18:02:03 <smooge> #startmeeting EPEL (2019-10-16)
18:02:03 <zodbot> Meeting started Wed Oct 16 18:02:03 2019 UTC.
18:02:03 <zodbot> This meeting is logged and archived in a public location.
18:02:03 <zodbot> The chair is smooge. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:02:03 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
18:02:03 <zodbot> The meeting name has been set to 'epel_(2019-10-16)'
18:02:03 <smooge> #meetingname epel
18:02:03 <zodbot> The meeting name has been set to 'epel'
18:02:03 <smooge> #topic aloha
18:02:03 <smooge> #chair nirik smooge tdawson bstinson Evolution pgreco sgallagh
18:02:03 <zodbot> Current chairs: Evolution bstinson nirik pgreco sgallagh smooge tdawson
18:02:03 <smooge> #topic Intros
18:02:16 <sgallagh> .hello2
18:02:17 <zodbot> sgallagh: sgallagh 'Stephen Gallagher' <sgallagh@redhat.com>
18:02:17 * tdawson is here
18:02:20 <nirik> morning...
18:02:23 <bstinson> hi all
18:02:46 <smooge> #info Meeting is run from https://board.net/p/epel
18:02:59 <pgreco> hi everybody
18:03:10 <smooge> smooge is sick
18:03:50 <smooge> #topic Old Business
18:04:38 <smooge> Is there any outstanding old business? I didn't have anything. For people wanting to add items to this meeting, we are going to be using  https://board.net/p/epel for agendas and such.
18:06:32 <bstinson> i can't think of anything
18:06:54 <smooge> #topic EPEL-6
18:07:09 <smooge> #info EPEL-6 is End of Life  in 2020-11
18:07:47 <smooge> #info EPEL-6 will be moved to archives in 2020-12
18:09:13 <smooge> #info EPEL-6  has about 620,000 systems still on it
18:09:16 <smooge> #topic EPEL-7
18:09:38 <smooge> so I think python36 got dealt with last week
18:09:53 <smooge> and we are now just working through some issues in both el6 and el7 with it
18:10:32 <smooge> the el6 are some people assuming that python34 provides python3 items when it provides python34 items
18:10:52 <smooge> I think tdawson was doing some work on testing what packages in EL7 install?
18:11:00 <tdawson> Correct
18:11:24 <tdawson> The only python related "wont installs" were because packages were retired, nothing because of the overall python36 issue.
18:11:38 <nirik> python34 in epel6 is not intended to be a user stack. It's only there for testing things against. (IIRC)
18:12:04 <tdawson> We only had about 20 real "won't install" issues.  Bugs have been filed, about half are being delt with.
18:12:39 <smooge> nirik, I think so. I also expect it is probably being used in production a lot
18:12:46 <tdawson> So, next week at some time, I'll start working on any that haven't been dealt with.
18:12:53 <smooge> tdawson, thanks!
18:13:19 <nirik> smooge: too bad for such people.
18:13:21 <carlwgeorge> nirik: are you talking about the "flatpackage" thing in fedora, where there are no subpackages?  only fedora does that, epel doesn't.
18:13:44 <nirik> no?
18:14:15 <carlwgeorge> i haven't heard anyone refer to python34 as "not intended for users"
18:14:28 <nirik> https://apps.fedoraproject.org/packages/python34/overview/
18:14:43 <nirik> "This package exists to allow developers to test their code against an older
18:14:43 <nirik> version of Python. This is not a full Python stack and if you wish to run
18:14:43 <nirik> your applications with Python 3.4, see other distributions
18:14:43 <nirik> that support it, such as CentOS or RHEL with Software Collection"
18:14:54 <nirik> same for python35 python36, etc.
18:15:07 <carlwgeorge> that's pulled from the description in master
18:15:21 <nirik> ok. so thats not the intent in epel?
18:15:23 <carlwgeorge> if you check the description in epel6 it doesn't say that
18:15:29 <nirik> it definitely is in the other fedora branches
18:15:45 <carlwgeorge> not that i've heard of, and i know lots of people rely on python34 in el6
18:15:48 <nirik> for epel7... python34 was intended to be the full stack back in the day
18:16:03 <nirik> huh, ok...
18:16:13 <nirik> sorry to hear it. And I could definitely be wrong.
18:20:42 <smooge> ok next up
18:20:49 <smooge> #topic EPEL-8
18:21:04 <smooge> oh one thing..
18:21:47 <smooge> #info EPEL-7 has about 1.2 million users and 200,000 of them seem to have yum -y update set as a daily cron job
18:22:19 <smooge> we pushed out a new epel-release last week and about 200k systems started using the new format within 36 hours it seemed
18:23:12 <nirik> fun
18:23:19 <tdawson> Is that a good thing?  bad thing?  Or just an information thing?
18:23:26 <smooge> just an information thing
18:23:31 <tdawson> Cool
18:23:38 <sgallagh> EPEL 8 is currently blocked on https://pagure.io/releng/issue/8865
18:23:54 <smooge> EPEL-8 or EPEL-8.1?
18:24:14 <sgallagh> And on me being able to disengage from the ongoing Fedora Modularity discussions long enough to re-review the status of grobisplitter and how we're doing buildroots in EPEL 8
18:24:17 <sgallagh> smooge: EPEL 8.1, sorry
18:24:25 <smooge> ok cool
18:24:42 <smooge> was worried for a minute
18:25:08 <smooge> I mean I am worried.. but that was a 'aaaaagh' versus a 'agh' type worry
18:25:34 <sgallagh> ...
18:25:35 <pgreco> I've been mostly offline last week, any packages that the epel-7 owner doesn't want to pick up?
18:26:13 <nirik> hard to say...
