[12:10] <lifeless> Burglaptop: thats ideal.
[12:10] <Burglaptop> lifeless: which is ideal?
[12:10] <lifeless> Burglaptop: forwarding as attachment preserves gpg sigs.
[12:10] <lifeless> but if you want to change, check the composer preferences
[01:09] <ryanpg> daniels, hi there... I just wanted to drop by and let you know that building and installing xf86-video-ati-X11R7.0-6.5.7.3 gets rid of bug #21382
[01:15] <pitti> good night everyone
[01:18] <TerminX> is there any fix for bug 21055 (evdev breaks xkb and gnome-settings-daemon) other than not using evdev?
[01:21] <daniels> ryanpg: awesome
[01:22] <daniels> ryanpg: dunno wtf happened there -- must be a missing b-d on render or something.  the code didn't change a bit.
[01:22] <ryanpg> daniels, yeah I don't know what happened either... so I guess I'll just close the bug I opened? or should I leave it until dapper is updated?
[01:22] <daniels> ryanpg: leave it until it's fixed in the archive, I guess ;) i'll check over all the b-ds more carefully
[01:23] <ryanpg> will do... I'll comment that it's fixed for me though and how
[01:32] <dobwan> exit
[01:32] <dobwan> .exit
[02:27] <crimsun_> elmo: please sync cegui-mk2, debsums, enlightenment, tqsllib, and teg from Sid, overriding Ubuntu changes. Thanks.
[02:29] <daniels> TerminX: not at this stage
[03:03] <mdke> Riddell, around? I was wondering if we can get the -docs update in before xmas?
[03:03] <mdke> or have people disappeared already for the holidays?
[03:04] <jsgotangco> heh
[03:04] <jsgotangco> don't be surprised
[03:04] <mdke> jsgotangco, Riddell doesn't take holidays or sleep
[03:05] <daniels> a few people have already disappeared for the holidays; most are disappearing at the end of the day today)
[03:05] <floam> TerminX: not that I know of
[03:05] <floam> TerminX: It's killing me too :(
[03:05] <jsgotangco> -_-;
[03:06] <jsgotangco> i guess he hibernates a few times
[03:07] <mdke> daniels, thanks
[03:07] <mdke> daniels, btw, the synaptics driver still not installed by default here, are the seeds fixed yet?
[03:07] <TerminX> floam: :(
[03:07] <floam> daniels: I just added a backtract to #21055, but it doesn't really look useful
[03:07] <floam> backtrace, even
[03:07] <daniels> mdke: no, I've been crap
[03:08] <mdke> daniels, i'm sure it's not the only bug on your plate :)
[03:08] <floam> plus it's probably a dupe of 20052, same issue
[03:08] <TerminX> yeah, can't even use a damn GTK2 theme without starting gnome-settings-daemon and halting the process before it gets a chance to screw up
[03:08] <daniels> floam: correct
[03:08] <daniels> mdke: far from it
[03:09] <mdke> indeed
[03:30] <lguerra> Hi, 
[03:30] <mdke> hi lguerra 
[03:31] <lguerra> Im test Linux ubuntu 2.6.15-8-386 #1 PREEMPT Tue Dec 13 03:21:39 UTC 2005 i686 GNU/Linux and application menu dont work
[03:31] <mdke> lguerra, you'll need to file a bug in bugzilla... there are not many people around right now
[03:32] <lguerra> my english is too bad, im read but i dont write too much
[03:32] <mdke> lguerra, it's fine :)
[03:32] <lguerra> how i put a bug in bugzilla?
[03:33] <lguerra> mdke
[03:34] <jdub> BenC: ping
[03:34] <mdke> lguerra, you go to http://bugzilla.ubuntu.com, register and then search the database for a similar bug.
[03:34] <mdke> then if you don't find one, you click "New" and follow the on-screen instructions
[03:34] <lguerra> ok thks
[03:34] <BenC> jdub: pong
[03:35] <jdub> BenC: is this what we're using for UP/SMP? http://lwn.net/Articles/163810/
[03:35] <mdke> lguerra, a decent guide here https://wiki.ubuntu.com/BugReports
[03:35] <lguerra> ok thks again
[03:36] <BenC> jdub: no, but that code looks damn promising and much better than what I am using
[03:36] <lguerra> mdke: another question
[03:37] <jdub> BenC: rock!
[03:38] <jdub> BenC: so hard to see what everyone's doing in kernel land :|
[03:38] <BenC> it handles the complex structures, which mine doesn't (so full functions that would normally be nops on non SMP are actually take care of
[03:38] <BenC> yeah
[03:38] <BenC> I'll take a look at thi
[03:38] <BenC> s
[03:38] <BenC> it's not too late to try it
[03:39] <desrt> BenC; no love with the firewire stuff :/
[03:39] <Burglaptop> lguerra: shoort
[03:39] <BenC> desrt: feel like compiling the ieee1394 dev code?
[03:39] <desrt> BenC; can i build it using linux-headers?
[03:39] <BenC> it's on kernel.org in a git repo
[03:39] <BenC> yeah
[03:40] <desrt> BenC; or do i have to patch the kernel proper?
[03:40] <BenC> make -C /usr/src/linux-headers-2.6.15-9-686 O=`pwd`
[03:40] <lguerra> anyone speak spanish? i cant how ask mi question
[03:40] <BenC> or something like that
[03:40] <lguerra> in english
[03:40] <BenC> desrt: cd ieee1394; and then that command
[03:40] <mdke> lguerra, you can try #ubuntu-es
[03:40] <desrt> actually
[03:41] <lguerra> nobody answer me 
[03:41] <mdke> lguerra, they are probably asleep :)
[03:41] <lguerra> jeje
[03:41] <lguerra> mdke: any problem if i write in spanish the bug????
