20:00:14 <tdawson> #startmeeting EPEL (2022-04-20)
20:00:14 <zodbot> Meeting started Wed Apr 20 20:00:14 2022 UTC.
20:00:14 <zodbot> This meeting is logged and archived in a public location.
20:00:14 <zodbot> The chair is tdawson. Information about MeetBot at https://fedoraproject.org/wiki/Zodbot#Meeting_Functions.
20:00:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
20:00:14 <zodbot> The meeting name has been set to 'epel_(2022-04-20)'
20:00:14 <tdawson> #meetingname epel
20:00:14 <zodbot> The meeting name has been set to 'epel'
20:00:14 <tdawson> #chair nirik tdawson pgreco carlwgeorge salimma dcavalca
20:00:14 <tdawson> #topic aloha
20:00:14 <zodbot> Current chairs: carlwgeorge dcavalca nirik pgreco salimma tdawson
20:00:19 <carlwgeorge> .hi
20:00:20 <zodbot> carlwgeorge: carlwgeorge 'Carl George' <carl@redhat.com>
20:00:25 <pgreco> .hi
20:00:26 <zodbot> pgreco: pgreco 'Pablo Sebastian Greco' <pablo@fliagreco.com.ar>
20:00:41 <tdawson> Hi carlwgeorge
20:00:45 <tdawson> Hi pgreco
20:00:51 <nirik> morning wonderfull epel peoples.
20:01:01 <dcavalca> .hi
20:01:02 <zodbot> dcavalca: dcavalca 'Davide Cavalca' <dcavalca@fb.com>
20:01:07 <tdawson> Hello wonderful nirik
20:01:14 <tdawson> Hi dcavalca
20:01:14 <dherrera> .hi
20:01:14 <zodbot> dherrera: dherrera 'None' <dherrera@redhat.com>
20:01:20 <tdawson> Hi dherrera
20:03:09 <salimma> .hello
20:03:09 <zodbot> salimma: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1".
20:03:17 <tdawson> Hi salimma
20:03:33 <salimma> .hi
20:03:34 <zodbot> salimma: salimma 'Michel Alexandre Salim' <michel@michel-slm.name>
20:04:50 <tdawson> Hmm ... I wonder if starting the actual business early will trigger another un-named person to suddenly appear ...
20:05:22 <tdawson> #topic EPEL Issues https://pagure.io/epel/issues
20:05:22 <tdawson> https://pagure.io/epel/issues?tags=meeting&status=Open
20:05:26 <tdawson> Guess not.
20:05:53 <salimma> I'll bite, who?
20:05:54 <tdawson> The only thing we have marked for a meeting is the CVE's.
20:06:00 <nirik> hum?
20:06:08 <tdawson> smooge
20:06:09 <rsc> .hello robert
20:06:10 <zodbot> rsc: robert 'Robert Scheck' <redhat@linuxnetz.de>
20:06:14 <tdawson> Hi rsc
20:06:58 <tdawson> salimma: CVE's is next week, correct?
20:07:26 <salimma> I think so, yeah
20:07:34 <salimma> last month it was discussed on the 20-something
20:07:57 <tdawson> That's what I was thinking.  OK, I've marked it for next week, and I'll do a better job of making sure I know which week.
20:07:57 <salimma> FWIW I am recovering from dealing with taxes so haven't done anything CVE related :)
20:08:03 <tdawson> :)
20:08:10 <tdawson> Then moving on
20:08:15 <tdawson> #topic Old Business
20:08:58 <tdawson> Wow, have I not been taking good notes, or do we really not have any old business.
20:09:19 <nirik> shocking
20:09:29 <salimma> last week's meetings have a lot of topics, but I can't remember them now
20:10:08 <tdawson> I think most everything we brought up last week, we finished last week.
20:10:14 <carlwgeorge> i've seen any activity for the imagemagick incompat update we approved
20:10:23 <carlwgeorge> *i haven't
20:11:11 <tdawson> I have made an imporovement to will-it ... it now does all the bugzilla's (not just 20), and shows the numbers on each page, and on the package page it lists them with links to the bugs.
