[12:23] <lamont> sigh
[12:26] <wasabi> heh. nice.
[12:26] <wasabi> "oops"
[12:27] <wasabi> Looks like scott did it. =)
[12:27] <wasabi> 2 years ago heh
[02:23] <RAOF> Guest starring the Autobots!
[02:31] <visik7> how can I browse code for a specific package in bzr ?
[02:31] <visik7> I can't find it on launchpad
[02:31] <visik7> I'm here https://launchpad.net/hal
[02:32] <visik7> but I can't find something like svnweb or cvsweb or gitweb
[02:32] <visik7> where is bzrweb ?
[02:33] <visik7> without downloading it from packages.ubuntu.com
[02:33] <visik7> ok I found it
[05:56] <TheMuso> c
[05:56] <TheMuso> ugh
[05:56] <RAOF_> Doing that in new and different channels now :)
[05:58] <StevenK> Heh
[05:58] <StevenK> infinity: Ping, re: libnss-db again
[06:00] <infinity> StevenK: I know I shouldn't pong unless I'm prepared to respond immediately, but I can't resist the little purple number in irssi.
[06:01] <RAOF_> :)
[06:01] <infinity> StevenK: I'm trying to make time to look at it today (it's even flagged in my mail client), but I'm also trying to get lpia to the point where it can be debootstrapped. :)
[06:02] <StevenK> infinity: Ahh, but since I've uploaded libapache2-mod-auth-p{lain,am}, as soon as libnss-db is published, yada can be kicked back to universe.
[06:03] <StevenK> infinity: Surely that's a good incentive? :-)
[06:03] <infinity> StevenK: Yeah, I noticed that. :)
[06:06] <StevenK> infinity: Do you have time to do two givebacks for me? :-)
[06:13] <ajmitch> StevenK: best news I've heard for awhile (yada in universe)
[06:15] <StevenK> ajmitch: Almost. :-)
[06:15] <ajmitch> now what's needed to be able to drop it from the archive entirely?
[06:15] <StevenK> The 24 odd packages in universe to stop using it.
[06:16] <ajmitch> I'm sure noone will miss them
[06:16] <Amaranth> err, the rdepends only show yada-doc
[06:17] <Hobbsee> Amaranth: that'd be because packages build-dep on yada
[06:17] <Hobbsee> Amaranth: apt-cache show yada
[06:17] <StevenK> *Build-Depends* not Depends
[06:17] <Amaranth> ah
[06:17] <Amaranth> oh, that's evil
[06:18] <StevenK> Just try using it, it's like .spec files for Debian
[06:18] <StevenK> And the Maintainer of it is convinced that yada can do no wrong - mainly because he wrote it.
[06:19] <ajmitch> describing it as .spec files, so true & so evil
[06:19] <ajmitch> though I know they're popular & have their uses, I never grew to like them
[06:20] <Amaranth> it'd be cooler if it actually worked with .spec files
[06:20] <ajmitch> no, because then you'd just have alien, creating mutant .debs
[06:21] <ajmitch> there are some important differences between RH-like & debian-like systems for packages
[06:38] <infinity> StevenK: Of course.
[06:40] <infinity> StevenK: Of course, I can't give anything back unless you tell me what. :P
[06:40] <Hobbsee> infinity: the whole world will do
[06:41] <StevenK> infinity: Sorry, I was afk. Can you please give-back gabber on all arches, and glade-3 on ia64 and sparc.
[06:42] <infinity> Hobbsee: I'm not going to do any mass-give-backs until after lpia is bootstrapped.
[06:42] <Hobbsee> infinity: awww.  no fun :)
[06:44] <infinity> StevenK: What's the magic word?
[06:44] <StevenK> infinity: I did say please. :-)
[06:45] <infinity> StevenK: The magic word is "now", not I suppose I can cope with politenes..
[06:45] <StevenK> EPARSE
[06:45] <infinity> As in, "give these back, NOW"
[06:45] <infinity> "slave" also works.
