17:03:17 <rbergeron> #startmeeting F17 Final Go No Go Meeting 17:03:17 <zodbot> Meeting started Thu May 17 17:03:17 2012 UTC. The chair is rbergeron. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:03:17 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 17:03:25 <rbergeron> #meetingname F17 Final Go No Go Meeting 17:03:25 <zodbot> The meeting name has been set to 'f17_final_go_no_go_meeting' 17:03:30 <rbergeron> #chair tflink 17:03:30 <zodbot> Current chairs: rbergeron tflink 17:03:39 <rbergeron> #chair dgilmore nirik 17:03:39 <zodbot> Current chairs: dgilmore nirik rbergeron tflink 17:03:43 <rbergeron> #topic Gathering the Peeps 17:03:53 <rbergeron> tflink: is adamw joining or is he passed out 17:04:08 <rbergeron> spot: hellooooo 17:04:09 <tflink> I haven't heard from him this morning 17:04:13 * nirik waves 17:04:21 * rbergeron hasn't heard from spot this morning either 17:04:25 <rbergeron> hey nirik. 17:04:50 <rbergeron> #topic Purpose of this meeting 17:05:21 <rbergeron> #info Purpose of this meeting is to see whether or not F17 Final is ready for shipment, according to the release criteria. 17:05:29 <rbergeron> (to put it very briefly) 17:05:37 <rbergeron> #info This is determined in a few ways: 17:05:46 <rbergeron> #info No remaining blocker bugs 17:06:01 <rbergeron> #info Test matrices are fully completed 17:06:26 <rbergeron> #link https://fedoraproject.org/wiki/Current_Release_Blockers 17:06:57 <tflink> the wiki page doesn't seem to be updated properly 17:07:01 <rbergeron> #link https://fedoraproject.org/wiki/Fedora_17_Final_Release_Criteria 17:07:04 <rbergeron> even with a refresh? 17:07:41 <tflink> wait, maybe I'm just getting my bugs mixed up 17:08:26 <nirik> I think it's not up to date. 17:08:28 <rbergeron> #link https://fedoraproject.org/wiki/Category:Fedora_17_Final_RC_Test_Results 17:09:05 <rbergeron> Okay. tflink: do you ant to take us through what you know to be the right blockers? 17:09:06 <tflink> it was last updated about an hour ago 17:09:22 <rbergeron> or are you implying that perhaps we have none now? :) 17:09:31 <rbergeron> #topic Blocker Bugs 17:09:37 * spot is here now, sorry 17:09:45 <tflink> #info 5 proposed blockers 17:09:45 <tflink> #info 2 accepted blockers 17:10:05 <tflink> any preferences on proposed or accepted first? 17:10:29 <rbergeron> meh... accepted? 17:10:38 <tflink> works for me 17:10:41 <rbergeron> :) 17:10:47 <tflink> #topic (821077) Firstboot window does not fit the screen after install from F17 TC4 Final KDE Live 17:10:50 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=821077 17:10:53 <tflink> #info Accepted Blocker, NEW 17:11:20 <tflink> there is some disagreement on how to fix this 17:11:58 <tflink> but the issue is that firstboot is creating screens before kwin is up and running 17:12:11 <tflink> so that when kwin starts up, it messes with the layout of the existing screen 17:13:04 <nirik> yeah, sleep seems nasty, but easy to do to just work around it for now. 17:13:22 <rbergeron> but will it ever actually get fixed instead of just worked around :) 17:13:38 <nirik> it's unclear to me if anyone is working on the thing in c... 17:13:41 <rbergeron> i can see clumens' point about how that varies from hardware to hardware 17:13:45 <rbergeron> nirik: i agree 17:13:49 <spot> nirik: there is a working C program on that bug 17:13:54 * nirik looks again. 17:14:08 <tflink> nirik: https://bugzilla.redhat.com/attachment.cgi?id=585074 17:14:13 <nirik> ok. 17:14:32 <spot> https://bugzilla.redhat.com/attachment.cgi?id=585074 17:14:36 <rbergeron> working as in "has a heartbeat" or is it actually functional 17:14:37 <nirik> so then the question is where that should live. 17:14:42 * rbergeron looks at bug 17:14:55 <spot> the firstboot maintainer wants it to be in its own package 17:14:58 <rbergeron> err, attachment 17:15:02 <spot> so i'm inclined to simply package it 17:15:21 <nirik> yeah, might be generically useful for other things too. 17:15:38 <nirik> then firstboot can try and call it if it's installed, or ignore if it's not or something. 17:15:45 <rbergeron> spot: how well is it tested? 17:16:30 * spot has no idea 17:16:46 <tflink> rbergeron: the c code? not much AFAIK? kamil has tested with a wait() in firstboot but I'm not sure if many people have been testing the c code 17:16:51 <spot> the code is rather simple and sane though. 