20:00:09 <tdawson> #startmeeting EPEL (2022-10-19)
20:00:09 <zodbot> Meeting started Wed Oct 19 20:00:09 2022 UTC.
20:00:09 <zodbot> This meeting is logged and archived in a public location.
20:00:09 <zodbot> The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
20:00:09 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:09 <zodbot> The meeting name has been set to 'epel_(2022-10-19)'
20:00:11 <tdawson> #meetingname epel
20:00:11 <zodbot> The meeting name has been set to 'epel'
20:00:12 <tdawson> #chair nirik tdawson pgreco carlwgeorge salimma dcavalca dherrera gotmax[m] smooge
20:00:12 <zodbot> Current chairs: carlwgeorge dcavalca dherrera gotmax[m] nirik pgreco salimma smooge tdawson
20:00:14 <tdawson> #topic aloha
20:00:22 <pgreco> .hi
20:00:22 <zodbot> pgreco: pgreco 'Pablo Sebastian Greco' <pablo@fliagreco.com.ar>
20:00:24 <carlwgeorge> .hi
20:00:26 <zodbot> carlwgeorge: carlwgeorge 'Carl George' <carl@redhat.com>
20:00:32 <tdawson> Hi pgreco and carlwgeorge
20:00:42 <davide> .hello dcavalca
20:00:42 <gotmax[m]> .hello gotmax23
20:00:42 <zodbot> davide: dcavalca 'Davide Cavalca' <dcavalca@fb.com>
20:00:45 <zodbot> gotmax[m]: gotmax23 'Maxwell G' <gotmax@e.email>
20:00:57 <tdawson> Hello davide and gotmax[m]
20:01:08 <rsc> .hello robert
20:01:09 <zodbot> rsc: robert 'Robert Scheck' <redhat@linuxnetz.de>
20:01:16 <tdawson> Hello rsc
20:01:40 <dherrera> .hi
20:01:43 <zodbot> dherrera: dherrera 'Diego Herrera' <dherrera@redhat.com>
20:01:47 <tdawson> Hi dherrera
20:03:19 <tdawson> I believe nirik[m] is on vacation, so I don't expect him here.
20:03:27 <Ebeneezer_Smooge> gwew
20:03:30 <Ebeneezer_Smooge> here
20:03:46 <tdawson> Hello Ebeneezer_Smooge
20:05:03 <tdawson> Now as many as last week, but looks like majority of us are here
20:05:13 <tdawson> #topic End Of Life (EOL)
20:05:14 <tdawson> RHEL 7 will go EOL on 2024-06-30
20:05:16 <tdawson> CentOS Stream 8 goes EOL in 2024-05-31
20:05:17 <tdawson> CentOS Stream 9 goes EOL in 2027-05-31
20:05:48 <tdawson> #topic EPEL Issues https://pagure.io/epel/issues
20:05:50 <tdawson> https://pagure.io/epel/issues?tags=meeting&status=Open
20:06:39 <tdawson> So ... do we still need modularity  (#198) listed as a meeting topic?  Do we have anything left to discuss about it?
20:07:04 <Ebeneezer_Smooge> only thing I was going to mention was I read the preview and it looked good
20:07:25 <tdawson> Cool.  Thanks.
20:07:31 <Ebeneezer_Smooge> updated and cleaned
20:07:36 <Ebeneezer_Smooge> the ticket that is
20:07:55 <tdawson> Thank you
20:07:59 * carlwgeorge sets reminder to push epel-release update to stable on halloween
20:08:21 <tdawson> Good idea, and thanks carlwgeorge
20:09:51 <tdawson> Of the other 4 ... #200 (Investigate getting EPEL packagers acces to BR repo) and #39 ... both are sortof stalled if I remember.
20:10:51 <tdawson> That leaves #203
20:11:00 <tdawson> .epel 203
20:11:01 <zodbot> tdawson: Issue #203: Automating adding epel-packagers-sig to packages - epel - Pagure.io - https://pagure.io/epel/issue/203
20:12:13 <tdawson> The summary that carlwgeorge wrote in the last comment was pretty good. - https://pagure.io/epel/issue/203#comment-821144
20:12:40 <carlwgeorge> anyone opposed to changing the title of the issue to more accurately reflect where we're at with it?
20:13:16 <tdawson> That's fine with me.
20:13:45 <Ebeneezer_Smooge> +1
20:14:45 <tdawson> I'm not really sure where to go from here.  We've talked about the different things we could do ...
20:14:47 <gotmax[m]> Sure
20:14:53 <carlwgeorge> there was some discussion around the auto-branching being based on eln extras workloads versus the existence of epel9 branches/builds.
