[12:05] <medwards_> A subset is going to go on a small box (probably ARM or MIPSel) and I'm prodding them along towards free Java.
[12:07] <mdke> ok sent to the list
[12:07] <mdke> someone check it ot
[12:07] <lamont> mdz: you want the bitchy message gone, or just the exit? :-)
[12:07] <medwards_> lamont: any reason not to use killall -clean?
[12:08] <mdz> lamont: both
[12:08] <mdz> medwards_: it really just needs to write a @#$@ pid file
[12:08] <pitti> night everybody
[12:08] <lamont> medwards_: I don't see a -clean...
[12:08] <mdz> but failing that, it shouldn't break
[12:09] <ogra> night pitti 
[12:09] <medwards_> lamont: the killall would probably do a better job of avoiding race conditions when it's spawning rapidly.
[12:09] <medwards_> -clean is -TERM, 5 seconds, -KILL iirc
[12:09] <lamont> make them smaller anyway
[12:10] <medwards_> I think killall contains logic to kill parents before children
[12:10] <medwards_> (in the -KILL scenario)
[12:11] <medwards_> if it doesn't, it probably should.
[12:11] <medwards_> when one is resorting to -9, there's not much point in letting parents wait for children.
[12:14] <medwards_> lamont: pidfile and decent handling of children on -TERM would be a good thing, all right.
[12:17] <medwards_> lamont: but see http://bugs.debian.org/298040 for handling of chroot installs
[12:17] <lamont> well, upgrade to the new version didn't die...
[12:18] <lamont> mdz: thoughts? do we mind waiting until breezy to fix 298040?
[12:18] <medwards_> The postinst kill code should probably also be made conditional on /etc/default/cupsys.
[12:18] <medwards_> lamont: 298040 is closed upstream
[12:18] <lamont> it should really just use a *&))^_& pidfile
[12:19] <lamont> medwards_: in -7, hoary is derrived from -1
[12:19] <medwards_> lamont: but the code in the postinst probably reintroduces it if it isn't conditional on FORCE_RESTART.
[12:19] <medwards_> lamont: oh.  I see.
[12:19] <lamont> start-stop-daemon is your friend. :-)
[12:20] <mdz> lamont: is it too much to ask for proper preinst/postinst handling of the daemon, like every other package?
[12:20] <mdz> at the very least it can kill reliably based on the inode
[12:20] <lamont> inode?
[12:20] <medwards_> yeah, I use start-stop-daemon in postrm and if I really need it dead I killall in postinst
[12:21] <medwards_> lamont: inode of binary
[12:21] <medwards_> lamont: none of this path BS
[12:21] <lamont> mdz: it's a question of how invasive a change you want in the postinst...
[12:22] <medwards_> killall in preinst, that is.
[12:22] <lamont> given how familiar I am with it (not very), I'm disinclined to change it now for hoary, other than to make it not fail...
[12:22] <smurfix> lamont: inode of the server process, i.e. it doesn't inadvertently kill ~me/foo/bin/thing instead of /usr/sbin/thing
[12:22] <medwards_> smurfix: and handles jailed processes correctly, too
[12:23] <wasabi_> oh here i was talking to myself
[12:23] <wasabi_> somebody said ant!
[12:23] <wasabi_> oh not to mention somebody said my actual name too
[12:23] <lamont> pidof uses the inode.  cooll
[12:24] <medwards_> lamont: does it attempt to order by parentage?
[12:25] <lamont> not that I see in 20 seconds of bouncing around in the code, nol.
[12:25] <lamont> this should fix both issues (using pidof, and not checking)
[12:27] <mdz> lamont: for hoary, the most important thing is that it not break upgrades
[12:28] <mdz> lamont: if in some oddball circumstances the user needs to reboot because cupsys is horked, so be it
[12:28] <lamont> mdz: ok.
[12:28] <lamont> -1ubuntu11 should fix that..
[12:28] <lamont> hrmpf.  should have put the bug # in the changelog. :-(
[12:29] <medwards_> lamont: how does this fix get submitted to debian?
[12:29] <lamont> by me updating the bug in debian...
[12:29] <lamont> unless you want the pleasure...
[12:30] <lamont> of course, step 1 is to verify that the bug still exists in sid, etc.
[12:30] <medwards_> I looked for a corresponding bug#, but it probably hadn't been reopened.
[12:30] <lamont> and, whether to blow up the upgrade or not may or may not be a maintainer decision in debian...
[12:30] <xerox> Hi. I need to install a more recent version of SBCL than ubuntu's default one. I tryed with checkinstall but I didn't succeded. Debian unstable has a more recent .deb, how can I grab it? Would it be a good thing to do?
[12:31] <medwards_> right, but using pidof and sleeping between -KILL and testing again will probably help.
[12:31] <lamont> yes
[12:31] <lamont> you want the pleasure?
[12:31] <lamont> or shall I put it on my list?
[12:31] <medwards_> will do.
[12:32] <mdz> xerox: your first question would be more appropriate in #ubuntu, and the second question is answered in the FAQ
[12:32] <dholbach> good night mvo
[12:32] <xerox> mdz: #ubuntu told me to try #ubuntu-devel :) I'll try with the faq..
[12:32] <mdz> xerox: if there is a bug in the version in Hoary that ought to be fixed, tell #ubuntu-motu
[12:33] <mvo> night all
[12:33] <xerox> mdz: the real problem is that sbcl version is rather old, and it has some bad bugs.
[12:33] <dholbach> xerox: yeah... join #ubuntu-motu
[12:33] <xerox> Reasking the same thing would be okay there?
[12:33] <dholbach> yes
[12:33] <ogra_dogwalk> yep
[12:34] <mdz> ogra_dogwalk: walking your dog while you're on IRC?
[12:34] <ogra_dogwalk> *g* just came in again :)
[12:35] <mdz> oh, wonderful, I can't login to ubuntulinux.org anymore
[12:35] <mdz> I wonder if that means ubuntu.com works now
[12:36] <medwards_> lamont: want me to look at 4987, or is there something else I can do to help (and buy a little goodwill WRT LiveCD coaching)?  :)
[12:36] <ogra> mdz, does it tell you soure actually logged in ? 
[12:36] <jdub> hrm, thully can even be a pain when he closes bugs...
[12:36] <ogra> argh
[12:36] <mdz> nope
[12:36] <ogra> jdub, we have him in -motu quite often recently
[12:36] <mdz> Sign-in failure
[12:36] <mdz> You are not currently logged in. Your user name and/or password may be incorrect. 
[12:36] <jdub> mdz: oh man :|
[12:37] <mdz> jdub: has someone made changes in the past few hours?
[12:37] <ogra> mdz, http://www.grawert.net/wiki_error.png
[12:37] <mdz> ogra: no, that's the old ubuntu.com/ubuntulinux.org error
[12:37] <mdz> this is different; it doesn't even accept my login/password
[12:37] <lamont> medwards_: 4987 just needs a sync... /me composes mail
[12:37] <jdub> mdz: not that i know of
[12:37] <ogra> i always log in through https and ubuntulinux.org, never used ubuntu.com
[12:39] <mdz> ogra: everyone does, because ubuntu.com doesn't work
[12:39] <mdz> it never has
[12:39] <ogra> ah, ok
[12:39] <mdz> but now ubuntulinux.org is borked too
[12:39] <ogra> yep
[12:39] <mdz> if you're logged in, don't log out ;-)
[12:39] <ogra> lol
[12:39] <jdub> "the ssh method"
[12:42] <smurfix> Didn't that login problem happen a few weeks ago already?
[12:42] <ogra> smurfix, months ago already
[12:42] <dholbach> did anyonce perceive a slight out-of-sync-ness of the archive?
[12:42] <dholbach> like: <dholbach> i have   Version: 0.12.2-1
 and   http://people.ubuntu.com/~lamont/buildLogs/l/linphone/0.12.2-2ubuntu1/   claims they all were successful
[12:43] <dholbach> xerox states something quite similar about sbcl
[12:43] <dholbach> and i'm not talking about the usual interval
[12:45] <lamont> but not for ppc or amd64
[12:45] <ogra> dholbach, fyi Package: linphone, Version: 0.12.2-2ubuntu1
[12:45] <ogra> seems ok here
[12:45] <dholbach> on an amd64
[12:45] <dholbach> daniel@bert:~/blue$ apt-cache show linphone | grep Version
[12:45] <dholbach> Version: 0.12.2-1
[12:45] <dholbach> daniel@bert:~/blue$
[12:45] <ogra> dholbach, i am on amd64
[12:46] <lamont> claimes they became installed about 2 hours ago
[12:46] <dholbach> xerox talks about  Version: 1:0.8.17.4-1   of sbcl, where 0.8.18-1 should be recent
[12:46] <ogra> argh, sorry, showsrc instead of show
[12:47] <ogra> dholbach, youre right
[12:48] <xerox> (I also note that 0.8.20 is stable now)
[12:53] <lunitik> Any chance of getting http://higgs.djpig.de/ubuntu/www/ incorporated into Ubuntu site officially? its quite useful  :)
[01:00] <lunitik> Note on previous message... not mine... guy has a link to contact him where you could ask permission... it really would be nice though... I still go to p.d.o to see whats in the archive when I'm not at my machine for instance, but its rather inacurate  :(
[01:12] <mjg59> Gagh.
[01:13] <mjg59> Ok, I've found how how Windows does suspend-to-disk when battery is low in suspend-to-ram
[01:13] <jbailey> mjg59: OOoo.. Howhowhow?
[01:13] <jbailey> Frequently wakeups?
[01:13] <mjg59> It sets a wakeup time
[01:13] <thom> mdz: take it no kubuntu?
[01:13] <mjg59> I don't know how it calculates the wakeup time
[01:14] <mjg59> So yeah, possibly it wakes up every 24 hours or so and checks the battery stats
[01:14] <jbailey> mjg59: Could be every 30 minutes, check battery level?
[01:14] <mjg59> Yeah
[01:14] <jbailey> Hmm, would 24 hours be often enough?
[01:15] <mjg59> Quite possibly not. 
[01:16] <dholbach> sounds like a statistical solution would be a possible approach :-)
[01:17] <Riddell> thom: kubuntu is good to go as far as I'm concerned
[01:18] <tseng> amu: have you addressed boost like we discussed?
[01:20] <medwards_> lamont: can you explain the relationship between initrd.list and initrd.gz?
[01:22] <ogra> tseng, he might be asleep, its 1:20am here
[01:22] <thom> Riddell: you have cds and torrents and stuff?
[01:22] <tseng> well im guessing he'll see the message eventually.
[01:22] <thom> Riddell: (I only care about the torrents, tbh)
[01:23] <medwards_> Does that get rebuilt when making a new livecd?  And if so, using what tool?
[01:23] <Riddell> thom: http://cdimage.ubuntu.com/kubuntu/daily-live/20050317.1/
[01:23] <Riddell> http://cdimage.ubuntu.com/kubuntu/daily/20050317.1/
[01:23] <mdz> medwards_: initrd.list and initrd.gz come from the installer builds
[01:24] <medwards_> mdz: the installer also builds the livecd initrd?
[01:25] <mdz> medwards_: the installer and the live CD use identical initrds
[01:25] <mdz> thom: we're making that determination on #kubuntu-devel
[01:25] <lamont> medwards_: in fact, I think if you had a 1GB CD, and included casper and the livecd cloop on it, you'd have an install/live cd
[01:25] <lamont> mdz: is true?
[01:26] <medwards_> thom: gtk boog?
[01:26] <mdz> lamont: the extra magic would be to add a boot parameter to switch casper on
[01:26] <lamont> right
[01:26] <mdz> in fact the DVD image is exactly that
[01:26] <dholbach> thom: nice :-)
[01:27] <thom> mdz: okey
[01:27] <seb128> mdz: gtk2-engines-industrial uploaded
[01:27] <mdz> thom: can you still see straight? ;-)
[01:27] <mdz> seb128: wonderful, thanks
[01:27] <seb128> you're welcome
[01:28] <thom> mdz: heh, yeah :-)
[01:28] <thom> the pub was so rammed that getting to the bar was the limiting factor ;-)
[01:28] <tseng> mdz: oh, ive run a quick test of the weekly dvd on qemu btw
[01:28] <tseng> nothing exploded
[01:28] <mdz> bar bottleneck
[01:28] <medwards_> thom: or the use of avi as a gdb stub? :)
[01:28] <mdz> tseng: wow, that must have taken a long time
[01:28] <thom> medwards_: the use of avi :-)
[01:28] <tseng> hah yeah.
[01:29] <thom> mdz: the centre point of the local irish community, on st patricks day. messy :-)
[01:29] <mdz> thom: let the kubuntu torrents fly
[01:29] <mdz> Kamion: are you around to do publishing magic?
[01:30] <mdz> thom: do you need for me to make the metafiles on little?
[01:30] <zenwhen> is there going to be a gtk1 clearlooks engine?
[01:31] <thom> mdz: i can do it on orcadas, not a problem; are we just going with the daily Riddell posted, or republishing as a release?
[01:31] <mdz> thom: I think I can decipher Kamion's scripts well enough to publish it as a release
[01:31] <thom> ok; if you want to do that magic i'll wait for a bit
[01:31] <jdub> mdz: (is this kubuntu build going to be announced more widely?)
[01:32] <mdz> jdub: this is the kubuntu preview :-)
[01:32] <jdub> excellent
[01:32] <mdz> jdub: it is, you should check out a live CD
[01:33] <medwards_> mdz: so if I want to try a unionfs livecd, I need to hack on ubuntu-installer (presumably the input to initrd.list as well as linuxrc)?
[01:34] <zenwhen> I have to say... for a KDE dist... Kubuntu is really sweet.
[01:34] <seb128> jdub: I've uploaded a fixed gtk2-engines-industrial
[01:34] <jdub> seb128: ta
[01:34] <seb128> jdub: bah, the effet on the desktop is not weird dude :p
[01:34] <seb128> that's the same behaviour as all the other windows
[01:35] <jdub> seb128: it totally is. bigarse 1px flashy shit on the left.
[01:35] <jdub> it looks very pronounced on the desktop
[01:35] <jdub> so many people have mentioned it
[01:35] <jdub> even some french people
[01:35] <jdub> ;-)
[01:35] <seb128> oh right
[01:36] <seb128> in fact it doesn't with my background
[01:36] <medwards_> hmm .. how do I draw "smiley with beret" again?
[01:36] <seb128> but it does with the calendar one :p
[01:36] <jdub> seb128: ah, yeah, that makes a big difference ;)
[01:36] <Riddell> ogra: tested kubuntu yet?
[01:36] <jdub> ogra: dude. *dragon* leather boots.
[01:36] <mdz> thom: so should I make-torrents or no?  last time there was some confusion with the torrents being present in one place and not in another
[01:36] <ogra> LOL
[01:36] <jdub> ogra: this is kde pimping!
[01:36] <ogra> LOOOL
[01:37] <thom> mdz: if you're doing the release, do torrents to
[01:37] <medwards_> jdub: the KDE dragon is a bastard child of Disney's Figment, neh?
[01:37] <ogra> Riddell, not the newest, nope
[01:37] <medwards_> (if you've been to EPCOT)
[01:37] <jdub> it is not my place to comment on the kde mascot's parentage or lack thereof
[01:38] <ogra> Riddell, i'll start the download before going to bed, test them tomorrow
[01:38] <cc> hmm, so is the latest array 7 (?) livecd all working well?
[01:38] <mdz> thom: ok
[01:39] <mdz> cc: yes, it's pretty solid
[01:39] <mdz> ogra: if you're leaving it on overnight, wait for a torrent ;-)
[01:39] <cc> mdz: so the X issues are fixed. thanks. i shall resync now
[01:39] <medwards_> cc: dunno yet, my torrent has said 3 hours to go for the last 3 hours.  :-(
[01:40] <ogra> mdz, k
[01:40] <medwards_> would be nice if torrents were distinguishable from one another on the tracker.
[01:40] <cc> medwards_: isn't it easier to get it from cdimage itself, rather than a torrent?