17:17:24 <tflink> it would be nice to have someone who knows more about X review the code, though 17:18:16 * nirik is happy to help review, etc... it sounds like thats going to be the best way forward since the sleep is hacky and firstboot doesn't want to carry the c 17:19:08 <tflink> well, is it a better solution than c59? 17:19:24 <tflink> nvm, that's not packaged in fedora 17:19:24 <spot> tflink: yes. 17:19:30 <nirik> "There's XFixes support in Python XLib but only in trunk, not packaged in 17:19:30 <nirik> Fedora..." 17:19:40 <nirik> yeah, that seems worse to me (more complex, less tested) 17:19:47 <tflink> yep 17:19:53 <spot> not to mention ajax's comment about not mixing the two 17:20:00 <nirik> and ... that yeah 17:20:29 <rbergeron> so: solution is... probably to have spot package said attachment, hopefully have people test it in the interim, and hope it works and move to a new RC? 17:20:41 * rbergeron doesn't want to override maintainer here but ... 17:20:48 <tflink> yeah, that sounds about right 17:21:12 <rbergeron> spot: do you know of anyone else who could sanity check that patch offhand before you run off and do it 17:21:19 <tflink> #info packaging the C code outside of firstboot seems like the best solution for now 17:21:23 <spot> rbergeron: ajax should be able to help 17:21:46 <nirik> we also need changes in firstboot too, right? to call it? 17:22:07 <nirik> and it needs to be on media, etc. 17:22:19 <spot> nirik: yep. 17:22:30 <tflink> firstboot doesn't need to be on media to test it 17:22:40 <tflink> but before we call it solved, yes 17:22:44 <nirik> true. 17:23:02 <rbergeron> #action spot to get patch packaged, get ajax to sanity check patch 17:23:13 <rbergeron> re: changes in firstboot - plans? 17:24:00 <tflink> not sure how all that would work - do python bindings need to be written for the C code? 17:24:33 * rbergeron wonders if jreznik is around 17:24:42 <nirik> it should be able to just look for the executable, if it's there call it... if not ignore and drive on... 17:24:57 <rbergeron> he's not, fooey 17:25:24 <tflink> mgracik is probably the better person for firstboot changes 17:25:57 * rbergeron nods 17:26:23 <rbergeron> tflink or spot or nirik: does one of you want to tackle asking him in bz and maybe also reaching out via email/irc 17:26:43 * tflink just attempted to ping him in IRC 17:26:52 <tflink> but I doubt that he's still around 17:27:11 <rbergeron> yeah, me too 17:27:23 * tflink can coordinate 17:27:27 <rbergeron> tflink: thank you 17:27:40 <rbergeron> #action tflink to coordinate with mgracik re: firstboot changes 17:27:44 <nirik> yeah, sorry, happy to help too 17:27:48 <tflink> you beat me to it :) 17:27:54 <rbergeron> so this remains blockery :) 17:27:59 <tflink> yep 17:28:01 <rbergeron> tflink: i guess we can move on to next bug 17:28:11 <tflink> #topic (822123) Graphical grub is unusable slow 17:28:11 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=822123 17:28:11 <tflink> #info Accepted Blocker, MODIFIED 17:28:15 <rbergeron> unless there's something else you wanted to cover 17:28:18 <rbergeron> but i guess not :) 17:28:26 * rbergeron shuts up and lets tflink do his excellent driving as usual 17:28:31 <tflink> nope, there isn't a whole lot more to say on that one 17:28:58 <tflink> this one should be fixed, just waiting new compose 17:29:17 <spot> not fixed, workedaround 17:29:27 <spot> but same difference from a go-nogo perspective 17:29:31 <tflink> true, I'm not being careful with my words 17:29:46 <tflink> commenting out theme != fixed 17:30:12 <tflink> but it works in the testing so far 17:30:22 * nirik nods. 17:30:24 <tflink> so I don't think there is much to say/do on this other than wait for RC2 17:30:34 * rbergeron nods 17:30:50 <rbergeron> okay 17:31:00 <rbergeron> #info workedaround, just need to wait for RC2 17:31:01 <tflink> #info anaconda-17.28-1 should workaround this to the point where it's no longer a blocker 17:31:16 <tflink> that's all of the accepted blockers right now 17:31:18 <rbergeron> your version sounds far superior :) 17:31:21 <rbergeron> okay! 17:31:37 <tflink> but we have a proposed blocker that is a problem 17:31:42 <tflink> well, at least one 17:31:47 <rbergeron> let's take a look at them 17:31:54 <tflink> #topic (794927) PackageKit