20:11:32 <tdawson> But that just happened, and it's still churning through everything.
20:12:37 <carlwgeorge> nice
20:13:08 <tdawson> carlwgeorge: Hmm .. I'm noticing that the person who originally reported it, wasn't who got the email.  I'll see about making sure he get's it.
20:13:09 <nirik> whats the link again? or is it a report type thing?
20:13:18 <tdawson> https://tdawson.fedorapeople.org/epel/willit/status-overall.html
20:13:50 <tdawson> carlwgeorge: Meaning the person who initially sent the email about ImageMagick
20:15:10 <nirik> man, there's a lot of packages in epel7-testing
20:15:14 <tdawson> Hmm .. interesting ... looks like the overall page didn't get the change ... I'll have to fix that ... but the different repo's now have the changes.
20:16:04 <tdawson> OK, looks like I have some bugs to work out, but all the repo pages should look like this one - https://tdawson.fedorapeople.org/epel/willit/epel9/status-repo.html
20:17:34 <tdawson> Well, I think that's it for old business.
20:17:42 <tdawson> #topic EPEL-7
20:17:43 <tdawson> CentOS 7 will go EOL on 2024-06-30
20:18:56 <tdawson> Anything for epel7 this week?
20:19:16 <carlwgeorge> if anyone uses gitpython from epel7, i've got a pending update i could use karma on
20:19:23 <tdawson> I get the feeling this is going to be a hosrt meeting.
20:19:27 <nirik> just noticed all the updates in updates-testing. :) wonder if we should do anything about that
20:20:12 <pgreco> nirik, are they stalled for bad karma?
20:20:17 <nirik> nope
20:20:19 <tdawson> nirik on EPEL7?  Ya ... that's alot.
20:20:20 <nirik> https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2016-a5199f34b3
20:20:24 <nirik> 6 years old.
20:20:29 <nirik> just no karma at all
20:20:41 <nirik> https://bodhi.fedoraproject.org/updates/FEDORA-EPEL-2016-2f4a8e068f
20:20:50 <tdawson> Huh ... I thought I cleaned those out about a year ago.
20:20:53 <pgreco> so those should have been auto-pushed, unless the owner configured them not to somehow
20:20:59 <pgreco> tdawson: I had that feeling too
20:21:04 <salimma> do we need a 'stale update' report?
20:21:08 <nirik> yes, the submittor unset 'stable by time'
20:21:14 <salimma> ah
20:21:25 <salimma> we don't have critpath updates for EPEL, right? by definition
20:21:39 <nirik> but I'd say we should push them all to stable as long as they don't have - karma... perhaps
20:21:41 <nirik> right.
20:21:59 <tdawson> Well, it depends on how old they are ... but I think so.
20:23:08 <nirik> anyhow we don't have to derail the meeting, perhaps a topic for the list?
20:23:11 <tdawson> I'll take a look at them ... I used to be able to do bodhi queries to show things like that, but nowdays all those scripts I had are broken and I haven't been able to get them working again.
20:23:28 <salimma> yeah, might be worth discussing in the list
20:23:42 <tdawson> Yep.
20:23:49 <tdawson> Anything else epel7 related?
20:23:57 <salimma> e.g. should we do like an orphan report, and indicate who pushed which stale updates, and if they don't push/unpush them then they get pushed out?
20:24:58 <nirik> bodhi updates query --releases EPEL-7 --status testing --rows 175
20:25:02 <nirik> should show them all.
20:25:23 <nirik> I sent a email about ansible 2.9.x on epel7 going to be retired...
20:25:47 <salimma> --rows 175 seems very specific :)
20:26:14 <tdawson> Plus, rows can only be up to 100
20:26:32 <tdawson> nirik I saw that about ansible.  Have you gotten any negative feedback?