[03:41] <BenC> jdub: sweet, it handles modules too!
[03:41] <BenC> my module nop'ing is sort of broken
[03:42] <desrt> O=...
[03:42] <mdke> lguerra, i don't know if anyone will understand it, best to try english :) your english is fine
[03:42] <lguerra> ok thks
[03:43] <desrt> BenC; gonna first try desktop connected to laptop in target disk mode and laptop connected to FW enclosure (to rule out possibilities)
[03:44] <desrt> as useful as it might seem for laptops it's 10x more useful for xserves
[03:44] <desrt> interesting.
[03:44] <desrt> desktop + target disk mode = [4399369.911000]  sdc: assuming drive cache: write through
[03:44] <desrt> [4399369.911000]   sdc: [mac]  sdc1 sdc2 sdc3 sdc4 sdc5
[03:44] <desrt> ++
[03:45] <BenC> well, that doesn't rule out a bug in firewire
[03:45] <desrt> true.  but it does rule out the possibility of a faulty host controller
[03:45] <BenC> there's two different kinds of sbp2 node definitions
[03:45] <BenC> yeah
[03:46] <desrt> you were saying?
[03:46] <desrt> (about sbp2)
[03:51] <robitaille> jdub ping
[03:52] <jdub> robitaille: pong
[03:52] <jsgotangco> mr. fridge!
[03:53] <robitaille> jdub,  did you get my email the other day about a ubuntu-qc mailing list?
[03:53] <BenC> desrt: oh, just that target mode might be using one, and that your enclosure using another
[03:53] <BenC> and the other is broken
[03:53] <BenC> one is more rare than the other
[03:53] <BenC> I have two or three bug reports like yours
[03:53] <desrt> hmm
[03:54] <desrt> are you firewire upstream?
[03:54] <BenC> jdub: I'm seriously thinking about using this patch for the next upload, thanks
[03:54] <BenC> desrt: in title only, I rarely do work on it anymore
[03:54] <desrt> hmm
[03:54] <jdub> robitaille: i did; i should have replied immediately to say that i was procrastinating before creating it
[03:54] <desrt> macos is giving a weird error in dmesg when i plug the drive in
[03:55] <BenC> could just be bad firmware
[03:55] <BenC> maybe try updating that
[03:55] <robitaille> jdub,  no problem.  So how long you are planning in procrastinating about it?  :)
[03:55] <jdub> heh
[03:55] <desrt> FireWire (OHCI) apple id 31 built-in: handleSelfIDInt - received quads == 0.  issuing bus reset
[03:56] <BenC> that's just a tad ugly
[03:56] <desrt> bah
[03:56] <desrt> but it works.
[03:56] <desrt> ok.  so enclosure works in osx :(
[03:56] <jdub> robitaille: the problem is that we don't yet have a good standard for handling the rather complicated problems posed by national, language, provicial, city, locality loco team lists
[03:56] <desrt> BenC; any way to find out what 'type' of sbp2 this thing is?
[03:57] <jdub> robitaille: so whenever someone proposes a list that poses new problems in this scheme, i freeze up for a bit ;-)
[03:57] <robitaille> jdub,  I agree.     But we do have now http://ubuntu-qc.org
[03:58] <jdub> robitaille: so you'd prefer ubuntu-qc to ubuntu-ca-qc or ubuntu-quebec?
[03:58] <BenC> desrt: only way I know of is to use gscanbus and check out the config rom
[03:58] <jdub> robitaille: eeek
[03:58] <jdub> hrm
[03:58] <desrt> benc; installing
[03:58] <robitaille> jdub,  ubuntu-qc seems to follow existing scheme more than the other 2
[03:59] <jdub> robitaille: -qc is not a language code or top level domain code
[03:59] <jsgotangco> i thought it was "quality control"
[03:59] <jsgotangco> lol
[03:59] <desrt> hmm
[03:59] <jdub> ha ha ha
[03:59] <desrt> drive works fine with ubuntu on laptop.
[03:59] <BenC> really?
[04:00] <robitaille> so ubuntu-quebec?   Maybe we would like a ubuntu-quality-control one of these days
[04:00] <BenC> so one machine it works fine, and another it doesn't, same kernel?
[04:00] <jsgotangco> and i thought we've about to create a bugsquad team...*sigh*
[04:00] <desrt> i think so.
[04:00] <desrt> well
[04:00] <desrt> powerpc vs. i386
[04:00] <BenC> if that's the case, then is the broken machine amd64?
[04:00] <desrt> no.
[04:00] <BenC> so ppc works, or is broken?
[04:00] <desrt> ppc works
[04:00] <desrt> desktop (686-smp) is broken
[04:00] <BenC> that's an odd little bit
[04:01] <BenC> do any firewire devices work?
[04:01] <desrt> ya.. target disk mode worked
[04:01] <desrt> also.. this very enclosure worked when it contained a harddrive (but not CD)
[04:01] <BenC> so the i386 saw the ppc-target-mode?
[04:02] <desrt> yes.  and i386 also saw this enclosure when it contained a harddrive
[04:02] <BenC> hmm, so it's the cdrom
[04:02] <desrt> but the ppc laptop in both linux and macos sees the cdrom
[04:02] <BenC> wow, that's a sumper
[04:02] <BenC> stumper
[04:02] <desrt> quite.
