21:00:24 <rbergeron> #startmeeting F16 Beta Go No Go Meeting Round two
21:00:24 <zodbot> Meeting started Wed Sep 28 21:00:24 2011 UTC.  The chair is rbergeron. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:24 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
21:00:45 <rbergeron> #meetingname F16 Beta Go No Go
21:00:45 <zodbot> The meeting name has been set to 'f16_beta_go_no_go'
21:00:47 <jdulaney> rbergeron:  What's up?
21:00:53 <rbergeron> #topic Who is here?
21:00:54 * nirik waves.
21:01:03 <rbergeron> #info present: jdulaney, nirik, rbergero
21:01:22 <rbergeron> #chair adamw tflink
21:01:22 <zodbot> Current chairs: adamw rbergeron tflink
21:01:35 * rbergeron waits around for our illustrious qa peeps
21:01:43 <rbergeron> in the meantime...
21:01:47 <rbergeron> #topic Why are we here?
21:01:52 * hno observes
21:01:54 <tk009> */me watches from the bleachers
21:02:28 <rbergeron> #info Dearly beloved, we are gathered here today to see whether or not the beta release criteria have been met.
21:02:43 <rbergeron> #link http://fedoraproject.org/wiki/Fedora_16_Beta_Release_Criteria
21:03:02 * jdulaney plays some Hammond Organ
21:03:25 <rbergeron> Sooooo
21:03:27 * rbergeron whistles
21:04:04 <dgilmore> rbergeron: im here
21:04:12 <rbergeron> dgilmore: hi
21:04:14 <dgilmore> rbergeron: sitting next to spot
21:04:20 <rbergeron> tell spot I hope he brought his liver
21:04:22 <rbergeron> I MEAN
21:04:25 <rbergeron> UMMM
21:04:27 <rbergeron> hi spot!
21:04:50 * rbergeron is a-waiting on tflink/adamw
21:04:59 <adamw> yo
21:05:09 <rbergeron> ah, there's one half o that equation.
21:05:17 <adamw> sorry, busy working on the bug that means the release criteria haven't been met (BIG SPOILER!)
21:05:22 <rbergeron> Okay, so... let's... begrudgingly... start.
21:05:29 <rbergeron> #topic Current release blockers
21:05:37 <jdulaney> adamw:  You want to be QA?
21:05:38 <rbergeron> #link https://fedoraproject.org/wiki/Current_Release_Blockers
21:05:45 * jdulaney could do so if needed
21:05:47 <rbergeron> adamw: want to take us through my favorite number?
21:05:55 * jsmith is here, slightly late
21:06:17 <adamw> can you guys give me like 2 minutes?
21:06:22 <adamw> i just need to confirm a fix for pjones
21:06:33 * nirik turns on the elevator music.
21:06:33 <rbergeron> adamw: yes
21:06:40 <rbergeron> I will wait even longer than that :)
21:06:44 <pjones> yeah, and I just need to run to the liquor store
21:06:56 <rbergeron> pjones: join the club
21:07:05 <dgilmore> pjones: is it that bad?
21:07:06 <jdulaney> LOL
21:07:10 <jdulaney> Bring me some rum
21:07:12 <adamw> dgilmore: worse!
21:07:17 <jdulaney> Lovely
21:07:27 * adamw slugs back another jack-and-codeine
21:07:28 * rbergeron senses a grouphug coming
21:07:30 <pjones> dgilmore: miscompilation is not my idea of a good day.
21:07:45 <pjones> dgilmore: also I like booze
21:07:57 * jsmith twiddles his thumbs and looks out the window
21:08:00 <adamw> okay
21:08:11 <rbergeron> jsmith: at least your view is of milan :)
21:08:18 <dgilmore> pjones: :)
21:08:29 <adamw> let's get the small fry out of the way here
21:08:31 <jsmith> rbergeron: Aye, but it's also 11:00pm, and it has been a long day :-)
21:08:32 <rbergeron> adamw: yes? /me notes you have a chair pound symbol
21:08:36 <dgilmore> my view is logan airport
21:08:48 <adamw> okay
21:08:57 <adamw> #topic http://bugzilla.redhat.com/show_bug.cgi?id=727522
21:09:05 <adamw> long story short, this should probably be set VERIFIED
21:09:22 <adamw> the claimed reproductions with rc2 and rc3 looks to be pilot error.
