20:00:26 <tdawson> #startmeeting EPEL (2023-10-25)
20:00:26 <zodbot> Meeting started Wed Oct 25 20:00:26 2023 UTC.
20:00:26 <zodbot> This meeting is logged and archived in a public location.
20:00:26 <zodbot> The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
20:00:26 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:26 <zodbot> The meeting name has been set to 'epel_(2023-10-25)'
20:00:28 <tdawson> #meetingname epel
20:00:28 <zodbot> The meeting name has been set to 'epel'
20:00:29 <tdawson> #chair nirik tdawson pgreco carlwgeorge salimma dcavalca dherrera gotmax23 smooge
20:00:30 <zodbot> Current chairs: carlwgeorge dcavalca dherrera gotmax23 nirik pgreco salimma smooge tdawson
20:00:31 <tdawson> #topic aloha
20:00:34 <neil> .hi
20:00:35 <zodbot> neil: neil 'Neil Hanlon' <neil@shrug.pw>
20:00:38 <pgreco> .hi
20:00:39 <zodbot> pgreco: pgreco 'Pablo Sebastian Greco' <pablo@fliagreco.com.ar>
20:00:39 <neil> afternoon, tdawson :)
20:00:46 <michel-slm> .hello salimma
20:00:46 <smooge> .hi
20:00:47 <zodbot> michel-slm: salimma 'Michel Lind' <michel@michel-slm.name>
20:00:50 <zodbot> smooge: smooge 'Stephen Smoogen' <smooge@redhat.com>
20:00:51 <tdawson> Afternoone neil
20:00:52 * neil goes to get a drink
20:00:55 <dcavalca> .hi
20:00:56 <zodbot> dcavalca: dcavalca 'Davide Cavalca' <davide@cavalca.name>
20:01:05 <smooge> my first time I think ever using that
20:01:07 <tdawson> Hi pgreco and smooge and dcavalca
20:01:13 <tdawson> Hello michel-slm
20:01:43 <michel-slm> hello all
20:01:49 <tdawson> smooge: You real name is now revealed.
20:02:00 <carlwgeorge> .hi
20:02:01 <zodbot> carlwgeorge: carlwgeorge 'Carl George' <carl@redhat.com>
20:02:57 <nirik> morning
20:03:54 <smooge> oh no.. and now my real name be used by anyone in dark magic
20:04:27 <tdawson> Morning nirik
20:04:34 <tdawson> Hi carlwgeorge
20:05:12 <tdawson> #topic End Of Life (EOL)
20:05:13 <tdawson> RHEL 7 / epel-7 will go EOL on 2024-06-30
20:05:15 <tdawson> https://endoflife.date/rhel
20:05:16 <tdawson> CentOS Stream 8 / epel-8-next goes EOL in 2024-05-31
20:05:18 <tdawson> CentOS Stream 9 / epel-9-next goes EOL in 2027-05-31
20:05:19 <tdawson> https://endoflife.date/centos-stream
20:05:42 <tdawson> #topic EPEL Issues https://pagure.io/epel/issues
20:05:43 <tdawson> https://pagure.io/epel/issues?tags=meeting&status=Open
20:06:24 <tdawson> I guess we'll start with the old issue first.
20:06:29 <tdawson> .epel 253
20:06:30 <zodbot> tdawson: Issue #253: Move weekly meetings to Matrix - epel - Pagure.io - https://pagure.io/epel/issue/253
20:06:51 <tdawson> As many of you might have noticed ... we're on iRC this week. :)
20:07:55 <tdawson> There isn't really a rush for the move to matrix, and nirik found some features that were still missing.
20:08:32 <tdawson> nirik: Is there other things besides the email to meetingminutes?
20:09:58 <nirik> I don't think so, but not 100% sure...
20:10:10 <Son_Goku> .hello ngompa
20:10:11 <zodbot> Son_Goku: ngompa 'Neal Gompa' <ngompa13@gmail.com>
20:10:20 <tdawson> I know there isn't !chair ... but I'm not sure if that is needed.
20:10:21 <nirik> anyhow, hopefully soon all will be well.
20:11:22 <tdawson> OK ... well, let's keep the ticket open, and put the status in there.
20:11:35 <tdawson> Hello Son_Goku
20:12:01 <tdawson> Anything else before we move on?
