[12:29] <dieman> holy crap, samsung released amd64 binaries for their printer
[12:31] <Chipzz> nice :)
[12:32] <jdub> dieman: their printer is amd64? ;-)
[12:48] <wasabi_> That was too easy.
[12:54] <JanC> dieman: most samsung printers work OOTB on Ubuntu?
[12:55] <JanC> I guess you mean their proprietary GUI installer ?  ;-)
[01:47] <floam> BenC: is there any obvious reason an initrd might fail to be created with no message upon install of the 2.6.17 kernel-image packages?
[01:48] <floam> update-initramfs -c -k 2.6.17-4-k7 seems to make one (havn't checked to see if it works yet)
[02:05] <infinity> floam: Because, afaict, the postinst for the -4- kernel just plain doesn't create an initramfs at all.
[02:05] <jdub> infinity: massive!
[02:08] <jsgotangco> hey jdub
[02:08] <jdub> yo jsgotangco 
[02:11] <tseng> hi jsgotangco 
[02:12] <jsgotangco> hey tseng how's it going?
[02:12] <tseng> fine, you?
[02:12] <jsgotangco> pretty good starting the week though *groan*
[02:15] <tseng> does anyone know the current method of getting a dapper-updates approved?
[02:16] <tseng> mail mdz, file bug assigned to foo, ...?
[02:16] <Riddell> tseng: send diff to mdz and kamion.  upload when approved
[02:16] <tseng> Riddell: cheers.
[02:21] <BenC> floam, infinity: Yeah, the kernel-package used to create the -4 kernel was screwed
[02:59] <labuser> I have a fix for a buggy package in ubuntu, where do I send it to? (It's a one-line diff)
[02:59] <Chipzz> labuser: I guess file a bug in launchpad ?
[03:00] <Chipzz> you can mark your attachment as being a patch I think
[03:00] <labuser> it's been on launchpad for 2 months but the package is still broken
[03:01] <labuser> someone else filed it here: https://launchpad.net/distros/ubuntu/+source/mldonkey/+bug/42632
[03:01] <Ubugtu> Malone bug 42632 in mldonkey "startup script is broken" [Medium,Confirmed]  
[03:02] <labuser> it's not a minor thing either... attempting installation of the package breaks apt alltogether
[03:03] <labuser> http://rafb.net/paste/results/sonrqK93.html
[03:03] <labuser> I pasted the fix there... maybe someone can commit it
[03:03] <Fujitsu> Which package? mldonkey-gui
[03:04] <labuser> mldonkey-server
[03:04] <Fujitsu> Ah.
[03:04] <Fujitsu> OK.
[03:04] <Fujitsu> Works fine for me...
[03:04] <Fujitsu> How does it break it?
[03:06] <Fujitsu> It doesn't actually install the mldonkey binary... But it doesn't break.
[03:06] <labuser> Installation of any other package fails. I forgot the exact message
[03:07] <labuser> I did an apt-get source and applied the fix, then did dpkg -i with that, and everything was ok
[03:07] <JanC> sounds like a failed install script?
[03:07] <labuser> yes that's right
[03:07] <Fujitsu> Works fine for me, still.
[03:07] <labuser> Fujitsu, the installation ?
[03:08] <Fujitsu> labuser, yes... I installed both mldonkey-server and mldonkey-gui, and it still works.
[03:08] <labuser> dapper right
[03:09] <Fujitsu> Hmm.
[03:09] <Fujitsu> Yeah.
[03:09] <Fujitsu> BOOM
[03:09] <labuser> I don't know then
[03:09] <Fujitsu> I remove mldonkey-gui, try to remove mldonkey-server and it explodes.
[03:09] <labuser> there you go :)
[03:10] <labuser> I guess it really did go BOOM ... :)
[04:27] <floam> BenC: oh, okay
[04:44] <Hobbsee> morning all
[05:00] <dieman> JanC: the ml-2010 driver isn't actually in foomatic yet
[05:00] <dieman> ml-1710 supposedly works, but margins are off and you can't use the 1200x600 mode
[05:17] <Hobbsee> hi Burgundavia 
[05:17] <Burgundavia> hey Hobbsee
[05:48] <bluefoxicy> wow that's nice, I just typed <tseng> tab and xchat-gnome printed:   tseng tseng tseng
[05:49] <bluefoxicy> anyway.  tseng:  If you're still up do you know what glibc detects of corrupted heap elements?
[05:51] <bluefoxicy> I'm thinking it's entirely possible to detect double free(), free() of a corrupt heap element, and free() when the next and previous elements in the chain of heap allocations are corrupt; but I'm not sure if glibc detects all of these AND differentiates between double free() and other weird corruption (which is doable simply by marking an entry in a special way when it's free()'d)
[05:51] <bluefoxicy> i know it'll catch some level of dirtiness.
[06:43] <fabbione> morning
[06:47] <Hobbsee> hi fabbione!
[06:48] <fabbione> hey Hobbsee 
[06:48] <Hobbsee> fabbione: edgy isnt actually broken that badly :)
[06:49] <fabbione> it's all because we are too cool :)
[06:49] <Hobbsee> fabbione: of course :)
[06:51] <fabbione> hmmm
[06:51] <fabbione> i need more bw
[06:57] <Hobbsee> fabbione: bw?
[06:58] <fabbione> bandwidth
[07:00] <fabbione> SCORE
[07:01] <jsgotangco> how cheap/expensive is bw there in dk?
[07:01] <bluefoxicy> https://wiki.ubuntu.com/AutomatedSecurityVulnerabilityDetection  This would be awesome, if https://wiki.ubuntu.com/AutomatedProblemReports ever happens.
[07:01] <Hobbsee> fabbione: ahhhh...
[07:01] <bluefoxicy> Actually the double free()/heap corruption/stack smash detection and reporting can be done without the kernel-level facilities
[07:01] <fabbione> well who did activate the libata shit didn't prepare the transition properly
[07:02] <fabbione> jsgotangco: not that expensive..
[07:02] <bluefoxicy> also it may be possible to load default signal handlers and modify signal() to unload them when new handlers are installed and load ours when signal handlers are cleared.
[07:04] <bluefoxicy> ours would have to be able to tell signal() to load no handler *reliably* (ours would be in glibc though so no problem) and then send the signal to self to die though.  Hrm.  Or something creepy like that.
[07:36] <Hobbsee> do we  have any current ETA for l-r-m for 2.6.15-5-686?
[07:40] <Hobbsee> ah, i'ts oddly named, and it FTBFS.  fun.
[07:43] <floam> Hobbsee: it already exists
[07:43] <floam> it's probably sitting in line to be built
[07:44] <Hobbsee> floam: i'm meaning the one that depends on linux-image-2.6.17-5-686
[07:44] <Hobbsee> which is l-r-m-2.6.17-6-686, which FTBFS, it seems.
[07:44] <bluefoxicy> FTBFS?
[07:44] <Hobbsee> bluefoxicy: failed to build from source
[07:44] <bluefoxicy> oh
[07:45] <basanta> does ubuntu use different technique than debian to get flash drives automounted? 
[07:45] <jsgotangco> :P
[07:45] <floam> basanta: does debian use HAL/pmount?
[07:45] <bluefoxicy> .... for signoff?
[07:46] <basanta> floam, yes
[07:47] <floam> then they probably do the same thing
[07:47] <Hobbsee> bluefoxicy: no, those have other acronyms
[07:47] <basanta> yes, but ubuntu can automount flash drives that debian can't 
[07:48] <bluefoxicy> Hobbsee:  yes but I assumed you wouldn't say 'fubar' if you intended somebody repair it :P
[07:48] <Hobbsee> bluefoxicy: well...
[07:48] <Hobbsee> bluefoxicy: stuff that's counted as fubar'd usually gets fixed, eventually, and after much swearing and cursing
[07:49] <Hobbsee> whee!  where's that darned autoconf patch.
[07:49] <bluefoxicy> Hobbsee:  You guys need your own reality TV, seriously.  Swearing and cursing is amusing.
[07:49] <floam> basanta: there's a good chance ubuntu has patched things up a bit or has something newer, but I think they use the same general stuff
[07:50] <bluefoxicy> Did you see when Keybuk spent like 15 minutes spouting 'fuck' every other word because he tried to build openoffice.org and it installed over a gigabyte of build deps?
[07:50] <basanta> floam, ok
[07:50] <bluefoxicy> He was after poor doko's blood :P
[07:51] <Hobbsee> i heard about it, yeah
[07:52] <basanta> floam, do you think ubuntu-developers would share this patch? and do you know where to ask?
[07:52] <fabbione> Hobbsee: l-r-m will be sorted once 5.13 is built
[07:53] <Hobbsee> fabbione: 5.13 of linux-image?  or of l-r-m?
[07:53] <floam> basanta: well you can look through the packages yourself or ask in ubuntu-devel
[07:53] <floam> I'm sure they would share it
[07:53] <fabbione> Hobbsee: the former
[07:53] <Hobbsee> floam: everything's GPL'd...it's  likely to be around here somewhere.
[07:54] <floam> I assume by share he meant they'd find him for it and hand it over
[07:54] <Hobbsee> fabbione: right, know off the top of your head if that's about to be built?
