21:00:30 <tdawson> #startmeeting EPEL (2023-11-08)
21:00:30 <zodbot> Meeting started Wed Nov  8 21:00:30 2023 UTC.
21:00:30 <zodbot> This meeting is logged and archived in a public location.
21:00:30 <zodbot> The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
21:00:30 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
21:00:30 <zodbot> The meeting name has been set to 'epel_(2023-11-08)'
21:00:31 <tdawson> #meetingname epel
21:00:31 <zodbot> The meeting name has been set to 'epel'
21:00:33 <tdawson> #chair nirik tdawson pgreco carlwgeorge salimma dcavalca dherrera gotmax23 smooge
21:00:33 <zodbot> Current chairs: carlwgeorge dcavalca dherrera gotmax23 nirik pgreco salimma smooge tdawson
21:00:34 <tdawson> #topic aloha
21:00:37 <carlwgeorge> .hi
21:00:39 <zodbot> carlwgeorge: carlwgeorge 'Carl George' <carl@redhat.com>
21:00:41 <michel-slm> .hello salimma
21:00:42 <zodbot> michel-slm: salimma 'Michel Lind' <michel@michel-slm.name>
21:00:48 <tdawson> Hi carlwgeorge
21:00:54 <pgreco_> .hi
21:00:55 <tdawson> Hello michel-slm
21:00:56 <zodbot> pgreco_: Sorry, but user 'pgreco_' does not exist
21:00:58 <pgreco> .hi
21:00:59 <tdawson> Hi pgreco
21:01:00 <zodbot> pgreco: pgreco 'Pablo Sebastian Greco' <pablo@fliagreco.com.ar>
21:01:01 <michel-slm> I need to be chaired until we move back to Matrix :)
21:01:04 <tdawson> #chair michel-slm
21:01:05 <zodbot> Current chairs: carlwgeorge dcavalca dherrera gotmax23 michel-slm nirik pgreco salimma smooge tdawson
21:01:11 <pgreco> hey, trying to update my nick...
21:01:32 <nirik> morning
21:01:39 <michel-slm> or actually, since we'll never re-bridge maybe I should just unpair my account and use 'salimma' here next week
21:01:41 <tdawson> Morning nirik
21:01:53 <michel-slm> afternoon y'all
21:02:28 <neil> .hi
21:02:29 <zodbot> neil: neil 'Neil Hanlon' <neil@shrug.pw>
21:02:30 <neil> heya folks
21:02:33 <smooge> .hey
21:02:43 <tdawson> Hi neil
21:02:46 <tdawson> Hey smooge
21:03:03 <Son_Goku> .hello ngompa
21:03:06 <zodbot> Son_Goku: ngompa 'Neal Gompa' <ngompa13@gmail.com>
21:03:13 <dherrera> .hi
21:03:14 <zodbot> dherrera: dherrera 'Diego Herrera' <dherrera@redhat.com>
21:03:17 * Son_Goku is double-stacked with a CentOS meeting too
21:03:30 <tdawson> Hello Son_Goku
21:03:35 <tdawson> Hi dherrera
21:03:55 <dherrera> hi tdawson :)
21:04:01 <tdawson> Son_Goku: Double stacked is ok, We'll take as much of you as we can get.
21:04:05 <smooge> there is a centos meeting?
21:04:20 <Son_Goku> the CentOS Board meeting collides with this one once a month
21:04:55 <smooge> ah so nothing important
21:05:12 <tdawson> #topic End Of Life (EOL)
21:05:13 <tdawson> RHEL 7 / epel-7 will go EOL on 2024-06-30
21:05:14 <tdawson> https://endoflife.date/rhel
21:05:16 <tdawson> CentOS Stream 8 / epel-8-next goes EOL in 2024-05-31
21:05:17 <tdawson> CentOS Stream 9 / epel-9-next goes EOL in 2027-05-31
21:05:19 <tdawson> https://endoflife.date/centos-stream
21:05:41 <Son_Goku> smooge: lol
21:05:50 <tdawson> #topic EPEL Issues https://pagure.io/epel/issues
21:05:51 <tdawson> https://pagure.io/epel/issues?tags=meeting&status=Open
21:06:48 <tdawson> Oh ... there are more topics for the meeting than I expected.
