From Fedora Project Wiki

< QA‎ | Meetings

Attendees

  • adamw (75)
  • cmurf (67)
  • Viking-Ice (26)
  • handsome_pirate (12)
  • satellit (11)
  • roshi (9)
  • zodbot (5)
  • spstarr_work (5)
  • kparal (5)
  • jreznik (5)
  • tflink (3)
  • robatino (3)
  • drago01 (2)
  • nirik (1)
  • mkrizek (1)
  • pjones (1)
  • pschindl (1)

Agenda

  • Previous meeting follow-up
  • Fedora 20 Beta status
  • Test Days
  • Open floor

Previous meeting follow-up

  • roshi to write up recap of the graphics test week - this was done

Fedora 20 Beta status

Open floor

N/A

Action items

  • viking-ice to check his possible resize issue in at least TC6 (ideally RC2)

IRC Log

adamw #startmeeting Fedora QA meeting 16:00
zodbot Meeting started Mon Nov 4 16:00:48 2013 UTC. The chair is adamw. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00
zodbot Useful Commands: #action #agreed #halp #info #idea #link #topic. 16:00
adamw #meetingname fedora-qa 16:00
zodbot The meeting name has been set to 'fedora-qa' 16:00
adamw #topic Roll call 16:00
adamw so, i'm around for once :) 16:01
cmurf shocking 16:01
* nirik is lurking around in the back 16:01
* roshi is here 16:01
* tflink is quaking in his boots due to presence of adamw 16:01
* kparal says hi 16:01
* jreznik needs a few more minutes ;-) 16:01
* Viking-Ice is also lurking in the back 16:01
* mkrizek is here 16:01
* cmurf has coffee but has to drive to ABQ in 3 hours, not here for long. 16:01
adamw so while roll call's going on - my apologies for disappearing last week, and the shim snafu 16:03
adamw i'd told so many people i was going on vacation i was kinda assuming everyone knew, but turns out it was all IRL and no-one here was aware. and shim 0.5 going in ahead of votes was my bad, sorry again 16:03
* pschindl is here 16:03
cmurf badaboom 16:04
* handsome_pirate waves 16:05
* handsome_pirate isn't on drugs this week 16:05
tflink adamw: no worries, it happens 16:05
adamw handsome_pirate: then all is lost' 16:05
tflink handsome_pirate: DARE finally worked for someone? 16:05
handsome_pirate heh 16:05
handsome_pirate I'm off the vicodan 16:05
cmurf kparal: Fricaceed = stewed and served in sauce. As in dead. Kaput. Axed. Killed. Squashed. Squished. Strangled. 16:05
handsome_pirate I actually don't remember last week very well 16:06
handsome_pirate Especially last Monday 16:06
kparal cmurf: thanks. now we need to teach google to know that as well 16:06
cmurf kparal: well i was trying to go for obscure as well as amusing if not weird 16:06
adamw alrighty folks 16:07
adamw #topic Previous meeting follow-up 16:07
adamw let's see, doing this by the seat of my pants... 16:07
adamw "roshi to write up recap of the graphics test week" 16:07
adamw roshi? 16:07
roshi done :) 16:08
* roshi finds link 16:08
adamw cool, did you use my super-lame math stuff? :) (haven't had a chance to read the list yet) 16:08
roshi http://scriptogr.am/roshi/post/graphics-test-week-recap 16:08
roshi I used the recap script - if that's what you mean 16:09
pjones kparal: cmurf: "fricassee" will probably get you better google results 16:09
adamw no, for graphics test weeks i have this thing where i calculate the number of bugs getting fixed and stuff 16:09
adamw no biggie, can go back and do it later 16:09
adamw #info "roshi to write up recap of the graphics test week" - this was done, see http://scriptogr.am/roshi/post/graphics-test-week-recap 16:09
roshi cool, I'd like to see that 16:10
adamw ok, anything from last week that isn't obvious from the minutes? 16:11
* handsome_pirate admits to not remembering last week 16:12
cmurf adamw: Possibly that I volunteered you and me to work out an exception policy for Btrfs and LVM thinp. 16:12
adamw oh goody. so, unless i borrowed pirate's drugs, i'm assuming we haven't done that yet. 16:13
cmurf adamw: Maybe we have and you did borrow his drugs. 16:13
Viking-Ice what was that supposed to be about 16:14
Viking-Ice write exception policy for that from the current release criteria 16:14
cmurf yes because in practice when the going gets tough we try to find exceptions anyway 16:14
adamw Viking-Ice: i believe the idea is to update the criteria to only cover certain areas of LVM and btrfs functionality, not implicitly cover everything they're capable of 16:14