[07:54] <floam> of course they have to allow him to use it
[07:54] <lifeless> floam: we share all out patches
[07:55] <fabbione> Hobbsee: https://launchpad.net/+builds
[07:55] <lifeless> erm, I meant basanta: we share all the patches
[07:55] <johanbr> basanta: Try looking at http://patches.ubuntu.com .
[07:55] <basanta> lifeless, johanbr thank  you .
[07:55] <Hobbsee> fabbione: ahh :)  that's the page i was looking for :)
 basanta: well you can look through the packages yourself or ask in ubuntu-devel  <-- uh floam?
[07:56] <floam> uh bluefoxicy?
[07:56] <bluefoxicy> this IS ubuntu-devel
[07:56] <floam> bluefoxicy: this is #ubuntu-devel
[07:56] <floam> there is also ubuntu-devel the mailing list
[07:56] <bluefoxicy> you said in, not on ubuntu-devel@
[07:56] <floam> whatever
[07:56] <floam> clearly we are here so there is only one other, I thought it made sense
[07:57] <floam> I will be sure not to mistake my prepositions in the future though
[08:04] <Hobbsee> floam: to be fair, #ubuntu-devel is usually the irc channel, and ubuntu-devel is the mailing list.
[08:04] <Hobbsee> bluefoxicy: ^
[08:51] <pitti> Good morning
[08:51] <Hobbsee> hi pitti!
[09:24] <Hobbsee> Mithrandir: i think britney is confused - that's not the version of amarok in the repositories.  oh.  lovely.  hardcoded dependancy on amarok in amarok-arts, therefore the bugger wont install anyway.
[09:24] <bluefoxicy> pitti:  not so hot on that idea?
[09:25] <Mithrandir> Hobbsee: oh well, it's just a first milestone, it's not that important if it's not perfect. :-)
[09:26] <Hobbsee> Mithrandir: hehe that is true.  i'm just merely annoyed that it's taken me this long to notice, and that a few packages which i was told/thought were uploaded werent.  
[09:26] <Mithrandir> Hobbsee: become a core dev and you can fix it yourself. :-)
[09:27] <pitti> bluefoxicy: when I played with my preloaded-library approach (which is essentially the same thing) I had some problems with it as well, like gdb not being able to cleanly attach and so on; but that might be unrelated to the crashed process' state
[09:27] <Hobbsee> Mithrandir: heh.  yeah.  reckon you can give me dev and core dev in the same meeting?  :P
[09:27] <Hobbsee> s/you/the TB
[09:27] <pitti> bluefoxicy: anyway, we can consider modifying glibc later, since it is totally independent from the actual spawned agent which collects data
[09:28] <Mithrandir> Hobbsee: I don't think that's being done, no..  But you're active and seems clueful enough that making core-dev shouldn't take long anyway.
[09:28] <Hobbsee> Mithrandir: hehe.  actually, it was funny.  i was told when i originally started packaging, by two separate people, that they could see me going for core dev soon.
[09:29] <bluefoxicy> pitti:  nod.  At any rate the program will sleep if you SIGSTOP it.
[09:32] <bluefoxicy> pitti:  i am still looking at this thing for automatically prioritizing things that look like new security vulns, by the way.
[09:33] <bluefoxicy> pitti:  do you think you'll have something ready by Edgy?
[09:37] <pitti> bluefoxicy: (btw, you can't ptrace() a stopped process)
[09:37] <bluefoxicy> pitti:  (shit)
[09:39] <pitti> bluefoxicy: yes, that's the goal
[09:40] <pitti> bluefoxicy: (having something for edgy)
[09:48] <bluefoxicy> pitti:  gdb seems to be able to attach to less after I ctrl-z less
[09:49] <pitti> bluefoxicy: not here, it just hangs indefinitely at 'Attaching to program:'
[09:50] <pitti> bluefoxicy: and after fg'ing the target process, gdb suffers a horrible death
[09:50] <bluefoxicy> ah right, it is hanging.
[09:50] <bluefoxicy> I see :o
[09:53] <pitti> bluefoxicy: it makes sense, since waitpid() on a  stopped process will never return
[09:53] <bluefoxicy> it looks like you can attach with ptrace(), just gdb doesn't understand why you would try.
[09:53] <bluefoxicy>  /build/buildd/gdb-6.4.90.dfsg/gdb/linux-nat.c:1025: internal-error: linux_nat_attach: Assertion `pid == GET_PID (inferior_ptid) && WIFSTOPPED (status) && WSTOPSIG (status) == SIGSTOP' failed.
[09:53] <bluefoxicy> ^^^ this is gdb bitching that the target process didn't return the status it was expecting
[09:53] <bluefoxicy> or that it just couldn't attach I guess.
[09:53] <bluefoxicy> it's one of the two, maybe you're right.
[09:53] <bluefoxicy> pitti:  want me to ping the gdb devs tomorrow on that?
[09:53] <pitti> bluefoxicy: if you feel like it, but I'm not sure whether it's merely a gdb restriction or a principal restriction of ptrace()
[09:54] <bluefoxicy> well, I'll find out.
[09:54] <pitti> thanks
[09:55] <bluefoxicy> night.
[09:55] <Hobbsee> night bluefoxicy 
[10:01] <bluefoxicy> pitti: https://wiki.ubuntu.com/AutomatedSecurityVulnerabilityDetection by the way.
[10:02] <bluefoxicy> (this would be the only reason I'm as interested as I am in this :)
[10:06] <pitti> bluefoxicy: thanks, will look at it
[10:43] <Mithrandir> ogra: you're aware that e-d is installable now?  Do you want me to spin you CDs?
[11:09] <ogra> Mithrandir, oh, yes please :)
[11:09] <Mithrandir> ogra: edubuntu livefs-es building.  Do you need new alternate CDs too?
[11:09] <fabbione> mdz: i am ready to upload the new redhat-cluster-suite. am I still go or do you want me to send email for UVF exception?
[11:10] <mdz> fabbione: what's in it?
[11:10] <ogra> Mithrandir, if you like :)
[11:11] <Mithrandir> ogra: alternate cds building.
[11:11] <fabbione> mdz: it's the natural upgrade from dapper (gfs1) to edgy (gfs1 and gfs2) userland. gfs2 is already in the kernel
[11:11] <ogra> Mithrandir, thanks ! 
[11:11] <fabbione> mdz: i only didn't manage to upload it before becuase of the big changes in the infrastructure
[11:11] <fabbione> mdz: (as i reported to the last meeting)
[11:12] <mdz> fabbione: backward-compatible?
[11:12] <fabbione> mdz: gfs2 no, the userland with gfs1 is modulo one line change in the config
[11:13] <fabbione> mdz: there is a proper debconf message for that
[11:13] <fabbione> mdz: that points to the wiki where we will build proper upgrade paths and stuff
[11:13] <mdz> fabbione: I mean is the package backward-compatible, not the filesystem
[11:13] <mdz> i.e., does it still support gfs1?
[11:13] <fabbione> mdz: yes it does
[11:14] <fabbione> it does support both
[11:14] <Kamion> argh, how do I get rid of the "Slow Keys Alert" without being able to click on it?
[11:14] <fabbione> but i need to merge the new gfs1 in the kernel.
[11:14] <mdz> ok, go ahead
[11:14] <Kamion> my mouse is already grabbed by vmware
[11:14] <fabbione> the new one uses shared lock manager with gfs2 to avoid code duplication
[11:14] <fabbione> mdz: perfect thanks
[11:14] <Kamion> and ctrl-alt does not work while the slow keys alert is up, apparently
[11:15] <mdz> Kamion: kill $random_gnome_component?
[11:16] <Kamion> yeah, which one though ...
[11:16] <Kamion> ah, gnome-settings-daemon maybe
[11:17] <Kamion> no, the bastard just comes back and leaves my keyboard screwed
[11:19] <Riddell> Mithrandir: kubuntu-desktop is broken because of koffice which is waiting on ruby1.8 compiling on powerpc
[11:20] <Riddell> doko: any idea what the gcc bug is that's cauing ruby1.8 to fail on powerpc?
[11:22] <bluefoxicy> mdz:  I'd wanted to bring your attention to https://wiki.ubuntu.com/CommunityEdgyIdeas/Install-CD earlier due to the /home directory thing.
[11:22] <infinity> Riddell: Are you sure it's a GCC bud and not a ruby bug?
[11:22] <bluefoxicy> I think.  I know there was something I wanted to bring your attention to specifically.
[11:23] <bluefoxicy> anyway I'm sleeping, it's 5am
[11:23] <Riddell> infinity: look at the build log, seems to be a gcc segfault
[11:23] <infinity> I don't see a GCC segfault.
[11:23] <dholbach> hey infinity! how are you?
[11:23] <infinity> /build/buildd/ruby1.8-1.8.4/build-tree/ruby-1.8.4/lib/mkmf.rb:804: [BUG]  Segmentation fault
[11:23] <infinity> ^^^ That's ruby segfaulting.
[11:24] <Riddell> hmm, right
[11:24] <infinity> Of course, it could be GCC miscompiling it still, but I'd like to blame ruby until I know for sure.
