17:06:14 <roshi> #startmeeting F22-blocker-review
17:06:14 <zodbot> Meeting started Mon Feb  2 17:06:14 2015 UTC.  The chair is roshi. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:06:14 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
17:06:14 <roshi> #meetingname F22-blocker-review
17:06:14 <zodbot> The meeting name has been set to 'f22-blocker-review'
17:06:15 <roshi> #topic Roll Call
17:06:15 * danofsatx taps his foot
17:06:23 <roshi> who's still around?
17:06:27 <danofsatx> me!
17:06:33 <roshi> here, have a chair
17:06:37 <roshi> #chair danofsatx
17:06:37 <zodbot> Current chairs: danofsatx roshi
17:06:40 <danofsatx> pero necesito mas cafe
17:07:05 * pschindl is here
17:07:11 <danofsatx> just a sec while I go find a cup (read: pitcher)
17:07:11 * kparal here
17:07:15 <roshi> you too win a chair!
17:07:22 <roshi> #chair pschindl kparal
17:07:22 <zodbot> Current chairs: danofsatx kparal pschindl roshi
17:08:56 <roshi> alright, onto the boiler plate :)
17:09:07 <roshi> #topic Introduction
17:09:08 <roshi> Why are we here?
17:09:08 <roshi> #info Our purpose in this meeting is to review proposed blocker and nice-to-have bugs and decide whether to accept them, and to monitor the progress of fixing existing accepted blocker and nice-to-have bugs.
17:09:12 <roshi> #info We'll be following the process outlined at:
17:09:14 <roshi> #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting
17:09:17 <roshi> #info The bugs up for review today are available at:
17:09:19 <roshi> #link http://qa.fedoraproject.org/blockerbugs/current
17:09:22 <roshi> #info The criteria for release blocking bugs can be found at:
17:09:24 <roshi> #link https://fedoraproject.org/wiki/Fedora_22_Alpha_Release_Criteria
17:09:27 <roshi> #link https://fedoraproject.org/wiki/Fedora_22_Beta_Release_Criteria
17:09:30 <roshi> #link https://fedoraproject.org/wiki/Fedora_22_Final_Release_Criteria
17:09:43 <roshi> we've got 1 proposed Alpha and 1 proposed Beta
17:09:49 <roshi> #topic (1185999) [Gtk3] Text on various bits of chrome (tab titles, menus, buttons...) is white with GTK+ 3.15.4
17:09:52 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1185999
17:09:53 * adamw brb, call of nature
17:09:55 <roshi> #info Proposed Blocker, gtk3, NEW
17:10:06 <roshi> best to heed it's call....
17:10:08 <adamw> this is borderline blocker for me, i've been living with it here for the last week or so, but it is pretty annoying
17:11:14 <danofsatx> ok, coffee is brewing. We're on proposed #1? I'm not sure...does it actually prevent the use of the browser, or just make it unwieldy?
17:11:16 * nirik agrees with anoying
17:11:25 <pschindl> I think it is more FE. I use it every day with this. It is really annoying, but it can be used.
17:11:26 <nirik> just makes it very hard to read things.
17:12:08 <kparal> pschindl: FE for what release?
17:12:16 <kparal> are we OK releasing Final like this?
17:12:31 <roshi> I don't think it violates the criterion
17:12:46 <pschindl> I'm not ok, but I don't think it should block release.
17:12:46 <danofsatx> I think it violates the polish criteria, but not usability.
17:12:55 <roshi> kparal: for final, I might want this under the catch-all "polish" criterion
17:13:24 <kparal> can you use the browser when you can't read some of its UI?
17:13:44 <kparal> imho it violates the criteria
17:13:47 <roshi> well, I use the keyboard - so the visual of the UI doesn't affect me much
17:13:59 <pschindl> kparal: Yes I can and I do :)
17:14:04 <roshi> but then again, I haven't seen it, either
17:14:16 <danofsatx> mouseover changes the background so you can see the titles,
17:14:16 <roshi> we've got a couple people who are still using it
17:14:20 <kparal> could somebody upload some screenshots to the bug report?
