15:01:10 #startmeeting Fedora QA meeting 15:01:10 Meeting started Mon Mar 23 15:01:10 2015 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:10 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:15 #topic Roll call 15:01:21 * roshi is here 15:01:23 #meetingname fedora-qa 15:01:23 The meeting name has been set to 'fedora-qa' 15:01:30 ahoyhoy! who's around to a some q 15:01:32 * satellit listening have to leave early 15:01:38 * tflink is here 15:02:03 * kparal is here 15:02:41 * FranciscoD|Uni is hanging around too 15:03:50 hi satellit, thanks for all the testing over the weekend 15:03:58 : ) 15:04:43 * pschindl is here 15:05:53 alright alright alright 15:05:59 #topic Previous meeting follow-up 15:06:14 so in fact we should go back a couple of meetings - I don't think we did the follow-up for 03-02 yet either 15:06:44 so, first up: "roshi to break down the migration testing into doable chunks" 15:06:49 (this was the dnf testing) 15:06:51 roshi? 15:06:55 I put tickets in the trac 15:07:05 * roshi grabs a link 15:07:15 * kinokoio is here 15:07:33 * jreznik is here 15:07:37 https://fedorahosted.org/fedora-qa/query?keywords=~dnf_migration&status=!closed 15:07:45 those are the 4 15:08:13 awesome 15:08:24 i have a topic for this scheduled later, so we can go back into it in detail then 15:08:40 sgtm 15:08:52 #info "roshi to break down the [dnf] migration testing into doable chunks" - this was done - tickets can be found via this search: https://fedorahosted.org/fedora-qa/query?keywords=~dnf_migration&status=!closed 15:09:11 "kparal to adjust fedocal to tie blocker review meetings to New York DST (as the other two are)" 15:09:17 i believe we did that too, right? 15:09:21 sure 15:09:37 there seem to be some side effects in fedocal, but in general it works 15:10:02 https://fedorahosted.org/fedocal/ticket/149 15:10:52 if people have troubles with it, I can delete the whole series and create it anew 15:11:13 has anyone noticed problems? 15:11:18 * roshi hasn't 15:11:54 #info "kparal to adjust fedocal to tie blocker review meetings to New York DST (as the other two are)" - this was done, but kparal notes that it seems to have caused some side effects in fedocal, notify him if the event causes any issues for you 15:12:00 ok, that's 03-2 15:13:06 #info "adamw to schedule openQA runs on nightlies until then [Beta TC1 release]" - this was kind of overtaken by events, as releng cut Beta TC1 super-early, and I think we've done a decent job of keeping up with testing since then 15:13:55 and that's all the follow-up, unless there's any other follow-up topics? 15:14:40 alrighty then 15:15:23 #topic Fedora 22 Beta status 15:15:45 so it seems like the whole 'get testing in early' idea is getting sandbagged a bit by a constant stream of more prominent bugs... 15:16:06 it happens :/ 15:16:06 are folks hitting the libhawkey/libedit crasher in TC4? 15:16:21 (that's the 'screen mysteriously goes blank shortly after hitting the hub' bug) 15:16:25 * satellit seems like it 15:17:58 fedup does not work at all in F21 15:18:18 the fedup bug is known and just waiting on a systemd update for f21... 15:18:22 unless you found a different one? 15:18:30 no, its the same 15:18:36 * kinokoio apologizes 15:18:49 but yeah, we're missing coverage of fedup, nfs repo tests, and the annoying kickstart tests everyone hates, mostly 15:18:56 kinokoio: no need to apologize, just clarifying :) 15:19:01 wish that systemd update would show up... 15:19:28 we're also missing a bunch of the Beta desktop tests, those we could knock out with live images so we could get onto that 15:20:43 bvincent and roshi knocked out a bunch of the server tests for TC1 which is great, but we do need to do the database role tests, right after server WG (i.e. me :/) writes some database role tests :P 15:21:03 Cloud doesn't look like it's been done since Alpha, what's the status of cloud images with beta roshi 15:21:05 ? 15:21:34 cloud images have been having some issues 15:21:54 I think most of them are getting rectified if they haven't been already 15:22:22 i see a bug in initscripts on the blocker list 15:22:25 our outputted images were qcow3 for a while, and the infra cloud instances most of us have access to only supports qcow2 15:22:34 * adamw didn't even know there was a qcow3 15:22:36 so those should be now formatted as qcow2 15:22:43 it's qcow2 v3 15:22:51 which is also known as qcow3 15:22:53 aiui 15:23:07 but it made it so we couldn't test images on openstack 15:24:08 but now that that's being fixed cloud stuff can be tested 15:24:39 ok, cool 15:24:41 that's about it for cloud, afaik 15:25:08 so looks like we should be able to make progress with beta testing if we work around the libedit bug one way or another (lives, use tc3 and work around the network bug by installing from dvd or whatever) 15:25:52 if folks can help try and get us to full coverage this week that would be awesome. i guess i'll also try and get an early blocker status email out this time to get the devs onto the blocker treadmill 15:26:23 #action adamw to send out early blocker status email 15:26:35 * roshi forgot to send the blocker review announce - but we'll still have it after the meeting 15:26:39 did anyone have any other notes on 22 beta? particular areas of concern? 