17:01:05 #startmeeting F17-blocker-review 17:01:05 Meeting started Fri Jan 27 17:01:05 2012 UTC. The chair is tflink. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:05 Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:01:14 #meetingname F17-blocker-review 17:01:14 The meeting name has been set to 'f17-blocker-review' 17:01:21 #topic roll call 17:01:33 who's ready for some blocker bug review awesomeness? 17:01:37 * nirik is lurking around 17:01:39 #chair adamw 17:01:39 Current chairs: adamw tflink 17:02:21 yo 17:02:34 * adamw is blockerrific 17:03:16 * tflink has an image of tony the tiger saying "it's blockerrrrrrific!" 17:03:59 that's the one 17:05:07 * tflink waits a bit more so we can hopefully have 3 people 17:08:06 pschindl, jskladan: here to help with the blocker reivew? 17:08:16 tflink: just lurking :) 17:08:23 we'll take that as a 'yes' 17:08:28 :'( 17:08:49 alrighty, it's almost 10 after. let's get this party started 17:09:03 tflink: I'm interested on what's going on here. 17:09:03 #topic Introduction 17:09:19 #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. 17:09:34 #link https://fedoraproject.org/wiki/QA:SOP_Blocker_Bug_Meeting 17:09:45 #link https://fedoraproject.org/wiki/Current_Release_Blockers 17:09:56 #info 5 proposed blockers 17:10:07 #info 1 proposed nth 17:10:18 oh yay, the pain will be short. 17:10:29 unless there are any objections, I'm going to get started with the proposed blockers 17:10:41 adamw: depending on the level of debate, yes :) 17:10:50 #topic (696482) Protective MBR entry on GPT drives must be marked active for some machines to boot, but this is a violation of GPT spec 17:10:53 #link http://bugzilla.redhat.com/show_bug.cgi?id=696482 17:10:54 Bug 696482: urgent, unspecified, ---, anaconda-maint-list, NEW, Protective MBR entry on GPT drives must be marked active for some machines to boot, but this is a violation of GPT spec 17:10:55 #info Proposed Blocker, NEW 17:11:02 oh, god, this one. 17:11:16 like I said, depending on the level of debate ... 17:11:24 i think i saw some discussion of this in #anaconda yesterday 17:11:52 * tflink has a machine that hits this 17:12:11 it's supposed to be an EFI machine, too 17:12:56 however, I'm not really sure what we can do on this without more input 17:13:11 yes, I'm working on fixing up mjg59's patch for that 17:13:12 which is bug 754850 17:13:13 Bug https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=754850 unspecified, unspecified, ---, anaconda-maint-list, NEW, Some systems won't boot off GPT disks if the protective MBR entry isn't flagged bootable 17:14:05 looks like they're going to go ahead and set the flag. 17:14:07 does that mean that anaconda will set the MBR boot flag? 17:14:16 yeah. 17:14:36 so we can probably close the old bug as a dupe of the new one, since that's where the action is. 17:14:48 is that the #2 option of your comment in the 696482, adam? 17:14:50 so it sounds like a fix is on the way 17:15:02 o hai 17:15:11 jskladan: indeed 17:15:30 maxamillion: http://2.bp.blogspot.com/_HEgnDdxt_5o/S5x1sGE_-3I/AAAAAAAAAy0/pDyugx5dh7M/s320/o+hai.jpg 17:16:18 proposed #agreed 696482 - AcceptedBlocker - 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 includes correctly accessing any encry 17:16:27 jskladan: ha! 17:16:35 ack/nak/patch? 17:16:48 nack 17:16:56 i just made it a dupe of 754850 17:17:01 so we can make *that* one a blocker if we like 17:17:27 proposed #agreed 754850 - AcceptedBlocker - 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 includes correctly accessing any encry 17:17:37 adamw: not sure they're dupes, though 17:17:53 754850 is a potential solution to 696482 17:18:08 either way, though 17:18:12 ack/nak/patch? 17:18:48 ack 17:19:03 ack 17:19:12 tflink: bcl said to make them a dupe, as 754850 is where the action's going to be. 17:19:20 #agreed 754850 - AcceptedBlocker - 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 includes correctly accessing any encrypted part 17:19:34 adamw: I still don't agree but don't care enough to argue :) 17:19:46 that's the qa spirit! 17:19:55 #topic (742207) No usable bootloader option during a text mode f15->f16 upgrade 17:19:58 #link http://bugzilla.redhat.com/show_bug.cgi?id=742207 17:19:59 Bug 742207: high, unspecified, ---, anaconda-maint-list, NEW, No usable bootloader option during a text mode f15->f16 upgrade 17:20:01 #info Proposed Blocker, NEW 17:20:04 oh, i'll secretaryize 17:20:14 adamw: appreciated 17:20:37 did we ever float a proposal to test@ regarding a release requirement around text-mode installs? 