cannot import GPG keys - Fatal error: GDBus.Error:org.gtk.GDBus.UnmappedGError.Quark._pk_5ftransaction_5ferror.Code14: Invalid input passed to daemon: char '<' in text! 17:31:58 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=794927 17:32:01 <tflink> #info Proposed Blocker, MODIFIED 17:32:31 <tflink> this is a rather nasty one that is a consequence of a PK fix that we pulled in for TC5 17:32:58 <tflink> the build does fix the exact issue that is reported but it goes back to requiring an admin password for key importing 17:34:06 <nirik> which leads to bug 748320 17:34:36 <nirik> which was not a blocker in the past. 17:34:51 <rbergeron> it was NTH 17:34:53 <rbergeron> it looks like 17:34:58 <nirik> yep 17:35:15 <tflink> for F16 17:35:16 <spot> mclasen: 794927 is fixed, but it re-exposed 748320 17:35:18 <rbergeron> have we heard anything from dgilmore on this? 17:35:50 <tflink> I don't think so, no 17:35:53 * mclasen reads up on those bugs 17:35:55 <nirik> which also leads back to bug 998. ;) 17:36:09 <tflink> 998? 17:36:17 <rbergeron> that is exactly what i was about to type 17:36:26 <nirik> https://bugzilla.redhat.com/show_bug.cgi?id=998 17:36:31 <drago01> tflink: the anaconda does not check signatures bug 17:36:50 <drago01> tflink: has been opne like ... forever ;) 17:37:02 <tflink> wow - open since 1999 17:37:04 <nirik> I think it's likely the oldest open fedora bug. 17:37:28 <nirik> anyhow... 17:37:34 <rbergeron> MEMORIES 17:37:47 * bcl waves 17:37:58 <nirik> should we vote on 748320 being a blocker here? or ? 17:38:38 <tflink> we can, but let's do 794927 first 17:38:41 <rbergeron> yep 17:38:49 * tflink and kamil are already +1 blocker 17:38:52 <spot> +1 17:38:53 <rbergeron> to 794927 17:38:57 * nirik missed that was only proposed. 17:39:07 <nirik> +1 blocker on 794927 17:39:23 <drago01> which part of the bug 17:39:26 <drago01> the error message 17:39:32 <rbergeron> +1 blocker to 794927 as well 17:39:32 <drago01> or the "it requires admin password" 17:39:33 <drago01> ? 17:39:40 <mclasen> spot: not sure I understand how 748320 was 're-exposed' - it was never fixed, was it ? just worked around in the livecd kickstart ... 17:39:41 <tflink> proposed #agreed - 794927 - AcceptedBlocker - Violates the Fedora 17 alpha release criterion "The installed system must be able to download and install updates with yum and the default graphical package manager in all release-blocking desktops" 17:39:55 <spot> mclasen: yeah, thats what i mean. 17:39:59 * adamw cites whoever scheduled this meeting at 10am for cruel and unusual punishment 17:40:16 <tflink> drago01: this bug is the error that PK gives when it tries to import a key 17:40:43 <drago01> tflink: ah ok ... this is clearly a blocker 17:41:00 <tflink> ack/nak/patch on the proposal? 17:41:08 <nirik> ack 17:41:10 <drago01> a 17:41:19 <rbergeron> adamw: well we need to get it in before release readiness 17:41:43 * adamw catches up on logs 17:42:13 <rbergeron> "it's big, it's heavy, it's wood" 17:42:18 <rbergeron> ack 17:42:27 <tflink> #agreed - 794927 - AcceptedBlocker - Violates the Fedora 17 alpha release criterion "The installed system must be able to download and install updates with yum and the default graphical package manager in all release-blocking desktops" 17:42:37 <tflink> #topic (748320) automatically import default GPG keys 17:42:37 <tflink> #link https://bugzilla.redhat.com/show_bug.cgi?id=748320 17:42:39 <tflink> #info Proposed Blocker, NEW 17:43:03 * adamw is -1 blocker because we've been shipping this way for ages. we already explicitly rejected it as an f16 blocker (on the basis we shipped f15 this way.) 17:43:07 <drago01> -1 17:43:07 <nirik> -1 blocker, +1 NTH 17:43:24 <adamw> but really, we should fix it like next week so we don't wind up back here for f18. 17:43:29 * nirik nods 17:43:31 <tflink> probably -1 blocker unless this is the only way we can fix the PK issue 17:43:47 <tflink> other NTH votes? 17:43:57 * rbergeron is +1 NTH, -1 blocker 17:44:28 <adamw> tflink: i think you messed up a bit there 17:44:46 <adamw> this bug isn't really related to 794927 at all. kamil just got reminded of it once that bug was fixed. 