20:00:31 <tdawson> #startmeeting EPEL (2023-06-14)
20:00:31 <zodbot> Meeting started Wed Jun 14 20:00:31 2023 UTC.
20:00:31 <zodbot> This meeting is logged and archived in a public location.
20:00:31 <zodbot> The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
20:00:31 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:31 <zodbot> The meeting name has been set to 'epel_(2023-06-14)'
20:00:33 <tdawson> #meetingname epel
20:00:33 <zodbot> The meeting name has been set to 'epel'
20:00:34 <tdawson> #chair nirik tdawson pgreco carlwgeorge salimma dcavalca dherrera gotmax23 smooge
20:00:34 <zodbot> Current chairs: carlwgeorge dcavalca dherrera gotmax23 nirik pgreco salimma smooge tdawson
20:00:36 <tdawson> #topic aloha
20:00:37 <rsc> .hello robert
20:00:38 <zodbot> rsc: robert 'Robert Scheck' <redhat@linuxnetz.de>
20:00:40 <jonathanspw> .hi
20:00:41 <zodbot> jonathanspw: jonathanspw 'Jonathan Wright' <jonathan@almalinux.org>
20:00:45 <rcallicotte> .hi
20:00:46 <zodbot> rcallicotte: rcallicotte 'Robby Callicotte' <rcallicotte@mailbox.org>
20:01:01 <dherrera> .hi
20:01:01 <tdawson> Hello rsc
20:01:01 <zodbot> dherrera: dherrera 'Diego Herrera' <dherrera@redhat.com>
20:01:13 <tdawson> Hi jonathanspw rcallicotte and dherrera
20:01:23 <carlwgeorge> .hi
20:01:24 <zodbot> carlwgeorge: carlwgeorge 'Carl George' <carl@redhat.com>
20:01:25 <michel> .hello salimma
20:01:27 <zodbot> michel: salimma 'Michel Alexandre Salim' <michel@michel-slm.name>
20:01:34 <jonathanspw> howdy tdawson
20:01:37 <tdawson> Hi carlwgeorge
20:01:43 <tdawson> Hello michel
20:02:27 * michel traded schedule with his better half so can finally make this meeting again
20:02:53 <michel> Davide sends his regards, he has the CentOS board meeting
20:04:58 <tdawson> #topic End Of Life (EOL)
20:05:00 <tdawson> RHEL 7 / epel-7 will go EOL on 2024-06-30
20:05:01 <tdawson> CentOS Stream 8 / epel-8-next goes EOL in 2024-05-31
20:05:03 <tdawson> CentOS Stream 9 / epel-9-next goes EOL in 2027-05-31
20:05:22 <tdawson> #topic EPEL Issues https://pagure.io/epel/issues
20:05:23 <tdawson> https://pagure.io/epel/issues?tags=meeting&status=Open
20:05:57 <tdawson> This week we're down to just one issue marked with meeting.
20:06:12 <tdawson> .epel 230
20:06:13 <zodbot> tdawson: Issue #230: old epel7/8 updates in testing - epel - Pagure.io - https://pagure.io/epel/issue/230
20:06:50 <neil> o/
20:06:52 <neil> .hi
20:06:53 <zodbot> neil: neil 'Neil Hanlon' <neil@shrug.pw>
20:06:54 <tdawson> dherrera: How well is this doing?
20:06:59 <tdawson> Hi neil
20:07:14 <dherrera> i'm slowly working on it, found more orphaned projects and I have a list of retired that will upload later this afternoon after I finish dealing with the rest
20:07:49 <dherrera> I need to confirm those are retired though, they might just be projects that never got an epel branch
20:08:03 <neil> BTW ive seen the messages in #epel re: mactel-boot.. will wait til open floor to bring up
20:08:54 <tdawson> dherrera: Is there anything that you need help with?
20:09:14 <pgreco> .hi
20:09:15 <zodbot> pgreco: pgreco 'Pablo Sebastian Greco' <pablo@fliagreco.com.ar>
20:09:23 <tdawson> Hi pgreco
20:09:26 <pgreco> I'm sorta here today ;)
20:09:29 <neil> Heya Pablo
20:09:35 <dherrera> not that I think of, I just need to deal with it
20:09:50 <neil> I'm still hearing your voice when I read your messages btw. it's concerning
20:10:08 <tdawson> dherrera: Thank you very much for doing that work.  Feel free to reach out if you need help with it.
20:10:15 <pgreco> hehehe, I have that annoying effect
20:10:21 <dherrera> 👍️
20:10:53 <tdawson> Anything else on this before we move on?
20:11:53 <tdawson> #topic Old Business
20:12:54 <tdawson> I don't have anything on Old Business that needs to come up unless others bring them up. ... any Old Business that people want to bring up?