Viking-Ice is that something we would apply now or $next_release? 16:15
adamw dunno? /me only just got here :P 16:15
cmurf and as dlehman put it, LVM is the emacs of storage, so not everything LVM can be supported in the installer anyway. So actually, critieria could be modified. 16:15
adamw i suppose now, if installer team considers it impractical to cover the entirety of lvm functionality as blocker stuff now 16:16
cmurf right 16:16
cmurf i think in the short term we keep doing what we've been doing 16:16
Viking-Ice then we just keep it simple btrfs lvm thinp issues wont block the release this cycle 16:16
adamw my initial thought was along the lines that the installer only has to read/interpret stuff it's capable of writing itself, but eh 16:17
* satellit join late 16:17
adamw we can always do proposal/revision through th elist 16:17
Viking-Ice mean either anaconda supports this stuff properly or they dont we cant be running around every release cycle to adjust our criteria based on their whims 16:17
cmurf Viking-Ice: well no it's more complicated than that it depends on the issue because i don't think it's appropriate to have bugs that qualify for e.g. final block violations, to be available in Guided partitioning where people get into trouble easier 16:17
cmurf Viking-Ice: on the other hand, we can't do major changes between beta and final because that risks more bugs 16:18
cmurf hence an exception that actually allows some blatantly obvious and subjective judgement calls 16:18
Viking-Ice I think it's best that we avoid "exceptions" at all cost in the release criteria 16:18
drago01 adamw: does this code gets rewritten every release? 16:19
drago01 adamw: I mean it should have matured at some point 16:19
adamw Viking-Ice: i don't think it'd be written as a 'this except this' kind of way 16:19
cmurf it would not 16:19
adamw drago01: it gets touched quite a lot every cycle, yeah, but i don't know if the bug that's prompted this is even a new bug 16:19
Viking-Ice drago01, one would have thought so but the feeling of that is no it does not... 16:19
adamw it came up in thinp testing, and we didn't really touch thinp before this release 16:19
cmurf it's not 16:19
cmurf and there are lots of other bugs 16:20
cmurf and there are maturity issues 16:20
cmurf and there are UI/UX concerns adamw brought up on anaconda list over a week ago that relate 16:20
adamw if we're going to get philosophical about it i'd still argue the installer just tries to be too much of a partitioning tool, but hey 16:20
cmurf on that point we agree 16:20
adamw cmurf: which no-one's replying to. 16:20
Viking-Ice adamw, as well as alot of other things geoip wtf 16:20
adamw anaconda-devel-list: where threads go to die 16:20
cmurf i've argued custom needs to be pulled out into its own app, decoupled from the installer 16:20
cmurf it makes the installer unstable 16:20
adamw Viking-Ice: at least geoip's simple and the scope doesn't change over time... 16:21
cmurf and the real custom partitioning should be essentially as you described it, which is what i call an enhanced point and shoot installer 16:21
Viking-Ice adamw, leave that stuff up to firstboot or initial setup 16:21
adamw anyhoo, we're kinda in the weeds 16:22
adamw cmurf/me will post any proposals we come up with to the list so they can be discussed then 16:22
cmurf yeah i think a brief convo with dlehman is in order then proposals for discussion 16:22
adamw #topic Fedora 20 Beta status 16:22
Viking-Ice I personally frown upon any criteria changes in the midst of the cycle and prefer we just tackle these issues on issues bases 16:22
cmurf Viking-Ice: well read the minutes from last week's go/no-go where thinp came up 16:23
adamw so, uh, what is F20 beta status? 16:23
* adamw hasn't checked in since thursday or so 16:23
adamw #chair cmurf roshi 16:23
zodbot Current chairs: adamw cmurf roshi 16:23
kparal we will need another RC https://bugzilla.redhat.com/show_bug.cgi?id=1022206#c13 16:23
kparal also, I can't verify https://bugzilla.redhat.com/show_bug.cgi?id=1023556 16:24
* adamw brb, call of nature 16:24
adamw #chair kparal 16:25
zodbot Current chairs: adamw cmurf kparal roshi 16:25
Viking-Ice I've not yet filed that blocker bug that causes anaconda to crash when resizing ( large partition ) in the default autoparted stuff 16:25
