15:03:36 #startmeeting Fedora QA meeting 15:03:36 Meeting started Mon May 7 15:03:36 2012 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:03:36 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:03:39 #meetingname fedora-qa 15:03:39 The meeting name has been set to 'fedora-qa' 15:03:43 #topic roll call 15:03:50 oy! roll! c'mere! 15:03:54 * mkrizek is here 15:04:00 * satellit_ listening 15:04:07 * Cerlyn is here 15:04:30 * tflink is here 15:05:16 alrighty 15:05:26 now, chaaaarge 15:05:41 #topic previous meeting follow-up 15:05:49 we just had one here - "pschindl to poke gnome-boxes devs about Thursday's test day" 15:06:01 since that got postponed (again?) to this week, i guess it happened. 15:06:36 * kparal is in afk mode 15:07:06 he pinged me last week about the test day 15:07:26 I guess that there was some miscommunication on who was actually going to write the test cases and it didn't look as if everything was going to be done on time 15:07:58 I haven't heard anything else since we moved the test day back a week, though 15:07:59 so they pushed it out again, okay. 15:08:19 #info pschindl followed up on Boxes test day, event was not prepared in time so was pushed out another week, needs checking up again 15:08:29 #action adamw to check in on Boxes test day once more 15:08:38 #topic Fedora 17 Final status/planning 15:08:53 so...we're supposed to be rolling an RC tomorrow. this is looking somewhat unlikely =) 15:09:04 of course, the blocker list is https://fedoraproject.org/wiki/Current_Release_Blockers 15:09:28 * tflink has a list of blockers that need to be retested w/ tc3 15:09:40 hasn't finished fleshing out the 'how to test' part of it, though 15:10:13 14 blocker and NTH bugs that need to be retested w/ TC3 if they haven't been already 15:10:51 shall we go through the proposed blockers that weren't covered on friday? 15:11:20 have they moved much? I didn't see much activity on bz over the weekend 15:11:21 * pschindl is here (train has delay) 15:12:00 the list seems to have re-populated, though 15:12:12 there's 8 i can see 15:12:24 pschindl: hiya, you didn't miss a lot, we covered your action item 15:12:41 so I'm late :( 15:13:21 adamw: any additional information needed? 15:13:25 pschindl: nope, it was fine 15:13:30 okay, so let's go for a mini blocker-review 15:13:41 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=819563 15:14:12 i don't see a huge amount of info here, kparal... 15:14:14 I reported this just right now 15:14:22 the logs are there 15:14:42 this sounds like another flavor of the time bug we saw earlier 15:14:51 WRT fsck errors 15:15:03 I need someone to re-test 15:15:04 it seems like it's having some problem with an existing root partition / VG? 15:15:13 I used "use all space" 15:15:17 there's errors about /dev/mapper/vg-lv_root in storage.log and program.log 15:15:29 tflink: which bug's that? 15:15:42 * tflink doesn't remember off hand, searches 15:17:32 .bug 811706 15:17:36 tflink: Bug 811706 fsck errors during install from livecd if system time is too far behind - https://bugzilla.redhat.com/show_bug.cgi?id=811706 15:18:07 I doubt that they're closely related - the timing of the error message just sounded familiar 15:19:06 my VMs don't have system time shifted, I just checked 15:19:28 those fsck errors are different, anyways 15:19:59 sorry, dealing with a medical emergency 15:20:15 adamw: I can continue the review if you #chair me 15:20:42 that's weird - anaconda formats the lv then fsck says it's not ext4 when it attempts to mount it 15:21:02 maybe it's mounted while formatting? 15:21:33 doesn't look like it - there's some lv monkeying after formatting 15:21:57 #chair tflink kparal 15:21:57 Current chairs: adamw kparal tflink 15:22:03 can somebody do just a quick install of i686 live in VM? 15:22:08 seems like an anaconda logic error 15:22:20 don't have it here, and dl.fp.o is slow atm 15:22:20 yeah, I can get one started once I download the live 15:22:40 anyway, shall we agree needs more info and move on? 15:22:45 let's just wait a day until somebody confirms or not 15:22:53 propose #agreed need more info to determine state of 819563 15:22:56 yeah, sounds liek a plan to me 15:22:58 ack 15:23:06 ack 15:23:07 #agreed need more info to determine state of 819563 15:23:15 #topic https://bugzilla.redhat.com/show_bug.cgi?id=819371 