[01:41] <medwards_> e. g. hoary-array-7-live-i386.iso
[01:41] <thom> medwards_: yes
[01:41] <medwards_> cc: 8 KB/s
[01:41] <thom> there are many, many ways that the tracker could be improved
[01:42] <maswan> medwards_: you mean like this? http://cdimage.debian.org:6969/
[01:42] <thom> medwards_: well, "./hoary-install-i386.iso.torrent": "seeding" (100.0%) - 5P3s0.165D u358.7K/s-d0.0K/s u1097012K-d598874K ""
[01:42] <medwards_> thom: torrent is currently 28 kb/s (vs. 8 for cdimage.ubuntu)
[01:43] <Burgundavia> where is the current kbuntu torrent
[01:43] <Burgundavia> ?
[01:43] <thom> Burgundavia: not in place yet
[01:43] <Burgundavia> hmm
[01:44] <medwards_> maswan: similar tracker, yes, but ubuntu livecds aren't named uniquely (which hoary-livecd-i386.iso.torrent is array-7?)
[01:44] <maswan> medwards_: ah
[01:46] <Riddell> medwards_: rumour has it that one of konqi's footprints started it's own desktop environment
[01:47] <maswan> medwards_: well, that's kind of hard to solve in the tracker then.
[01:47] <medwards_> So let me get this straight.  cloop is writable, and the livecd uses device-mapper to route written blocks to a ramdisk overlay?
[01:48] <thom> maswan: you could probably get more useful metadata into the .torrent
[01:48] <medwards_> Riddell: several desktop environments could live happily in one of konqi's (memory) footprints.  :)
[01:48] <thom> Keybuk: you really are on .au time, eh?
[01:48] <Keybuk> pretty much :-/
[01:48] <maswan> thom: Well, you do have a text comment field
[01:49] <jdub> Keybuk: western australian time ;)
[01:49] <thom> maswan: (and you'd have to display that somehow in the tracker, but yeah)
[01:49] <maswan> thom: these days I timestamp the weekly debian iso torrents with the date -u of when they are made.
[01:50] <Keybuk> actually, I don't think I'm in any real timezone since I think it's Thursday
[01:50] <jdub> heh
[01:50] <medwards_> If unionfs worked (I can still trigger oopses with the Knoppix 3.8 version), it seems to me that it would be preferable;
[01:51] <medwards_> stack jff2 on squashfs, for instance.
[01:51] <medwards_> s/jff2/jffs2
[01:52] <cc> when ubuntu goes selinux, jffs2 might be a bottleneck
[01:53] <ogra> hey only about an hour to unix time 1111111111 
[01:53] <tseng> jffs3 is being developed and I hear that xattr is on the gameplan
[01:53] <jdub> the status is: 1111107212
[01:53] <tseng> jdub: are we there yet?
[01:53] <elmo> root     21673  109  6.3  38428 32808 pts/0    R+   00:53   0:04  |                   \_ /usr/sbin/apache2 -k start -DSSL
[01:53] <ogra> nearly
[01:54] <elmo> 109%... neat
[01:54] <medwards_> cc: I'm thinking in terms of server uses; one could afford to use a fresh chroot every sbuild run if it were just several overlaid package sets with a ramdisk on top.
[01:55] <maswan> hmpf. elmo's apache is neater than mine.
[01:55] <elmo> god damn I wish apache had some serious competition in the httpd market
[01:56] <medwards_> jffs2 comes in when I want to persist, say, a client's data (including apt-getted packages) on a USB stick but get most of my bits off the livecd.
[01:56] <maswan> inetd + echo&cat? :)
[01:56] <medwards_> elmo: tomcat?
[01:57] <medwards_> don't think there's a mod_php, though.
[01:57] <maswan> ehm. tomcat? isn't that an apache webservice thingie?
[01:58] <medwards_> maswan: java servlet container
[01:58] <HrdwrBoB> tomcat isn't a realistic alternative in terms of a multifunction httpd
[01:58] <maswan> medwards_: well, yeah, webservices is where I've been in contact with it.
[01:58] <medwards_> maswan: sucks less than you would think for static content
[01:59] <jdub> elmo: twisted. :-)
[01:59] <tseng> elmo: hey would you mind syncing gtkpod 0.88?
[02:00] <medwards_> HrdwrBoB: of course not -- when I have to deal with tomcat, I front-end it with apache+mod_jk{2,}
[02:00] <elmo> tseng: done
[02:00] <tseng> elmo: you rock!
[02:00] <ogra> tseng, dont forget gnupod, and all the other pods ;)
[02:00] <tseng> heh.
[02:00] <thom> jetty isn't awful but is still java; apache is the only thing i'd trust for serving large data
[02:01] <ogra> tseng, thanks for the thully work btw
[02:01] <tseng> ogra: yeah np
[02:01] <jdub> heh, thully work ;)
[02:01] <thom> tseng: you poor guy
[02:01] <medwards_> thom: agreed, although jetty is even more of a pain to build than jakarta stuff.
[02:01] <HrdwrBoB> medwards_: yeah
[02:01] <tseng> thom: better then HostingGeek at least
[02:01] <thom> tseng: maybe
[02:02] <jdub> thom: "ASF member announces: 'Apache only httpd I would trust'."
[02:02] <lifeless> news @ 12
[02:02] <mdz> jdub: in other news, Ubuntu developers running Ubuntu on critical systems
[02:02] <medwards_> elmo: depending on how you define "competition", IIS  :-/
[02:03] <HrdwrBoB> medwards_: IIS on win2k3 server is in some ways superior
[02:03] <thom> IIS is great, barring the fact that having a web server in ring 0 is UTTER CRACK
[02:03] <maswan> well, or ftp.se.d.o or ftp.acc.umu.se. :)
[02:03] <jdub> mdz: hoary has been running admirably on my critical systems, btw.
[02:03] <mdz> jdub: I run Warty on my servers
[02:03] <mdz> and Hoary on my development systems
[02:03] <HrdwrBoB> it's an embarrassment that IIS can run 500 websites all with their own users and apache can't
[02:04] <medwards_> mdz: now if ubuntu ran on sparc, debian would have an infrastructure plan ... :-/
[02:04] <thom> HrdwrBoB: it's irritating
[02:04] <mdz> medwards_: sparc.ubuntu.com
[02:04] <jdub> like warty's dovecot.
[02:04] <thom> HrdwrBoB: since perchild/metux has the potential to really kick ass
[02:04] <HrdwrBoB> thom: yeah it looks great
[02:04] <tseng> jdub: i use warty dovecot
[02:05] <tseng> whats the deal with that?
[02:05] <thom> but perchild is dead and the metux guys have no inclination to get their stuff upstream
[02:05] <medwards_> HrdwrBoB: that's actually the sort of thing I would look to apache2+mod_jk2+tomcat5 for.
[02:05] <jdub> tseng: hurts on lots of users, hurts on huge folders
[02:05] <tseng> ah.
[02:05] <thom> medwards_: why? it should be way lower level than application server
[02:06] <HrdwrBoB> medwards_: all the file access should run as the user for that website
[02:06] <medwards_> mdz: sparc64 strategy?  biarch?
[02:06] <HrdwrBoB> so the permissions can be sane
[02:06] <thom> medwards_: 64bit kernel, 32bit userland
[02:06] <mdz> medwards_: following the same road as Debian so far; I don't think fabbione has given it much thought yet
[02:07] <thom> HrdwrBoB: yeah, it'll be nice when it happens
[02:07] <HrdwrBoB> I worked around this partially by using suexec and making php work by php calling php cgi
[02:07] <HrdwrBoB> but you lose the mod_php features
[02:07] <thom> "features"
[02:07] <HrdwrBoB> and you still don't access the files the right way
[02:07] <thom> nod
[02:08] <mdke> *cough* website
[02:09] <HrdwrBoB> oh, and I had to add a post upload script that made php files executable
[02:09] <mjg59> daniels: Stock Hoary preview install, and I have no working dri on an i915
[02:09] <mjg59> It says that it's initializing the SW cursor, which is probably a bad sign
[02:09] <thom> HrdwrBoB: bleah
[02:09] <thom> mdke: hrm?
[02:10] <mdke> thom, something is up with the website. we can't log in
[02:11] <medwards_> HrdwrBoB: tomcat is not jboss.  :-)
[02:11] <mjg59> daniels: Argh, cock. It's because it hasn't picked enough video RAM for DRI
[02:13] <jdub> fontconfig/libfcontconfig1 postinst b0rkage is known?
[02:13] <jdub>  libfontconfig1 depends on fontconfig; however:
[02:13] <jdub>   Package fontconfig is not configured yet.
[02:13] <mdke> maybe someone has hacked our website
[02:14] <HrdwrBoB> never attribute to malice...
[02:14] <mdke> HrdwrBoB, heh
[02:14] <mdke> HrdwrBoB, maybe it needs a reboot
[02:14] <mdke> *crosses fingers*
[02:15] <mjg59> daniels: Something very wrong here. I set VideoRAM and it still claims that it needs more RAM.
[02:15] <medwards_> mjg59:  that word is currently reserved for ij, iirc.
[02:15] <elmo> the website should work now
[02:15] <mdke> thanks elmo 
[02:16] <mdke> elmo, nope :(
[02:17] <elmo> mdke: works for me
[02:17] <elmo> and I can see it working for others
[02:17] <mdke> elmo, ok
[02:17] <mdke> hmm
[02:18] <mdke> elmo, yes sorry
[02:18] <mdke> thanks a lot
[02:22] <Burgundavia> jdub: I was working a user through that, I don't see it on my system though
[02:22] <Burgundavia> jdub: didn't resolve anything
[02:33] <ogra> lol
 mark shutleworth is from space
[02:33] <thom> may explain a few things
[02:33] <ogra> heh
[02:46] <HrdwrBoB> haha
[02:53] <mdz> jdub: someone commented about it on another bug, but no proper bug yet
[02:53] <mdz> jdub: mvo went to sleep; fixes appreciated
[02:57] <HrdwrBoB> 1111111101
[02:57] <HrdwrBoB> 1111111103
[02:57] <ogra> yeah
[02:58] <ogra> HrdwrBoB, not here yet
[02:58] <HrdwrBoB> heh
[02:58] <ogra> 1111111098
[02:58] <tseng> 1111111102
[02:58] <tseng> brandon@lappy:~$ date +%s
[02:58] <tseng> 1111111111
[02:58] <tseng> wee.
[02:58] <maswan> whee
[02:59] <ogra> yeah, happy 1111111111
[03:02] <mdke> ogra, man you shouldn't be able to work that out
[03:02] <ogra> :)
[03:02] <mdke> heh
[03:03] <Keybuk> mdke: *shrug* one assumes ogra was born in 1970
[03:03] <ogra> yup
[03:03] <ogra> mid of feb, to be precise
[03:03] <ogra> man, i'm old...
[03:04] <ogra> ...time for bed....
[03:04] <ogra> night all
[03:04] <Keybuk> (and, by inference, ogra)
[03:04] <mdke> nite
[03:04] <ogra> like nearly everyone around
[03:04] <mdke> how old is that?
[03:04] <mdke> 111111111?
[03:04] <thom> Keybuk: damn youngster ;-)
[03:05] <Keybuk> mdke: roughly 332793000
[03:05] <mdke> *head spins*
[03:05] <ogra> hehe
[03:06] <mdke> what date is unix year dot?
[03:06] <ogra> mdke, Keybuk is not the man i would argue with about numbers ;)
[03:06] <mdke> ogra, no change, i don't have a clue
[03:06] <mdke> *chance
[03:06] <Keybuk> ogra: hmm?  I can barely count
[03:07] <thom> Keybuk: thanks :P
[03:07] <Keybuk> mdke: Sweetmorn, Chaos 1, 3136 YOLD
[03:07] <mdke> *head spins*
[03:07] <mdke> in christian time?
[03:08] <Keybuk> that's boring :p
[03:08] <Keybuk> and I suspect you really mean Gregorian time
[03:08] <mdke> "he Unix epoch is the time 00:00:00 UTC on January 1, 1970"
[03:09] <mdke> hmm
[03:09] <schweeb> Keybuk: 14.4 yrs younger than epoch here, heh
[03:09] <thom> 9.7 here
[03:10] <mdke> 11.9
[03:10] <mdke> you guys are clearly just having a good time in here, would anyone like to look at https://www.ubuntulinux.org/wiki/IdeasForNewFrontPageStructure and comment?
[03:11] <mdke> see also subpages
[03:15] <mdz> Keybuk: you are young
[03:24] <zul> hey
[03:31] <thom> g'night folks
[03:35] <Riddell> thom: not staying up to set up torrents?
[03:35] <thom> Riddell: already running
[03:37] <Riddell> thom: URL?
[03:38] <thom> http://torrent.ubuntu.com/kubuntu/preview/
[03:38] <Riddell> thom: excellent, many thanks
[03:39] <thom> np
[03:39] <thom> sleeptime now
[03:39] <infinity> 'Night thom.
[03:39] <thom> ciao
[03:50] <mdz> infinity: please squash https://bugzilla.ubuntu.com/show_bug.cgi?id=7819
[03:50] <mdz> infinity: highest priority
[03:50] <mdz> looks like a regression in mvo's upload
[03:53] <infinity> On it.  Was just catching up on bugzilla mail right now.
[03:53] <calc> thom: here?
[03:54] <usual> hey calc
[03:54] <calc> hi
[03:54] <usual> calc, do you have anything to do with kubuntu?
[03:54] <calc> no :\
[03:54] <usual> calc, ok, I remember you packaging kde in debian
[03:54] <calc> yea
[03:54] <calc> i tried to get a job with credativ but they don't appear to be hiring
[03:55] <usual> ahh
[03:55] <infinity> mdz : I don't suppose there's an ubuntu snapshot archive (like snapshot.debian.net), where one can easily grab old sources and diff to track regressions?
[03:55] <usual> calc, you did a great job
[03:55] <calc> thanks
[03:55] <mdz> infinity: morgue.ubuntu.com
[03:56] <mdz> calc: thom went to sleep
[03:56] <calc> ok
[03:56] <mdz> calc: credativ is not a prerequisite for kubuntu, you know
[03:56] <calc> true, but with my current job i very little time to do anything else
[03:56] <calc> at work 9hr+ and 3hr commute
[03:56] <mdz> I know how that goes
[03:56] <mdz> 3hr driving or transit?
[03:57] <calc> driving
[03:57] <mdz> ouch
[03:57] <calc> yea if i could sleep for 3hr that would be fine ;)
[03:57] <zul> calc: against traffic or with traffic
[03:57] <calc> obviously with traffic or it wouldn't take 3hr ;)
[03:58] <zul> calc: meh..
[03:58] <calc> the place i work is only ~ 40 miles away
[04:00] <mdz> calc: someone was trying to coerce me into an updates libvorbis in hoary the other day
[04:00] <mdz> s/updates/updated/
[04:00] <calc> heh
[04:00] <calc> jdub asked me about it a few days ago and so i tried rebuilding a chroot
[04:01] <calc> the first time it took my box several hours just to download it must have hit a really bad mirror
[04:01] <mdz> the vagaries of a time-based release schedule
[04:01] <zul> mdz: like the ipw2200 driver?
[04:01] <mdz> zul: yep, only more persistent than thully
[04:01] <zul> more? oi vey...
[04:02] <calc> mdz: so ubuntu hoary releases in about 2 weeks right?
[04:05] <mdz> calc: April 6th
[04:05] <mdz> release candidate on march 30th
[04:06] <mdz> in otherwords, we're solidly in stabilization mode and not kowtowing to requests for new upstream versions ;-)
[04:06] <zul> like ipw2200
[04:07] <zul> oops..:)
[04:07] <calc> mdz: ok
[04:08] <calc> hmm yea a newer ipw2200 than from last dec would have been useful
[04:08] <calc> its broken on amd64
[04:11] <calc> fsck
[04:11] <calc> i uploaded libao a debian native
[04:12] <calc> stupid .orig renaming
[04:14] <Amaranth> hey, doesn't anyone know where the gnome menus get their icons from?