21:07:03 * neil tells himself to keep quiet so we can get through
21:07:23 <tdawson> I'll start with (hopefully) the quick ones.
21:07:31 <tdawson> .epel 247
21:07:32 <zodbot> tdawson: Issue #247: Revisting conflicts policy for EPEL compat packages - epel - Pagure.io - https://pagure.io/epel/issue/247
21:07:59 <tdawson> carlwgeorge: We're just double checking that the pull request is ok, correct?
21:08:09 <carlwgeorge> yup, ready to merge unless there any objections
21:08:22 <tdawson> #link https://pagure.io/epel/pull-request/257
21:09:09 <neil> lgtm; +1
21:09:21 <dherrera> lgtm :)
21:09:25 <smooge> lgtm
21:09:28 <tdawson> We're already gotten several +1's on that.  Does anyone have any negatives?
21:09:35 <carlwgeorge> i count three favorable comments in the pr, not that we have any explicit bar it has to meet.  there just isn't a rush so i figured i'd give people a chance to speak up here at the meeting.
21:10:21 <michel-slm> +1
21:10:31 <tdawson> I personally was very glad for this, it allowed me to get a gpgme compat package into epel9
21:10:51 <pgreco> +1
21:11:29 <carlwgeorge> tdawson: jumping the gun slightly before we merged this?  :P
21:11:51 <tdawson> #info Pull request approved by at least 8 positive votes and no negatives.
21:12:23 <tdawson> carlwgeorge: True, but we had approved the spirit, if not the wording of the proposal.
21:12:52 <carlwgeorge> for sure, not a big deal at all.  especially considering there were other existing conflicts in other packages.
21:13:02 <tdawson> Anything else before we move on?
21:13:07 <carlwgeorge> nope, merging now
21:13:57 <tdawson> .epel 253
21:13:58 <zodbot> tdawson: Issue #253: Move weekly meetings to Matrix - epel - Pagure.io - https://pagure.io/epel/issue/253
21:14:22 <nirik> we can do this, there's some minor things still missing, but if you don't mind that, go for it.
21:14:27 <nirik> it should be usable at least.
21:14:35 <carlwgeorge> see the last comment, i'm not in a rush for this one but yselkowitz asked if we could discuss it again
21:14:41 <carlwgeorge> hence me tagging it
21:14:55 <carlwgeorge> hmm, he asked for it but i don't see him here
21:14:59 <neil> i participated in a meeting this morning using the new bot. seemed fine enough, in-meeting functionality wise
21:15:02 <michel-slm> we might as well dogfood it if there's a fallback plan to switch back here if it fails
21:15:15 <neil> carlwgeorge: he said his availability is tough at this time thru spring
21:15:25 <nirik> There's no mail to meetingminutes list yet, talked with ryan about that yesterday, not sure he has had a chance to figure out the best way forward on that.
21:15:44 <nirik> no chair system, but thats fine for a first cut IMHO
21:16:10 <tdawson> I'm ok without the meetingminutes email.
21:17:09 <tdawson> I'm pretty sure Son_Goku would prefer that we move the meeting to Matrix.
21:17:17 <Son_Goku> yes please
21:17:32 <dherrera> me to :)
21:17:41 <tdawson> Anything else need to be discussed before we vote on moving to Matrix?
21:17:49 <pgreco> I have no preference to move it, but I won't oppose it ;)
21:17:58 <carlwgeorge> i'm in the same boat as pgreco
21:18:18 <michel-slm> I'm in favor of moving too
21:18:19 * nirik is fine either way too, moving it will get more testing. ;)
21:18:33 <tdawson> Sounds good.   Let's vote - +1 = move to matrix  0 = no preference -1 = stay on IRC for now.
21:18:42 <carlwgeorge> 0
21:18:48 <nirik> 0
21:18:50 <dherrera> +1
21:18:53 <neil> 0
21:19:02 <tdawson> Non-committee members should vote too, just so we know how others feel.
21:19:04 <tdawson> +1
21:19:39 * neil has gotten over his fear of voting
21:19:43 <smooge> 0
21:19:56 <neil> even if it's *un*wanted 😈
21:20:03 <smooge> i am not against nor for
21:20:15 <michel-slm> +1
21:21:11 <tdawson> OK, looks like it passed
21:21:43 <nirik> cool
21:21:59 <Son_Goku> +1
21:22:01 <carlwgeorge> which room are we doing it in?