* satellit minor bug? https://bugzilla.redhat.com/show_bug.cgi?id=1004621 16:25
kparal Viking-Ice: 1013586 ? 16:25
cmurf hmm yes resize bugs should not crash but I think the criteria was inteded for resize to not for sure work until final; this has come up before 16:26
adamw yeah, that was the intent 16:27
adamw still, crash when resizing any large partition sounds bad 16:27
cmurf if it crashes in the UI before "begin installation" is less bad by far than after 16:28
Viking-Ice kparal, well not sure this is the same one I think mine was due to altering 3 digit ( 100+GiB ) partition ( altering 2 digit partitions <99GiB seemed to work ) 16:28
cmurf i've been resizing 3 digit windows/NTFS volumes for the Windows matrix test for some time and it's always worked in guided 16:29
cmurf it was busted in custom until TC5 I think, but was definitely working in TC6 16:29
Viking-Ice cmurf, I just took the default chosen standard partition layout and was resizing it ( anaconda puts way to much in / then is needed ) 16:29
cmurf kparal's resize bug that happens with ext4 as well is a different bug instigated by something unique about the partition size that's confusing blivet 16:30
Viking-Ice cmurf, that was tc5 have not retested it with more recent bits 16:30
cmurf Viking-Ice: oh, ok so changing the size within custom, not resizing an existing volume 16:30
cmurf i think that's fixed 16:30
cmurf i haven't run into it with TC6 16:30
cmurf RC2 is being really fussy in qemu/kvm, I keep getting soft lockups on reboots but the installations are clean 16:31
adamw satellit: that one looks quite bad... 16:31
adamw at least worth an FE nomination 16:31
satellit adamw: KDE lives only works once installed 16:31
cmurf oh dear 16:32
satellit ok not my bug someone else should nominate it... 16:32
adamw #info there is at least one blocker known not to be resolved in Beta RC2: https://bugzilla.redhat.com/show_bug.cgi?id=1022206#c13 16:32
satellit cannot update live 16:32
adamw #info we also need information on https://bugzilla.redhat.com/show_bug.cgi?id=1023556 from dlehman 16:33
adamw #info https://bugzilla.redhat.com/show_bug.cgi?id=1023556 may possibly be worth considering for FE or blocker status 16:33
adamw #action viking-ice to check his possible resize issue in at least TC6 (ideally RC2) 16:34
adamw think that ought to cover everything we have? 16:34
jreznik jgrulich as far as I know is going to be away for a few days, just fyi for the plasma-nm one 16:34
adamw jreznik: sounds to be more in sddm than plasma-nm itself 16:34
adamw (i'm assuming it's a logind issue) 16:34
cmurf doesn't seem like it's consistent or it'd have been a big problem all of october yet nothing in the comments during that time 16:36
adamw true 16:36
spstarr_work hello 16:37
satellit I saw it in testing https://fedoraproject.org/wiki/Test_Results:Fedora_20_Beta_RC2_Desktop?rd=Test_Results:Current_Desktop_Test 16:37
cmurf someone volunteer to boot the current KDE live desktop and see if they have an internet connection? 16:37
adamw cmurf: it would be wireless-only, as I read it 16:38
roshi I'll do it - just let me d/l it 16:38
cmurf oh well isn't that normal? 16:38
satellit wireless works in other DE's live 16:38
cmurf i never have wireless working from live desktop because the proprietary firmware isn't installed 16:38
roshi really cmurf? 16:39
cmurf really 16:39
satellit and in anaconda 20.25.5-1 here 16:39
roshi it's always worked for me from live images 16:39
spstarr_work cmurf: I can test this 16:39
cmurf now, if i create a litd usb stick of live, with an overlay, and install the firmware, i persistently get working wireless between boots 16:39
adamw cmurf: the bug isn't to do with firmware afaict 16:39
adamw and most adapters don't need it these days 16:39
spstarr_work wifi? 16:39
* satellit permissions to connect? 16:40
satellit dbus? 16:40
cmurf well i only have Macs here that have Broadcom adapters and even the 2012 model has proprietary firmware required 16:40
cmurf so that may be why my wireless expectations are so low 16:40
spstarr_work i can boot my W500 in a KDE Live CD 16:40
spstarr_work but its a intel wifi chip 16:40
Viking-Ice cmurf, I had a dell with that broadcome problem but that hw has died on me so no worries no more about that ;) 16:41
satellit system 76 here with no proprietary drivers 16:41
