[12:03] <seb128> elmo: nautilus-cd-burner sync please
[12:11] <herzi> seb128: i work on a gaim-devel now
[12:11] <seb128> have you contacted the debian maintainers ?
[12:12] <herzi> not yet
[12:12] <herzi> i want to have a patch ready that works on my box
[12:12] <herzi> then i will contact them
[12:12] <seb128> you should have started by this
[12:13] <seb128> I think they have perhaps the change ready
[12:13] <seb128> they were waiting for the debian freeze for this change apparently
[12:15] <seb128> oups
[12:23] <sabdfl> night all
[12:24] <Matt|> nite
[12:25] <seb128> 'night
[12:25] <sivang> night sabdfl
[12:51] (lamont/#ubuntu-devel) elmo?
[01:59] <zul> hey
[04:02] (Kamion/#ubuntu-devel) lamont: can you try kicking off another set of builds of debian-installer 20041118ubuntu2? I fixed the hw-detect bug that broke it just after uploading it
[05:35] <elmo> Kamion: done
[05:35] <elmo> night all
[06:39] <fabbione> morning guys
[06:43] <mdz> morning
[06:43] <fabbione> hey mdz
[06:45] (lamont/#ubuntu-devel) Kamion: will do
[06:49] <fabbione> Kamion: are you still around?
[06:52] (lamont/#ubuntu-devel) ENOKEYBUK
[06:52] <fabbione> heheh
[06:57] <fabbione> mdz: 3745... not yet :P
[07:05] <fabbione> ehehe
[07:29] (lamont/#ubuntu-devel) no daniels either.  hrm.
[07:29] <fabbione> lamont: something about X?
[07:31] (lamont/#ubuntu-devel) just gonna pester him about xorg upload with ia64 support.. :)
[07:31] <fabbione> lamont: it's pending upload
[07:31] <fabbione> i will kick him today
[07:32] (lamont/#ubuntu-devel) There are 37 (or so) main packages remaining to be built, most of which are either d-w xorg, or blocked by gcc-3.4
[07:32] <fabbione> lamont: yeah.. i know..
[07:32] <fabbione> i will talk to him to do an upload today
[07:33] <fabbione> or max tomorrow
[07:33] (lamont/#ubuntu-devel) np - I was just quoting my activity report.. :)
[07:33] <fabbione> ahah
[07:35] <mdz> good night, all
[07:37] <fabbione> night mdz!
[07:49] <doko> morning all.
[07:49] <doko> lamont: please let's sort out the libunwind problems for ia64 first, before doing the gcc upload
[07:51] (lamont/#ubuntu-devel) doko: it won't build until something happens...
[07:51] (lamont/#ubuntu-devel) cc'ed you on the mail to t-bone telling him what he could work on.
[07:52] (lamont/#ubuntu-devel) that is, I'm trying to let t-bone do all the real work.
[07:56] <doko> lamont: ok
[07:57] (lamont/#ubuntu-devel) night all
[08:04] <fabbione> night lamont
[09:05] <pitti> Morning
[09:28] <jdub> Keybuk, mdz: http://lwn.net/Articles/111859/
[09:30] <Keybuk> jdub: cute ... though udevd looks more "interesting"
[09:30] <Keybuk> as it actually deals with the various races hotplug can introduce
[09:31] <Keybuk> (remove beating add, for example)
[09:31] <jdub> yeah, where can i find out more about it?
[09:31] <Keybuk> lkml, lwn, etc.
[09:31] <jdub> hrm
[09:31] <jdub> must of missed it in lwn
[09:32] <Keybuk> is part of the udev package these days
[09:32] <Keybuk> it's even in hoary
[09:32] <Keybuk> syndicate scott% whatis udevd
[09:32] <Keybuk> udevd (8)            - udev event serializer daemon and udev event sender
[09:33] <Keybuk> you basically use udevsend instead of hotplug, and udevd does the work in the right order
[09:33] <jdub> cool
[09:34] <smurfix> http://linux-hotplug.sourceforge.net/; kernel.org/pub/linux/utils/kernel/hotplug/
[09:35] <jdub> maybe hotplug-ng is gregkh's attempt to fix the hotplug problems though
[09:35] <Keybuk> jdub: reading his code, it isn't even trying
[09:35] <Keybuk> he's basically just written run-parts
[09:36] <jdub> hrm, udevd is not mentioned even as far back as october 28th
[09:36] <Keybuk> I don't what the ideal final solution is yet
[09:37] <Keybuk> udevd seems to be the fore-runner, with hotplug-ng C scripts replacing the existing shell hotplug agents
[09:37] <Keybuk> (you still need all the stuff to do module and firmware loading)
[09:37] <fabbione> hey Keybuk 
[09:37] <fabbione> is daniels around there?
[09:37] <Keybuk> fabbione: he was still in bed
[09:38] <jdub> Keybuk: so /sbin/hotplug calls udevsend?
[09:38] <fabbione> Keybuk: ok
[09:38] <Keybuk> jdub: no, you set the kernel to call udevsend instead of /sbin/hotplug
[09:38] <Keybuk> udevsend informs udevd of the new event
[09:39] <Keybuk> and udevd makes sure /sys is populated, creates the device node and runs the /etc/hotplug* scripts for it
[09:39] <jdub> hrm, positive review of mepis from ladislav in lwn
[09:39] <jdub> Keybuk: ahh
[09:40] <Keybuk> udevd also makes sure the events actually happen in the right order
[09:41] <pitti> Keybuk: btw, /sys/ is automatically populated by the kernel; you mean populating /dev/?
[09:41] <pitti> Hi sivang!
[09:41] <Keybuk> pitti: no, udevd *makes sure* /sys is populated
[09:42] <Keybuk> when the kernel fires the hotplug event, it usually hasn't actually got around to populating /sys yet
[09:42] <pitti> Keybuk: ah, you mean it waits for the sysfs entries?
[09:42] <Keybuk> so the hotplug event runs, and has to sit and sleep until the right /sys path shows up
[09:42] <pitti> Keybuk: this is a nice feature; so far we have too many races because of still missing /sys entries
[09:42] <Keybuk> indeed
[09:42] <Keybuk> there's also the "add event takes longer than the remove event" issue
[09:43] <Keybuk> where you plug a device and unplug it quickly
[09:43] <Keybuk> and the add event is still finishing after the remove event has been fired and finished
[09:43] <jdub> Keybuk: could udevd use inotify on /sys? :)
[09:43] <Keybuk> jdub: dunno whether inotify works on /sys -- possibl
[09:55] <mojo> hi all devel
[09:55] <mojo> I found prob of RealPlayer that make it not run on Hoary
[09:56] <mojo> if remove the swfformat.so and swfformat.so in plugins folder
[09:56] <mojo> and that will fix it
[09:56] <mojo> just move the swfformat.so and swfformat.so in plugins folder in RealPlayer folder
[09:56] (bob2/#ubuntu-devel) you should add that to the wiki
[09:56] <mojo> it's still a bug
[09:57] <mojo> I'm wondering y swf has prob with Debian
[09:57] <mojo> ..
[09:57] (bob2/#ubuntu-devel) you mean helixplayer, right?
[09:57] <fabbione> hey daniels 
[09:57] <mojo> no
[09:57] <mojo> RealPLayer
[09:57] <fabbione> daniels: when do you plan to upload the next X?
[09:57] <mojo> is there any relation of swfrender and swfformat in Debian and Ubuntu?
[09:57] <fabbione> daniels: since you have planned some major changes it is a good idea to start now
[09:57] <mojo> sorri
[09:58] <mojo> I say remove not move
[09:58] <mojo> just remove 2 files
[09:58] <fabbione> daniels: early breakage will give you more time to fix stuff around
[09:58] <fabbione> daniels: and ia64 buildds are waiting for you :-)
[09:58] (bob2/#ubuntu-devel) mojo: where can one get the source for realplayer?
[09:58] <mojo> http://forms.helixcommunity.org/helix/builds/
[09:59] <mojo> get the latest src
[09:59] (bob2/#ubuntu-devel) erm. that's helixplayer.
[09:59] <mojo> and can u please check the swf plugins
[09:59] <mojo> there's RealPlayer there
[10:00] (bob2/#ubuntu-devel) ok.  I'm at work atm, sorry, might have a look later.
[10:00] (bob2/#ubuntu-devel) someone has ITP'd it for debian, tho, you might want to talk to them
[10:00] <mojo> it's just acutally Helix Player that is on respo
[10:00] <mojo> I can't find any RealPLayer there
[10:00] <daniels> fabbione: today
[10:00] <daniels> lamont: ping
[10:01] <mojo> if u know any Debianized package of RealPlayer, please give me the URL
[10:01] (bob2/#ubuntu-devel) mojo: google: "helixplayer itp"
[10:01] <mojo> ok
[10:01] <mojo> what is itp?
[10:01] (bob2/#ubuntu-devel) intent to package
[10:02] <mojo> k
[10:08] <pitti> Hi silbs, hi mvo_!
[10:08] <mvo_> hi pitti 
[10:08] <mvo_> hi silbs 
[10:09] <seb128> morning
[10:20] <mojo> bob2: it
[10:21] <mojo> seems to me that's a bug of RealPlayer
[10:21] <mojo> bob2: and helix-player itp doesn't get involve in Realplayer so they won't give me a hand
[10:22] <jdub> Our plan is to produce a usable version of beagle that can be shipped
[10:22] <jdub> as part of SUSE 9.3.  This almost exactly corresponds to the timeline
[10:22] <jdub> for GNOME 2.10.  We call this goal "Milestone One," and the tasks
[10:22] <jdub> required to reach that goal are labeled as such in
[10:22] <jdub> bugzilla.gnome.org.
[10:22] <jdub> 
[10:22] <jdub> Keybuk: ^
[10:22] (bob2/#ubuntu-devel) yeah, robert love went and assigned a bunch of novell people tasks
[10:22] (bob2/#ubuntu-devel) including "make it not crash all the bloody time"
[10:24] <Keybuk> jdub: cool, so you think we can sneak that into hoary too? 
[10:25] <jdub> Keybuk: perhaps we should seriously consider it :)
[10:25] <jdub> not a lot of dependencies we don't have
[10:25] <jdub> kernel will be sorted
[10:25] <jdub> i'm sure we can inspire a new d-bus release
[10:26] <Keybuk> jdub: yeah, I think it sounds like a goal to me
[10:26] <jdub> no gsf cil stuff
[10:26] <jdub> hrm, no evolution sharp
[10:26] <jdub> weird
[10:27] <Keybuk> no evo# where?
[10:27] <jdub> hoary
[10:28] <Keybuk> ah, is it in Debian?