[06:45] <StevenK> That strikes me as rude, not polite. :-)
[06:46] <Hobbsee> StevenK: but infinity may like it, so...
[06:46] <infinity> Oh, the parse error was s/not/but/
[06:46] <infinity> My bad.
[06:46] <infinity> I'm having some brain::finger disconnects today.
[06:46] <StevenK> Today? :-P
[06:46] <infinity> Shush.
[06:47] <elkbuntu> StevenK, see how far you can go. oil the whip :)
[06:47] <infinity> I've been checking every package in main for special-casing of i386, I'm going cross-eyed.
[06:47] <StevenK> Heh, ouch
[06:47] <infinity> Finished up all of Required, Important is next...
[06:47] <infinity> Then Standard and Build-Essential.
[06:48] <StevenK> Ohhh
[06:48] <infinity> Then I beg someone else to do Optional while I go sob in the corner in the foetal position.
[06:48] <StevenK> infinity: sejong is showing NOT OK : <socket.timeout instance at 0x2aaab2ba1710> (AUTO), too
[06:48] <infinity> StevenK: sejong hates me.
[06:48] <StevenK> I thought it hated everyone?
[06:48] <infinity> No, it hates me specifically.
[06:48] <infinity> It just takes it out on you guys.
[06:49] <infinity> Crashed building glibc.  WHAT A SHOCK.
[06:49] <infinity> BenC: I'll give you hookers and beer if you can provide me with a stable kernel for my sparc buildds.
[06:49] <infinity> BenC: Unlimited supplies.
[06:53] <infinity> Okay, I really am not awake...
[06:53] <infinity> This confused me for a full 30 seconds:
[06:53] <infinity> adconrad@drescher:~$ sudo su - buildd
[06:53] <infinity> Password:
[06:54] <infinity> (There's no buildd user on drescher)
[06:55] <StevenK> Heh
[06:56] <infinity> Hrm, I wonder if I can recruit MOTU manpower for https://wiki.ubuntu.com/MobileAndEmbedded/Bootstrap/
[06:58] <Hobbsee> infinity: there's got to be a way you can automate that....?
[06:59] <infinity> Hobbsee: Sort of, but not really.
[06:59] <infinity> Hobbsee: The bits mentioned in the wiki could be checked for, but I give a once-over to debian/rules to make sure people aren't doing cute and clever things, etc.
[06:59] <infinity> Hobbsee: It needs people who have A) a deep understanding of packaging software, and B) can read very quickly. :P
[07:00] <Hobbsee> heh :)
[07:00] <infinity> So far, the intersection of those two traits appears to be... Me.
[07:00] <infinity> Which is less than ideal. :)
[07:01] <infinity> Oh well, off to apt-get source all of Important.
[07:05] <infinity> I'm offended that I now have to use "-y" with "apt-get source" thanks to the new "OMG, REVISION CONTROL, ARE YOU SURE?!" feature...
[07:08] <StevenK> Morning pitti
[07:08] <infinity> Guten Tag, pitti.
[07:10] <Hobbsee> hiya pitti
[07:11] <infinity> pitti: Up early, aren't you?
[07:12] <StevenK> Bwahaha, gabber finally succeeded to build.
[07:12] <infinity> StevenK: If that was in error, I can reject the binaries. :P
[07:13] <StevenK> Hmph
[07:13] <infinity> StevenK: I wouldn't want your worldview shattered.
[07:14] <StevenK> infinity: It involved fixing 3 seperate Gnome 1.x packages to not use libdb3. I don't want to think that pain was for nothing. :-)
[07:14] <pitti> Good morning everyone
[07:14] <pitti> infinity: slightly, yes; hayfever didn't let me sleep any more
[07:14] <infinity> pitti: :(
[07:15] <pitti> but I went to bed early, so that's ok :)
[07:15] <infinity> pitti: I feel your pain, though.  I slept about 2 hours last night before I decided to just get up and work at 6am.