21:09:50 <adamw> tflink and i both confirmed we can do nfs installs when we set up the nfs server right and be sure to pass the correct path, and so did southern_gentleman just now.
21:10:02 <adamw> i can see obvious things that the testers who 'reproduced the bug' did wrong in their logs.
21:10:06 <adamw> so, not worried there.
21:10:08 * rbergeron doesn't even see it in blockers list
21:10:12 <rbergeron> anymore
21:10:19 <adamw> it's #2 in approved blockers.
21:10:24 <rbergeron> oh
21:10:26 <rbergeron> i'm blind
21:10:29 <rbergeron> PLEASE PROCEED
21:10:42 <jdulaney> So, we good here
21:10:58 <jsmith> WORKSFORME
21:11:05 * nirik nods. good
21:11:25 <adamw> okay
21:11:31 <adamw> sorry, had to divert attention to pjones for a sec
21:11:50 <adamw> #agreed 727522 is really fixed, the apparent rc2/rc3 reproducers were pilot error
21:11:53 <rbergeron> hooray
21:11:53 <adamw> #topic http://bugzilla.redhat.com/show_bug.cgi?id=737731
21:12:18 <adamw> this one is a valid blocker and is not fixed yet. however, reserve your panickin' pants for later, because it's a preupgrade bug, and preupgrade bugs are Special.
21:12:39 <adamw> they get fixed in the previous releases, not the release under preparation
21:12:44 * rbergeron nods
21:12:48 <adamw> so this does not block f16 image creation. it will be fixed by an f15 update.
21:13:06 <rbergeron> #info 737731 doesn't block f16 image creation, will get fixed by an F15 update.
21:13:15 <adamw> this is something we've done several times before, so it's not something to worry about at all. i should probably write an exception for preupgrade bugs explicitly into the relevant procedures, though.
21:13:17 <dgilmore> ok
21:13:30 <jdulaney> adamw +1
21:13:43 <adamw> one more small fry on the proposed blocker list:
21:13:45 <adamw> #topic http://bugzilla.redhat.com/show_bug.cgi?id=741655
21:14:12 <adamw> this is clearly a nutty layout that doesn't block beta. actually i think we've gathered enough votes and -1ed it as a beta blocker, but current_release_blockers hasn't updated.
21:14:42 <tflink> but isn't updated, updating now
21:14:46 <tflink> bug
21:14:50 <adamw> okay.
21:14:58 <rbergeron> ack
21:15:02 <nirik> ack here
21:15:08 <rbergeron> yeah, we have -5 on blocker
21:15:08 <jsmith> iack
21:15:12 <adamw> no, i did update it
21:15:19 <adamw> i changed it to F16Blocker rather than F16Beta
21:15:32 <adamw> as it could be considered a final blocker, since we have a fairly optimistic final criterion that any vaguely sane partition layout should work.
21:15:38 <adamw> anyhoo, we don't need to worry about it now.
21:15:41 <rbergeron> right.
21:15:44 <rbergeron> next!
21:15:51 <adamw> #agreed 741655 is not a beta blocker as it's a crazy custom partition layout, moved to final
21:15:55 <rbergeron> yeah, that.
21:15:55 <adamw> so, onto the very big fry
21:16:14 <adamw> #topic http://bugzilla.redhat.com/show_bug.cgi?id=741781
21:16:25 <adamw> this is the big bug that exploded yesterday morning just as things were looking nice.
21:16:46 <adamw> as a very simple approximation, EFI-based install in 16 Beta RC3 (and all beta candidate builds) is utterly screwed. there is about zero chance of any EFI install working.
21:16:52 <dgilmore> adamw: is this the last one?
21:17:04 <adamw> there's anothe EFI-related bug we can vote on after it, but this is the big one to discuss.
21:17:23 <adamw> this turns out to be another fun bugs-behind-bugs-behind-bugs scenario.
21:17:30 * jdulaney is reading through it, looks like a def +1 for blocker
21:17:36 <adamw> pjones fixed the immediately obvious bug, which was to do with EFI video mode setting, last night.
21:17:53 <adamw> bcl and I tested it all night and this morning, and established that the initial bug was fixed but there were two other significant bugs hiding behind it.
21:18:02 <adamw> we have played out and, hopefully, fixed those bugs within the last hour or so.
21:18:29 <adamw> https://bugzilla.redhat.com/show_bug.cgi?id=741781#c19 provides an explanation of the two bugs, if you want a quickish catch-up
21:19:14 <rbergeron> which leaves us where?