[10:28] <Keybuk> that's going to be a tomboy dep too, isn't it?
[10:29] <jdub> doubt it
[10:30] <seb128> jdub: is there any plan to upload polypaudio in Debian one day ?
[10:31] <jdub> seb128: was planning to upload 0.7, but there are problems with it.
[10:32] <seb128> ok
[10:32] <jdub> seb128: so working those out with lennart, then it'll go straight into debian and hoary.
[10:32] <jdub> seb128: want to sponsor it? :)
[10:32] <seb128> what kind of problem ?
[10:33] <jdub> b0rked autofoo
[10:34] <seb128> jdub: for the sponsoring if you want, but I thought it was better to find a non-canonical people ?
[10:35] <jdub> seb128: i think that's *very* important for becoming a DD, but less important for sponsoring.
[10:36] <jdub> seb128: i wanted to do it, but it can be harder. :)
[10:36] <seb128> ok
[10:55] <azeem> jdub: I can still sponsor polypaudio for you if you want, btw
[11:10] <fabbione> seb128: thanks for fixing gdm, is that enough for the other gnome-suite as well?
[11:10] <fabbione> seb128: i have the problem with panels too
[11:11] <seb128> no panel problem is a libwnck bug, I'm fixing it for Debian right now it should be fixed in hoary as soon as libwnck is synced after that
[11:11] <fabbione> cool thanks!
[11:11] <seb128> s/no panel/no, the panel/
[11:11] <seb128> np
[11:11] <fabbione> seb128: did i ever tell you that you rock?
[11:11] <fabbione> ;)
[11:11] <seb128> thanks :)
[11:11] <seb128> (you rock too ;)
[11:12] <fabbione> :P
[11:16] <seb128> :)
[11:16] <seb128> hey rburton 
[11:16] <rburton> hi seb128 
[11:25] <fabbione> if you think that compiling X is slow.. build the kernel
[11:29] <jdub> $ gpg --list-keys jdub
[11:29] <jdub> gpg: WARNING: using insecure memory!
[11:29] <jdub> ^ should that happen with our gpg?
[11:29] <fabbione> nope
[11:30] <azeem> it happens if gnupg is not suid, I think
[11:30] <jdub> but we have user mlock
[11:30] <fabbione> azeem: iirc out gpg is patched for that
[11:30] <azeem> nicey
[11:30] <jdub> hrm, maybe mjg59's kernel doesn't do that
[11:30] <fabbione> s/out/our
[11:31] <azeem> but gnupg drops the priviledges as soon as it got hold of secure memory anyway
[11:38] <fabbione> and there we go... my laptop has been ubuntified too
[11:38] <jdub> i love reading people getting angry about "all this attention on a new distribution, that ubuntu thing"
[11:39] <sivang> jdub : well, eat their (red) hats out :) 
[11:39] <sivang> ;-)
[11:42] <fabbione> daniels: did you upload all the xorg goodies in the archive? like composite manager?
[11:42] <daniels> fabbione: NEW
[11:44] <pitti> Hey, I just got my Warty CD shipment
[11:44] <pitti> They really look nice
[11:44] (bob2/#ubuntu-devel) woo
[11:47] <fabbione> i still didn't.. but hounestly.. i can't wait :P
[11:48] <sivang> shipment cds
[11:48] <pitti> they came too late for the Linux fair here
[11:48] <pitti> but still there are enough linux freaks in our University :-)
[11:52] <seb128> elmo: libwnck sync please
[12:12] <daniels> Kamion: ping
[12:18] <elmo> seb128: done
[12:20] (thom/#ubuntu-devel) morning
[12:20] <fabbione> morning elmo, thom
[12:21] <fabbione> elmo: do you have any objections to start looking at sparc.u.c ?
[12:22] <fabbione> elmo: i have started building the "golden debs"
[12:22] <fabbione> and theoretically i can also upload them
[12:22] (Mithrandir/#ubuntu-devel) fabbione: crack!  I want my craaaaack.
[12:22] <fabbione> Mithrandir: soon :-)
[12:22] <fabbione> Mithrandir: it will take another week or so to finish
[12:22] <fabbione> + still missing the kernel stuff
[12:23] <fabbione> and in order to get everything in the proper place we will need to do a bit more stuff after that
[12:23] <fabiand> hey, someone experienced problems with vconfig and/or the 8021q module?
[12:23] <fabbione> jdub: after you get main, start building universe you lazy guy :P
[12:24] <seb128> elmo: thanks
[12:24] <seb128> hey thom 
[12:25] <daniels> fabbione: how well set up is linux-restricted-modules to deal with multiple nvidia versions?
[12:26] <fabbione> daniels: no idea.. i did only the first packaging.. 
[12:26] <daniels> fabbione: i386/amd64 are 6629, while ia64 is 5336
[12:26] <daniels> ok
[12:26] <fabbione> daniels: but i guess pretty well
[12:26] <fabbione> daniels: 6629 is buggy
[12:27] <fabbione> you need to patch it
[12:27] <fabbione> or wait for the next release (better
[12:27] <fabbione> )
[12:27] <daniels> how's it buggy?
[12:28] <lupus_> I know ubuntu does not support sun jvm but a lot of users are using it
[12:29] <lupus_> http://news.com.com/Java+flaw+could+lead+to+Windows%2C+Linux+attacks/2100-1002_3-5464872.html?tag=nefd.top
[12:31] (Mithrandir/#ubuntu-devel) daniels: l-r-m handles it ok-ish.
[12:31] <fabbione> daniels: something about agp
[12:31] <lupus_> A flaw in Sun Microsystems' plug-in for running Java on a variety of browsers and operating systems could allow a virus to spread through Microsoft Windows and Linux PCs.
[12:31] <lupus_> should ubuntu warn it's users for this or not?
[12:34] (Kamion/#ubuntu-devel) daniels: pong
[12:37] <daniels> Kamion: can you please chat to sideshow for a sec about powerpc stuff?
[12:37] <daniels> fabbione: hooray for nvidia :P
[12:38] <daniels> Kamion: he's having severe troubles booting his iBook -- both 2.6.8.1 (ours) and his self-compiled 2.6.9 refuse to find his root device?
[12:38] <daniels> s/.$//
[12:41] <fabbione> daniels: did i ever claim that they are cool?
[12:41] <daniels> fabbione: you seem to like them
[12:42] <fabbione> Kamion: nice work with that ddetect thingy :-)
[12:42] <fabbione> daniels: gimme multihead without and i will get rid of them NOW
[12:42] <daniels> fabbione: ati :)
[12:42] <fabbione> daniels: than gimme the money to change gfx
[12:43] (Kamion/#ubuntu-devel) daniels: filesystem?
[12:43] <fabbione> Kamion: this morning i discussed with mdz about creating udebs from linux-source and he convinced me to take the goal for hoary :-)
[12:43] (Kamion/#ubuntu-devel) daniels: sounds like an initrd problem anyway; I don't know the answer to his problem offhand, but I'd boot with init=/bin/sh and step through it by hand
[12:43] (Kamion/#ubuntu-devel) fabbione: cool
[12:43] <fabbione> Kamion: he also told me that you already have some ideas on how to do it
[12:43] <fabbione> Kamion: should we take a look to it later today?
[12:44] <fabbione> (i am still building the ccache for the kernel source..)
[12:44] <fabbione> and i would like to get some lunch before
[12:45] (Kamion/#ubuntu-devel) ideas> I do?
[12:45] (Kamion/#ubuntu-devel) it's mostly a linux-source-* build system job, and I don't know it well at all
[12:46] <daniels> Kamion: ext3, presumably
[12:46] <fabbione> Kamion: well ok.. but the i think mdz was talking about a way to simply the management of the module lists for the udebs (or something like that)
[12:46] <daniels> Kamion: the kernel won't find the root FS (i.e. i have no root and i must scream)
[12:46] <fabbione> s/simply/simplify
[12:47] <daniels> Kamion: while specifying root=, initrd= by hand also
[12:47] <fabbione> Err http://people.ubuntulinux.org ./ xcompmgr 1.1.1+cvs.20041109-0ubuntu1
[12:47] <fabbione>   404 Not Found
[12:47] <fabbione> daniels: mind to fix you Packages file?
[12:47] <daniels> fabbione: fix your head :P
[12:47] <daniels> fabbione: deb http://people.ubuntu.com/~daniels/ xcompmgr/
[12:48] <fabbione> bah
[12:48] <daniels> fabbione: alternately, they just got ACCEPTED about half an hour ago ...
[12:48] <daniels> fabbione: i have fdclock/ also, and a broken transset/
[12:48] <fabbione> daniels: ok...
[12:48] <daniels> but working versions of all of the above should be in hoary
[12:49] (Kamion/#ubuntu-devel) daniels: boot from CD and poke around to see if the initrd's sane?
[12:49] (Kamion/#ubuntu-devel) fabbione: we have to keep with approximately the layout of linux-kernel-di-*, otherwise merging becomes totally impossible
[12:50] <fabbione> Kamion: sure, make sense
[12:50] (Kamion/#ubuntu-devel) fabbione: we should resist attempts to "simplify" it; besides which kernel-wedge is really clever and does a lot of things we want, like converting kernel module dependencies into package dependencies
[12:50] <fabbione> Kamion: yes.. i was checking kernel-wedge..
[12:51] (Kamion/#ubuntu-devel) let me dig up Herbert's e-mail from when we last discussed
[12:51] (Kamion/#ubuntu-devel) this
[12:51] <fabbione> Kamion: the "hardest" thing to do is to integrate 
[12:51] <fabbione> #!/usr/bin/make -f
[12:51] <fabbione> include /usr/share/kernel-wedge/generic-rules
[12:51] <fabbione> debian/rules (END) 
[12:51] <fabbione> this ^
[12:51] <fabbione> into debian/rules for the kernel
[12:52] <fabbione> Kamion: sure.. please fwd them to me
[12:53] <fabbione> i really need to cook some lunch
[12:53] <fabbione> later
[12:53] (Kamion/#ubuntu-devel) fabbione: there's apparently some complication with depmod
[12:53] (Kamion/#ubuntu-devel) fabbione: remember that you need a modules.dep for this to work
[12:54] <fabbione> Kamion: yup.. i read that in one of the notes
[12:55] (Kamion/#ubuntu-devel) elmo,lamont: build logs on yellow are still in super-noisy mode due to the locale the buildd is using not being configured in the chroot
[12:56] <elmo> fixed
[12:57] <elmo> (it was king last time)
[12:57] <elmo> I either have an old amd64 install cd, or this is broken in warty final
[12:58] (Kamion/#ubuntu-devel) it's the chroot
[12:58] (Kamion/#ubuntu-devel) just running debootstrap won't generate locales
[12:58] (Kamion/#ubuntu-devel) so LANG=<anything but C> will cause this, and we do generally set LANG!=C
[12:58] <elmo> you sure it's not /etc/environment in the base ?