[04:03] <BenC> what kernel is the ppc running?
[04:03] <desrt> -8
[04:03] <desrt> i386 is -9
[04:03] <desrt> but i386 was also broken with -8
[04:03] <BenC> any chance you could bump the ppc to -9?
[04:03] <desrt> well -9 is the one you downgraded back to stable tree, right?
[04:03] <BenC> yeah
[04:03] <desrt> that experiment was not a success :)
[04:03] <desrt> considering -9 is broken and -8 works
[04:04] <desrt> (but obviously downgrading wasn't the cause of brokenness)
[04:04] <BenC> -8 works on the i386?
[04:04] <desrt> on the ppc only
[04:04] <desrt> nothing works on 386.  everything works on ppc.
[04:04] <BenC> does -9 work on ppc?
[04:04] <desrt> upgrading now :)
[04:04] <BenC> just want to make sure, for my own sanity
[04:04] <BenC> thanks
[04:05] <desrt> it's pretty weird
[04:05] <desrt> everything works except for the exact combination of my PC + the drive enclosure + CD-ROM drive
[04:05] <desrt> all other combinations are fine
[04:06] <desrt> and it appears to not depend on the kernel version
[04:06] <desrt> (although i know this used to work... circa 2.6.8ish)
[04:19] <desrt> neat
[04:19] <desrt> i just tossed a different CD-ROM drive in there and it's working now
[04:20] <desrt> so uh... uh....
[04:21] <desrt> i choose to blame either the cd-rom drive or the enclosure.....
[04:21] <desrt> it probably behaves it a strange racey sort of way with the one drive
[04:22] <jdub> robitaille: still here?
[04:22] <robitaille> jdub,  yes
[04:23] <jdub> robitaille: you ok with ubuntu-quebec?
[04:23] <robitaille> jdub,  yes
[04:23] <robitaille> are you?
[04:23] <jsgotangco> jdub, how different is the love tour from the business tour? purely advocacy?
[04:23] <jdub> i'm going to settle for a few weird historical artifacts :-)
[04:24] <robitaille> jdub:  let's say that our group had some problem with a name than contains both a "ca" and a "qc" together :)
[04:24] <jdub> jsgotangco: yeah, non-business, loco/lug community focused, etc.
[04:24] <jdub> robitaille: ha ha ha
[04:25] <jdub> crazy quebecistanis!
[04:25] <jsgotangco> haha
[04:25] <jdub> jsgotangco: basically, i do the biz/gov stuff on the side, while the asia tour is almost exclusively about business goals :)
[04:25] <jsgotangco> jdub, right hmmmm
[04:26] <jdub> jsgotangco: it's the kind of thing you and i could do in asia :)
[04:26] <jsgotangco> except europe is winning heh
[04:27] <jdub> (hrm, which is why the word 'advocates' should totally not be in the asia tour story on the fridge)
[04:27] <jdub> north america is doing well too
[04:27] <jdub> greedy!
[04:27] <robitaille> jdub,  many thanks for the list
[04:27] <jdub> what most voters don't know is that the next tour will indeed be europe ;)
[04:28] <mdke> jdub, ah cool
[04:28] <rob1> not comming as far south of asisa as australia is it?
[04:28] <rob1> hot pie is burnies
[04:29] <jsgotangco> asia-oceana
[04:46] <desrt> daniels; ping
[05:18] <jdub> BenC: why doesn't linux-source-2.6.15 build-dep on gcc-3.4
[05:18] <jdub> ?
[05:18] <jdub> oh
[05:18] <jdub> 2.6.12 does
[05:18] <jdub> BenC: we're building 2.6.15 with gcc 4?
[05:22] <infinity> jdub : Yup.
[05:22] <infinity> jdub : On all but hppa, where it's still using 3.4
[05:24] <daniels> desrt: pong
[05:24] <jdub> infinity: wow
[05:27] <Amaranth> infinity: kick ass, that'll make life much easier
[05:27] <infinity> That's the hope.
[05:27] <Amaranth> i won't have to explain to people how to make a kernel module build with gcc-3.4
[05:27] <infinity> Unless it makes life much crashier, but so far that doesn't seem to be the case.
[05:28] <infinity> (We may still have to do a last-minute switch back to -3.4 if we find a showstopper, so don't get too excited yet, but the whole 2.6.15 series has been building with 4.0, and it doesn't seem much worse than our 2.6.12 kernels with 3.4)
[05:28] <Amaranth> gotta love those gcc bugs
[05:28] <infinity> It's not just GCC bugs.
[05:29] <Amaranth> things it used to warn on now fail?
[05:29] <infinity> The kernel is some pretty touchy code that often relies on GCC misbehaviour.
[05:29] <Amaranth> ah
[05:29] <infinity> It's a constant battle to keep improving both the compilers and the kernel and keep them lockstep.
[05:29] <Amaranth> don't people compile it with icc?
[05:30] <infinity> Linux has been compiled with a variety of compilers, but I wouldn't trust some subsystems and drivers much further than I can throw an elephant unless you're using GCC (and a recommended version, no less)
[05:36] <jdub> infinity: can't wait for grumpy + icc - that'll be fun ;-)
[05:39] <fabbione> morning
[05:40] <infinity> jdub : For some unknown value of "fun" that I'm not yet familiar with.
[05:42] <fabbione> icc?