adamw heh 16:41
adamw okay, we should ook into it a bit more i guess 16:41
adamw so anyone have any worries about 20 beta that haven't been covered yet? 16:41
cmurf no 16:42
cmurf well not more than the usual 16:42
* cmurf is out for the rest of this week at a conference, 50/50 chance of checking in 16:44
adamw rogert 16:44
adamw welp, then it looks like it's time for... 16:46
adamw #topic Open floor 16:47
adamw any other business? 16:47
robatino do beta FEs get automatically promoted to final FEs after beta gold? 16:47
robatino see https://lists.fedoraproject.org/pipermail/test/2013-November/118626.html (no response) 16:47
* handsome_pirate would be +1 for that 16:48
* handsome_pirate also thought he responded :( 16:48
* handsome_pirate blames the drugs 16:48
adamw that seems to come up every week 16:48
robatino maybe promote proposed FEs to proposed, and accepted to accepted 16:48
adamw but in practice, no, mainly because we have no mechanism for it and no-one wants to do it. 16:48
cmurf adamw: i forgot to respond to that 16:48
Viking-Ice well we should just skip promotion if we are going down that road 16:48
cmurf i think alpha to beta automatic is OK 16:48
adamw i wouldn't be opposed to it if someone wanted to write a way to do it and look after it. 16:49
cmurf beta to final is risky 16:49
Viking-Ice as in and FE remains an FE through milestones 16:49
jreznik and also fe that could be still ok for beta could cause more issues for final in some cases 16:49
cmurf jreznik: exactly my thought 16:49
Viking-Ice cmurf, FE are FE we dont have to pull them in 16:49
cmurf Viking-Ice: yes but they're allowed in, and can cause an RC to blow up 16:49
jreznik cmurf: but allows it 16:49
handsome_pirate It seems that accepted should be auto-promoted 16:50
cmurf so it's like, yeah we can regress but it can be a problem 16:50
jreznik cmurf: exactly my thought :) 16:50
cmurf if anything 16:50
cmurf FE's should be done by the last TC 16:50
Viking-Ice cmurf, but since people have started to pull in proposed FE regardless if they have been approved FE on blocker bug meetings why bother with this process et all 16:50
cmurf and then no more 16:50
adamw Viking-Ice: like i said that was a mistake 16:50
cmurf Viking-Ice: yes that horse has been beated to death and apologized from the grave at least two or three times now 16:50
adamw not intended to be a new policy or something 16:51
cmurf adamw: maybe we need to flog you, strip you, put you in the middle of a busy intersection and painted blue 16:51
Viking-Ice adamw, cmurf when and where ? 16:51
adamw earlier in this meeting, for a start, and in the bug report 16:51
cmurf Viking-Ice: adam apologized in the bug, and in here, and maybe on a list 16:51
* cmurf votes against flogging but in favor of painting blue 16:52
handsome_pirate tar and feather 16:53
cmurf overdone 16:53
* adamw checks if he has any more vacation time left 16:53
Viking-Ice ah it's there buried in comments 16:53
cmurf right, buried with all the other buried comments 16:53
cmurf okay! so we're done? 16:53
adamw i don't think you'd have liked it if i used non-existent secret Red Hat powers to make it bright yellow and bold, would you? :P 16:53
Viking-Ice well thunderbird craps it'self more then usual on large inboxes 16:54
cmurf adamw: gold plated and polished 16:54
Viking-Ice adamw, honestly I thought you had done this to try to not slip beta a week again 16:54
Viking-Ice due to release pressure 16:54
adamw Viking-Ice: nope, it wouldn't have blocked beta in the first place 16:54
cmurf well that didn't work out very well 16:54
adamw honestly i was just half-assing it before going on vacation 16:54
cmurf half asser 16:55
adamw 'oh shit i have to throw together a new TC request, this one this one this one DONE', screwed it up. 16:55
Viking-Ice well there was that 50/50% chance it would go un-noticed 16:55
adamw yeah, if it had actually worked no-one would've given a crap :P 16:55
cmurf 0% chance something not bootable would have gone unnoticed 16:55
cmurf ok wrap? 16:56
adamw sounds lik eit 16:56
adamw thanks for coming, folks 16:56
adamw #endmeeting 16:57

Generated by irclog2html.py 2.12.1 by Marius Gedminas - find it at mg.pov.lt!

Retrieved from "https://fedoraproject.org/w/index.php?title=QA/Meetings/20131104&oldid=360123"