20:13:07 <tdawson> .epel 242
20:13:08 <zodbot> tdawson: Issue #242: Formalizing the EPEL Steering Committee voting process - epel - Pagure.io - https://pagure.io/epel/issue/242
20:13:53 <tdawson> carlwgeorge: Anything needed on this?  Or is it still in the "get the documentation" stage?
20:14:06 <carlwgeorge> i finally put up the pr with the wording for voting
20:14:12 <carlwgeorge> #link https://pagure.io/epel/pull-request/254
20:14:33 <tdawson> Cool
20:14:49 <tdawson> I haven't read it yet ...
20:15:02 <carlwgeorge> we don't have to vote now, happy to give everyone time to review
20:15:31 <Son_Goku> looks good to me
20:15:39 <Son_Goku> it's pretty straightforward
20:15:47 <carlwgeorge> it's mostly the fesco wording, adjusted for how we already work
20:15:57 <tdawson> Sounds good.  Thanks for getting that written ... I'll make sure we vote on it in next weeks meeting.
20:16:02 <jonathanspw> .hi
20:16:03 <zodbot> jonathanspw: jonathanspw 'Jonathan Wright' <jonathan@almalinux.org>
20:16:42 <tdawson> Hi jonathanspw
20:17:17 <tdawson> Anything else before we move on?
20:17:29 <carlwgeorge> not from me
20:17:30 <smooge> not from me
20:17:43 <tdawson> #topic Old Business
20:18:20 <tdawson> Does anyone have any Old Business ?
20:18:58 <michel-slm> kinda sorta
20:19:08 <tdawson> michel-slm: Go for it.
20:19:34 <michel-slm> I have a COPR now to hold mailing list packages that have not made it to epel9 yet https://copr.fedorainfracloud.org/coprs/salimma/mailman-epel/packages/
20:20:08 <michel-slm> so that way once we get to hyperkitty - hoping to get there today - we unblock nirik from working on deploying it while we get these in
20:20:30 <tdawson> That's a very good idea.
20:20:32 <michel-slm> finally got round to adding some features to ebranch to make this easier too, so once I add a few more things I'll cut a new release next week
20:20:37 <nirik> oh? do we have postorious too? I haven't looked lately
20:21:26 <michel-slm> postorius is already in the COPR, and I think I've filed branch requests for everything leading up to it
20:22:03 <michel-slm> (ebranch can now link existing requests to the dependency report, then when you file a request for the topmost package it automatically link in the dependent reports too)
20:22:29 <nirik> cool.
20:22:35 <nirik> thanks again for working on this.
20:22:44 <michel-slm> but I need to make it parse bugzilla CLI's output after filing a bug, and also search for existing issues - the latter is the most annoying one (look at how noisy Debian reportbug's search is)
20:22:50 <michel-slm> np! sorry it took so long
20:24:46 <tdawson> michel-slm: Thank you for that.
20:24:52 <tdawson> Is there any other Old Business?
20:26:19 <tdawson> #topic General Issues / Open Floor
20:26:28 <tdawson> Anything for the Open Floor?
20:26:35 <carlwgeorge> i've got something
20:27:02 <tdawson> carlwgeorge: Go for it
20:27:19 <carlwgeorge> a maintainer is trying to add redhat-lsb to epel9, and there are quite a few problems.  the specific one i wanted to bring attention to here is using recommends instead of requires for dependencies that are not available in rhel9 or epel9.
20:27:23 <carlwgeorge> https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2023-336dbb57e0
20:28:06 <smooge> hahahahahaha (sorry that was outloud)
20:28:41 <carlwgeorge> i've been leaving comments and karma to try to nudge the maintainer in the right direction, but not having much success
20:28:54 <smooge> so seriously.. should it be 'epel-lsb' instead?
20:29:07 <michel-slm> smooge: huh, maaaybe
20:29:17 <Son_Goku> lol no
20:29:23 <tdawson> sorry ... my mind is hurting on that.
20:29:24 <carlwgeorge> oh the package name being wrong-ish is another problem entirely
20:29:25 <Son_Goku> this package shouldn't enter epel or fedora
20:29:27 * michel-slm is of the opinion this is an opportunity to refactor anyone still depending on lsb
20:29:35 <michel-slm> it's like python3-mock but worse
20:29:49 <Son_Goku> this package's existence is why we can't get rid of qt3 or qt4 in fedora
20:30:00 <carlwgeorge> in fact, that might be the easiest fix.  retire redhat-lsb and have the maintainer create a new lsb package that goes through a full review.
