15:01:45 #startmeeting Fedora QA meeting 15:01:45 Meeting started Mon May 20 15:01:45 2013 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:45 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:01:51 #meetingname fedora-qa 15:01:51 The meeting name has been set to 'fedora-qa' 15:01:55 #topic roll call 15:02:19 *beep* No signal. Unable to call "roll". 15:02:30 here 15:02:33 * satellit_e listening 15:02:47 * Viking-Ice here 15:02:51 * pschindl is here 15:03:30 morning folks 15:03:31 anyone else? 15:03:55 * jreznik is here 15:04:02 morning jreznik 15:04:09 #chair jreznik viking-ice 15:04:09 Current chairs: adamw jreznik viking-ice 15:04:31 #topic Previous meeting follow-up 15:05:01 #info "martix or adamw to write a Graphics Test Week recap email" - yeah, failed that again. I'll just move it to a personal todo and stop boring the meeting with it 15:05:02 * mkrizek is here 15:06:05 #info "tflink to write a list post announcing the blocker bug proposal webapp" - I don't believe he did this yet 15:06:18 oh, that is my fault, but feel free to do it 15:06:19 unfortunately, no 15:06:22 #info "adamw to look into Test_Day:2013-04-30_MariaDB lack of participation and see if we should re-run it" - continuing the festival of fail: nope 15:06:37 "brunowolff to re-test bug #909473 and re-open if it's still bad" 15:07:04 i don't see any comments in there' 15:07:12 wow, well we're clearly doing a bang-up job :) 15:07:27 #info "brunowolff to re-test bug #909473 and re-open if it's still bad" - not sure if he did this yet 15:07:36 well, festivals are fun 15:07:55 and that's the important lesson, kids. 15:08:20 anyhow, moving swiftly on! 15:08:21 #topic Fedora 19 Beta status 15:08:38 so, in "less sucky" news...beta looking pretty decent 15:09:07 I did a quick retest with the games spin on the intel machine and things looked OK. I haven't retested on AMD/AMD machine yet. 15:09:19 '#info Beta RC2 is currently in testing, some likely blocker bugs have been found but not too many 15:09:24 Guest15468: aha, thanks 15:09:28 #info Beta RC2 is currently in testing, some likely blocker bugs have been found but not too many 15:09:53 #info Go/No-Go is scheduled for Thursday 2013-05-23 15:10:15 anyone have anything to note/discuss on beta? 15:10:37 * satellit_e still trying to fix soas spin 15:10:55 thanks for mentioning go/no-go, adamw 15:11:03 corner cases still bad: 955779 no desktop on RV280; 962086 cannot skip bootloader; n-i-bz: grub2 requires LBA 15:11:27 yeah, i've been trying to get bcl to fix the bootloader thing 15:12:15 * jskladan lurks 15:12:20 jreiser: that rv280 one of yours look nutty 15:12:52 Yeah rv280 is nutty. I can run gdb, but need direction on effective debug 15:13:57 jreiser: is it actually using the radeon driver? 15:14:12 because I think RV280 should be blacklisted 15:14:17 it claims to be using ati. 15:15:00 It should be using llvmpipe when using gnome-shell. 15:15:12 jreiser: hmm. do you have it handy? can you fpaste 'glxinfo'? 15:15:13 yes, llvmpipe is ON. 15:15:16 so, I have a thing at some point to mention... ;) The desktop updates tests are pretty gnome centric. ;) Would there be any interest in alternate tests for other desktops? 15:15:22 or genericizing the existing one 15:15:23 I have seen the generated sse2 code. 15:15:47 nirik: genericizing the existing one would be great, assuming you can play the balance between genericness and usefulness 15:15:53 The box is not up now; will boot, then try for glxinfo. 15:16:07 ok, I can try and take a stab at it. 15:16:10 My machine is different in that I don't have SSE2 on my CPUs. 15:16:10 propose to the list? 15:16:31 nirik: if the changes are significant, sure 15:16:38 brunowolff: that was one of the bugs that came up earlier 15:16:51 right? 15:16:52 * nirik nods 15:16:57 tests for sugar-desktop would be very different than for other DE's 15:17:10 jreiser: well what we're saying is it should be using llvmpipe *not native acceleration* 15:17:24 satellit: the update tests are greyed for SoaS on the basis it doesn't have a graphical updater 15:17:35 ok 15:17:37 satellit: is there a 'rule' about how you update sugar? 