20:26:52 <nirik> salimma: there's 168 updates, so I wanted more than that. ;)
20:27:03 <salimma> --rows 9999 ;)
20:27:04 <nirik> nope, no feedback at all.
20:27:38 <carlwgeorge> hopefully most people are running their playbooks from el8 or newer.  i bet a few who are doing it on el7 will start complaining after it's retired there.
20:27:39 <nirik> oh yeah, can't go over 100, have to --page 2
20:27:50 <nirik> yep. I am sure they will.
20:28:58 <salimma> someone who really needs it can always start a COPR
20:29:01 <tdawson> Anything else for epel7?
20:29:12 <carlwgeorge> not from me
20:29:52 * nirik has nothing more for epel7
20:29:57 <tdawson> #topic EPEL-8
20:29:57 <tdawson> CentOS Stream 8 goes EOL in 2024-05-31
20:30:34 <salimma> good news for vi users: neovim 0.7.0 just made it to testing for both EPEL 8 and 9
20:30:50 <tdawson> Ya!!
20:32:35 <salimma> thanks to everyone here for sorting out the luajit issue here a few weeks ago, esp tdawson who unmuddled the situation by properly retiring luajit from the gitlab repo
20:33:00 <salimma> (though that part was epel9, heh, jumping the gun here)
20:33:38 <tdawson> I found about 770 packages that has been improperly retired.  Hopefully they are all now properly showing if they are a dead package or not.
20:33:44 <tdawson> I also fixed our script.
20:34:14 * SSmoogen walks in and everyone goes quiet
20:34:36 <tdawson> Quick ... everyone hide.
20:34:44 <tdawson> Anything else for epel8 ?
20:34:46 <pgreco> tdawson: improperly how?
20:35:46 <tdawson> pgreco: When we "retired" them, we created a main branch and put a dead.package file in there, but left the c9s branch be.  We now clean out the c9s branch and put a dead.package file in there, and in the main branch we have a dead.package.c9s
20:36:04 <salimma> ah
20:36:07 <pgreco> ack
20:36:07 <tdawson> Before, if someone went to the c9s branch, it still looked like the package was being worked on.
20:36:32 <nirik> :(
20:36:36 <tdawson> #topic EPEL-9
20:36:44 <pgreco> ok, I was interpreting the "improperly retired" as if they had bee retired when they shouldn't have
20:36:45 <tdawson> Just noticed I didn't ever change the topic. :)
20:36:53 <pgreco> not that the process for doing it was "improper"
20:37:23 <tdawson> pgreco: Ah ... Ya. ... That would be different.
20:37:35 <carlwgeorge> we hit a notable epel9 milestone
20:37:38 <tdawson> We have "unretired" one package that will be in RHEL 9.1
20:37:50 <carlwgeorge> there are now more source packages in epel9 than there are in c9s
20:37:59 <nirik> lacking a 27stroke-b form
20:38:01 <tdawson> Cool!!  That's great.
20:38:37 <carlwgeorge> so we get to brag extra at the rhel9 launch that epel has more packages than rhel already
20:39:08 <nirik> nice
20:40:02 <salimma> what was the equivalent for C8?
20:40:11 <salimma> e.g. how long did it take post-release to reach this milestone
20:40:48 <tdawson> At least 6 months.  Because it took a long time just to get started.
20:40:54 <nirik> yeah.
20:41:07 <nirik> since things didn't even start really until GA
20:41:45 <tdawson> And we had the modularity stuff we had to figure out.
20:41:48 <carlwgeorge> i'm not sure anyone ever tracked it for epel8.  hypothetically someone could look through the epel8 snapshots, but i don't think those go all the way back to 8.0.
20:43:01 <tdawson> Actually, looking at the koji history of the epel8 tag would give you that information.