17:14:44 <kparal> roshi: we're talking about the average joe here
17:14:54 <pschindl> But if there is some polish criteria then it should block
17:15:02 <adamw> the most annoying thing for me is the right-click menu
17:15:09 <adamw> kparal: it's hard to take screenshots of menus :/
17:15:16 <kparal> how do you download a file, if not right click?
17:15:22 <kparal> adamw: it's not, set up a delay
17:15:29 <adamw> oh, yeah, always forget that
17:15:37 <kparal> or a video
17:15:39 <adamw> so on the right click menu you can only read the entry that's currently highlighted
17:15:39 <pschindl> I can send some, but not today.
17:15:42 <kparal> ctrl+alt+shift+r
17:15:46 <adamw> so you have to wash down it to read each entry one at a time
17:16:12 <kparal> I'm definitely +1 Final. you can talk me out of the other milestones
17:16:14 <roshi> hrm
17:16:27 <roshi> +1 Final - under polish
17:16:39 <danofsatx> ^^ what he said.
17:16:41 <adamw> when you say 'polish', to what exactly are you referring?
17:16:41 <roshi> does it do the same thing under kde?
17:16:51 <adamw> roshi: i don't know if anyone's tried, yet.
17:16:53 <roshi> the criteria that should not be named
17:16:59 <kparal> :D
17:17:05 <danofsatx> I thought there was an entry in the bug that said it did
17:17:49 <roshi> we talked about this in the past, but don't know if it got properly codified
17:18:09 <danofsatx> comment #11
17:18:34 <roshi> about having a polish criteria, for catching stuff like this - where it still technically *works*, but is ugly or we wouldn't want to ship with it
17:18:47 <roshi> despite it not actually *violating* a criteria
17:19:18 <kparal> I think very few of our users would describe it as "technically works"
17:19:41 <kparal> I think they would use other expressions
17:20:16 <adamw> we have some criteria that can be described as 'polish criteria', but none of them is really related to this.
17:20:18 <kparal> I'd use the proposed criterion, just postpone it for a later milestone because it's not that a critical problem
17:20:50 <adamw> i'd just say it's a conditional violation of the 'browser has to work' criterion and we say it's serious enough to block whichever milestone we want to block.
17:21:08 <adamw> i'd be fine with beta blocking, probably, i just think for alpha it's liveable-with / workaroundable.
17:21:13 <roshi> wfm
17:21:24 <roshi> I was just going off of: "The web browser must be able to download files, load extensions (if applicable), and log into FAS"
17:21:31 <adamw> (i think you can mess with gtk+ themes or something to avoid it, probably, i was just hoping it'd get fixed before i have to bother.)
17:21:36 <adamw> yeah, that one.
17:21:48 <roshi> if all that works, it's a non-blocker in my head
17:22:02 <roshi> hard to use != not working, IMO
17:22:20 <roshi> otherwise emacs and vim wouldn't work, for a lot of people :p
17:22:20 <pschindl> roshi: +1
17:22:49 <roshi> did someone have screenshots or a video to look at?
17:23:12 <kparal> hard to use != works as expected
17:23:18 <adamw> roshi: no, we're lazy. get a vm. :P
17:23:25 <pschindl> I have but on the computer which is 20 km far away right now :(
17:23:28 <roshi> "as expected" is a moving target :p
17:23:43 <roshi> lol
17:23:48 <danofsatx> yeah, I currently expect to hit this bug, so yes - it works as expected
17:24:11 <adamw> roshi: https://www.happyassassin.net/temp/firefox-white.webm
17:24:13 <roshi> I won't be upset if I get outvoted - just doesn't seem like it actually violates the criteria to me
17:24:53 <roshi> yeah, that looks annoying
17:25:03 <roshi> thanks, btw :)
17:25:07 <roshi> votes?
17:25:20 <roshi> -1 (or -.5)
17:25:25 <adamw> roshi: it's always difficult to absolutely nail down wording. i mean, frequently, it's *technically* possible to do the thing somehow or other.
17:25:43 <adamw> i'm -1 alpha, +1 final, beta's a bit harder...probably +1 though.
17:25:46 <roshi> can you get to all the places you need with keyboard shortcuts?
17:26:10 <kparal> +1 beta or final
17:26:11 <roshi> true, which is my .5
17:26:24 <adamw> roshi: i dunno all the firefox keyboard shortcuts.
