QA/Meetings/20110815

From FedoraProject

Jump to: navigation, search

Contents

Attendees

  • tflink (130)
  • adamw (106)
  • kparal (21)
  • maxamillion (16)
  • Viking-Ice (11)
  • clumens (10)
  • nb (6)
  • jsmith (6)
  • bcl (5)
  • rbergeron (4)
  • mkrizek (4)
  • dgilmore (3)
  • jskladan (2)
  • zodbot (2)
  • athmane (2)
  • Southern_Gentlem (1)
  • pjones (1)

Agenda

Previous meeting follow-up

  1. Bug #728657 successfully split up into separate bugs by adamw

Current Alpha Blockers

  1. Review of all open blocker bugs conducted: for details, see MeetBot summary

RC3 Testing Status

  1. KDE desktop validation not done
  2. All other testing completed, and bugs identified requiring RC4

Fedora 16 Alpha RC4 ready-ness

  1. Waiting for all fixes to RC3 bugs
  2. Due to all of the rebuilds needed for RC4, we should avoid carrying forward results from RC3

Open Discussion

  1. AutoQA 0.6.1 released
  2. AutoQA 0.7.0 in planning - 0.7.0 tasks

Action items

  1. robatino, adamw or tflink - do RC4 announcement once the compose is complete

IRC Log

tflink #startmeeting 2011-08-15 Fedora QA Meeting 15:00
zodbot Meeting started Mon Aug 15 15:00:35 2011 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00
tflink #topic roll call 15:00
* kparal here 15:00
* athmane is here 15:00
* mkrizek is present 15:01
tflink kparal, athmane, mkrizek, pschindl: welcome 15:01
adamw yo 15:02
adamw sorry folks, was on hairball patrol... 15:02
tflink adamw: that sounds like fun 15:02
* tflink waits another couple minutes to see if we get more people 15:03
adamw sorry for no meeting announcement, either, was focusing on alpha 15:04
tflink no worries, I was in the same boat :) 15:05
* tflink forgot until this morning 15:05
tflink ok, let's get this started 15:05
tflink #topic Previous Meeting Follow-Up 15:05
tflink AFAIK, we don't have anything to follow up on 15:05
tflink is there anything I'm missing? 15:05
adamw were there any #action items last week? 15:06
* jskladan lurks in 15:06
adamw the 'previous meeting followup' is mostly a checkin to make sure #actions were...actioned 15:06
tflink adamw to split up 728891 into separate bugs and assign them appropriately 15:06
adamw done! 15:06
adamw that was the file conflict bug, it's been solved for a while now 15:06
tflink #info 728891 successfully split up into separate bugs 15:07
tflink that's the only one I'm seeing from last week 15:07
adamw okay 15:07
tflink on to the next topic 15:07
tflink #topic Current Alpha Blockers 15:07
tflink #link https://fedoraproject.org/wiki/Current_Release_Blockers 15:08
tflink we have 2 new proposed blockers that it'd be nice to go over quick 15:08
tflink #link https://fedoraproject.org/wiki/Fedora_16_Alpha_Release_Criteria 15:09
tflink #topic https://bugzilla.redhat.com/show_bug.cgi?id=730415 15:09
tflink #info kickstart with user --name=blah results in traceback 15:09
tflink talking with bcl on friday, this should be fixed soon 15:10
tflink IIRC, a patch has been posted to anaconda-devel 15:10
adamw this doesn't really make any blocker criteria 15:10
tflink nope, probably not 15:10
adamw we don't have a specific 'kickstart install should work' criterion but if we did i bet it'd be beta 15:10
tflink proposed #agreed 730415 - RejectedBlocker, AcceptedNTH - Does not hit any alpha release criteria but it is kind of ugly. Tested fix would be accepted. 15:11
adamw i'm not sure about nth at this point 15:11
tflink ack/nack/patch? 15:11
adamw that'd involve respinning anaconda again to do nothing but fix this... 15:12
Southern_Gentlem can you stay its a beta blocker 15:12
adamw do we have an anaconda dev in the house? 15:12
* tflink thought that there was another bug that they wanted to fix 15:12
adamw Southern_Gentlem: it can be proposed as one and then we'd discuss it on friday 15:12
tflink oh, it's proposed nth 15:12
tflink proposed #agreed 730415 - RejectedBlocker - Does not hit any of the alpha release criteria - repropose as beta blocker 15:13
clumens THUD 15:14
adamw ack for that 15:14
adamw clumens: so we're split on whether 730415 should be nth 15:14
