15:01:37 <tflink> #startmeeting fedora-qadevel
15:01:37 <zodbot> Meeting started Mon Feb 15 15:01:37 2016 UTC.  The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:01:37 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:01:37 <zodbot> The meeting name has been set to 'fedora-qadevel'
15:01:37 <tflink> #topic Roll Call
15:01:49 * mkrizek 
15:02:20 * kparal 
15:02:41 <tflink> #chair kparal mkrizek jskladan
15:02:41 <zodbot> Current chairs: jskladan kparal mkrizek tflink
15:02:50 <tflink> since I usually forget to do that
15:03:24 * linuxmodder sits in back row adn learns
15:03:24 * jskladan enjoys the chair
15:03:32 <tflink> I assume no jan or lukas today?
15:04:03 <kparal> doesn't seem so
15:04:32 <tflink> linuxmodder: welcome, feel free to ask questions as you have them
15:04:56 <tflink> ok, let's get started
15:05:14 <tflink> I suspect this'll be a shorter meeting since there's not a whole lot to discuss ATM
15:05:33 <tflink> #topic Announcements and Information
15:05:44 <tflink> nothing was put up in the wiki page - anything to add?
15:06:11 <jskladan> not really, I was mostly slacking on the end of the week :/
15:06:20 <mkrizek> #info massive disposable clients explosions during mass rebuil, should be at least partially fixed
15:06:36 <kparal> #info posted a patch for forwarding cli options to minions
15:07:04 <kparal> mkrizek: it actually looks completely fixed, yay
15:07:26 <tflink> mkrizek: the fix is 100% in ansible, right? I think you said it was but wanted to confirm
15:07:44 <linuxmodder> getting back into the f24 test swing as is so figured I'd hang out and straddle back in tflink
15:07:49 <mkrizek> the config is in ansible, yes
15:08:17 <tflink> awesome, glad to see that fixed
15:08:25 <mkrizek> but I saw a few "couldn't boot in x seconds" crashes on Friday :/
15:08:35 <mkrizek> I haven't investigated further though
15:09:13 <tflink> a couple is better than the hundreds we were seeing :)
15:09:43 <mkrizek> true
15:09:52 <tflink> not sure how much effort it's worth, though. i'd rather put energy into a system for catching crashed (our fault) jobs and rescheduling them
15:10:11 <tflink> whether that ends up being part of execdb, trigger or something new
15:10:30 <mkrizek> agreed, that's been on my todo list for some time now
15:10:49 <tflink> ah, the todo list of doom :)
15:11:16 <tflink> any other announcement/information items?
15:11:18 <tflink> oh
15:11:34 <kparal> hey, I reserve the right to have the todo list of doom for myself
15:11:52 <tflink> #info qadevel changes coming soon - phab will be moving to phab.qa.fedoraproject.org
15:12:08 <kparal> we might need to fix some hyperlinks
15:12:15 <kparal> in the docs, on wiki, etc
15:12:16 <tflink> kparal: if you really want it, I can send you the todo lists that I've been compiling over the last month
15:12:19 <tflink> yeah
15:12:28 <mkrizek> kparal: everyone has a right to be doomed
15:12:37 <tflink> there should be good enough redirects that it shouldn't be an issue but there's that S-word again
15:12:46 <tflink> twice, even
15:13:16 <tflink> moving on, then
15:13:29 <tflink> #topic Upcoming Work
15:14:10 <tflink> we ended up spending quite a bit of time around devconf discussing the roadmap and priorities
15:14:34 <linuxmodder> for any doc stuff feel free to ping me
15:14:48 <linuxmodder> I am working on several of the guides for  24  release
15:14:48 <tflink> most of which hasn't made its way into phab yet but the rough outline is in an org-mode doc that I've been working off of
15:15:32 <tflink> http://paste.fedoraproject.org/322896/55493271/
15:15:47 <tflink> that's the current state of the "draft priority" doc I have
15:16:03 <tflink> I'm planning to get that taken care of today, I hope once phab is upgraded in prod
15:16:57 <tflink> for the moment, I think that folks have tasks
15:17:18 <tflink> jskladan: any thoughts on when the image bits will be wrapped up? IIRC, it was pretty close last week
15:17:36 <tflink> kparal: do you have any tasks lined up after T596 is done?