17:26:54 <roshi> Ctrl-o, Ctrl-t, Ctrl-s
17:27:09 <roshi> that's most of the ones you need, but this is a digression :p
17:27:32 <adamw> none of those seems totally relevant to the things you might want to do, but yeah.
17:27:37 <roshi> so, looks like +2 for final
17:27:51 <adamw> any other votes ?
17:28:08 <danofsatx> -1
17:28:37 * kparal pokes pschindl
17:29:39 <roshi> I'm fine with +1 to final
17:29:47 <roshi> I wouldn't want to ship with this, by any means
17:29:52 <roshi> proposed #agreed - 1185999 - AcceptedBlocker Final - This bug is a conditional violation of the Alpha criterion, "t must be possible to run the default web browser and a terminal application from all release-blocking desktop environments..." but isn't deemed bad enough to block until a later Milestone as FF *technically* works.
17:30:03 <pschindl> I can vote +1 for final, if you want :)
17:30:15 <adamw> ack, we can argue about beta if it's still broken then
17:30:16 <roshi> patch
17:30:28 <roshi> proposed #agreed - 1185999 - AcceptedBlocker Final - This bug is a conditional violation of the Alpha criterion, "It must be possible to run the default web browser and a terminal application from all release-blocking desktop environments..." but isn't deemed bad enough to block until a later Milestone as FF *technically* works.
17:30:40 <roshi> s/"t/"It/
17:30:46 <kparal> ack
17:30:52 <pschindl> ack
17:31:11 <roshi> #agreed - 1185999 - AcceptedBlocker Final - This bug is a conditional violation of the Alpha criterion, "It must be possible to run the default web browser and a terminal application from all release-blocking desktop environments..." but isn't deemed bad enough to block until a later Milestone as FF *technically* works.
17:31:38 <roshi> onto the next?
17:32:03 <kparal> yes
17:32:05 <roshi> #topic (1187742) rebuild openldap with support for moznss
17:32:06 <roshi> #link https://bugzilla.redhat.com/show_bug.cgi?id=1187742
17:32:06 <roshi> #info Proposed Blocker, openldap, NEW
17:32:29 <roshi> who wants to pull secretary duty?
17:33:06 <adamw> i gots it
17:33:08 <danofsatx> based on sgallah's statement in the bug, I am +1 on the grounds that it breaks Server's primary deliverable.
17:33:09 <adamw> (unless anyone else wants it)
17:33:27 <adamw> this is proposed for what milestone?
17:33:27 <kparal> the description is bit terse for me
17:33:31 <adamw> beta
17:33:40 <adamw> we should probably adjust blockerbugs to specify the milestone for each nomination...
17:34:29 <danofsatx> I agree, the description doesn't give much. I was going to deploy a couple test instances to see WTF was going on.
17:34:41 <roshi> it does, I just forgot to say we were moving to beta proposals
17:34:56 <roshi> under this criterion: "Release-blocking roles and the supported role configuration interfaces must meet the core functional Role Definition Requirements..."
17:35:20 <adamw> yeah, a bit more detail on what it breaks would be good, but i'm ok with 'sgallagh says it's broken' really
17:35:20 <kparal> that's a beta criterion?
17:35:35 <danofsatx> heh....it is now ;)
17:35:44 <adamw> danofsatx: i think that's one of the ones from f21
17:35:51 <adamw> or is that one of my new proposals?
17:35:53 <adamw> where am i?
17:35:55 <adamw> where's my drink?
17:35:55 <roshi> it is, kparal
17:36:06 <kparal> adamw: why is your bottle empty?
17:36:09 <roshi> http://fedoraproject.org/wiki/Fedora_22_Beta_Release_Criteria#Roles
17:36:14 <adamw> good lord, this is an emergency
17:36:40 <kparal> alright, +1 beta
17:36:42 <roshi> got white russians over here, if you're in need of an emergency infusion of vodka and cream :)
17:36:54 <adamw> you could also cite one of the specific domain controller role requirements, but eh, i think that's fine for now
17:36:56 <roshi> +1 from me
17:37:08 <adamw> +1 beta but i'll ask for a bit more detail on what's broken so we can confirm and test
17:37:31 <pschindl> +1
17:37:49 <roshi> proposed #agreed - 1187742 - AcceptedBlocker Beta - This bug is a clear violation of the Beta criterion: "Release-blocking roles and the supported role configuration interfaces must meet the core functional Role Definition Requirements..."