17:45:16 <tflink> adamw: if the keys are already imported, we don't need to worry about PK handling it, though 17:45:24 <adamw> the relationship is that 794927 was either caused or exposed by the fix to 821015 17:45:39 <adamw> you seem to have inadvertently transposed that relationship onto 794927/748320 instead 17:45:46 <adamw> at least, by my reading of the log 17:46:18 <tflink> proposed #agreed - 748320 - RejectedBlocker, AcceptedNTH - This doesn't directly hit any of the Fedora 17 release criteria and we've been shipping like this for a while now - rejected as blocker. However, this would improve the user experience and a tested fix would be considered for inclusion past code freeze 17:46:27 <mclasen> in any case, the fix for 794927 is really obvious, and makes importing work 17:46:57 <rbergeron> ack. 17:46:59 <adamw> yeah. point is, 748320 and 794927 aren't at all related. 748320 is just the same bug we've had for years, with the stupid security discussion in 998. nothing at all has changed about it for 17. 17:47:03 <adamw> ack 17:47:05 <nirik> ack 17:47:12 <tflink> #agreed - 748320 - RejectedBlocker, AcceptedNTH - This doesn't directly hit any of the Fedora 17 release criteria and we've been shipping like this for a while now - rejected as blocker. However, this would improve the user experience and a tested fix would be considered for inclusion past code freeze 17:47:20 <tflink> #topic (822430) PK is stuck after installing a single update 17:47:20 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=822430 17:47:20 <tflink> #info Proposed Blocker, NEW 17:47:48 <tflink> I was hoping to try reproducing this before the meeting but ran out of time 17:48:13 <adamw> hughsie told me he'd look at it but i guess he got no time either. :/ 17:48:26 * tflink doesn't really want to accept this without repros and more info 17:49:00 <adamw> as described it's obviously crappy behaviour but only really borderline blocker 17:49:08 <adamw> the most common case - just install all the updates and proceed - apparently works... 17:49:13 <rbergeron> has anyone else had this happen? 17:49:25 <tflink> it was just reported earlier today 17:49:26 <adamw> no, but i don't think i tried the repro steps lately 17:49:46 * adamw installing 17:50:38 * rbergeron wonders how many people just cherry-pick one update 17:50:57 <adamw> and if it happens when you just cherry-pick *two* updates... 17:51:01 <tflink> through PK? not sure 17:51:22 <tflink> either way, I'd say punt or reject as a blocker 17:51:51 <nirik> I suspect it's not too common, but not sure. 17:52:04 <rbergeron> adamw: yeah 17:52:44 <rbergeron> i don't think this prevents according to the criteria 17:53:10 <rbergeron> it just makes it suck pretty bad :) 17:53:32 <tflink> proposed #agreed - 822430 - RejectedBlocker - This doesn't directly hit any of the Fedora 17 release criteria and has a simple workaround - install all of the updates. Thus it can be fixed by an update and does not need to block release 17:53:34 <nirik> yeah, it's nasty, but -1 blocker I think. 17:53:42 <nirik> ack 17:54:01 <rbergeron> i think it would be different if it was stuck while installing but it seems that it gets stuck after the install 17:54:07 <rbergeron> makes for a crappy experience 17:54:32 <adamw> are we nth? we do need an rc2 17:54:35 <rbergeron> for those who only do one package of cherry picking which I suspect is not the average user :) 17:54:53 <rbergeron> i would be NTH on this, yes 17:55:00 <tflink> I'm wary of taking PK NTH this late when we already have PK issues 17:56:17 <tflink> if we're clear that it isn't a guaranteed freeze break, I'd be OK with NTH though 17:56:52 <nirik> I'd be ok with NTH, making sure tho that this is commited/produced seperately from the blockers so we don't have to take it if it's too invasive. 17:57:18 <adamw> good plan. 17:57:43 * rbergeron agrees with that 17:58:01 <rbergeron> separate patches plz 17:58:05 <adamw> wow. we have no de-select all button in update? that seems silly. 17:58:12 <adamw> excuse me while i de-select 107 packages... 17:58:34 <tflink> proposed #agreed - 822430 - RejectedBlocker, AcceptedNTH - This doesn't completely hit any of the Fedora 17 release criteria and has a simple workaround - install all of the updates. Thus it can be fixed by an update and does not need to block release. However, it does cause problems with certain update use cases and would be considered for freeze break. Separate patches for the different PK issues are requested so that they can be evaluated individ 17:58:40 <tflink> nice and short 17:58:42 <rbergeron> adamw: well, i guess at least it's even more unlikely people will do that :) 17:58:49 * tflink should write a novel ... in IRC :) 17:59:06 <drago01> adamw: why? it is not like deselecting updates is a common case ... 