[05:42] <infinity> Intel CC.
[05:42] <jdub> fabbione: the intel one
[05:42] <fabbione> oh
[05:42] <fabbione> you kidding..
[05:42] <jdub> meanwhile, vmware is being oom-killed on this machine ;-)
[05:43] <fabbione> speaking of which.. i need to build the new modules
[05:44] <jdub> the installer script is sooooo ugly
[05:46] <Amaranth> jdub: poke chipx86 or something ;)
[05:46] <fabbione> hmmm
[05:46] <fabbione> go quickcam!
[05:46] <Amaranth> btw, broadcom driver in dapper == i can get off OS X
[05:46] <Amaranth> yay
[06:17] <mpt> fabbione, have you considered logging #ubuntu-desktop?
[06:18] <BenC> it compiled, which is always a good sign
[06:18] <infinity> alternative_smp?... Does it come with tattoos and army boots?
[06:18] <BenC> just a silly wig
[06:19] <fabbione> mpt: no, becuase if people don't tell/ask me about channels, it's unlikely i get to know about them. in any case my bot is in too many channels and i can't join anymore
[06:21] <mpt> ok
[06:21] <BenC> well, it booted, and gdm+nvidia started
[06:22] <fabbione> new pack!
[06:22] <fabbione> hey BenC 
[06:22] <fabbione> how big is the new pack?
[06:22] <jdub> BenC: rock
[06:23] <BenC> new pack?
[06:23] <fabbione> pack/pack-2dae6bb81ac4383926b1d6a646e3f73b130ba124.pack
[06:23] <BenC> I don't know
[06:23] <BenC> must be automatic from kernel.org
[06:23] <fabbione> probably
[06:23] <BenC> rsync sucks for git
[06:23] <BenC> especially when it repacks
[06:23] <fabbione> rsync sucks.
[06:24] <fabbione> i start my u.c rsync at 400K
[06:24] <fabbione> after a few 100MB i am down at 20K/s
[06:24] <fabbione> it exponetially slows down
[06:24] <BenC> "SMP alternatives: switch to UP code"
[06:24] <BenC> rock
[06:25] <fabbione> BenC: is that SMP2UP on amd4?
[06:25] <fabbione> amd64
[06:25] <BenC> no, that's a new alternative_smp bit jdub pointed me to
[06:25] <BenC> much better than smp2up
[06:25] <fabbione> ah
[06:25] <infinity> What's this one do?
[06:26] <BenC> has two advantages
[06:26] <BenC> 1) It does the module stuff cleanly
[06:26] <BenC> 2) It even accounts for cases where UP would have been a nop and SMP introduces an entire function (semaphores for example)
[06:26] <BenC> in the second case, it will nop the entire function
[06:26] <infinity> Nice.
[06:27] <infinity> Is it more likely to work cross-platform, so we can do this on amd64 and powerpc too?
[06:27] <BenC> plus, it works with hotplug cpu, meaning it will reenable SMP functions if you hotplug a second cpu :)
[06:27] <BenC> it's a bit more complex
[06:27] <infinity> You're kidding.  That's sexy.
[06:27] <desrt> that sounds pretty .... odd?
[06:27] <BenC> but since it does the function stuff, it should be easier for things like ppc
[06:28] <desrt> what if a critical section is executing while you swap in the 2nd CPU?
[06:28] <desrt> (but you have no way of telling that you're in a critical section since there is no locking for UP)
[06:28] <BenC> define critical?
[06:28] <desrt> holding locks
[06:28] <desrt> or would-be-holding-locks
[06:28] <desrt> (in the case that locks are NOP'd out)
[06:28] <BenC> not sure, but I haven't looked at how to enables SMP after the fact
[06:29] <BenC> how it enables I mean
[06:29] <infinity> Could be a small race there in theory.  In practice, I don't see that it would hurt much.
[06:29] <BenC> it should be easy to do this for amd64 atleast
[06:29] <desrt> maybe processor swapping requires all CPUs to come to a synchronisation point before proceeding
[06:30] <daniels> add it to a waitqueue?
[06:30] <infinity> Rather.
[06:30] <BenC> well, everything appears to be stable
[06:30] <infinity> I can't imaging that UP->SMP hotplugging would be a common use case anyway.  Still, cool if it works.
[06:30] <BenC> could be that cpu hotplug already does enough so that the smp enabler doesn't have to worry about it
[06:30] <infinity> imagine, either.
[06:31] <desrt> daniels; remember that bug where you got windows leaving traces on the background and the panel appeared at the bottom of the screen?  you fixed it with an upload of xorg shortly following flight 2
[06:31] <desrt> the top panel appeared at the bottom, i mean
[06:33] <daniels> desrt: er, no shit?
[06:33] <daniels> desrt: on r300 with ppc?
[06:33] <daniels> (r300 -> r3xx, r4xx)
[06:33] <desrt> daniels; ya.  it's not fixed.
[06:33] <daniels> oh.  no, it's not.
[06:33] <desrt> oh.  excellent.
[06:33] <fabbione> desrt: known problem
[06:33] <daniels> it works with exa
[06:33] <desrt> perfect
[06:33] <daniels> it *seems* to be a bug in the xaa core
[06:33] <fabbione> daniels: didn't you say that EXA is BAD BAD BAD
[06:34] <desrt> ah.
[06:34] <daniels> fabbione: sure
[06:34] <fabbione> to the definition of BAD like in ghostbuster?
