17:01:11 <smooge> #startmeeting EPEL (2018-10-31)
17:01:11 <zodbot> Meeting started Wed Oct 31 17:01:11 2018 UTC.
17:01:11 <zodbot> This meeting is logged and archived in a public location.
17:01:11 <zodbot> The chair is smooge. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:01:11 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:01:11 <zodbot> The meeting name has been set to 'epel_(2018-10-31)'
17:01:12 <smooge> #meetingname epel
17:01:12 <smooge> #topic Chair and Introductions
17:01:12 <zodbot> The meeting name has been set to 'epel'
17:01:12 <smooge> #chair avij bstinson Evolution nirik smooge
17:01:12 <zodbot> Current chairs: Evolution avij bstinson nirik smooge
17:01:22 <nirik> morning
17:01:45 <avij> oh, hello
17:02:19 <smooge> hello
17:02:39 <smooge> I am checking on bstinson and Evolutions availability
17:03:46 <smooge> #topic RHEL-7.6 available
17:04:02 <smooge> So RHEL-7.6 came out this week and it has a LOT of changes from 7.5
17:05:01 <smooge> new packages and packages with major bumps
17:05:10 <pgreco> hello
17:05:26 <smooge> I sent a list of packages which I know are 'broke' currently one way or another
17:05:53 <smooge> we need to drop one package and also fix our sync of aarch64 because it seems some packages are supposed to be in it but we can't sync them
17:06:08 <pgreco> smooge, some of those were already broken in 7.5
17:06:16 <smooge> probably
17:06:34 <smooge> I shouldn't say that.. they were definitely
17:06:46 <smooge> but I would like to get them fixed now
17:07:04 <pgreco> great
17:07:14 <avij> I filed some "please recompile" bugs back then. some got recompiled, some others not. maybe this needs a provenpackager approach.
17:07:27 <smooge> I think it is going to need a provenpackager
17:07:43 * nirik is happy to work on it as time permits.
17:08:34 <avij> that'd probably work better
17:08:41 <avij> than bugging individual developers
17:09:05 <tdawson> I might have some time as well.
17:09:23 <smooge> nirik, the first item that needs fixing is removing the libglvd package
17:09:40 <smooge> should I put that in as a releng ticket after this meeting?
17:09:41 <pgreco> we'd need to know which ones are in sync with master, which ones just need a version bump
17:10:05 <nirik> smooge: well, that breaks aarch64 then right? and/or did puiterwijk already do this?
17:10:07 <pgreco> and which ones require missing pkgs, like beets
17:10:42 <puiterwijk> The package is removed. But we didn't get the aarch64/ppc64le ones from RHEL7 yet, it seems
17:10:45 <smooge> nirik, that is the part where we need to figure out what is wrong with our scripts
17:11:07 <nirik> ok. note that 7.6 alt might not in fact be out. ;)
17:11:23 <puiterwijk> RHN says the packages did get released
17:11:25 <smooge> nirik, I asked about that and there wasn't going to be one
17:11:27 <puiterwijk> But anyway, sure
17:11:41 <nirik> right, that too.
17:12:13 <smooge> so after this meeting I am going to try and figure out why we can't and fix it
17:12:20 <nirik> sounds good
17:12:37 <smooge> however at the moment people can't install MATE or rebuild a lot of packages because of this
17:14:03 <smooge> so I would like to get a plan of action approved: 1. I try to get downloads. 2. If downloads do not exist for some reason I work with a proven packager to build correct libglvd for aarch64/ppc64le
17:14:30 <smooge> 3. we either remove or replace libglvd and put in the buildroot appropriately
17:14:39 <nirik> sounds good... but if we build it it will need to be used by all arches.
17:14:49 <smooge> 4. we get some proven packagers to go over the list of broke packages and fix
17:17:16 <avij> yes, I trust that smooge knows what he's doing
17:17:45 <tdawson> smooge: When you get to step 4, give me a call, I'm a proven packager, and would like to help.
17:18:16 <Evolution> better late than never. sorry.
17:18:20 <smooge> thanks tdawson . I am looking to see what the exact package is that needs possibly rebuilt
17:18:24 <smooge> processing...
17:19:05 <smooge> libglvnd-1.0.1-0.8.git5baa1e5.el7.src.rpm
17:20:12 <smooge> nirik, to follow the rules of not over-riding .. we would rebuild  libglvnd-1.0.1-0.8.git5baa1e5.el7.src.rpm with a release of ?
17:20:23 <smooge> 0.7999999 ?
17:20:46 <nirik> 1.0.1-0.0.8.git5baa1e5
17:20:52 <nirik> add leading 0
17:21:33 <smooge> hmmm
17:21:40 <smooge> the current one in EPEL is /srv/web/pub/epel/7/x86_64/Packages/l/libglvnd-1.0.1-0.1.git5baa1e5.el7.x86_64.rpm
17:23:14 <smooge> I am guessing that the compose may get confused over 1.0.1-0.0.8.git5baa1e5 vs 1.0.1-0.1.git5baa1e5
17:23:44 <nirik> it shouldn't care. it composes everything every updates push.
17:23:57 <nirik> users however may not see the newer one.