17:38:03 <danofsatx> ack
17:39:01 <kparal> ack
17:39:03 <adamw> ack
17:39:06 <pschindl> ack
17:39:35 <roshi> #agreed - 1187742 - AcceptedBlocker Beta - This bug is a clear violation of the Beta criterion: "Release-blocking roles and the supported role configuration interfaces must meet the core functional Role Definition Requirements..."
17:39:43 <roshi> that's it for proposed blockers
17:40:03 <kparal> discussion time?
17:40:08 <roshi> want to talk about tagging specific people as bug prodders with the QA-contact field?
17:40:14 <kparal> sure
17:40:17 <roshi> #topic Discussion Time
17:40:34 <roshi> kparal: want to give us a tl;dr of the idea?
17:40:39 <kparal> so, the idea is, every accepted blocker would get a single qa person assigned
17:40:56 <kparal> the qa person would make sure the issue is getting resolved
17:41:13 <kparal> first of all, it would inform the relevant developer that this bug was accepted as a blocker
17:41:16 <roshi> I would also say the assigned person secretarializes their bug
17:41:29 <kparal> very often, the developers/maintainers don't know it at all
17:41:44 <kparal> e.g. gnome devs usually don't follow rh bugzilla
17:41:50 <adamw> hum. it seems a bit inefficient
17:42:00 <roshi> which does, adamw ?
17:42:10 <adamw> kparal: i think that's overstated. at least some of the gnome folks do, and they know as a team when they have blockers. mclasen follows the blocker list
17:42:10 <adamw> the idea of one-person-per-bug
17:42:21 <adamw> it seems more efficient for one person to follow up on all blockers
17:42:41 <kparal> I thought it would be better the distribute the load
17:42:56 <roshi> I was thikning we'd split it up like this: we have a Gnome person, a KDE person, and anaconda person, etc etc
17:43:00 <roshi> or several people for each
17:43:36 <kparal> roshi: also possible
17:43:37 <roshi> for instance, I'm going to make sure cloud bugs get handled by the cloud folks
17:43:50 <danofsatx> what happens when we get a bug that keeps getting tossed back and forth, such as the GTK3 bug we just discussed? Is it a gnome bug, or a Firefox bug?
17:44:18 <danofsatx> or upcoming bugs between anaconda and dnf-3
17:44:19 <roshi> I'd say it stays with whoever had it first, since they have the history
17:44:29 <adamw> well, for our purposes we'd just pick one person or the other to 'have' it from the blocker process PoV, it wouldn't matter too much
17:44:51 <kparal> to finish the proposal, the qa person would make sure there are some updates in bugzilla. so we don't always end up with "on updates, will wait another week" during accepted blocker reviews. so he would ping devs to provide some updates, if there were none in bugzilla for e.g. a week or two
17:45:02 <roshi> I think the idea was to have some way of knowing that the bug was going to get poked, and knowing who that person was
17:45:22 <danofsatx> agreed, just playing devil's advocate
17:46:14 <kparal> the current problem is that I somewhat assume that adamw or roshi pokes the devs, but I'm not sure, so I don't know whether I should do it or not. of course I can always ask first, but it doesn't seem to efficient, mainly due to timezone issues
17:46:24 <adamw> i guess it just somehow 'feels' to me like something better for one person to do the whole lot - when i used to prioritize it it really wasn't *that* much work, an hour or two
17:46:50 <adamw> i sort of feel like if we do one-person-per-bug everyone's going to spend 15 minutes on it and that's more person-hours
17:46:53 <adamw> but it's only a feeling
17:47:10 <adamw> also, practically speaking, what does the QA Contact field *do*? we've never used it for anything, have we?
17:47:17 <kparal> I don't think so
17:47:25 <kparal> it's unused at the moment, I believe
17:47:25 <roshi> my plan this release was to do the poking - and I'm fine with keeping that idea
17:47:38 <adamw> i'm already CCed on just about every blocker in existence, so i don't know how i'd conveniently check which blockers were 'mine' to follow up on. well, i suppose it'd be easy to throw together a custom search
17:47:41 <roshi> and I can reach out if I need someone else to cover it
17:47:51 <adamw> have you been doing it so far?