[06:34] <daniels> fabbione: in that it may well kill your system, and good luck with anything being even vaguely performant
[06:34] <jdub> wow, my father in law just plopped this totally bizaare piece of hardware down on the table
[06:34] <infinity> jdub: Does it vibrate?
[06:34] <fabbione> aahha
[06:34] <daniels> when anholt's done fixing the memory manager, I'll reconsider exa as the default for r3xx/r4xx
[06:34] <jdub> it's a 486 cyrix machine with 4MB ram, deigned for pos use, with a pen interface
[06:34] <jdub> fabbione: ha ha ghostbusters
[06:35] <daniels> jdub: the mexican place I was at a couple of days ago appears to be using zauruses
[06:35] <daniels> your backpack has protons in it?? that's too cool!
[06:35] <jdub> i'm glad ubuntu runs on 486
[06:35] <daniels> mine only has neutrons :(
[06:35] <fabbione> ehe
[06:35] <jdub> dunno about this $MB ram issue though ;)
[06:35] <jdub> 4MB
[06:35] <jdub> *cough*
[06:36] <daniels> jdub: you'll want a 1GB swap partition
[06:36] <daniels> also afaict you can't be using more than 640x480x8, unless you want to have your framebuffer image being swapped out
[06:37] <jdub> wow
[06:37] <desrt> or stored in video ram....
[06:37] <jdub> windows for pen computing version one
[06:38] <daniels> desrt: if you can't accelerate all your ops, then you need to have a software image
[06:39] <daniels> desrt: i'm betting whatever's in there does nothing more advance than solid fills, point-to-point lines, and blits
[06:39] <desrt> daniels; but if you disable accel entirely....
[06:39] <desrt> daniels; or is that not supported these days?
[06:39] <daniels> desrt: yeah, I guess
[06:40] <daniels> desrt: Option "NoAccel" is supported everywhere
[06:40] <desrt> well
[06:40] <desrt> a 486 with 4MB of ram is supposed to be painful
[06:40] <daniels> yeah
[06:40] <desrt> so it may as well be very painful :)
[06:40] <daniels> the lines will be fine, but fills and blits will cane you
[06:40] <daniels> slow framebuffer reads/writes + slow bus == oh my god
[07:21] <jdub> http://www.computercloset.org/DauphinDTR1.htm
[07:21] <jdub> ^ ha ha, that's it
[07:21] <fabbione> jdub: neat toy
[07:22] <fabbione> you can probably install a reduced ubuntu on it
[07:22] <fabbione> for sure you want a stripped down to death kernel
[07:22] <jdub> yeah
[07:23] <jdub> will probably try netbooting a tiny kernel
[07:23] <fabbione> i don't think you can netboot that thingy :)
[07:23] <fabbione> probably using one of these floppy from netrom.something
[07:24] <fabbione> that have a pxe client on floppy
[07:24] <fabbione> with only the specific enet driver for that machine
[07:24] <jdub> "The unit will run for two continuous hours on a battery charge without the keyboard attached. If the keyboard is attached, it will run for an hour and a half."
[07:24] <jdub> ha ha ha
[07:25] <infinity> Nice. :)
[07:25] <daniels> orright, no x11r7 final 'till next year
[07:25] <daniels> but the only changes between rc4 and final were pure doc, so whatevz
[07:31] <jdub> hrm, the ethernet port may only be a header, too...
[09:11] <pitti> Hi
[09:19] <jsgotangco> pitti, good morning
[09:19] <pitti> hi jsgotangco 
[09:19] <jsgotangco> pitti, happy holidays =)
[09:23] <pitti> jsgotangco: thank you, same to you :)
[09:23] <pitti> enjoy christmas, family, and vacation
[09:26] <jsgotangco> pitti, yup but before that, are locales still broken? =)
[09:26] <jsgotangco> sorry i had to ask heh
[09:26] <jsgotangco> haha
[09:26] <jsgotangco> enjoy the weekend =)
[09:27] <pitti> jsgotangco: consider it a present for xmas - find out about the deeper quirks of dpkg ;)
[09:59] <Amaranth> was the workaround in cairo for an X bug removed?
[09:59] <Amaranth> because i think i'm seeing that bug again on ppc
[10:00] <Amaranth> it's fun watching icons dance up the desktop though
[10:04] <ogra> Amaranth, are you sure youre not accidentially using KDE ? :P
[10:04] <Amaranth> heh
[10:15] <dholbach> hellas
[10:16] <jsgotangco> hey
[10:16] <jsgotangco> dholbach, happy holidays
[10:16] <dholbach> thank you
[10:17] <jsgotangco> ho ho ho
[10:18] <dholbach> happy holidays and merry christmas to everbody else too!
[10:18] <robitaille> thanks
[10:19] <fabbione> i hate you all :)
[10:19] <fabbione> i want to be vac too!
[10:19] <dholbach> morning seb128
[10:20] <fabbione> hey dholbach 
[10:20] <dholbach> fabbione: you'll be working 27-31?
[10:20] <ogra> fabbione, i thought we are supposed to (from tomorrow on)
[10:20] <fabbione> dholbach: no
[10:20] <fabbione> dholbach: we are not allowed...