[07:15] <pitti> two hours? erk, that sucks
[07:17] <sbalneav> Night all
[07:21] <Hobbsee> oh darn.  i'm out of tuna.
[07:24] <Hobbsee> no more tuna for hobbsee's :(
[07:24] <TheMuso> Salmon is a good substitute.
[07:26] <Hobbsee> dont have any of that either.
[07:27] <Burgundavia> salmon has the issue of travelling an obscene distance to Aus
[07:28] <StevenK> pitti: Do you mind ripping out the kernel stuff from NBS? :-)
[07:29] <infinity> I have the technology to do that.
[07:29] <StevenK> I'm unsure if the new kernels have passed NEW yet.
[07:29] <infinity> I also have that technology. :P
[07:30] <pitti> StevenK: I NEWed kernels yesterday
[07:30] <pitti> StevenK: except for lrm which FTBFSed partially
[07:31] <infinity> Oh, he's alive enough to respond.
[07:31] <infinity> pitti: I'll let you play ftpmaster, then. :)
[07:31] <infinity> (Ws just looking for an excuse to avoid lpia for a few minutes, to be honest)
[07:31] <StevenK> infinity: I figured that. :-)
[07:32] <StevenK> infinity: If you want to avoid lpia for a few minutes, look at libnss-db. :-P
[07:32] <infinity> What I really need to do is hate on soyuz for a bit.
[07:32] <infinity> Manually scheduling all these builds is driving me mad.
[07:33] <infinity> The problem with hating on soyuz is that it leads to hacking on soyuz...
[07:34] <Hobbsee> infinity: this is beneficial - makes you avoid lpia, of course
[07:36] <infinity> On balance, I think I'm better off just getting all of lpia built.
[07:36] <infinity> Then I can quietly slink off somewhere and strangle myself.
[07:40] <pitti> StevenK: I won't NBS the kernel before lrm is sorted out and linux-meta points to -9.
[07:40] <pitti> StevenK: no need to deliberately break the daily CDs
[07:41] <StevenK> pitti: Fair enough.
[07:41] <StevenK> pitti: libapache-mod-auth-radius and liberis-1.3-11-dbg can NBS'd at your leisure.
[07:43] <pitti> StevenK: thanks, done
[07:43] <StevenK> pitti: Thanks.
[07:43] <StevenK> pitti: libgladeui-1-6 should also be able to be killed in a few hours.
[07:43] <StevenK> pitti: But not yet. :-)
[07:45] <pitti> StevenK: you really fell in love with cleaning the house, didn't you? :-)
[07:47] <StevenK> Cleaning the archive is much nicer. :-)
[08:32] <Burgundavia> pitti: the only bit that sucks is that g-c-m has a much nicer UI
[08:33] <Dessan> works decently well, however it thought my printer could only print post cards...
[08:33] <Dessan> That I wasn't so happy about.
[08:33] <pitti> Burgundavia: I agree
[08:33] <pitti> Burgundavia: and that's my (only) major concern about s-c-p; it's UI is horribly complicated
[08:34] <Burgundavia> Dessan: bugs.launchpad.net/ubuntu/+filebug
[08:35] <Burgundavia> pitti: the UI hasn't changed since FC2
[09:13] <pitti> hey mvo
[09:22] <mvo> good morning pitti
[09:22] <pitti> mvo: WDYT about SRU verification of bug 112803? I don't know how to reproduce the situation easily
[09:22] <ubotu> Launchpad bug 112803 in dbus "MASTER [Feisty]  cupsd leaking file descriptors (was: Multiple jobs are not printed)" [High,Fix released]  https://launchpad.net/bugs/112803
[09:23] <pitti> mvo: I reproduced it by hacking the dbus code only
[09:23] <pitti> mvo: but there have been confirmations that it works now, and the patch is obvious
[09:24] <mvo> pitti: checking now, but if you reproduced it earlier and can no longer with the new version I think it will be fine (but let me test if install/removal work) before moving to -updates
[09:25] <pitti> mvo: yeah, I reproduced the fd leak in cupsd and verified that it was fixed with the new libdbus
[09:26] <pitti> mvo: dbus has three fallback methods of acquiring a dbus connection, and calling dbus-launch is the third one
[09:26] <pitti> mvo: so I disabled the first two (check for unix sockets and environment variables IIRC), so that dbus-launch was tried immediately
[09:27] <pitti> mvo: but I don't know the setup on the reporter's machines well enough to reproduce it with the unmodified package
[09:27] <pitti> mvo: it just happens for an awful lot of people apparently
[09:32] <mvo> yeah, the bugreport has a lot of comments
[09:36] <carlos> pitti: hi
[09:36] <pitti> hey carlos, how are you?