17:20:58 pardon the noob question, I've been out of the bugzapper game for a while due to silly school stuff ... what's ack/nack/patch definitions? 17:21:13 maxamillion: not sure what you're asking 17:21:33 ack - ok with the proposal 17:21:38 nak - not OK with the proposal 17:21:48 tflink: i don't think we got to it yet. 17:21:54 patch - change to proposal 17:22:00 tflink: ah, ok ... perfect, thanks 17:22:05 maxamillion: np 17:22:21 tflink: i think it's on pschindl's todo list as part of the criteria concordance 17:22:34 yeah, I haven't gotten all the way through that email yet 17:22:34 we really need anaconda team's opinion on whether text mode upgrade is 'supported' 17:22:47 we could go with "The installer must be able to complete an installation using the text, graphical and VNC installation interfaces" 17:23:02 I send a mail to anaconda-devel-list where I ask them for their meaning 17:23:16 * opinion 17:23:18 I'd argue that bootloader issues make for an incomplete installation 17:23:20 sry 17:23:34 tflink: this issue is upgrade specific 17:23:40 tflink: fresh text install worked okay, we checked that 17:23:59 pschindl: i don't see that mail - it may be stuck in moderation 17:24:08 these are alpha blockers, yes? 17:24:11 I remember seeing it, looking for a link 17:24:20 ah i see it now 17:24:24 didn't have 'text' in the subject :) 17:24:25 (sorry, I'm in a meeting at $dayjob with my laptop ... trying to juggle both) 17:24:33 #link https://www.redhat.com/archives/anaconda-devel-list/2012-January/msg00207.html 17:24:34 maxamillion: yeah 17:24:41 so for alpha, i'd say this is fairly clearly not a blocker 17:24:49 imho not an alpha blocker 17:24:49 upgrades are beta stuff - even graphical upgrades 17:24:53 aaa ninja'd 17:24:57 =) 17:24:57 adamw: awesome, thanks 17:25:01 nack 17:25:01 we could bump it to a beta proposal 17:25:15 maxamillion: there is no proposal yet :) 17:25:21 * maxamillion is confused 17:25:23 ok 17:25:33 maxamillion: it's kinda silly 17:25:37 I thought the proposal was to make that bug a blocker 17:25:44 maxamillion: we usually discuss it for a bit 17:25:45 or is that not inferred? 17:25:47 before the proposal, we're voting +1/-1 blocker/nth 17:25:48 ah, ok 17:25:53 you can vote +1 or -1 during that time 17:25:56 -1 17:26:12 I think at alpha phase, upgrades are a NTH 17:26:18 then whoever's running the meeting will propose a complete action - make it a blocker or not a blocker, with justification - and we vote ack/nack/patch on that 17:26:20 agreed that it isn't a blocker for alpha but it might be good to poke for movement before beta 17:26:34 it's important that everyone's okay with the precise text of the action because it winds up in the bug report and the meeting log. 17:26:39 yay for bureaucracy! 17:26:51 I'd rather avoid the text-mode limbo that we had for F16 17:26:59 yeah, we should definitely sort it out for beta 17:27:06 either fix it or block text mode upgrades 17:27:09 adamw: rgr 17:27:11 that's what we need anaconda team to decide 17:27:27 * tflink is wondering about nth for alpha 17:27:51 or just push it to beta and get out the pointy stick 17:28:20 thoughts? 17:28:22 nth for alpha...hum, prolly okay with me 17:29:05 * tflink is ok with either as long as we figure it out for beta 17:29:25 yeah, i'm kinda either way on nth for alpha, but definitely beta blocker. 17:30:01 +1 for beta blocker 17:30:23 +1 for beta blocker 17:30:27 * jskladan is not sure, what is the impact of nth, so stays silent 17:30:36 pschindl, maxamillion : any thoughts on alpha nth? 17:30:45 'nth' is nice-to-have 17:30:51 yeah, I'd +1 it as an alpha nth 17:30:56 what it means is that a fix for it can be pulled through the freeze 17:30:58 which is a horribly named type, I think 17:31:05 tflink: probably so, yes :) 17:31:18 I don't think it's nice to have. I rather see it as nice to do not have :) 17:31:24 packages can't be pulled through the freeze without an associated nth bug 17:31:37 see https://fedoraproject.org/wiki/QA:SOP_nth_bug_process 17:31:52 * maxamillion checks the link 17:32:10 nth or blocker bug 17:32:22 the difference being that we hold release