=== zul [n=chuck@ubuntu/member/zul]  has joined #ubuntu-boot
[12:10] <zul> hey
=== BenC [n=bcollins@debian/developer/bcollins]  has joined #ubuntu-boot
[03:05] <BenC> any chance of there being enough time for a -X.11 kernel upload before flight 2 cd's are ready?
[03:07] <jbailey> BenC: You might want to try a nick highlight for him.
[03:07] <jbailey> Unless you're trying to hide to avoid him killing you for the thought. =)
=== BenC bucks up
[03:08] <BenC> Kamion: ping
[03:09] <Kamion> BenC: hey - what's the issue? I was about to build images like *right now*
[03:09] <Kamion> if it's not world-shattering-kills-babies I'd prefer not
[03:10] <BenC> nah, go ahead
[03:10] <BenC> I didn't want to stall the build, but if it was going to be > 24 hours, I was going to try to get another kernel out
[03:13] <Kamion> ah, no, I want to do the builds now and if possible test a few before crashing, then release in the morning
=== jbailey [n=jbailey@modemcable139.249-203-24.mc.videotron.ca]  has joined #ubuntu-boot
[03:13] <Kamion> just waiting for this cron.daily to finish so that I have an xserver-xorg-driver-via that works on my i386 test laptop, and then I'll start
[03:18] <BenC> ok
[03:18] <BenC> will there be some place I can download these from before morning?
[03:18] <BenC> and is ppc expected to work?
[03:19] <Kamion> http://cdimage.ubuntu.com/daily/current/, http://cdimage.ubuntu.com/daily-live/current/
[03:19] <Kamion> powerpc should work, modulo any screwups in the fixes lamont and I hacked together earlier
[03:20] <Kamion> I'll try to check powerpc before crashing
[03:21] <Kamion> note that the new live CD is very light on feedback as it's booting, for the moment
[03:21] <Kamion> it'll eventually use usplash but at the moment there's a long period of no output
[03:22] <Kamion> and then some fun slews of kernel traces :)
[03:25] <BenC> is there a faster place to get it from? :)
[03:26] <BenC> I'm only get like 5-8k/sec from there
[03:26] <BenC> kernel traces are from unionfs?
[03:27] <Kamion> think so, and no, no faster place
[03:27] <Kamion> if you keep up to date fairly regularly then you can rsync
[03:27] <Kamion> cdimage.ubuntu.com::cdimage/daily-live/current/dapper-live-powerpc.iso etc.
[03:28] <Kamion> both the install and live CDs are fairly well rsyncable
[03:29] <Kamion> well, actually, you could try some of the ones on https://wiki.ubuntu.com/Archive
[03:29] <Kamion> not many people mirror all of cdimage though
[03:30] <Kamion> mostly those are just releases.u.c mirrors
[08:13] <fabbione> Kamion: i just tested rescue with the latest installer.. i still can't get to it. It appears something broke it
[08:14] <fabbione> and yes, rescue/enable is set to true
=== HiddenWolf [n=HiddenWo@136.17.dynamic.phpg.net]  has joined #ubuntu-boot
[08:29] <fabbione> Kamion: never mind.. it still misses the rescue udeb
[08:30] <fabbione> (in the build)
[08:31] <fabbione> but i didn't bother to check it, because iirc you said that it was on the list of fixed stuff... i might be 100% wrong tho :)
[09:31] <Kamion> CD install?
[09:33] <infinity> 'Morning, Kamion.
[09:33] <infinity> How did your battles go last night?
[09:33] <Kamion> GAH THREE HOURS SLEEP
[09:33] <infinity> Dude.
[09:33] <infinity> Back.  To.  Bed.
[09:34] <Kamion> can't, wife needs me to drive her and child to school
[09:34] <Mithrandir> yeah, get back to bed.
[09:34] <Kamion> (grr)
[09:34] <Mithrandir> drive, then bed.
[09:34] <infinity> Give keys to wife.
[09:34] <infinity> Bed.
[09:34] <Kamion> will be awake by then, no point
[09:34] <Mithrandir> or taxi.
[09:34] <Kamion> she can't drive yet
[09:34] <Kamion> image candidates are up, please test them :)
[09:34] <fabbione> Kamion: nah never mind me please.. it is still the monolith build.. it doesn't pull in the rescue-mode udeb and i forgot to add it manually.