[10:20] <ogra> heh
[10:20] <dholbach> fabbione: *phew* yeah
[10:20] <fabbione> not like i won't be playing with my toys
[10:20] <fabbione> i need to build my console server to plug hppa/a64/an extra sparc
[10:20] <fabbione> and some other piece of equpment
[10:21] <fabbione> fix my tivo box
[10:21] <fabbione> all kind of stuff
[10:21] <fabbione> not that i expect my wife to allow me to do so
[10:21] <fabbione> but at least i will try :)
[10:21] <dholbach> fabbione: that's what i thought ;)
[10:21] <ogra> heh
[10:35] <jsgotangco> caio
[11:05] <jettana> hello
[11:06] <jettana> i had problem about remastering Ubuntu install cd
[11:20] <jordi> pitti: ping
[11:20] <pitti> fast pong
[11:20] <jordi> those are the ones I like
[11:21] <jordi> pitti: when a package in launchpad says it has no template to translate, usually it means the src package had no pot distributed?
[11:21] <jordi> pitti: https://launchpad.net/products/rosetta/+bug/2022
[11:22] <pitti> sounds right
[11:26] <pitti> jordi: we need to fix all these packages to produce a pot
[11:29] <pitti> slomo_: ping
[11:31] <pitti> seb128, slomo_: I updated the avahi main inclusion report
[11:31] <seb128> pitti: thanks for looking on it
[11:32] <fabbione> pitti: i just tried a breezy->dapper update for the locales stuff
[11:32] <fabbione> and it's no go
[11:32] <pitti> fabbione: what breaks?
[11:32] <fabbione> it looks like /var/$something/ is missing
[11:32] <fabbione> quite a long path..
[11:32] <pitti> fabbione: does the locale package break?
[11:32] <pitti> or the langauge packs?
[11:32] <fabbione> yes
[11:32] <fabbione> the locale
[11:32] <pitti> /var/lib/locales/supported.d I assume
[11:32] <fabbione> ye
[11:32] <pitti> ok, thanks
[11:33] <Lathiat> pitti: ooh nifty
[11:33] <fabbione> it says in preinst or something that it is missing
[11:33] <fabbione> and error hard
[11:33] <pitti> fabbione: could you please file a major bug against me? I can't fix it now (EBUSY)
[11:33] <Lathiat> pitti: dropping qt4 is easy
[11:34] <pitti> Lathiat, slomo_, seb128: ok, packaging and debs look fine, security looks fine, too, but the RC bugs need to be solved
[11:34] <Lathiat> pitti: qt4 in universe?
[11:34] <fabbione> pitti: meh sure ok
[11:34] <Lathiat> pitti: what RC bugs?
[11:34] <pitti> Lathiat: http://bugs.debian.org/cgi-bin/pkgreport.cgi?src=avahi
[11:34] <pitti> and the FTBFS on i386
[11:34] <Lathiat> ubuntu build logs?
[11:34] <pitti> the important and normal Debian bug look pretty serious, too
[11:35] <Lathiat> looking now
[11:35] <Lathiat> didnt realise they were there
[11:35] <seb128> pitti: http://people.ubuntu.com/~lamont/buildLogs/a/avahi/0.6.1-1ubuntu1 ...
[11:35] <pitti> well, these bugs probably come and go on a daily basis
[11:35] <seb128> pitti: all successful
[11:35] <pitti> seb128: no idea, there are no 0.6.1-1ubuntu1 i386 debs
[11:35] <Lathiat> most of this stuff is a quick/minor
[11:36] <fabbione> pitti: 21436
[11:36] <pitti> seb128: hmm, odd
[11:36] <pitti> fabbione: thanks
[11:36] <seb128> pitti: I'm not sure the archive is updated since yesterday
[11:37] <seb128> pitti: gst-plugins-good0.10 built yesterday and no sign of the deb neither
[11:37] <seb128> seems that elmo broke the archive before running in VAC :p
[11:37] <seb128> I got no update today neither
[11:37] <seb128> (out of an openoffice stuff from yesterday)
[11:39] <jordi> pitti: so what do we do in this case?
[11:39] <pitti> jordi: well, fix the package to build a POT :)
[11:39] <pitti> guys, I have to leave now, I need to visit my grandparents
[11:39] <jordi> ok. even for breezy?
[11:39] <pitti> no, not for breezy
[11:40] <pitti> we can manually build a breezy POT and feed it into Rosetta
[11:40] <pitti> as we did for hoary
[11:40] <pitti> I might be online again in the afternoon
[11:40] <seb128> pitti: enjoy your holidays, happy christmas/new year :)
[11:40] <pitti> seb128: for you, too!
[11:40] <pitti> jordi: happy christmas to you
[11:45] <Nafallo> what happened at ubz? everyone has been hugging everyone else since then :-P.
[11:45] <dholbach> Nafallo: there was some love potion in the food
[11:45] <mvo> SUPERHUG powers
[11:45] <JaneW> dholbach: chinese aphrodisiacs
[11:46] <Nafallo> hehe :-)
[11:47] <dholbach> did somebody apart from the ubuntu world comment on that already? ;-p
[11:47] <Nafallo> hehe, no idea ;-)
[11:49] <Nafallo> hmm
[11:49] <Nafallo> mp3 support out of the box for dapper?
[11:49] <Nafallo> http://blogs.gnome.org/view/uraeus/2005/12/23/0
[11:49] <dholbach> Nafallo: we already talked about that - there's some "if your a distributor..." clause somewhere on those pages
[11:51] <Nafallo> right...
[12:03] <jdub> Nafallo: we can't ship it, but fluendo will be able to
[12:04] <Nafallo> indeed. I actually read more than the blog now :-).
[01:36] <slomo_> jbailey: ping?
[01:37] <jbailey> slomo_: pong?