21:22:03 <tdawson> #info Move to Matrix passed Committee: +1(4) 0 (neutral)(2)  no vote 1
21:22:22 <Son_Goku> we can reuse #meeting:fedoraproject.org, no?
21:22:44 <smooge> or do it in epel
21:22:48 <Son_Goku> it has the matrix bots and whatnot
21:22:57 <nirik> please use the meeting rooms
21:23:05 <smooge> then meeting
21:23:11 <Son_Goku> I would prefer to use the equivalent matrix room
21:23:15 <tdawson> I'd prefer we do it in the regular fedora meeting room - https://chat.fedoraproject.org/#/room/#meeting:fedoraproject.org
21:23:35 * nirik is pretty against doing meetings in team rooms, but I know some people want to do that... but it shouldn't be default at least
21:23:35 <Son_Goku> we should probably get all the rooms upgraded to clean out the IRC puppets and stuff
21:24:24 <Son_Goku> (or whatever is needed to clean out the IRC stuff)
21:25:05 <michel-slm> does anyone have a quick guide to unlink one's Matrix account from IRC bridging?
21:25:19 <tdawson> I've updated the ticket.  I'll send out an email and update the Fedora Calendar.
21:25:21 <michel-slm> e.g. the opposite of kparal's excellent writeup for enabling bridging, auto-identing etc
21:26:03 <tdawson> Anything else before we move on?
21:26:08 <nirik> you don't have to do anything, the bridge is not there.
21:26:37 <carlwgeorge> element is on the struggle bus trying to join that room
21:27:16 <michel-slm> nirik: oh nice
21:27:26 <nirik> I can upgrade em... thats kinda distruptive, but sure... would clean up old stuff
21:27:39 <Son_Goku> we might as well
21:27:48 <michel-slm> upgrading won't delist the room from Fedora Space right?
21:27:51 <Son_Goku> clearing out the appservice and dead puppets will speed the rooms up a lot
21:28:08 <Son_Goku> michel-slm: the listing is based on the published address, iirc, so it should be fine
21:28:12 <nirik> it makes a new room, moves any aliases and perms and drops a 'this room is gone, go here' tombstone in the old room
21:28:27 <neil> the hyperscale upgrade went great for me in my matrix homeserver
21:28:47 <carlwgeorge> i saw that in action earlier today for the centos hyperscale room
21:28:49 <neil> fedora support channel not so much, but i know there were issues w/ it
21:28:53 <tdawson> We have two other issues, so I'm going to move on.
21:28:57 <neil> tdawson++
21:29:00 <tdawson> .epel 256
21:29:01 <zodbot> tdawson: Issue #256: Older epel9-next build would shadow newer epel9 build when used as dependency - epel - Pagure.io - https://pagure.io/epel/issue/256
21:29:22 <Son_Goku> whoa
21:29:25 <Son_Goku> that's a weird effect
21:30:00 <michel-slm> yeah. basically for buildroot if there's an epel9-next build the epel9 one won't be seen at all
21:30:10 <neil> weird
21:30:15 <nirik> right, koji only knows last tagged thing.
21:30:18 <michel-slm> it .. sort of makes sense but everyone who encountered it find it unintuitive
21:30:26 <tdawson> I didn't get the pull request done ... and I left this tagged as "meeting" ... I just wanted to double check ... are we ok with this just being a "documentation thing" ?
21:30:31 <michel-slm> oh.. it's about who tags last, not about one repo always winning?
21:30:48 <neil> maybe documentation + backlog ticket to run periodic check?
21:31:04 <nirik> right. If there's nothing tagged in epel9-next it uses epel9 version. if there's anything at all tagged in epel9-next, it uses that because that tag is higher in the inheritence
21:31:06 <michel-slm> yeah, I think it's fine if it's documented, especially if we document that every new minor release cycle epel9-next gets purged
21:31:17 <pgreco> Need to drop,  bye guys!
21:31:23 <nirik> bye pgreco
21:31:24 <tdawson> pgreco: Bye
21:31:25 <neil> peace pgreco
21:31:34 <michel-slm> nirik: ah ok, so even if there's a later epel9 build, any epel9-next build will win right? that's my initial understanding
21:31:36 <michel-slm> bye pgreco !