20:13:50 <michel> does ansible count as old business? otherwise we can wait until open floor
20:14:21 <tdawson> michel: Well, we've brought ansible up before ... is it relating to the old stuff, or something new?
20:14:42 <michel> might be new given the version I have in mind is currently only in Rawhide ... so I'll wait
20:14:51 * michel wondering if gotmax23 is here
20:14:52 <tdawson> ok.
20:15:50 <gotmax> pong
20:15:51 <tdawson> I'm going to move on then
20:15:56 * gotmax will be out of another meeting in 5ish minutes
20:15:59 <tdawson> #topic General Issues / Open Floor
20:16:03 <gotmax> tdawson: sounds good
20:16:23 <tdawson> neil: I think you were first in line for Open Floor
20:17:47 <tdawson> Though, before you say to much ... if you want an execption from the two week rule for here - https://docs.fedoraproject.org/en-US/epel/epel-package-request/ ... the answer is going to be no.
20:18:55 <neil> yeah I don't want an exception, just want to discuss what the right thing to do. it seems as though the upstream software is gone, but it's still in fedora (and AFAIK required for Mac-type things to work)
20:19:16 <neil> for clarity, we're talking about mactel-boot
20:19:30 <tdawson> Ah ... ok
20:20:06 <carlwgeorge> it's up to the maintainer to sort out, either by retiring the package or switching to a maintained fork
20:20:38 <carlwgeorge> if the maintainer doesn't answer a bug requesting resolution for that problem, then you can follow https://docs.fedoraproject.org/en-US/fesco/Policy_for_nonresponsive_package_maintainers/
20:20:47 <neil> well, the maintainer here is Matt, then. who you've said isn't part of Fedora.
20:21:05 <carlwgeorge> yeah i fully expect radio silence from him
20:21:06 <neil> so I guess I'm also wondering how this package isn't orphaned.
20:21:46 <carlwgeorge> if someone just stops participating in fedora, nothing happens to their packages until they fail to mass rebuild or someone follows the non-responsive process
20:22:00 <carlwgeorge> might be failed two mass rebuilds in a row, i think
20:22:11 <tdawson> Looks like florien put in a ... "Port to C99" for the package.
20:22:25 <tdawson> So at least someone is keeping it buildable.
20:22:34 <neil> well, don't we clean up orphaned maintainers? we just had a bunch of packages orphaned because the maintainer was inactive
20:22:55 <carlwgeorge> this package isn't orphaned
20:23:05 <tdawson> "we" as in the EPEL committe don't, but Fedora does.
20:23:15 <michel> we clean up provenpackage access only, right?
20:23:16 <carlwgeorge> orphaned packages are retired after 6 weeks iirc
20:23:21 <michel> we as in Fedora
20:23:50 <michel> but for this particular case I'd say just use the EPEL escalation process to get access
20:23:52 <carlwgeorge> michel: yes, provenpackagers with no activity for some amount of time have that perm revoked.  i think like a year.
20:24:11 <carlwgeorge> i disagree, adding it to epel is a bad idea if it's dead upstream
20:24:28 <carlwgeorge> first that problem should be resolve, hopefully by finding a maintained fork
20:24:33 <carlwgeorge> *resolved
20:24:34 <jonathanspw> What's the true definition of "dead upstream" though.
20:24:38 <michel> ah, I missed that part, sorry. let me ask the Asahi folks
20:24:39 <jonathanspw> Could be looked at a bunch of different ways.
20:24:49 <carlwgeorge> jonathanspw: it doesn't exist anymore
20:24:52 <michel> but I'm guessing from the name this is only for Intel Macs?
20:24:54 <carlwgeorge> (in this case)
20:25:11 * tdawson laughs when he reads the bug for the fix Florian put in  "Filing this here because the upstream URL leads to nowhere, and the mjg59 Github organization does not contain mactel-boot, either."
20:25:26 <jonathanspw> But if it has no known security vulns or major bugs how's it different from tons of other mostly-abandonware that's packages and might have a source site that hasn't been touched since 2005?
20:25:34 <carlwgeorge> oh so florian noticed this problem and then walked away slowly, lol
20:26:13 <carlwgeorge> jonathanspw: it's not different, and as those are found they should be actioned as well, either retired or switched to a maintained fork
20:26:25 <tdawson> It sounds like abandonware that is living on it's own inertia in various linux distributions.
20:26:44 <carlwgeorge> regardless, adding it to an epel branch when it's maintainability is in question is not a good idea
20:26:51 <jonathanspw> Then we're back to the question of what's abandonware/dead?  Say the site still works but no updates in 10 years...
20:27:10 <jonathanspw> I'm playing devil's advocate for more specifics here is all.  We're talking in vague terms about what's dead and what's not.
20:27:11 <michel> ah, description: "Files for booting Fedora on Intel-based Apple hardware using EFI."