[04:14] <Amaranth> like the icon for the "Accessories" menu
[04:16] <calc> it should be fairly easy to find out
[04:16] <Amaranth> oh?
[04:17] <Amaranth> appearently not, i'm lost ;P
[04:17] <calc> look in /etc/xdg/menus
[04:17] <calc> which would then point you to /usr/share/desktop-directories/Accessories.desktop
[04:17] <calc> which then points you to the icon "gnome-util"
[04:18] <Amaranth> *facepalm*
[04:18] <calc> perhaps not everyone has ingrained the fdo specs into their brain ;)
[04:18] <Amaranth> this is what i needed when i started on this stupid thing
[04:18] <Amaranth> i hate fdo specs :P
[04:18] <Amaranth> oh, you're the one responsible for that annoying Debian menu showing up last week?
[04:19] <Amaranth> :P
[04:19] <calc> hehe
[04:19] <calc> i wrote it many months ago
[04:19] <calc> i guess you installed it last week
[04:19] <Amaranth> yeah
[04:19] <calc> hey it works right
[04:19] <calc> it took me a while to debug it to where it worked according to policy ;)
[04:19] <Amaranth> the fdo menu spec is a pita, i'd just like to point that out
[04:20] <calc> someone could easily make nice looking icons for it as well, i am not an artist though so didn't do that
[04:20] <Amaranth> no wonder GNOME 2.10 didn't have a menu editor
[04:20] <calc> its not that difficult to make changes to the menu
[04:20] <Amaranth> my editor more or less works but it has some issues
[04:20] <calc> though someone has to read the spec and actually grok wtf its talking about
[04:21] <calc> well i mean wrt the user editting it in their $HOME and it actually merging them
[04:21] <calc> manually editing the file in /etc/xdg is not a good idea
[04:21] <calc> that makes it change for all users
[04:21] <Riddell> http://www.kubuntu.org.uk/  Preview Released
[04:21] <calc> Riddell: cool
[04:21] <tseng> Riddell: go go gadget crack pipe
[04:22] <tseng> good show.
[04:22] <Riddell> :)
[04:24] <mdz> Riddell: what is KGX?
[04:24] <Riddell> KDE/GNU/Unix
[04:24] <Riddell> the term never caught on for some reason
[04:25] <mdz> is it in common use in the KDE community?
[04:25] <calc> hmm can you call something Unix without paying someone?
[04:25] <Riddell> mdz: only occational
[04:25] <Riddell> calc: that's why it's just an X :)
[04:26] <calc> heh
[04:41] <jammcq> mdz: Hey, did you tell me that the Ubuntu live cd will look for a swap partition on a harddrive and use it?
[04:41] <infinity> mdz : Do you want debdiff patches to the bug(s) in question, or should I find a sponsor and upload?
[04:42] <mdz> infinity: the latter
[04:42] <mdz> jammcq: that's correct
[04:43] <jammcq> mdz: does it also create a swapfile on a FAT filesystem if it finds it?
[04:43] <mdz> jammcq: no, it doesn't
[04:43] <jammcq> we're discussing this over in #ltsp, to look for a harddrive and use it for swap, but we don't want to destroy any existing data
[04:43] <mdz> infinity: you are not in the keyring yet?  did you mail elmo?
[04:44] <mdz> jammcq: using an existing swap partition is quite safe; I'm not sure that I would create a swap file
[04:44] <infinity> mdz : Unless he's added me silently, I'd go with 'no'... I'll bug him right now (Didn't do so earlier, due to the whole start date fiasco).
[04:44] <mdz> if nothing else, it could be left behind occupying lots of space, in the event of a crash
[04:44] <mdz> infinity: he's asleep if he has any sense
[04:44] <infinity> mdz : His idle time disagrees.
[04:44] <jammcq> ok, it just seems that it would be kind of rare for it to find a swap partition.  at least outside of the linux community
[04:45] <mdz> infinity: I qualified that statement
[04:45] <infinity> Heh.
[04:48] <mdz> infinity: if no one else is around, i can sponsor you
[04:48] <bob2> cool
[04:48] <bob2> "rmmod hci_usb" hangs rmmod at 100% cpu
[04:48] <infinity> mdz : elmo's alive.  I'm bugging him as we speak.
[04:48] <bob2> and fails
[04:50] <mdz> bob2: "then don't do that"
[04:50] <srbaker> what's a good tool for non destructively resizing partitions?
[04:51] <mdz> srbaker: parted (glad to talk about it further in #ubuntu...)
[04:51] <bob2> mdz: hah
[04:51] <infinity> srbaker : In theory, or practice?...  In practice, the best tool is backing up and doing repartitioning destructively. :)
[04:51] <srbaker> infinity, hehe
[04:56] <calc> ok all the new ogg/theora stuff is sitting in debian incoming now
[05:04] <lamont> mdz: hoary-test is building on all 4 architectures.  we did leave one buildd of each _not_ trying, since there are some long builds near the top, and even at 1-per-run, they'd bottle neck everything occasionally....
[05:05] <lamont> hoary-test is the last chosen, and only builds if nothing else wants to build.
[05:05] <lamont> but it takes 10 things at a time
[05:06] <mdz> lamont: sounds good; how long do you figure it will take to try everything?
[05:07] <lamont> last time around it was on the order of 70 hours or so, using 3 buildd's, and no ccache..
[05:07] <lamont> so probably about 2-3 days
[05:07] <infinity> mdz : fontconfig fix up.
[05:07] <lamont> possibly 4-5
[05:10] <lamont> mdz: of course, I expect _main_ to be done sometime tomorrow
[05:11] <lamont> would be sooner, but oo.o, oo.o2, kernel, xorg kinda add up :)
[05:11] <lamont> i386 will be the leader, for sure
[05:11] <elmo> I haven't even imported universe yet - that'll take like a quarter of a day, easy, I reckon
[05:12] <lamont> woot
[05:13] <lamont> mdz: if you want the test build to be done sooner, we can include that last buildd, but it might mean that you have to wait an hour or two for, say, a casper build...
[05:13] <lamont> elmo: duh.  1400 pkgs would be !universe, wouldn't it... :-(
[05:14] <calc> 1400 sounds less than even main
[05:14] <calc> maybe just what fits on the cd?
[05:15] <elmo> source packages
[05:15] <calc> oh
[05:16] <lamont> and it's really closer to 1500.  /me rounded agressively
[06:04] <fabbione> morning
[06:08] <fabbione> night lamont 
[06:10] <Cym> I have a question about the libavcodec package..
[06:11] <Cym> it appears that libpostproc is not part of the packaging rules (which is makes sense to build at the same time as libavcodec) IMHO
[06:13] <Cym> i figure that when I do a dpkg-buildpackage after "apt-get source libavcodec", then it should create packages for ffmpeg, libavcodec, and libpostproc
[06:13] <Cym> then migrating these rules to cvs would be much easier to maintain
[06:29] <fabbione> ah neat
[06:30] <fabbione> sparc is only lagging 13 packages for main
[06:35] <elmo> caught up fairly fast
[06:39] <fabbione> with 15 days of downtime.. yeah
[06:39] <fabbione> it was pretty good
[07:07] <dilinger> what kind of build machines do you guys use for sparc?
[07:08] <fabbione> my netra t1 :-)
[07:10] <dilinger> 400 or 500mhz ultrasparc IIi?
[07:12] <fabbione> 433 iirc
[07:15] <dilinger> cool
[07:15] <fabbione> mdz: any objection for a console-data port upload that does NOT touch any of the existing code?
[07:16] <fabbione> it adds one rule in debian/rules
[07:16] <fabbione> and sparc in debian/control for one package
[07:17] <fabbione> jdub: ^^
[07:25] <mdz> infinity: #7819 can be closed if you have uploaded the fix
[07:27] <infinity> mdz : Just did so.
[07:27] <mdz> infinity: thanks
[07:32] <elmo> hmm, Assigned to Me is broken for me in bugzilla
[07:35] <Cym> does the maintainer for ffmpeg/libavcodec visit this room?
[08:06] <doko> good morning
[08:06] <amu> moin2
[08:08] <infinity> Mornin', doko.
[08:11] <mdz> distrowatch says that linuxcd.org is selling kubuntu CDs already
[08:14] <amu> mdz: heh 
[08:18] <pitti> Morning
[08:27] <doko> morning pitti
[08:27] <doko> mdz: I assume the OOo upload should be safe now?
[08:28] <pitti> doko: morning
[08:28] <mdz> doko: yes, I have not uploaded it yet, feel free
[08:33] <doko> mdz: ok, done
[08:34] <doko> this 1.1.3, haggai didn't feel to comfortable upgrading to 1.1.4 at this point, and the major reports are fixed
[08:48] <pitti> Riddell/amu: ping
[08:52] <fabbione> Mithrandir: where are you now?
[08:52] <Mithrandir> Oslo
[08:52] <Mithrandir> my father's
[08:52] <fabbione> did you have a nice trip?
[08:52] <fabbione> i got the msg btw.. if the address is correct, it is very close to where i live
[08:53] <Mithrandir> cool
[08:53] <Mithrandir> nah, horrible trip.  Cold, people talking, uncomfy seats.
[08:53] <Mithrandir> but I'm managing.  Karianne is sleeping a bit
[08:53] <Riddell> pitti: hi
[08:53] <fabbione> suckage
[08:54] <pitti> Hi fabbione 
[08:54] <fabbione> hey pitti
[08:54] <pitti> Riddell: CAN-2005-0396 is fixed in Warty, but not Hoary
[08:54] <pitti> fabbione: no reason to, I just wanted to say hello :-)
[08:54] <fabbione> ;)
[08:55] <pitti> Riddell: will you still do this?
[08:55] <fabbione> only when you talk about CAN's
[08:55] <Riddell> pitti: CAN-2005-0396 is "KDE version prior to KDE 3.4"
[08:56] <Riddell> pitti: I'll not that in the changelog 
[08:56] <Riddell> note
[08:56] <pitti> Riddell: ah, ok
[08:56] <pitti> Riddell: no need to, if it didn't require a change
[08:56] <pitti> Riddell: I'll add it to the nonvuln list then
[08:57] <pitti> Riddell: OTOH, a note in the changelog would be easier if you have to do another upload anyway
[08:57] <Riddell> pitti: cool.  how come e.g. CAN-2005-0237 didn't set off your alarm?
[08:57] <Riddell> pitti: I don't have another upload planned (I plan to go to sleep)
[08:57] <pitti> CAN-2005-0237 	
[08:57] <pitti> kdelibs	(hoary/main, warty-security/universe)
[08:57] <pitti> ^ this is from my security radar
[08:57] <pitti> so everything is in order for this
[08:58] <pitti> ... it seems
[09:00] <pitti> ogra: here?
[09:22] <HiddenWolf> mvo?
[09:28] <dholbach> gooood morning
[09:31] <d3vic3> morning
[09:32] <HiddenWolf> seb128: is it correct that that gtk-engine fix you did only fixes the ubuntu-themed mousecursor for the session?
[09:32] <pitti> Hi seb128
[09:32] <seb128> it does what is written in the changelog
[09:32] <seb128> hi
[09:32] <pitti> dholbach: Morning
[09:32] <dholbach> morning seb128, pitti, d3vic3 
[09:33] <pitti> dholbach: do you want to merge the Debian changes of grip to fix the security bug?
[09:34] <dholbach> pitti: if it has time until i got some coffee, i'll do it :-)
[09:36] <pitti> dholbach: oh, I'm not hurrying you :-)
[09:36] <pitti> dholbach: just wanted to tell you about it
[09:52] <HiddenWolf> Ugh. Firefox won't download anything anymore
[10:03] <sivang> morning all
[10:03] <dholbach> *grmbl* is gamin still on crack?
[10:04] <dholbach> morning sivan!
[10:04] <pitti> Hi sivang
[10:05] <sivang> morning dholbach 
[10:05] <sivang> pitti: Hey Martin 
[10:05] <sivang> pitti: just saw the thread about the users-admin tool, I dojn
[10:05] <sivang> pitti: suppose there a sane way to make an upgrade path upgrade if the admin didn't do custom configs?
[10:06] <sivang> pitti: (or maybe we do not want this to be available at all as an upgrade path)
[10:07] <dholbach> pitti: we can just sync it
[10:07] <dholbach> elmo: could you please sync  grip  from sid?
[10:08] <dholbach> morning mvo
[10:08] <mvo> hi dholbach 
[10:08] <mvo> morning all
[10:09] <Kamion> morning
[10:10] <fabbione> morning Kamion
[10:10] <fabbione> hi everybody
[10:10] <dholbach> hi fabbione, Kamion 
[10:13] <sivang> mvo: so tired you cannot low five back? :)
[10:13] <mvo> sivang: I hadn't had my first cup of tea yet :)
[10:14] <sivang> mvo: hehehe , you're excused :-)
[10:14] <Amaranth> hey, you guys wanna help me beat some bugs out of this menu editor i wrote? :)
[10:16] <seb128> Amaranth: don't ask to ask, just ask
[10:16] <Amaranth> http://dev.realistanew.com/menu-editor/menueditor_0.1-1_i386.deb
[10:16] <Amaranth> heh
[10:16] <Amaranth> it's not really i386, i just suck at making packages
[10:17] <ajmitch> Amaranth: btw, did you use python-xdg for it?
[10:17] <Amaranth> *facepalm*
[10:18] <Amaranth> I probably will for the next version...
[10:18] <ajmitch> I was hoping to work on an editor but real life got in the way :)
[10:18] <seb128> http://denu.sourceforge.net/
[10:18] <seb128> and there is one in C on the GNOME CVS
[10:19] <seb128> that's just for information
[10:19] <Amaranth> yeah, denu was just weird the the on in CVS doesn't really do anything
[10:19] <Amaranth> and i don't know C :)
[10:19] <seb128> denu works ?
[10:20] <seb128> your one looks nice, but doesn't really do anything atm
[10:20] <Amaranth> it can edit current entries and add new ones
[10:21] <seb128> it doesn't display anything else than categories here
[10:21] <seb128> Traceback (most recent call last):
[10:21] <seb128>   File "./menu-editor", line 343, in editEntry
[10:21] <seb128>     entry = self.entries[model.get_value(iter, 0)] 
[10:21] <seb128> KeyError: 'Games'
[10:21] <Amaranth> yeah, if you don't run it as root (gksudo or sudo) it uses ~/.local/share/applications/
[10:21] <seb128> it should merge both
[10:21] <seb128> you really want to use pyxdg
[10:21] <Amaranth> seb128: Not if you aren't running as root...
[10:22] <seb128> sure
[10:22] <Amaranth> How could you edit them?
[10:22] <seb128> read the spec
[10:22] <seb128> by creating a new .desktop in ~/.local
[10:22] <seb128> that's user first, then system
[10:22] <Amaranth> oh, *facepalm*
[10:22] <Amaranth> i only read enough of it to make what i have work :P
[10:22] <seb128> ie: it takes the user change first and then merge the system entries
[10:22] <seb128> ah ah
[10:23] <Amaranth> 0.2 will be out tomorrow using python-xdg and doing that :P
[10:23] <Amaranth> well, maybe saturday
[10:23] <seb128> sudo apt-get install python-xdg
[10:23] <seb128> python /usr/share/doc/python-xdg/examples/test-menu.py /etc/xdg/menus/applications.menu
[10:23] <seb128> python-xdg already does a lot
[10:24] <Amaranth> neat, but it gave me the german version
[10:24] <seb128> ie this example builds the full application menu, you just have to display what it returns
[10:24] <seb128> menu.setLocale("de") is used in the sources
[10:24] <Amaranth> ah
[10:24] <seb128> you can change that for whatever you want
[10:24] <Amaranth> hey, where is that stored?