[12:59] <elmo> 'cos that's all I did on king.. ;)
[12:59] <elmo> what I mean is, both king and yellow had an /etc/environmment after a freash warty install
[12:59] <elmo> (an empty one)
[01:00] (Kamion/#ubuntu-devel) you've got an old install CD
[01:01] (Kamion/#ubuntu-devel) I fixed that in base-config 2.44ubuntu25 on 9 October, before final
[01:01] (Kamion/#ubuntu-devel) however, it should be mostly harmless
[01:02] (Kamion/#ubuntu-devel) elmo: you need to *generate* the locale as well as just setting it; all /etc/environment does is the latter
[01:03] (Kamion/#ubuntu-devel) 'dpkg-reconfigure locales' generates them; prebaseconfig does the equivalent of that in d-i
[01:03] <elmo> no, I just want to wipe it out
[01:03] <elmo> so we're using C
[01:03] <elmo> we don't have locales installed on most of our machines
[01:04] (Kamion/#ubuntu-devel) ah; the installer won't do that for you
[01:05] (Kamion/#ubuntu-devel) changing it in /etc/environment should certainly do for just wiping out the locale, yeah
[01:05] (Kamion/#ubuntu-devel) ok, sorry if I was teaching granny to suck eggs, no idea how much you ignore locales in general :-)
[01:13] <daniels> Kamion: do you have a working dhcpd/tftpboot setup for ibooks?
[01:14] <elmo> uh - hasn't he got a CD?
[01:15] <daniels> matter of fact, he has, actually
[01:16] <elmo> I was going to say, if he doesn't, there's a bazillion lieing around the flat
[01:17] (Kamion/#ubuntu-devel) daniels: yeah, somewhere
[01:17] (thom/#ubuntu-devel) down to 2500 unread messages *sigh*
[01:17] (Kamion/#ubuntu-devel) daniels: booting from CD probably way less hassle though
[01:19] <daniels> Kamion: jane/lulu had a couple of powerpc cds, as it turned out
[01:19] <daniels> thom: how many of them were 'where's firefox 1.0'?
[01:19] <daniels> elmo: 'bazillion', 'two', close enough
[01:32] (Kamion/#ubuntu-devel) pitti: so, do you make it general consensus that we should at least drop mozilla-browser from the CD?
[01:32] <pitti> Kamion: yes
[01:33] <jdub> Kamion: *kill!*
[01:33] <pitti> Kamion: but there still seem to be too many users/usages for it
[01:33] <pitti> Kamion: so I would at least leave it in main
[01:33] (Kamion/#ubuntu-devel) so move to supported?
[01:33] <pitti> Kamion: the remaining question is what to do with the locale packages?
[01:33] <pitti> Kamion: currently they are in universe
[01:33] (bob2/#ubuntu-devel) Kamion: thanks for your help dude
[01:34] <pitti> Kamion: if we propagate them to main, our future language packs should include them
[01:34] (bob2/#ubuntu-devel) not actually sure what fucked up, but rerunning ybin worked
[01:34] <pitti> Kamion: which is kind of a waste
[01:34] (Kamion/#ubuntu-devel) pitti: dunno; I'm not wearing the mdz hat any more ;)
[01:34] <daniels> Kamion: that's a pretty big hat
[01:34] (Kamion/#ubuntu-devel) bob2: ah, ok, cool
[01:35] <pitti> Kamion: we have this problem regardless of whether moz is in shipped or supported
[01:35] <pitti> Kamion: so I don't think it's a problem to move it to supported
[01:35] (Kamion/#ubuntu-devel) indeed ...
[01:35] (Kamion/#ubuntu-devel) jdub: from your comment above can I assume you approve? :)
[01:36] <jdub> hell yeah
[01:38] <fabbione> Kamion: i will test the new cd tomorrow :-)
[01:38] <fabbione> Kamion: no problem about it ;)
[02:14] <fabbione> Kamion: do you mind to check if the current kernel-di-i386 builds properly?
[02:14] <fabbione> I am getting an error here and i would like to know if it is just local
[02:14] <fabbione> or something more than that
[02:14] (Kamion/#ubuntu-devel) fabbione: sure, let me just make a hoary-i386 chroot
[02:14] <fabbione> Kamion: thanks
[02:14] (Kamion/#ubuntu-devel) fabbione: what's the error?
[02:14] <fabbione> debian/ppp-modules-2.6.8.1-3-386-di lib/modules/2.6.8.1-3-386/kernel/lib/crc-ccitt.ko
[02:14] <fabbione> debian/nic-shared-modules-2.6.8.1-3-386-di lib/modules/2.6.8.1-3-386/kernel/lib/crc-ccitt.ko
[02:14] <fabbione> some modules are in more than one package
[02:14] <fabbione> command exited with status 255
[02:14] <fabbione> make: *** [binary-arch]  Error 2
[02:15] <fabbione> Kamion: i think the integration is much easier than we think
[02:15] <fabbione> Kamion: kernel-wedge is well designed
[02:15] (Kamion/#ubuntu-devel) fabbione: that usually means that crc-ccitt needs to be added to a common package, but I thought I fixed that particular one ages ago; I'll have a look
[02:16] <fabbione> Kamion: probably a left over from the recent merges?
[02:17] <fabbione> anyway.. i "just" need a working kernel-di-i386 that i can use to verify my changes
[02:17] (Kamion/#ubuntu-devel) fabbione: shouldn't think so, it built last time
[02:17] (Kamion/#ubuntu-devel) fabbione: I'm looking, ok :)
[02:17] <fabbione> sure
[02:17] <fabbione> no rush
[02:18] (Kamion/#ubuntu-devel) fabbione: what version are you using? the changelog says I fixed this in 0.64ubuntu6
[02:18] <fabbione> -0.67ubuntu3
[02:18] (Kamion/#ubuntu-devel) and of the kernel?
[02:18] (Kamion/#ubuntu-devel) fabbione: and of kernel-wedge?
[02:19] <fabbione> -17
[02:19] <fabbione> Version: 1.25ubuntu2
[02:19] <fabbione> in the same order as requested :-)
[02:24] <queuetue> Hi.  I'm a reasonably good python developer interested in ubuntu - is there a list of low-hanging fruit that needs to be worked on so I could get my toes wet and learn about the ubuntu development process?
[02:24] (Kamion/#ubuntu-devel) fabbione: that's weird; I can see why it happens, but it didn't happen for me nor for the buildd; investigating ...
[02:25] <fabbione> Kamion: just tell me if there is anything i can test here
[02:25] <fabbione> Kamion: only one note: i am not building in a chroot and i have other kernels installed
[02:25] <fabbione> (if that's relevant)
[02:27] <sabdfl> queuetue: do you want to become a maintainer, or a contributor on docs and translations?
[02:27] (Kamion/#ubuntu-devel) queuetue: I'm not sure we have particularly good lists of low-hanging fruit at the moment, unfortunately, other than trawling bugzilla
[02:27] (Kamion/#ubuntu-devel) queuetue: is there something in particular you're interested in?
[02:29] <queuetue> Kamion, I'm not picky, but I happen to have a lot of interest in getting jack and professional audio recording/editing easy to use on a linux workstation atm...
[02:30] <queuetue> sabdfl, I could do the first two, if someone could walk me through the process - translation, probably not - I'm English only. :)
[02:30] <sabdfl> queuetue: are you a maintainer in any other distro, or upstream on any open source projects?
[02:31] <jdub> seb128: dude
[02:31] <jdub> seb128: duuuuude
[02:31] <jdub> seb128: nautilus 2.9.1 :-)
[02:31] <queuetue> sabdfl, No, I've contributed over the years to mozilla, python, and the kernel (via kernel-janitors) but nothing large or serious.
[02:31] <seb128> jdub: I know, I've packaged the CVS this morning and checked with alex that all was ok :)
[02:32] (Mithrandir/#ubuntu-devel) queuetue: you could always look at the bounty list.
[02:32] <jdub> seb128: that's going to make hoary fun :)
[02:33] <sabdfl> what's rocking in the new nautilus?
[02:33] <jdub> sabdfl: boNObo -> gone :-)
[02:33] <sabdfl> queuetue: ok, to become a maintainer you basiaclly need
[02:33] <sabdfl> to demonstrate your ability to package upstream code
[02:34] <sabdfl> pick a piece of software you have a real interest in improving the package of, or which has not yet been pacakged for ubuntu
[02:34] <sabdfl> learn about the packaging system, and produce a package
[02:34] <sabdfl> get feedback from the existing maintainers
[02:34] <seb128> jdub: not totally :)
[02:35] <seb128> but yeah, no view anymore, less crappy code :)
[02:35] <jdub> seb128: it's so close :-)
[02:35] <sabdfl> queuetue: then, when it's good enough, it goes in and you get to upload packages to universe
[02:35] <sabdfl> after a while you may also get to upload to main
[02:40] <elmo> ?! how has a random console based app I alien'ed ended up linked to libarts...
[02:41] <lifeless> hardlinked ?
[02:41] <lifeless> or soft?
[02:41] <pitti> Mithrandir: FYI, I just modified sysklogd to run syslogd as normal user "syslog" instead of root
[02:41] <pitti> Mithrandir: do you see any problem with this?
[02:41] <pitti> Mithrandir: it opens the log files as root and then drops to user "syslog"
[02:41] (bob2/#ubuntu-devel) lifeless: unix filesystems have these things that are kinda like "short cuts" on windows
[02:43] <elmo> lifeless: no, as in, 'when you run ldd, it shows the binary is linked to the libarts library'
[02:44] <azeem> elmo: perhaps it was part of some KDE project, and the build process involved linking against libarts for all binaries without further thinking?
[02:44] <elmo> aiee, OO.o has some sort of word-completion on-by-default?   how satanic
[02:44] (Mithrandir/#ubuntu-devel) pitti: who owns /var/log/syslog?
[02:44] (thom/#ubuntu-devel) OOo is satanic
[02:44] <elmo> azeem: it's a RAID monitor/configuration program :P
[02:44] <pitti> Mithrandir: root
[02:44] (bob2/#ubuntu-devel) 
[02:44] (bob2/#ubuntu-devel) haha
[02:44] (Mithrandir/#ubuntu-devel) pitti: note that sysklogd rotates its own logs.
[02:44] <pitti> Mithrandir: that's the default and I don't want to chage it
[02:44] (Mithrandir/#ubuntu-devel) thom: no shit.