18:27:13 <nirik> it might be nice to have some way to track things, but not sure how would work.
18:27:38 <smooge> sgallagh, I can help you on the epel-8 grobisplitter tomorrow or Friday
18:27:47 <sgallagh> smooge: Yes, please
18:28:46 <smooge> so does anyone have a suggestion for a better way to track things?
18:29:06 <pgreco> smooge, I'm ok with the bugzilla/branch flow
18:29:33 <pgreco> I'm asking about the cases where the actual owner of the packages doesn't want to maintain epel-8 also
18:29:52 <pgreco> so we can look for alternative maintainers
18:30:09 <smooge> most of the packages I have run into are probably ones with no active maintainer
18:30:13 <smooge> puppet is one
18:30:18 <pgreco> I noticed this because znc was updated a few days ago
18:30:27 <pgreco> for epel7, and there is no epel8 branch
18:30:32 <smooge> ah yeah..
18:30:36 <pgreco> and found strange that nobody asked for it
18:30:42 <smooge> I thought i did
18:30:51 <smooge> but again we aren't tracking things clearly
18:30:54 <nirik> well, the problem with the bug / branch flow is...
18:31:00 <nirik> we have no way to find all the bugs
18:31:02 <nirik> do we?
18:31:26 <nirik> and when people mail maintainers directly, we have less than no way to know whats been requested.
18:31:59 <smooge> yeah agreed
18:32:13 <smooge> sorry I need to step away and get some more tissues
18:32:27 <nirik> we could try some searches and get a lot perhaps... subject has epel8 and request in it? but dunno if that would be helpfull.
18:32:51 <tdawson> I think it would grab a bunch.
18:33:47 <nirik> 112 from a quick query
18:34:04 <nirik> bugzilla query -s NEW -t 'epel8'
18:35:25 <smooge> cool I didn't know that command
18:35:53 <tdawson> So, what would we do if we look through those and find some where the maintainer hasn't replied for "some amount of time"?
18:35:55 <smooge> why yes I worked with this place for 24 years
18:36:12 <nirik> bugzilla query -s NEW -t 'epel8'  --outputformat "%{id}: %{component}: %{summary}"
18:36:12 <tdawson> Put them on a tracker bug?
18:36:23 <smooge> yeah. I will do this tomorrow
18:36:34 <nirik> well, in epel7 I think we gave maintainers 2 weeks?
18:36:44 <smooge> I believe so
18:36:54 <nirik> we could do similar or more or less...
18:37:14 <smooge> I have some people who have been interested in becoming 'packagers' for EPEL-8 items
18:37:19 <nirik> but yeah, after that time I guess it would be 'does someone want to step up and maintain it'
18:38:11 <pgreco> yeap, I think that would help speed up things
18:38:19 <nirik> (those in that query are all NEW, so probibly maintainers haven't answered, but they could have and left it in new'
18:39:29 <tdawson> Looking at that list, I see two that I know aren't built because of dependencies, or ... well mongodb's license no longer allows it in EPEL.
18:39:53 <smooge> yeah.. I think anything related to mongodb is 'CAN NOT/WILL NOT FIX'
18:39:55 <pgreco> is mongodb allowed anywhere these days....?
18:40:14 <smooge> Sadly MongoDB is just a pawn in the game of VC life
18:40:44 <nirik> I closed a few that were in appstream too...
18:41:03 <nirik> anyhow, if folks want to play around with queries we can discuss what to do with them on the list?
18:41:32 <tdawson> That sounds good ... bring it up next week with proposed solutions?
18:41:57 <smooge> .. wonders what package he has to install to get bugzilla
18:42:24 <tdawson> python-bugzilla-cli
18:42:40 <smooge> ah.. cool. I was looking at a large helping of perl
18:43:11 <smooge> sgallagh, other than round tuits anything else we can do on EEPL-8 for you?
18:43:28 * nirik nods, although I will not be around next week. :)
18:43:41 <smooge> #info Will revisit open bugs next week
18:43:51 <smooge> #info nirik is out next week
18:43:58 <smooge> #topic Open Floor
18:44:03 <sgallagh> round tuits?
18:44:09 <nirik> so I have one item:
18:44:31 <nirik> kanarip noted that our docs are... really old/bad/poor. We should try and fix that...
18:44:41 <smooge> sgallagh, http://www.lightspeedtokens.com/customroundtuits.html
18:45:01 <sgallagh> ah
18:45:03 <nirik> perhaps when you go over grobisplitter this week you could document the setup in detail so we at least have that for people who wonder what the setup is?
18:45:06 <smooge> nirik, I agree with him on this
18:45:12 <sgallagh> nirik: +1
18:45:14 <smooge> I will do so that also
18:46:44 <nirik> and we should try and move our docs out of the wiki...
18:47:05 <smooge> yep
18:47:23 <nirik> I think we should look at going direct to docs.fedoraproject.org...
18:48:27 <smooge> so I was looking at that previously and I thought it needed to sit in a git repo for them to get it from
18:48:31 <smooge> did that change?
18:49:06 <nirik> yes, it needs a repo, but I think thats setup in docs namespace for us when we request it?
18:49:15 <nirik> last cut we were going to use docs.pagure.io
18:50:17 <smooge> so I started putting things into the actual epel file repo
18:50:42 <smooge> so pagure.io/epel has a couple of RST files but I can put them wherever they are needed (or use what format is needed)
18:51:52 <nirik> yeah, I think it's better to add them to the main fedoraproject docs. I think we should ask pbokoc about where and what and such?
18:52:03 <smooge> yes I will do that and cc you
18:52:54 <smooge> anything else for today?
18:53:06 <nirik> thats all I got
18:58:27 <smooge> #endmeeting