[10:24] <Amaranth> in gconf or whatever
[10:24] <seb128> /usr/share/doc/python-xdg/examples/test-menu.py
[10:24] <Amaranth> so i can load the correct language
[10:25] <seb128> sudo editor /usr/share/doc/python-xdg/examples/test-menu.py
[10:25] <Amaranth> yeah, i've already got that open
[10:25] <seb128> menu.setLocale is in the file
[10:25] <seb128> it's not that big
[10:25] <seb128> should be easy to find
[10:26] <Amaranth> how can i see what language the user is using so i can show the right info?
[10:26] <seb128> use the locale settings
[10:27] <Amaranth> where do i get those from? :)
[10:29] <seb128> python -c "import locale; print locale.getdefaultlocale()"
[10:29] <seb128> by example
[10:30] <Amaranth> bleh, so simple
[10:30] <Amaranth> why can't i find these things at 3:30am?
[10:30] <sivang> seb128: what does that script do? (test-menu.py)
[10:31] <seb128> sivang: just run it and see
 python /usr/share/doc/python-xdg/examples/test-menu.py /etc/xdg/menus/applications.menu
[10:31] <sivang> seb128: k, thanks :)
[10:32] <sivang> ah! extracts the menu layout of the desktop
[10:32] <pitti> dholbach: what about the howl removal from grip?
[10:33] <dholbach> pitti: a rebuild sufficed :-)
[10:33] <pitti> dholbach: ah, there were no actual changes? good
[10:33] <dholbach> pitti: it's a dependency expanded by shlibs:Dependencies and libgnomevfs
[10:33] <dholbach> pitti: just newer config.{guess,sub} 
[10:33] <DM_Rado> hi everyone
[10:35] <sivang> seb128: is there a way to make a toplevel menu when set as a checkbox act nicely? I tried using one as a checkbox but it requires 2 click to make it checked/unchecked which is kinda bad..
[10:36] <Treenaks> sivang: toplevel menu? as in a checkbox in the menu bar?
[10:36] <DM_Rado> i have an acer laptop, travelmate 730 series, ubuntu seems too have alot of errors when installing
[10:36] <sivang> Treenaks: exactl
[10:36] <seb128> sivang: I don't get the question. You want to put a checkbox in the menu bar ? that's ugly
[10:36] <sivang> Treenaks: exactly, even
[10:37] <sivang> seb128: true, nevermind.
[10:37] <Treenaks> not only is it ugly, it's against the HIG :)
[10:37] <DM_Rado> is there any way around this?
[10:38] <sivang> Treenaks: shooosh, don't let anybody hear I suggested that :)
[10:47] <Kamion> seb128: any chance you could make GNOME stuff depend on gs-esp | gs rather than just gs?
[10:47] <doko> kamion: is there an easy way to determine the release date of a install/live cd?
[10:47] <mvo> doko: there is /cdrom/.disk/info
[10:47] <Kamion> seb128: I thought this was done for warty, but either it was incomplete or it's regressed
[10:47] <doko> mvo: thanks
[10:48] <Kamion> yeah, what mvo said
[10:49] <Kamion> seb128: the reason for this is that otherwise DVD installations prompt in the second stage to reinsert the CD in order to install gs and gs-gpl; gs is a mixed virtual packages, and apt picks the real gs package rather than gs-esp in desktop
[10:51] <Kamion> seb128: actually it's just gnome-gv and ijsgimpprint I think, so I'll do it if you want ...
[10:52] <seb128> Kamion: yes please
[10:52] <Alessio> sorry, who can i ask for add a link to this page http://www.ubuntulinux.org/community/forums/ ??
[10:56] <Kamion> elmo: has hoary-changes been missing mails? I don't see anything about fontconfig 2.2.3-4ubuntu{6,7}
[10:58] <fabbione> Kamion: neither do i
[11:39] <dholbach> smurfix: i feel with you very much
[12:04] <ogra> pitti, now
[12:05] <kiko> HERE COMES NOISE
[12:05] <ogra> morning
[12:05] <mpt_switzerland> rah rah rah
[12:05] <kiko> morning ogra 
[12:05] <pitti> ogra: nevermind, already dealt with
[12:05] <kiko> how are these ubuntu people doing?
[12:05] <pitti> ogra: Morning :-)
[12:05] <ogra> :)
[12:05] <pitti> kiko: one is banging his head with flawed PHP fixes
[12:05] <kiko> pitti, just drop php from main. next problem?
[12:05] <pitti> kiko: it seems that I just discovered that a security fix introduced another security problem *sigh*
[12:06] <pitti> kiko: seconded :-)
[12:06] <ogra> kiko, +++
[12:08] <seb128> mvo: nice work on the icon when apt is working :)
[12:09] <dholbach> seb128: yeah
[12:09] <dholbach> morning ogra
[12:09] <mvo> seb128: thanks. it's not perfect because it can't track the real apt lockfile (no permissions), but it tries hard to get it right
[12:09] <ogra> hey dholbach 
[12:10] <seb128> mvo: seems to work fine here
[12:19] <kiko> pitti, that's what you get when blackhats supply your security fixes 
[12:19] <Kamion> so who knows anything about mozilla-locale-{da,it,ptbr}?
[12:20] <Kamion> looks like they need to be updated for new mozilla-browser
[12:20] <Kamion> likewise mozilla-thunderbird-locale-nb
[12:21] <pitti> kiko: the flaw is not in _my_ fix, but in upstream's :-)
[12:21] <pitti> kiko: I just reviewed it before I applied it, the one I will release is correct :-)
[12:21] <kiko> that's what I said! :)
[12:21] <pitti> Kamion: I already asked and mailed elmo about it
[12:22] <Kamion> elmo: xfree86-driver-fglrx and xfree86-driver-fglrx-dev should be demoted to multiverse, I think
[12:22] <Kamion> since xfree86 is in universe
[12:23] <pitti> Kamion: it's a matter of sync
[12:23] <Kamion> pitti: ok, thanks
[12:29] <Kamion> GNOME folks: shouldn't #7741 be the responsibility of the theme, or something?
[12:30] <Kamion> and is there a better way to get that UTF-8 character than sticking UTF-8 characters in the source code?
[12:31] <seb128> Kamion: http://bugzilla.ubuntu.com/show_bug.cgi?id=7714
[12:31] <d3vic3> fabbione, ping 
[12:31] <seb128> Kamion: I'm thinking to do that, but not sure if that's an issue for some fonts ...
[12:32] <fabbione> d3vic3: pong
[12:32] <d3vic3> fabbione, hdparm can it be uploaded ?
[12:32] <Kamion> seb128: ok, shall I duplicate #7741 onto #7714 then? the font issues would be the same either way
[12:32] <seb128> yep
[12:33] <fabbione> d3vic3: it depends what are the changes
[12:33] <Kamion> seb128: done, thanks
[12:33] <seb128> np
[12:33] <d3vic3> fabbione, Ubuntu #7829, pretty small change 
[12:34] <ogra> crimsun around ?
[12:35] <crimsun> ogra: pong
[12:35] <ogra> yeah
[12:35] <fabbione> d3vic3: yes. it sounds reasonable
[12:35] <d3vic3> fabbione@ubuntu.com ?
[12:35] <fabbione> yes
[12:35] <d3vic3> incoming 
[12:36] <ogra> crimsun, what do you think if we both sit down after release and wirte up a explanation how sound works (for dummies) on the wiki, people dont seem to understand it at all wich causes a lot of frustration
[12:36] <crimsun> ogra: absolutely
[12:36] <ogra> crimsun, great :-)
[12:37] <crimsun> ogra: we'll discuss hardware->alsa->esd->gst->application, correct?
[12:37] <ogra> yop
[12:37] <crimsun> ok
[12:37] <ogra> explaining the layer architecture and why its silly to switch from esd to alsa in ubuntu :)
[12:38] <ogra> if they understand that they will not fuck it up (i hope)
[12:44] <ogra> oops
[12:44] <Kamion> install CD builds for today running now
[12:45] <dholbach> fabbione: shall i remind you of http://ubuntu.gplan.info/fuse/ today or tomorrow? :-)
[12:45] <dholbach> fabbione: there's actually no haste, just wanted to inform you, like you requested :-)
[12:45] <fabbione> dholbach: tomorrow is saturday :-) do you want me to free the dark side of the force from my wife?
[12:46] <dholbach> fabbione: no... i'll ask on monday again ;-)
[12:46] <fabbione> i am looking at it now
[12:46] <dholbach> fabbione: don't want to unleash any domestic problems ;-)
[12:47] <fabbione> ah but that's a brand new upstream version...
[12:47] <dholbach> fabbione: it's needed for a newer python-fuse needed by gmailfs
[12:47] <fabbione> amen
[12:47] <seb128> pitti: here ?
[12:47] <dholbach> dunno how gmailfs made it into the archive in the first place
[12:47] <crimsun> gmailfs is uninstallable right now because of it
[12:48] <dholbach> yes
[12:48] <pitti> seb128: yep
[12:48] <dholbach> fabbione: i don't need it either
[12:48] <seb128> pitti: do you know what version of the gnome-control-center mo file is in the language-pack-da ?
[12:49] <crimsun> I don't have an acct myself, but I am supposed to transition it to python 2.4 ;-)
[12:49] <dholbach> fabbione: but that's the unfortunate duty of a MOTU
[12:49] <seb128> pitti: https://bugzilla.ubuntu.com/show_bug.cgi?id=7836, he's right, that's fine with my build and the langpack screw it
[12:49] <tseng> dholbach: is gmailfs still have suid?
[12:49] <fabbione> yes.. i am checking...
[12:50] <pitti> seb128: it should be 1:2.10.0-0ubunu1
[12:50] <pitti> seb128: s/unu/untu/
[12:50] <tseng> dholbach: when it first came out a friend of mine reviewed it and was able to mount/remount anything as a normal user. id be interested to hear if thats still the case
[12:50] <dholbach> tseng: i can't tell, crimsun is to blame, whatever gmailfs does wrong
[12:50] <dholbach> ;-)
[12:50] <crimsun> oh my
[12:50] <crimsun> ;-P
[12:51] <seb128> pitti: 
[12:51] <seb128> -rw-r--r--  1 root root 58475 2005-03-08 02:45 /usr/share/locale/da/LC_MESSAGES/control-center-2.0.mo
[12:51] <seb128> that's the 2.10.0 one
[12:52] <seb128> -rw-r--r--  1 root root 60131 2005-03-10 15:13 /usr/share/locale-langpack/da/LC_MESSAGES/control-center-2.0.mo
[12:52] <seb128> that's the langpack one
[12:52] <dholbach> tseng, crimsun, fabbione: i consider gmailfs to be a hack of the category look-it-works--geeky-eh?
[12:52] <fabbione> dholbach: checking now... i need sometime to be confident in it
[12:52] <pitti> seb128: 
[12:52] <dholbach> fabbione: take your time... it's not that urgent
[12:52] <pitti> pitti@rookery:/srv/language-packs.ubuntu.com/sources-base/language-pack-da-base/data/da/LC_MESSAGES $ md5sum control-center-2.0.po
[12:52] <pitti> e057b4b3abc9b3221d0bfe4ba7b81df0  control-center-2.0.po
[12:52] <crimsun> fabbione: many thanks.
[12:53] <pitti> seb128: indeed:
[12:53] <pitti> $ grep -A 3 "Separate _group" control-center-2.0.po
[12:53] <pitti> msgid "Separate _group for each window"
[12:53] <pitti> msgstr ""
[12:53] <seb128> $ grep -A 3 "Separate _group" da.po
[12:54] <seb128> msgid "Separate _group for each window"
[12:54] <seb128> msgstr "Separat _gruppe for hvert vindue"
[12:54] <seb128> WTF ?
[12:54] <pitti> seb128: this is from which version?
[12:54] <seb128> control-center-2.10.0/po
[12:54] <seb128> I've just apt-get source the current archive one
[12:55] <seb128> 1:2.10.0-0ubuntu1
[12:55] <pitti> seb128: langpack contains rookery:/home/lamont/public_html/translations/20050308/control-center_1:2.10.0-0ubuntu1_translations.tar.gz
[12:55] <seb128> apt-get source on this version give a po/da.po with the translation
[12:55] <seb128> and it works here with my non-stripped package before installing the langpack
[12:56] <pitti> seb128: hmm, the tarball contains the translation. odd...
[12:56] <seb128> should I reassing the bug to you ?
[12:56] <pitti> seb128: yeah, please do
[12:56] <seb128> thanks
[12:56] <pitti> seb128: the tarball is correct
[12:57] <seb128> I know :p
[12:58] <seb128> apt-get source gives a po/da.po with the translation as said before
[12:58] <seb128> so the source in the archive is correct
[12:58] <pitti> seb128: argh, I know what's wrong
[12:58] <seb128> are you sure you have this version in the language-pack ?
[12:59] <pitti> seb128: the build accidentially used the version from 2.9.something
[12:59] <seb128> ah ? what is it ?
[12:59] <seb128> how ?
[12:59] <pitti> seb128: my build scripts don't take epochs into account
[12:59] <seb128> oh
[12:59] <pitti> seb128: so it regarded 2.9 as newer than 2.10
[12:59] <seb128> but the epoch is not new
[12:59] <pitti> seb128: yeah, but somehow it was renamed s/1:2/1.2/
[12:59] <pitti> seb128: was just an accident
[01:00] <seb128> k
[01:00] <pitti> seb128: I removed the broken tarball
[01:00] <seb128> thanks
[01:01] <seb128> jdub: dude, the bug is to close, the artwork part is #7715
[01:01] <seb128> jdub: or dup it :)
[01:02] <jdub> seb128: oh
[01:03] <mvo> jdub: any news from the update-notification icon?
[01:04] <seb128> an from the desktop files for g-a-i ? :)
[01:04] <jdub> mvo: i hopefully have it in a tarball that arrived today - i will check :)
[01:04] <jdub> seb128: had meeting today :|
[01:05] <seb128> I can work on that if you want
[01:05] <fabbione> dholbach: fuse looks good.
[01:05] <dholbach> fabbione: rocking
[01:06] <mvo> jdub: great, thanks
[01:06] <dholbach> fabbione: thanks alot
[01:06] <fabbione> no problem
[01:06] <mjg59> Hrm. Why is ppp being odd?
[01:06] <crimsun> fabbione: thanks!
[01:07] <mjg59> It's not reading back any of the things it's expecting, but if I cat the modem node I can see the modem sending them back
[01:08] <mjg59> And minicom seems very broken
[01:09] <fabbione> i got the problem with minicom but i didn't bother to check it
[01:09] <fabbione> since it's not the first time that does that
[01:10] <medwards_> Should xvncviewer, xutils, and rsync suggest openssh-client, openssh-server, and openssh-client respectively instead of ssh?
[01:10] <mjg59> If I do echo ATDT0845blah >/dev/modem, it dials and connects
[01:11] <mjg59> If I try to use ppp, it doesn't read any of the stuff it expects
[01:11] <mjg59> Hm. Nor does wvdial.
[01:11] <mjg59> Something wrong here.
[01:11] <fabbione> could it be related to the ppp security fix?
[01:12] <mjg59> Dunno. Trying some debug now.
[01:12] <fabbione> that would really make my day
[01:12] <fabbione> btw.. it seems that the acpi change doesn't break the ABI
[01:12] <mjg59> Cool
[01:12] <crimsun> medwards_: xutils-> -server seems a bit odd, but I suppose to maintain the Suggests, both openssh-server and openssh-client would be present
[01:13] <fabbione> no, it doesn't
[01:13] <Kamion> medwards_: rsync for one might well want both
[01:13] <Kamion> seeing as rsync runs at both client and server end
[01:14] <thom> Kamion: did mdz tell you about the "fun" he was having trying to push kubuntu as a release?
[01:14] <Kamion> medwards_: I'd prefer people to use the ssh-client and ssh-server virtual packages, unless they *really* need openssh in particular
[01:14] <Kamion> thom: nope
[01:14] <mjg59> fabbione: Can you add a patch to add ICH6 PCI ids to the intel8x0m driver?
[01:14] <kiko> thom, what fun?