17:59:11 <tflink> ack/nak/patch? 17:59:13 <nirik> might be 's/patches/patches and builds, updates/' 17:59:15 <mclasen> adamw: context menu ? I thought there was one... 17:59:20 <adamw> oh, didn't try that. 17:59:30 <tflink> proposed #agreed - 822430 - RejectedBlocker, AcceptedNTH - This doesn't completely hit any of the Fedora 17 release criteria and has a simple workaround - install all of the updates. Thus it can be fixed by an update and does not need to block release. However, it does cause problems with certain update use cases and would be considered for freeze break. Separate patches and/or builds for the different PK issues are requested so that they can be eva 17:59:36 <adamw> hey, he's right. reproduced exactly as described. 18:00:14 <adamw> squish it! 18:01:04 <rbergeron> eva ... ? 18:01:16 <rbergeron> evaluated 18:01:29 <nirik> ack 18:01:30 <rbergeron> ... /me doesn't know if that's the end of your statement, tflink? 18:01:44 * rbergeron acks it otherwise, unless the end says "evaluated and then we shall eat kittens" 18:02:09 <tflink> evaluated individually 18:02:13 <adamw> mmmm....kittens 18:02:19 <tflink> ... and then we shall eat chili made of kittens 18:03:13 <tflink> any more ack/nak/patch? I see 2 18:03:40 <rbergeron> everyone else is already devouring kittens. 18:03:56 <bcl> akk 18:03:59 <tflink> #info no kittens will be harmed during the triage and fix for this bug ... if all goes according to plan 18:04:06 <rbergeron> a kute kitten! 18:04:12 <tflink> #agreed - 822430 - RejectedBlocker, AcceptedNTH - This doesn't completely hit any of the Fedora 17 release criteria and has a simple workaround - install all of the updates. Thus it can be fixed by an update and does not need to block release. However, it does cause problems with certain update use cases and would be considered for freeze break. Separate patches and/or builds for the different PK issues are requested so that they can be evaluated in 18:04:28 <tflink> #topic (820478) [ta_IN] Without autohinting tamil strings rendering is badly broken. 18:04:31 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=820478 18:04:34 <tflink> #info Proposed Blocker, MODIFIED 18:04:44 <tflink> this one seems kind of borderline to me 18:05:30 <adamw> well, the proposer is okay with not a blocker. looking at the screenshots i have to say not a blocker. it's not unreadable, just a bit fuzzy. 18:05:35 <tflink> huh, I thought it looked worse when I looked at the pictures last nice 18:05:37 <tflink> night 18:05:43 <adamw> (with my long years of experience in reading tamil, etc etc) 18:05:45 <rbergeron> do we have an answer as to whether or not this affects install? 18:05:55 <adamw> you can install in tamil, so, i would say yes. 18:05:56 <rbergeron> or is it just after things are up and running and installed 18:06:41 <adamw> i kinda go with kamil's take on it, really. 18:06:44 <adamw> if we slip, we can pull it in. 18:06:55 <nirik> yeah, +1 NTH, -1 blocker 18:07:00 * adamw running an install in Tamil. it is a little ugly. 18:07:15 <adamw> tiny font size, for some reason. 18:07:35 <rbergeron> I am +1 NTH ... -1 blocker 18:07:42 <adamw> +1 NTH but not if we're going to do some kind of crazy hero rc2 run today. 18:08:37 <tflink> proposed #agreed - 820478 - RejectedBlocker, AcceptedNTH - This doesn't directly hit any of the Fedora 17 release criteria but it does impact readability and user experience for installations in Tamil. The fix is built and tested, will accept if more composes are done. 18:08:43 <tflink> ack/nak/patch? 