20:30:24 <Son_Goku> I was kind of surprised ovasik gave him redhat-lsb
20:30:28 <smooge> I understand but this is an enterprise issue and sometimes the software which 'works' on EL9 but was built for EL5 is still needed for payroll
20:30:50 <carlwgeorge> michel-slm: in several of the bugzillas about this i strongly advised people to port to python-distro
20:31:12 <carlwgeorge> Son_Goku: was it "given" or orphaned and adopted?
20:31:12 <smooge> i can see getting rid of it in Fedora
20:31:15 <michel-slm> inspired by carlwgeorge mentioning this last week ---
20:31:32 <nirik> I would be happy to git rid of it everywhere... but I guess people have some use for it.
20:31:35 <michel-slm> Son_Goku and I were discussing how nice it would be to have a reboot of LSB/FHS that, you know, actually works
20:31:48 <smooge> but Enterprise Linux tend to require things which no one likes
20:31:50 <Son_Goku> carlwgeorge: it seems like the latter
20:32:03 <michel-slm> be less prescriptive about what should be in (like qt3, qt4) and be more multiarch friendly
20:32:18 <michel-slm> smooge: yeah... because standard(TM) :(
20:32:40 <nirik> It might be useful to have the people wanting this note why... the bug just asks for it without any further info
20:32:46 <smooge> well also because the software which runs the printer that does the paychecks has no updates since EL4
20:33:20 <Son_Goku> looks like the original redhat-lsb repo in fedorahosted was never imported into pagure
20:33:23 <Son_Goku> he just made a new one
20:33:24 <smooge> sorry.. [checks notes] EL2.1
20:33:52 <michel-slm> making sure people got paid does sound important
20:34:09 <smooge> that aside.. I think we can say for multiple reasons this package as named and imported may not be a good idea.
20:34:11 <Son_Goku> generally, these days the only thing anyone needs from redhat-lsb is lsb_release
20:34:20 <nirik> yeah.
20:34:27 <Son_Goku> and a few years ago, SUSE ported lsb_release to work with os-release(5)
20:34:31 <carlwgeorge> even just a plain lsb_release package would be better than this
20:34:52 <nirik> right. if thats the use case
20:35:19 <Son_Goku> I packaged one to unbreak puppet when I worked at datto: https://copr.fedorainfracloud.org/coprs/ngompa/lsb_release-shim-el9/
20:35:46 <carlwgeorge> maybe if you package that into epel9 i can convince this maintainer to give up reviving redhat-lsb
20:35:51 <michel-slm> yeah, having lsb_release for compatibility reason would be nice
20:35:54 <Son_Goku> sure
20:35:56 <Son_Goku> I'll do it
20:36:00 <Son_Goku> idgaf
20:36:06 <Son_Goku> it's trivial and I already did it
20:36:09 <michel-slm> esp if we can add a comment there / a loud message in %posttran to say "please port to os-release"
20:36:25 <nirik> you are right that this should require the things it should require according to the spec it purports to use
20:36:42 <carlwgeorge> i try to be correct in my advice :)
20:36:50 <smooge> proposal: carlwgeorge opens a ticket to epel committee on this.
20:36:59 <Son_Goku> smooge: +1
20:37:02 <nirik> :) yeah, I meant I agree and that should block it from going out.
20:37:09 <smooge> we review and say 'this isn't a good idea for EPEL and we have a better idea to use instead'
20:37:11 <Son_Goku> I actually would like to go a step further
20:37:23 <Son_Goku> proposal: carlwgeorge opens a ticket to FESCo about this.
20:37:39 <carlwgeorge> because whatever makes sense for epel in this case also makes sense for fedora
20:37:43 <Son_Goku> ultimately, this is a serious issue that I would rather FESCo weigh in because it also affects Fedora
20:37:55 <smooge> ah I thought this was an epel only package
20:37:58 <Son_Goku> nope
20:38:01 <Son_Goku> it's fedora and epel
20:38:18 <Son_Goku> putting my fedora kde hat on, this package is a bane for us
20:38:26 <Son_Goku> anywhere it exists causes trouble
20:38:30 <nirik> does the fedora one at least have all the requirements it should? (I have not looked)
20:38:33 <Son_Goku> yes
20:38:37 <Son_Goku> and that's the problem
20:38:46 <smooge> ok how about this
20:38:49 <carlwgeorge> no
20:38:50 <nirik> well, it's a different problem. ;)
20:39:00 <Son_Goku> yup :)
20:39:08 <smooge> 2 tickets. one for EPEL and we get done with that quick. And 2 to FESCO and that gets dealt with in a while
20:39:10 <Son_Goku> carlwgeorge: no? then this is even more serious
20:39:22 * Son_Goku goes and looks
20:39:22 <carlwgeorge> my last comment in the bodhi update has several missing things, and i only checked the core module, not all of them
20:40:05 <nirik> for the epel9 update, but thats not checking the fedora packages is it?