adamw is there any other reason to rebuild anaconda at this point? 15:14
clumens no 15:14
tflink 729599 would be the other one, but that's nth too 15:15
clumens right 15:15
tflink and we were having trouble reproducing that on friday 15:15
tflink well, in the way that we were fearing anyways 15:15
tflink any other votes? 15:16
adamw so...yeah, i'm kinda -1 nth at this point. anyone else have a vote? 15:16
clumens i'm fine with kicking it on down to final 15:16
* tflink is taking that as an ack so that we can move on 15:16
tflink #agreed 730415 - RejectedBlocker - Does not hit any of the alpha release criteria - repropose as beta blocker 15:16
tflink #topic https://bugzilla.redhat.com/show_bug.cgi?id=729500 15:17
tflink #info Error while installing updates on Fedora 16 Alpha RC3 15:17
tflink adamw: any luck reproducing this? 15:17
adamw not seen it again 15:17
adamw and no-one else has either 15:17
adamw so let's reject it for now 15:17
tflink rejected blocker? 15:17
adamw and if others hit it for rc4...add it again 15:17
tflink proposed #agreed - 729500 - RejectedBlocker - Have not been able to reproduce, rejecting as blocker. Re-propose if it shows up again. 15:18
tflink ack/nack/patch? 15:18
adamw ack 15:18
* tflink would prefer to have 3 votes ... 15:19
tflink #agreed - 729500 - RejectedBlocker - Have not been able to reproduce, rejecting as blocker. Re-propose if it shows up again. 15:19
* maxamillion is here ... late but here 15:19
tflink maxamillion: welcome 15:20
tflink #topic https://bugzilla.redhat.com/show_bug.cgi?id=730438 15:20
maxamillion tflink: thanks :) 15:20
tflink #info SSL CA errors when reporting an installer bug to bugzilla 15:20
* adamw notes once more for the record that you don't need to pass an exam to vote on blockers 15:20
tflink this is something that I hit on friday - it was covered up by another libreport bug 15:20
adamw being an idiot on the internet with an opinion is enough 15:20
adamw right, nice catch tflink 15:20
maxamillion adamw: lol 15:20
pjones but maybe we should start? 15:21
tflink #info fix available, test boot.iso is being built for karma now 15:21
adamw if we haven't formally voted yet: +1 blocker, prevents bug reporting from anaconda (again) 15:21
* maxamillion isn't entirely sure how to respond that 15:21
adamw maxamillion: i recommend blithely ignoring it 15:21
tflink proposed #agreed - 730438 - AcceptedBlocker - he installer must be able to report failures to Bugzilla, with appropriate 15:21
tflink information included. 15:21
adamw ack 15:21
maxamillion adamw: rgr that 15:21
tflink proposed #agreed - 730438 - AcceptedBlocker - he installer must be able to report failures to Bugzilla, with appropriate information included 15:21
tflink ack/nack/patch ? 15:22
tflink don't make me start calling people out by name :) 15:22
adamw ack. with an extra t. 15:22
adamw (no, not 'tack'.) 15:22
tflink kparal, jskladan: votes? 15:22
rbergeron attack? 15:22
jsmith ACK 15:23
tflink finally :) 15:23
adamw ubuntu is attacking?! 15:23
adamw man the defences! 15:23
tflink #agreed - 730438 - AcceptedBlocker - he installer must be able to report failures to Bugzilla, with appropriate information included 15:23
rbergeron OMG 15:23
tflink OK, proposed NTH time 15:23
* Viking-Ice joins in 15:23
* kparal was slow to vote 15:23
tflink #topic https://bugzilla.redhat.com/show_bug.cgi?id=729599 15:23
adamw tflink, to the Negative Review Cannon! 15:23
tflink #info PartitionException: msdos disk labels do not support partition names. 15:23
adamw rbergeron, to the FUD Dispenser! 15:23
rbergeron adamw: only if it has a flip-back yead 15:24
rbergeron head 15:24
adamw you got it 15:24
* bcl pokes his head in 15:24
* maxamillion is pretty confident he missed something 15:24
tflink I think that I'm -1 nth on this one for alpha - the impact doesn't seem to be quite as bad as we first thought 15:24
adamw oh, right, this one... 15:24
jsmith +1 for NTH, -1 for blocker 15:24
jsmith (alpha blocker, that is) 15:24
adamw tflink: is ther news that's not in the bug? 15:24
tflink adamw: it was your testing, you tell me 15:25
adamw ah 15:25
adamw well, i don't wanna rely too much on my little test 15:25