20:28:14 <neil> I think it'd help if I write up in the bugzilla what exactly I'm trying to solve, and why.. but that is , I guess, a different question from what we do about the abandonware. personally I have no skin in the game, really, but it does seem people in fedora are probably relying on it, and folks have asked for it in rhel/rocky, which is why I filed
20:28:14 <neil> the rfe
20:28:22 <michel> this discussion seems to be something more appropriate for FESCo by the way? (whether we should scan for and prune software with dead upstreams)
20:28:36 <neil> I think so, michel
20:28:39 <carlwgeorge> jonathanspw: for the sake of the meeting lets limit this convo to mactel-boot, and for the wider discussion and possible guidelines file an fpc ticket
20:28:56 <jonathanspw> carlwgeorge: sounds good, +1 for michel's idea of letting fesco figure it out
20:29:00 <jonathanspw> (more broadly)
20:29:06 <tdawson> +1
20:29:27 * carlwgeorge starts digging in the guidelines to see if this is already covered in some form
20:29:33 <smooge> jonathanspw: in this case there is no place to download the source code except from koji
20:29:45 <pgreco> https://github.com/detly/mactel-boot/commits/master
20:29:52 <neil> BTW, thank yall for putting up with my insanity 😉
20:29:53 <jonathanspw> smooge: i guess that's a pretty good definition that could contribute to defining dead then
20:30:00 <pgreco> apparently someone at least put it somewhere with the intention of doing something
20:30:13 <pgreco> it even includes Florian's patch (merged 2 years ago)
20:30:47 <tdawson> I was just about to point to that.
20:30:59 <smooge> pgreco: sigh google and github search failed me when I looked
20:31:02 <smooge> thanks
20:31:28 <tdawson> If there is a repo that can accept patches ... in my opinion, it's not dead.
20:31:36 <pgreco> it took more than 2 years to merge the PR, but still :)
20:31:46 <neil> doesn't matter still merged /s
20:31:48 <tdawson> Otherwise my favorite editor (nedit) would be dead by now.
20:32:11 <tdawson> Not exactly healthy, but not dead.
20:32:16 * michel recalls fortune cookies, where he thinks the upstream now is just the Debian repo
20:32:28 <smooge> it was when I was dealing with that package
20:32:31 <smooge> anyway..
20:32:31 <pjones> tbh someone probably ought to just write a patch for hfsutils to make some subcommand there do the *one line of code* that's actually meaningful here
20:32:41 <carlwgeorge> and no activity in almost 2 more years
20:32:58 <neil> pjones: I think that might be the smarter play
20:33:05 <pjones> why would there be any activity? It's literally one ioctl() call.
20:33:14 <smooge> so my understanding from what sgallagh found was that this is all incorporated into grub2-macbless provided by `grub2-tools-efi`
20:33:47 <pjones> smooge: yeah, we haven't used mactel-boot for a long long time AFAIK
20:34:01 <smooge> so its a tool which is only needed for older EL releases
20:34:06 <pjones> this probably could just get killed
20:34:08 <sgallagh> Everything except an absolutely trivial bit of shell script that Anaconda has been exec()-ing for a decade and could easily incorporate.
20:34:09 <carlwgeorge> pjones: the context is if it's an actively maintained upstream to determine if the package should be retired or not
20:34:26 <neil> I'm happy to own it for the rest of el8/9, if necessary...
20:34:45 <smooge> neil: its in Rocky as /usr/sbin/grub2-macbless
20:35:05 <smooge> s/its in/it should be in/
20:35:19 <smooge> I don't know if EL8 has it
20:35:42 <tdawson> carlwgeorge: Actually, the real question is whether it should be in epel or not ... retiring is a Fedora thing.
20:36:01 <smooge> actually it looks like it is in EL8 and EL9 as that
20:36:06 <carlwgeorge> and in my mind the epel questions should be put on hold until the fedora status is resolved
20:36:38 <neil> #link https://forums.rockylinux.org/t/error-installing-rocky-8-4-on-macbook-pro/2935/2
20:36:54 <carlwgeorge> if that fork that was mentioned is determined to be good enough, fedora can switch to it then neil can do the stalled epel request to add it to epel8
20:37:36 <tdawson> If/when you do add it to epel8 ... be sure to update the URL
20:38:49 <tdawson> How about we move on to michel's ansible stuff.
20:38:54 <michel> yay!
20:39:00 <tdawson> is max here?
20:39:04 <neil> can do. I'll file a pr for it in rawhide and we'll go from there. thank you, all.