21:19:15 <adamw> if we were not to fix those two bugs, a very clean EFI install to a pretty simple system config has a shot at working; more complex configurations, and any install to a system which already has a Fedora entry in its EFI boot loader, will fail.
21:19:18 <rbergeron> ah
21:19:34 <adamw> so, any install over a previous Fedora EFI install, basically.
21:19:45 <nirik> fix(es) are in anaconda?
21:20:05 <dgilmore> or grub and anaconda?
21:20:09 <adamw> the fix for the initial video issue is in grub. the fix for problem #2 as defined in my comment 19 is in grub. the fix for problem #1 is in anaconda.
21:20:12 <jsmith> both, right?
21:20:21 <adamw> there's also a possible problem #3 in anaconda.
21:20:28 <adamw> pjones, am i missing anything?
21:20:37 <nirik> so, no hope of a updates.img fix. needs fixed grub too.
21:20:51 <adamw> nirik: well, it would actually probably have been better if we could keep the fix entirely in grub
21:20:59 <pjones> hmmm
21:21:02 <adamw> because it's a lot safer to fudge testing if the new build doesn't need a new anaconda
21:21:07 <adamw> but, we can't, so that ship has sailed.
21:21:10 <nirik> true.
21:21:17 <nirik> also grub1 right?
21:21:22 <adamw> yeah, exactly
21:21:29 <adamw> grub2 isn't touched by any of this
21:22:07 <adamw> that's pretty much the issue as i understand it right now.
21:22:37 <adamw> the relevant criterion is "The installer must boot and run on systems using EFI other than Apple Macs " (combined with any old 'the installed system must actually freaking boot' criterion you like)
21:22:39 <dgilmore> ok
21:22:49 <rbergeron> so: let me ask ....
21:22:58 <adamw> so this is a pretty clear-cut criteria buster unless we fudge it hugely and horribly.
21:23:01 <rbergeron> we have fixes for all things, but we need to rerun through all tests, correct?
21:23:15 <dgilmore> or just all the installer tests
21:23:17 <rbergeron> assuming we do a no-fudge scenario
21:23:31 <dgilmore> desktop validation having not changed is ok?
21:23:31 <adamw> yeeeees. with the proviso that the fixes are extremely new and it's at least possible something else might jump out and bite us in a half hour.
21:23:42 <adamw> dgilmore: we can probably keep the desktop results, yeah.
21:23:51 * nirik wonders about pulling grub1 into a new rc, and fixing anaconda in updates.img/release note? or no chance?
21:24:04 <nirik> I guess it breaks critera.
21:24:10 <pjones> it really does break criteria
21:24:12 * nirik withdraws the suggestion.
21:24:13 <adamw> it's a pretty icky fudge, in that it means we now have two somewhat different versions of beta.
21:24:20 <rbergeron> yeah
21:24:26 * dgilmore notes he is getting on a plane in ~30 mins
21:24:31 <rbergeron> dgilmore: when do you land?
21:24:38 <rbergeron> tomorrow?
21:24:39 <rbergeron> basically
21:24:40 <dgilmore> rbergeron: yes
21:25:14 <rbergeron> so... since we're floating things like fudge and whatnot...
21:25:15 <dgilmore> 12 hours or so
21:25:24 <rbergeron> and dgilmore is getting on a plane...
21:25:33 <adamw> btw, if we want to talk fudges, the obvious one is to take a mulligan on caring about EFI and just release RC3 and say 'EFI is broken. Sorry.' I think that's the only fudge that would really make much sense. any kind of 'let's sort of fix it a bit but not really' fudge is just more trouble than it's worth. either we decide EFI doesn't matter and we just forget about it, or we fix it properly, whether that means slipping or not.
21:25:41 <rbergeron> do we need dennis to kick off a new rc now (or someone)?
21:25:49 <dgilmore> adamw: i agree there
21:25:53 <adamw> for the record, qa would not be very happy with such a fudge.
21:26:00 <nirik> adamw: +1
21:26:02 <rbergeron> adamw: yeah
21:26:06 <adamw> rbergeron: we need an anaconda build first.
21:26:08 <dgilmore> adamw: i think it comes down to saying sorry efi is know broke we will fix asap
21:26:11 <nirik> I think it's bad to change our criteria when we try and use it.