[09:34] <infinity> No time for testing right now, I'm pitti's bitch today.
[09:34] <fabbione> Kamion: it also asks for casper-checks (that i had to remove from the list)
[09:34] <infinity> When he's done with me, you can have me back, though.
[09:35] <Kamion> casper-check will disappear soon
[09:35] <fabbione> Kamion: downloading images as we speak
[09:35] <Mithrandir> Kamion: just the daily-live/current, right?
[09:35] <fabbione> Kamion: yup.. no problem.. i just killed it.. i don't need it at all for rescue testing
[09:35] <Kamion> Mithrandir: yes
[09:41] <infinity> Kamion : Are you expecting to want buildd response time and CPU power over the next day?
[09:41] <infinity> Kamion : I assumed yes, hence I've been holding off on doing anything like mass-give-backs.
[09:52] <Mithrandir> Kamion: /scripts/casper is not in the initramfs, at least for amd64.
[09:52] <Mithrandir> so, "completely broken"
[09:54] <infinity> My fault, or yours?
[09:56] <Mithrandir> yours, most likely.
[09:56] <infinity> LIES.
[09:57] <Mithrandir> root@xoog:/$ dpkg -l casper
[09:57] <Mithrandir> No packages found matching casper.
[09:57] <Mithrandir> casper is not installed in the image at all
[09:57] <infinity> Oh, then that livecd was built using an old livefs cloop.
[09:57] <Mithrandir> quite conceivably.
[09:58] <infinity> Let me just check recent logs and make sure casper's definitely in there now.
[09:59] <infinity> Err, oh.
[09:59] <Mithrandir> please do.
[09:59] <Mithrandir> can you kick off live cd builds too?
[09:59] <infinity> Yeah, but.  No one re-uploaded *-meta after Colin seeded casper, I bet.
[09:59] <infinity> And the build process uses the metapackages, not the seeds directly.
[10:00] <Mithrandir> ok, he should be back in a bit anyway.
[10:00] <Mithrandir> well, off to work.
[10:05] <Mithrandir> Kamion: shouldn't the disc label include something like "prerelease" and not just be "Ubuntu 6.04"?
[10:12] <Kamion> infinity: I thought casper had been added in manually
[10:13] <Kamion> Mithrandir: yes
[10:13] <Kamion> I thought it said "Alpha" somewhere
[10:13] <Kamion> infinity: -live isn't sufficient because that doesn't affect the base livefs
[10:13] <infinity> Kamion : Looks like lamont added it to "base" (the one place where seeds can't help, cause there's no "base-live" metapackage), but not to the other builds.
[10:14] <Kamion> oh, sigh, do I really have to upload ubuntu-meta?
[10:14] <infinity> Of course, ubuntu-live-base (which all live images could include) would work... Or I can just put it in manually.  I don't care.
[10:14] <Kamion> if you could grit your teeth and hardcode it just for today I'd really appreciate it
[10:14] <infinity> Oh, I'll just hardcode it for now.
[10:14] <Kamion> thanks
[10:14] <infinity> No big deal.
[10:14] <infinity> Just thinking out loud about the most "elegant" solution for later.
[10:14] <infinity> I'll hardcode it and kick off builds momentarily.
[10:18] <infinity> If "ubuntu-base-live" would be nothing but casper, may as well unseed it from all the live seeds, and I'll just leave it hardcoded.
[10:21] <infinity> Oh baz, why do you hate me so?
[10:24] <infinity> Kamion : New builds kicked off.
[10:30] <Kamion> thanks
=== thundr [n=thundr@66-168-50-185.dhcp.mdsn.wi.charter.com]  has joined #ubuntu-boot
=== lbm [n=lbm@cpe.atm4-0-1301006.0x50a0824e.vgnxx6.customer.tele.dk]  has joined #ubuntu-boot
[11:30] <Kamion> infinity: livefses seem to have built - all ok?
[11:31] <infinity> No... I'll need to re-do a few. :/
[11:32] <Kamion> oh, which broke?
[11:33] <infinity> Oh, feh.  lamont's hack to copy multiple nitrds out for architectures with more than one broke.
[11:33] <infinity> So powerpc is broken.