20:39:17 <tdawson> sorry, getmax
20:39:21 <gotmax> 👋
20:39:28 <michel> gotmax23 should be here, last he said he had 5 mins left on another meeting
20:39:31 <michel> oh hai
20:39:48 <michel> so -- Ansible 8 / ansible-core 2.15 is in Rawhide only ATM. everything else is on 7.x / 2.14
20:40:23 <gotmax> Fedora Rawhide and EPEL 8/9 Next are on Ansible 8
20:40:28 <michel> but 7.x is EOL next year, so ... some of our internal folks are asking what's the timeline for getting 8 in EPEL 9. and... this would be considered an incompatible update, I guess?
20:40:52 <michel> gotmax: ah! thanks
20:41:03 <gotmax> Ansible 8 will be in RHEL 8.9 and 9.3, as RHEL will update to 2.15
20:41:03 <michel> hard to check with the infra ddos happening right now
20:41:22 <gotmax> 8.9 and 9.3 are the next ones right?
20:41:29 * gotmax can never keep the numbers straight
20:41:29 <jonathanspw> yes
20:41:32 <jonathanspw> EOY-ish
20:41:44 <gotmax> right
20:41:52 <michel> that sounds like a good timeline. thanks
20:42:05 <gotmax> sure!
20:42:34 <tdawson> Well ... that was quick.
20:42:39 <michel> just confirmed with rpmdistro-repoquery centos-stream 9 ansible - yup I see 8
20:43:16 <tdawson> Anything else for Open Floor?
20:43:56 <michel> one more from me - I'm working on getting awscli2 into EPEL 9
20:44:10 <tdawson> michel: Oh ... nice
20:44:13 <michel> everything builds locally, some tests fail (14 out of 64k so I can check them) - but this involves updating python-flit-core and python-flit
20:44:30 <michel> doesn't seem to be an incompatible update from the changelog though
20:44:36 <tdawson> Ouch ... how backwards compatible are those?
20:44:46 <michel> https://flit.pypa.io/en/stable/history.html
20:44:54 <gotmax> It should be semver compatible, but is it actually necessary?
20:44:57 <michel> would appreciate someone else taking a look too
20:45:38 <michel> gotmax: yeah, it strictly wants flit-core >= 3.7.1 - and epel9 has 3.4.0
20:46:14 <michel> back to the Ansible thing - the feedback from the internal folks is they tried to install 8 and had issues (they're on stream, not on alma/rhel). let me follow up after this
20:47:47 <gotmax> Okay. I wonder if you could relax that pin, but meh. I would do any impact check (https://hackmd.io/@python-maint/rJSm5WC9Y) for the packages that depend on python3-flit-core to make sure it won't introduce incompats/break things
20:48:40 <gotmax> michel: Once you figure out the ansible thing, feel free to file a bug and I'll take a look :)
20:49:29 <michel> gotmax: thanks. I suspect the ansible thing is due to someone mixing and matching our internal snapshots with the latest epel-next, so it's probably on our side :)
20:50:02 <tdawson> Any other Open Floor items?
20:50:05 <michel> nothing more from me
20:50:11 * gotmax has nothing either
20:50:44 <smooge> nothing from me
20:50:53 <smooge> oooh one little tidbit
20:51:05 <smooge> EL6 is still available in COPR to build against
20:51:15 <tdawson> Oh ... interesting
20:51:24 <carlwgeorge> that seems...suboptimal
20:51:28 <smooge> and with that I will go cackling into the night
20:51:33 <tdawson> *laughs*
20:51:37 <gotmax> There was a lot of talk about it on devel@
20:51:49 <gotmax> I guess the LEAPP team still wants it?
20:51:50 <rcallicotte> 6?
20:51:51 <neil> scream test?
20:52:16 <gotmax> They switched it to use RHEL 7 chroots instead of CentOS 7 chroots to fix a package signing problem
20:52:30 <carlwgeorge> neil: https://botsin.space/@scream
20:52:39 <gotmax> The usual ranters were unhappy that we supported old RHEL but not old Fedora
20:52:43 <neil> random: if any of yall are in Vancouver for openinfra... find me ☺
20:53:10 <gotmax> *They switched it to use RHEL 6 chroots instead of CentOS 6 chroots to fix a package signing problem
20:53:59 <neil> carlwgeorge++
20:54:00 <tdawson> Well, since we are end users of COPR, I'm just going to say I'm glad we don't have to deal with that.
20:54:13 <neil> tdawson++
20:54:30 <rcallicotte> tdawson++
20:54:30 <zodbot> rcallicotte: Karma for tdawson changed to 7 (for the release cycle f38):  https://badges.fedoraproject.org/tags/cookie/any
20:54:51 <tdawson> Anything else before we close the meeting?
20:55:02 <smooge> not from me
20:55:07 <neil> not from I. thank you, folks!
20:55:36 <tdawson> Thank you all for the good discussions.  And thank you all for all the work you do on EPEL.  Talk to you next week.
20:55:50 <rcallicotte> bye. thanks
20:55:51 <tdawson> #endmeeting