21:26:11 <dgilmore> or we slip
21:26:21 <jsmith> nirik: +1
21:26:36 <adamw> dgilmore: we could very vaguely possibly fix it and not slip if we work like fuck and delay the go/no-go call till tomorrow.
21:26:57 <dgilmore> adamw: we could. i could kick off the compose now
21:27:04 <nirik> although we in the past have said any no-go is a week slip. ;)
21:27:06 <adamw> dgilmore: you would have trouble, as there isn't a fixed anaconda yet
21:27:08 <dgilmore> and nirik could put the bits in place when done
21:27:11 <jsmith> dgilmore: Assuming you had an anaconda build
21:27:15 <adamw> they're looking into problem #3
21:27:18 <dgilmore> ok
21:27:28 <adamw> problem #3, btw, is that currently anaconda doesn't write a splash location to the grub config file
21:27:36 * nirik could try and make a rc if needed. or we could draft notting
21:27:43 <adamw> which sounds harmless, but apparently can have major consequences if your firmware doesn't have a vga driver or something.
21:27:47 <jsmith> adamw: I thought they weren't doing any grub splashes at all?
21:27:52 <jdulaney> an rc would be good
21:27:55 <jsmith> adamw: Ah, there's the bit I was missing
21:28:04 <adamw> there's a blank grey image, which is somewhat different from no splash. pjones would have the technical details.
21:28:19 <jsmith> Makes sense
21:28:21 <jdulaney> You know, come to think of it
21:28:25 <jdulaney> Hmm
21:28:28 <rbergeron> okay, so, I'm looking for proposals before dennis takes off.
21:28:32 <rbergeron> not to rush y'all.
21:28:32 * jdulaney just had a thought and lost it
21:28:40 <pjones> jsmith: we can't ship with no image on uefi (because that equals having no console for various implementation reasons)
21:28:52 <rbergeron> We have a proposal out there to delay the actual end of this meeting... for... about a day.
21:28:58 <pjones> so we've got a grey one to show instead, or something.
21:29:04 <pjones> (what the image looks like isn't my department)
21:29:09 <jsmith> rbergeron: With a fallback of slipping?
21:29:12 <adamw> i think if we can get a hopefully-fixed anaconda build in the next $VERY_SHORT_AMOUNT_OF_TIME and spin up an rc4 that is an unconditionally good thing to do.
21:29:14 <rbergeron> This assumes we have a working anaconda build, and that someone else can build an RC, and that it can get tested, and that problem #3 gets fixed.
21:29:34 <jsmith> rbergeron: That's a lot of assumptions, but I'm willing to give it a shot
21:29:35 <rbergeron> adamw: so maybe we make that call in a few hours of whether or not, and dennis can designate someone to do some RC?
21:29:37 <adamw> rbergeron: a day plus as many hours as can possibly be managed.
21:29:45 <jdulaney> Could we do another go/no go at the end of the week, rather than slip a whole week?
21:29:49 * rbergeron notes: WE CAN ONLY USE THIS METHOD ONCE
21:29:58 <rbergeron> or else we will do it forever.
21:30:02 <rbergeron> becuase a day is always "the solution"
21:30:05 <jdulaney> Indeed
21:30:09 <jsmith> rbergeron: +100
21:30:11 * rbergeron notes she feels pretty dirty about it as is
21:30:20 <jsmith> rbergeron: You're not alone
21:30:23 <rbergeron> but - I feel like EFI, bless its heart... well, yeah.
21:30:24 <nirik> I can try and rc... dgilmore is giving me a rundown now.
21:30:27 <adamw> rbergeron: we are not repeating anything from this motherfucking beta cycle ever again, or else i am going to quit and start a farm in saskatchewan.
21:30:30 <nirik> and we have a sop
21:30:39 <rbergeron> I recognize the importance, obviously
21:30:52 <adamw> yeah, the importance of EFI is a slippery issue.
21:31:03 <jsmith> nirik: If you get an RC out, I can do some (limited) testing early in the morning, while you folks are sleeping
21:31:04 <adamw> it's clearly _theoretically_ highly important, and we have actually agreed to make it alpha critical for the next release.
21:31:05 <rbergeron> nirik: other than mirrors, which we've discussed, and i'll look at everything else too - we're not going to foul any other bits up by not making this call until tomorrow?
21:31:18 <rbergeron> EFI hardware needs to get obtained
21:31:22 <adamw> otoh, the fact that no-one made more fuss about the fact that it's been completely broken in every beta compose so far indicates not many people actually use it.