[02:44] <azeem> elmo: with a qt-frontend? ;)
[02:44] <pitti> Mithrandir: oh, since when?
[02:44] (Mithrandir/#ubuntu-devel) pitti: since forever.?
[02:45] <pitti> Mithrandir: I always had to add a stanza to /etc/logrotate.conf
[02:45] <elmo> and CLIPPY!!
[02:45] <elmo> RUN AWAY
[02:45] <lifeless> elmo: garh
[02:45] (Mithrandir/#ubuntu-devel) : tfheen@vawad ~ > grep messages /etc/logrotate.d/*
[02:45] (Mithrandir/#ubuntu-devel) : tfheen@vawad ~ >
[02:45] (Mithrandir/#ubuntu-devel) pitti: and I have rotated messages in /var/log
[02:45] <pitti> Mithrandir: my logs were never rotated on my box
[02:46] <pitti> Mithrandir: in logrotate.conf?
[02:46] (Mithrandir/#ubuntu-devel) pitti: I don't have _anything_ touching files such as /var/log/syslog and /var/log/messages in /etc on my unstable nor my warty box.
[02:47] <pitti> Mithrandir: /etc/cron.daily/sysklogd 
[02:47] <pitti> Mithrandir: that's doing it
[02:48] (Mithrandir/#ubuntu-devel) pitti: ah, true.
[02:48] <pitti> Mithrandir: it calls "/etc/init.d/sysklogd reload-or-restart "
[02:48] <pitti> Mithrandir: I modified sysklogd's init script to really restart the daemon
[02:48] <pitti> Mithrandir: so this works
[02:48] <pitti> Mithrandir: I already checked that before
[02:48] (Mithrandir/#ubuntu-devel) pitti: I thought it did it itself, but that ought to work, yes.
[02:48] (Mithrandir/#ubuntu-devel) fabbione: oh, why?
[02:48] (bob2/#ubuntu-devel) fabbione: can I have your 4 monitors?
[02:48] <pitti> Mithrandir: I was more concerned about the fact that I write root files as normal suer
[02:48] <fabbione> daniels: i swear... .. it installed by itself..
[02:48] <pitti> Mithrandir: I don't see a problem with it right now, but I wanted to have a second opinion
[02:48] (Mithrandir/#ubuntu-devel) pitti: it drops priviledges after it has opened the files?
[02:48] <fabbione> daniels: AHHH XPRT ON MY DESKTOP
[02:49] <pitti> Mithrandir: yes, before it does not work :-)
[02:49] <fabbione> bob2: forget it
[02:49] <fabbione> ;)
[02:49] <pitti> Mithrandir: unless I change the postinst to do some chown magic, but I don't want that
[02:49] (Mithrandir/#ubuntu-devel) pitti: hmm.
[02:49] (bob2/#ubuntu-devel) hah
[02:49] (Mithrandir/#ubuntu-devel) pitti: have you talked with Joey about it?  I don't see any big problem with it per se, at least.
[02:49] <pitti> Mithrandir: theoretically the syslogd user should be as hard to break as the root user
[02:50] <pitti> Mithrandir: not yet
[02:50] <pitti> Mithrandir: I will send him a patch, but Iwanted to have something working 
[02:50] (Mithrandir/#ubuntu-devel) pitti: I think it's a good idea
[02:50] <pitti> Mithrandir: btw, if we are at it
[02:50] <pitti> Mithrandir: I also tried to run klogd as nonroot
[02:50] (Mithrandir/#ubuntu-devel) pitti: but it won't be able to reopen the files when they are rotated, so you have to restart it.
[02:50] <pitti> Mithrandir: yes, that's why I modified the init script
[02:51] <pitti> Mithrandir: I also documented that in the manpage
[02:51] (Mithrandir/#ubuntu-devel) ok
[02:51] <pitti> Mithrandir: I tried to drop privs in klogd, too
[02:51] <daniels> fabbione: it's better than discovering someone INSTALLED XPRINT ON YOUR SERVER
[02:51] (Mithrandir/#ubuntu-devel) but it blew up?
[02:51] <pitti> Mithrandir: it opens /proc/kmsg as root and then drops to user
[02:51] <pitti> Mithrandir: but then I get a permission denied if it reads from the file descriptor
[02:51] (Mithrandir/#ubuntu-devel) pitti: sounds like a bug
[02:52] <pitti> Mithrandir: this works with normal files, but obviously not with files in /proc
[02:52] <pitti> Mithrandir: I thought that, too
[02:52] <pitti> Mithrandir: but it's hard to circumvent
[02:52] (Mithrandir/#ubuntu-devel) pitti: fix the kernel? :)
[02:52] <pitti> Mithrandir: doesn't work for Debian, where half of the users run their own kernels
[02:52] (Mithrandir/#ubuntu-devel) get the fix upstream, then.
[02:52] <pitti> Mithrandir: and I don't want to have it break on custom kernels, too
[02:53] <pitti> Mithrandir: still it will take a while until everybody has the fix
[02:53] (Mithrandir/#ubuntu-devel) pitti: true enough
[02:53] <elmo> seriously, does anyone know how to disable OO.o's auto-word-complete insanity?  it's driving me Krazy
[02:55] <azeem> elmo: please use #ubuntu for general support questions :p
[02:55] <daniels> elmo: hm, I got duplicate ACCEPTED notifications for xcompmgr
[02:55] <pitti> Mithrandir: thanks!
[02:57] <elmo> To: Daniel Stone <daniel.stone@canonical.com>,
[02:57] <elmo>         Daniel Stone <daniels@debian.org>
[02:57] <seb128> this is annoying, why do we get 2 mails ?
[02:58] <azeem> wrong use of dpkg-genchanges I guess
[02:58] <seb128> I don't care to get a mail on my debian email for an hoary upload
[02:58] (Kamion/#ubuntu-devel) Maintainer: != Changed-By:
[02:58] (Kamion/#ubuntu-devel) if the Maintainer: field is legal it'll get a mail
[02:58] (Kamion/#ubuntu-devel) legal i.e. known to Ubuntu's katie
[02:58] <seb128> ok, so the right way is to use my @debian.org for the uploads ? :)
[02:59] (Kamion/#ubuntu-devel) either that or don't ever use your @debian.org address for hoary uploads at all and get it taken out of katie's allowed list
[02:59] <daniels> elmo: nevermind me
[02:59] (Kamion/#ubuntu-devel) which is what most of us do
[02:59] <elmo> seb128: I can take your debian.org address out of the whitelist if you want
[02:59] <daniels> Kamion: hm, I was advised to set Maintainer: daniels@d.o and C-B: d.s@c.c
[02:59] <daniels> elmo: if you could take daniels@d.o out, that would be phat
[03:00] (Kamion/#ubuntu-devel) daniels: I meant wrt the allowed list really
[03:00] <daniels> Kamion: oh, right
[03:00] (Kamion/#ubuntu-devel) fabbione: ok, something very evil is going on in kernel-wedge here, may take me a while to untangle it
[03:00] <azeem> daniels: set Maintainer to daniels@d.o in .changes, or in debian/control?
[03:00] <seb128> elmo: hum, no I prefer to keep it in fact, but thanks. I'll do a mail filter for these rather :)
[03:01] <daniels> azeem: control
[03:01] <fabbione> Kamion: ah.. ok.. do you have a simple workaround i can use in the meanwhile?
[03:01] <azeem> daniels: it was the Maintainer in .changes which triggered the double mail I guess, you can override that
[03:01] <daniels> azeem: yeah, but ugh
[03:02] (Kamion/#ubuntu-devel) fabbione: no, I don't understand it yet, sorry
[03:03] <azeem> daniels: dpkg-buildpackage reads $DEBEMAIL, so you could set that to @canonical.com perhaps
[03:03] <fabbione> Kamion: ok no problem.
[03:03] (Kamion/#ubuntu-devel) fabbione: I've come up with a way to reproduce your problem, and it seems to be a bug in kernel-wedge that I couldn't reproduce it; although I completely don't understand why it fails for you with the current kernel-wedge code
[03:03] <fabbione> Kamion: do you want access to my box?
[03:04] <fabbione> Kamion: it's just my devel workstation, if you want to play just tell me
[03:05] (Kamion/#ubuntu-devel) nah, it's ok, I'll be able to fix the bug without that
[03:06] <fabbione> ok
[03:09] (Kamion/#ubuntu-devel) fabbione: oooohhh!
[03:09] (Kamion/#ubuntu-devel) fabbione: it's locale-dependent; try LC_COLLATE=C
[03:09] <fabbione> Kamion: AH
[03:10] <fabbione> Kamion: BINGO!
[03:11] (Kamion/#ubuntu-devel) [cjwatson@cairhien /tmp] $ sort test
[03:11] (Kamion/#ubuntu-devel) -b
[03:11] (Kamion/#ubuntu-devel) a
[03:11] (Kamion/#ubuntu-devel) c
[03:11] (Kamion/#ubuntu-devel) [cjwatson@cairhien /tmp] $ LC_ALL=it_IT.UTF-8 sort test
[03:11] (Kamion/#ubuntu-devel) a
[03:11] (Kamion/#ubuntu-devel) -b
[03:11] (Kamion/#ubuntu-devel) c
[03:11] <fabbione> <- en_DK.UTF8
[03:11] (Kamion/#ubuntu-devel) that's basically the root cause why you see the problem and I don't; the problem still exists though
[03:11] <fabbione> ;)
[03:11] <fabbione> yeah i agree
[03:12] <fabbione> i can live with C for testing.
[03:14] <zul> hey
[03:23] (Kamion/#ubuntu-devel) fabbione: kernel-wedge 1.25ubuntu3 uploading, fixes both the reason I didn't see this and the dependency bug itself
[03:23] <fabbione> Kamion: ROCKING!
[03:26] <fabbione> Kamion: i was thinking that for the first merge i will create a dri struct like this to keep merging simple
[03:26] <fabbione> Kamion: linux-source-<ver>/debian/d-i/arch/
[03:26] (Kamion/#ubuntu-devel) dri?
[03:26] <fabbione> s/dri/dir
[03:26] (Kamion/#ubuntu-devel) ah
[03:27] <fabbione> and create symlinks to the proper files according to the arch
[03:27] (Kamion/#ubuntu-devel) if you could basically just copy each linux-kernel-di-* package except for the debian/ directory in there, that would be ideal
[03:27] <fabbione> Kamion: you can't :(
[03:27] (Kamion/#ubuntu-devel) why not?