20:40:16 <Son_Goku> oh god
20:40:23 <Son_Goku> I wish I hadn't looked
20:40:30 <carlwgeorge> #link https://refspecs.linuxfoundation.org/LSB_5.0.0/LSB-Core-generic/LSB-Core-generic/command.html
20:40:31 <nirik> ah yes. Neat
20:40:40 <carlwgeorge> the spec isn't specific to fedora or epel
20:41:09 <nirik> right, but our packages are... but if it can't conform we should probibly drop it, or get it to conform
20:41:21 <Son_Goku> oh dear
20:41:29 <Son_Goku> didn't we break the ncurses thing recently?
20:41:47 <smooge> michel-slm: if you do go down the route of new lsb.. let me know. I would be interested in signing up
20:42:15 <michel-slm> yup!
20:42:32 <michel-slm> I think this will be something that ... a lot of people cross-distro would be happy to see someone do
20:42:43 <carlwgeorge> i don't even know what install/remove_initd is, or where it comes from.  that may involve new/unretired packages.  but i fully agree with nirik, the package should be dropped or corrected.
20:42:53 <michel-slm> I'll try and get some bandwith for it in a few months, because the current situation is sad
20:43:05 <michel-slm> huh, LSB 5 ?!
20:43:14 <nirik> it requires ncurses-compat-libs but that package was dropped
20:43:22 <Son_Goku> yeah that's what I thought
20:43:27 <Son_Goku> this package is officially totally broken
20:43:33 <nirik> oh no, it's there... hum.
20:43:49 <Son_Goku> it shouldn't be? I remember a ticket about it being dropped
20:44:03 <michel-slm> ncurses-compat-libs... didn't it just get added to EPEL recently? something got added recently anyway, might be something else
20:44:17 <dcavalca> I think I was the one adding that, or requesting it, or something
20:44:18 <nirik> I meant it's in fedora...
20:44:20 <carlwgeorge> so plan of attack:  file a fesco issue about the overall state of the package in fedora.  then file an epel issue saying we want to see fesco's guidance before allowing it in epel9.
20:44:24 <dcavalca> because we definitely need it at Meta :)
20:44:28 <Son_Goku> carlwgeorge: +1
20:44:38 <Son_Goku> dcavalca: please don't tell me you need redhat-lsb too :'(
20:44:49 <dcavalca> yeah no fuck that
20:44:53 <Son_Goku> XD
20:44:54 <smooge> carlwgeorge: +1
20:45:04 <nirik> carlwgeorge: +1
20:45:18 <Son_Goku> as an aside, I'll go ahead and open a package review for lsb_release
20:45:21 <smooge> sorry I forgot I am not to vote anymore
20:45:47 <michel-slm> +1
20:45:50 <carlwgeorge> smooge: i can adjust my voting pr to give you and you alone a perma-vote :P
20:46:06 <pgreco> +1
20:46:09 * tdawson comes out of hiding ... smooge voted for me
20:46:10 <smooge> no no no.. I get to run for office again
20:46:32 <nirik> although it might be better as a devel list thread/discussion than tickets... because I suspect many people will have... opinions
20:46:33 <carlwgeorge> smooge will make epel great again
20:47:09 <tdawson> Anyway ... it sounds like we have a plan of attack.
20:47:27 <tdawson> That sounds wrong .... a plan.
20:47:33 <carlwgeorge> yup, will proceed and yield the rest of the open floor time to anyone else with something to bring up
20:48:12 <tdawson> Anyone with anything else for Open Floor?
20:49:49 <tdawson> If nobody has anything, does anyone mind commenting on how was LinuxFestNW Lite ?
20:51:16 <carlwgeorge> i imagine most of y'all have seen previous versions of it, but here was my talk about epel9 and epel10 https://youtu.be/-7VwIV-Qnww
20:51:33 <Son_Goku> it went pretty well
20:51:49 <dcavalca> it was great to see folks and hang out in person for once
20:51:53 <tdawson> Oh, I didn't know they had the talks on youtube.