21:31:25 <rbergeron> or we need to locate some
21:31:28 <rbergeron> adamw: yeah, precisely
21:31:45 <rbergeron> dgilmore: are you cool with this idea?
21:31:49 <pjones> adamw: but we do expect that to be changing quite a bit in the next year or two
21:31:57 <dgilmore> rbergeron: sure
21:32:03 <adamw> also the fact that i posted a 'testing request' thread in the forums yesterday which got precisely no takers, which is a first for me. (sometimes you get no takers who actually manage to do any useful testing, but someone will always actually _try_.)
21:32:04 <rbergeron> alright.
21:32:09 <adamw> pjones: yup, indeed.
21:32:14 <dgilmore> rbergeron: we usually dont stage until thurs/fri, mirrors should be ok
21:32:18 <nirik> FYI, I was going to propose next cycle moving the go/no-go out a day.
21:32:30 <nirik> so, I am ok with trying that now, but no more than a day.
21:32:45 <rbergeron> Maybe we just allow one and one only "get out of jail free card" per cycle and keep it on wednesdays otherwise ;)
21:32:54 <rbergeron> anyway. Does someone want to make the official proposal we can all agree to?
21:32:54 <jsmith> rbergeron: I like that better
21:33:10 * jdulaney notes that he did test on EFI hardware pre-Alpha, but hasn't since then
21:33:11 <adamw> btw, i should note that it's entirely on my shoulders that this was not caught until yesterday, and i'm very very sorry for that
21:33:12 <rbergeron> If we don't have an official anaconda build by... when, we just won't try?
21:33:23 <rbergeron> Or working, i should say?
21:33:25 <brunowolff> Are there concerns with marketing having enough lead time, or is that really just a "final" issue?
21:33:30 * rbergeron looking for some love here
21:33:31 <jsmith> proposed agreed: Postpone go/no-go decision until tomorrow, pending another RC with updates to anaconda and grub
21:33:44 <adamw> rbergeron: if we don't have a compose by, say, pre-midnight tonight there's just no point.
21:33:51 <adamw> (my time)
21:33:56 <adamw> at that point we either release rc3 or slip.
21:34:02 <rbergeron> adamw: a full compose, or just anaconda built?
21:34:05 <nirik> and I think we slip.
21:34:06 <adamw> a compose.
21:34:07 <rbergeron> a full compose to test with, RC4
21:34:46 <rbergeron> Okay, so: proposed agreed: If we have a built anaconda and RC4 by midnight pacific tonight, opt to do go/no-go tomorrow. Otherwise, slip one week.
21:34:47 <jsmith> Do we have clear instructions on how to exercise these particular bits to get good testing?
21:35:06 <jsmith> rbergeron: +1 from me (althought I still feel slightly dirty)
21:35:17 <adamw> jsmith: bcl and I are fairly clear on the bug and have usable test hardware (though I have an annoying separate EFI bug which confuses things).
21:35:18 <jdulaney> rbergeron:  I'm also +1
21:35:29 <adamw> jsmith: i also have one other tester who's not around right now but will be tonight.
21:35:38 * jdulaney will test no matter what, but thinks he can dig up EFI hardware
21:35:53 <jsmith> adamw: OK... I've got a spare hard drive here in Milan, but I don't know if my hardware will be any good for testing
21:35:54 <adamw> with the 'fixed' bits, it's relatively easy for anyone with an EFI-capable system to test this specific fix, if you can find such a system, please contact me for instructions.
21:35:56 <rbergeron> dgilmore: are spot's eyes rolling in his head right now?
21:36:16 <adamw> jsmith: google your system model and/or motherboard model plus 'efi' (or 'uefi') and the results should tell you. or read the manual.
21:36:25 * nirik is +1 to the proposal
21:36:29 <tflink> +1
21:36:52 <adamw> +1
21:36:54 <dgilmore> rbergeron: he is not watching
21:36:57 <jdulaney> +1 formal
21:37:06 <rbergeron> dgilmore: he's smart :)
21:37:07 <rbergeron> okay
21:37:08 <pjones> I can be +1 to that
21:37:21 <adamw> dgilmore: is he going LALALALALALALA with his fingers stuffed in his ears?
21:37:28 <rbergeron> #agreed If we have a built anaconda and RC4 by midnight pacific tonight, opt to do go/no-go tomorrow. Otherwise, slip one week.
