15:34:18 <dgilmore> #startmeeting RELENG (2014-06-16)
15:34:18 <zodbot> Meeting started Mon Jun 16 15:34:18 2014 UTC.  The chair is dgilmore. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:34:18 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:34:24 <dgilmore> #meetingname releng
15:34:24 <zodbot> The meeting name has been set to 'releng'
15:34:30 <dgilmore> #chair dgilmore nirik tyll sharkcz bochecha masta
15:34:30 <zodbot> Current chairs: bochecha dgilmore masta nirik sharkcz tyll
15:34:41 <dgilmore> #topic init process
15:34:45 <dgilmore> who is here?
15:34:52 <nirik> morning
15:34:56 <dgilmore> I know tyll said he would be away
15:35:07 * sharkcz is here
15:35:29 <dgilmore> as we get started, I wrote a wiki page last week
15:35:32 <dgilmore> https://fedoraproject.org/wiki/ReleaseEngineering/Meeting_Process
15:35:42 <dgilmore> I documented a process for meetings
15:35:59 <dgilmore> which i did not totally follow this week, I failed to send an agenda on Friday
15:39:36 <dgilmore> feedback and changes to the page welcome
15:40:34 <dgilmore> #topic #5877 Schedule and coordinate mass rebuild for F21
15:40:40 <dgilmore> https://fedorahosted.org/rel-eng/ticket/5877
15:40:58 <dgilmore> so i just closed this but we should probably do a recap
15:41:19 <dgilmore> one issue was that make 4 was not built before the mass rebuild
15:41:23 <nirik> I think it went pretty smoothly overall.
15:41:26 <nirik> yeah.
15:41:48 <dgilmore> for the most part it was faster and pretty smooth
15:42:12 <dgilmore> we do need to rewrite the script to send builds, but I think we talked about that last week
15:42:26 * masta looks in
15:42:36 <masta> howdy - sry late
15:42:50 <dgilmore> I guess we need to haver a step in the process, make sure everything that is supposed to be in before the mass rebuild is
15:43:14 <nirik> yep.
15:43:39 <nirik> There were only a very few jobs that stalled... much better than the last one where I recall a number getting stuck.
15:44:36 <dgilmore> yeah
15:44:55 <dgilmore> overall it was pretty good
15:45:30 * nirik nods.
15:45:36 <dgilmore> anyway, i closed the ticket
15:45:59 <dgilmore> #topic #5870 rawhide signing
15:46:05 <dgilmore> https://fedorahosted.org/rel-eng/ticket/5870
15:46:20 <dgilmore> not sure we can talk about much without tyll
15:47:00 <dgilmore> nirik: what is the status of the tickets for access to secondary arch setups?
15:47:17 <nirik> dgilmore: I think thats done... can doublecheck
15:47:39 <dgilmore> okay cool. when tyll is back from holiday we can move forward with it again
15:47:43 <nirik> yeah
15:47:45 <masta> yep
15:47:54 <dgilmore> #topic #5914 Move fedmsg based blocking service to Fedora Infrastructure
15:48:02 <dgilmore> https://fedorahosted.org/rel-eng/ticket/5914
15:48:11 <dgilmore> this is something with tyll also
15:48:23 <dgilmore> so not much to discuss today
15:48:34 <nirik> right
15:48:52 <dgilmore> #topic Secondary Architectures updates
15:48:54 <dgilmore> #topic Secondary Architectures update - ppc
15:49:03 <dgilmore> masta: want to give a ppc update?
15:49:10 <masta> dgilmore: sure.
15:50:07 <masta> so I'm really busy, but looking forward to get my hands dirty with ppc64 support.... this week I'm going to start spending more time with them as other tasks tone-down.
15:50:46 <masta> Today I got a report of broken mash config for debug packages, so I look forward to that kinda... will get me into the mix a bit
15:51:03 <masta> I do not believe they have rebuilt yet though
15:51:25 <masta> that's all I know for ppc64 right now. Any questions?