for blockers but not for nth 17:32:41 anyway 17:32:45 let's just make it a beta blocker and move on 17:32:46 so it sounds like were vaguely for nth on alpha 17:32:47 this is taking too long 17:32:52 that sounds better 17:33:18 it should be blocker. We have beta criterion for this 17:33:27 nth for alpha, blocker for beta .... is my opinion 17:33:35 proposed #agreed - 742207 - AcceptedBlocker (beta) - The installer must be able to successfully complete an upgrade installation from a clean, fully updated default installation (from any official install medium) of the previous stable Fedora release, either via preupgrade or by booting to the installer manually. The upgraded system must meet all release criteria 17:33:48 ack/nak/patch? 17:33:49 ack 17:33:54 ack 17:34:04 ack 17:34:08 ack 17:34:19 proposed #agreed - 742207 - RejectedBlocker, RejectedNTH (alpha), AcceptedBlocker (beta) - The installer must be able to successfully complete an upgrade installation from a clean, fully updated default installation (from any official install medium) of the previous stable Fedora release, either via preupgrade or by booting to the installer manually. The upgraded system must meet all release criteria 17:34:25 just being more specific 17:34:30 #agreed - 742207 - RejectedBlocker, RejectedNTH (alpha), AcceptedBlocker (beta) - The installer must be able to successfully complete an upgrade installation from a clean, fully updated default installation (from any official install medium) of the previous stable Fedora release, either via preupgrade or by booting to the installer manually. The upgraded system must meet all release criteria 17:34:42 #topic (784677) image installs erroneously change system hostname setting 17:34:45 #link http://bugzilla.redhat.com/show_bug.cgi?id=784677 17:34:46 Bug 784677: unspecified, unspecified, ---, anaconda-maint-list, NEW, image installs erroneously change system hostname setting 17:34:48 #info Proposed Blocker, NEW 17:35:43 is this a sw raid issue? 17:35:44 * jskladan can not find alpha or beta criterion which would make this a blocker 17:36:23 it sounds like RAID labels won't be created correctly if the hostname isn't set in the right place 17:36:24 "Adding to F17Alpha since this should be a fairly quick fix and needs to make F17." - that's not what the blocker process is for. 17:37:09 this might be beta blocker material but I'm -1 blocker, -1 nth on this for alpha 17:37:16 anyone know what dlehman means by 'disk image install'? 17:37:19 i'm a bit confused. 17:37:32 adamw: I think it's an issue with sw raid installs 17:37:49 anaconda isn't setting the hostname correctly and that's messing with installs that create RAID volumes 17:37:52 i don't think it is. 17:38:00 the reproduction steps don't mention raid. 17:38:12 the change that *causes* the bug was related to raid, but it doesn't mean the bug only happens on raid installs. 17:38:18 yeah, I misread the bug 17:38:20 i'm not 100% clear though, as i'm not sure what 'disk image install' means. 17:38:34 it seems like if you'd be doing some installation from running system to other disk 17:38:39 (maybe some cloning?) 17:38:54 and you set the hostname for the _new_ system in anaconda UI 17:39:13 which then sets it on the 'host' system from which you run the installation 17:39:17 unless I'm missing something, this sounds like not blocker material 17:39:26 tflink: +1 not a blocker 17:39:37 +1 not a blocker 17:39:41 well, i know i'm missing something, so i'd like to know what the 'something' is before voting :) but if no-one can explain, i'd say we punt and ask for clarification 17:40:58 but I guess we can agree on "not an alpha blocker" and move on 17:41:01 or not? :) 17:41:06 * tflink plays elevator music while we wait for a response 17:41:21 tflink: muzak ftw! 17:41:23 dlehman says " installing to a file that contains what could be used for, eg: a virt disk" 17:41:34 i must be particularly dense today, because i'm still not entirely sure what the hell that means. 