15:17:53 yum update from root 15:18:00 g-i-s gets alpha-blended shadows. after that, gnome-shell eats CPU for 5 minutes before timing out with no change in graphics display 15:18:13 nirik: well, all test cases are more gnome-centric - I'm trying to use some common sense :) 15:18:43 jreznik: the updates ones are even more so tho. ;) For example, Xfce doesn't even ship PK... 15:18:50 adamw: 'glxinfo' requires access to display. Cannot get display. now what? 15:18:56 adamw: actually there is a sugar control panel to update the activities in sugar 15:19:05 satellit: oh, interesting 15:19:12 so we could probably do with looking at that too... 15:19:31 jreiser: you have to run glxinfo from within an x session, yeah...maybe try a KDE one :) but we can debug in the bug I guess. 15:20:08 jreznik: nirik: in general, genericizing them is a good thing. i'm aiming to make revising a lot of the existing test cases my Big Project for post-f19 pre-f20 15:20:30 there's quite a lot of other stuff needs doing, like writing some partitioning test cases, adding a yum test case, a buncha stuff. 15:20:36 http://wiki.sugarlabs.org/images/thumb/e/e3/Screenshot_of_Software_Update.png/700px-Screenshot_of_Software_Update.png 15:20:45 if anyone wants to work on any of that now, of course :) 15:20:48 * nirik nods again. 15:21:03 satellit: does that only update sugar activities, though? not the base system? 15:21:47 yes the update for sugar-desktop is in sugar-terminal root yum update 15:22:28 FYI http://wiki.sugarlabs.org/go/Tutorials/Introduction_to_the_Sugar_Interface 15:23:49 okay 15:24:00 so adding a yum testcase to 'base' is indicated there as well. 15:24:27 #info we could do with a 'update with yum' test case in the base matrix 15:24:37 xfce (and I think lxde and possibly others) use yumex... it just has a update thing... but it doesn't do notifications, so none of the gui notify stuff applies. 15:24:39 #action nirik to work on revising the 'desktop updates' test case to be more generic 15:24:57 nirik: it's easy enough to just make that 'if your desktop provides...' 15:24:58 mate uses yumex also 15:25:41 yeah, ok. so, update that one and make a seperate non gui/yum one? 15:26:40 adamw: rv280 glxinfo is http://paste.fedoraproject.org/13201/90635761 15:26:54 nirik: that's where i'd go with it, yeah 15:27:01 nirik: the yum test would be in the Base matrix rather than the Desktop matrix 15:27:10 will try and do so, if anyone else wants to first, feel free. 15:29:39 alrighty 15:29:43 anything else on Beta, folks? 15:32:04 well then 15:32:05 #topic Test Days 15:32:30 so, frantic work appears to be underway on https://fedoraproject.org/wiki/Test_Day:2013-05-21_AnacondaNewUI_Followup 15:33:13 speaking of partitioning test cases...there's some stuff to yoink 15:34:13 so looks like 'mganisin', who is Marian Ganisin from RH in brno, is doing a lot of work on this one 15:34:30 jreznik / pschindl, are you in touch with that? 15:35:57 * nirik is reminded to look at the feature page and try and help get it fully finished. 15:36:41 adamw: I talked to them when they started planning it, I'll try to get in touch to see what are the plans right now 15:36:44 #info https://fedoraproject.org/wiki/Test_Day:2013-05-21_AnacondaNewUI_Followup looks ready or close to ready 15:37:13 #info NewUI_Followup test day also has some useful test cases for anaconda that we should adopt into the matrices (though we need to revise them into correct format) 15:38:08 if people can try to spread the word a bit about the test day that'd be great 15:39:22 * jreznik is going to do some marketing over our asocial networks groups 15:39:38 hehe 15:39:45 i'll post in my usual places 15:41:18 https://fedoraproject.org/wiki/Test_Day:2013-05-23_Thermostat looks to have a shell in place, but they haven't written most of their test cases yet 15:41:31 has anyone been working with them on that? 15:42:08 Martix: do you