21:31:43 <tdawson> michel-slm: OK, I was planning on putting that in.
21:31:51 <nirik> yep. any tagged build in epel9-next will win over any epel9 build.
21:31:59 <nirik> koji has no idea about 'versions'
21:32:25 <tdawson> michel-slm: As the person who usually purges them ... there is alot of rust stuff in there and sometimes I'm nervous about that.  Can I send you a list for a "go ahead" ?
21:33:09 <michel-slm> tdawson: best post in the Rust Matrix channel if you want to ask, since I think decathorpe would know better
21:34:07 <tdawson> michel-slm: Sounds good.
21:34:11 <michel-slm> but in general, yes, it should be fine. we want these packages purged, a heads up will be nice so people know they shouldn't use epel9-next once the cleanup happens
21:34:16 <michel-slm> thanks!
21:35:23 <tdawson> OK.  I do seperate them into "mine" and "everyone else" ... I'll do what I can to send out emails so people know what's getting purged.
21:36:00 <tdawson> I'll try to get the pull request ready for next week.
21:36:11 <tdawson> Any ... now moving on to what I think is going to take the longest time.
21:36:21 <smooge> repotags?
21:36:27 <tdawson> *laughs*
21:36:50 <tdawson> Sorry ... but ... I've been working on repotags all week ... that was quite funny to me.
21:36:57 <smooge> keeping el7 open for another 4 years?
21:36:58 <tdawson> .epel 255
21:36:59 <zodbot> tdawson: Issue #255: redhat-lsb in EPEL - epel - Pagure.io - https://pagure.io/epel/issue/255
21:37:12 <smooge> ahhhh
21:37:14 <tdawson> Close ... but not quite.
21:37:27 <carlwgeorge> i tagged this one just to give an update on the fesco ticket, and ask for one small thing
21:37:50 <smooge> sanity?
21:37:57 <carlwgeorge> lol
21:38:06 <tdawson> carlwgeorge: Go for it.
21:39:05 <carlwgeorge> pretty much all of fesco agrees it should be retired from fedora, but i think two members don't want to override the maintainer who wants to keep it.  i'm not sure how that will play out in the end.  one of those -1 members did say he's fine with epsco blocking it from epel.
21:39:23 <carlwgeorge> > If EPEL Steering Committee wants to reject the package from EPEL, that's up to them.
21:39:32 <carlwgeorge> #link https://pagure.io/fesco/issue/3089#comment-882014
21:40:36 <carlwgeorge> i think it will be discussed again at the next fesco meeting, with hopes that the maintainer will show up to discuss.
21:41:40 <michel-slm> we don't really have solid policies around this huh. e.g. what happens if the maintainer can't or won't show up
21:41:59 <Son_Goku> I don't know
21:42:01 <carlwgeorge> my ask here is that we unpush the update from epel9-testing.  if we decide later to allow it in epel9 a new update can easily be created.
21:42:17 <Son_Goku> +1 with that proposal
21:42:33 <michel-slm> that seems the minimal change we can make, +1
21:42:35 <nirik> it's not possible to go stable is it? I guess thats what you are trying to avoid?
21:42:36 <carlwgeorge> even with the current negative karma the update has met the time requirement and could be manually pushed by the maintainer, which i'd like to avoid
21:42:43 <nirik> yeah.
21:42:52 * michel-slm would also agree with retiring it but maybe we should only do that after it's retired from Rawhide
21:43:23 * smooge thinks unpushing it is a good idea
21:43:24 <nirik> two more -1's will unpush it. ;)
21:43:40 <nirik> https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2023-336dbb57e0
21:43:41 <smooge> ah.. where is it?
21:43:44 <carlwgeorge> fwiw, Son_Goku did a similar action to prevent it going to f39 stable prematurely
21:44:00 <michel-slm> +1 here, -1 there... hard to keep track :)
21:44:14 <Son_Goku> https://bodhi.fedoraproject.org/updates/FEDORA-2023-8e90030e60
21:44:20 <Son_Goku> this was the update I unpushed
21:44:57 <michel-slm> oh, it's a new epel9 package, I see. yeah... just -1-ed it
21:45:25 <michel-slm> need one more -1
21:45:27 <tdawson> I'm torn on this.  On the one hand, I don't like the committee being against a package that legally can be in epel.  On the other hand, I also feel like sgallagh in this comment - https://pagure.io/fesco/issue/3089#comment-882250  - the redhat-lsb is supposed to represent the distribution.