[03:27] <fabbione> there is the package-list that is clashing :(
[03:27] <fabbione> otherwise i need to move it somewhere else
[03:28] <fabbione> or rename it like: package-list.<arch>
[03:28] (Kamion/#ubuntu-devel) not if you have per-arch directories
[03:28] (Kamion/#ubuntu-devel) package-list does have per-arch support, but the merging would get complicated
[03:28] <fabbione> exactly
[03:28] (Kamion/#ubuntu-devel) I thought you were suggesting just having i386/package-list, i386/kernel-versions, i386/modules/, etc., for each arch
[03:29] <fabbione> my point was to copy linux-kernel-di-/* in that debian/di/<arch> di
[03:29] <fabbione> dir
[03:29] (Kamion/#ubuntu-devel) right
[03:29] <fabbione> ywah
[03:29] (Kamion/#ubuntu-devel) if you build it after the linux-image-* packages have been constructed, it should be easy
[03:29] (Kamion/#ubuntu-devel) set SOURCEDIR to the linux-image-* temporary tree
[03:30] <fabbione> yup
[03:30] <fabbione> exactly what i was thinking about
[03:30] (Kamion/#ubuntu-devel) but depmod has to be run, and that's what Herbert was talking about
[03:30] (Kamion/#ubuntu-devel) kernel-wedge needs the modules.dep file
[03:30] <fabbione> yes i read the mails
[03:30] <fabbione> that's easy
[03:31] (Kamion/#ubuntu-devel) in fact I notice that there's already a modules.dep file in the linux-image package
[03:31] <fabbione> correct
[03:31] <fabbione> but it is simple...
[03:31] <fabbione> you just call depmod with an arg or 2
[03:31] <fabbione> i did it before.. i just have to RTFM again to remember how to do it
[03:32] (Kamion/#ubuntu-devel) um
[03:32] (Kamion/#ubuntu-devel) the modules.dep file is already in the package, therefore it must be generated as part of the package build process
[03:32] <fabbione> it's like directory and kernel version
[03:32] (Kamion/#ubuntu-devel) unless I'm mistaken you shouldn't have to do anything
[03:32] <fabbione> uh could be yes
[03:32] <fabbione> s/could/should
[03:33] <fabbione> so is it ok for you if i use a debian/di subdir?
[03:33] <fabbione> i don't want to clutter the toplevel since debian lives in the kernel tree already
[03:34] (Kamion/#ubuntu-devel) yeah, fine by me, make it d-i rather than di I think, or maybe "udebs"
[03:34] <fabbione> sure
[03:34] <fabbione> d-i sounds good to me
[03:34] <fabbione> ok i am off for today
[03:34] <fabbione> cya later guys
[03:34] (Kamion/#ubuntu-devel) cool
[03:35] (Kamion/#ubuntu-devel) heh, I'm just off for lunch
[03:35] (Kamion/#ubuntu-devel) bye
[03:37] <daniels> seeya fabbione
[03:39] <pitti> smurfix: I'm currently discussing the cscope patch with Astharot 
[03:39] <pitti> smurfix: I have some other objections
[03:40] <pitti> smurfix: wait for his 3rd patch :-)
[03:43] <elmo> pitti: btw, <hint type=v. subtle>ntpd still runs as root</> ;)
[03:44] <pitti> elmo: oh, nice idea
[03:44] <pitti> elmo: I'll put that onto my todo list
[03:44] (Mithrandir/#ubuntu-devel) elmo: uhm, is that possible to change?
[03:44] <pitti> Mithrandir: why it shouldn't?
[03:44] (Mithrandir/#ubuntu-devel) pitti: it needs to be able to call adjtimex(2)
[03:44] <pitti> Mithrandir:        CAP_SYS_TIME
[03:44] <pitti> Mithrandir: should work
[03:45] (Mithrandir/#ubuntu-devel)        EPERM  buf.mode is non-zero and the user is not super-user.
[03:45] <pitti> "Allow   modification  of  system  clock(3,n)  (settimeofday(2),  adj-
[03:45] <pitti>               timex(2)); allow modification of real-time (hardware) clock"
[03:45] (Mithrandir/#ubuntu-devel) not according to the man page for adjtimex.
[03:45] <pitti> capabilities(7)
[03:45] (Mithrandir/#ubuntu-devel) but it might
[03:45] <pitti> Mithrandir: yes, all manpages only talk about root, not capabilities
[03:45] <pitti> Mithrandir: but that's an error of the manpages
[03:45] (Mithrandir/#ubuntu-devel) file bugs!
[03:45] (Mithrandir/#ubuntu-devel) :)
[03:46] <pitti> Mithrandir: I'm currently trying to depriv klogd
[03:46] <pitti> Mithrandir: I play around with various capabilities
[03:47] <elmo> Mithrandir: RH has a patch for it - I believe it's even in current Debian source, it just doesn't run as a non-root user for some reason
[03:47] <pitti> even nicer
[03:48] (Mithrandir/#ubuntu-devel) pitti: ok :)
[03:49] <pitti> Mithrandir: next one is X.org :-))
[03:49] <daniels> elmo: are our kernels NX?
[03:49] <pitti> Mithrandir: no, not seriously
[03:49] <daniels> pitti: HA HA HA
[03:49] <elmo> daniels: no idea
[03:49] <pitti> my ps aux also shows gpm, but that's not supported
[03:49] <daniels> elmo: 'kay
[03:49] (Mithrandir/#ubuntu-devel) pitti: you. are. on. crack. :)
[03:49] <daniels> does anyone know if our kernels are NX?
[03:49] <pitti> Mithrandir: with the root slash axe in my hands, I am :-)
[03:49] <daniels> Mithrandir: there's no WRITE_RANDOM_SHIT_INTO_DEV_MEM capability, sadly
[03:50] (Mithrandir/#ubuntu-devel) daniels: NX as in nonexecutable?
[03:50] (Mithrandir/#ubuntu-devel) daniels: it's so wrong that X servers do that.
[03:52] <elmo> if it still needs ingo's patch, then it won't be in 
[03:53] <elmo> I'm not sure if mainline does, esp. in 2.6.8.1 era
[03:53] (Kamion/#ubuntu-devel) Mithrandir: not as bad as the READ_RANDOM_SHIT_FROM_DEV_MEM that they used to do in order to get "randomness"
[03:53] <daniels> Mithrandir: non-executable
[03:54] <daniels> Mithrandir: yeah, writing to the video card is for suckahs :P
[03:55] (Mithrandir/#ubuntu-devel) Kamion: if they happen to hit the area covered by /dev/random, it shouldn't be too bad. :P
[03:56] <smurfix> pitti: the last version goes into an endless loop if TMPDIR is too long, not to mention that the buffer isn't null terminated.
[03:57] <pitti> smurfix: argh, you are right
[03:59] <pitti> Astharot: I'd recommend checking strlen(tempname)
[03:59] <Astharot> okay
[03:59] <pitti> Astharot: before strcpy'ing 
[04:05] <pitti> Mithrandir: argh
[04:05] <pitti> Mithrandir: reading /proc/kmsg requires CAP_SYS_ADMIN
[04:05] <pitti> Mithrandir: WTF invented this???
[04:07] (Mithrandir/#ubuntu-devel) er, that's just Wrong.
[04:07] (Mithrandir/#ubuntu-devel) CAP_SYS_ADMIN is ~root, isn't it?
[04:08] <pitti> Mithrandir: nearly as good as root, yes
[04:08] <pitti> Mithrandir: you can boot, mount, change hostnames, and so on
[04:08] (Mithrandir/#ubuntu-devel) yeah
[04:08] (Mithrandir/#ubuntu-devel) reading from kmsg does empty the buffer, though
[04:08] <pitti> Mithrandir: I tried with less crucial caps, but it doesn't work
[04:09] <pitti> Mithrandir: it should be something like CAP_DAC_READ_SEARCH
[04:09] <pitti> Mithrandir: oh sorry, SYS_ADMIN does not allow to boot
[04:09] <pitti> Mithrandir: but mounting devices is bad enough
[04:09] <pitti> Mithrandir: I'm not sure whether this is worth the effort then
[04:10] (Mithrandir/#ubuntu-devel) sounds like an oversight in the kernel
[04:10] <pitti> Mithrandir: actually there should be an extra capabilitiy for reading kernel messages
[04:10] <pitti> Mithrandir: it doesn't really fit in any existing one
[04:11] (Mithrandir/#ubuntu-devel) probably why it was added to SYS_ADMIN, then
[04:12] <pitti> Mithrandir: it's still slightly better than root
[04:12] <pitti> Mithrandir: but I think you can get root privileges with mount, too
[04:12] <pitti> Mithrandir: one could loopback mount a new /etc and thus overwrite shadow or so
[04:13] (Kamion/#ubuntu-devel) far easier than that
[04:13] (Kamion/#ubuntu-devel) make a filesystem with a setuid copy of /bin/sh on it, loopback-mount that
[04:13] <pitti> Kamion: yes, right
[04:13] <pitti> so I suppose I just leave it as root then
[04:13] (Mithrandir/#ubuntu-devel) pitti: it'll stop script kiddies, but nobody who knows their unix.
[04:13] <pitti> not worth the fuss
[04:27] (bob2/#ubuntu-devel) speaking of which, are we getting a pimp new ppc glibc in hoary?
[04:31] <elmo> bob2: no
[04:33] (Kamion/#ubuntu-devel) there's a new glibc in experimental isn't there?
[04:33] (Kamion/#ubuntu-devel) at least, gotom has been talking about doing that
[04:33] (Kamion/#ubuntu-devel) hm, not uploaded yet
[04:33] <azeem> the glibc guys are good at talking :)
[04:34] <daniels> w
[04:34] (Kamion/#ubuntu-devel) azeem: that's not really fair to them
[04:34] <azeem> heh, just joking
[04:34] (Kamion/#ubuntu-devel) ok :)
[04:34] <azeem> oh, jbailey is not around, so the pun was not recieved anyway
[04:35] <[Clint] > they're just on a different timescale than some people
[04:35] (Kamion/#ubuntu-devel) [Clint] : it's not their fault, it's the release team being evil
[04:35] (Kamion/#ubuntu-devel) we told them in no uncertain terms to stop pulling from CVS ages ago
[04:36] (Kamion/#ubuntu-devel) (for sarge)
[04:36] <[Clint] > they're still on a different timescale
[04:36] <azeem> yeah, but they could upload to experimental, no?
[04:36] (Kamion/#ubuntu-devel) azeem: gets difficult as soon as shlibdeps change
[04:36] <[Clint] > there are other people on that timescale too
[04:36] (bob2/#ubuntu-devel) but do they want to maintain 3 branches of glibc?