[11:33] <infinity> Well, the extracted initrd is "broken", the cloop is fine.
=== infinity looks over other logs.
[11:35] <Kamion> that was my hack ...
[11:35] <Kamion> what's wrong with the initrds? they looked OK last night
[11:35] <infinity> Oh, the nyour hack was wrong. :)
[11:36] <Kamion> doesn't surprise me
[11:38] <infinity> Ugh.  Bear with me.  Wireless is acting up.
[11:38] <infinity> ln: creating symbolic link `livecd.kubuntu.initrd' to `livecd.kubuntu.initrd-powerpc\n2.6.15-8-powerpc64-smp': File exists
[11:38] <infinity> Oh, that's proabbly harmelss.
[11:38] <infinity> harmless too.
[11:38] <infinity> Assuming the two got copied out, the link logic is just broken.
[11:39] <infinity> Of ocurse, it trips the set -e
=== infinity tidies.
[11:39] <infinity> Do you care about base images?
[11:40] <infinity> i386/base failed cause it ran during cron.daily.
[11:40] <infinity> Same with i386/edubuntu.
[11:40] <infinity> I'll re-run both of those.
[11:40] <infinity> And fix up powerpc and re-run it.
[11:41] <Kamion>     if [ "${KVERS% *}" = "$KVERS" ] ; then
=== Kamion wonders why that didn't trip
[11:41] <Kamion> base> no
[11:41] <Kamion> edubuntu> yes
[11:41] <infinity> You were too late.  Ran both.
[11:41] <infinity> Base is fast anyway.
[11:42] <Kamion> ah, I see, not a space, a newline
[11:42] <infinity> <nod>
[11:42] <infinity> How do you split on newline in shell?
[11:43] <infinity> \n?
[11:43] <Kamion> NUMKVERS="$(set -- $KVERS; echo $#)"
[11:43] <Kamion> if [ "$NUMKVERS" = 1 ] ; then
[11:43] <Kamion> easiest to just let the shell do it like that
[11:44] <Kamion> (tested better this time)
[11:49] <infinity> Okay, powerpc running gain.
[11:49] <infinity> again, too.
[11:54] <infinity> So, right now, ubuntu and kubuntu should be a go on i386/amd64
[11:54] <infinity> Waiting on edubuntu for i386/amd64, and all 3 for powerpc.
[11:55] <Kamion> */amd64 has to wait for my new gfxboot-theme-ubuntu to hit the archive, otherwise rescue mode is broken
[12:07] <infinity> I really wish these machines had faster disks...
[12:13] <Mithrandir> or just a huge ramdisk.
[12:13] <Mithrandir> AWTY?  As in, any images I can test, yet?
[12:15] <Kamion> hmm, I can probably build a subset and build amd64 later
[12:16] <Kamion> except even Ubuntu/powerpc isn't done yet
[12:16] <infinity> It's clooping it right now.
[12:16] <infinity> (powerpc/ubuntu, that is)
=== infinity gets out and pushes.
[12:18] <HiddenWolf> infinity, there is a solution where you put an extention card on your ide controller and fill the card with ram. PC will then see the ram as a harddisk,
[12:18] <HiddenWolf> infinity, perhaps worth consideration.
[12:18] <infinity> ...
[12:18] <Mithrandir> HiddenWolf: it's slow-ish, though.
[12:18] <infinity> That's way too much effort. :)
[12:19] <HiddenWolf> Mithrandir, slower than flash, faster than hdd's
[12:19] <infinity> RAM disks would be best, except we don't have enough RAM in these boxes to shuffle around 2 gig images.
[12:19] <Kamion> infinity: whoa, why is it publishing livecd.* (base, edubuntu, kubuntu, ubuntu) under /LiveCD/dapper/ubuntu/?
[12:20] <infinity> ...
[12:20] <Kamion> http://royal.buildd/~buildd/LiveCD/dapper/ubuntu/20051214.3/
[12:21] <infinity> Because something, or someone, is special.
[12:21] <Kamion> anyway, i386/powerpc building
[12:21] <Mithrandir> HiddenWolf: http://www.tomshardware.com/2005/09/07/can_gigabyte/
[12:22] <infinity> Or, specifically, royal is special?