18:09:03 <nirik> ack 18:09:05 <adamw> ack 18:09:07 * rbergeron would really rather not do a crazy run - there are multiple folks who need to know when things are happening to get them prepped for tuesday or not, but... 18:09:13 <rbergeron> ack 18:09:16 <tflink> #agreed - 820478 - RejectedBlocker, AcceptedNTH - This doesn't directly hit any of the Fedora 17 release criteria but it does impact readability and user experience for installations in Tamil. The fix is built and tested, will accept if more composes are done. 18:09:16 <adamw> we can figure that later 18:09:20 <rbergeron> adamw: yeah 18:09:32 <tflink> #topic (815473) Preugprade from F16 to F17 Beta doesn't successfully set the boot menu default 18:09:35 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=815473 18:09:37 <tflink> #info Proposed Blocker, NEW 18:09:57 <tflink> note that preupgrade issues are not spin-blocking issues 18:10:08 <tflink> so this does not affect our ability to spin a new RC 18:10:56 <adamw> uh...bad freenode. 18:11:05 <tflink> huh? 18:11:11 <adamw> "--- card.freenode.net has changed the topic to: (820478) [ta_IN] Without autohinting tamil strings rendering is badly broken. (Meeting topic: F17 Final Go No Go Meeting)" 18:11:27 <sgallagh> hubbard.freenode.net did the same 18:11:34 <tflink> freenode is running the meeing now? sweet! I can take the day off! 18:11:37 <adamw> ghosts in the machine!@ 18:11:40 * tflink saw verne 18:11:51 <adamw> they're stealing our damn release names now? the bastards. 18:11:55 <adamw> beefymiracle.freenode.net.... 18:11:58 <bcl> hopefully they don't decide to split 18:12:45 <rbergeron> adamw: what? 18:13:19 <rbergeron> okay, back to 815473? 18:13:22 <tflink> ok, so thoughts on blockeryness? 18:13:30 * rbergeron notes we have release readiness in 45 minutes in here 18:13:31 <tflink> supposedly this is fixed in git 18:13:44 <tflink> we need a build in F16, though 18:14:40 <tflink> and have been waiting for a build for a month and a half? 18:15:03 * tflink looks for a sharp, pointy stick 18:15:08 <rbergeron> why so long? 18:15:18 <sgallagh> tflink: If we think it's a blocker issue, we can have a provenpackager push the upstream patch and rebuild 18:15:42 <adamw> this does hit the criteria as written. 18:15:49 <adamw> since we say the reboot direct to anaconda should work. 18:15:53 <adamw> so i guess, by the process, +1/ 18:16:01 <tflink> it does? which criterion? 18:16:23 <adamw> oh. hum. i think i misremembered. 18:16:44 <adamw> uh, the topic in my client is still 820478. you might want to re-do #topic just to be sure. 18:17:06 <tflink> #topic (815473) Preugprade from F16 to F17 Beta doesn't successfully set the boot menu default 18:17:09 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=815473 18:17:10 * rbergeron grins 18:17:11 <tflink> #info Proposed Blocker, NEW 18:17:11 <rbergeron> win 40 18:17:22 <adamw> ohh, josef quoted the _test case_, not the criterion. 18:17:35 <tflink> the criteria in question is "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" 18:17:51 <adamw> d'oh. yeah, the criterion doesn't specify, it just says preupgrade ought to be viable. soo...since this is easy to workaround and missing the boot option doesn't really hurt anything, probably -1. it's kinda academic anyhow, really. 18:18:23 <tflink> yeah, I'm probably -1 blocker, +1 sharp pointy stick to "encourage" a new build 18:18:36 <sgallagh> -1 blocker, +1 NTH 18:18:54 <tflink> nth doesn't really make sense for this since preupgrade doesn't go into the compose 18:19:07 <sgallagh> ah, true. I wasn't thinking 18:19:12 <sgallagh> Strike that NTH vote 18:19:21 * rbergeron is with tim on the sharp pointy stick 18:19:34 <nirik> ditto what tflink said. 18:20:15 <tflink> proposed #agreed - 815473 - RejectedBlocker - This doesn't affect the ability of preupgrade to actually work and thus, doesn't violate the Fedora 17 release criteria. The workaround is relatively painless but a new preupgrade build is very much desired as the issue has been fixed in git 18:20:20 <tflink> ack/nak/patch? 18:20:44 <rbergeron> ack 18:20:51 <adamw> ack 18:21:02 <tflink> #agreed - 815473 - RejectedBlocker - This doesn't affect the ability of preupgrade to actually work and thus, doesn't violate the Fedora 17 release criteria. The workaround is relatively painless but a new preupgrade build is very much desired as the issue has been fixed in git 18:21:03 <adamw> mads pointed out that just the preupgrade fix isn't enough, apparently grub2-reboot needs fixing too... 18:21:15 <sgallagh> I'm unclear on the process. Does only nirik get to vote on dev's behalf? 