* athmane is sorry, need to go 15:25
tflink I'm not too strongly -1 15:25
tflink athmane: thanks for joining us 15:25
adamw if bcl is sufficiently worried about the impact of this one, i'm still +1 15:25
Viking-Ice this happens only on upgrade right ? 15:25
tflink the theory is that it could happen on clean install, too 15:26
adamw yeah 15:26
Viking-Ice with msdos partition ? 15:26
bcl It happens any time you have a msdos labeled disk with the /boot partition on it 15:26
tflink if you have msdos disk labels 15:26
adamw the theory is that any install to a disk with an msdos disk label, which isn't being entirely reformatted, could be in trouble 15:26
Viking-Ice which we clean out if default partitioning scheme is chosen in anaconda ? 15:26
adamw i tried a test to confirm this and the install worked, but it's entirely possible i screwed something up. 15:26
adamw Viking-Ice: default depends on exactly what's on the disk already, i think 15:27
tflink proposed #agreed - 729599 - Doesn't hit any alpha criteria but could affect clean installs to disks with existing msdos partitions 15:27
tflink ack/nack/patch? 15:27
tflink whoops, that was supposed to say AcceptedNTH 15:27
tflink proposed #agreed - 729599 - AcceptedNTH - Doesn't hit any alpha criteria but could affect clean installs to disks with existing msdos partitions 15:27
adamw bcl: how messy is the fix for this? 15:27
clumens so, i'll be doing an anaconda after all? 15:28
clumens https://www.redhat.com/archives/anaconda-devel-list/2011-August/msg00181.html 15:28
clumens looks straightforward to me 15:28
adamw yeah... 15:28
adamw lemme see, what's the worst that could possibly happen... 15:29
adamw if the boot partition doesn't get a name, would it still work? 15:29
bcl adamw: the fix is clean, I just had to check to make sure the disk label supports the feature before using it. 15:29
adamw (i.e. if somehow that conditional was never satisfied) 15:29
bcl yes, everything would probably be fine. The change was primarily for EFI 15:30
adamw okay... 15:30
Viking-Ice I'm +1 nth if there is no risk of breaking partitioning if there is -1 nth 15:30
adamw i guess, based on the simplicity of the fix, +1 nth 15:30
adamw so yeah, new anaconda 15:30
maxamillion +1 nth 15:30
bcl there is always risk. the question is -- is the risk better to take in Alpha or Beta? 15:30
tflink ok, sounds like we have an agreement 15:30
clumens can do. 15:30
tflink #agreed - 729599 - Doesn't hit any alpha criteria but could affect clean installs to disks with existing msdos partitions 15:30
* tflink will wait for new anaconda build before putting out a test boot.iso 15:31
clumens bcl: i thin this is a low risk change. 15:31
tflink ok, accepted blocker time 15:31
tflink wait, do we need to do this? 15:31
adamw prolly worth reviewing the rpm fix 15:32
tflink eh, there's only one and that'll come up a little later (rpm issue) 15:32
tflink or we can do it now :) 15:32
tflink #topic https://bugzilla.redhat.com/show_bug.cgi?id=728707 15:32
tflink #info on package upgrade RPM is removing empty directories accidentally 15:33
adamw so, good news, there's a 'fix' 15:33
adamw bad news, it involves pulling in a medium-size bunch of package rebuilds 15:33
adamw still, if we gotta do it, we gotta do it 15:33
tflink yep 15:33
Viking-Ice yup 15:33
* adamw is still not entirely sure why panu doesn't think rpm should be fixed. 15:33
maxamillion what are the chances that the fix could break the package rebuilds? or cause $other? 15:34
tflink it sounds like rpmbuild was part of the problem 15:34
adamw the fix in rpm? negligible 15:34
maxamillion adamw: cool 15:34
adamw but in general, every time you rebuild a package even with no changes, there's a small-but-existent chance of it screwing _something_ up 15:34
maxamillion well ... ture 15:35
maxamillion true even 15:35
tflink so do we want to try pushing for a fix in rpm? 15:35
maxamillion bleh .... typing is hard 15:35
adamw i just poked the bug with that. but panu already said the right thing is the rebuilds, and we're short on time 15:35
tflink very true 15:36
maxamillion what would the estimated ETA on the rebuilds be? 15:36