[01:14] <Kamion> thom: I can imagine though - I generally end up hacking publish-release for nearly every release I do
[01:15] <Kamion> thom: he said something worrying about md5sum mismatches?
[01:15] <fabbione> mjg59: i think so...
[01:15] <medwards_> point being, they all suggest the transitional package ssh.
[01:15] <thom> Kamion: see query
[01:16] <fabbione> mjg59: usual story.. send me the patch :-)
[01:16] <fabbione> s/patch/Dpatch/
[01:16] <medwards_> which depends on the openssh server+client.
[01:16] <thom> and yeah, the amd64 preview iso in simple/.pool had changed a few hours before
[01:17] <mjg59> Sent
[01:17] <mjg59> How odd. Restarting slmodemd has fixed it.
[01:17] <mjg59> So, uh, rock. Working modem.
[01:18] <fabbione> mjg59: patch is ok..
[01:20] <pitti> daniels: ping?
[01:21] <medwards_> not exactly RC, I suppose.
[01:22] <medwards_> And should I be compiling with g++ 3.3 or 3.4 when using cppunit and libboost?
[01:23] <medwards_> hmm, nevermind ... those aren't in ubuntu.
[01:23] <medwards_> But generally, are C++ libs in ubuntu compiled with one or the other or a mix?
[01:24] <pitti> Kamion: time for an USN review?
[01:25] <pitti> Kamion: s/^/do you have/
[01:25] <Kamion> medwards_: ssh> yeah, I know
[01:25] <Kamion> pitti: sure
[01:26] <seb128> lamont: here ?
[01:26] <medwards_> BTW, it's really nice to start dselect on a livecd and not have to undo 50 defaults before I can do anything useful.
[01:26] <doko> medwards_: g++ 3.3, 3.4 has an incompatible ABI.
[01:27] <seb128> lamont: if you are around just wondering about the gst-plugins0.8 build
[01:27] <medwards_> doko: right.  I'm hoping that linking against shared libraries written in C doesn't create problems, and that hoary is systematically compiled with one or the other.
[01:28] <Kamion> yes, we haven't switched C++ ABI in hoary or anything
[01:28] <medwards_> g++ 3.4 is in for bootstrapping breezy?
[01:29] <Kamion> Debian has g++ 3.4 too
[01:29] <Kamion> as did warty
[01:29] <Kamion>    g++-3.4 |    3.4.3-6 |       testing | i386, powerpc
[01:29] <Kamion>    g++-3.4 |   3.4.3-12 |      unstable | i386, powerpc
[01:29] <Kamion>    g++-3.4 | 3.4.2-2ubuntu1 |         warty | amd64, i386, powerpc
[01:29] <Kamion>    g++-3.4 | 3.4.3-9ubuntu3 |         hoary | amd64, i386, powerpc
[01:29] <trukulo> xfree86 4.5 released, just to inform (is not really important as we use xorg)
[01:29] <medwards_> check.  b-d on g++ still gets 3.3 though, right?
[01:29] <Kamion> yes
[01:30] <Kamion> g++ defines the default
[01:30] <Kamion> when we switch to 3.4/4.0/whatever we'll have to go around changing C++ library package names
[01:30] <Kamion> Debian's done this before so the path is well-trodden, if long
[01:30] <medwards_> So is there another c102-like agony in our near future?
[01:31] <Kamion> at some point we'll have to have c103 I believe
[01:31] <medwards_> any reason modutils isn't on the livecd?
[01:31] <medwards_> 2.4 kernels only, I suppose.
[01:32] <jdub> seb128: does red hat use krb5 or heimdal?
[01:32] <crimsun> it's not needed for 2.6
[01:32] <doko> medwards_: no, a c1003 agony ;)
[01:32] <seb128> jdub: I've no idea, why ?
[01:32] <medwards_> crimsun: just beat you to it.  :)
[01:32] <Kamion> smurfix: kbd-chooser 1.09ubuntu9 has BitKeeper metadata in it again
[01:33] <jdub> seb128: worth knowing considering our choice ;)
[01:33] <Kamion> smurfix: reviewing debdiff between source packages before uploading is a good way to avoid that, I find
[01:35] <seb128> jdub: <jpr> suse uses heimdal
 evo should work with either
 do you know for redhat ?
 mit
[01:36] <seb128> jdub: that's it
[01:36] <jdub> "hmm" :)
[01:36] <seb128> what ?
[01:36] <seb128> we have the issue than kde/firefox/... use krb5, and openoffice.org2 use kde and evo and BOOM
[01:37] <seb128> and according to jbailey krb5 is nice :)
[01:37] <medwards_> doko: just for kicks, I'll try to massage cppunit to generate a c103 binary.
[01:37] <thom> anyone see any reason not to remove powernowd from ia64 seeds? 
[01:38] <medwards_> I see the hoary livecd contains perl but not python-2.3.  :-/
[01:38] <jbailey> seb128: =)
[01:38] <Kamion> medwards_: it'll have python2.4
[01:38] <seb128> hey jbailey :)
[01:38] <crimsun> medwards_: python2.4 please
[01:39] <medwards_> optimists, are we?
[01:39] <doko> optimists look for python 2.4.1 :)
[01:40] <crimsun> 2.4.1a0. c'mon doko, you're slackin'! ;-P
[01:40] <medwards_> I really ought to get around to cleaning up my multithreaded hotspot profiler hack and pitching it upstream.
[01:40] <doko> crimsun: 2.4.1 release candidate 2
[01:41] <medwards_> s/hotspot/hotshot/
[01:41] <fabbione> thom: i am not completely sure that 7788 is kernel...
[01:42] <medwards_> I was hoping to finish the thread-aware time sampling first (gettimeofday bites)
[01:42] <pitti> Keybuk, Kamion: dpkg doesn't use debhelper, so it wasn't stripped. any objection against an upload which adds pkgstriptranslations to debian/rules?
[01:43] <pitti> Kamion: same for coreutils (4.3 MB worth of translations), and sooner or later we have to fork the package anyway for pkgstriptranslations
[01:43] <thom> fabbione: ah, no, it's not
[01:44] <fabbione> thom: ok :-)
[01:44] <Kamion> pitti: note that stripping dpkg means losing translations of stuff displayed to the user while doing the second stage install, since we don't have the ability to put a progress bar over the top of that yet
[01:45] <Kamion> coreutils is probably ok
[01:45] <pitti> Kamion: hmm, right
[01:45] <pitti> Kamion: so we defer dpkg, but I can do coreutils?
[01:45] <Kamion> I'd love to fix that but I just haven't quite managed to convince debconf's newt frontend into displaying progress bars yet
[01:46] <Kamion> pitti: yeah, think so
[01:46] <pitti> ok
[01:48] <pitti> Kamion: oh, coreutils does use debhelper, so it's a no-change upload (-0ubuntu0)
[01:50] <Kamion> ok
[01:51] <pitti> Kamion: btw, I compiled a list of unstripped packages sorted by translation size here: http://people.ubuntu.com/~pitti/unstripped-hoary-main.txt
[01:51] <pitti> Kamion: mdz agreed to do uploads for the biggest couple of packages which are shipped on CD
[01:53] <Kamion> he agreed to do them, or he agreed to them being done? :)
[01:54] <pitti> Kamion: the latter :-)
[01:54] <pitti> "I think we could get a majority of the benefit by rebuilding these packages:"
[01:54] <fabbione> yay for doko and ooo
[01:54] <fabbione> doko: sparc hates you
[01:54] <Kamion> pitti: I'd appreciate iso-codes staying unstripped please
[01:55] <pitti> Kamion: of course
[01:55] <Keybuk> pitti: for now, I'd keep the dpkg ones -- it's kinda a very core component; but then I guess we've stripped the apt ones?
[01:55] <pitti> Kamion: this one should even be blacklisted
[01:55] <Kamion> pitti: definitely
[01:55] <pitti> Keybuk: yeah, we already agreed to defer this until we have nice progress bars on installation :-)
[01:55] <Kamion> half the point of iso-codes is to provide those translations, and it appears in build-dependencies of packages that use the translations in their build processes
[01:56] <medwards_> gnome-terminal line wrapping is b0rked.
[01:56] <medwards_> (on array-7)
[01:57] <medwards_> Most likely to be gnome-terminal, readline, or bash, I wonder?  Or maybe termtype?
[02:00] <doko> fabbione: did ooo build on sparc/hoary before?
[02:00] <fabbione> doko: yes. always
[02:00] <fabbione> and it keeps building
[02:00] <medwards_> OK, only happens when the previous cmd's stdout doesn't end in newline, resulting in misplaced prompt.  bash, probably.
[02:00] <doko> hmm, powerpc failed :(
[02:00] <fabbione> ask lamont to kick it back
[02:00] <fabbione> ppc sometimes dies for no reasons
[02:01] <doko> ok
[02:01] <Kamion> uh, maybe check why it died first?
[02:01] <fabbione> [   ]  openoffice.org-bin_1.1.3-2.3ubuntu9_sparc.deb 
[02:01] <fabbione> Kamion: that too
[02:01] <doko> Making: ../unxlngppc.pro/slb/set.lib
[02:01] <doko> nm: 'Illegal': No such file
[02:01] <doko> nm: 'instruction': No such file
[02:01] <fabbione> but i was already doing it
[02:02] <Kamion> right, sounds like the usual powerpc thing then
[02:02] <fabbione> yup
[02:15] <mjg59> Hrm. Is someone able to add a patch to i810switch?
[02:15] <mjg59> (It's in universe)
[02:16] <crimsun> sure, what does it need to do?
[02:16] <dholbach> mjg59: i can apply the patch and upload it for you, but i 1) most likely won't understand, what it is about, 2) can't even test it with my amd64
[02:17] <mjg59> http://www.codon.org.uk/~mjg59/tmp/i810switch.diff
[02:17] <mjg59> It just adds an extra PCI ID
[02:18] <medwards_> hmm.  mount did not auto-detect that /dev/hda4 contained an ext3 fs.
[02:18] <dholbach> crimsun: shall i? do you want to?
[02:19] <medwards_> mounted it as ext2.  wonder what I have to do to fix that.
[02:19] <crimsun> dholbach: I'm kinda busy atm, so if you don't mind, you can take it
[02:19] <dholbach> crimsun: right
[02:19] <crimsun> thanks
[02:19] <zul> mornin
[02:20] <dholbach> hi zul
[02:20] <zul> hey dholbach 
[02:22] <pitti> Kamion: re m-thunderbird-locale-nb, I removed it from l-support-nb long ago, but this needs to be demoted to universe (or removed)
[02:22] <pitti> Kamion: it's not seeded and nothing depends on it 
[02:30] <Kamion> pitti: ok
[02:37] <medwards_> oy.  apt-get build-dep wants to pull in build-essential, including g++ 3.3.  And cppunit wants libqt3-mt-dev and hence libqt3c102-mt.  Don't think I'll be charging on to c103 right now.
[02:41] <Kamion> er, yeah, naturally current build-deps are c102ish
[02:42] <Kamion> and you can't start a C++ transition with packages high up in the dependency graph; you need to start from the bottom
[02:42] <medwards_> Kamion: yes, I've modified build-essential before to get the g++ I wanted in a pbuilder run.
[02:42] <medwards_> Kamion: I didn't expect cppunit to build-depend on Qt.
[02:43] <Kamion> http://lists.debian.org/debian-devel-announce/2003/01/msg00002.html
[02:43] <medwards_> (I don't generally run the GUI front end, which winds up in a separate binary package)
[02:44] <Kamion> anyway it's not like going through the transition buys you much
[02:44] <Kamion> from your POV anyway
[02:45] <medwards_> right, I was actually thinking of sneaking a g++ 3.4 build of this ugly bugger through the day job's QA system as a compiler test.  :-)
[02:46] <medwards_> ("this ugly bugger" being a very thread-intensive network management client/server app)
[02:50] <medwards_> Kamion: I will still do a c103 build, but I'm going to have to cut the Qt out of the cppunit build first, so it'll wait until the next cycle.
[02:51] <ogra> amu, Riddell, haggai, just had a look at kubuntu, its marvelous, really rocking (no, i wont switch ;) )
[02:52] <haggai> ogra: aw, switch switch switch :)
[02:52] <jdub> ogra: where is your national pride? :)
[02:52] <amu> ogra: hehehe
[02:52] <ogra> *g*
[02:52] <amu> we'll see :)
[02:52] <medwards_> (cppunit is otherwise on the bottom of the graph, and boost is a template library)
[02:53] <medwards_> OTOH, I might jump straight to g++-4.0, if it's even remotely stable.
[02:54] <ogra> one thing btw, how am i supposed to unlock the screen after i lockaed it ?
[02:54] <ogra> -a
[02:54] <thom> oh, man. FUCK YOU VERY MUCH, FIREFOX
[02:55] <Treenaks> thom: what happen?
[02:55] <amu> ogra: the kubuntu one ? 
[02:55] <ogra> yep
[02:55] <thom> i can reproduce #7552 with my french test user, but not with my english main user
[02:55] <ogra> amu, is there a password ? 
[02:55] <amu> ogra: it could be a bug, nice testcase 
[02:55] <thom> i do hope this is not a locales bug
[02:55] <thom> if it is i shall cry
[02:56] <amu> ogra: there's no pass, try ubuntu/ubuntu
[02:56] <ogra> amu, tried, didnt work
[02:56] <ogra> amu, but it might occur in the gnome version as well, just rsyncing my ubuntu live image
[02:56] <amu> how you locked your screen? from menu?
[02:56] <ogra> yep
[02:56] <ogra> lock session
[02:57] <pitti> ogra: now that you say it, I remember seeing this bug, too
[02:57] <amu> ogra: could you add a bug
[02:57] <ogra> pitti, ku- or ubuntu ?
[02:57] <pitti> ogra: !k
[02:57] <ogra> ok
[02:57] <pitti> ogra: normal gnome ubuntu
[02:58] <ogra> so it occurs in both 
[02:58] <pitti> ogra: yeah, why not?
[02:58] <pitti> ogra: the ubuntu user does not have a password
[02:58] <ogra> dunno
[02:58] <pitti> ogra: but screensaver does not accept an empty password
[02:58] <ogra> but then screenlocking should be disabled completely....
[02:59] <ogra> since it doesnt make much sense to lock yourself out completely ;) 
[02:59] <ogra> hey, dholbach gets famous :) http://lwn.net/Articles/125666/
[02:59] <amu> ogra: right but how you'll do it, till it's the same packagebase? 
[03:00] <ogra> amu, disable screenlocking in the package for the live seed would be a guess.....
[03:00] <ogra> s/seed/seeds
[03:01] <jdub> dholbach: that's an awesome report, too :-)
[03:01] <jdub> dholbach: that should totally go to ubuntu-news :)
[03:01] <ogra> funny that lwn needed 17 days to put it up :)
[03:01] <amu> ogra: apt-cache rdepends xscreensaver
[03:02] <dholbach> jdub: i plan doing it every month, to let everyone FEEL the motu love :-)
[03:02] <ogra> yeah
[03:03] <ogra> amu, i see...
[03:03] <amu> ogra: a userfriendly solution could be checking if we are on a liveCD, before locking the screen, than ask for a userpass
[03:03] <ogra> yeah
[03:03] <ogra> great idea
[03:06] <amu> hmm dialog isnt install by default  
[03:06] <zul> eww...dialog
[03:08] <amu> zul: ok, kdialog :) 
[03:08] <ogra> amu, #7843
[03:08] <zul> oh thats better
[03:10] <amu> ogra: could be, modify the desktop file and run a script which checks if we are on a livesys, if yes, (k)dialog ask user to set a password, if we are on a real sys nothing happens. Is there something wrong with it?   
[03:11] <ogra> amu, sounds ok....
[03:11] <amu> ignoring this by default, educate people to set a password, is another solution
[03:12] <ogra> amu, you mean setting one before session ? (i.e. on bootup)
[03:14] <amu> ogra: yep, this would be a saver, faster, cleaner, better way 
[03:14] <Kamion> the password is disabled nowadays, AFAIK
[03:14] <Kamion> so it's easier than that: just check whether the password is disabled, rather than having to check if you're on a live CD
[03:15] <amu> Kamion: see orga's bug, it looks not. Oh yes, that would be easier.     