18:21:19 <sgallagh> If not, ack :) 18:21:33 <nirik> sgallagh: anyone can vote I think, we seek consensus. ;) 18:21:39 <nirik> and ack 18:21:56 <tflink> sgallagh: anyone can vote - does it seem like I'm ignoring you? 18:22:11 <sgallagh> tflink: Oh no, I just wasn't sure if we only had one "official" rep 18:22:32 <tflink> yeah, it took me a minute to figure out what you were asking 18:22:42 <adamw> yeah, tflink and i use a highly sophisticated neural networking based algorithm for weighting and sorting the votes 18:22:44 <tflink> on to the last one 18:22:48 <tflink> #topic (822495) Fedora 17 DVD build does not contain device-mapper-multipath 18:22:48 <adamw> by which i mean we pull it out of our ass 18:22:51 <tflink> #link http://bugzilla.redhat.com/show_bug.cgi?id=822495 18:22:53 <tflink> #info Proposed Blocker, NEW 18:23:27 <adamw> well, d'oh. 18:23:32 <tflink> I think this is a pretty clear blocker 18:23:41 <tflink> "The installer must be able to complete an installation using any network-attached storage devices (e.g. iSCSI, FCoE, Fibre Channel)" 18:23:52 <adamw> but..but..that doesn't say multipath. :P 18:24:04 <adamw> still, probably +1. you could cite the 'valid partition layout' criterion too i guess. 18:24:04 <tflink> imho, anyone running FCoE or FC w/o multipath is a moron 18:24:11 <adamw> strong words! 18:24:32 <sgallagh> I agree with tflink here. (On both the blocker and moron comments) 18:25:51 <tflink> proposed #agreed - 822495 - AcceptedBlocker - Violates the Fedora 17 final release criterion "The installer must be able to complete an installation using any network-attached storage devices (e.g. iSCSI, FCoE, Fibre Channel)" and multipath is very much the norm for higer end networked storage 18:26:01 <adamw> ack 18:26:04 <rbergeron> ack 18:26:09 <sgallagh> ack 18:26:15 <adamw> so for this we just need to put it in spin-kickstarts or something, right? 18:26:20 <tflink> #agreed - 822495 - AcceptedBlocker - Violates the Fedora 17 final release criterion "The installer must be able to complete an installation using any network-attached storage devices (e.g. iSCSI, FCoE, Fibre Channel)" and multipath is very much the norm for higer end networked storage 18:26:22 <adamw> anaconda has code to install it when needed, but it needs to be available 18:26:26 <nirik> acl 18:26:29 <nirik> ack even 18:26:33 <tflink> adamw: wouldn't comps be a better place for it? 18:26:59 <tflink> it sounds like anaconda has dm-multipath in the installer but it isn't installed to the target system 18:27:59 <adamw> well i think we put stuff that needs to be on the DVD but not in any package group into spin-kickstarts 18:28:11 <adamw> that's what we did with the last thing. crap, what was it. mactel-boot. 18:28:29 <adamw> if you put it in comps it has to go in some package group. well, there was the idea of creating an 'advanced storage tools' comps group or something. 18:28:40 <tflink> where are the rest of the DM pkgs? 18:28:59 <sgallagh> adamw: I'm not sure I'm comfortable adding a new comps group this late in the process 18:29:07 <adamw> nowhere i can see. nothing with 'device-mapper' is in comps since f15. 18:29:10 * tflink notes that this was found on ppc but is likely to exist on x86* too 18:29:17 <adamw> i think the last comment confirms that. 18:29:36 <adamw> anyhow, we're short on time i guess, so we can figure out how to fix it later... 18:29:43 <tflink> yep 18:29:53 <tflink> that's all of the blocker bugs unless another was proposed since we started 18:29:58 <rbergeron> please god no 18:30:03 <tflink> doesn't look like it 18:30:21 <adamw> last one i see was 2.5 hours ago, so no. 18:31:18 <rbergeron> okay, shall we switch gears into the Go/No-Go portion 18:31:45 * rbergeron doesn't know if it's worth covering test matrices at this point, though it would be helpful to know how far we got 18:31:55 <adamw> pretty close to covered. 18:32:03 <tflink> I don't see any holes that concern me much 18:32:23 <rbergeron> #topic Go or No Go? 18:32:24 <adamw> we're missing...SCSI, which we're always missing. and might want to drop. 18:32:29 <adamw> Xen, which is bad. 18:32:34 <adamw> i should've got konrad on that. 18:32:45 <sgallagh> adamw: Is that Xen Dom0 or DomU? 18:32:46 <tflink> it sounds like there has been some testing in the xen community 18:32:47 <adamw> one updates source...eh, we probably covered that in a TC. 18:32:51 <rbergeron> #info Test matrices are pretty much covered, except SCSI, which we are always missing. 