15:18:16 <kparal> atm I'm looking at how our stdout and stderr streams interleave quite badly when they're pushed through paramiko. I have some ideas to improve that
15:18:33 <tflink> kparal: is there a ticket filed for that?
15:18:36 <kparal> nope
15:18:47 <kparal> I just noticed it today, so I started looking into that
15:18:53 <jskladan> I'm going to set up the cron jobs later today, hopefully, and then I'll hook it up in taskotron tomorrow (automatic image discovery, with config as overload)
15:18:56 <tflink> could you file one if you start spending more than an hour or two on it?
15:19:06 <kparal> ok
15:19:24 <tflink> jskladan: i forget, is all the code we need to load the latest image from a dir already in libtaskotron?
15:19:25 <jskladan> kparal: this again... has it been a year already?
15:19:44 <jskladan> tflink: IIRC it is, just not being used
15:19:52 <tflink> even the config option?
15:19:59 <jskladan> nope, that I need to add
15:20:12 <jskladan> we have a general "find a newest image in a dir" function
15:20:15 <kparal> jskladan: it's the same problem in a different place (paramiko instead of buildbot). and for this particular case, I think I can fix it :)
15:20:16 <tflink> either way, so long as it all gets done before we throw any config switches :)
15:20:30 <jskladan> adding a config override is on my plate for tomorrow
15:20:49 <tflink> sounds good
15:20:50 <linuxmodder> the dist-git is  for  rpm building bits right?
15:20:59 <tflink> linuxmodder: not sure I understand your quesiton
15:22:13 <linuxmodder> on above paste
15:24:15 <tflink> dist-git is the system of git repos used to build packages, yes
15:24:15 <linuxmodder> the tasks under Dist-Git
15:24:16 <tflink> we're planning to add a "task-dist-git" repo so that packagers and interested testers can add automated tasks for packages that they're interested in
15:24:16 <tflink> but the exact details around that are still rather fuzzy
15:25:18 <linuxmodder> so not just  maintainers /co maintainers then?
15:25:18 <tflink> a superset of maintainers
15:25:19 <linuxmodder> kinda like cc yourself on rhbz
15:25:20 <tflink> not really - the "task maintainers" would be able to at least update, create and change the automated tasks run against a package
15:25:21 <jskladan> tflink: /me needs to learn to type faster...
15:25:23 <tflink> jskladan: am I answering questions right before you do?
15:25:24 <jskladan> eh, about five seconds faster than me :D
15:25:24 <tflink> mkrizek: do you have enough tasks to keep you busy for a bit?
15:25:24 <tflink> since this seems to have morphed into "tasking" :)
15:26:19 <mkrizek> tflink: for a bit yeah, will ping you once that changes :)
15:26:19 <tflink> sounds good
15:27:06 <tflink> that's pretty much all I had for today, which brings us to ...
15:27:10 <tflink> #topic Open Floor
15:27:23 <tflink> anything that folks want to bring up which wasn't addressed already?
15:27:42 <mkrizek> not that I can think of
15:28:00 <tflink> linuxmodder: I'm not trying to cut you off if you had more questions, just figured that the conversation could easily continue outside the meeting
15:28:36 * jskladan has nothing to add now
15:28:57 * tflink dusts off the patent-pending non-deterministic fuses which haven't been used in a while
15:29:04 <tflink> and lights one
15:29:15 <linuxmodder> tflink,  totally understood no prob
15:29:40 <tflink> and ... BOOM
15:29:41 <jskladan> http://www.mememaker.net/static/images/memes/3534272.jpg
15:29:47 <tflink> thanks for coming, everyone
15:29:59 * tflink will send out minutes shortly
15:30:23 <tflink> jskladan: some days, yes but that's the wonderful/terrifying thing about the fuse being non-deterministic ... you never know :)
15:30:26 <tflink> #endmeeting