[03:15] <doko> d3vic3: ping?
[03:16] <ogra> Kamion, but setting one right before locking would be a pretty cool faeture, else you cant lock at all if you are on the live CD
[03:16] <Kamion> ogra: sure, I don't see where I disagreed with that?
[03:16] <Kamion> ogra: I was *agreeing* with you
[03:16] <ogra> :)
[03:17] <pitti> Keybuk: here?
[03:17] <Keybuk> yup
[03:17] <Kamion> do i386/amd64 installations with JFS / fail for anyone here?
[03:17] <Kamion> (#7666, can't reproduce)
[03:17] <thom> Kamion: yeah, it did for me with preview; i can try again now
[03:18] <thom> well, nowish
[03:18] <thom> seb128: about? 
[03:18] <seb128> yep ?
[03:18] <Kamion> thom: thanks
[03:18] <Kamion> could be a race or something
[03:18] <thom> seb128: help?! #7552 seems to be blowing up in gtkfilechooserdefault.c; but i dunno why
[03:19] <seb128> is that the firefox download stuff ?
[03:19] <seb128> oh, no, a different one :)
[03:21] <Kamion> hmm, interesting idea, time zero-question kickstart installations with each of the major filesystems
[03:22] <seb128> thom: http://bugzilla.gnome.org/show_bug.cgi?id=170755 has the same bt
[03:23] <thom> yeah, that looks identical
[03:23] <seb128> iz gtk boog
[03:23] <d3vic3> doko, pong 
[03:23] <seb128> :(
[03:23] <thom> hm, fun
[03:24] <thom> seb128: care to add yourself to the cc list?
[03:24] <pitti> jdub: it seems that I'm the moderator of ubuntu-hardened
[03:24] <pitti> jdub: however, I never received a moderator password
[03:24] <pitti> jdub: can you please gpg-mail it to me?
[03:24] <seb128> thom: yep, I'll probably reassign to gtk
[03:24] <thom> ok
[03:24] <thom> i'll see if i can blame the download problem on gtk too :-)
[03:25] <pitti> thom: which download problem?
[03:25] <thom> pitti: 7711
[03:25] <smurfix> Kamion: Bleh. Forgot. Again. :-( You may give me the evil eye in Sydney.
[03:25] <jordi> morning
[03:25] <jordi> well
[03:25] <jordi> for me
[03:26] <thom> jordi: even spain is usually not this slack!
[03:26] <Kamion> smurfix: heh. :) was trying to track down that debian-installer/keymap thing by debdiff ...
[03:27] <jordi> thom: shuddup, this hangover is not fun at all :)
[03:27] <thom> jordi: over enthusiastic St Patrick's Day partying? :P
[03:28] <medwards_> never seen this before:  "Package expat is not available, but is referred to by another package. ..."
[03:28] <jordi> thom: I never drink rum, whisky or shit like that, but yesterday I had a bit
[03:28] <jordi> + the usual beers and wine
[03:29] <jordi> thom: nope, it was during the Alternative Falles celebration. We burnt our falla two days early. :)
[03:31] <medwards_> expat seems to be in universe, but some package in main must refer to it in some way.  Probably libexpat1, which seems to be in main and built from the expat source.
[03:31] <dholbach> ok... i'm off - see you later
[03:33] <medwards_> I'm a little surprised to find debian packages in ubuntu main that don't seem to have been recompiled (libexpat1, for one).
[03:35] <thom> they'll all have been recompiled; if they have debian version numbers it means we've not modified them
[03:35] <jdub> medwards_: everything is rebuilt
[03:37] <medwards_> jdub: without any indication in the changelog?
[03:38] <jdub> medwards_: that's right
[03:38] <jdub> medwards_: everything in every repository is rebuilt
[03:38] <medwards_> jdub: that's certainly a good thing.  :)
[03:40] <medwards_> jdub: but it makes it a little trickier for the debian maintainer if bug reports come in against a package that ubuntu rebuilt.  How does that interact with binary NMUs?
[03:41] <Nafallo> is it just me or is cdimage.u.c slow as hell today?
[03:42] <medwards_> hmm, looks like reportbug would send a report to ubuntu-users instead of the debian maintainer.
[03:43] <Nafallo> medwards_: good then :-)
[03:44] <medwards_> but I assume that's a feature of ubuntu's reportbug.  If debian users start adding ubuntu to their sources.list, some odd things may happen.
[03:45] <Nafallo> medwards_: therefore that action isn't supported AFAIK :-).
[03:48] <medwards_> I try to add a changelog entry even when just rebuilding locally, to avoid bug reports going to the wrong place.  I'm envisioning a situation where I have, say, ubuntu's libexpat1 but debian's expat, same version number, dep skew.  How will I even know?
[03:49] <kiko> so
[03:49] <kiko> doko?
[03:50] <doko> doko: kiko!
[03:50] <medwards_> Maybe reportbug should add the md5sum of the md5sums file for each package in the depends.  :-/
[03:51] <kiko> doko@
[03:51] <kiko> ack
[03:51] <kiko> doko!
[03:51] <doko> kiko?
[03:51] <kiko> doko, we're getting reports on #lp-dev of SRE.py conflicting in the latest hoary apt-get update
[03:52] <kiko> doko, I can get you a traceback if that's helpful -- has any important python module or package changed from yesterday to today?
[03:53] <doko> not that I am aware of. which package is SRE.py in?
[03:53] <kiko> it's part of standard python
[03:53] <doko> ahh, sre.py ...
[03:53] <kiko> and SRE is a module with symbols, IIRC
[03:54] <kiko> I'll get you the traceback, or rather, carlos will in a minute
[03:54] <doko> ok, I'll have a look then
[03:55] <kiko> doko, http://www.rafb.net/paste/results/BlrELh33.html
[03:56] <doko> hmm, interesting. maybe throwing away the compiled modules helps? 
[03:57] <kiko> carlos, ping?
[03:57] <carlos> kiko: pong
[03:57] <carlos> doing it
[03:57] <carlos> ohh, to remove compiled modules?
[03:57] <carlos> doko: should I remove all .pyc?
[03:57] <kiko> carlos, doko: can you guys coordinate the fix? I need to take a call
[03:58] <jordi> kiko!
[03:58] <kiko> hombre!
[03:58] <jordi> hombre!
[03:58] <jordi> qu pasa!
[03:58] <kiko> qu, mucho trabajo carajo
[03:58] <doko> maybe first try to see, which pyc files are wrong: magic = string.join(["\\x%.2x" % ord(c) for c in imp.get_magic()] ,"")
[03:59] <doko> these should be the first bytes in the .pyc file.
[03:59] <jordi> kiko: dude I haven't been training at all.
[03:59] <jordi> kiko: I've run 3 times since we last met.
[03:59] <carlos> doko: sorry, I did the removal already ...
[03:59] <carlos> doko: I'm too fast :-(
[03:59] <kiko> jordi, I've run dozens of times, jesus, you need to get there
[03:59] <kiko> doko, carlos: could it not be a pythonpath issue? I suggested running a -v
[04:00] <lamont> morning
[04:00] <pitti> Hi lamont
[04:00] <jordi> kiko: I think my motivation is coming back more or less.
[04:00] <Nafallo> lamont: hi there :-)
[04:00] <jordi> I hope I'll be back soon.
[04:00] <carlos> jordi: is Belen your motivation's name?
[04:00] <doko> kiko: yes, I assume .pyc files mixed from 2.3 and 2.4 were found.
[04:00] <jordi> carlos: no
[04:01] <carlos> doko: same problem after .pyc removal
[04:01] <kiko> jordi, plan to run a marathon in this semester, and train for it
[04:01] <jordi> carlos: team mates doing competitions and I only seeing the results in webpages mostly.
[04:01] <kiko> agh
[04:01] <jordi> kiko: nah, marathons are way too much. half, maybe.
[04:01] <jordi> not a full one
[04:02] <jordi> kiko: I want to prepare the olympic triathlons decently.
[04:03] <carlos> bored people...
[04:03] <carlos> :-)
[04:09] <Kamion> medwards_: in practice that's not been a problem; we're building off the same base so we haven't experienced dep skew in anything that matters
[04:10] <Kamion> medwards_: we warn VERY VERY CLEARLY not to mix the Debian and Ubuntu repositories; it confuses apt
[04:10] <smurfix> Hoboy. Trying to burn the PowerPC hoary live-cd from OS X crashes Apple's Disk Utility program.
[04:10] <medwards_> Kamion: OK, was just puzzled by libexpat1 in main and expat in universe.
[04:11] <Kamion> medwards_: I imagine libexpat1 was needed by something in main and expat wasn't, then
[04:12] <Kamion> smurfix: well-known bug in Apple's hdiutil
[04:12] <Kamion> smurfix: I've had both a success report and a failure report with one recent version of Mac OS X, so it doesn't appear to be entirely consistent
[04:12] <Kamion> but clearly a segfault on any input data must be a bug in the software, not the data :)
[04:13] <smurfix> Kamion: Maybe it'd help not to have a slash in the volume name ..?
[04:13] <Kamion> smurfix: does it help if you change that?
[04:14] <pitti> meh, wrong key
[04:14] <Kamion> smurfix: um - what slash?
[04:14] <Kamion> Volume id: Ubuntu 5.04 ppc Bin-1
[04:15] <smurfix> Kamion: I don't know yet, the ibook managed not to boot from the latest livecd I burned (may be CD drive problems)
[04:15] <smurfix> Kamion: Mounting the current image says "Ubuntu/PowerPC_hoary" as the volume name
[04:16] <Kamion> ah, the hfs-volid
[04:16] <medwards_> Kamion: expat pkg consists of one program, xmlwf, 19K in size.  All of the bugs reported against it in debbugs are actually libexpat1 bugs, except one that is probably python.  :)
[04:17] <Kamion> medwards_: nevertheless we only put stuff in main if we need it to be in main
[04:17] <medwards_> Kamion: sure, I just found the results of inspecting the BTS amusing.
[04:18] <Kamion> smurfix: would be surprising though; I'd've thought that : would be more of a problem than / on Macs
[04:20] <Kamion> smurfix: there's only one occurrence of "Ubuntu/PowerPC_hoary" in the binary, though, so you could hex-edit it to something else and try :)
[04:21] <Kamion> smurfix: hm, actually, no, there are quite a few towards the end. Want me to do you a CD with a different volume id?
[04:30] <zul> mjg59: ping 
[04:30] <medwards_> Last upload that fixed an expat bug was on 2003-03-17.  Might be worth at least touching up the BTS, and maybe actually updating to the might-be-expat-2.0 snapshot from January and looking over upstream's BTS.  Too late for hoary but perhaps not for sarge.
[04:39] <lamont> jdub: did you care that flumotion is ftbfs?
[04:39] <lamont> configure: error: No suitable version of python found
[04:39] <lamont> but it valiantly tried 2.2 all the way down to 1.5 :-)
[04:40] <lamont> and hoary-test gets it's _FIRST_WINNER_!!!!  and the winner iiiiiiissss.......
[04:40] <lamont> graphviz
[04:41] <seb128> lamont, fix gst-plugins0.8
[04:41] <seb128> I want to give it some testing before hoary
[04:41] <lamont> seb128: _I_ can't...
[04:41] <seb128> need elmo ?
[04:41] <lamont> elmo: seb needs the build-deps for gst-plugins0.8 promoted to main....
[04:42] <lamont> seb128: yeah, is archive issue
[04:42] <seb128> usually that works fine, somebody automatically handles such issues :)
[04:43] <seb128> but right
[04:44] <lamont> seb128: probably best thing to do is send elmo email, cc jdub/mdz requesting the promotion.  then jdub/mdz can ack it, and elmo has his history trail;
[04:46] <seb128> right
[04:47] <smurfix> Kamion: Please do
[04:47] <smurfix> Kamion: Place it somewhere rsync'able if possible
[04:49] <mdz> morning
[04:49] <zul> morning mdz
[04:50] <mvo> hey mdz 
[04:50] <seb128> hey mdz 
[04:51] <kiko> morning mdz 
[04:51] <seb128> hey kiko
[04:52] <kiko> hey seb128, how are you
[04:53] <seb128> kiko: fine, and you ?
[04:53] <Kamion> smurfix: I'm just doing a new daily install CD build, that's simplest
[04:53] <sivang> kiko dude ! :)
[04:53] <kiko> seb128, overworked but happy
[04:53] <kiko> hey sivan!
[04:54] <lamont> morning mdz
[04:54] <mvo> Kamion: will that fix the keyboard issue?
[04:54] <Kamion> mvo: not yet
[04:55] <Kamion> mvo: this is to see if changing the volume label stops Disk Utility crashing
[04:55] <mvo> Kamion: ok
[04:55] <mvo> Kamion: just wanted to know if it's worth rsyncing it for testing :)
[04:55] <smurfix> Dear Apple, please let my poor old ibook boot from usb, KTHXBYE
[04:56] <pitti> smurfix: that's probably a matter of yaboot extension, isn't it?
[04:59] <lamont> mdz: what severity do you want the hoary-test ftbfs's to go at?
[04:59] <lamont> major or crit?
[04:59] <smurfix> pitti: That's a matter of extending Apple's OpenFirmware
[04:59] <mdz> lamont: critical at this stage
[04:59] <pitti> Morning mdz
[05:00] <thom> morning mdz
[05:13] <mdz> seb128: my panel has been crashing every night while I am asleep
[05:14] <mdz> seb128: the bt looks the same as the one I submitted to bugzilla
[05:15] <mdz> there is an Error dialog which comes up at the same time, but no text is displayed in it as the process is stopped
[05:15] <mdz> seb128: is there any information I can collect besides the bt?
[05:19] <Kamion> smurfix: the instructions in the d-i manual don't help?
[05:19] <seb128> good question ... I don't really have an idea on the crash and the backtrace is weird. I'll ping one of the gnome-panel upstream to know if he has some ideas and let you know
[05:19] <haggai> guys how to undelete a wiki page?  Someone just deleted Kubuntu
[05:20] <Kamion> still seems to be there?
[05:20] <Kamion> oh, maybe not
[05:22] <haggai> did someone just fix it?
[05:23] <apokryphos> Perhaps deletions doesn't come up under the Recent Changes
[05:24] <Kamion> history shows an all-caps rant about how Kubuntu shouldn't be a separate distribution
[05:25] <haggai> Kamion: yup, followed by a delete
[05:25] <Kamion> ok, bored of bugs about groff hyphens in UTF-8, I give up and will change that to render as ASCII 0x2D
[05:26] <lamont> Kamion: that's using the old sledge hammer!
[05:26] <seb128> somebody is using a ppp (RTC) internet access here ?
[05:27] <lamont> seb128: I _could_ be, if you need me to
[05:27] <lamont> RTC?
[05:27] <seb128> mvo: you have a standard ppp or isdn ?
[05:28] <seb128> lamont: hum, is that french ? a modem access like using the modem applet (or kppp), that's because of #7677 in fact
[05:28] <lamont> ah, ok.
[05:28] <lamont> want I should grab the laptop?
[05:28] <seb128> I know that mvo has worked on some modem stuff
[05:29] <seb128> let's wait if he has an idea first
[05:29] <lamont> ok
[05:29] <seb128> I'll ping you if he doesn't, thanks
[05:30] <lamont> np
[05:48] <mvo> seb128: modem stuff? 
[05:48] <seb128> mvo: #7677
[05:49] <mvo> seb128: looking, thanks
[05:50] <seb128> mvo: I don't know, you have looked on some applets/ppp bugs, haven't you ?
[05:50] <seb128> or that's only for isdn ?