18:32:56 <tflink> they don't seem to want to update the matrix, though 18:32:59 <tflink> sgallagh: DomU 18:33:15 <adamw> and "QA:Testcase_install_repository_NFS_graphical", which i can only assume is cos no-one saw it. 18:33:16 <tflink> Dom0 issues are explicitly excluded from being a blocker 18:33:32 <adamw> we're missing a few desktop tests, included all of KDE. 18:33:38 <adamw> so...we're probably not really there on matrix coverage, tbh. 18:33:46 <rbergeron> So basically at this point we have... three approved blockers. yes? 18:33:54 <rbergeron> including the one we just approved? 18:34:04 <adamw> yeah. 18:34:09 <adamw> i think we're looking pretty no-go. 18:34:20 <tflink> 4 approved blockers, no? 18:34:24 <adamw> matrices aren't quite there, and we clearly need an RC2 spin that isn't ready and isn't likely to be in the next 20 minutes. 18:34:28 <tflink> the currrent 2 and the 2 we just took 18:35:12 <rbergeron> oh, we took two? i thought the others were NTH 18:35:22 <tflink> the PK one and the dm-multipath one 18:35:55 <nirik> yeah, sadly no go looking. ;( 18:36:12 <rbergeron> #info 4 blockers - 2 previously accepted, 2 new 18:36:52 <rbergeron> #info some desktop tests, including KDE, are missing, we need an RC2 spin. 18:37:07 <rbergeron> Is everyone in agreement that we are a no-go? 18:37:14 * spot agrees 18:37:36 <nirik> yep. 18:37:38 <tflink> yep, no-go 18:37:45 <sgallagh> Yes, no-go 18:38:10 <rbergeron> #agreed F17 Final (from RC1) is a no-go; we slip one week, for a GA of May 29. 18:38:33 <tflink> and we hit the rhbz upgrade :( I was hoping not to have to deal with that 18:38:40 <rbergeron> crap 18:38:46 <tflink> not the same day 18:39:03 <tflink> it's just the QA scripts that are affected 18:39:05 <rbergeron> okay, any further comments? 18:39:29 * rbergeron nods 18:39:38 <rbergeron> #info Release readiness meeting in this channel in 20 minutes 18:39:51 <rbergeron> #info rbergeron to send out slip mail with a sad, unbeefy face 18:39:55 <rbergeron> #undo 18:39:55 <zodbot> Removing item from minutes: <MeetBot.items.Info object at 0x25a31050> 18:40:02 <rbergeron> #action rbergeron to send out slip mail with a sad, unbeefy face 18:40:15 <adamw> tflink: and bodhi, right? 18:40:18 <adamw> or is that fixed? 18:40:23 <rbergeron> spot: how long do you think it will take to get that package up and going? 18:40:30 <tflink> it sounds like lmacken has that under control 18:40:37 <spot> rbergeron: bcl actually thinks he found a native python fix 18:40:46 <spot> rbergeron: so it may not be needed at all 18:40:56 <tflink> a fix for python-bugzilla was posted yesterday 18:41:14 <adamw> yeah, we're testing bcl's fix at the moment. 18:41:23 <adamw> by which i mean bcl is testing it and i'm holding the whiskey bottle. 18:41:55 <rbergeron> spot: orly 18:42:17 <bcl> so far it is looking good. 18:42:30 <bcl> Desktop, KDE, LXDE all work fine. 18:42:32 <rbergeron> okay. so we just need to basically ping dgilmore when we think we've got everything 18:42:40 <rbergeron> bcl: you are awesome 18:42:40 <adamw> bcl: i'll see if i can reproduce your Xfce install fail 18:43:29 <adamw> just to let everyone know, since we've got some time now, i'll probably screw off golfing this afternoon. 18:43:45 <nirik> so, when is the next go/no-go? a week from today? 18:43:53 <rbergeron> nirik: yes 18:44:01 <rbergeron> nirik: per evilbob's email - is this really a huge conflict with the irc meeting 18:44:05 <rbergeron> i mean it's in a different channel 18:44:08 <rbergeron> and ... yeah. 18:44:17 <nirik> not really, IMHO 18:44:35 <rbergeron> yeah, MHO2 18:45:10 <bcl> I like thursdays much better. 18:45:40 <rbergeron> #info Next Go/No-Go will be 2012-05-24, same bat time, same bat channel (17:00 UTC, 13:00 eastern, 10:00 pacific) 18:45:41 <nirik> I can propose moving those later or eariler... 18:46:16 <rbergeron> nirik: i don't think it's a big deal, it's not that often, and i don't think there is a huge overlap in attendees 18:46:33 <rbergeron> Okay, anything else? 18:46:46 * rbergeron puts a bottle of something yum in adamw's golf bag thingy 18:47:18 <adamw> christ, i'm bad enough without that 18:47:49 <rbergeron> it's supposed to improve things 18:48:06 <rbergeron> Okay, going once, going twice.... 18:48:13 * rbergeron thanks everyone for coming today 18:48:34 <rbergeron> #endmeeting