have time to check in and help them complete the page or delay the event, or are you busy? 15:44:14 adamw: I can ask them in ticket on their status 15:44:51 it might be best to try a bit more proactively to get in touch, at least if they don't respond there 15:45:16 then poke them on IRC :-) 15:48:19 yeah, he's not online atm 15:48:35 #action martix to co-ordinate with Thermostat test day runners to make sure they'll be ready or postpone the event 15:49:23 #info https://fedoraproject.org/wiki/Test_Day:2013-05-23_Thermostat page shell is in place, but many test cases missing 15:49:29 #topic Open floor 15:49:34 anything for open floor, folks? 15:49:50 we have a QA calendar 15:49:54 https://apps.fedoraproject.org/calendar/QA/ 15:50:04 not much on it ATM 15:50:13 so, there's a new bugzilla version. 15:50:58 one change I thought I would mention because I bet lots of people will find out about it soon: needinfo bugs will now mail reminders to the suppy the info. ;) 15:51:08 yep 15:51:11 that'll be fun 15:51:44 weekly. 15:51:55 yeah 15:52:17 #info Bugzilla has been updated to 4.4. Among the notable changes: longer alias names, and needinfo bugs will now email the 'needinfo'd party weekly 15:54:11 on the up side, we can use the intended alias names for freeze exception bugs now 15:54:30 (F19BetaFreezeException not F19BetaFreezeExcept for e.g.) 15:54:37 so i'll change that 15:54:40 can the calender app be hooked up to various mail clients and google calender etc or is this just completely standalone app? 15:54:55 ( which makes it totally useless ) 15:54:57 Viking-Ice: it can export ICS at least, which you can import into anything else 15:55:01 i don't know if it has caldav support 15:55:29 adamw: just a note, I don't see any devels on Anaconda Test Day, only QA 15:55:42 maybe we would be better of just to use google calender instead 15:55:53 yes, you can hook it up to anything that uses ical... google calendar can just fine. 15:55:53 which hooks up to tablets and phones 15:55:57 Martix: yeah, i'll try and get bcl to come along 15:56:04 I think mkolman would volunteer 15:56:25 Martix: or, yeah, if there's gonna be a lot of testing in brno, someone from brno 15:56:41 mkolman: feel free to write yourself as developer on this page: https://fedoraproject.org/wiki/Test_Day:2013-05-21_AnacondaNewUI_Followup 15:56:42 Viking-Ice: and of course Skynet :) 15:57:11 nirik: do we have a QA fedocal yet? tflink was working on it with you, right 15:57:23 adamw: yes, as he noted above it's there now. ;) 15:57:34 you can also subscribe to all calendars 15:57:48 goddamnit, i'm not drunk enough yet. 15:58:11 #info QA has a calendar in the fedocal system at https://apps.fedoraproject.org/calendar/QA/ 15:58:23 are we planning to migrate the schedule stuff to fedocal? F20? F21? 15:58:53 jreznik: ^^^ 16:00:29 so what's up with version 20 being offered in bugzilla when we have not even branched yet? 16:00:35 yeah, I dunno about that 16:00:38 adamw: it's uploaded there 16:00:40 might be fallout from the 4.4 migration 16:00:59 adamw: in fedocal - not ideal right now, I have to take a loko 16:01:32 oop, we're over time 16:01:49 we're doing blocker review in #fedora-blocker-review next, though I forgot to announce it (sorry) 16:01:50 * nirik noticed the 20 as well, I didn't add it, perhaps we should remove it? 16:01:57 nirik: probably 16:02:05 I think that happened before the migration (at least the downtime), as I noticed that about 2 days ago. 16:02:05 nirik: yes please 16:02:36 #action nirik to remove F20 from bugzilla for now 16:03:01 wont removing it affect bugs if assigned to that? 16:03:16 they should just get moved to rawhide, that shouldn't be hard 16:04:57 yep, rawhide and I have a script if it will be needed (not sure it still works with a new bz) 16:05:42 i guess you'll find out :) 16:05:50 ok, let's all head over to #fedora-blocker-review for some blocker review 16:05:53 thanks for coming, folks 16:06:57 * nirik looks to see if there's actually any in it yet. 16:07:21 #endmeeting