14:00:26 #startmeeting fedora-qadevel 14:00:26 Meeting started Mon May 9 14:00:26 2016 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:26 Useful Commands: #action #agreed #halp #info #idea #link #topic. 14:00:26 The meeting name has been set to 'fedora-qadevel' 14:00:26 #meetingname fedora-qadevel 14:00:26 The meeting name has been set to 'fedora-qadevel' 14:00:26 #topic Roll Call 14:00:40 who all's here for the always exciting qadevel meeting? 14:00:59 * kparal is here 14:01:46 * mkrizek here 14:01:57 * lbrabec is here 14:02:54 ok, let's get started 14:03:07 I suspect that this may end up being a short meeting 14:03:17 #topic Announcements and Information 14:03:27 #info F23-25 base images for disposable minions are building on dev - jskladan 14:03:27 #info deployed task-libabigail in dev - mkrizek 14:03:27 #info deployed distgit style tasks support in dev - mkrizek 14:04:01 any additions/comments/questions? 14:04:57 * kparal has none 14:05:25 nope 14:05:33 nope 14:06:15 ok, moving on 14:06:17 #topic Docker Testing 14:06:28 just wanted to quickly touch base on this 14:06:48 we're mostly waiting for D842 and some examples ATM, right? 14:07:14 we'll need to figure out how to download layered images but I'm not sure that can be done quite yet since the build process isn't finalized 14:07:48 xunitparser package is on review, i hope it will be ok 14:08:26 that's waiting on another package, no? 14:08:44 I'd rather not push that change until python-xunit is packaged, at least in COPR 14:08:54 and added to our list of "stuff to package for fedora" 14:09:06 * tflink is OK with that diff otherwise 14:09:39 well already packaged the xunitparser, I'm waiting for sponsor 14:10:02 oh, didn't know that 14:10:17 can you post the bz# for the review? 14:10:32 * tflink will poke some sponsors, didn't realize that's what we were waiting on 14:10:51 sponsor or reviewer? 14:10:59 #link https://bugzilla.redhat.com/show_bug.cgi?id=1332999 14:11:13 sponsor, it is my first package 14:11:51 https://bugzilla.redhat.com/show_bug.cgi?id=1333525 - someone also had the same idea apparently 14:12:28 within a day of eachother, too 14:13:15 anything else on this? 14:13:20 not here 14:13:41 ok, moving on 14:13:46 #topic ABI Checking 14:13:55 Another topic that I just wanted to touch base on 14:14:12 #info ABI Checking has been deployed to dev, appears to be working so far 14:14:20 * tflink cheers 14:14:41 is there anything else that we know of which needs to be done on this? 14:15:08 https://github.com/sinnykumari/task-libabigail/issues/1 14:15:08 * kparal just requested the task to use /var/tmp instead of /tmp for extracting rpms 14:15:12 ^^ 14:15:31 yeah, extracting to /tmp seems like a bad idea 14:15:45 so we might see a larger number of OOM-killed runs in the meantime 14:15:46 especially since we're not sure if 6GB is gonna be enough RAM to start with 14:16:14 other thing would be to figure out what builds we want to exclude from critpath 14:16:40 and of course let the task run for a day or two to see if there are any other explosions :) 14:16:58 what are we currently running on, all packages? 14:17:02 all critpath? 14:17:04 critpath 14:17:08 ok 14:17:31 did sinny or dodji (hoping i spelled their names right) have any more input on what to run, or what not to run 14:17:47 IIRC, anything webkit or llvm should be avoided for now 14:18:28 but I also think that was an incomplete list 14:18:51 mkrizek: I will look into list of packages available in critpath and will try to figure out which all packages may take more than 10GB of RAM while running libabigail task 14:18:51 it's a big list, so a bit time consumig work 14:19:07 later we clarified we only have 6 GB RAM 14:19:19 I hope, if I'm reading scrollback correctly 14:19:20 if we need to reconfigure some stuff, that's an option 14:19:44 alternatively, we can try to flag the ones which die from OOM (after the /tmp thing is fixed) 14:20:00 * tflink wonders if it would be wise for us to disable tmpfs on the taskotron images 14:20:29 not sure 14:20:41 let's see how things go 14:20:49 I'd rather not spend time on that if we don't have to right now 14:21:04 +1 14:21:32 anything else on this? it sounds like things are progressing well 14:22:49 * tflink takes silence as a "no" 14:22:58 #topic Packaging 14:23:05 Another topic which should be quick 14:23:30 * kparal wonders why he initially read "#topic Panicking" :-) 14:23:47 There was a thread on qa-devel@ but as a quick summary: we've been asked to package all the libtaskotron bits in Fedora instead of building things outside koji 14:23:53 kparal: bad experiences packaging? 14:24:28 has to be my subconsciousness 14:24:58 I've committed to getting everything packaged in Fedora by F25, so it's not a "NOW! NOW! PANIK!!11!!" kind of thing but I'd like to get started sooner than later 14:25:12 i've/we've ... yay, English pronouns! 14:25:38 speaking of packaging, libtaskotron build finally got a builder in copr after 3 hours of waiting in a queue \o/ 14:25:43 :) 14:26:18 by F25 you mean by branch? 14:26:19 in the long run, we need to get libtaskotron into the Fedora repos and I think doing builds there will be better for us in the long run, anyways 14:26:56 mkrizek: it wasn't very specific, I'd like to shoot for that, though in case we hit problems 14:27:36 the biggest thing from infra was more that we have plans to do it 14:28:27 there was a miscommunication where they heard "yeah, we might get around to scheduling that task in the next month and might finish it in the next year" and that led to the "no more non-koji builds in the infra repo" 14:28:42 miscommunication/misunderstanding/whatever 14:29:43 it's been sorted out now, AFAIK 14:30:11 if something like this comes up again in the future, please let the team know 14:30:57 * tflink doesn't recall hearing anything more than "we won't be able to push builds into infra repo" but it's very possible that he just forgot :( 14:31:26 not blaming, don't care otherwise now that it's been figured out, just bringing up the "please tell folks" bit :) 14:31:49 moving on 14:31:58 if there's nothing else on this topic? 14:32:23 nothing here 14:32:30 #topic Tasking 14:32:44 Other than jskladan, is anyone else in need of tasks right now? 14:33:25 * kparal is not 14:33:35 other than potential update of stg/prod, I don't have tasks now 14:34:43 mkrizek: can you start taking a few of the packaging tasks, then? 14:35:13 unless there are remaining dist-git tasks that I'm forgetting 14:35:45 I don't think there are 14:35:47 ok, will do 14:36:12 we still need guinea pigs for the distgit :) 14:36:13 it'd be nice to have more of the dist-git stuff documented, though 14:36:17 i have a few in mind 14:36:28 #action tflink to find guinea pigs for distgit style tasks 14:37:13 thanks 14:37:43 anyone else in need of tasks? 14:39:44 * tflink assumes not 14:39:49 if you do, let me know :) 14:39:55 #topic Open Floor 14:40:01 Anything else that folks wanted to bring up? 14:40:51 nothing here 14:41:02 nothing here 14:41:15 nothing here 14:41:31 * tflink lights the magical patent-pending non-deterministic fuse 14:43:04 thanks for coming, everyone! 14:43:08 * tflink will send out minutes shortly 14:43:11 #endmeeting