adamw they're all done 15:36
maxamillion oh 15:36
Viking-Ice  ;) 15:36
adamw need karma, but that's about all 15:36
adamw so we should just spin rc4 with the rebuilds of all packages on the dvd, and hope 15:36
maxamillion then +1 to the rpm fix 15:36
* maxamillion likes his package manager to be as bug free as possible 15:37
Viking-Ice adamw, yup 15:37
jsmith +1 to the rpm fix as well 15:37
tflink proposed #agreed - 728707 - unless something new comes up, go with the rebuilds as a fix for the rpm issue and make sure RC4 has the new builds 15:37
Viking-Ice ack 15:38
adamw ack 15:38
tflink #agreed - 728707 - unless something new comes up, go with the rebuilds as a fix for the rpm issue and make sure RC4 has the new builds 15:38
jsmith ACK 15:38
tflink ok, done with the bug review party for now 15:38
adamw great 15:38
tflink #topic RC3 Testing Status 15:38
adamw so, bcl and clumens, if we could get a new anaconda build asap that'd be great, then we can compose rc4 and get to testing 15:38
tflink I'm probably not the best person for this update, anyone care to take it on? 15:39
adamw well, it's pretty straightforward...we covered most everything that needs covering, and hit some bugs that should be fixed in rc4 15:39
adamw i guess the only thing missing is kde desktop validation 15:39
tflink at this point, might as well wait for RC4 15:40
clumens yeah i can do that real quick 15:40
tflink #info still missing KDE desktop validation, may wait for RC4 for those tests 15:40
tflink anything else RC3 related? 15:41
tflink then moving on ... 15:41
adamw any issues anyone's worried about that haven't been reviewed as blockers? 15:41
adamw speak now or forever hold thy peace... 15:41
tflink #topic Fedora 16 Alpha RC4 ready-ness 15:42
tflink As I understand it, we're pretty much ready for RC4 15:42
tflink just waiting on an anaconda build and hopefully some karma 15:42
tflink I assume that the plan is to spin up RC4 today and get testing 15:43
adamw yeah 15:43
adamw definitely 15:43
Viking-Ice yup 15:43
adamw we're pretty tight on time; we need at least alpha validation tests complete by wednesday 15:43
tflink #info next go/no-go meeting is on Wednesday (2011-08-17) 15:43
adamw given all these damn rebuilds landing, and changed anaconda, we should try to avoid relying on pulled-forward rc3 tests as much as possible 15:43
* jsmith agrees with adamw 15:44
tflink #info due to all of the rebuilds needed for RC4, we should avoid carrying forward results from RC3 15:44
adamw note that the Alpha, Beta, Final column on the validation matrix itself is somewhat out of whack - if in doubt, criteria take priority 15:44
adamw (so if something's marked 'Alpha' on the matrix but the matching release criterion is Beta, then it's a Beta test and we should fix the table at some point) 15:45
tflink alpha, beta ... they're all just greek letters :-D 15:45
adamw if robatino's around when rc4 compose is done he'll do the announcing, otherwise myself or tflink will take care of it 15:46
tflink #action robatino, adamw or tflink - do RC4 announcement once the compose is complete 15:46
tflink overall, I think that the message is - be ready for testing. Let's not slip alpha another week 15:47
tflink not that anyone needed reminding 15:47
tflink but I think that about covers it for RC4 stuff until it's released 15:47
tflink any concerns with RC4 that we didn't cover? 15:47
Viking-Ice nope not from me 15:48
tflink OK, I think that covers the agenda that I had in mind 15:48
tflink #topic Open Discussion 15:48
tflink any other topics that should be discussed? 15:48
adamw i don't see anything else major upcoming 15:49
* jskladan needs to catch the bus home. see you tomorrow gang! 15:49
adamw there's a test day slot thursday but it's empty 15:49
adamw cya jskladan! 15:49
adamw which is probably a good thing given all of this 15:49
adamw do we have an autoqa update from anyone? 15:49
kparal do we have an autoqa update at all? :) 15:49
tflink we don't have much to say 15:50
kparal we releases 0.6.1 15:50
adamw have you all beek working on Project Colada? :) 15:50
kparal that fixes some bugs 15:50