[11:24] <ogra> does anyone have experience with fakechroot ? i dont seem to be able to set up an environment ...
[11:26] <Mithrandir> ogra: your dailies are served.
[11:26] <ogra> thanks !
[11:27] <mdz> bluefoxicy: I reviewed those pages a couple of weeks ago and commented on things where I had an opinion
[11:27] <bluefoxicy> mdz:  nod, I did it a couple hours ago.
[11:28] <bluefoxicy> mdz:  short version:  some people I know actually have 500-750 gig hard drives and fill them with questionable content (for example, downloads of anime series they don't own...); that's like 100 DVDs or so.  It'd be nice to be able to nuke everything but /home at install if you're honestly going to make /root a big 500 gig partition just because the disk is 500 gigs wide.
[11:29] <bluefoxicy> er. /, not /root
[11:29] <bluefoxicy> sleep time for me.
[11:30] <Kamion> use manual partitioning then. there's no way for autopartitioning to keep everybody happy here.
[11:30] <Kamion> there's a reason we provide manual partitioning in non-expert mode
[11:33] <mdz> right, I think that's a fairly uncommon use case
[11:35] <Kamion> not so much uncommon use case as that there are N different common use cases. at least with the present situation it's easy to see that we chose it because it was the simplest alternative.
[11:36] <Kamion> Mithrandir: live CD's buggered - "User not known to the underlying authentication module" all over the place
[11:37] <Mithrandir> Kamion: gnr
[11:38] <Kamion> huh, it worked a second time, without quiet splash - except now my gdm login screen is ENORMOUS
[11:38] <Kamion> like 2048x1536 or something
[11:38] <Mithrandir> maybe it's a bad CD, then.
[11:38] <Kamion> it's vmware
[11:39] <Kamion> and autologin fails
[11:39] <Mithrandir> uh, ok
[11:39] <Mithrandir> I'll burn a dvd and see what happens on real hardware.
[11:39] <Mithrandir> this is i386, then?
[11:39] <Kamion> yes
[11:40] <Kamion> Mithrandir: since I'm here, anything I can do to debug it? I've forgotten how to debug casper
[11:41] <Mithrandir> Kamion: boot with break=bottom and poke /root to see if the ubuntu user is there.
[11:41] <Kamion> 'k
[11:44] <doko> Riddell: already checked, it's not a gcc bug, rechecked rebuilding with gcc-3.4 and gcc-3.4
[11:44] <infinity> s/3.4$/4.0/ ?
[11:45] <Kamion> bash: line 2: pwck: command not found
[11:45] <Mithrandir> Kamion: uh.
[11:45] <Kamion> it's in /usr/sbin; I suspect $PATH
[11:45] <Mithrandir> I thought casper set path, but maybe not?
[11:45] <Kamion> oh, never mind, I'm running this manually
[11:46] <Mithrandir> we should just make sbin a symlink to bin and get rid of that sillyness. :-P
[11:46] <Kamion> what's the break= to let me edit casper-bottom scripts before they run?
[11:47] <Mithrandir> break=casper-bottom
[11:47] <Kamion> ta
[11:47] <Mithrandir> (surprise!) :-)
[11:50] <Mithrandir> it would also have been nice if udev hadn't suddenly removed udevplug without telling me.
[11:50] <Kamion> ah yes, try udevtrigger && udevsettle
[11:51] <Kamion> I thought Scott was going to leave a wrapper :(
[11:51] <Mithrandir> apparently not.
[11:54] <Kamion> Mithrandir: you need to do 'set passwd/make-user false' too.
[11:54] <Kamion> er, true, not false
[11:54] <Kamion> dunno why that didn't break dapper
[11:54] <Mithrandir> why doesn't it default to true?
[11:54] <Kamion> er. it does. I haven't had breakfast yet.
[11:55] <Mithrandir> may I suggest eating breakfast and coffee, then? ;-)
[11:55] <Kamion> yeah
[12:05] <Mithrandir> Kamion: ok, it seems like casper doesn't set the path correctly, somewhere.  I'll just export it at the top of the casper script and let it be with that, I think.
[12:06] <Kamion> oh, you saw the same thing then?
[12:07] <Kamion> ./scripts/casper:301:    PATH=/root/usr/bin:/root/usr/sbin:/root/bin:/root/sbin:$PATH run_scripts /scripts/casper-bottom
[12:07] <Mithrandir> yeah
[12:07] <Mithrandir> well, it works if I pass PATH=/usr/sbin:/usr/bin:/bin:/sbin on the kernel command line
[12:10] <Kamion> that creates the user for me too, yes
[12:10] <Kamion> I have no xorg.conf now
[12:10] <Kamion> I think I didn't on one of my previous boots too, so it's not due to PATH
[12:10] <Mithrandir> me neither, but I wonder if that's just X being b0rken.
[12:11] <Mithrandir> export PATH=/root/usr/bin:/root/usr/sbin:/root/bin:/root/sbin:/usr/bin:/usr/sbin:/bin:/bin
[12:11] <Mithrandir> should be sane?
[12:11] <Kamion> oh, I understand the problem now
[12:11] <Kamion> /root/usr/sbin is not useful once you've chrooted
[12:11] <maswan> Mithrandir: shouldn't you have a /sbin in there btw?
[12:12] <Mithrandir> maswan: thanks, fixed.  The last one should have been /sbin
[12:12] <Kamion> just do PATH=/root/usr/bin:/root/usr/sbin:/root/bin:/root/sbin:/usr/bin:/usr/sbin:/bin:/sbin:$PATH run_scripts /scripts/casper-bottom
[12:12] <Mithrandir> Kamion: well, no, but if you do break=bottom and echo $PATH it includes sbin
[12:12] <Mithrandir> iirc
[12:13] <Kamion> dpkg-reconfigure xserver-xorg -> still no xorg.conf
[12:13] <Mithrandir> iz X bug, then
[12:14] <Mithrandir> oh, thepthul.  break=top gives me PATH=/usr/local/bin:/usr/bin:/sbin:/bin
[12:14] <Kamion> score
[12:14] <Kamion> initramfs-tools bug?
[12:15] <Mithrandir> it seems it's the PATH coming from the kernel
[12:15] <infinity> yeah, initramfs-tools doesn't define a PATH.
[12:15] <infinity> Probably should.
[12:15] <infinity> Though the kernel shouldn't be broken in this regard either.
[12:16] <Mithrandir> Kamion: if you fix xorg, I'll do casper.
[12:17] <Kamion> you're on
[12:30] <Kamion> xorg fixed
[12:30] <Kamion> what's up with the lack of wallpaper in a default edgy install?
[12:32] <infinity> Kamion: I'm running the publisher by hand right now to speed up the ABI switch.  Poke me when that xorg is uploaded, so I can publish again.
[12:32] <infinity> Mithrandir: Ditto, for casper.
[12:34] <Kamion> infinity: it's already uploadd
[12:34] <Kamion> +e
[12:34] <Kamion> if you mean source
[12:35] <Mithrandir> infinity: cheers.
[12:36] <simira> Mithrandir: that was quick
[12:37] <infinity> Kamion: Kay, you win.
[12:37] <infinity> Mithrandir: And you?
[12:37] <Mithrandir> infinity: just a sec, I want to test this locally first.
[12:37] <infinity> 5 minutes of testing, or 30?
[12:37] <Mithrandir> closer to 30 than 5
[12:37] <Mithrandir> so just push Colin's stuff through first.
[12:38] <infinity> Okay, I'll publish while I wait, then.
[12:38] <simira> do I feel a Knot approaching?
[12:38] <ogra> smells like it
[12:38] <ogra> :)
[12:38] <fabbione> Mithrandir: mind if i upload openais? it's -server/cluster stuff
[12:39] <Mithrandir> fabbione: please do.
[12:39] <fabbione> Mithrandir: thanks
[12:46] <Kamion> simira: ubiquity doesn't work yet, so it'll take a while ...
[12:49] <pitti> mdz: what do you think about updating mutt from a cvs snapshot to the stable 1.5.12 version? mainly bug fixes, translation updates, and code cleanup, and two small new features; I can send you the full changelog if you want
[12:52] <mdz> pitti: edgy or dapper?
[12:52] <pitti> mdz: edgy
[12:56] <pitti> hi slomo 
[12:56] <slomo> hi pitti 
[01:18] <doko> Mithrandir, Kamion: please could you sync dmake (needed to build OOo, doesn't touch the CD's)
[01:18] <pitti> wb mdz
[01:18] <pitti> mdz: in case you sent an answer after '<pitti> mdz: edgy', I didn't get it
[01:19] <Mithrandir> doko: I can't do syncs.
[01:19] <doko> Mithrandir: yeah, but you're current the lock holder ;)
[01:19] <Mithrandir> doko: if it doesn't touch the CDs, I have no objection.
[01:21] <Kamion> ok, I'll start a sync run
[01:21] <Kamion> doko: please assign sync requests for single packages to the relevant package - it makes things slightly easier to find
[01:21] <infinity> Mithrandir: ETA on the casper upload?
[01:22] <Mithrandir> infinity: sorry, had to bootstrap a bit here, if stuff works, in ten minutes.