[01:38] <slomo_> jbailey: do you know of any great problems with breezy and flight2 live cds on ppc64? the mono guy refused to debug it any further because he had lockups, kernel panics, etc :/
[01:39] <jbailey> slomo_: Yup, all of the above.
[01:39] <jbailey> Fixed in current Dapper.
[01:39] <slomo_> perfect... i'll tell him :) thanks
[01:39] <jbailey> Well, loose version of fixed.  Don't use opengl at all
[01:39] <jbailey> But if you avoid that, the rest of it is now mostly fine.
[01:40] <slomo_> do we have daily live cds and are they working currently? ;)
[01:41] <apokryphos> slomo_: http://cdimage.ubuntu.com/daily/
[01:41] <jbailey> slomo_: I don't test the live CDs much.  I do try to run current dapper, though.
[01:42] <slomo_> ok, thanks
[02:05] <lucas> hi
[02:33] <mhz> hi all
[02:35] <Pygi> Hi hi
[03:42] <mdke> Riddell, ping? is there a reason the kubuntu docs are building into trunk/kubuntu/build rather than trunk/build/kubuntu?
[03:42] <pvanhoof> libc6-dev: Depends: libc6 (= 2.3.5-1ubuntu12) but 2.3.5-1ubuntu17 is to be installed
[03:42] <pvanhoof> on dapper
[03:42] <pvanhoof> is that situation going to get fixed soon? :)
[03:42] <pvanhoof> I just reinstalled my laptop, but that one is holding back any -dev packge install
[03:43] <pvanhoof> making it nearly impossible to get a working development environment ;), that sucks
[04:55] <jbailey> pvanhoof: EErr.  What arch is that?
[04:55] <jbailey> pvanhoof: I've seen no other bug reports saying that, so it seems like you've got something else going on.
[05:00] <pvanhoof> i386
[05:01] <pvanhoof> jbailey, standard setup breezy (from cd) and immedialtely after installed upgraded to dapper
[05:01] <pvanhoof> did it that way because the todays dapper daily cd didn't work
[05:01] <pvanhoof> but gtg :-(. catch me later on this, will yha?
[05:05] <jbailey> pvanhoof: Sure.  Ping me when you're around.
[05:33] <stockholm> mdz is not here. will he be online later?
[05:33] <dholbach> stockholm: no, he's on vacation
[05:34] <stockholm> ah, right, christmas!
[05:34] <stockholm> i forgot. 
[05:34] <stockholm> when is he back?
[05:34] <dholbach> i dunno exactly, but i suppose beginning of next year
[05:36] <stockholm> ok, fine
[05:36] <stockholm> thanks
[06:36] <slomo_> jbailey: fyi current ppc64 daily live cd doesn't work :/
[06:38] <jbailey> slomo_: =(  Can you file a bug?
[06:38] <jbailey> It's unlikely to get looked at until the new year.
[06:40] <slomo_> jbailey: ok... hm, should flight2 without X work?
[06:40] <jbailey> I have no idea, sorry.  I haven't tested flight2.
[06:41] <slomo_> ok... hmm, i'll ask this guy to file bugs... i don't know what exactly doesn't work
[06:41] <slomo_> i need to get more different machines :)
[07:01] <sivang> wow, nice I didn't know debconf questions were being GTK'd in synaptic
[07:01] <Pygi> hehe :)
[07:02] <mvo> sivang: :)
[07:06] <mvo> sivang: I (well, biased) find using synaptic nice for big upgrades because I get a rough idea how long it will take because of the progressbar
[07:06] <sivang> mvo: agreed, but I now seem to be running into an endless loop with locales installation failure -
[07:06] <sivang> mvo: do I need to do anything then change my sources.list file for dist-ugprading in synaptic?
[07:07] <mvo> sivang: a endless loop? hm, I had a problems with the locales too, it solved itself after a second "apply"
[07:07] <mvo> sivang: no, changing your sources.list is enough
[07:07] <sivang> mvo: ok, I will retry
[07:12] <sivang> mvo: weird. I can't get out of thie loop. It repeats the question about which dictionaris to install, displays the locale LC_ALL=C fallback errors on the terminal fails again and starts from the beginning
[07:14] <mvo> sivang: I remember getting a lot of questions about the dictinoaries as well. I had to confirm it 10-20 times (upgrade glitch). you may try to upgrade to breezy first before going to dapper. that works usually better 
[07:14] <mvo> going only a single version at a time
[07:20] <jbailey> mvo, sivang: IIRC, skipping versions isn't supported for upgrades.
[07:21] <zul> heh...i did that as well
[07:21] <mvo> jbailey: yes, it's not supported. but sivang is a power-user. I did such upgrades as well, it requires a bit more hand-holding
[07:22] <jbailey> Right.  I didn't know if anyone had ever bothered even trying it.
[08:12] <khermans_> I would like to make a Ubuntu document, but I want it to stay updated by using SOAP to connect to my webserver to get the latest info
[08:12] <khermans_> how can I add a SOAP connection to my PDF document?
[08:12] <neuralis> khermans_: you can't.
[08:13] <khermans_> ?
[08:13] <khermans_> supposedly you can in Acrobat 6+
[08:13] <khermans_> http://www.planetpdf.com/developer/article.asp?ContentID=6324
[08:13] <tseng> that sounds like a question for the acrobat manual
[08:14] <neuralis> khermans_: i'm pretty sure none of the unix pdf viewers support such a thing.