20:15:13 <carlwgeorge> one is forward looking, one is backwards looking (not necessarily a negative)
20:15:35 <Ebeneezer_Smooge> I would like to say go with eln extras
20:15:38 <carlwgeorge> i don't think many maintainers actually look at eln or configure any workloads
20:15:46 <Ebeneezer_Smooge> and put all the things which are in EPEL in that
20:16:26 <Ebeneezer_Smooge> I think the only people who look at ELN are people who work on ELN. That said, getting things working or attempted to working is good now.
20:16:28 <tdawson> The one thing I like about the eln-extras is that it would mean that the maintainer has already thought about what they want.
20:16:43 <carlwgeorge> i'm leaning towards both, if there was epel9 builds or an eln extras workload, then autocreate the epel10 branch
20:17:13 <carlwgeorge> limited it to just things in eln extras would severely limit the impact
20:17:16 <gotmax[m]> carlwgeorge: Seems reasonable to me
20:17:34 <tdawson> One thing to note is that putting all of epel in one workgroup would be bad ... if there are conflicts and particularly bad dependency problems, the whole workgroup will get dropped.
20:18:02 <carlwgeorge> and the drawback of creating a branch that is later retired or never used is trivial imo, so i'd defer to making more branches than fewer
20:18:37 <tdawson> I'm not against the eln stuff, just noting that putting everything in one workload can backfire.
20:19:26 <tdawson> I do like having both, like carlwgeorge said ... that sounds good to me.
20:19:50 <gotmax[m]> I might be thinking too far ahead, but I think we should also notify the affected maintainers and allow them to opt out
20:20:18 <gotmax[m]> And then just auto branch those who sad yes or didn't respond
20:21:33 <tdawson> That sounds reasonable.  We have enough time we should be able to send notifications.
20:21:38 <Ebeneezer_Smooge> tdawson, are you meaning like having eln-extras-kde eln-extras-mailprogs eln-extras-ansible ?
20:21:53 <tdawson> Ebeneezer_Smooge: Correct.
20:22:32 <carlwgeorge> i was asked this week about creating an eln-extras-rhel-packager workload with centpkg and openvpn, so rhel maintainers can dogfood centos 10 as early as possible
20:22:41 <tdawson> It's fine if they are all eln, that's great.  But when my eln-extras-kde has a wierd failure (like one of the packages was renamed), then all my kde stuff drops out.
20:23:13 <tdawson> So having a eln-extras-all-epel  ... that would be bad.
20:24:30 <tdawson> carlwgeorge: That would be good.
20:25:35 <carlwgeorge> so combo approach seems to be fine with everyone.  i guess we can put this on the backburner until someone is ready to actually write the code for this logic.  we'd probably want to look at how fedora does mass branching.
20:26:16 <tdawson> I'm pretty sure mohan has a script.
20:27:15 <tdawson> I think we need to make a tag "back-burner" or something like that, and mark this with it.
20:27:32 <Ebeneezer_Smooge> will go make that tag
20:27:35 <gotmax[m]> I could probably help with getting a list of packages and notifying maintainers, but one of the overlords would need to do the actual branching
20:27:55 <carlwgeorge> yeah it would need to be run by releng i imagine
20:29:02 <Ebeneezer_Smooge> back-burner tag should be there on reload
20:29:33 <tdawson> Cool, thanks
20:29:52 <tdawson> Anything else we want to discuss on this before we move on?
20:30:34 <carlwgeorge> on a related note, https://bugzilla.redhat.com/show_bug.cgi?id=2005139 needs to get resolved before epel10
20:30:38 * gotmax[m] has nothing
20:30:49 <davide> This all sounds good to me
20:31:09 <carlwgeorge> the latest-CentOS-Stream symlink we use to valid if an epel branch can be created doesn't have a major version
20:31:13 <tdawson> carlwgeorge: Do you mind putting a summary on the issu again?
20:31:35 <carlwgeorge> tdawson: sure
20:32:16 <carlwgeorge> in fact let me rephrase, that symlink issue needs to be resolved well before epel10, it needs to be resolved before the first cs10 "production" compose
20:32:30 <tdawson> carlwgeorge: That bug will probruably fall to me, or I'll at least have a hand in it.
20:32:32 <carlwgeorge> because it will start giving us the wrong results for epel9 branch requests
20:32:45 <carlwgeorge> both fedpkg and fedscm-admin use it
20:33:38 <tdawson> Actually ... it's probrubly very urgent ... because CentOS Stream 8 is moving to that infrastructure ... so it needs to get fixed sooner.
20:34:23 <carlwgeorge> currently epel8 branch requests use a json file of rhel8 content, the check using centos metadata is only for >=9
20:34:37 <carlwgeorge> but i don't disagree that fixing it sooner than later is desirable
20:34:50 <tdawson> Oh, I wasn't thinking in urgent for epel, I meant the centos stream team.
20:34:56 <carlwgeorge> https://pagure.io/fedpkg/blob/master/f/fedpkg/utils.py#_389
20:35:01 * carlwgeorge nods
20:35:51 <tdawson> OK, moving on ...