20:51:55 <carlwgeorge> the rest are here https://www.youtube.com/@LinuxFestNorthwest/videos including Son_Goku and dcavalca's asahi talk
20:52:00 <dcavalca> our talk on asahi went well and it's on youtube already
20:52:03 <carlwgeorge> yeah they were pretty quick
20:52:16 <dcavalca> carlwgeorge talk also went well and got good discussion afterwards
20:52:30 <michel-slm> wow the talks upload are fast
20:52:41 <michel-slm> I think carlwgeorge converted someone to CentOS Stream
20:52:59 <tdawson> That's always a good thing. :)
20:53:09 <carlwgeorge> at least made him give it proper consideration after previously ruling it out for his usage
20:53:25 <Son_Goku> carlwgeorge: https://bugzilla.redhat.com/show_bug.cgi?id=2246225
20:53:42 * carlwgeorge mutters "fucking rolling release" under his breathe
20:54:54 <Son_Goku> please take a look, it should be trivial
20:54:59 <carlwgeorge> assigned
20:55:33 <tdawson> Son_Goku: Isn't lsb up to version 5, but that is version 3.2 ... or is lsb-release_os-release different?
20:55:46 <carlwgeorge> https://github.com/thkukuk/lsb-release_os-release/tags
20:55:56 <Son_Goku> the original repository was an SVN monorepo
20:56:01 <carlwgeorge> it's technically a different upstream
20:56:09 <Son_Goku> lsb_release was a separate subrepo
20:56:19 <Son_Goku> all of that stuff got screwed up with the move to git
20:56:31 <tdawson> Ah, ok.
20:56:52 <pgreco> every thing you guys keep saying about lsb makes me more confused
20:57:04 <carlwgeorge> it's not just you
20:57:18 <smooge> pgreco: congratulations you now are also on the replacement committee
20:57:20 <Son_Goku> think of lsb like flatpak/snap runtimes
20:57:23 <nirik> it's basically trying to say "if you install this package you conform to version X of the spec'
20:57:40 <Son_Goku> except codified with a spec that everyone was supposed to follow
20:57:41 <nirik> but it is old and useless. ;)
20:57:46 <Son_Goku> and that :)
20:58:00 <neil> Son_Goku: so, with hatred then? :P
20:58:01 <nirik> if you need qt4... you have some issues.
20:58:12 <Son_Goku> if you need *qt3* you have issues
20:58:20 <carlwgeorge> useless for it's original intent, but lots of software has built in hard requirements on the lsb_release command being present
20:58:22 <Son_Goku> redhat-lsb was the reason we can't get rid of qt3
20:58:23 <smooge> like I said.. payroll software
20:58:31 <Son_Goku> until lsb v5, qt3 was the hard dep, not qt4
20:58:51 <nirik> right. old stuff. ;) anyhow....
20:58:55 <michel-slm> when we make our own let's not call it LSB anymore
20:58:55 <smooge> and various other backoffice software or research software
20:59:04 <michel-slm> I get a headache thinking of what LSB means
20:59:10 <Son_Goku> Linux Standard Base? :)
20:59:13 <michel-slm> LSBFHSTheRealDealIPromiseHonorToGod
20:59:20 <smooge> [that james webb is old enough to require something written in qt3 ]
20:59:33 <neil> Love Some Bludgeoning
20:59:53 <carlwgeorge> relevant https://access.redhat.com/solutions/6973382
21:00:01 <smooge> neil: I see you have been on a committee meeting for the LSB :)
21:00:09 <tdawson> Before peoples minds explode with LSB goodness ... it looks like our time is up.
21:00:13 <neil> smooge, no but if you hum a few bars...
21:00:20 <neil> ;)
21:00:46 <smooge> tdawson: but then I have to go back to real work
21:00:48 <tdawson> Thank you all for the good discussions, and for this exciting flashback that I had tried to block out. :)
21:01:05 <smooge> LSB the LSD of Linux
21:01:06 <tdawson> Sorry, meant to say for all the work you do for EPEL.
21:01:14 <michel-slm> LSD sounds like the Canonical rewrite of LSB
21:01:20 <Son_Goku> XD
21:01:21 <michel-slm> too soon? ;)
21:01:25 <tdawson> I'll talk to you all next week.
21:01:26 <michel-slm> thank you all!
21:01:26 <smooge> too soon
21:01:29 <smooge> thanks tdawson
21:01:29 <carlwgeorge> linux standard definition
21:01:37 <tdawson> #endmeeting