21:45:55 <carlwgeorge> yup, sgallagh is spot on
21:46:03 <michel-slm> I mean.. if the package actually works, sure, but I agree with most of FESCo that it's... misrepresenting compliance
21:46:07 <Son_Goku> this is one of those weird edge cases
21:46:16 <smooge> i would be ok with epel-lsb or distro-lsb or something
21:46:16 <Son_Goku> and also, the package effectively doesn't work
21:46:35 <nirik> it would need to mean something else...
21:46:36 <Son_Goku> carlwgeorge documented fairly extensively the breakages
21:46:56 <carlwgeorge> smooge: yeah the redhat- in the name is a weird artifact of history i'd also like to get rid of, but without an active upstream it's not worth renaming
21:47:08 <tdawson> I personally have not liked "LSB" since it's inception.
21:47:22 <tdawson> I really don't want an epel-lsb
21:47:30 <nirik> they annoyed a lot of people by requiring rpm. :)
21:47:39 <Son_Goku> that isn't what bothers me about it :P
21:47:53 <carlwgeorge> reminder that we got in this situation because the former red hat maintainers orphaned it, and the current maintainer snatched it up.  that would have been a good opportunity to just create a new lsb package that would go through package review.
21:48:30 <carlwgeorge> ideally it would have been blocked on the "functions as advertised" check
21:48:50 <Son_Goku> this was also a bit of an X/Y
21:49:04 <Son_Goku> redhat-lsb is unusual among lsb implementations in that lsb_release was never a separate subpackage
21:49:18 <carlwgeorge> for those unaware https://xyproblem.info/
21:49:35 <Son_Goku> most distributions that did have lsb conformance also made lsb-release a distinct package that could be installed separately from lsb modules
21:50:04 <Son_Goku> but we never did that, so I suspect it never occurred to anyone to just... do that in response to all the bug reports requesting the tool to exist
21:50:57 <Son_Goku> everyone focused on having redhat-lsb when the actual solution was just to ship lsb_release(1) itself
21:51:02 * michel-slm rather regrets the situation where so many packages suddenly get abandoned this cycle :(
21:51:08 <smooge> that said, I can understand some of the needs.. a friend of mine who is a 'no one else knows linux so I got stuck' sysadmin was complaining that he couldn't install the TLA company name software he needed on EL9 because it didn't have redhat-lsb which was required
21:51:13 <carlwgeorge> so is everyone fine with me just clicking unpush on this, then we can move on to open floor before we run out of time
21:51:39 <michel-slm> sure
21:51:42 <Son_Goku> yup
21:51:45 <Son_Goku> go for it
21:51:46 <tdawson> I'm going to state no-preference
21:51:56 <michel-slm> well, I voted down the package so it's close to being unpushed anyway :)
21:52:17 <carlwgeorge> to rephrase, no one is opposed to me unpushing it, at least for now, right?
21:52:30 <tdawson> Correct
21:52:34 <Son_Goku> yup
21:52:42 <carlwgeorge> will do and will link to the meeting minutes.  let's move one.
21:52:44 <carlwgeorge> *on
21:53:03 <tdawson> I'd like to hear how this turns out in Fesco, so I'm not opposed to having this stay with the meeting tag.
21:53:16 <tdawson> but ya ... we're low on time
21:53:24 <tdawson> #topic General Issues / Open Floor
21:53:29 <michel-slm> yeah we should keep the meeting tag
21:53:31 * carlwgeorge 's element browser tab is still spinning on joining the meeting room
21:53:34 <tdawson> Does anyone have anything for Open Floor ?
21:53:42 <carlwgeorge> yup
21:53:49 <tdawson> carlwgeorge: Go for it.