[04:37] (Kamion/#ubuntu-devel) azeem: anyone installing packages from experimental would have to install experimental glibc too, and before you know it we'd end up effectively trying to support it
[04:37] <azeem> bob2: woody isn't really a branch
[04:37] (bob2/#ubuntu-devel) hm, s'pose so
[04:37] <azeem> Kamion: the trick is to tell sbuild/whatever to only pull from experimental if the Build-Depends explicitely require a version from there I guess
[04:38] <pitti> smurfix: Astharot just sent another patch
[04:38] <pitti> smurfix: since I proposed much of this code, would you mind taking a look at it?
[04:38] <azeem> in any case, did you guys ask the glibc maintainers whether they would perhaps start working on glibc now, get it into hoary first and then into sid once sarge is done?
[04:39] (lamont/#ubuntu-devel) daniels: this channel's proabbly better...
[04:39] (lamont/#ubuntu-devel) and yes, I can get you the build tree after
[04:39] (lamont/#ubuntu-devel) do you want it even for success?
[04:39] <daniels> lamont: i'm expecting failure, tbh
[04:39] (Kamion/#ubuntu-devel) azeem: well, they already have started working on it; there's stuff in svn and I see a lot of stuff in my #debian-glibc scrollback
[04:40] <azeem> ah, cool
[04:40] <azeem> then bribe them to work faster :)
[04:40] (tseng/#ubuntu-devel) ergh
[04:40] (lamont/#ubuntu-devel) daniels: given my email this morning, I understand fully. :-)
[04:40] (tseng/#ubuntu-devel) does anyone know where to find the debian cramfs-initrd patch broken out?
[04:40] (bob2/#ubuntu-devel) I'm pretty sure it's upstream in 2.6
[04:41] (tseng/#ubuntu-devel) ah, thatll do
[04:42] <daniels> lamont: heh :)
[04:43] (lamont/#ubuntu-devel) daniels: btw, I see that you already fixed the other 2, so I won't mention them. :-)
[04:43] (lamont/#ubuntu-devel) anyway, lob me a pointer to source, and I'll throw it against the wall and see if it sticks
[04:45] <daniels> lamont: just building it now for you
[04:45] (lamont/#ubuntu-devel) cool
[04:50] <daniels> thom: if the screenshot is no use, please reassign to xserver-xfree86+d.s@c.c and ask for lspci output and /var/log/XFree86.0.log
[04:52] (thom/#ubuntu-devel) righto
[04:52] (lamont/#ubuntu-devel) daniels: eta?  just trying to figure out when I should bounce back in front of my computer from the grand install fest in the house...
[04:54] <smurfix> pitti: I'd dr
[04:54] <smurfix> oops
[04:54] <pitti> smurfix: i'd only change the permission to 0600
[04:54] <pitti> smurfix: otherwise it looks good now
[04:55] <daniels> thom: (by 'no use', I mean, shows no corruption -> it's a rendering bug)
[04:55] <smurfix> pitti: OK. I'd use strcpy now, since the length has been tested -- not much point in filling the whole buffer with zeroes, but that's just my preference.
[04:55] <daniels> lamont: on the way now
[04:57] (thom/#ubuntu-devel) daniels: you're getting this bug, don't worry ;-)
[04:58] <daniels> thom: did I say 'my shout tonight'?  whoops ...
[05:00] <smurfix> pitti: assuming he has actually tested the patch, unlike his first one :-) looks OK otherwise.
[05:01] <jdub> lifeless: ping
[05:04] (Kamion/#ubuntu-devel) ok, ifrename stuff is definitely not working properly in hoary
[05:12] (Kamion/#ubuntu-devel) Warning: Interface name is `eth0' at line 2, can't be mapped reliably.
[05:12] (Kamion/#ubuntu-devel) Error: cannot change name of eth1 to eth0: File exists
[05:15] <smurfix> Kamion: well, if eth0 exists, that sortof makes sense...
[05:15] (Kamion/#ubuntu-devel) smurfix: the real question's why eth0 hasn't already been renamed to something else
[05:16] (Kamion/#ubuntu-devel) which I think is because it didn't get detected by the installer for some reason, or netcfg just couldn't be arsed to write out the iftab line, or something
[05:17] (Kamion/#ubuntu-devel) I'd also like to know why events sometimes seem to get lost between hotplug and udev :(
[05:17] (bob2/#ubuntu-devel) is hal meant to be tcp-reliable?
[05:19] <daniels> bob2: ... ?
[05:20] <daniels> bob2: <hal> ah rad, a new bit of hardware! <hal> dbus: yo <dbus> hal: sup <hal> dbus: new mouse, yo <dbus> ALRIGHT SUCKERS, THERE'S A NEW MOUSE <hal> dbus: kthxbye
[05:21] (bob2/#ubuntu-devel) you should write a book
[05:21] <daniels> about what?  the fd.o platform?
[05:22] (bob2/#ubuntu-devel) hah
[05:22] (bob2/#ubuntu-devel) speaking of which...
[05:23] (Mithrandir/#ubuntu-devel) daniels: I think we should have an IRC client interface to DBUS.
[05:23] (Mithrandir/#ubuntu-devel) DBUS-over-IRC.
[05:24] <daniels> wouldn't be terribly difficult
[05:24] (Mithrandir/#ubuntu-devel) with aussie dialect?
[05:24] <ironwolf> daniels: it worked, thanks.
[05:24] <daniels> 'yo' and 'sup' are not aussie dialect :P
[05:25] <daniels> ironwolf: rockin :)
[05:25] <daniels> ironwolf: we'll fix it for hoary, thanks for the debugging help
[05:25] (thom/#ubuntu-devel) writing unit tests considered boring
[05:26] <rburton> thom: unit tests are good for your karma, or do you want to come back as a winxp administrator?
[05:26] (Mithrandir/#ubuntu-devel) thom: you just failed your unit test.
[05:26] (thom/#ubuntu-devel) just because they're "cool" doesn't make writing tests for correct md5 generation fun :P
[05:27] <daniels> rburton: shouldn't you have someone pick up the book, hand it to someone who does the backswing, and then another independent person to take care of the foreswing, with unit-tested interfaces between each?
[05:27] (thom/#ubuntu-devel) Mithrandir: as long as that's not the same as SteveA failing his, that's fine
[05:27] <rburton> daniels: yes
[05:27] (Mithrandir/#ubuntu-devel) thom: do I want you to elaborate?
[05:27] <rburton> daniels: with a mock thom object to confirm that the force of the slap was sufficient
[05:27] (Kamion/#ubuntu-devel) thom: in the SHOWER
[05:27] <daniels> rburton: remind me to bring a very heavy book next week ;)
[05:27] (lamont/#ubuntu-devel) daniels: where?
[05:27] <daniels> rburton: why a mock one?
[05:28] <rburton> daniels: just for checking. then replace with real thom in the real world
[05:28] (thom/#ubuntu-devel) Mithrandir: almost certainly not
[05:29] <ironwolf> daniels: Where do I suggest changing maxtaptime to 130 as default in xorg.conf for synaptic pads?  bugzilla? or just bonk you? :)
[05:29] (thom/#ubuntu-devel) no bonking on channel, please
[05:30] <daniels> lamont: p.u.c/~daniels/xorg
[05:30] <daniels> lamont: but!! remove debian/patches/000_stolen_from_fedora.diff from the source package
[05:30] <daniels> lamont: rm != baz rm
[05:31] <daniels> ironwolf: hmm.  any negative side-effects this could have?
[05:31] <RubenV> for those interested
[05:31] <daniels> ironwolf: i assume this is for people who can't/don't complete a tap within 100ms(?)
[05:31] <RubenV> i've just build cvs packages for beep-media-player
[05:31] <RubenV> http://files.lambda1.be/linux/
[05:32] <ironwolf> daniels: well, it works now, didn't really work well before.  It made it so that faster/*normal for me* taps actually were acknowledged.
[05:32] <ironwolf> daniels: so no, no bad effects.
[05:32] <daniels> ironwolf: sorry?  i would presume it meant taps could last long
[05:33] <daniels> ironwolf: e.g. if you tapped for 120ms, it would work now, but not before
[05:33] (lamont/#ubuntu-devel) daniels: and you want the build dir even for success, yes?
[05:33] (lamont/#ubuntu-devel) build running.
[05:33] <ironwolf> daniels: appears to be opposite. *no clue why*
[05:38] <daniels> ironwolf: bong
[05:38] <ironwolf> daniels: bong?
[05:39] <daniels> ironwolf: 'that's weird'
[05:39] <ironwolf> daniels: tell me about it... :)
[05:39] <ironwolf> daniels: that's why I asked.
[05:42] <eruin> smells like hash
[05:43] <lifeless> jdub: pong
[05:43] <lifeless> jdub: shiny gnome syncs
[05:51] <azeem> " A quiet revolution is taking place on the home desktop market and it's called MEPIS Linux."
[05:51] <azeem> dudes, you're out of the game
[05:51] (bob2/#ubuntu-devel) yeah, dunno why we wasted so much time
[05:51] (bob2/#ubuntu-devel) or why LWN is running press releases
[05:55] (Kamion/#ubuntu-devel) gnnnnnnnnn
[05:55] (Kamion/#ubuntu-devel) why does init=/bin/sh give me a blank screen, eh?
[05:56] (Kamion/#ubuntu-devel) ah, vga=771 seems to be causing issues; yet without it I get a screen too big for my display :(
[05:57] (Kamion/#ubuntu-devel) I hate Via and all its spawn
[05:59] <eruin> http://www.mepis.org/node/view/1735
[05:59] <eruin> take a look at that menu
[06:00] <eruin> reminds me of some crude mix of win95alpha and a modern linux desktop
[06:01] <mdz> thom: welcome back
[06:01] (bob2/#ubuntu-devel) ubuntu's menu just can't compete with that
[06:01] <eruin> no
[06:01] <mdz> morning, folks
[06:01] <robtaylor> eruin: oh my god
[06:02] <eruin> hehe yeah
[06:03] <eruin> pretty horrible innit
[06:03] <mvo__> good morning mdz 
[06:03] <chrisa> They made KDE look worse
[06:03] <robtaylor> that takes some skill ;)
[06:04] <eruin> yeah, hehe
[06:06] <daniels> bashing MEPIS considered offtopic
[06:06] (thom/#ubuntu-devel) mdz: thanks :-)
[06:06] <daniels> mdz: 'morning boss
[06:06] (thom/#ubuntu-devel) mdz: i tried to reply to your text, but it seems it got et
[06:09] <mdz> thom: oh, you received that? interesting
[06:09] <mdz> thom: I tried to call as well, but it went straight to voicemail
[06:10] (thom/#ubuntu-devel) yeah, i got your voice mail when i landed
[06:10] <mdz> but you got the text stateside?
