20:36:22 15:35 me and dherrera are gathering data on the impact to epel based on last week's news from red hat. of course i hope all epel contributors feel welcome to continue participating, but i understand if anyone can no longer justify as much or any time on it, or just has lost motivation. 20:36:25 skipped the bridge 20:36:29 carlwgeorge: personally my only concern will be with supporting epel8 once c8s is EOL 20:36:56 ahh... no wonder there's confusion. yeah I prefer taking these meetings from the IRC side these days 20:37:07 carlwgeorge: i guess count me in. To avoid dramatics the tldr is that if rebuilds are dead it's very likely my day job will have to shift directions to the debian or similar ecosystem at which time i can't justify my time to maintain RHEL/EPEL things. Of course I hope that doesn't happen. 20:37:20 I've been thinking about the time I spend maintaining ansible in EPEL 8 20:37:27 yup, i wasn't gonna call you out but jonathanspw's scenario is exactly what i'm talking about 20:37:48 It would be a move of necessity more than desire - feelings about RH aside I just wouldn't have time for it. 20:38:02 I don't directly have a use there, and I'm not super inclined to spend time on it for a platform I don't see as particularly open anymore 20:38:12 gotmax23: if i'm reading you right you still want to be involved in epel9 and the future but epel8 is a bridge to far, right? 20:38:15 I'm not making any firm decision yet, but wanted to be open 20:38:37 carlwgeorge: Yeah, I guess 20:38:41 * nirik can take back over epel8 if desired. but again, I'm pretty swamped. ;) 20:38:54 at this point i'll say any consideration of reduction is fair game for this, we're talking potential impact 20:39:20 Thanks, I'll let you know 20:39:25 well, reduction specifically related to last week's changes, to keep things honest 20:39:42 no change for me. 20:39:55 just wish i had more time to contribute is all 20:40:06 so we'll count jonathanspw and gotmax23's packages, and anyone else that speaks up. i saw one or two on the mailing list and i need to go back and recheck those. 20:40:35 e.g. someone just requested fd-find for EPEL 8 pointing out RHEL 8 will be around for many more years. was happy to branch it, but if there's no Alma/Rocky it will be hard to maintain going forward 20:40:35 i'm honestly just dreading looking at any lists right now 20:41:15 michel-slm: i know everyone hates subman, but does the devsub not work for you for that? 20:41:46 I reluctantly installed subman on my Fedora laptop and I really don't like it 20:41:51 not that i'm trying to change your mind, we can sidebar that later 20:42:21 It does work with mock and podman, but it also sends a bunch of telemetry to Red Hat 20:42:56 on the epel8 maintenance after c8s eol thing, no promises, but there is talk of continuing commits in gitlab for c8s after the eol, just no c8s builds. i hope it happens. 20:43:13 carlwgeorge: i certainly hope so too 20:43:17 I... localize my build work in containers, so having to deal with subman will be a bit annoying, iirc I tried and it wants to be installed on the host 20:43:31 Yes, that's right 20:43:37 Also, I don't think --forcearch works 20:44:07 gotmax23: i haven't looked closely but i would expect subman to be not that bad. insights on the other hand is all about telemetry. 20:44:51 reminder that koji scratch builds use real rhel and don't require touching anything related to subscriptions 20:45:28 COPR does too, right? 20:45:33 Yes 20:45:38 yes for el8 and el9, no for el7 20:46:06 but that goes away next year anyways, and rhel7/c7 isn't affected, so moot point 20:47:11 that's it from me i guess 20:47:38 carlwgeorge: koji scratch build is great... but it's slower iteration than local builds, esp if you want to test installability (mock --postinstall) 20:47:45 carlwgeorge: Thank you for looking into that. I know it's sortof depressing, but it's good to find out those numbers. 20:47:55 michel-slm: yeah, exactly 20:48:11 but yeah, I'll have to look into a decent way to use subman. there's also the option of having an internal mirror I guess? 20:48:14 Does anyone have anything else for Open Floor? 20:49:33 michel-slm: Eighth_Doctor has a thing for that 20:50:09 yep 20:50:21 Would it be possible to set up maintainer test machines for RHEL 8 and RHEL 9? 20:50:22 https://github.com/datto/rhel-reposync-playbook 20:50:30 https://fedoraproject.org/wiki/Test_Machine_Resources_For_Package_Maintainers 20:50:34 Conan Kudo: Cool 20:50:52 I'm sure the current maintainers of that playbook would welcome contributions 20:50:58 Dalton M: ^ 20:51:08 we do have 9... we didn't have 8 for some reason, I can't recall what 20:51:47 Yeah, but 9 is CentOS Stream 20:51:52 is that a setup like the fedora s390x shared machine people can troubleshoot on? 20:52:00 at least as it currently stands, it looks like AlmaLinux is going to be fine, judging by what themayor has said, the public statement, and what's going on in their git servers 20:52:30 ah, hum. well, yeah, we can sort that or try to for sure. 20:53:09 * nirik makes a note/todo 20:53:31 Rocky is apparently using the actual SRPMs and seems up to date with yesterday's RHEL 8.8.z release while Almalinux is not and already missing things 20:53:48 yeah well... uhhh 20:53:57 I'm going to leave that alone 20:54:09 to give people a glimmer of hope, the shitstorm has renewed conversations about how to give away more free rhel. i'll leave it vague like that because no one knows what that will ultimately look like. but the conversations are happening. 20:54:50 also of course getting a fresh round of "everyone hates subman" feedback reminds people that it needs more work (even though it has improved a lot in the past few years) 20:56:55 Time is coming to an end, anythine else before we close the meeting? 20:57:05 on a brighter note, related to this (the CRB reenablement problem) - now that DNF5 is out, I'm going to look into implementing repo overrides 20:57:25 neat! 20:57:46 tdawson: Nothing from me :). Thanks for running the meeting 20:58:04 * nirik gets ready for the next meeting on the meeting marathon day 20:58:10 gotmax23: oops, no comment there 20:58:22 thanks tdawson and everyone for coming 20:58:52 nirik: would appreciate any help re: mailman, mind getting added to a (low-frequency, promise) Matrix room? it's where we post updates 20:58:59 nirik, I still say you are a machine man! 20:59:19 Thank you all for all your work on EPEL and it's community. And for keeping the discussions civil, even when some of them are on the slightly sad side. 20:59:28 Talk to you next week. 20:59:39 #endmeeting