[01:22] <infinity> Okay, I'll wait.
[01:22] <infinity> (Well, I'll build xorg and massage it through while I'm waiting)
[01:22] <mdz> pitti: <mdz> pitti: sure, yes
[01:22] <doko> Kamion: 53158
[01:23] <Kamion> yes I've found it now, just for the future
[01:23] <Kamion> main reason is that if it's assigned to the right package then launchpad tells us right on the bug page what component it's in
[01:24] <ogra> i'm trying to bootstrap a fakechrrot here, seems it fails on initscripts installation because its not able to bind mount / to /.root anybody having a hint what i could look for to find ot why mount says "no such file or directory" ? (/.root exists) 
[01:24] <ogra> *fakechroot
[01:25] <ogra> meh 
[01:26] <ogra> Mithrandir, its developemtn related, i need the fakechroots on the DC machines :P
[01:27] <Kamion> fakechroot's probably just broken. I don't think it's ever had any love in Ubuntu.
[01:27] <ogra> nope, it hadnt
[01:27] <Kamion> and it does some invasive things which are likely to break
[01:27] <ogra> its also in universe ... and we dont have a debootstrap variant (i'm just creating one)
[01:27] <Lathiat> hey wow thats kinda cool
[01:28] <ogra> Kamion, elmo wants me to use it ...
[01:28] <Kamion> fix it, then :)
[01:28] <Kamion> er ... debootstrap variant?
[01:28] <ogra> funnily not even the sarge.fakechroot variant bottstraps cleanly... seems it had not more love in debian 
[01:29] <Kamion> please run that by me before uploading, some of the proposed debootstrap hacks are really nasty
[01:29] <Kamion> that's probably because aj thinks it's horrible crack too
[01:29] <ogra> Kamion, i didnt plan to add it to the package (unless you want fakechroot in main)
[01:29] <Kamion> hell no
[01:29] <ogra> :)
[01:30] <Kamion> if the debootstrap script is in the fakechroot package I guess that's ok
[01:30] <ogra> but a review would be appreciated :)
[01:31] <ogra> (i simply took the edgy script and changed device, ldd functions and proc mounting to the fakechroot functions debootstrap provides)
[01:31] <ogra> *device creation
[01:32] <Mithrandir> ok, casper seems to work now.
[01:34] <Mithrandir> infinity: ok, uploaded, so do your thing.
[01:35] <infinity>    72793 | S- | casper               | 1.61                 | 5 seconds
[01:35] <infinity> Heh.  Wasn't there when I looked 10 seconds ago.
[01:35] <Mithrandir> :-)
[01:35] <Mithrandir> thanks
[01:40] <Kamion> hmm. ubiquity works (ish), gparted --installer doesn't.
[01:41] <Kamion> I guess that wasn't really a surprise
[01:41] <mdz> fabbione: how is sparc looking for a milestone?
[01:41] <mdz> Kamion: the i386 desktop CD i built on Saturday booted but failed to start X
[01:41] <mdz> had to leave for the airport at that point
[01:42] <mdz> did someone look into powerpc?
[01:42] <Kamion> mdz: that's sorted nonw
[01:42] <Kamion> now
[01:42] <Kamion> xserver-xorg.postinst was broken
[01:42] <mdz> Kamion: was there anything besides gnome-applets hurting powerpc?
[01:42] <Kamion> apparently not
[01:42] <mdz> yay
[01:43] <Kamion> not dependency-wise anyway
[01:44] <Kamion> doko: bug 50311 is another reason why you should assign to the proper package; then launchpad will do typo validation for you
[01:44] <Ubugtu> Malone bug 50311 in wajig "wajig grabs information from debian.org instead of ubuntu.com" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/50311
[01:44] <Kamion> doko: er, bug 53011
[01:44] <Ubugtu> Malone bug 53011 in Ubuntu "sync request" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/53011
[01:44] <Kamion> speaking of typos :)
[01:44] <Kamion> build-bot doesn't exist
[01:44] <doko> Kamion: buildbot, yes, I will do so for the future
[01:44] <ogra> hmm, intresting ... a DD who recompiles his packages for hoary and breezy http://people.debian.org/~dexter/dists/fakechroot/
[01:45] <Kamion> thanks
[01:45] <thom> nrrrgh dexter
[01:48] <basanta> G0SUB, bg in disguise??
[01:50] <infinity> We should totally hire dexter and convert the entire archive to yada.
[01:51] <ogra> heh
[01:52] <thom> infinity: is that before or after we shift to using rpms?
[01:52] <ogra> i thought we'd stay with deb but use alien ?
[01:52] <ogra> ;)
[01:54] <ogra> Setting up initscripts (2.86.ds1-14.1ubuntu1) ...
[01:55] <ogra> mkdir: cannot create directory `/.root': File exists
[01:56] <ogra> ah, thats the second attempt ...
[01:56] <ogra> Setting up initscripts (2.86.ds1-14.1ubuntu1) ...
[01:56] <ogra> mount: No such file or directory
[01:56] <ogra> grmbl
[02:01] <fabbione> mdz: no specific sparc work has been done. So whatever is in the other arches is on sparc
[02:03] <Mithrandir> Kamion: what's the state of ubiquity, etc wrt knot-1?
[02:04] <G0SUB> Kamion: bug 53238
[02:04] <Ubugtu> Malone bug 53238 in ubiquity "Ubiquity doesn't complain about bootable XFS partitions" [Untriaged,Confirmed]  http://launchpad.net/bugs/53238
[02:04] <pitti> Hi G0SUB 
[02:04] <G0SUB> pitti: hello :) I came back today
[02:06] <Kamion> G0SUB: duplicate of about a million other bugs; see DapperReleaseNotes/UbiquityKnownIssues
[02:08] <Kamion> Mithrandir: need to sort out gparted
[02:08] <G0SUB> Kamion: ok, thanks for the pointer
[02:08] <Kamion> 12:40 < Kamion> hmm. ubiquity works (ish), gparted --installer doesn't.
[02:08] <Kamion> 12:41 < Kamion> I guess that wasn't really a surprise
[02:08] <Kamion> G0SUB: it'll be fixed in my first edgy upload, too
[02:08] <Kamion> well, worked around anyway
[02:13] <pitti> G0SUB: can we meet today?
[02:14] <G0SUB> pitti: tomorrow at 14:00 UTC?
[02:14] <pitti> G0SUB: ok
[02:14] <G0SUB> fine
[02:24] <Kamion> mdz: have you done the validation for the ubuntu-archive e-mail change?
[02:35] <Kamion> Hmm. Now gparted at least plugs in correctly but you can't do anything in it.
[02:37] <Fujitsu> Kamion, can I ask you a quick question about the carpaltunnel sync?
[02:39] <zul> heylo
[02:40] <snorre> Take a look at this: http://pastebin.ca/90171
[02:40] <Hobbsee> Fujitsu: what's the question?
[02:40] <Kamion> Fujitsu: sure
[02:41] <Fujitsu> Would it be acceptable to just rebrand carpaltunnel 0.0.9-0.1 as 0.0.9ubuntu2, without putting in the rest of the changelog entries, etc?
[02:41] <Fujitsu> Because merging is silly if it increases the delta more than necessary because a sync isn't possible.
[02:42] <Kamion> Fujitsu: yes
[02:42] <Fujitsu> OK, thanks.
[02:42] <Kamion> just note it in the changelog so that we know to sync it if 0.0.9.1 or 0.0.10 or whatever shows up
[02:43] <Fujitsu> OK, shall do.
[02:43] <snorre> Can someone that's into kernel compiling please look at this issue: http://pastebin.ca/90171
[02:46] <Hobbsee> Mithrandir: did you ever find out what the problem with ppc kubuntu-desktop was?
[02:46] <Hobbsee> hi Keybuk 
[02:46] <Mithrandir> Hobbsee: Riddell told me it was koffice.
[02:47] <Hobbsee> Mithrandir: right.  how bizarre
[02:47] <Hobbsee> it's not.
[02:47] <Keybuk> Hobbsee: heyhey
[02:48] <Hobbsee> Keybuk: i've been breaking things :D
[02:49] <Hobbsee> well, not really...
[02:49] <Keybuk> so have I :)
[02:49] <Hobbsee> Keybuk: hehe..  what'd you break?
[02:50] <Kamion> TTOTD: dpkg-genchanges > foo.deb is not very useful
[02:50] <Kamion> (meant foo.changes obviously)
[02:50] <Hobbsee> Kamion: TTOTD?
[02:50] <Kamion> top tip of the day
[02:51] <Riddell> Hobbsee: ruby on powerpc isn't compiling, which is blocking koffice
[02:51] <Hobbsee> Riddell: ah.  but why's koffice in k-d anyway?
[02:52] <Riddell> Hobbsee, Mithrandir: I uploaded kubuntu-meta without koffice so once that's in the archive we should be able to start making CDs
[02:52] <Riddell> Hobbsee: for krita
[02:52] <Hobbsee> Riddell: ahh.  right
[02:53] <Hobbsee> Riddell: do we know what happened to wlassistant?
[02:55] <Riddell> Hobbsee: looks like it got uploaded to dapper, let me reupload to edgy
[02:55] <Hobbsee> Riddell: oops?  right.