[06:10] (thom/#ubuntu-devel) yup
[06:10] (thom/#ubuntu-devel) on the train
[06:11] <mdz> mvo__: your underscores are growing :-)
[06:11] <mdz> mvo__: is there a good way for upgrade-notifier to take advantage of APT::Periodic::Download-Upgradeable-Packages ?
[06:11] <mvo_> :)
[06:11] (thom/#ubuntu-devel) two hours of fun on the public "transport" system
[06:12] <mvo_> mdz: sure! I'll add a option for this to the coming preferences for it
[06:13] <pitti> Hi mdz!
[06:13] <mdz> mvo_: I guess it will need to prompt for the root password to modify the config
[06:13] <mdz> thom: which one?
[06:13] <mvo_> mdz: yes, that's going to be needed
[06:14] (thom/#ubuntu-devel) mdz: hollywood/western -> LAX
[06:14] <mdz> thom: oh god
[06:14] (thom/#ubuntu-devel) yeah
[06:14] <mdz> I did that once
[06:14] <mdz> red -> blue -> green or whatever
[06:14] (thom/#ubuntu-devel) "oh god" covers that pretty well
[06:14] (thom/#ubuntu-devel) yeha
[06:14] <mdz> I live near the universal city station
[06:14] (thom/#ubuntu-devel) and then a free bus
[06:14] (thom/#ubuntu-devel) oh, right
[06:15] (thom/#ubuntu-devel) i was trying to work out where studio city was :-)
[06:15] <mdz> just about 2 hours as I recall
[06:15] <mdz> and a 20-minute wait at one of the transfers
[06:15] <mdz> LA transit does have the benefits of the "honor system", though
[06:18] (thom/#ubuntu-devel) nod
[06:18] (thom/#ubuntu-devel) and then the fun of the Trillions Standing Around at LAX
[06:19] (bob2/#ubuntu-devel) 1718 is pretty close to beer o'clock in london
[06:19] <daniels> honor system?
[06:19] (bob2/#ubuntu-devel) aiui
[06:19] <daniels> thom: not to mention the numerous GWB portraits
[06:19] <mdz> daniels: you just walk on
[06:20] <mdz> and they randomly ask to see tickets
[06:20] (thom/#ubuntu-devel) yeah, i didn't get asked in 4 days
[06:21] (thom/#ubuntu-devel) the bus to santa monica was interesting, too
[06:21] <daniels> mdz: ah, rad
[06:21] <daniels> mdz: if you get enough inspectors, it works better than barriers or whatever
[06:23] (Kamion/#ubuntu-devel) oops, incredibly stupid hw-detect mistake by me ...
[06:24] (Kamion/#ubuntu-devel) set variable $HOTPLUG_TYPE, test $hotplug_type
[06:24] <daniels> mmm, i've done that a few times
[06:24] (thom/#ubuntu-devel) bob2: where is best for you guys?
[06:25] <mdz> daniels: where are you, anyway?  London?
[06:25] (bob2/#ubuntu-devel) everyone who's anyone is in london
[06:25] (thom/#ubuntu-devel) yeah, they're lowering house prices
[06:26] (Kamion/#ubuntu-devel) the best people are a few dozen miles north
[06:26] (bob2/#ubuntu-devel) thom: erm, dunno
[06:26] (bob2/#ubuntu-devel) thom: earl's court high street is right near the hotel tho :)
[06:27] <daniels> mdz: yah, been here for a week
[06:28] <daniels> thom: we're at south ken now, we'll be at earl's court later
[06:31] <daniels> thom: so logically anywhere near those two points, or between, is phat
[06:31] <daniels> and dope
[06:31] (thom/#ubuntu-devel) and k-rad?
[06:31] (thom/#ubuntu-devel) shall i -> sarf ken, then?
[06:32] <daniels> thom: how long will it take?
[06:33] (thom/#ubuntu-devel) some time
[06:33] (thom/#ubuntu-devel) uh, some random time
[06:33] (thom/#ubuntu-devel) less than an hour
[06:33] <daniels> thom: if my memory serves me correctly, and it's ~30min, fo'shizzle
[06:33] <daniels> post-6pm sounds reasonable
[06:35] (thom/#ubuntu-devel) yeah, that's about right
[06:36] <rburton> thom: you'll need one of these if you are going down to da hood with daniels -- http://www.theregister.co.uk/2004/11/24/iboom.jpg
[06:36] <eruin> lul
[06:36] <eruin> that pod looks so out of place
[06:36] (thom/#ubuntu-devel) *g*
[06:37] <daniels> rburton: oh dear
[06:37] <daniels> ghetto lyf
[06:44] (Kamion/#ubuntu-devel) hmm, I think I may need to create a pciutils-udeb
[06:45] (Kamion/#ubuntu-devel) I can't present the names of network interfaces otherwise
[06:47] <smurfix> Rich , using 
[06:52] (thom/#ubuntu-devel) right
[06:53] (bob2/#ubuntu-devel) rock
[06:54] <daniels> thom: dope, seeya in a bit
[07:03] <nmf> dir
[07:05] <mdz> oh yay, de-rootified syslogd
[07:06] <daniels> mdz: yeah, pitti went off about half an hour ago, he mumbled something about starting on derootifying X
[07:44] <azeem> executive review of MEPIS: 12(!) modified packages (mostly alsa, the kernel [without source] , cloop) three new binary-only packages (control/system center, installer) and a themes package
[07:44] <daniels> kernel modified sans source?
[07:44] <azeem> well
[07:45] <azeem> there are a couple of patches in /usr/src
[07:45] <azeem> the dorks in #mepis told me I should get the source at kernel.org
[07:45] <azeem> but couldn't even tell me it's modified from vanilla or debian
[07:45] <azeem> of course, they don't ship any of the source for their packages (except glibc, incidently), they refer to ftp.debian.org
[07:46] <daniels> smooth move
[07:46] <azeem> 19:14 < darkstego> I am guessing azeem doesn't have access to the private dowload
[07:46] <azeem> anyway, no match, they just seem to be good at bribing "authors"
[07:47] (Kamion/#ubuntu-devel) ooh, anything I can GPL clause 4 them for?
[07:47] <azeem> roblimo wrote a book about MEPIS, but couldn't tell me off-hand their APT-source line ("eh? what for? I'm using a GUI client...")
[07:47] <azeem> it's 2004.mepis.org/mepis, if you want to have a look
[07:48] <azeem> (of course, they might have all the source package somewhere else and I did not find it yet, but I doubt that)
[07:53] (lamont/#ubuntu-devel) daniels: you around still?
[07:55] (lamont/#ubuntu-devel) seb128: you around?
[07:55] (lamont/#ubuntu-devel) Preconfiguring packages ...
[07:55] (lamont/#ubuntu-devel) Can't exec "lspci": No such file or directory at /tmp/libglide3.config.429423 line 74, <STDIN> line 2.
[07:58] <daniels> lamont: represent
[08:01] <mxpxpod> seb128: ping
[08:04] (lamont/#ubuntu-devel) represent?
[08:04] (lamont/#ubuntu-devel) oh
[08:05] <elmo> lamont: daniels is stuck in his time warp again - he thinks he's a wapper from the 80's
[08:06] <elmo> clearly, David Bellamy wasn't insulting enough - I think we'll have to step it up a notch and start calling him Vanilla or VI
[08:06] <elmo> \o/ [yay for 200 miles of physical distance :)] 
[08:06] (lamont/#ubuntu-devel) lol
[08:08] <daniels> elmo: it's been done -- someone did that like two years ago.  mr time warp. :P
[08:09] <mjg59> They can't just refer to ftp.debian.org
[08:09] <mjg59> That would only work if we distributed under 3b, which we don't
[08:31] <eruin> hmm
[08:32] <eruin> permissions on /dev/nv* are too restrictive all of a sudden
[08:32] <eruin> any recent updates that might cause that?
[08:32] <seb128> lamont: yes ?
[08:32] <seb128> mxpxpod: pong
[08:32] (lamont/#ubuntu-devel) seb128: was griping about glide
[08:33] <seb128> I've nothing to do with glide afaik
[08:33] (lamont/#ubuntu-devel) ah, ok.  It starts with 'g' you see... :-)
[08:34] <seb128> yep, but this one is not mine :)
[08:34] <mxpxpod> seb128: have you looked into making a gaim-dev package?
[08:35] <seb128> mxpxpod: herz1 is working on that, #3959
[08:35] <mxpxpod> seb128: awesome, thanks
[08:35] <seb128> np
[08:40] <herz1> mxpxpod: i think we can close the bug at the weekend
[08:40] <herz1> .oO(today is criawips hacking day for me)
[08:40] <mxpxpod> herz1: that's awesome!
[08:40] <herz1> yes, i want to build a gaim-encryption package
[08:40] <mxpxpod> herz1: hopefully, that will lead to gaim plugin packages
[08:40] <mxpxpod> herz1: and a gaim-evolution-plugin package
[08:41] <herz1> yep
[08:41] <mxpxpod> herz1: there should probably be a standard naming scheme for gaim plugins
[08:41] <mxpxpod> herz1: like gaim-plugin-foo
[08:41] <herz1> yes
[08:41] <herz1> sounds good
[08:41] <mxpxpod> cool
[08:42] <herz1> apt-cache should be happy if we have upstreams package names in the long descrption
[08:42] <mxpxpod> yeah
[08:43] <herz1> mxpxpod: 4079
[08:44] <seb128> feel free to package/maintain it guys
[08:44] <seb128> herz1: the bug is about getting the package done by somebody or you're saying you'll maintain it ?
[08:47] <herz1> i can create and maintain it
[08:47] <seb128> herz1: cool :)
[08:54] <mdz> Changed-By: Brandon Hale <brandon@localhost.localdomain>
[08:54] <mdz> tseng: might want to change that :-)
[09:10] <Keybuk> mdz: Yeah, people who upload with someone else's name in Changed-By are odd :p
[09:12] <mdz> Keybuk: which of the photo->{html,thumbnails} generators do you use?
[09:12] <mdz> tseng: if I confused nicks, ignore me
[09:12] <Keybuk> mdz: NIH
[09:12] <mdz> gah
[09:12] <mdz> there are about 50 in Debian, surely one of them is decent
[09:13] <mdz> I just want something less scary than gallery
[09:13] <Keybuk> dunno, they all seem to pre-suppose MySQL, PostgreSQL, PHP, etc.
