21:00:14 #startmeeting EPEL (2022-01-26) 21:00:14 Meeting started Wed Jan 26 21:00:14 2022 UTC. 21:00:14 This meeting is logged and archived in a public location. 21:00:14 The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions. 21:00:14 Useful Commands: #action #agreed #halp #info #idea #link #topic. 21:00:14 The meeting name has been set to 'epel_(2022-01-26)' 21:00:14 #meetingname epel 21:00:14 The meeting name has been set to 'epel' 21:00:14 #chair nirik tdawson pgreco carlwgeorge michel dcavalca 21:00:14 #topic aloha 21:00:14 Current chairs: carlwgeorge dcavalca michel nirik pgreco tdawson 21:00:25 .hello rcallicotte 21:00:26 rcallicotte_: rcallicotte 'Robby Callicotte' 21:00:30 .hi 21:00:31 dcavalca: dcavalca 'Davide Cavalca' 21:00:42 .hi 21:00:43 carlwgeorge: carlwgeorge 'Carl George' 21:00:46 morning 21:00:46 .hi 21:00:47 dherrera: dherrera 'None' 21:00:47 Hi rcallicotte 21:00:53 Hi dcavalca 21:00:59 Hi carlwgeorge 21:01:08 Hi nirik 21:01:17 Hi dherrera 21:01:40 .hello robert 21:01:41 rsc: robert 'Robert Scheck' 21:02:03 Hi rsc 21:02:54 .hello salimma 21:02:57 michel: salimma 'Michel Alexandre Salim' 21:03:01 Hi michel 21:03:59 hello 21:04:02 pgreco lt me know that he wouldn't be here. 21:04:09 Hi Ebeneezer_Smooge 21:05:03 #topic EPEL Issues 21:05:03 https://pagure.io/epel/issues 21:05:03 https://pagure.io/epel/issues?tags=meeting&status=Open 21:05:17 Looks like there isn't anything marked for the meeting. 21:05:34 There isn't anything that is new there from last week actually. 21:05:49 https://bugzilla.redhat.com/show_bug.cgi?id=2032208 21:06:00 oh sorry wait on that 21:06:03 * salimma has a conflicting talk, will try to chime in but might miss things 21:06:34 salimma: OK. If there is anything specific you want to bring up, let us know. 21:06:47 #topic Old Business 21:08:00 pgreco: didn't have any update about the macros, other than they are progressing 21:08:53 I've started working again on Will-It-Build, now that we have fedora-minimal in EPEL9 ... but other things keep coming up, but expect something this upcoming week. 21:09:12 Other than those two things, I don't have anything for old Business. 21:10:07 There are a couple conference talks, which might be considered old business, but I'm going to bring them up in the General Issues / Open Floor. 21:10:10 #topic EPEL-7 21:10:20 Anything dealing with EPEL 7 this week? 21:11:13 I'll take the resounding silence as a no. 21:11:19 #topic EPEL-8 21:11:50 i haven't been checking lists lately but where are we at on the playground sunset? 21:12:08 tdawson started it and got some feedback from someone 21:12:41 i think it was just to review the list of repos needing package.cfg fixes? 21:12:42 I have tickets open on releng for turning things off there. I'm also working on the package.cfg step 21:12:54 there's a PR, but I haven't reviewed it yet 21:13:04 this upgrade request blocking one of my epel 8 package is interesting: https://bugzilla.redhat.com/show_bug.cgi?id=2035143 - the maintainer needinfo-ed Josh Boyer, then Josh Boyer cleared it. but it's slated for RHEL 8.6 so ... hopefully it lands in Stream soon? I don't have access to the internal Jira 21:13:11 https://pagure.io/releng/issue/10555 21:14:45 There is still a few more things dealing with playground decommissioning - https://pagure.io/epel/issue/136#comment-774581 21:15:43 I was thinking of removing the playground config from epel-release tomorrow (or at least get the pull request for it) ... unless someone beats me to it. 21:16:04 +1 21:16:32 thanks for all the work tdawson 21:17:25 salimma: Looking at the bug, they are discussing the impact the update will have on other things, some of that discussion may or may not deal with partners and other things, so I won't go into more detail. 21:17:56 tdawson: thanks, I don't really want to know the details beyond... any idea when this lands in Stream? 21:18:12 when it's approved and gets attached to errata 21:18:23 I'm also OK with the plan B of just packaging gflags22 as an EPEL8 only package 21:18:29 As much as I was excited about playground initially, I'm glad to see it retired a dignified death ... then burned with fire. :) 21:18:55 i would hold off on a gflags22 if you can, the bz seems to be making progress. stuff like this never happens quickly. 21:19:01 basically - if this takes more than a few months, by that time we might as well target EPEL 9 instead 21:19:10 tdawson: agreed. if only to get rid of package.cfg 21:19:26 carlwgeorge: yeah, I'd rather wait too 21:20:01 can you see the "target release" field on that bug? 21:20:06 Yep, although I said I was only going to remove the epel9-playground from the files ... but, it's fairly easy to see which files have been edited and which are the default ... I think I'm going to remove all the defaults ... but I'll put those in the email. 