17:41:40 I still don't understand the impact of this 17:41:54 adamw: then there are at least 2 dense people here :) 17:41:56 adamw: IMHO that's what I've been talking about: you have a running system 17:42:04 from which you perform some installation 17:42:05 add me to the list of dense 17:42:12 to a file 17:42:14 or whatever 17:42:23 and the anaconda has an "hostname" text box 17:42:38 which sounds like not-blocker-material and non-standard-workflow to me 17:42:39 which you file in order to set the hostname for the newly installed system 17:42:52 yeah, i think you're right 17:43:08 it sounds like an issue for an anaconda dev environment, not actual usage 17:43:15 yup 17:43:41 imho not a blocker not a nth 17:44:20 adamw: that's one of the methods that livemedia-creator uses. 17:44:21 you have to have anaconda installed on the same system it's meant to install, then you run something like 'anaconda --image=/path/to/disk/image/file' 17:44:33 oh 17:44:36 well that's interesting 17:45:04 proposed #agreed - 784677 - RejectedBlocker RejectedNTH - Doesn't hit any of the alpha release criteria 17:45:10 ack 17:45:12 ack 17:45:13 ack 17:45:21 ack 17:45:22 #agreed - 784677 - RejectedBlocker RejectedNTH - Doesn't hit any of the alpha release criteria 17:45:35 #topic (750376) nss 3.13 breaks sssd TLS 17:45:35 #link http://bugzilla.redhat.com/show_bug.cgi?id=750376 17:45:35 #info Proposed Blocker, MODIFIED 17:45:36 Bug 750376: urgent, unspecified, ---, emaldona, MODIFIED, nss 3.13 breaks sssd TLS 17:46:15 * jskladan does not know a bit about this, back to the muzak... 17:47:43 it sounds like there might have been some yum breakage at some point 17:48:28 so this looks to be fixed in current rawhide 17:48:51 * tflink still isn't 100% sure of the impact 17:48:58 broken ssl, ssh? 17:49:50 so this was proposed by sgallagh back last nov 17:49:55 yeah i'm trying to follow it, seems like a complex bug 17:49:55 I think this has more to do with TLS+LDAP authentication for accounts .... but I'm still only about 2/3 of the way through reading the bug report 17:49:59 sssd is remote auth stuff 17:50:09 but i want to be sure there isn't actually some worse breakage buried in here 17:51:52 i'm leaning towards ask for clarification and punt, here 17:52:01 don't want to miss an important bug in a complex report like this 17:52:09 yep, sounds like a plan to me 17:53:00 proposed #agreed - 750376 - We need more information on the impact of this bug and whether or not it has been fixed in rawhide before making a decision 17:53:01 if I'm not mistaken, this bug looks like its close to resolution ... comment #65 looks positive 17:53:08 but yeah, I think more information is likely a solid plan 17:53:12 ack 17:53:24 ack 17:53:36 #agreed - 750376 - We need more information on the impact of this bug and whether or not it has been fixed in rawhide before making a decision 17:53:47 #topic (725219) anaconda should run in clone not span mode 17:53:48 #link http://bugzilla.redhat.com/show_bug.cgi?id=725219 17:53:48 #info Proposed Blocker, NEW 17:53:49 Bug 725219: low, unspecified, ---, ajax, NEW, anaconda should run in clone not span mode 17:54:03 yay, my 'favourite one' 17:54:08 adamw: is the whiteboard your handywork? 17:54:23 yeah 17:54:30 but i also re-proposed this 17:54:49 because it got kinda worse in f16 17:55:10 you mean the "guess the invisible button" game? :) 17:55:19 yup 17:55:30 comment #48 17:55:44 the workaround is 'unplug a monitor', which kind of sucks. 17:55:46 yeah, I just read that one ... sounds kinda sketchy 17:56:20 I think the workaround is tolerable if extremely necessary, but if the patch wouldn't be an act of pulling teeth I certainly think this would be nice to have 17:56:27 not sure we have a criteria for this, though 17:56:32 I'm conflicted on it 17:57:03 imho there is not a criterion for this, even though this is extremely unpleasan bug 17:57:10 yeah 17:57:12 that's a fair point 17:57:23 yeah, there isn't really a criterion that fits it great, we'd have to stretch one 17:57:35 * maxamillion gets out the yoga mat 17:57:37 we could attempt to use "The installer must be able to complete an installation using the text, graphical and VNC installation interfaces" 17:57:39 it'd just be the 'you have to be able to install' criterion with a situational qualification 17:57:44 yeah, that'd be the one 17:58:04 but i'm okay with rejecting it, really 17:58:22 I think it could be nth 17:58:23 +1 on rejecting it as blocker 17:58:30 +1 to reject as blocker 17:58:35 +1 for NTH 17:58:57 +1 on nth, even though I think that anaconda guys won't spend much time on it 17:58:59 +/- 0 on blocker 17:59:18 I'm not even sure this qualifies as NTH 17:59:27 why would we take a fix for this past freeze? 