15:23:25 seems like an obvious blocker to me (note: calligra is the new name for koffice) 15:23:34 so these are stock apps on the KDE live, crashing on launch 15:24:06 +1 blocker, patch is coming 15:25:34 proposed #agreed - 819371 - AcceptedBlocker - Violates the following F17 final release criterion: "All applications listed under the Applications menu or category must withstand a basic functionality test and not crash after a few minutes of normal use" 15:25:50 ack 15:26:23 ack 15:26:46 #agreed - 819371 - AcceptedBlocker - Violates the following F17 final release criterion: "All applications listed under the Applications menu or category must withstand a basic functionality test and not crash after a few minutes of normal use" 15:27:23 in retrospect, I shouldn't have done that. Now it's not clear who's driving this thing :) 15:27:25 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=819140 15:27:32 * adamw elbows tflink out of the driver's seat 15:27:44 repoclosure issue, +1 blocker. 15:28:02 * tflink wonders where zodbot is, though 15:28:08 +1 blocker 15:28:30 propose #agreed 819140 is accepted as a blocker per criterion "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (DVD) install" 15:29:20 ack 15:29:26 ack 15:29:31 ack 15:29:46 #agreed 819140 is accepted as a blocker per criterion "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (DVD) install" 15:30:08 there's two more repoclosure bugs, let's do those quick 15:30:16 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=819139 15:30:28 tflink: zodbot never seems to work when we do a blocker review in here not -bugzappers 15:30:35 ack ack 15:30:46 propose #agreed 819139 is accepted as a blocker per criterion "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (DVD) install" 15:30:52 ack 15:30:58 #agreed 819139 is accepted as a blocker per criterion "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (DVD) install" 15:31:03 repoclosure bugs are automatic acks I think 15:31:04 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=819138 15:31:10 propose #agreed 819138 is accepted as a blocker per criterion "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (DVD) install" 15:31:17 i don't think we have process for that, but i've done it before... 15:31:49 ack 15:31:53 #agreed 819138 is accepted as a blocker per criterion "There must be no file conflicts (cases where the files in some packages conflict but the packages have explicit Conflicts: tags are acceptable) or unresolved package dependencies during a media-based (DVD) install" 15:31:55 ack 15:32:02 satellit_: please quit ccing yourself to bugs, it's giving me inflight collisions =) 15:32:24 k 15:33:11 satellit_: it's okay, just kidding 15:33:13 adamw: says the man who has a habit of updating bugs while we talk about them 15:33:16 i may have overridden some of them though 15:33:17 tflink: heheh 15:33:23 three to go 15:33:34 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=818935 15:33:53 hum, didn't we talk about this friday? but i don't see a record 15:33:57 it's kparal 15:34:04 kparal's mysterious 'someone else is logged in' bug 15:34:13 I think it's a split off of the two issues in the other bug 15:34:24 so it's the same issue, just separated out into a new bug 15:34:33 well, part of the same issue 15:34:40 I still believe it's directly connected to https://bugzilla.redhat.com/show_bug.cgi?id=814690 15:34:51 and that one bug received some development progress 15:35:09 halfline found some issues in systemd 15:35:54 but 818935 is sooo hard to reproduce 15:36:50 i'm still probably -1 blocker if it's so hard to hit. though actually, i think i hit it yesterday. 15:37:20 I think it would be ok to downgrade it to NTH 15:37:24 i was very, very tired, so i'm not going to commit to everything. but afair, i got off the plane, booted up my system, ran yum update, read some mails and stuff, and went to shut down, and got the 'other people are logged in' dialog. oh, wait. i might have had a facebook open. 15:37:31 * adamw runs facebook as another user. 15:38:28 does it help wrt privacy? :) 15:38:29 so yeah, i guess i'm -1... 