17:25:11 <smooge> ok I was worried the system would say "1.0.1-0.1 > 1.0.1-0.0" so include 1.0.1-0.1
17:25:22 <smooge> we can cross that bridge if it happens
17:25:42 <smooge> does that sound like a workable plan for people?
17:25:43 <nirik> koji doesn't care about versions... only order things are tagged in. ;) but yeah, we might want to make it newer than the old one and older than the new one
17:25:57 <nirik> but ideally we can get the aarch64 stuff synced and not deal with it
17:27:09 <smooge> I will work on getting that after this meeting
17:27:35 <smooge> Evolution, avij do you have the numbers on how big of a delta 7.6 is to 7.5?
17:29:04 <Evolution> 520 packages, roughly
17:29:10 <Evolution> source packages that is.
17:31:00 <smooge> with GNOME bumped to a much newer version and some other bumps.. it might be a good idea to 'ok' any related communities to do bumps
17:31:01 <avij> right. I don't know if there is some other metric that would be meaningful.
17:31:20 <smooge> that is my usual metric
17:32:27 <smooge> #topic python3.6
17:33:08 <smooge> so I haven't heard much from python people on when they would be available, but I would like to look at doing a VFAD in mid/late November
17:34:01 <nirik> sounds good
17:34:07 <smooge> How does November 22nd sound?
17:34:29 <nirik> is that thanksgiving? oh no, the week before... sure.
17:35:20 <avij> should be possible
17:36:24 <smooge> nirik, yeah.. it would seem so. How does any day earlier in that week sound?
17:36:56 <smooge> It takes about 2 weeks to get a VFAD together.. so the earliest we can do it is the 15th
17:37:15 <nirik> thursdays have a lot of meetings... wed would be better for me, but we could do thursday if others prefer
17:37:25 * avij does not care about thanksgiving
17:37:33 <smooge> Thu is my hell day. I was thinking a Tuesday
17:38:04 <smooge> Oh we could do the 20th.
17:38:11 <pgreco> does it complicate things that centos 7.6 might not be already out yet?
17:38:30 <smooge> pgreco, for the python rebuild update?
17:38:32 <smooge> possibly
17:38:34 <pgreco> yes
17:39:02 <smooge> wait.. no
17:39:10 <avij> C7.6 packages would most likely be in the CR repo by then
17:39:12 <smooge> we are rebuilding python36
17:39:23 <smooge> which isn't in RHEL at all
17:39:34 <smooge> so shouldn't be in CentOS at all
17:39:39 <smooge> *cough*
17:39:42 <orc_fedo> nirik thanksgiving is on the 26th actuall, that week -- it is as early as it may be with calendar jitter
17:40:10 <pgreco> smooge, I'm thinking about what devs are using to build/test before upload
17:40:57 <smooge> orc_fedo, I thought Nov 26th was a Monday. Am I using a Julian calendar?
17:40:59 <pgreco> but avij is right, CR should be already out
17:41:12 <orc_fedo> whoops
17:41:15 <orc_fedo> my bad
17:41:34 <orc_fedo> holidat then is: 22
17:41:59 <smooge> ok so I am looking at Nov 16th (Friday always bad) or Nov 20th (maybe?) or Nov 27/28
17:42:51 <smooge> how do any of those dates look pgreco avij Evolution nirik et al?
17:43:32 <pgreco> no problem for me, I won't be of much use for python though ;)
17:43:58 <smooge> well it will mostly be rebuild the src.rpm in whatever arm buildsystem you have
17:43:58 <nirik> 20th sounds good
17:44:23 <smooge> It would be good for me also
17:44:56 <avij> Friday Nov 16th I won't be home; Tue 20th and Tue 27th / Wed 28th would work better. on the other hand, I don't know if my presence matters much in this.
17:45:56 <smooge> I would like to have as much of the steering committee available to make sure if we have any 'oh well fudge that means we need a policy' to answer
17:46:36 <avij> sure, I can participate in such decisions
17:48:14 <smooge> okie dokie. I am putting this down for the 20th starting at 17:00 UTC until 23:59 UTC
17:48:58 <bcotton> smooge: shoot me the details when you're done. i'll add it to my weekly post on the community blog
17:49:01 <smooge> nirik, is that too early?
17:49:09 <nirik> nope, sounds good
17:49:25 <smooge> #topic US Daylight Savings HELL Ends this weekend
17:49:37 <smooge> #info Meeting Times are in UTC
17:50:32 <smooge> So 17:00 UTC is whenever 17:00 UTC is when DST is over
17:51:04 <smooge> #topic Open Floor
17:51:30 <smooge> OK I don't have anything else for this meeting.
17:51:47 <smooge> Does anyone else?
17:51:57 <smooge> oh pgreco everything going ok on the arm epel?
17:52:12 <pgreco> smooge, not much
17:52:27 <pgreco> rust got the fixes I sent this week, still waiting on the others
17:52:45 <pgreco> slowly, but going ;)
17:53:42 <pgreco> I'm hoping that the need for rebuilds in 7.6 will help me
17:54:35 <smooge> ok cool
17:54:55 <smooge> bcotton, I will do so tomorrow
17:55:10 <smooge> if no one has anything else.. I am going to turn into a duck
17:55:18 <smooge> #endmeeting