[02:56] <Hobbsee> Riddell: why?
[02:56] <Kamion> ah, gparted is SIGSEGVing; that would explain why it disappears ...
[02:56] <Riddell> Hobbsee: the changelog had dapper in it not edgy
[02:56] <Hobbsee> Riddell: oops.
[02:57] <Kamion> it's not in dapper's unapproved queue
[02:57] <Kamion> apparently got randomly auto-rejected; I wish soyuz would stop doing that
[02:58] <Hobbsee> hehe
[02:58] <Kamion> oh, there was no .orig.tar.gz in the .changes (which would have been fine for edgy but not for dapper)
[03:18] <dholbach> i'd be so happy if there was a "xgl or whatever" strike force - we get a bunch of bugs just because our versions of xgl/compiz/... are broken and some random packages on the net seem to work
[03:18] <dholbach> (apart from that I'd be happy, if I could subscribe an xgl team to a bug to get some input) :)
[03:19] <seb128> Mithrandir: is that ok to upload epiphany-extensions (just a rebuild), it's not on the CD
[03:19] <Mithrandir> seb128: please do
[03:19] <seb128> thank you
[03:39] <lucas> is there a know problem with /usr/include/scsi/sg.h being in both libc6-dev and linux-kernel-headers ?
[03:40] <dholbach> Gloubiboulga, Riddell: we thought about having a "fix your favourite desktop" hug day on wednesday
[03:41] <simira> :)
[03:41] <Riddell> dholbach: sure
[03:41] <dholbach> Gloubiboulga, Riddell: if you want to add / fix things on wiki.ubuntu.com/UbuntuBugDay wiki.ubuntu.com/UbuntuBugDay/Draft - that'd be nice
[03:41] <Gloubiboulga> dholbach, would be nice
[03:42] <Riddell> dholbach: poke me in half an hour after the kubuntu meeting
[03:42] <dholbach> Gloubiboulga, Riddell: just look at the pages and add whatever you think might be useful
[03:45] <Hobbsee> dholbach: that'd be cool :)
[03:52] <sladen> Keybuk / infinity: could you represent wxwidgets2.6 to the buildds
[03:52] <Keybuk> sladen: I'm on holiday today
[03:53] <sladen> Keybuk: holiday!   I didn't know Canonical /did/ holiday :)
[03:53] <Hobbsee> surely they dont.
[03:53] <Hobbsee> there's no such thing as a holyda
[03:53] <Hobbsee> *holiday
[03:54] <jdub> see, no such thing, Hobbsee can't even spell the word
[03:54] <ogra> Kamion, the installer on ppc tells me it finds no kernel modules
[03:54] <Hobbsee> jdub: hehe!  i cant typing-spell anyway
[03:59] <simira> Mithrandir: hey, time to get home to make pancakes! I am hungry!
[03:59] <^robertj> what's the correct method for requesting a package from non-free be brought into multiverse?
[04:00] <pitti> ^robertj: a sync request, I'd say
[04:00] <Hobbsee> pitti: heh, surely not...
[04:00] <^robertj> pitti: hrmm, where?
[04:00] <fabbione> pitti: apr sounds good.. like libapr.. and make it clash with libapr from apache :P
[04:01] <pitti> ^robertj: file a bug against an unknown package, ask for the sync, give the package name and source and subscribe ubuntu-archive
[04:01] <^robertj> pitti: aha, thanks
[04:01] <pitti> fabbione: the current name is crash-reporter, but that feels like namespace trampling
[04:01] <Mithrandir> pitti: call it apport, which means to fetch something (like, a dog which fetches a stick you throw) as well as the description on http://en.wikipedia.org/wiki/Apport
[04:02] <pitti> Mithrandir: hm, I like 'apport'
[04:02] <Mithrandir> "Category: Occult"
[04:02] <Mithrandir> ;-)
[04:02] <pitti> Mithrandir: has some connotation to a watchdog
[04:02] <Mithrandir> simira: yes, dear.
[04:02] <ogra> Kamion, hw-detect: Missing modules 'ide-core, ide-mod, ide-probe-mod, ide-detect, ide-generic'
[04:02] <jsgotangco> jdub: do you have the fridge drupal theme handy with you?
[04:02] <Mithrandir> pitti: it's also APpoRt so you have the right letters in there. :-P
[04:03] <pitti> Mithrandir: heh @ asport
[04:03] <fabbione> simira: isn't about time to call Mithrandir when pancakes are ready instead? ;)
[04:03] <pitti> Mithrandir: now I know a verb for this well-known effect :)
[04:03] <pitti> fabbione: you have been married for too long already :-P
[04:03] <Mithrandir> pitti: haha. :-)
[04:04] <pitti> Mithrandir: 'crapport' :)
[04:04] <fabbione> pitti: ahaha
[04:04] <pitti> Mithrandir.apport(pancake)
[04:05] <Mithrandir> wuff
[04:05] <^robertj> pitti: bug #53245
[04:05] <Ubugtu> Malone bug 53245 in Ubuntu "sync request for tremulous" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/53245
[04:05] <pitti> ^robertj: and now subscribe ubuntu-archive
[04:06] <^robertj> looking for that op[tion now
[04:06] <Kamion> ogra: not surprising
[04:06] <Kamion> ogra: probably built against the wrong kernel version or something
[04:06] <pitti> ^robertj: 'Subscribe someone else'
[04:06] <^robertj> done
[04:06] <pitti> ^robertj: looks good now
[04:06] <^robertj> thanks a heap, great game btw :)
[04:06] <^robertj> Vicious though...
[04:07] <ogra> Kamion, likely... are these modules coming from a udeb or should i see them in /lib/modules somewhere ?
[04:07] <Kamion> ogra: udebs get unpacked into /lib/modules
[04:07] <Kamion> at least those udebs do
[04:07] <Kamion> modprobe does not magically know how to unpack udebs ;-)
[04:07] <ogra> yes, but thats at the very first screen after selecting the language and scanning the CD
[04:07] <Kamion> ogra: alt-f2, uname -a
[04:07] <ogra> so there are no modules with these names at all
[04:08] <Kamion> the initramfs is also built out of udebs
[04:08] <ogra> 2.6.17-5-powerpc
[04:08] <ogra> which matches /lib/modules
[04:08] <Kamion> ogra: look in /cdrom/pool/main/l/linux-source-2.6.17; see if there are any udebs there with the right kernel versions
[04:09] <ogra> ah, well
[04:10] <ogra> all 2.6.17-4.6 
[04:10] <ogra> so we'll need a d-i rebuild i bet
[04:10] <Kamion> ogra: no
[04:10] <Kamion> ogra: is this edubuntu?
[04:11] <ogra> yes
[04:11] <Kamion> ogra: you need to do a seed merge to pull in my most recent change to the installer seed
[04:11] <ogra> built some hours ago
[04:11] <ogra> eek, ok
[04:11] <Kamion> if the udebs are out of date, that's never something that needs a d-i rebuild
[04:11] <Kamion> because the debian-installer source package does not build any udebs - it is strictly a consumer of udebs
[04:12] <ogra> oh, right they come from the kernel source, i forgot 
[04:12] <Kamion> if the kernel/initramfs are out of date, then *that's* due to d-i being out of date
[04:14] <mxpxpod> are the vmware modules for the new dapper kernel scheduled for compile?
[04:23] <Kamion> ^robertj: FYI we sync source packages not binaries, so it's best to just give the source package name
[04:24] <^robertj> yeah, should have it's just called tremulous so I think if anyone tackles it they will find it without much difficulty :)
[04:26] <Kamion> nod
[04:26] <Kamion> s/if/when/, sync requests are processed relatively promptly
[04:26] <^robertj> ooh, good :)
[04:28] <Hobbsee> Kamion: that they are.  thanks :)
[04:34] <Kamion> Hobbsee: which one was that?
[04:34] <Kamion> I think I did all of yours so I probably overlooked that
[04:34] <Hobbsee> Kamion: ah, it was a gnome one....camorama
[04:35] <Hobbsee> Kamion: hehe thankyou :)
[04:35] <Kamion> I didn't process that; I assume ubuntu-archive wasn't subscribed
[04:36] <Hobbsee> Kamion: okay, i'm not sure why.  i'll grab you a link
[04:36] <Kamion> oh, it's because you assigned it to ubuntu-archive too
[04:36] <Hobbsee> Kamion: oh dear.  did i really do that?
[04:36] <Kamion> that causes it not to show up on the list we normally use (probably due to a Malone bug); I suggest you leave syncs unassigned, as a general rule
[04:37] <Kamion> https://launchpad.net/distros/ubuntu/+source/camorama/+bug/53153/+activity says so :)
[04:37] <Ubugtu> Malone bug 53153 in camorama "[Edgy MoM]  Please sync camorama  0.17-5 from Debian Sid" [Untriaged,Confirmed]  
[04:37] <Hobbsee> Kamion: i thought i subscribed it, sorry about that.  i suspect i might have done that with another one too
[04:38] <Kamion> I'll hunt all other such bugs down now
[04:38] <Kamion> just two
[04:39] <Hobbsee> yeah.  i didnt think i'd done it on the one. sorry about that.  i'd only recently started subscribing the archive myself, rather than getting a MOTU to do it.