kparal *released 15:50
adamw cool 15:50
kparal and planned 0.7 15:50
tflink #info AutoQA 0.6.1 released and deployed to fix some bugs 15:50
kparal https://fedorahosted.org/autoqa/query?status=new&status=assigned&status=reopened&milestone=0.7.0 15:50
adamw sounds like everything's rolling 15:51
adamw what are the major goals for 0.7.0? 15:51
tflink #info AutoQA 0.7.0 has been planned, work is progressing 15:51
kparal we would like to concentrate on the infrastructure for the next release 15:51
kparal resultsdb and staging/testing are the general themes I believe 15:51
kparal adamw: wasn't it Project Coconut? 15:52
adamw it has many names 15:53
kparal I see 15:53
adamw but only one goal 15:53
* kparal imagining it vividly 15:53
kparal any other thoughts on autoqa updates? 15:54
tflink we need a package review (soon to be 2) 15:54
tflink and a sposor for mkrizek 15:54
kparal hongqing says he'll provide a new test for autoqa soon - mediakit_sanity. it should test ISO images of Branched release 15:54
tflink s/sposor/sponsor 15:54
adamw the package review thing is looking like a bit of a roadblock 15:54
adamw i've been distracted by alpha, but one thought i had was to keep it in the family 15:55
kparal there has been some progress for mkrizek I believe. the future now looks brighter 15:55
mkrizek tflink: good news, I might get a sponsor this week hopefully 15:55
adamw we can exchange (or just provide) reviews within the qa group; we have, or should have, enough packagers to do that 15:55
* tflink makes note to remember that 15:55
kparal adamw: the roadblock was the packaging _sponsor_ I believe 15:56
kparal they are scarce 15:56
tflink but it sounds like mkrizek might have found one 15:56
* nb is a sponsor 15:57
kparal here we go, remember that nick, mkrizek :) 15:58
nb mkrizek, have you already found someone? or are you still looking? 15:58
tflink cool, progress :) 15:58
* kparal has dinner on the table. let's speed it up! :) 15:59
mkrizek nb: I might, I have been contacted with one 15:59
adamw two is better than one... 15:59
nb mkrizek, ok, if that doesn't work, let me know 15:59
mkrizek nb: ok, thanks! 15:59
adamw thanks a bunch nb 15:59
nb adamw, no problem 15:59
adamw do we have anyone else who needs sponsoring? 15:59
* tflink was already sponsored but can't quite do reveiws yet 16:00
tflink odd situation 16:00
nb tflink, you can't? 16:00
adamw well, reviews shouldn't be a big deal 16:01
adamw i can do those if no-one else can 16:01
tflink nb: long story short, I got sponsored to take on a to-be-orphaned package and I still have a mentor 16:01
adamw let's not let kparal's dinner go cold... 16:01
tflink adamw: where were you when I was waiting 4 months for a review on py.test? 16:01
tflink  :) 16:01
* kparal already moved his notebook to kitchen, no worries 16:01
nb tflink, oh ok 16:02
tflink anyhow, unless we have other topics I'm setting the #endmeeting fuse for 5 minutes 16:02
adamw tflink: on project colada! 16:02
tflink nb: so its not so much can't as shouldn't until I have more experience 16:02
tflink any volunteers to do the bug updating? 16:04
adamw i can, if no-one else wants to 16:04
adamw dgilmore: i'll do an rc4 recipe update on the trac ticket once the anaconda update is up 16:05
adamw then we can all get to testing... 16:05
tflink adamw: think it's worth the effort to get a boot.iso out for karma on lorax, anacanda etc. ? 16:06
tflink or can we just pull all that in to RC4 without karma 16:06
dgilmore adamw: ok, what about rpm? 16:06
adamw tflink: probably worth a sanity test, yeah. 16:06
adamw dgilmore: the fix for the rpm issue does not appear to be in rpm. 16:06
tflink k, just making sure I wasn't wasting my time :) 16:06
adamw dgilmore: the fix is to rebuild all the affected packages, it seems 16:06
dgilmore tflink: we can pull it in without, but it will not hurt to get it 16:06
adamw dgilmore: so we'll have to pull all those rebuilds. 16:06
dgilmore adamw: fun 16:06
tflink alrighty, thanks for coming everyone 16:07
tflink time to test the crap out of alpha RC4 16:07
* tflink will send out minutes shortly 16:07
tflink #endmeeting 16:07