20:36:39 <tdawson> .epel 199
20:36:40 <zodbot> tdawson: Issue #199: ensure EPEL bugzilla assignees point to valid packagers - epel - Pagure.io - https://pagure.io/epel/issue/199
20:37:13 <tdawson> This one just sorta puttered out.  Did we have any action items for this and/or more discussion that we needed?
20:38:19 <tdawson> Although I have an idea on how to script this, I currently have too much on my plate(s)
20:40:01 <tdawson> I believe someone commented that this would be something that Fedora in general needs, not just EPEL.  last time we talked about it
20:40:32 <tdawson> So ... since we're all so talkative about it ... I'm fine with setting this one on the back-burner as well.
20:41:52 <tdawson> Moving on ...
20:42:05 <tdawson> I don't have anything for Old Business, so moving to open floor.
20:42:08 <gotmax[m]> Fine with me. I agree that this should not be an EPEL only effort
20:42:16 <gotmax[m]> (I think I said that in the first place :)
20:42:28 <tdawson> #topic General Issues / Open Floor
20:42:39 <tdawson> gotmax[m]: Ah ... yep, there it is in your comment.
20:43:18 <davide> My only thing is that I got an asterisk build working
20:43:27 <davide> Going to put up an update soon after some testing
20:43:51 <davide> One of the subpackages needs festival, so I'll need to get that branched too
20:43:54 <tdawson> Cool
20:44:04 <carlwgeorge> i proactively retired a few uninstallable packages that had bugs filed and ignored for multiple months
20:44:14 <davide> (oddly, it has an old c8 build but doesn't seem to be actually shipped)
20:44:34 <gotmax[m]> Last cycle, python-mock and python-nose were requested 10 times despite being deprecated in Fedora. I'd like to avoid repeating this with python-toml and python-tomli (I'm the EPEL maintainer for the latter).
20:45:01 <gotmax[m]> This is probably too early, but this has been on my mind :)
20:45:16 <dherrera> I noticed that last week opencv was brought to epel9 :D
20:45:22 <tdawson> gotmax[m]: Are they retired, or deprecated in Fedora yet?
20:45:30 <tdawson> dherrera: Ya!!   yes, I noticed that .
20:45:39 <Ebeneezer_Smooge> davide, festival was originally in the c8 buildroot as a build dependency of emacs.
20:45:58 <salimma> .hi
20:45:59 <zodbot> salimma: salimma 'Michel Alexandre Salim' <michel@michel-slm.name>
20:46:06 <davide> ...I have many questions...
20:46:10 <Ebeneezer_Smooge> it was cleaned up as best as possible I think by
20:46:11 <salimma> sorry I'm super late, had an emergency doctor's appointment
20:46:26 <tdawson> Hi salimma ... hope you are ok.
20:46:45 <gotmax[m]> There's https://fedoraproject.org/wiki/Changes/DeprecatePythonToml for python-toml. python-tomli is not being deprecated formally, but it is now part of the stdlib, and I agree with the python-maint people that it shouldn't be branched.
20:46:57 <davide> Ok, in that case I'll get festival branched normally
20:47:12 <salimma> tdawson: thanks, it's our toddler. he's ok but he'll be away from daycare long enough to need a doctor's note
20:47:31 <gotmax[m]> Also, there's a high severity ACE vulnerability in ipython that wasn't patched in EPEL 7 or 8 for 10 months :(
20:47:40 <gotmax[m]> https://src.fedoraproject.org/rpms/ipython/pull-request/45 is for EPEL 8
20:47:46 <tdawson> salimma: Good to hear it's nothing too serious.
20:48:06 <gotmax[m]> EPEL 7 will need to go through the incompat update policy or have the fix backported
20:48:15 <salimma> gotmax (He/Him): yeah, EPEL CVEs are worrying
20:48:18 <gotmax[m]> I'm not the maintainer, but I noticed and figured I'd point it out
20:48:39 <salimma> (I just fixed one in django3 but I don't even want to look at what's affecting python-django 2.x there)
20:49:46 <gotmax[m]> Michel Alexandre Salim 🎩, epel-packagers-sig, and python-sig are maintainers of ipython
20:50:06 <gotmax[m]> Does anyone from those SIGs want to merge the EPEL 8 PR or should I do it as a PP?
20:50:26 <gotmax[m]> * and python-packagers-sig are
20:51:10 <salimma> scratch build looks fine, I'd say it's fine
20:51:30 <salimma> I can merge it, then can you build an update and we can test it more thoroughly in Bodhi?
20:52:14 <salimma> merged
20:52:28 <gotmax[m]> You wanted me to build it?
20:54:33 <tdawson> Looks like our time is closing up, and the conversation is winding down.
20:55:07 <tdawson> Thank you all for coming this week, and for the good discussions we had.
20:55:21 <tdawson> I'll talk to you all next week, if not sooner.
20:55:34 <tdawson> #endmeeting