[05:50] <mvo> seb128: yes, I did quite a bit of debugging in this area, modem/ppp too
[05:50] <mvo> seb128: but it's horrible as all the backend stuff is written in perl
[05:50] <seb128> :(
[05:51] <seb128> bah don't, that's useful :)
[05:53] <thom> bah, firefox 1.1 renderer is *so* much better than 1.0
[05:53] <thom> i wish they were on a release schedule useful to us
[05:53] <mvo> seb128: I'll have a look
[05:54] <seb128> mvo: thanks
[05:54] <seb128> thom: what is the schedule for 1.1 ?
[05:55] <thom> PR comes out about the same time hoary does
[05:56] <seb128> they have 6 months between versions too ?
[05:57] <smurfix> Number of screws to unscrew to replace an iBook hard disk: 35
[05:57] <HiddenWolf> seb128: the way mozilla has been going, they have a roadmap they don't keep to. Too few developers.
[05:57] <smurfix> Tune in next installment for anaccount of how many are left lying around afterwards
[05:58] <wasabi_> you needa  p-p-p-powerbook
[05:58] <smurfix> wasabi: it's an old one, and for some reason was slightly cheaper than a big chunk of titanium
[05:59] <smurfix> The cutesy first-gen ibooks are *much* worse than that.
[06:00] <wasabi_> I have an iBook2. One of the white ones.
[06:00] <wasabi_> http://www.p-p-p-powerbook.com/
[06:01] <lamont> smurfix: sounds like you need an extension cable for the hdd - then you could just keep it external. :-)
[06:14] <mjg59> Are there instructions on building a lightly customised CD image?
[06:20] <pitti> doko: regarding the OO.o bibliography bug on ppc, what shall I check?
[06:21] <doko> pitti: if it crashes or not, as mentioned in the report
[06:21] <pitti> doko: oh, it still crashes, as it does for ages now
[06:21] <doko> which version?
[06:22] <sivang> doko: have you uploaded the fixed oowrite version as per #7803 ?
[06:22] <pitti> doko: just posted a bug followup
[06:22] <pitti> doko: 1.1.3-2.3ubuntu9
[06:23] <amu> mjg59: a livesys i guess? 
[06:23] <doko> pitti: that's the old one
[06:23] <pitti> doko: I don't have a newer one
[06:23] <mjg59> amu: No, install
[06:23] <pitti> doko: a new version from today?
[06:24] <doko> pitti: yes, it needs building on powerpc ...
[06:24] <pitti> oh :-)
[06:24] <pitti> doko: no newer version on http://patches.ubuntu.com/~lamont/buildLogs/o/openoffice.org/ either
[06:25] <doko> pitti: I know
[06:25] <pitti> doko: okay, I check when it is available
[06:27] <lamont> openoffice.org:         07:53:47 (6 entries, sigma 02:04:27)
[06:27] <lamont> and we're 2.5 hours into the build
[06:27] <amu> mjg59: saw some of them in d-i, probably the best if you ask Kamion
[06:27] <lamont> "today" may be a bit overagressive, given your tz
[06:28] <pitti> lamont: btw, is this sigma information available for other people than you? 
[06:28] <pitti> lamont: it would be interesting for security builds
[06:29] <Kamion> mjg59: not yet AFAIK, it's on my to-do list to write some
[06:29] <lamont> pitti: it's not even shared between the buildd machines for a given architecture
[06:29] <pitti> lamont: okay
[06:32] <mjg59> Kamion: Ah, ok - I probably need to hack one up next week
[06:32] <doko> sving: it's built, but maybe not yet in the archive
[06:32] <mjg59> (A cd, that is, not the docs)
[06:32] <doko> sivang: it's built, but maybe not yet in the archive
[06:32] <sivang> doko: ok
[06:32] <pitti> sivang: how far is your patch?
[06:33] <pitti> sivang: I have a fixed g-c-m pending, but I would like to avoid uploading twice
[06:33] <pitti> sivang: (I fixed another bug)
[06:33] <sivang> pitti: I see, which bug? I'm actually preparing source packages and a debdiff for you to view
[06:33] <sivang> pitti: (done with coding)
[06:34] <pitti> sivang: #7842
[06:34] <sivang> pitti: (and translations)
[06:34] <Kamion> mjg59: I'll try; I just keep getting snowed under every time it nears the top of the pile :(
[06:34] <pitti> sivang: it'd be nice to finish that today
[06:34] <sivang> pitti: I'm done actually, just preparing stuff for your to review...
[06:34] <pitti> sivang: I just need the debdiff
[06:35] <sivang> pitti: k
[06:36] <sivang> pitti: in about 5 minutes
[06:37] <Mitario> hi everyone
[06:37] <pitti> Hi Mitario 
[06:37] <sivang> hey Mitario 
[06:43] <pitti> sivang: I have to leave in half an hour; shall we do this at Monday?
[06:43] <sivang> pitti: nrealy done, 1 minute :)
[06:43] <pitti> ok :-)
[06:43] <lamont> Kamion: what is a '10' from debconf again?
[06:45] <sivang> pitti: debdiff online, same location
[06:45] <lamont> mvo: you around?
[06:46] <lamont> Setting up fontconfig (2.2.3-4ubuntu5) ...
[06:46] <lamont> dpkg: error processing fontconfig (--configure):
[06:46] <lamont>  subprocess post-installation script returned error exit status 10
[06:46] <sivang> pitti: packages refreshed as well
[06:47] <sivang> pitti: (on my /g-c-m/ folder in muse)
[06:47] <pitti> sivang: already downloaded, inspecting now
[06:47] <pitti> sivang: did you chagne any code?
[06:47] <pitti> change, even
[06:48] <pitti> sivang: I fixed the indentation, btw
[06:48] <Kamion> lamont:         badparams => 10,
[06:48] <sivang> pitti: ah ok, thanks , you edited the patch inline?
[06:48] <Kamion> lamont: try DEBCONF_DEBUG=developer
[06:48] <sivang> pitti: I didn't change any code
[06:48] <lamont> that fits with things.
[06:48] <pitti> sivang: yes
[06:48] <lamont> Kamion: cool
[06:49] <sivang> pitti: just hardcoded messages as you wanted for the sudo message
[06:49] <sivang> pitti: (this were the only modifications I did for the code)
[06:50] <lamont> Kamion: that gave me what I need for the bug.
[06:50] <Kamion> good stuff
[06:51] <Kamion> lamont: don't get too radical, now
[06:55] <sivang> pitti: let me know the verdict :-)
[06:55] <pitti> sivang: oh, patch is nice, I'm adding the .de translations right now
[06:56] <Mitario> mvo, ping?
[06:56] <sivang> pitti: phew finally..:)
[06:57] <doko> sivang: the OOo build is in the archives
[06:58] <lamont> Kamion: all it needed was an || RET=false :-)
[06:59] <sivang> doko: ok, I'm eager to test :)
[06:59] <lamont> Mithrandir: you arouhd?
[06:59] <lamont> dh_testdir
[06:59] <lamont> /bin/bash: error while loading shared libraries: libdl.so.2: cannot open shared object file: No such file or directory
[07:00] <sivang> doko: in incoming?
[07:00] <sivang> doko: (I am using the .de mirror, hasn't hit it yet)
[07:02] <lamont> mdz: I wonder if elmo is waiting for an ack on the sgml2x sync...
[07:04] <seb128> Kamion: is there a place with the files to translate for the installer ? Some screen have no french version and I want to translate them
[07:07] <seb128> pitti: here ?
[07:08] <pitti> seb128: yeah
[07:08] <seb128> I would like to get libshout in main (build requirement for gst-plugins0.8)
[07:08] <seb128> according to mdz "Given a supportability review from Martin, yes."
[07:09] <seb128> so if you can put that on your todolist ... :)
[07:09] <pitti> seb128: I'm taking a look at it ASAP
[07:09] <seb128> thanks
[07:09] <seb128> atm gst-plugins0.8 ftbfs
[07:10] <seb128> there is no real hurry but still good to fix the ftbfs
[07:10] <pitti> seb128: it's ftbfs because of the missing lib in main?
[07:10] <seb128> correct
[07:14] <mvo> bbl
[07:17] <Kamion> seb128: http://people.ubuntu.com/~cjwatson/installer-po/ (may be missing some messages, I need to check it)
[07:20] <pitti> seb128: this thing does not actually contain any mp3 codec stuff?
[07:24] <seb128> pitti: I don't think it does the codec, rather the streaming part
[07:25] <pitti> seb128: do you think #219027 will affect us?
[07:25] <pitti> seb128: (not compiled thread-safe)
[07:26] <seb128> pitti: not sure, I can check if you want
[07:27] <pitti> seb128: otherwise, packaging, debs and bugs are fine for me
[07:27] <seb128> nice, thanks
[07:27] <pitti> seb128: it's a network-related app, so it is security sensitive somewhat
[07:27] <pitti> seb128: I didn't do a source audit :-), but I did not audit KDE either :-)
[07:28] <seb128> if you are not cumfortable with it I can drop this part of the gst-plugins0.8 build
[07:28] <pitti> seb128: oh no, just go ahead
[07:28] <seb128> thanks
[07:28] <pitti> seb128: however, can you please check the thread bug?
[07:29] <seb128> I'm looking the hoary build logs
[07:29] <seb128> checking for the pthreads library -lpthreads... no
[07:29] <seb128> checking whether pthreads work without any flags... no
[07:29] <seb128> hum
[07:29] <seb128> Kamion: thanks for the po files
[07:30] <pitti> bye folks, gotta go now
[07:39] <Mitario> brb testing preview
[07:45] <robertj> If you have refresh rates that work when the autoprobed rates don't, is there some place that info should go so it can be auto-detected for other people?
[07:47] <dholbach> hello
[07:47] <zul> hey dholbach 
[07:47] <dholbach> hi zul 
[07:47] <_d4vid> hi all
[08:06] <medwards_> speaking of autoprobed graphics, any way to get 1920x1200 on this Dell M60?  (screen=1920x1200 works on knoppix)
[08:06] <Treenaks> 1920x1200? is that 17" (I hope so)
[08:08] <ogra> robertj, hey
[08:08] <robertj> heya
[08:09] <ogra> whatsup ?
[08:09] <robertj> Is there a database for monitor refresh rates for stuff that can't be autoprobed?
[08:09] <robertj> Or more particularly, i've got a monitor that seemed to probe fine but gave me a black screen until I googled up some better refresh rates
[08:09] <ogra> heh, not yet, currently my inbox is the collection place
[08:09] <robertj> hehe
[08:10] <robertj> any plans on that front?
[08:10] <ogra> yup
[08:10] <robertj> Wiki page ?
[08:10] <ogra> thats what hwdb is for, but the recieving part is not in place yet
[08:10] <dholbach> php-site, hm? :-)
[08:10] <robertj> (btw, check out http://l3ktr0n.homeip.net if you like muds and dhtml ;)
[08:11] <ogra> robertj, first of all we will need the DB and the data recieving part in place, then we can build a website or something
[08:12] <robertj> oh, I get you
[08:12] <robertj> I totally forgot that the hardware utility had a mechanism for reporting back, or at least the beginnings of one
[08:12] <ogra> but currently i focus on getting the data and converting it to something ueseful
[08:13] <ogra> yup, it should hand you a token after the transfer, so you can look up your own data and point someon who wants to support you there
[08:16] <robertj> I think its particularly nasty how this monitor responded to a probe it didn't support
[08:17] <robertj> How do ADC monitors play well with others?
[08:18] <dholbach> hey Micksa 
[08:18] <dholbach> oops
[08:18] <dholbach> hey Mitario  :-)
[08:18] <Mitario> thanks, hi ;-)
[08:18] <dholbach> how are you?
[08:18] <Mitario> yeah fine :) you?
[08:18] <dholbach> a bit tired, but a black tea will do the trick, i guess :-)
[08:18] <dholbach> so fine myself, thanks
[08:19] <Mitario> hehe ok, nice :-)
[08:22] <medwards_> Treenaks: wide laptop screen.  (And sorry; that question probably belonged on #ubuntu.)
[08:23] <Treenaks> medwards_: yes, but 17" I hope? I'm going to buy a 1680x1050 one and I think it has VERY tiny pixels already
[08:25] <medwards_> and yest, they are very tiny pixels.
[08:25] <Treenaks> I'll stick to the 1680 one :)
[08:26] <medwards_> I often use teeny-weeny-eyestrain-o-vision text anyway; so it makes for smoother eyestrain.
[08:27] <Treenaks> medwards_: LOL
[08:27] <Treenaks> I like the ultra-highres-ness of my Palm
[08:27] <Treenaks> I have a few apps that use absolutely microscopic fonts
[08:29] <medwards_> The M60 seems to get 1600x1200 in array-7 (black stripes left and right) and the default font is just about right.
[08:30] <medwards_> However, it appears just to have locked up hard.  Bugger.
[08:31] <Treenaks> hm, not too promising
[08:31] <medwards_> and things were going so well ...
[08:31] <mroth> medwards_: do you know what the video hardware is on it?
[08:31] <medwards_> NVidia Go700, I think
[08:31] <Treenaks> I'm looking at an OEM version of this one: http://www.uniwill.com/products/performance/259ia3/259ia.php?HL=1&P=1
[08:33] <medwards_> Probably an OOM.  Totally wedged it, though.
[08:34] <medwards_> (I was pushing my luck with apt-get and casper.)
[08:34] <lamont> mdz: question of priorities...
[08:35] <mdz> lamont: shoot
[08:35] <lamont> there are 34 packages (main and universe - didn't do that breakdown) that will fail to remove if defoma happens to be removed before them (the Depend: defoma)...  The fix is a no-change NMU
[08:35] <lamont> how much do we care at this point?
[08:35] <lamont> (basically, all the font family of packages...)
[08:35] <lamont> fontconfig needed it, will upload that in a second.
[08:35] <lamont> actually, the fix is to bump the build-depends: defoma to the right version
[08:36] <lamont> debian #300278
[08:36] <medwards_> Love that startup sound.
[08:38] <lamont> mdz: font-arhangai fontconfig grace grace6 lmodern msttcorefonts pango1.0 scalable-cyrfonts scigraphica tipa ttf-* vflib2 vflib3 x-ttcidfont-conf xfonts-baekmuk xfonts-scalable-nonfree xfonts-thai-ttf
[08:38] <mdz> lamont: sounds like a good idea for main
[08:38] <lamont> ok.  will do those after lunch
[08:38] <mdz> lamont: for universe, let MOTU decide what to do
[08:40] <dholbach> lamont: could you give me a list of the universe packages that will need the bumped defoma build-dependency?
[08:40] <dholbach> lamont: of the entire list, i'll put it on the wiki and start working
[08:41] <lamont> dholbach: yeah - I'll let you know once I do the split
[08:41] <lamont> actually...
[08:41] <dholbach> lamont: or give me the entire list... no problem
[08:46] <Mitario> brb
[08:49] <medwards_> I don't suppose mdz has looked at adding overlay on USB stick to casper.
[08:50] <lamont> dholbach: sorry for the spewage in the other window...
[08:50] <lamont> mdz: 8 packages in main, more in universe.
[08:50] <dholbach> lamont: no problem
[08:51] <lamont> (in town, back in a couple hours)
[08:51] <dholbach> lamont: i'll start working
[08:51] <lamont-away> dholbach: Build-Depends: defoma (>=0.8.11ubuntu2)
[08:51] <mdz> medwards_: I'm already planning to implement it for Breezy
[08:52] <medwards_> mdz: can device-mapper do flash-friendly load balancing?
[08:52] <dholbach> lamont-away: thanks
[08:52] <mdz> medwards_: that type of feature was one of the motivating factors for the choice of device-mapper as the basis
[08:52] <lamont-away> mdz: can't you do it today with a correctly formatted disk and the right boot cmdline?
[08:52] <mdz> medwards_: don't USB sticks do that internally?  CF devices do
[08:52] <mdz> lamont-away: load one, yeah.  saving it is the tricky part
[08:53] <mdz> well, almost
[08:53] <mdz> you'd have to do it by hand because you'd need to arrange to mount the media
[08:53] <mdz> it really needs work to make it into a usable feature
[08:54] <medwards_> mdz: don't know whether USB sticks are that smart; I suppose I can do the naive thing and see how long it lasts.