[09:14] <Keybuk> mine's just a shell script that does some imagemagick and writes out a static html page
[09:15] <mdz> I've written similar stuff too many times and want someone else to maintain it :-P
[09:16] <azeem> cthumb is what Keybuk described, except it is written in perl
[09:16] <Keybuk> heh, at some point I'll probably investigate pyblosxom gallery plugins; that seems to be a good way forward
[09:16] <azeem> otherwise, gthumb does a good job for publishing albums if you only need to do this once in a while, IMHO
[09:18] <azeem> it's at least being worked on, jimmac blogged about that IIRC
[09:20] <mdz> I'm trying curator, purely because it's python
[09:20] <mdz> aw, seems to be unmaintained
[09:21] <mdz> both upstream AND in Debian
[09:21] <mdz> bins, cthumb, curator, igal, imageindex and album are the ones I see
[09:21] <mdz> and galrey
[09:22] <elmo> the more featureful ones seem to have chequered pasts security wise :(
[09:23] <Keybuk> "Root me, root me, root me with your HTML-embedded scripting language!"
[09:28] <mdz> elmo: yeah, I just want something simple and non-interactive
[09:28] <mdz> which I think most of those are
[09:28] <mdz> curator produces reasonably nice output
[09:28] <lamont_> 0000:00:02.7 Multimedia audio controller: Silicon Integrated Systems [SiS]  Sound Controller (rev a0)
[09:28] <lamont_> should that work?
[09:29] <zul> it should
[09:29] <zul> i have one
[09:30] <mdz> lamont_: yeah, it's pretty common
[09:30] <elmo> I think last time I used igal because it was simple and produced nice multi-size options
[09:30] <lamont_> mdz: speaker icon in the panel has a red slash through it...
[09:31] <mdz> lamont_: I know you can dig deeper than that :-)
[09:31] <lamont_> yeah - digging dammit
[09:33] <zul> hint...intel8x0
[09:33] <lamont_> zul: is loaded
[09:33] <lamont_> SiS SI7012 [alsa mixer] 
[09:35] <lamont_> sigh.  once you unmute things, you still have to turn up the volume control on the speaker to > 0. :-(
[09:46] (tseng/#ubuntu-devel) mdz: you caught me
[09:52] (lamont/#ubuntu-devel) bonnie is now installing warty on her own computer, which has a second hard drive for doing hoary test installs.
[09:52] (lamont/#ubuntu-devel) that reminds me... gonna need a switch over there now...
[10:16] <RubenV> anything special 'bout it?
[10:16] (tseng/#ubuntu-devel) it seems pretty fast.
[10:16] (tseng/#ubuntu-devel) like, on hard drugs fast.
[10:17] <RubenV> if only my mirror wouldn't be broken
[10:18] <jdub> RubenV, tseng: no more bonobo! (mostly)
[10:18] <RubenV> ok
[10:18] <RubenV> let's fix my sources list
[10:18] <RubenV> gotta have that :)
[10:28] <sivang> jdub : already in hoary?
[10:28] <jdub> sivang: yes, seb128 is the master :)
[10:28] <sivang> jdub : yes he is :)
[10:29] <jdub> lamont: aren't we shifting to gcc-3.4 for hoary?
[10:31] <zul> latewr
[10:33] <sivang> jdub : does it use any new file search/index system? really fast..
[10:35] <eruin> new nautilus you say? /me aptgetupdates
[10:36] (lamont/#ubuntu-devel) jdub: uh, sure... that's a gcc-defaults change, I expect...
[10:36] (lamont/#ubuntu-devel) atm, we haven't yet.
[10:38] <eruin> any particular spot I can place flags like mcpu, etc?
[10:39] <eruin> sorry, wrong channel indeed
[10:41] <__daniel> anyone else having problems logging into the wiki?
[10:41] <__daniel> i always get "... connection terminated unexpectedly ..." error messages :-/
[10:41] <jdub> lamont: hrm, we should probably do that sooner rather than later
[10:41] (lamont/#ubuntu-devel) jdub: yeah.
[10:42] (lamont/#ubuntu-devel) that's a doko thing.
[10:46] <doko> jdub, lamont: we didn't decide on a gcc shift for hoary. with the C++ API change that would mean a lot of diversion in changing package names to reflect the new ABI and integration of somewhat 400 not yet applied patches to sucessfully build with 3.4. Not sure, if we really want to do that.
[10:48] <Mithrandir> doko: do you think we could have a talk about multiarch in Mataro?
[10:48] <Mithrandir> doko: like talking together, not holding a speech
[10:48] <Mithrandir> I want to move forward
[10:48] <mvo_> hi __daniel 
[10:48] <__daniel> hai mvo_
[10:49] <doko> Mithrandir: who was the guy working on multiarch?
[10:49] <Mithrandir> doko: darksatanic hacked a bit on the dpkg part of it.. apart from that, me. :)
[10:51] <doko> Mithrandir: yes, I know ;) yes, should be worth half a day or a day.
[11:31] <elmo> if the gnome desktop background change thing breaks - what's the most likely thing to kick/look at ?
[11:31] <seb128> elmo: define the breakage
[11:32] <pitti> Hi guys
[11:32] <seb128> you change it in the UI and there is no change at screen ?
[11:32] <seb128> evening pitti 
[11:32] <elmo> seb128: right
[11:32] <jdub> seb128: just uploaded vino + libxdamage-dev
[11:32] <jdub> seb128: just about to upload howl and gnome-vfs
[11:33] <seb128> elmo: that's #909, gconf's bug
[11:33] <seb128> elmo: killall -HUP metacity should do the trick as a workaround
[11:33] <elmo> pitti: please switch to uploading directly to jackass
[11:33] <jdub> seb128: unfortunately, the howl change is going to be scary :-)
[11:33] <seb128> jdub: what's new in gnome-vfs ?
[11:33] <seb128> oh, howl/gnomevfs
[11:33] <jdub> (howl dependency)
[11:34] <seb128> yeah
[11:34] <seb128> jdub: go go go, we need some scary changes, hoary is getting boring :p
[11:34] <jdub> this is really scary though
[11:34] <jdub> because we link so... liberally
[11:34] <jdub> (libtool's fault?"
[11:34] <jdub> everything that links to gnomevfs (which means EVERYTHING) will stop working when howl upgrades :-)
[11:35] <jdub>         libhowl-0.9.6.so.1 => not found
[11:35] <jdub> ^ FEAR
[11:35] <seb128> DOH
[11:35] <seb128> and while I'm thinking about that
[11:35] <seb128> if somebody know an app linking on the old libnautilus2.so let me know
[11:36] <seb128> since this lib has been dropped :p
[11:36] <seb128> (abiword was broken due to that)
[11:36] <elmo> seb128: doesn't seem to have worked :(
[11:36] <seb128> elmo: killall nautilus if you don't use it atm :)
[11:37] <jdub> seb128: so how should we deal with this mass b0rk?
[11:37] <elmo> seb128: that worked, thanks!
[11:37] <jdub> seb128: lots of package uploads sounds like teh suck to me.
[11:37] <seb128> elmo: np
[11:37] <seb128> jdub: thinking about it, that really sucks
[11:37] <pitti> elmo: I tried, but ftp does not work very well for me
[11:37] <seb128> jdub: move the issue to #gnome-debian :)
[11:37] <jdub> ok
[11:38] <pitti> elmo: but I can route it over my server and upload from there
[11:40] <jdub> seb128: our problem :-)
[11:41] <seb128> apparently :)
[11:42] <seb128> ok, so
[11:42] <seb128> $ apt-cache rdepends libhowl0 | wc -l
[11:42] <seb128> 55
[11:42] <seb128> DOH
[11:42] <jdub> :-)
[11:44] <seb128> jdub: so, what's the libhowl lib name now ?
[11:45] <seb128> libhowl.so.0.... ?
[11:45] <jdub> yeah
[11:45] <jdub> 0.0.0 :-)
[11:46] <seb128> I you did name the package correctly we wouldn't have this problem now :p
[11:46] <jdub> heh
[11:46] <jdub> it was .1 before! and totally broken! :)
[11:46] <seb128> yeah, crappy stuff should not go in the archive in the first place :p
[11:46] <jdub> heh
[11:47] <jdub> we hadn't identified the arse yet, unfortunately :|
[11:47] <seb128> BTW, the only right solution is the massive rebuild
[11:47] <jdub> FUN
[11:47] <seb128> but
[11:47] <seb128> is it binary compatible with the previous one ?
[11:47] <jdub> elmo: 
[11:47] <jdub> PGP/GnuPG signature check failed on vino_2.8.1-0ubuntu2_source.changes
[11:47] <jdub> gpg: Signature made Wed Nov 24 17:27:31 2004 EST using DSA key ID 565B38F9
[11:47] <jdub> gpg: Can't check signature: public key not found
[11:47] <jdub> (Exit status 2)
[11:48] <jdub> vino_2.8.1-0ubuntu2_source.changes has bad PGP/GnuPG signature!
[11:48] <jdub> Removing vino_2.8.1-0ubuntu2_source.changes, but keeping its associated files
[11:48] <jdub> for now.
[11:48] <jdub> 
[11:48] <jdub> seb128: i think so - i can check
[11:48] <seb128> jdub: in which case include a dummy libhowl-0.9.6.so.1 
[11:48] <seb128> time to rebuild
[11:49] <seb128> but we need to be sure to not build/link against the dummy :p
[11:49] <seb128> yeah, that's ugly ... but could avoid apps breakage
[11:49] <jdub> btw, is your totem crashing on startup atm?
[11:49] <seb128> gst or xine ?
[11:49] <jdub> xine
[11:50] <__daniel> jdub: are you talking about the hoary ones?
[11:50] <seb128> jdub: it crashes with empty playlists
[11:50] <seb128> jdub: workaround, start it with an file in argument and keep it in the list :)
[11:52] <jdub> heh
[11:52] <jdub> thanks :)
[11:53] <seb128> np
[11:55] <jdub> elmo: ping?
[11:57] <jdub> seb128: that's a reasonable workaround -> perhaps we should wait until after 2.9.2 comes out next week to reduce the rebuild load?
[11:58] <enrico> Hello.  It appears that we can't log into the wiki: is someone aware of it?
[11:59] <seb128> jdub: oh, I've a less-ugly solution :p
[11:59] <seb128> jdub: but longer to get
[11:59] <enrico> Like... the docteam people are a bit stuck without being able to log into the wiki...
[11:59] <seb128> jdub: upload a gnomevfs without howl, keep it the time to get the howl depends down and put the new back
[11:59] <jdub> enrico: please ping elmo
[12:00] <seb128> ie: 2.9.2 will not depends on howl
[12:00] <jdub> seb128: bah, that's no fun!