[12:22] <Mithrandir> they only get 100MB/sec or so out of it, which is silly
=== infinity scratches his head.
[12:22] <Mithrandir> the latency is excellent, though
[12:24] <infinity> Kamion : Oh, wait.  I know.  That's the script not cleaning up after itself, cause it SO doesn't expect to fail where we had it failing.
[12:24] <infinity> Kamion : So, on the next run, when it blindly does "mv livecd.*", it got all the old images from the previous runs.  I'll just delete them.
[12:25] <Kamion> hah, ok
[12:25] <Kamion> ta
[12:25] <infinity> There.  Like it never happened.
[12:26] <infinity> I guess that script needs a trap and a cleanup. :)
[12:30] <infinity> Yup, definitely need some cleanup there.
[12:31] <infinity> Or to never, ever fail in that spot, ever again.
[12:32] <Kamion> Mithrandir: try i386 or powerpc now
[12:32] <Mithrandir> ENPOWERPC, but I'm pulling the i386 image now.
[12:33] <Mithrandir> will take a while, though, as I don't have a fresh one around
[01:31] <Kamion> infinity: how're the other builds going?
[01:31] <infinity> It's all done, except for powerpc/base.
[01:31] <infinity> So, if you're not building base images, you're good to go.
[01:32] <infinity> powerpc/base will be done shortly anyway.
=== thundr [n=thundr@66-168-50-185.dhcp.mdsn.wi.charter.com]  has joined #ubuntu-boot
=== jbailey [n=jbailey@modemcable139.249-203-24.mc.videotron.ca]  has joined #ubuntu-boot
=== BenC [n=bcollins@debian/developer/bcollins]  has joined #ubuntu-boot
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin]  has joined #ubuntu-boot
=== zul [n=chuck@CPE0006258ec6c2-CM000a73655d0e.cpe.net.cable.rogers.com]  has joined #ubuntu-boot
=== HiddenWolf [n=HiddenWo@136.17.dynamic.phpg.net]  has joined #ubuntu-boot
=== BenC [n=bcollins@debian/developer/bcollins]  has joined #ubuntu-boot
=== BenC [n=bcollins@debian/developer/bcollins]  has joined #ubuntu-boot
=== thundr [n=thundr@66-168-50-185.dhcp.mdsn.wi.charter.com]  has joined #ubuntu-boot
=== BenC [n=bcollins@debian/developer/bcollins]  has joined #ubuntu-boot
[10:38] <BenC> Kamion: flight2 ppc64 booted fine on my G5
[10:39] <Kamion> great, thanks
[10:39] <BenC> I did have one oops from the vfs (guess it was unionfs related)
[10:39] <BenC> and on shutdown/reboot it oopses again on unmount, and locks up
[10:43] <Kamion> mm I noticed that too
[10:53] <BenC> doesn't do that on i386?
[10:53] <Mithrandir> it oopses a bit, but copes.
[10:53] <BenC> sounds the same as my boot on ppc64
[10:53] <Mithrandir> BenC: look at it as motivation to fix the bugs.
[10:53] <Mithrandir> :-)
[10:54] <BenC> if only it looked easily traceable :)
[10:54] <BenC> the oops is in the vfs, so I have to trace it into unionfs
[10:54] <Mithrandir> it's easily reproducable, at least.
[10:54] <BenC> got a repo case I can do on an existing install?
[10:54] <BenC> repro
[10:55] <Mithrandir> no, I haven't had the time to track it down, I just see it happens on each and every live cd boot
[10:59] <BenC> new version of unionfs, I'll try that first :)
=== Keybuk [n=scott@descent.netsplit.com]  has joined #ubuntu-boot
[11:57] <Nafallo> Keybuk: hey! will we see streamlined boot soon or do you need someone to test it first? :-)
[11:57] <Keybuk> yes, was waiting for Flight 2
[11:57] <Keybuk> now I can do it while Colin sleeps for the rest of the week <g>
[11:58] <Nafallo> hehe, nice :-)
[11:59] <Nafallo> will be in queue though :-P
=== Nafallo bets daniels xorg still builds on all arches ;-)
[11:59] <Keybuk> queue?
[11:59] <Nafallo> buildqueue :-)
[11:59] <Keybuk> sure, but that don't take long
[12:00] <Nafallo> if you say so :-)