21:37:34 <dgilmore> adamw: not quite
21:37:47 <jdulaney> Drunk out of his mind?
21:38:05 <rbergeron> Okay. So: Midnight tonight, I will check in with adamw/whoever else, and see where we are.
21:38:21 <rbergeron> Communicating status in the meantime: I assume we can do that on #fedora-qa?
21:38:29 <rbergeron> I'd like to make sure we all have a good idea of where we are in the process.
21:38:36 <rbergeron> nirik: you are going to RC, or is notting?
21:38:39 <adamw> also, if we do get an rc, it would help out immensely to have more people helping with validation tests.
21:39:17 * jdulaney will test
21:39:27 <nirik> rbergeron: I will.
21:39:32 <rbergeron> #info if we don't have what is needed by midnight pacific, Robyn will send out the No-Go Mail.
21:39:36 <nirik> so, let me know when anaconda/grub is ready.
21:39:48 <rbergeron> #action nirik will be coordinating the RC; he needs to be told when anaconda/grub is ready.
21:39:50 <adamw> nirik: we will
21:40:04 <adamw> nirik: if the builds land relatively fast we'll do a boot.iso test build before requesting the RC
21:40:12 <nirik> ok.
21:40:13 <rbergeron> adamw: are you going to be keeping track of when that happens, I assume so?
21:40:17 <adamw> if they don't we'll just do it lottery style and fire the RC compose directly
21:40:24 <adamw> rbergeron: i'm gonna be glued to this thing for the duration.
21:40:35 <rbergeron> adamw: awesome, we will have a fun night
21:40:46 * rbergeron wasn't going to sleep anyway, so now she has an excuse!
21:40:59 * jdulaney has insomnia
21:41:00 <rbergeron> so last - we'll just be keeping up to date in #fedora-qa?
21:41:09 <rbergeron> Any questions on status / etc should be happening there?
21:41:18 <adamw> yes
21:41:21 * jsmith has to get some sleep, but will be up in six-ish hours to help
21:41:42 <rbergeron> #info For status updates/questions, please check in #fedora-qa
21:41:54 <rbergeron> Okay.
21:42:03 <rbergeron> Anything else?
21:42:07 <rbergeron> Pins and needles?
21:42:46 <adamw> oh.
21:42:47 <adamw> yes.
21:42:49 <rbergeron> Oh. So
21:42:49 <jdulaney> Needles and pins?
21:42:49 <adamw> there is something else...
21:42:57 <rbergeron> yeah, i wonder if you have the same thought I do
21:42:58 <rbergeron> go for it
21:43:06 <adamw> because i'm a freaking idiot, i tested for extra credit
21:43:10 <adamw> #topic https://bugzilla.redhat.com/show_bug.cgi?id=741817
21:43:19 <rbergeron> adamw: how many times do i have to tell you
21:43:27 <adamw> i tried installing f15 via efi (which works, btw) and then doing an anaconda upgrade to f16.
21:43:30 <adamw> it explodes.
21:43:36 <pjones> okay, I have to go now. I'll be back on irc later.
21:44:14 <adamw> whether this is a blocker depends whether we want to be strict about the upgrade criteria...is an f15-f16 efi upgrade failing bad enough
21:44:34 <adamw> i feel kind of okay about failing this one. there really can't be that many damn f15 efi installs out there that want to upgrade to f16 beta.
21:45:07 <rbergeron> is the upgrade criteria pretty comprehensive or fairly loose?
21:45:11 <rbergeron> as in
21:45:22 <hno> Is EFI an officially supported system configuration in F15?
21:45:34 <adamw> pretty much.
21:45:45 <adamw> we've had the EFI beta criterion for a while, so EFI install was a beta blocker in F15 too.
21:45:56 <adamw> rbergeron: it says:
21:46:01 <rbergeron> "feed me"
21:46:04 <rbergeron> oh.
21:46:11 <adamw> "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 "
21:46:31 <adamw> so, really, this is a grey area issue: the bug is a violation of the criteria but only in a specific configuration, i.e., EFI install
21:46:38 <adamw> and we're pretty sure that specific configuration is pretty rare.
21:47:13 <hno> Does it explode as in crashing the existing system beyond repair, or just fails to upgrade?
21:47:49 <adamw> i didn't check. i think it would leave the existing system intact, but i'm not 100% sure.