[08:56] <schweeb> medwards_: the biggest thing would be to mount with noatime
[08:56] <medwards_> schweeb: good point.
[08:56] <mdz> casper already does that
[08:57] <schweeb> I'm pretty sure that USB sticks are intelligent... if not you need to use jffs2 or something for balancing
[08:57] <doko> elmo: please could you install on davis/hoary: libreadline4-dev tk8.4-dev libgdbm-dev blt-dev libbluetooth1-dev
[08:58] <medwards_> schweeb: yes, which is why I was thinking of trying a port to unionfs.
[08:58] <medwards_> (immature though it is; I oopsed Knoppix 3.8 with "vi /etc/hosts")
[09:00] <medwards_> not that device-mapper isn't a good solution, but for some purposes FS-level overlays fit better (e. g., ramdisk for volatile areas, jffs2 for most persistent bits, something else for mysql DB backing store)
[09:02] <schweeb> can't seem to find any info about how it spreads writes across a usb stick
[09:03] <medwards_> schweeb: unionfs doesn't, it's a sort of VFS loopback that can stack at FS rather than block device level.
[09:04] <medwards_> Has other advantages too; for instance, the bottom layer can be a truly read-only FS such as squashfs.
[09:05] <schweeb> yea, I've heard about unionfs
[09:10] <elmo> doko: done
[09:11] <doko> elmo: thanks
[09:15] <medwards_> schweeb: http://lists.infradead.org/pipermail/linux-mtd/2004-December/011177.html
[09:16] <medwards_> jdub (same jdub)? suggests just using ext3 (presumably noatime)
[09:16] <schweeb> no
[09:17] <schweeb> jdub = jeff waugh
[09:18] <tritium> seb128, libgtk2.0-0 is compiled with threading, isn't it?
[09:19] <seb128> no
[09:19] <seb128> who needs threading ?
[09:20] <medwards_> schweeb: also interesting: http://www.diskonkey.com/documents/Performance_reliability.pdf
[09:21] <schweeb> stupid okay button is grayed out to open with xpdf grr
[09:22] <medwards_> looks like ext3 noatime, possibly tuned for lazy sync, is the way to go.
[09:23] <schweeb> yea
[09:24] <medwards_> mount
[09:25] <medwards_> ww
[09:27] <medwards_> mdz: any particular reason for casper not to journalize the device-mapper block device (with a journal size appropriate to the ramdisk, not the whole FS) before mounting it?
[09:28] <mdz> medwards_: the journal eats snapshot space, and doesn't really buy us anything unless the backing store is on persistent media
[09:29] <mdz> that's why it's ext2
[09:29] <dholbach> mdz: do you happen to know if the  Depends  on  defoma  has to be set to  (>=0.8.11ubuntu2)  as well?
[09:29] <mdz> dholbach: no, I don't
[09:29] <dholbach> mdz: ok, so i'll stop the action until lamont returns
[09:31] <medwards_> mdz: understood, I was just wondering if there was a reason why it wouldn't be a one-liner.  Already seems to be mounted auto.
[09:33] <herve> hi, motu trainee for the defoma version bump!
[09:34] <dholbach> herve: lamont left, so we'll have to wait for him, because i don't know how to test if the newly built package is alright
[09:34] <herve> ok, I get back to that good ol' python transition :-)
[09:34] <dholbach> herve: hahaha, it's not _that_ "ol'" :-)
[09:35] <dholbach> herve: we're just so fast :-)
[09:35] <herve> it's one of the first, no? not to say the first
[09:35] <tritium> seb128, I'm building python-matplotlib, and the interactive gui needs threading
[09:35] <medwards_> Actually, I shouldn't even need to hack casper, except maybe to route the snapshot onto the USB stick; I should be able to create a new device-mapper device from a livecd shell (using whatever option starts by wiping the overlay), then add the journal, and it should Just Work.
[09:35] <seb128> tritium: ...
[09:36] <seb128> tritium: sure gtk is built with threading
[09:36] <tritium> seb128, okay, just double checking
[09:37] <tritium> seb128, based on these instructions: http://matplotlib.sourceforge.net/interactive.html
[09:38] <herve> seb128, is there a reason a package built with glade and gnome2.8 from Sid won't compile on Hoary?
[09:38] <herve> seb128, it fails on gtk/gnome calls
[09:38] <herve> at compilation time I mean
[09:38] <seb128> which one ?
[09:38] <seb128> glade is an app, not a lib
[09:38] <herve> gcompris :S
[09:38] <seb128> you mean libglade ?
[09:39] <herve> but the .c/.h generated by glade
[09:39] <seb128> urg
[09:39] <seb128> some apps do that ?
[09:39] <herve> maybe the build log would be more obvious
[09:39] <herve> the .c says "generated by GLADE"
[09:39] <seb128> I'm downloading the sources
[09:39] <herve> + "do not edit"
[09:40] <herve> I think gcompris is too big for me anyway
[09:40] <Mitario> hellohello
[09:40] <herve> hi Mitario 
[09:41] <seb128> hi
[09:41] <seb128> herve: according to the build logs it's looking for -lvga
[09:42] <herve> I added libsvag1-dev to build deps
[09:42] <herve> now the compilation goes further but fails on gtk/gnome references
[09:42] <seb128> k, so let me some min to download the deb-src
[09:43] <zul> later
[09:44] <dholbach> bye zul 
[09:48] <mdz> thom: around?
[10:07] <abelli> ciao everybody
[10:07] <abelli> is sabdfl around?
[10:10] <mdz> abelli: on holiday
[10:10] <abelli> mdz: thank you
[10:10] <abelli> when will he be back?
[10:10] <mdz> he is away for another week I think
[10:11] <abelli> right, thank you.
[10:11] <abelli> does he read emails?
[10:16] <medwards_> mdz: it looks like you already create the snapshot persistent in casper.
[10:20] <medwards_> so 10snapshot just needs code to check, say, casper-udeb/snapshot/overlay-file and, if it's non-null, add a second line of stdin to the initial dmsetup create.
[10:24] <medwards_> no, that's not quite right.  Perhaps it just needs to use dmsetup load instead of dmsetup create.  Experiment time.
[10:29] <mdz> medwards_: it's in persistent mode, but it's stored on a ramdisk at the moment (where a journal would just take up more space)
[10:30] <mdz> it's done persistent so that it could easily be dumped out to persistent storage
[10:30] <Mitario> mvo, hey, around? :-)
[10:31] <mvo> hey Mitario 
[10:31] <mvo> yes
[10:31] <dholbach> hi mvo
[10:32] <mvo> hey  dholbach 
[10:32] <Mitario> mvo, you got answer from the cvsadmins?
[10:32] <dholbach> mvo: what about a beer in the middle of us? like bochum-langendreer? ;-)
[10:32] <mvo> Mitario: no, you?
[10:33] <mvo> dholbach: I'm _so_ tired :)
[10:33] <Mitario> mvo, yes, they also didn't know the problem, they asked me if they wanted to import a tarball for us, so i'll just do that then
[10:33] <mvo> Mitario: all right
[10:33] <dholbach> mvo: even a nice and cold beer wouldnt get you outside? ;-)
[10:41] <ogra> dholbach, mvo, what about a beer in the middle of us ? like cologne ;)
[10:41] <dholbach> hahahah
[10:42] <ogra> *g* kidding....
[10:42] <dholbach> ogra: then we should stop calls for more people or we end in bavaria or something
[10:42] <ogra> hehe
[10:42] <herve> argh
[10:43] <mvo> haha
[10:43] <herve> I was about to suggesting Strasburg!
[10:43] <dholbach> i read an article about belgian beer today
[10:43] <ogra> india, if you invite jdub, or atlantis wrt mdz ;)
[10:43] <dholbach> what about that?
[10:43] <herve> dholbach, gaaa, belgian beer!
[10:43] <Amaranth> seb128: Do you know where I can find documentation for python-xdg?
[10:44] <herve> dholbach, I advise you to taste Leffe and Chimay
[10:44] <seb128> there is some example in the package
[10:44] <dholbach> herve: made a note :-)
[10:45] <herve> maybe "Bire des ours" (bears beer) but I'm not sure it's Belgian
[10:45] <Amaranth> I just need to get it to save. ;)
[11:00] <herve> doko, wasn't it RC2 today?
[11:01] <doko> oops, typo, yes it's rc2
[11:01] <herve> rock!
[11:01] <dholbach> doko: hope we won't need wiki/UniversePython2.4rc2TransitionTODO :-)
[11:02] <herve> glad I'm not cardiac!
[11:02] <herve> :-)
[11:05] <herve> dholbach, as I've seen, it's just rebuilding those packages with the new dh_installdefoma postinst/prerm scripts
[11:05] <dholbach> herve: i was wondering if we needed a Depends: on defoma (>= version) as well
[11:06] <herve> I don't think for the binary depends, no defoma behaviour changed
[11:07] <seb128> a danish guy is opening has a translation issue and is opening a bug on the differents apps :/
[11:07] <medwards_> mdz: I'm almost there.  "enter preinstalled session" failed, apparently because mount neither correctly auto-detected ext3 nor succeeded in mounting it ext2 ("couldn't mount because of unsupported optional features (4)").
[11:08] <mdz> medwards_: sounds like it should just need a "modprobe ext3"
[11:10] <doko> seb128: the danish are strange, three different OOo reports as well ...
[11:16] <medwards_> mdz: that seems to be right.  mount -t ext3 from a shell worked, presumably auto-probing the module, and then it all worked once I undid dmsetup and losetup steps.
[11:18] <medwards_> well, "all" is a little strong; it seems to have gone into the install rather than the livecd, and then decided to use ati rather than nv, etc., etc. :-/
[11:19] <medwards_> Perhaps you could add casper-udeb/snapshot/fstype defaulted to auto?
[11:22] <medwards_> mdz: looks like persistent snapshot on USB stick is a wiki page away.  :)
[11:23] <medwards_> I'll test with a fresh ext2 overlay -- unless there's already a probe-additional-modules debconf key?
[11:24] <Mithrandir> lamont_r: pong
[11:24] <lamont_r> yo
[11:24] <lamont_r> Mithrandir: http://people.ubuntu.com/~lamont/buildLogs/Test/d/db3/......-amd64-fail
[11:25] <Mithrandir> lamont_r: db3 sucks, it uses LD_ASSUME_KERNEL=2.4.1 which blows up.
[11:25] <Mithrandir> (on amd64)
[11:25] <Mithrandir> glibc is compiled with minimum kernel version = 2.6.0
[11:26] <Mithrandir> (it fails with problems loading libdl.so, right?
[11:26] <Mithrandir> )
[11:26] <lamont_r> yep
[11:26] <lamont_r> so can we either fix it, or drop it from the archive?
[11:26] <medwards_> It would be nice to integrate cryptoloop-aes into the chain, too.  Then I could throw away the old knoppix-mib CD that I've been using to do paranoid CD-booted GPG key management.
[11:26] <lamont_r> hrm... dropping it isn't really a good option...
[11:27] <Mithrandir> lamont_r: what needs it?
[11:27] <lamont_r>  apt-cache showpkg libdb3| wc
[11:27] <lamont_r>     177     354    5123
[11:27] <lamont_r> what doesn't?
[11:27] <Mithrandir> :P
[11:27] <lamont_r> hrm.. that's with universe, I suspect
[11:27] <lamont_r> exim4
[11:27] <lamont_r> libgnomeprint
[11:27] <Mithrandir> hm :/
[11:27] <lamont_r> libsasl7
[11:28] <lamont_r> bonobo
[11:28] <lamont_r> I think it stays... :-)
[11:28] <Mithrandir> yeah :/
[11:28] <Mithrandir> I could try fixing it on amd64.
[11:28] <lamont_r> pls fix.  kthxbye. :-)
[11:28] <Mithrandir> beers_owned["lamont"] ++ :P
[11:29] <lamont_r> lol
[11:31] <Mithrandir> elmo: could you please sync db3?
[11:31] <abelli> why if i try to install myspell-it it says it's not authenticated [gpg things] ?
[11:32] <schweeb> abelli: likely the GPG key the package was signed by isn't in your keyring
[11:33] <abelli> schweeb: huh? right... thank you.
[11:34] <schweeb> try apt-get update again, and see if the update complains that it's missing keys
[11:35] <abelli> ok
[11:38] <herve> schweeb, shouldn't the package be refused if the key is unknown?
[11:39] <schweeb> no, it asks if you want to use a untrusted package
[11:41] <herve> I mean, when it's uploaded
[11:41] <schweeb> herve: could be from a non-ubuntu archive
[11:41] <herve> right
[11:42] <lamont_r> herve: apt is checking the archive signature, not the individual package signature
[11:42] <schweeb> or, I've had a couple of times where ubuntu archives will say that, then I update again, and it works again
[11:43] <zul> hi
[11:50] <marcin_ant> hi developers
[11:50] <marcin_ant> I got a question - you'll propably send me to #ubuntu - but I'll try
[11:51] <marcin_ant> Are there any plans to reduce size of upgrades? 
[11:51] <marcin_ant> it's very nice that hoary is fresh and up to date
[11:51] <marcin_ant> but it needs huge downloads
[11:52] <wasabi_> That's because it's in development.
[11:52] <dholbach> marcin_ant: you should be able to upgrade from a hoary CD
[11:52] <marcin_ant> do we really need to replace whole packages to upgrade?
[11:52] <wasabi_> Yes. We do. ;)
[11:52] <marcin_ant> dholbach: I mean - daily upgrades
[11:52] <mvo> marcin_ant: see http://www.ubuntulinux.org/wiki/APTPackageDeltas
[11:53] <mvo> it's all in planing, no code yet
[11:53] <dholbach> marcin_ant: no you don't need to upgrade daily
[11:53] <wasabi_> i did an "implementation" of that a few months ago
[11:53] <wasabi_> was pretty fun
[11:53] <wasabi_> just distributed a diff of the new .deb to the old .deb though
[11:53] <ogra> marcin_ant, if hoary is stable the amount of upgrades will drop, as long as its in development the current is the way to do it...
[11:53] <wasabi_> oh yeah heh.
[11:53] <wasabi_> this is exactly what I made
[11:54] <marcin_ant> ogra: I don't want to reduce amount of upgrades
[11:54] <ogra> marcin_ant, but we :)
[11:54] <wasabi_> marcin_ant, read the page.
[11:54] <marcin_ant> ogra: it's really nice that ubuntu/linux/gnome development is dynamic
[11:54] <marcin_ant> ogra: but my question is generally about bandtwidh
[11:54] <marcin_ant> we have tools like rsync
[11:55] <ogra> marcin_ant,  if its stable there are only security upgrades, which reduces them drastically
[11:55] <marcin_ant> and I don't get why I need to download whole package in new version to upgrade
[11:56] <wasabi_> because, read the page.
[11:56] <dholbach> marcin_ant: momentarily there is no other way, we don't have half packages or something, you don't have to upgrade daily, this should save you bandwidth
[11:56] <ogra> marcin_ant, feel free to add the missing code for http://www.ubuntulinux.org/wiki/APTPackageDeltas, i'm sure mvo will happily review and use it
[11:59] <marcin_ant> dholbach: yes I know but I got apt in cron
[11:59] <marcin_ant> dholbach: and sometimes I'm on broadband but sometimes on gprs (which is slooow)
[12:00] <dholbach> it should only get the lists, unless you checked the checkbox in update-manager to get the packages as well
[12:00] <marcin_ant> dholbach: and this is why I found this uncomfortable to download whole package 
[12:00] <marcin_ant> anyway I got answer - I'll read about it
[12:02] <medwards_> mdz: any idea how to clobber the magic on a dm snapshot so that dmsetup create will make an empty snapshot?
[12:02] <mdz> medwards_: should work to zero it
[12:02] <medwards_> mdz: zero the whole device? ouch.
[12:02] <medwards_> mdz: let's try the undocumented "dmsetup clear".
[12:03] <medwards_> not useful.