[04:39] <Kamion> no problem
[04:48] <ogra> Kamion, does the seed tree have a LP page i can subscribe to to getnotified (its not on SeedManagement or i'm blind)
[05:04] <Kamion> ogra: I guess that would be https://launchpad.net/people/ubuntu-core-dev/+branch/ubuntu-seeds/ubuntu.edgy/+subscribe; I've never tried it, but let me know if notifications work for that and I'll add it to SeedManagement
[05:04] <Kamion> I think it only pulls every day though
[05:04] <ogra> Kamion, thanks, i'll tell you if it works 
[05:04] <Kamion> but probably better than nothing
[05:05] <ogra> yep, i wouldnt have missed the last changes that way ...
[05:05] <ogra> but i have not gotten any change notifications from the other branches, so tha feature might only be available in gui yet :)
[05:05] <ogra> *that
[05:09] <Kamion> ok, anyone who understands gtk and window managers around? gparted is creating dialogs that are marked transient for a GtkPlug, and they're appearing behind the window containing the corresponding GtkSocket
[05:09] <Kamion> I've tried both raise() and present() (although admittedly not in combination)
[05:10] <Kamion> present() has no effect whatsoever here as far as I can tell; raise() does cause the dialog to appear in front of a terminal window that's also on the screen rather than right at the back, but doesn't bring it to the front of the window-containing-the-socket
[05:12] <Riddell> groovy, kubuntu-desktop installable everywhere, I guess that means we can start making CDs
[05:14] <Riddell> Kamion: could you kick off a livefs build for kubuntu?
[05:14] <Kamion> Riddell: running
[05:14] <Kamion> you'll need at least one more before knot-1 of coursse
[05:15] <Riddell> Kamion: why?
[05:16] <Kamion> Riddell: ubiquity's not remotely ready
[05:17] <Kamion> I have a huge pile of stuff in bzr to upload
[05:17] <Riddell> fair enough
[05:17] <Kamion> but only when it works :)
[05:17] <Riddell> Kamion: how's the alternative installer doing?
[05:17] <Kamion> tolerably OK I think
[05:17] <Riddell> I'll start an alternative CD build for kubuntu
[05:29] <elmo> dpkg: error processing /var/cache/apt/archives/linux-kernel-headers_2.6.17-5.13_i386.deb (--unpack): trying to overwrite `/usr/include/scsi/scsi.h', which is also in package libc6-dev
[05:30] <elmo> so, err, do I need to run away screaming from this box now or what?
[05:31] <sladen> niiice.
[05:31] <tseng> i did --force-overwrite and lived happily ever after
[05:31] <Hobbsee> elmo: yeah, i got that.  i used --force-overwrite too
[05:31] <tseng> ignorance is bliss.
[05:31] <elmo> did either of you report bugs?
[05:31] <Hobbsee> elmo: no, i made a mental note to do it though.
[05:31] <tseng> no, I assumed it was so horribly obvious
[05:32] <Hobbsee> sheesh we're terrible
[05:32] <tseng> yeah.
[05:32] <fabbione> elmo: known bug
[05:32] <fabbione> elmo: we need to get Ben to upload a new kernel
[05:33] <Kamion> infinity noticed it and stuck l-k-h on hold in all the buildds' edgy chroots
[05:36] <dholbach> mjg59: hello! i came across bug 49503 and thought that the Power spec had an idea for solving cases like the one mentioned there. or am i wrong?
[05:36] <Ubugtu> Malone bug 49503 in gnome-session "Hibernate button being displayed and functions on remote connections" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/49503
[05:58] <doko> Keybuk: when are merges manual? is this status dropped again, if unstable has a new upstream?
[05:58] <Keybuk> doko: ?
[05:58] <doko> Keybuk: universe-manual.html
[05:58] <HiddenWolf> Keybuk: I have a tv tuner (pci) and a webcam (usb); when both are plugged in it seems random which device is /dev/video0. Is that expected with dapper's udev?
[05:58] <pitti> zul: ping
[05:58] <Keybuk> doko: those are the list of sources for which there is no version in Debian less than the version in Ubuntu
[05:58] <Keybuk> ie. Ubuntu packages 1.0-0ubuntu1 and Debian packages 1.0-1
[05:58] <Keybuk> there's no base version for comparison
[05:58] <Keybuk> HiddenWolf: yes.
[05:58] <doko> ahh, thanks
[05:58] <HiddenWolf> Keybuk: right. :(
[05:58] <Keybuk> HiddenWolf: modern software shouldn't care ... it should present you a list of the available video devices
[05:58] <Keybuk> and remember them by more useful details
[05:58] <HiddenWolf> Keybuk: so the bug is in tvtime. :)
[05:58] <HiddenWolf> Keybuk: thank you
[05:59] <sfllaw> The video layer under Linux is just horrible.
[05:59] <sfllaw> tvtime really can't do much.
[06:00] <Hobbsee> hi sfllaw :)
[06:00] <sfllaw> It's like the bad old days of DOS printers.
[06:08] <ogra> dholbach, the simple bug is gnome-session not caring for the /apps/gnome-power-manager/can_{suspend,hibernate} gconf key 
[06:09] <ogra> if it would do that, the buttons wouldnt be shown
[06:12] <mcquaid> hello i wanted to backport baobab from edgy to dapper.  however no version of baobab is currently in edgy
[06:12] <siretart> https://launchpad.net/distros/ubuntu/+source/baobab
[06:12] <siretart>    Removed on 2006-07-05 12:44:06 UTC
[06:12] <mcquaid> does that happen sometimes when the next version is building up or does that mean a pkg has been dropped?
[06:13] <siretart> why that?
[06:13] <mcquaid> hmm, no reason listed there..
[06:15] <ogra> its included in control-center or gnome-system-tools now you wont be able to backport that right away
[06:16] <mcquaid> huh? sorry not quite getting that. it's become part of gnome-system-tools?
[06:17] <dholbach> siretart: it's in gnome-utils now
[06:17] <ogra> ogra@edubuntu:~/seeds/edubuntu.edgy$ dpkg -S /usr/bin/baobab
[06:17] <ogra> gnome-utils: /usr/bin/baobab
[06:17] <ogra> sorry, gnome-utils ... 
[06:17] <dholbach> siretart: and we have a bug report marked as fix committed, which will add it to the packages description
[06:18] <siretart> ok
[06:18] <mcquaid> oh well, i guess i'll use the swartz
[06:18] <mcquaid> but why do that anyway? baobab has no gnome deps
[06:18] <siretart> I assume gnome-utils handles upgrades from baobab users in dapper gracefully.. 
[06:18] <mcquaid> seems limiting
[06:19] <siretart> thats right. what about xfce users, who want baobab without gnome-utils?
[06:19] <mcquaid> for ex, i use it in xfce
[06:19] <ogra> ask gnome about it ...
[06:19] <dholbach> bad luck... baobab upstream wants to maintain it in gnome-utils and asked me to get rid of the old package
[06:19] <ogra> it was an upstream decision by the looks of it
[06:20] <dholbach> debian will do that too once they have gnome 2.15
[06:20] <dholbach> as it makes no sense to have the same code in the archive twice
[06:21] <mcquaid> yes, but makes just as little to tie something to a particular de when it's not needed
[06:21] <LaserJock> Kamion: ping?
[06:23] <ompaul> No manual entry for pickups <--- this is cos its a real guitar :)
[06:23] <ompaul> woops
[06:24] <Kamion> LaserJock: hi
[06:25] <LaserJock> Kamion: doh, nvm. I wondered why you unassigned ubuntu-archive from some sync bugs.
[06:28] <Kamion> pitti: sudo 1.6.8p12-4ubuntu2 doesn't seem to have the "keep all environment variables for users with unrestricted sudo" behaviour you added
[06:29] <pitti> Kamion: oh, indeed
[06:29] <pitti> Kamion: thanks for spotting, I will fix that
[06:29] <Kamion> thanks - it breaks ubiquity
[06:29] <pitti> Kamion: (it still has the patch, but the Debian fix breaks it)
[06:29] <Kamion> well, slightly. it means that it can't talk to gnome-screensaver
[06:29] <pitti> Kamion: (I'm in a meeting now, will fix later)
[06:29] <Kamion> I figure you can fix it faster than I can :-) but if you don't have the time, let me know and I'll switch to that
[06:34] <ogra> BenC, would it soemhow be possible to change the load order for {e,o}hci_hcd ? seems my laptop only recognizes ohci (USB 1.1) if thats loaded first somehow slowing down my USB disks ...
[06:34] <bluefoxicy> ogra:  I recommend a bios update as well ;)
[06:35] <mjg59> ogra: No
[06:35] <mjg59> That's handled by udev
[06:35] <mjg59> We should fix the bug
[06:35] <ogra> Kamion, does lithium not publish automatically anymore ? i see no trace of  20060717.1
[06:35] <ogra> mjg59, yes, i wasnt aware it was udev 
[06:35] <Riddell> Kamion: where's the place to look for livefs buildlogs?