21:48:02 * rbergeron wonders if the upgraded system is going to need to have additional updates to make the upgrade work
21:48:30 <rbergeron> err
21:48:33 <rbergeron> the system to be upgraded
21:48:36 <adamw> it happens at the 'what to do with bootloader' selection step, which I *think* is before it writes anything.
21:48:52 <adamw> rbergeron: no, this just looks like an anaconda bug.
21:49:05 <rbergeron> hrmm
21:49:10 * jdulaney must go
21:49:11 <rbergeron> well
21:49:15 <rbergeron> jdulaney: thanks, as always
21:49:16 <jdulaney> Will be back to test later
21:49:20 <adamw> i'd say for now we could vote on it as if it leaves the f15 system bootable
21:49:29 <adamw> and i'll re-check it later and make sure that's the case
21:49:52 <jdulaney> -1 blocker
21:49:54 <jdulaney> Peace
21:50:52 <rbergeron> i'm -1 also - i can see efi as a fresh install blocking, but it seems like a pretty rare case for upgrading, not very standard
21:50:56 <adamw> so yeah, i'm probably -1 on that basis, it's just too niche a config to worry overmuch about upgrades at beta stage.
21:51:06 <adamw> we should probably fix it for final, though.
21:51:20 <adamw> or else 15/efi users might just be stuck. i didn't test a yum upgrade.
21:51:30 <adamw> (i'd guess it would work, but then, i'm an optimist.)
21:51:40 <rbergeron> codeine bottle is half full!
21:51:48 <adamw> not any more, it ain't.
21:51:48 <rbergeron> Okay.
21:51:51 <rbergeron> LOL
21:51:52 * rbergeron hugs adamw
21:52:14 <rbergeron> okay, so, make a proposal, I think hwe have 3 -1's, or just assume we're all -1 with you.
21:52:17 * dgilmore thinks he has nirik set
21:52:18 <VileGent> +1 final blocker not beta
21:52:22 * dgilmore needs to run
21:52:31 <rbergeron> dgilmore: have a good flight
21:53:00 * nirik is fine with this being -1 blocker.
21:53:02 <adamw> okay
21:53:14 <adamw> propose #agreed 741817 not a blocker as long as it leaves f15 install bootable, adamw will check that
21:53:20 <rbergeron> +1
21:53:24 <nirik> aCk
21:53:33 <VileGent> +1
21:53:43 <jsmith> ACK
21:53:56 <adamw> #agreed 741817 not a blocker as long as it leaves f15 install bootable, adamw will check that
21:54:04 <rbergeron> okay, i'm going ot move on to mine then
21:54:13 <rbergeron> #topic Re-No/Go Meeting time
21:54:34 <rbergeron> So assuming all our pieces fall into place tonight - we still need to, um, "finish" this go/no-go.
21:54:41 <rbergeron> Same time, same place tomorrow? Or?
21:54:46 <rbergeron> Super late tonight?
21:54:48 <rbergeron> Thoughts?
21:54:55 <VileGent> tomorrow
21:55:17 <nirik> proposal: go maybe. meet tomorrow. ;)
21:55:21 <VileGent> tomorrow still gives us time to seed by tuesday
21:55:48 <adamw> yes, qa votes postpone decision.
21:55:49 <rbergeron> nirik, adamw, tflink, jsmith: will one of you run this puppy this time tomorrow? I'll be on a plane.
21:56:00 <adamw> if i'm not snoring on my keyboard, yeah.
21:56:09 <rbergeron> okay.
21:56:09 <jsmith> I should be around as well
21:56:32 <rbergeron> #agreed will meet tomorrow, same bat time, same bat channel, to do final go/no-go decision.
21:56:44 <rbergeron> #info robyn will be out tomorrow
21:56:57 <rbergeron> Okay.l
21:57:00 <rbergeron> Anything else
21:57:00 <rbergeron> ?
21:57:02 * rbergeron counts backwards
21:57:08 <rbergeron> 88...
21:57:10 <adamw> let's catch all the journalists and shoot them so word of this never escapes.
21:57:10 <rbergeron> 67....
21:57:18 * adamw looks around for a 'larabel'
21:57:49 <rbergeron> 22...
21:57:52 <rbergeron> okay, kids.
21:57:56 * rbergeron is going to end meeting.
21:58:17 <rbergeron> adamw, tflink: I'll see you guys ALL NITE LONG BABY
21:58:19 <rbergeron> #endmeeting