21:20:25 it says 8.6, but I don't know if it will land earlier or later in that cycle 21:21:50 well 8.5 was released in november, so with a 6 month minor cycle that means 8.6 should show up in may, and it needs to be in c8s well before that. that's about as clear as it can be at this point. 21:22:21 fair enough. thanks carlwgeorge tdawson for looking this up 21:22:33 Anything else EPEL8 related? 21:22:40 unless on the off chance it takes too long and it gets bumped to 8.7, which my guess is would be unlikely, but always possible 21:23:08 #topic EPEL-9 21:23:32 ahllo epel9 21:23:46 Ebeneezer_Smooge: Yep, looks like your bug falls in with that. 21:24:25 yeah.. all I was going to ask was if the correct command to tell David Sommerseth was : fedpkg request-branch --repo openvpn --branch epel-9 21:24:37 yep. 21:25:04 I don't usually put the --branch on there ... just : fedpkg request-branch --repo openvpn epel9 21:25:30 and if you do it in a openvpn src.fedoraproject.org checkout it will figure thats the package you mean. 21:25:39 speaking of that, do we know if fedpkg request-branch is safe if the repo hasn't been created yet? 21:26:12 yes, it is. you can request a new dist-git repo and several branches before it's created. 21:26:15 e.g. can someone does 'fedpkg request-repo foo bznum && fedpkg request-branch --repo foo epel9' and be guaranteed it gets processed in the right order? 21:26:16 safe in what sense? 21:26:18 nice 21:26:34 * nirik nods 21:26:51 i imagine the folks processing those do the repo requests first, then the branch requests 21:26:54 sweet, thanks 21:27:45 Anything else dealing with epel9 ? 21:27:46 ok thanks I have added my data to that ticket 21:28:27 lots of package branch requests, stalled requests, and dependency sorting out 21:28:35 i need to go through maxamillion's list for getting fedpkg working (or not) and get myself added as the admin for those in epel8/epel9 21:29:02 then my conscience will be clean 21:29:10 :) 21:29:25 one heads up, python-mock maintainer has rightfully declined putting it in epel9, so if you have that as a br you'll need to switch to the stdlib mock 21:29:43 related https://fedoraproject.org/wiki/Changes/DeprecatePythonMock 21:30:35 ditto the linters... 21:30:40 https://docs.fedoraproject.org/en-US/packaging-guidelines/Python/#_linters 21:30:48 another adjustment i've found myself making to rawhide branch in prep for an epel9 branch is...what nirik just said 21:31:37 yeah.. I expect it will take a lot of cleaning and then some 'remove from epel8' 21:31:38 oh yeah...been bit by linters before 21:31:49 Ohh ... ya!! ... sorry ... but I really hated those, cuz they were always changing. 21:31:58 I will need some reviews when that happens. 21:33:04 patching to use stdlib mock is the right thing to do anyway :) 21:33:38 Talking about maxamillion's list. Is the python group against putting those packages in epel9? Or is it just that they don't want to do it? 21:33:48 yeah... I have some packages I need in epel9 where the deps are mostly in except the test deps that shouldn't be running anyway 21:33:50 python-pytest-runner is another similar one, it should always be removed as a br, it adds nothing over running %pytest directly 21:34:18 from my conversation with Miro, it seems like "if you want to maintain it, go ahead but I won't" 21:34:35 i don't blame him 21:34:35 OK, I'm fine with that. Miro has alot on his plate. 21:34:37 I was gonna ask about that in the open floor 21:35:30 Oh, since we've been basically talking about it ... 21:35:35 #topic EPEL-Packaging-SIG 21:36:05 Anything realated to the packaging sig, that we haven't already brought up? 21:36:57 nothing from me, though I can happily report one of the perl package that the kernel devs at work have been clamoring for is finally buildable (public-inbox / lei) 21:37:26 Ya!! 21:37:30 I filed a few more stalled package tickets 21:37:31 carlwgeorge: do we want to discuss what the hackfest will look like? 21:37:36 looks like things are moving forward 21:37:48 #topic General Issues / Open Floor 21:38:31 I have a question... How would one handle a package where python-sig co-maintains the fedora branches but avoids the epel ones? 21:38:53 rcallicotte_: I normally request epel-packagers-sig be added, and my fas personally 21:38:58 ok 21:39:07 Yep, what I was about to say. 21:39:16 Miro would normally assign me to be the EPEL bz contact, to indicate he doesn't want to deal with bugs 21:39:40 Is there a link to getting added to the epel-packagers-sig? Id like to know more about it 21:39:40 salimma: i didn't have a specific plan, was thinking more like an open q&a with people available to dive deep as needed 21:39:42 that is what I should have done with a lot of the ones I