15:38:52 yeah, -1 unless it starts happening more often 15:39:09 kparal: basically, that's the idea. i have a 'facebook' user whose firefox profile only ever logs into facebook. my main user is configured to reject everything at all from facebook. read about it on a blog somewhere, so obviously it's a good idea .;) 15:39:47 * tflink makes note to write blog post about how firing people all the time is a bad idea 15:39:54 heh 15:39:58 that's okay, i don't read your blog 15:40:15 ok, what about having it as NTH? any votes? 15:40:16 my plans, they have been foiled! 15:40:20 are we even NTH on this bug, though? i'm not sure it really hits live? 15:40:26 have you ever seen it in a live boot? 15:40:30 hmm 15:40:39 I think the concern is the possibility of someone hitting this before updating 15:40:46 you're right it doesn't really matter on live 15:40:46 I have hit this bug today 15:40:52 I'm +1 on NTH 15:41:02 pschindl: have you hit it before? 15:41:15 adamw: long time ago 15:41:19 on livecd you don't really care about it, liveuser is in wheel without password 15:41:26 it happened to me twice 15:41:32 as I remember 15:41:33 pschindl: so like kparal, you see it very occasionally 15:41:37 kparal: oh right 15:41:58 * kparal takes back NTH request 15:42:18 let's -1 blocker it then 15:42:36 and come on, I have to go soon 15:42:53 ok 15:43:10 i think i am late 15:43:12 propose #agreed 818935 is rejected as a blocker as it is fixable with an update and seems to occur very infrequently, some testers have never seen it 15:43:41 ack 15:43:55 ack 15:44:06 #agreed 818935 is rejected as a blocker as it is fixable with an update and seems to occur very infrequently, some testers have never seen it 15:44:33 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=819492 15:44:41 two doozies coming up, looks like :/ 15:45:23 I hit this one when I tried to copy a movie for my girlfriend. it never worked for her 15:45:23 so this would be "All known bugs that can cause corruption of user data must be fixed or documented at Common F17 bugs" (final) 15:45:38 seems pretty clearly +1 for me 15:45:47 +1 15:45:58 right, +1 15:46:18 +1 15:46:34 +1 15:47:05 propose #agreed 819492 is accepted as a blocker per criterion "All known bugs that can cause corruption of user data must be fixed or documented at Common F17 bugs" 15:47:12 ack 15:47:17 ack 15:47:21 #agreed 819492 is accepted as a blocker per criterion "All known bugs that can cause corruption of user data must be fixed or documented at Common F17 bugs" 15:47:33 #topic Mini blocker review: https://bugzilla.redhat.com/show_bug.cgi?id=818378 15:48:58 so, this is the 'can't install the i686 DVD' bug, basically 15:49:00 AIUI anyway 15:49:09 has everyone who's tried an i686 DVD install hit this? or is it more chancy? 15:49:21 this is kinda obscure bug 15:49:32 * satellit_ i have on vb 15:49:37 i tried every kind of 32-bit install and hit it every time 15:49:43 * tflink is still downloading i686 media 15:49:43 me too 15:49:58 akscram: '32 bit installs always explode' isn't exactly obscure =) 15:50:01 working god for me 15:50:16 I tried to install to i386 laptop and grub wasn't installed correctly everytime 15:50:17 akshayvyas: did you use an x86_64 or i686 image? 15:50:37 adamw:i686 as my old p4 supports 15:50:45 akshayvyas: with tc3? 15:50:51 yep 15:51:06 still, even if one person got success, if two get multiple failures, that's bad enough to be blocker for me... 15:51:43 it seems odd to be installing a 32bit distro to a DL560, though 15:51:58 * satellit_ has to do with OLPC tree? 15:52:00 IIRC, they didn't come with anything but 64bit capable processors 15:52:20 but I could be remembering wrong 15:53:15 yes, I am remembering wrong - ignore me 15:53:15 not sure we need to get too specific on this one, since grub and kernel folks are both on it 15:53:32 adamw: +1 15:53:36 yeah, sounds like blocker material for now 15:53:42 I'm +1 15:53:59 +1 15:54:38 propose #agreed 818378 is a blocker per criterion "In most cases (see Blocker_Bug_FAQ), a system installed according to any of the above criteria (or the appropriate Beta or Final criteria, when applying this criterion to those releases) must boot to the 'firstboot' utility on the first boot after installation, without unintended user intervention, unless the user explicitly chooses to boot in non-graphical mode." - this breaks functional installation 15:54:38 in lots of tested cases 15:54:53 propose #agreed 818378 is a blocker per criterion "In most cases (see Blocker_Bug_FAQ), a system installed according to any of the above criteria (or the appropriate Beta or Final criteria, when applying this criterion to those releases) must boot to the 'firstboot' utility on the first boot after installation, without unintended user intervention, unless the user explicitly chooses to boot in non-graphical mode." - this breaks 32-bit install 15:54:56 damn that criterion. 