[08:14] <khermans_> ok, fair enough -- thought someone might know
[08:14] <neuralis> khermans_: it also just sounds like a bad idea. it's much easier to put a nice box at the top of your document that provides a url to the newest version of your document.
[08:14] <tseng> i might suggest though that you are better off making an xml docbook source file and dynamically generating the pdf programmatically
[08:14] <tseng> or something.
[08:15] <neuralis> tseng: that isn't really what he's looking for. he wants a document to auto-update _after_ it reaches the end-user, which is non-trivial and rarely done.
[08:15] <khermans_> neuralis, exactly
[08:17] <khermans_> so like, even if I distributed a PDF to my friend, he would never have to naviigate to the site again -- just reopen the PDF and there are the new steps for doing some Dapper-specific stuff, or even Dapper+1
[08:17] <khermans_> oh well :-(
[08:17] <neuralis> khermans_: the easiest way to do that is to provide the documentation as HTML, and use XHR after the page loads to attempt to fetch a more recent version, then replace the contents via standard DOM manipulation.
[08:18] <neuralis> actually, no, that won't work. cross-site xhr. nevermind.
[08:18] <neuralis> khermans_: i think you're mostly out of luck.
[08:55] <dholbach> good night everybody - have a happy christmas and a good start into the next year
[08:58] <Treenaks> dholbach: you're leaving us for a week?
[08:58] <Treenaks> dholbach: (have a good christmas/newyear too :))
[08:59] <dholbach> Treenaks: thanks, i guess i pop in every now and then, but just wanted to say, when i see you all ;)
[08:59] <Treenaks> dholbach: blog it ;)
[08:59] <dholbach> :)
[09:30] <mjg59> With cdbs, what's the right approach to adding a patch that requires the package to be autoreconfed?
[09:31] <jbailey> Is it autoconf and automake together?
[09:32] <mjg59> jbailey: autoconf, automake and autoheader
[09:32] <mjg59> It's easy to just keep those in the patch, except that seems to defeat the entire point of having the patch
[09:32] <mjg59> Uhm. Defeat the entire point of having the patch be a split-out patch
[09:33] <jbailey> mjg59: Generally I find the best solution is to add AM_MAINTAINER_MODE to configure.{in,ac} and the regenerate the files once.
[09:33] <mjg59> jbailey: See above :)
[09:33] <jbailey> That way security updates don't depend on the autotools all working together with whatever version they're at in the future.
[09:33] <mjg59> Right, but then if there's more than one patch that does that, there's no point in having split out patches at all
[09:33] <jbailey> Robert Millan added some magic to make it automatically autoreconf and such.  I've never tried it.
[09:34] <mjg59> "CDBS can be asked to update libtool, autoconf, and automake files, but this behavior is likely to break the build system and is '''STRONGLY''' discouraged."
[09:35] <mjg59> (No mention of autoheader, though the code seems to do it)
[09:35] <jbailey> It looks like DEB_AUTO_UPDATE_AUTOCONF might just need to be added to the rules file.
[09:35] <mjg59> Yeah. This still sounds like a poor solution, though I can't see a better one
[09:35] <jbailey> Right.
[09:36] <jbailey> What it really needs is some clever way that says "At the last possible moment, run autoreconf -f -i -s"
[09:36] <jbailey> Yes.
[09:36] <mjg59> They're not automatically pulled in?
[09:36] <mjg59> The extent to which I hate cdbs grows by the day
[09:37] <jbailey> They're automatically pulled in if you're using cdbs to generate your build-deps.
[09:37] <mjg59> Doesn't look like it. Bah.
[09:37] <jbailey> mjg59: Tell me what you want to be different.
[09:37] <jbailey> I'm actually hacking a bit on cdbs again.  If I can make it useful to you, I'd rather do that.
[09:38] <jbailey> Especially given that enough other people like it that you can't really avoid it.
[09:38] <mjg59> jbailey: Heh :) I want it to deal well with patches that touch the autotools config
[09:38] <jbailey> That's reasonable.
[09:38] <mjg59> Since it seems to defeat the point of split out patches otherwise
[09:39] <jbailey> Can you think of a non-sucky way to handle the build-deps problem?
[09:39] <jbailey> Right now it requires a control.in file.
[09:39] <jbailey> Since build-deps have to be stable before upload.
[09:39] <mjg59> Have a pre-upload pass?
[09:40] <mjg59> Can it be done somehow during dpkg-buildpackage? I've never actually looked enough at what it does then
[09:41] <jbailey> Right now the maintainer has to generate it manually with debian/rules debian/control
[09:41] <mjg59> Right
[09:41] <jbailey> It's easy enough, but it breaks the assumption that debian/control is the canonical file to edit.
[09:41] <mjg59> Yeah
[09:42] <mjg59> A ${cdbs:deps} type statement that would just be for things that cdbs pulls in?
[09:42] <jbailey> Yup.  I think we use automake style @CDBS@ right now, but that's the idea.
[09:42] <jbailey> It's what you include in the control.in.
[09:42] <mjg59> Right
[09:43] <jbailey> One hack that I *will* be putting in shortly is a test that if debian/control is not up to date, that the build will fail right at the beginning with a note asking the maintainer to run debian/rules debian/control by hand.
[09:44] <mjg59> Hrm. What's the right way to build-depend on automake?
[09:44] <jbailey> automake1.9
[09:44] <mjg59> Heh :)
[09:44] <mjg59> Fair enough
[09:44] <jbailey> (assuming your automake-fu is sufficiently recent)
[09:44] <mjg59> Yeah