15:51:36 <sharkcz> rawhide is stopped until the merge is done, we are currently waiting on antrl3 to be fixed (FTBFS even on primary), the merger is tracked at https://fedoraproject.org/wiki/Architectures/PowerPC-merger
15:52:23 <sharkcz> masta: I'll catch you later for a talk about available LE hw
15:53:21 <masta> sharkcz: ok, great
15:55:03 <dgilmore> sharkcz: cheers
15:55:13 <dgilmore> masta: cheers
15:55:19 <dgilmore> #topic Secondary Architectures update - s390
15:55:25 <dgilmore> sharkcz: how goes s390?
15:56:11 <sharkcz> rawhide is currently blocked by texlive crashing during test-suite run, so mas rebuild not started yet
15:56:17 <dgilmore> fun
15:56:38 <dgilmore> I assume that people are working on it
15:56:55 <sharkcz> it's only me, not sure yet if it's buildroot issue due the glibc or some texlive issue
15:57:22 <dgilmore> okay
15:57:36 <dgilmore> I know aarch64 has had a few glibc regressions lately
15:57:55 <dgilmore> lets try get you some help
15:58:23 <sharkcz> yep,t we have a new junior team member who is helping with ppc/s390, so I'll probably ask hi
15:58:30 <sharkcz> him
15:58:40 <dgilmore> cool
15:58:43 <dgilmore> sounds good
15:58:58 <dgilmore> anything else you want to talk about s390 related?
15:59:01 <sharkcz> yeah, he already fixed couple big endian related issues
15:59:02 <sharkcz> nope
15:59:15 <dgilmore> fixing issues is good :)
15:59:18 <dgilmore> #topic Secondary Architectures update - arm
15:59:33 <dgilmore> okay, so mass rebuild has under 3000 packages to go
15:59:35 <sharkcz> especially when he got them into upstreams :-)
15:59:45 <dgilmore> there has been a few  glibc regressions :(
15:59:52 <dgilmore> sharkcz: thats even betetr
16:00:15 <sharkcz> I'm teaching him the proper workflow :-)
16:00:28 <dgilmore> kernel wise 3.16 should boot on the xgene hardware with efi, but is missing the ethernet driver afaik
16:00:42 <dgilmore> sharkcz: :) awesome, thats what I like to hear
16:01:23 <dgilmore> I am going to try make a remix installer compose with a patched kernel this week.
16:01:39 <dgilmore> aarch64 is coming along nicely
16:01:50 <dgilmore> f19-updates are moving along still
16:02:13 <dgilmore> I think thats it for arm
16:02:15 <dgilmore> #topic Open Floor
16:02:24 <dgilmore> anyone have anything to bring up?
16:02:46 * nirik has nothing off hand...
16:03:23 <masta> how would we test a new script to rebuilds all the things?
16:03:31 <masta> the threaded script idea
16:03:55 <masta> setup a private fedora infra at home to sandbox things?
16:04:01 <nirik> masta: we might be able to test in our staging setup
16:04:49 <nirik> would take a long time to do all the builds if we didn't cancel them or something tho
16:04:49 <dgilmore> masta: staging koji
16:05:17 <dgilmore> nirik: in theory we could just throw the koji away and start over
16:05:33 <nirik> dgilmore: true. would be a nice way to test the 'make a stg koji playbook'
16:05:45 <dgilmore> yep
16:06:41 <dgilmore> I had something but can not currently remeber
16:07:07 <nirik> those mash fixes would be nice to get done, but it's going to be a bunch of painfull debugging I fear.
16:07:23 <dgilmore> nirik: yeah. thats something we need worked on
16:07:31 <dgilmore> it will be running a bunch of mashes
16:07:40 <dgilmore> and adding debuging prints
16:07:55 <dgilmore> work out where its failing
16:08:57 <dgilmore> if nothing else. I need coffee
16:09:19 <dgilmore> #endmeeting