15:55:13 maybe we should stick in another that just says 'installed systems should damn well boot' for reasons of shortness. :) 15:55:14 unlcess our understanding changes, ack 15:55:34 ack 15:55:38 ack 15:55:39 ack 15:56:30 #agreed 818378 is a blocker per criterion "In most cases (see Blocker_Bug_FAQ), a system installed according to any of the above criteria (or the appropriate Beta or Final criteria, when applying this criterion to those releases) must boot to the 'firstboot' utility on the first boot after installation, without unintended user intervention, unless the user explicitly chooses to boot in non-graphical mode." - this breaks 32-bit install 15:56:34 okay, that's the lot 15:56:59 #topic Fedora 17 Final status/planning 15:57:03 so, back on the general topic... 15:57:11 #info tflink is planning a 'bugs that need re-testing' summary 15:57:14 thanks for that tflink 15:57:41 will be sending that out shortly 15:57:47 anything else on strategy? i'm trying to deal with the whipping devs into line side of things now, aside from that all we can do is re-test fixes i think 15:58:13 make sure that we get the test matrix filled out 15:58:21 but that may already be done 15:58:29 * tflink hasn't checked this morning 15:58:34 point 15:58:50 the area that I'd like to see focus on is USB installation media 15:58:51 it's pretty close, but still some more 15:59:08 #info we still need to complete the TC3 matrix to find any other lurking blockers, no point waiting till the RC 15:59:16 there have been a lot of changes there recently and I'm having a hard time keeping track of what works and what doesn't 15:59:21 tflink: bcl and I are planning to work on that 15:59:40 tflink: the high-level overview is that 17 should now be much like 16 only better (dd'ed images should boot via efi for e.g.) 15:59:51 and dd'ed DVD images should find the packages 16:00:02 all the malarkey about multiple partitions has gone, so just pretend it never happened 16:00:05 stuff like this makes me think I'm not the only one who's lost: http://ask.fedoraproject.org/question/1614/why-does-a-fedora-usb-made-from-the-dvd-iso-still 16:00:38 well, not lost but it would be nice to get everything straight soon 16:00:44 tflink: that's abut fc 16 16:00:48 exactly 16:01:04 but most of the responses are fixes for F17 16:01:12 yeah i see this since fc 15 16:01:17 tep 16:01:20 yep 16:02:58 it's always been the case. f17 is the _first_ release where it should work. 16:04:02 okay 16:04:07 so, i guess we all know what we need to do 16:04:46 #topic Upcoming QA events 16:05:04 so as mentioned the RC compose is scheduled for tomorrow but that's not looking terribly likely...but let's do all we can to clear out the blocker list 16:05:10 blocker review meeting on Friday, of course 16:07:00 when is go/no-go? 16:07:05 next week? 16:07:45 next tuesday 16:08:16 yeah 16:08:20 so it's not on this week's list 16:08:36 oh boy, lots to do before then 16:08:37 other thing is the Boxes test day, i already gave myself an action item for that; looks like it's still not in shape. 16:08:49 is boxes working yet? 16:09:00 * tflink still hasn't tried it 16:10:00 we are working on it 16:10:09 me either. we put the anaconda memory requirement back down to 512 for the last build, which should help. 16:10:10 but lot of things still isn't working 16:11:04 * kparal leaves 16:13:58 * akshayvyas is leaving,have a great day adamw and tflink 16:13:59 okay. so, anyway, i'll cover that 16:14:09 seems like we're running over 16:14:13 is there much autoqa news, tflink? 16:14:22 nope, we've been testing F17 16:14:37 okay 16:14:41 #topic AutoQA update 16:14:50 #info there is no news, autoqa team has been working on F17 validation 16:14:56 #topic open floor 16:15:01 anyone have anything for open floor? 16:16:24 nothing here 16:18:34 ok 16:18:37 thanks for coming, all 16:18:44 sorry for the overrun (pschindl, hope kparal doesn't mind :>) 16:18:49 #endmeeting