17:59:55 because installing without next button sux :D 18:00:23 yeah, just because it's a really visible and annoying bug, really. 18:00:35 I'm not arguing on whether or not it sucks but why does that justify taking a fix past freeze? 18:00:53 we've done it on that basis before. 18:00:54 http://rbergero.fedorapeople.org/schedules/f-17/f-17-quality-tasks.html <-- me casually drops that in here for when people get a chance 18:01:21 which is why I hate the NTH name 18:01:32 but it sounds like the consensus is +1 NTH? 18:01:38 rbergeron: the 18:01:59 rbergeron: the 'pre-beta acceptance test plan' and 'pre-rc acceptance test plan' shouldn't exist 18:02:22 rbergeron: other than that looks good, thanks! 18:02:27 proposed #agreed 725219 - AcceptedNTH - There aren't any specific criteria for this bug but it is rather visable and interferes with installations 18:02:28 tflink: yeah 18:02:32 ack 18:02:33 ack/nak/patch? 18:02:34 ack 18:02:35 oh, you're right 18:02:36 tflink: ack 18:02:50 #agreed 725219 - AcceptedNTH - There aren't any specific criteria for this bug but it is rather visable and interferes with installations 18:02:57 rbergeron: we replaced those rats runs with tcs, we didn't just bump everything up 18:03:14 OK, that's it for the proposed blockers 18:03:19 on to the one proposed NTH 18:03:22 #topic (741549) Login failure due to bad ~/.local/share/icc selinux file contexts 18:03:25 #link http://bugzilla.redhat.com/show_bug.cgi?id=741549 18:03:27 Bug 741549: unspecified, unspecified, ---, bnocera, NEW, Login failure due to bad ~/.local/share/icc selinux file contexts 18:03:28 #info Proposed NTH, NEW 18:04:26 ah, looks like this is basically fixed now. though i'll want to re-test an f15->f17 upgrade just to be sure. 18:05:27 I'm -1 NTH on this 18:05:36 if it's still an issue, it sounds like a blocker 18:05:58 nvm, I'm reading too fast again 18:06:12 partially of-topic - is there a criterion for upgrading from F-X to F-(X+2)? /me can't find it 18:06:17 it's a bit too specific to be a blocker 18:06:33 jskladan: no, there isn't; a two-release upgrade is specifically *not* blocking 18:06:37 * maxamillion has to run for about 15 minutes 18:06:37 yeah, I spoke too soon 18:06:45 though actually, if you upgraded f15->f16 then f16->f17, you'd probably still hit it. if it wasn't fixed. 18:06:52 maxamillion: hopefully we'll be done in 15 minutes :) 18:06:55 adamw: ah, seems legit :) 18:07:07 but it's still not really blocking, it depends on having done something specific prior to the upgrade. 18:07:18 +1 ^^ 18:07:29 since it's an upgrade issue, NTH beta? 18:08:41 sure, whatever. if it's already been changed i'm less worried about it. 18:09:31 * jskladan is rather indifferent to this 18:09:53 tell you what, for now i withdraw the proposal 18:10:00 i'll re-propose it if it seems to become urgent again' 18:10:01 proposed #agreed - 741549 - Upgrades are beta issues, RejectedNTH for alpha, re-propose as Beta NTH 18:10:30 nack 18:10:34 i just un-proposed it 18:10:37 so there's nothing to reject :) 18:10:54 propose #agreed 741549 - adam un-proposed this, so no action required 18:11:08 ack ^^ 18:11:15 #agreed 741549 - adam un-proposed this, so no action required 18:11:31 well, that's it for the blocker/nth bugs 18:11:36 #topic open floor 18:11:48 any other issues/bugs to bring up? 18:11:57 * tflink sets fuse for 5 minutes otherwise 18:11:59 * jskladan is hungry 18:12:12 nope, burn that fuse 18:12:16 jskladan: do you have a bz for that? 18:12:20 we need to do rats #2, dgilmore is composing it today 18:12:25 i wish :D 18:12:51 well, without a BZ, I can't make it a topic :-D 18:13:02 got it closed by the significant other as WONTFIX 18:13:40 jskladan: not sure I've ever thought of handling things that way 18:14:14 :D 18:14:25 ok /me runs for the bus 18:14:27 #info Next blocker bug review meeting - 2012-02-03 @ 17:00 UTC 18:14:41 * tflink will be sending out minutes shortly 18:15:19 secretaryizing is all done 18:15:28 annnnd I'm back 18:15:36 annnnnd it appears I missed the end of the show 18:16:13 maxamillion: not quite, there is still 1 minute left on the fuse 18:16:31 * maxamillion runs like hell 18:16:33 >.> 18:16:46 * tflink realizes that he's holding the fuse 18:16:54 :P 18:17:05 boom! 18:17:13 thanks for coming everyone! 18:17:16 #endmeeting