20:43:19 <tdawson> Hmmm .... graphs ... :)
20:43:22 <SSmoogen> so the first missing package that I saw run into in a video about the Alma-9 beta was no elinks
20:44:25 <SSmoogen> as of last week there were 11500 long term CS9 systems and 3800 of those using EPEL-9
20:44:47 <tdawson> Wow ... that's getting up there.
20:46:06 <tdawson> Anything else for epel9 ?
20:46:52 <tdawson> #topic General Issues / Open Floor
20:46:57 <SSmoogen> not from the peanut gallery
20:47:11 <carlwgeorge> i wanted to throw one thing out there
20:47:22 <salimma> SSmoogen: elinks should be easy to get into epel
20:47:31 <carlwgeorge> we switched the general bodhi time frame for epel updates from 2 weeks to 1 week
20:47:57 <carlwgeorge> the incompat process still says they must remain in testing for 2 weeks regardless of karma
20:48:10 <tdawson> Oh ya ...
20:48:28 <carlwgeorge> i'm fine leaving that, but it's at 2 weeks to match to old general time limit, so i wanted to check how people felt about shortening it to 1 week to match the current
20:48:42 <salimma> yeah... it's a hassle to keep it different, I think
20:48:54 <tdawson> I think we should shorted it to 1 week, to match the change.
20:49:12 <salimma> do we have language for indicating push by karma should be disabled?
20:49:30 <salimma> since otherwise it can get promoted earlier
20:49:30 <Eighth_Doctor> shortening it is fine with me
20:49:53 <nirik> 1 week seems short, but that shouldn't really be the time/place people should notice this... so yeah.
20:50:09 <tdawson> salimma: Yes, the wording specifically says to disable karma pushing.
20:50:21 <SSmoogen> oh one other things.. on total counts, there are 12000 aarch64 Fedora-3* users and there are 11000 EPEL-8 users.  There are more s390x EPEL-8 users but less ppc64le than in Fedora
20:51:17 <tdawson> Ha!   As much as I like Fedora ... it sorta blows me away when I see the epel numbers.
20:51:24 <salimma> lots of people running stream 8 on aarch, interesting
20:51:40 <pgreco> salimma: count me in that list ;)
20:51:55 <salimma> AWS and Oracle Cloud, maybe? Oracle Cloud's free tier gives you 4-CPU aarch64 with 24GB RAM
20:52:19 <nirik> azure now has arm too.
20:52:24 <tdawson> carlwgeorge: Would you mind doing a pull request updating the incompat page so that it says 1 week, and then we can give our thumbs up/down on that.
20:52:26 <SSmoogen> The most EPEL-8 aarch64 is alma at 3900, Oracle at 1700, Stream at 1100 and RHEL at 2200, Rcoky at 1324
20:53:43 <carlwgeorge> can do
20:54:13 <tdawson> One of the hidden gems on my will-it-bugs is this page - https://tdawson.fedorapeople.org/epel/willit/epel9/status-bugz-no-source.html
20:54:37 <nirik> yep.
20:55:02 <tdawson> I'm really curious what it will show for epel8 and epel7 ... thus ... I need to fix my bug(s)
20:55:07 <nirik> due to us not having seperate component lists for different epel versions.
20:56:02 <tdawson> Anything else people want to bring up before I close the meeting?
20:56:10 <salimma> tdawson: no source?
20:56:32 <salimma> ooh nice
20:56:37 <salimma> as in, branch requests
20:56:52 <tdawson> https://github.com/tdawson/tdawson-misc-scripts/tree/master/willit
20:57:31 <nirik> oh neat, we stil have el4/el5/e6 versions available in bugzilla.
20:57:45 <tdawson> Open?
20:57:55 <tdawson> Meaning we have open bugs?
20:57:56 <nirik> ah, no, they are not active
20:58:11 <tdawson> Whew ... good.
20:59:52 <tdawson> Thank you all for coming and participating in the discussions this week.  And thank you for all you do for EPEL and it's community.
21:00:08 <tdawson> I'll talk to ya'll next week, if not sooner.
21:00:16 <tdawson> #endmeeting