15:26:54 roshi: ah, was gonna ask about that :) 15:27:04 scientific spin is having compose problems 15:27:09 :D 15:27:47 for Beta - good thing is we have productized anaconda password policy, I forwarded it to desktop and server MLs 15:28:05 there is a little problem with crypto volumes in anaconda https://bugzilla.redhat.com/show_bug.cgi?id=1196397 15:28:08 ah, that was the way it got resolved in the end? 15:28:12 what does "productized" mean? 15:29:02 roshi: each product, sorry edition, can have it's own policy 15:29:15 overriding anaconda's default 15:29:15 kinokoio: that crash was about assigning mount points to existing LVs, wasn't actually caused by encryption 15:29:29 ah 15:29:30 kinokoio: it should be fixed since, if I recall the version numbers rightly, Beta TC2; what did you hit it with? 15:30:16 it happened when I wanted to upgrade my system to F22 using the KDE Live of TC4 15:30:53 published the last duplicated yesterday 15:31:09 hum, i'm not sure yours is actually the same crash then. did you file a bug for yours? 15:33:27 well, we can follow up on #fedora-qa if needed 15:33:38 #topic DNF test planning 15:33:59 so here we are again :) as roshi said earlier, we have tasks for DNF testing: https://fedorahosted.org/fedora-qa/query?keywords=~dnf_migration&status=!closed 15:34:31 all they're waiting on is for people to take them over 15:35:47 this machine runs on soylent green, folks 15:36:11 roshi: i was thinking at least some of them should fit a test day nicely 15:36:18 "All Ingredients Certified Turing Complete!" 15:36:22 :P 15:36:23 yeah 15:36:25 same here 15:36:26 do we have an abrt test day scheduled? 15:36:31 * roshi checks 15:36:39 we have a lot of testcases for abrt 15:37:22 nope 15:37:34 but tomorrow we have cockpit test day 15:38:18 you're thinking rolekit? 15:38:37 https://apps.fedoraproject.org/calendar/list/QA/?subject=Test+Day 15:38:42 says cockpit 15:39:32 i was trying to see how you were relating cockpit test day to this dnf stuff :) 15:39:55 oh, no link, just plugging it :) 15:40:00 carry on :D 15:41:32 ah, i see :P 15:42:08 so yeah, does anyone want to volunteer to set up test days for the tasks that make sense? we might even be able to do one per topic 15:42:17 * adamw looks at no-one in particular, definitely not roshi 15:42:28 * roshi looks around 15:42:30 * kinokoio is up to it 15:42:41 I'll help you out :) 15:42:45 kinokoio: awesome! is there one of those areas you're interested in in particular? 15:42:51 doing all four would be a lot of work for a volunteer 15:43:16 I'll work through them with him 15:43:27 I can help in both Virtualization and Cockpit 15:43:58 * satellit afk 15:44:01 awesome, that sounds good 15:44:17 #action kinokoio and roshi to work on setting up test days for the DNF testing topics 15:44:30 if any of them doesn't really work for a test day, don't try and force it - we'll get to that testing another way 15:45:17 sounds good 15:45:33 coolbeans 15:45:35 Are they for particular in F22? 15:45:53 kinokoio: yes, the idea is to try and get as much dnf testing as possible in ahead of f22 release 15:46:22 i don't know how many open thursday slots we have left, but if there aren't enough, we can always run these on tuesdays or whatever 15:46:33 we'll get it figured out 15:47:10 great 15:47:13 * kinokoio will go to install F22 15:47:20 #topic Open floor 15:47:26 so that's all we had for the agenda - any other business, anyone? 15:48:23 Testday tomorrow and blocker review in 12 minutes 15:49:13 yay blocker review, fun times for everyone 15:50:11 oh yeah 15:50:26 Blocker Review: Making Day Drinkers out of all of us 15:51:02 =) 15:51:19 ok then, sounds like we don't have anything else, so let's wrap up and let people get a "coffee" before the blocker meeting 15:51:24 thanks for coming everyone! 15:51:25 sgtm 15:51:28 * adamw sets meeting fuse 15:53:09 blocker review in #fedora-blocker-review in 7 minutes, folks - cya there! 15:53:11 #endmeeting