force forklifted into EPEL8 21:40:05 except I think the epel-packagers-sig was created after that 21:40:21 https://docs.fedoraproject.org/en-US/epel/epel-packagers-sig/ 21:40:43 thanks tdawson 21:41:17 Ebeneezer_Smooge: I think most of the packages required for building fedpkg should be epel-packagers-sig maintained. So while you are putting yourself on there, I think it would be good to add the sig. 21:42:36 IIRC a SIG can also be made the POC 21:42:46 just add a @ prefix 21:43:15 So, salimma already brought up one other thing ... the EPEL stuff at the CentOS Dojo 21:43:56 Ebeneezer_Smooge: oh yeah, I gave up on that for now .... but I did get a lot of stuff working 21:44:05 Did we want to talk about stuff for the hackfest portion? 21:44:36 Hi maxamillion 21:44:44 I'm hoping my ebranch will be in a state where I can put it in COPR, so we can use it to help people figure out what to package 21:44:59 well, it better be in COPR by then or my talk will flatline :) 21:45:08 Ha! 21:45:21 i'm open to suggestions but i think an unstructured q&a session would be best, and individuals can match up to dive into specific packages, mentor/mentee style 21:45:46 I like that approach 21:45:50 * nirik is happy to be available timeslot permitting. 21:46:00 Sounds good to me 21:46:05 salimma: a talk about packaging in epel that needs a package from a copr repo, instead of putting it in epel itself? :P 21:46:33 nirik: timeslots are settled https://wiki.centos.org/Events/Dojo/FOSDEM2022 21:46:47 the hackfest is 1900 utc that friday 21:47:03 carlwgeorge: haha. well, I don't think I would have time to get it in EPEL itself, given how I have to juggle around things to find time to work on it :) 21:47:07 and it's the last session, so if things are going well we can just keep going as long as people like 21:47:16 I'll try to have it at least submitted, but might need hotfixes 21:47:30 Hopin normally keeps going for quite a few hours after right? 21:47:37 yes they do 21:47:40 .hello2 21:47:41 maxamillion: maxamillion 'Adam Miller' 21:47:43 tdawson: hi :) 21:47:47 IIRC if you're in you can stay in, but you can't join anymore 21:47:49 yeah it doesn't shut off, and people hang around to chit hat 21:47:53 *chit chat 21:48:36 cool. I can make that I think 21:48:43 salimma: no worries, i was just giving you a hard time. totally valid to say "i will be submitting this to fedora and epel, but it's not quite ready for review yet so copr it is". 21:48:56 great use for copr in fact 21:49:30 its a "playground" if you will... 21:49:39 Ha! 21:50:16 Don't call it that ... as Matt noticed, everything "playground" in Fedora ends up fizzling out and dying. 21:50:29 * rcallicotte_ nods 21:51:07 Sorry, I meant to put a :) after that. 21:51:19 ok the upcoming events for EPEL are: the Fosdem Dojo https://wiki.centos.org/Events/Dojo/FOSDEM2022 anything at the DevConf? 21:51:44 There is also the Fedora Council SIG presentation ... 21:52:32 not upcoming anymore, but if anyone missed it and wants to watch i went on the rhel presents livestream earlier to talk about epel https://youtu.be/0hi-bfQxE2A 21:52:37 That is Thu Feb 10, 2022 ... but I have no idea where to find that outside of the invitation I got. 21:53:10 carlwgeorge: I thought that was a great job. Hopefully got rid of some of the misconceptions of EPEL. 21:54:19 Anything else before we wrap things up? 21:54:38 i forget, did we put in an epel talk for devconf? 21:55:04 Not that I know of. I thought it was already closed before we noticed. 21:55:28 yeah that's right 21:56:01 There's a hyperscale meetup though 21:56:08 So we can talk epel there 21:56:28 ok any other shows we need to look at? 21:56:46 s/we/you people doing work while I sit on my ass in the peanut gallery saying we/ 21:56:52 I wouldn't want to overrun the hyperscale people ... but if they don't mind us talking EPEL. 21:57:16 We use epel and try to direct people there, so... 21:57:33 Would love to have some non-hs epel people there so I don't have to juggle two hats 😅 21:57:45 Ebeneezer_Smooge We see you quietly doing things in the background ... 21:58:03 michel Do you have the link/time handy ? 21:58:22 Uh... 8:30 am PT second day? 21:58:41 I don't have it in front of me but search for hyperscale. Afk, sorry, baby fussy 21:58:45 PT is US Pacific Time or Prague Time? 21:58:50 Lol 21:59:00 Pacific 21:59:07 thanks 21:59:10 go play with baby 21:59:12 Ebeneezer_Smooge the epel legend is allowed to take things easy in the peanut gallery from time to time 21:59:24 they will be in college before you know it 21:59:26 :) 22:00:08 Looks like our time is up 22:00:23 Thank you everyone for the good discussions, and for all that you do for the EPEL community. 22:00:35 Talk to you all next week, if not sooner. 22:00:41 thanks all! 22:00:53 #endmeeting