[06:37] <BenC> ogra: It should matter...ehci should work still
[06:37] <Kamion> Riddell: http://people.ubuntu.com/~cjwatson/livefs-build-logs/
[06:37] <BenC> shouldn't matter
[06:37] <BenC> 2.6.7-5.14 coming to a buildd near you...
[06:37] <ogra> BenC, i have a dmesg entry that the device will only be used at low speed as long as ohci is loaded
[06:37] <Kamion> ogra: look at the end of the build log and the reason will be clear (i.e. I broke some code); fixing
[06:37] <ogra> ah
[06:37] <Kamion> ogra: please always look at the build log before asking me :)
[06:38] <ogra> Kamion, will do, sorry
[06:38] <mcquaid> btw, regarding packages, since i went to dapper i've noticed a fair amount of 'orphaned' packages listed in synaptic but can't actually be installed as it's no longer available
[06:39] <BenC> mcquaid: python stuff?
[06:39] <Kamion> only edubuntu was affected, so I'll fix it by handd
[06:39] <mcquaid> is there any way to clear those up, or is it other avail programs that referenced it in some dep
[06:39] <Kamion> mcquaid: deborphan
[06:39] <mcquaid> BenC, for example libgtop-2.5 is still listed
[06:39] <Kamion> oh, ones you don't have installed?
[06:39] <pitti> Kamion: for me, I can start fixing it in about 90 minutes; how urgent is it?
[06:39] <mcquaid> yes
[06:39] <Kamion> I think those are just ones that are referenced in dependencies
[06:40] <mcquaid> ok, 
[06:40] <Kamion> pitti: I guess that would be OK; do you know the fix already?
[06:41] <pitti> Kamion: shouldn't be hard to do, I guesstimate 45 minutes for fixing and proper testing
[06:43] <Kamion> ah, if you don't know the fix immediately I'll leave it to you; I have a lot of other knot-1-critical-path stuff to do
[06:43] <Kamion> pitti: could you make sure to roll in the fix in bug 53273? the consequences of that bug are a bit nasty
[06:43] <Ubugtu> Malone bug 53273 in sudo "prerm doesn't work with dash" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/53273
[06:43] <pitti> oh, another bashism?
[06:43] <pitti> sure
[06:44] <ogra> pitti,  we have tons of them
[06:44] <Kamion> argh, can we please at least put command editing into dash? it's really frustrating not having that
[06:44] <Kamion> ogra: the bulk of them are in scripts that say #!/bin/bash though
[06:44] <Kamion> which is not ideal but not a bug either
[06:44] <ogra> well, *my* scripts didnt until now :)
[06:45] <ogra> but i have seem some other incidents ...
[06:45] <ogra> like debian/rules failing because of "rm -f /blah/{targetfile1,targetfile2}
[06:45] <ogra> "
[06:46] <Kamion> yeah, there's not a lot of that left though
[06:46] <Kamion> most of that was dealt with in Debian a long time ago
[06:46] <ogra> but debian never switched ...
[06:47] <ogra> so it might have reappeared 
[06:47] <Kamion> there have been people building stuff with dash as /bin/sh to test for that
[06:47] <Kamion> and lintian warns about its use in maintainer scripts
[06:48] <Kamion> so it doesn't tend to reappear all that much
[06:48] <Kamion> yes, there is the odd exception
[06:48] <ogra> :)
[06:48] <Kamion> generally speaking, though, it's in dodgy upstream shell programs rather than debian/rules or maintainer scripts
[06:49] <Kamion> the bulk of Debian maintainers have been made well aware of the problems by very long mail threads and lots of bug reports
[06:50] <ogra> yu, i noticed that ... its hard to find info on bashisms since you only get a long list of debian bugs at google :)
[06:50] <ogra> *yup
[06:51] <Kamion> http://www.opengroup.org/onlinepubs/009695399/nfindex.html is the POSIX specification, which includes the shell
[06:52] <Kamion> avoid things marked [XSI]  in /bin/sh scripts
[06:52] <Kamion> and avoid things not listed at all
[06:52] <ogra> thanks a lot, i'll see if i can get ltsp dash clean ...
[06:53] <Kamion> the dash man page is a reasonable thing to program to, although not perfect
[06:53] <ogra> ltsp-build-client currently relies somewhat on bash stuff ...
[06:53] <Kamion> but it's a lot better than programming to the bash man page
[06:53] <Kamion> ogra: edubuntu 20060717.1 fixed
[06:53] <ogra> well, i'm an old abs-howto user :) i never programmed after manpages
[06:53] <ogra> Kamion, thanks
[06:54] <Kamion> I'd run away from all the howtos
[06:54] <Kamion> they are rarely written with an eye to portable shell, more with an eye to SHINY
[06:54] <ogra> well it was what i used to learn bash scripting ages ago ...
[06:55] <ogra> but indeed you see my code isnt bashism free now (in fact i ever cared before i saw dash the first time )
[06:56] <Kamion> http://www.greenend.org.uk/rjk/2001/04/shell.html is also worth reading, written by a friend of mine; it's not specifically about portability but is generally worthwhile reading for shell programmers
[07:02] <Keybuk> also get a copy of Blinn
[07:02] <ogra> blinn ? 
[07:08] <Keybuk> http://www.amazon.com/gp/product/0134514947/
[07:08] <ogra> ugh, dead trees
[07:09] <ogra> i even read my pratchetts on the laptop nowadays :)
[07:09] <ogra> but i'll get a copy ... )
[07:11] <mjg59> iwj: My fonts have changed in Epiphany again
[07:12] <iwj> mjg59: Oh ?
[07:14] <iwj> (I have to go soon; it's Monday and if I don't I'll miss my dinner ...)
[07:14] <mjg59> iwj: Back to blurry colour madness
[07:14] <mjg59> iwj: Sure, no urgency
[07:14] <iwj> Is it using Nimbus ?
[07:14] <iwj> (Easiest way to tell: select Nimbus from the font preferences and see what it does.)
[07:14] <iwj> Perhaps some of my fontconfig-related changes have been undone or not merged properly.
[07:15] <ogra> mjg59, on a lappie ? i noticed that subpixel rendering was disabled after an upgrade to edgy
[07:15] <ogra> just switching it on in the fonts ui solved it here
[07:18] <mjg59> iwj: Hm. Could be. The font preferences don't seem to let me change the main one that's being used
[07:18] <iwj> Ug.
[07:18] <Keybuk> Kamion: your friend wins a "useless use of $@ award"
[07:18] <Keybuk> sfere$ for x in "$@"; do  echo "[$x] "; done
[07:18] <Keybuk> ...
[07:18] <mjg59> ogra: If subpixel rendering was disabled, I don't think I'd be seeing coloured blur
[07:18] <Keybuk> why not just "for x; do" :p
[07:18] <ogra> mjg59, thats how it looks on my screen ...
[07:19] <ogra> if i switch on subpixel rendering its crisp again
[07:20] <mjg59> ogra: If there's no sub-pixel anti-aliasing, then there is no coloured blurring
[07:20] <mjg59> iwj: Interestingly, it's not nimbus. Hrm.
[07:23] <Kamion> Keybuk: given that it's in a section entitled "Special Behaviour of $@", I think it's semi-justifiable
[07:23] <ogra> mjg59, well, then the ption in the font settings gui might be mis-named ... i can just tell waht i see here 
[07:23] <Kamion> Keybuk: but you could always mail him and suggest he add a note
[07:23] <Keybuk> Kamion: I disagree ... a bad example, even when demonstrating a particular behaviour, is a bad example
[07:23] <Keybuk> people tend to copy and paste examples without reading the surrounding text
[07:23] <Kamion> then I repeat my second comment :)
[07:23] <Keybuk> cf. the famous "wrong example of wait()" in Stevens which has been copied into many UNIX programs and defended
[07:23] <Keybuk> I have just mailed him, actually
[07:23] <Kamion> ta
[07:24] <Keybuk> (Stevens gives an example of a SIGCHLD handler that calls wait() as a general example of the kinds of things you do in such handlers ... this handler has been used as the template for many a developer's handler
[07:25] <Keybuk> the problem is that it calls wait() only once ... when you in fact need to do it in a loop to consume all child processes, as you may have received more than one signal or had signals dropped as you were beginning processing of one)
[07:25] <Keybuk> he fortunately corrected the example in later books (UNP carries a correct one)
[07:30] <Kamion> Keybuk: urgh, speaking of dodgy wait code, look at /usr/lib/python2.4/site-packages/apt/progress.py and search for waitChild
[07:30] <Kamion> let's wait for children by SPINNING
[07:30] <Keybuk> *blink*
[07:31] <Keybuk> I can see why they've done that ... but ... eww
[07:31] <Kamion> what's wrong with read() and a signal handler
[07:31] <Keybuk> I'd select() on an empty list ;)  but yeah, same thing
[07:33] <mdz_> fabbione: that isn't what I mean; for example, is ubuntu-desktop on sparc installable?  are the livefses building?