21:54:12 <carlwgeorge> we said we were gonna piggy back on fedora elections, for which nominations start...today
21:54:17 <carlwgeorge> #link https://fedorapeople.org/groups/schedule/f-39/f-39-elections-tasks.html
21:54:32 <smooge> carlwgeorge: you are lucky that you could even see it. when I went looking for various channels on fedora.im element never lists them
21:54:52 <Son_Goku> smooge: it's listed on fedoraproject.org, not fedora.im
21:55:00 <tdawson> I thought elections were in the spring.
21:55:02 <smooge> ok so how many seats are open and where do we throw hats?
21:55:07 <carlwgeorge> smooge: i actually couldn't see it, i edited my url to get to it.  preview loaded, but that's it so far.
21:55:18 <michel-slm> smooge: you throw it in the ring :)
21:55:34 <carlwgeorge> tdawson: did we say the spring elections?  that's fine if so, i just had it in my head "next election"
21:55:44 <nirik> carlwgeorge: I see you as having joined... ;(
21:56:10 <carlwgeorge> i want to believe in matrix, but element just has so many rough edges
21:56:16 <tdawson> We might have said both/either.  But I know spring came up at least once, and that's what I had in my head.
21:57:08 <tdawson> https://docs.fedoraproject.org/en-US/epel/epel-policy-steering-committee/#committee_member_selection
21:57:26 <michel-slm> yeah. I'm hoping the new protocol they are trialing in Element X will improve things, but Element X is also unusable right now
21:57:28 <tdawson> The documentation says 2024 election.
21:58:05 <smooge> ok will pick up hat and wait til Spring release of Fedora 40
21:58:07 <carlwgeorge> works for me.  but that just kicks the can down the road, we still need someone to figure out how we get folks on the ballot.
21:58:09 <tdawson> When I wrote that, I thought there was just one election a year, in the spring ... but I don't say which one.
21:58:16 <nirik> element x isn't really a new protocol... but yeah, hopefully things get better.
21:58:28 <michel-slm> it has the sliding sync thing right
21:58:47 <michel-slm> not new external protocol, just new sync protocol with the server
21:58:56 <nirik> yes, which is implented as a proxy in front of a matrix server. ;)
21:59:09 <carlwgeorge> i tried element x, and it is certainly faster, but has so many missing features, like no unread indicators.  even @ mentioning people doesn't work.
21:59:19 <smooge> personally I just would like to be able to list available channels and see what other people see. Why is searching for CentOS only showing the Red Hat CPE channel and CentOS Alt Images.. what hyperscale channel?
21:59:51 <smooge> anyway.. I think tdawson is looking ofr a suggestion to clarify what election to do epel elections
22:00:13 <carlwgeorge> our first chance in 2024 is f40 elections, which is from 2024-04-24 to 2024-05-31
22:00:19 <nirik> if you search it defaults to your same homeserver (fedora.im or whatever). You can switch it to fedoraproject.org or matrix.org or whatever you want
22:00:21 <tdawson> Are we all ok waiting for the Spring election?   Or does someone really want to do the one right now?
22:00:44 <nirik> most of the fedora rooms should also have a fedora.im alias/address
22:00:59 <carlwgeorge> i've got no problem waiting, i was just worried about having to rush to get in on this one if that was the plan
22:01:11 <carlwgeorge> pushing back to the spring makes the most sense
22:01:33 <tdawson> OK, let's wait till spring, that will get us time to learn how to get on the elections.
22:01:50 * carlwgeorge sets calendar reminder
22:02:02 <smooge> Elections will occur during the first Fedora election cycle of a calendar year
22:02:32 <smooge> nirik: ah I didn't know that fedoraproject.org had different channels than fedora.im
22:02:47 <tdawson> Looks like our time is up.
22:03:09 <smooge> tdawson: does my "Elections will" look good for the doc?
22:03:09 <tdawson> Thank you all for the good discussions.  And thank you all for all the work you do for EPEL and it's community.
22:03:23 <neil> thank you tdawson!
22:03:25 <carlwgeorge> smooge: lgtm
22:03:27 <nirik> well, rooms can have addresses on multiple servers. Or none (then you have to be invited or know the roomid uuid)
22:03:32 <smooge> i was going to say 'Spring' but that is only northern hemisphere :)
22:03:34 <tdawson> smooge: Yes ... I copied it to do a pull request, but you can if you want.
22:03:49 * nirik is happy to talk more matrix over on... er... matrix?
22:03:56 <tdawson> #endmeeting