17:48:05 <roshi> if by "doing it" you mean, "meaning to" :p
17:48:23 <kparal> adamw: so are you volunteering to do all the poking, or do you prefer to split the work between a few people (I'm not saying many people, but a few)?
17:48:31 <roshi> we talked about it, but I hadn't paid too much attention since we haven't branched yet
17:48:34 <adamw> kparal: i was volunteering roshi. ;P
17:48:37 <kparal> :D
17:48:44 * adamw is getting good at that
17:48:50 <roshi> I'm fine with that, if you guys are
17:48:54 <adamw> roshi: you know you're filling in my tax returns too, right?
17:48:59 <roshi> I don't mind spot checks to make sure I did it
17:49:12 <roshi> is the canadian tax system better than ours? Ours is a nightmare
17:49:31 <adamw> shall we give roshi a chance to do it, and if it turns out not to work well we can consider the spread-the-load approaches?
17:49:42 <roshi> that works for me
17:50:13 <kparal> sure, why not, if roshi is ok with that
17:50:56 <roshi> I had already planned on doing it - so it's fine with me
17:51:11 <roshi> then we talked about this, and haven't really thought about it since
17:53:02 <kparal> ok, agreed.  roshi does all the work. deal
17:53:22 <adamw> =)
17:53:25 <roshi> where do I get my poking stick?
17:53:32 <adamw> roshi: give us a yell if you start drowning and we'll reconsider
17:53:37 <danofsatx> +1
17:53:50 <roshi> I'll at least try to give a gurgle
17:53:53 <roshi> :p
17:54:42 <roshi> should we go over accepted blcokers, or hold off?
17:55:06 <roshi> we've got 2/4/3 accepted in Alpha/Beta/Final so far
17:55:20 * roshi is fine with holding off on the review of those
17:55:27 <adamw> i'm looking at the alpha ones
17:55:30 <roshi> unless we just wanted to do alpha
17:55:32 <adamw> the SELinux one just went to modified
17:55:50 <adamw> the anaconda unicode one i actually wanted to post an update on, i tried to reproduce it last week but couldn't, even though i don't think it's been explicitly fixed
17:55:52 <adamw> i'm just checking that again
17:56:16 <roshi> sweet
17:57:23 <adamw> so...i'm on it
17:58:59 <roshi> so do we need any discussion on them?
17:59:11 * roshi is fine either way
18:00:12 * danofsatx is drinking coffee, doing homework. Fine either way.
18:01:19 <adamw> don't think so
18:01:25 <adamw> i've just asked nirik if he can try and find out why nightly composes keep failing lately
18:01:25 <adamw> it'd be good to get a new nightly nominated and try to knock out some more of the tests
18:01:34 <roshi> for sure
18:01:41 <roshi> which is the last working one?
18:01:57 <adamw> the current nominated nightly is about the latest we have a boot.iso for
18:02:02 <adamw> we have much newer lives
18:02:13 <adamw> we have a workstation live for today
18:03:27 <roshi> #info Last working boot.iso is the currently nominated nightly. Live images have more recent images to test
18:04:21 <roshi> if there's nothing else, I'll set the fuse
18:04:30 * roshi sets quantum fuse...
18:04:34 <roshi> 3...
18:04:36 <nirik> it's just different broken deps breaking the ones that are broen
18:04:38 <nirik> broken
18:06:09 <danofsatx> jreznik just sent the email that python-3 is retargeted for F23.
18:06:21 <roshi> good to know
18:06:28 <roshi> and one less thing to test this release...
18:07:11 <danofsatx> https://lists.fedoraproject.org/pipermail/devel-announce/2015-February/001530.html
18:07:46 <roshi> 2...
18:09:03 <roshi> 1...
18:09:41 <danofsatx> there's up date to openldap blocker already ;)
18:09:47 <danofsatx> the bug, anyhow, not the package
18:09:57 <danofsatx> more details on what it breaks
18:10:01 <roshi> updates are good
18:10:09 <roshi> well, I'm going to close the meeting down
18:10:12 <roshi> thanks for coming folks!
18:10:16 <danofsatx> kerblooey
18:10:22 <roshi> #endmeeting