[07:34] <Kamion> Keybuk: actually select() is indeed better because the reason it spins is that (in my case) read() is immediately returning EAGAIN
[07:37] <Kamion> -> bug 53282
[07:37] <Ubugtu> Malone bug 53282 in python-apt "InstallProgress.waitChild spins if read fails immediately" [Untriaged,Unconfirmed]  http://launchpad.net/bugs/53282
[07:43] <ogra> hmm, intresting 
[07:44] <ogra> that daily iso rsync had only changes on ppc
[08:10] <ogra> ARGH
[08:10] <ogra> Unknown terminal: bterm 
[08:11] <ogra> where does that come from in the installer ??? 
[08:12] <sitontony> i have a suggestion
[08:14] <sitontony> i have a suggestion for the developement
[08:15] <Keybuk> Kamion: I guess the Debian BTS LDAP gateway is down atm?
[08:24] <pitti> Kamion: sudo still purges if root once had a password which was then locked with passwd -l root
[08:25] <pitti> Kamion: so I'd like to fix the prerm to check whether getent shadow root *starts* with ! instead of checking whether it is equal to !
[08:25] <pitti> Kamion: are you fine with that?
[08:25] <fabbione> mdz_: no i didn't check because it was higher priority to finish other stuff. Also sparc desktop for knot-1 can be broken. It's not a regression and there for not so important
[08:34] <pitti> Mithrandir: ok to upload a new sudo with the env hanling fix Kamion needs for ubiquity? I also fixed a bashism and a bug in the prerm
[08:34] <fabbione> pitti: openais with your requested changes is up
[08:35] <pitti> thanks
[08:35] <fabbione> pitti: it's only waiting your magic touch to enter main
[08:35] <fabbione> pitti: so that i can unleash the new r-h-c-s
[08:35] <pitti> fabbione: -ENOMAGIC here
[08:35] <fabbione> pitti: and transit lvm2 to the new locking manager
[08:35] <pitti> fabbione: that requires ftpmaster powers (Kamion/Keybuk)
[08:35] <fabbione> pitti: well you still need to tell ftp-master to go ahead :)
[08:35] <pitti> fabbione: anyone can, they'll look at the MIR queue
[08:36] <pitti> Kamion, Keybuk: for the record, openais is good for main
[08:37] <pitti> grr, I'm a fool, I just locked myself out of sudo
[08:38] <fabbione> pitti: you can use one of the unpatched exploit :P
[08:38] <pitti> heh, good idea :) the prctl one should work
[08:38] <fabbione> yeah they actually do.. i think elmo knows :P
[08:40] <pitti> fabbione: the advantage of the prctl one is that the exploit works fine on all arches; that's pretty rare usually :)
[08:46] <Mithrandir> pitti: please
[08:46] <ogra> xorg postinst fails
[08:51] <Burgwork> ajmitch, do you have a blog for your network auth stuff?
[08:52] <Burgwork> Petaris, I just asked him
[08:52] <Petaris> Burgwork: Ok
[09:04] <ogra> Jul 17 18:45:04 in-target: Setting up xserver-xorg (7.0.22ubuntu5) ...
[09:04] <ogra> Jul 17 18:45:05 debconf: Obsolete command TITLE Configuring xserver-xorg called
[09:04] <ogra> Jul 17 18:45:08 in-target: xserver-xorg postinst warning: not updating /etc/X11/X; file has been
[09:04] <ogra> Jul 17 18:45:08 in-target:    customized
[09:04] <ogra> Jul 17 18:45:08 in-target: [: 31: 
[09:04] <ogra> Jul 17 18:45:08 in-target: ==: unexpected operator
[09:04] <Keybuk> BASHISM!
[09:04] <ogra> ideas anyone ? i have no clue what that should tell me ... but it breaks edubuntu installs 
[09:05] <ogra> *sigh*
[09:05] <jdub> :o
[09:05] <ogra> funnily i cant find any == in there
[09:05] <Mithrandir> ogra: I think Colin fixed that later today.
[09:05] <ogra> nope, its the recent package
[09:06] <ogra> it installs fine (even though it shows the error) in the normal system ... but breaks ltsp chroots 
[09:06] <sladen> s/==/=/
[09:08] <ogra> i would have sworn there is no == in that file at all .. but i just found it :D
[09:08] <ivoks> why do we have /var/lock and /dev/shm mounted as tmpfs with default size (each is half of RAM), and every user can write in those dirs
[09:10] <Keybuk> ivoks: why does it matter?
[09:10] <Keybuk> any user can malloc() the equivalent size anyway
[09:10] <Keybuk> it's just page cache
[09:11] <ivoks> i know, but...
[09:11] <ivoks> do we really need 256MB for /var/lock? :)
[09:11] <Keybuk> no, you could always add an appropriate limiting option in /etc/fstab if you needed it
[09:12] <ivoks> sure... well, there's no harm in leaving it like this...
[09:12] <Keybuk> note that you can set this in /etc/default/tmpfs
[09:13] <Keybuk> (for /dev/shm, that is)
[09:17] <ogra> Mithrandir, o if i upload a bashism free package for xorg ? 
[09:18] <ogra> *ok even
[09:18] <Mithrandir> ogra: can I look at the diff, please?
[09:18] <ogra> heh, sure
[09:19] <ogra> +++ /tmp/GfWNPqHI0J/xorg-7.0.22ubuntu6/debian/xserver-xorg.postinst.in  2006-07-17 21:13:09.000000000 +0200
[09:19] <ogra> @@ -383,7 +383,7 @@
[09:19] <ogra> 
[09:19] <ogra>    # Set a sane monitor id default if probe fails
[09:19] <ogra>    if [ -z "$MONITOR_IDENTIFIER" ] ; then
[09:19] <ogra> -      MONITOR_IDENTIFIER=="Generic Monitor"
[09:19] <ogra> +      MONITOR_IDENTIFIER="Generic Monitor"
[09:19] <ogra> and a changelog entry
[09:19] <ogra> Mithrandir, ^^
[09:21] <Mithrandir> that's not even a bashism.  It's just illegal shell.  Looks good to me.
[09:22] <ogra> i wonder if it finishes before all kernel builds :)
[09:28] <bluefoxicy> pitti:  utrace?
[09:28] <bluefoxicy> pitti:  http://www.redhat.com/archives/fedora-cvs-commits/2005-October/msg00448.html as well.
[09:36] <pitti> bluefoxicy: that looks useful, thanks for digging it out
[09:37] <tseng> bluefoxicy: i met that jjohnson guy 2 years ago
[09:37] <tseng> with russel
[09:39] <bluefoxicy> pitti:  yeah, also <drow> you do know other people have written code to automatically attach gdb to things, right?  Last time I looked, gnome and kde could do it. <DannyB> drow: it's not good unless you've written it yourself <DannyB> don't you know anything
[09:42] <bluefoxicy> tseng:  interesting.  BTW mdz wants to put the libgcrypt11 update into dapper-backports instead of dapper-updates.
[09:42] <tseng> bluefoxicy: er, ok.
[10:12] <Kamion> Keybuk: BTS LDAP> no idea, sorry, I've never really followed that
[10:12] <Kamion> ogra: export TERM=linux and get on with your life :)
[10:12] <Kamion> ogra: the installer uses a terminal called 'bterm' and unfortunately not much else really understands that terminal type
[10:12] <ogra> Kamion, hmm, but why does it do that suddenly ? it worked before
[10:12] <Kamion> pitti: sudo.prerm> sounds reasonable
[10:13] <Keybuk> Kamion: I did a BeautifulSoup version instead -- cf. ubuntu-archive
[10:13] <Mithrandir> Kamion: what bts ldap thing?
[10:14] <Kamion> ogra: while that MONITOR_IDENTIFIER fix was correct, it's not what the warning was about
[10:14] <Kamion> Mithrandir: there's an LDAP gateway to bugs.debian.org which aba (IIRC) wrote
[10:15] <Mithrandir> Kamion: oh, that's down pending the cronjobs on merkel being reenabled.
[10:15] <ogra> Kamion, whats it then ? i didnt understand it 
[10:15] <Mithrandir> AIUI
[10:15] <Kamion> ogra: just some stuff complains about a $TERM it doesn't understand, that's all; various things have warned about it in rescue mode for years
[10:15] <Kamion> Mithrandir: sounds plausible
[10:15] <Kamion> ogra: well, I don't know yet or I'd have said :)
[10:15] <ogra> Kamion, ok, i'll add a TERM=linux to the top of the ltsp-build-client script, even i find it weird that it always worked until now
[10:17] <Kamion> ogra: you didn't give me any context about where the "Unknown terminal" message was coming from before now
[10:17] <Kamion> you should check what you're using that's trying to open a terminal; nothing in ltsp-build-client should be trying to talk to the terminal surely
[10:17] <ogra> Kamion, well, from xserver-xorg prostinst ...
[10:18] <Kamion> the unknown terminal message is coming from xserver-xorg.postinst?
[10:18] <ogra> ltsp-build-client is supposed to fail if the second stage install (xorg and ldm) in the chroot fails
[10:18] <ogra> yes
[10:18] <Kamion> I've found the == bashism; it's in xresprobe. fixing
[10:18] <ogra> i have an error with configuring xorg and xserver-xorg failed
[10:19] <ogra> ah
[11:23] <Kamion> world conspiring against me. now apt is behaving weirdly.