[09:37] <carlos> pitti: fine, thanks, and you?
[09:37] <seb128> hey carl
[09:37] <seb128> hey carlos
[09:37] <pitti> carlos: reasonably well
[09:37] <carlos> pitti: how many days left? :-P
[09:37] <carlos> seb128: hey!
[09:37] <pitti> carlos: 18
[09:38] <carlos> pitti: :-)
[09:39] <carlos> pitti: I wonder whether you would have time this week to discuss .xpi translations move to language pack infrastructure
[09:39] <pitti> carlos: sure; we can do it in ~ 10 minutes if you want (just want to finish my current code)
[09:40] <carlos> pitti: sure
[09:40] <carlos> pitti: please, ping me when you are ready
[09:40] <pitti> will do
[09:40] <pitti> ah, cool, tests pass now
[09:45] <stgraber> morning
[09:45] <Hobbsee> morning stgraber!
[09:45] <Tonio_> morning :)
[09:47] <pitti> hi stgraber
[09:49] <StevenK> pitti: libgladeui-1-6 can be NBS'd at your leisure.
[09:49] <pitti> StevenK: you rock, thanks
[09:50] <pitti> killed
[09:51] <mvo> pitti: if you are hapy with my comment on 112803, I will set it to verification-done
[09:52] <seb128> bug #112803
[09:52] <ubotu> Launchpad bug 112803 in dbus "MASTER [Feisty]  cupsd leaking file descriptors (was: Multiple jobs are not printed)" [High,Fix released]  https://launchpad.net/bugs/112803
[09:52] <pitti> mvo: I am, thanks a lot
[09:52] <pitti> mvo: I'll copy it on Thursday, to meet the seven day rule
[09:53] <seb128> pitti: what would be the way to go to make nautilus-cd-burner build?
[09:53] <pitti> infinity: is there an easy way to find out why hal doesn't start in the buildd chroots? it seems that avoiding the dependency is hard
[09:54] <pitti> infinity: would it be reasonable to install a policy-rc.d which disables the startup of daemons? as build dependencies they are hardly useful, I figure?
[10:00] <pitti> carlos: I'm ready
[10:00] <carlos> pitti: ok
[10:00] <carlos> pitti: Launchpad already has XPI import support (not available in production yet)
[10:01] <carlos> pitti: and in this cycle, we should finish the XPI export support
[10:01] <carlos> so I would like to start preparing Ubuntu Gutsy to be able to provide updates for Firefox, Thunderbird and others using current language packs
[10:02] <carlos> right now, would be a matter of use the standard .xpi files, later, Launchpad will provide you with the updated .xpi just like we do for .po files
[10:03] <carlos> later == end of August (if there is no problem with the code deployment)
[10:04] <pitti> carlos: not sure whether we can directly put them into langpacks, though
[10:04] <pitti> asac: AYT?
[10:05] <pitti> carlos: right now, mozilla-firefox-locale-* needs the contents of the XPI in /usr/lib/firefox/extensions/ and a configuration snippet in /var/lib/mozilla-firefox/extensions.d/
[10:05] <carlos> pitti: isn't that generated from the .xpi file itself?
[10:05] <pitti> carlos: so, it's of course possible to do this unpacking and shipping in the langpack source itself, that would be no problem
[10:06] <carlos> as far as I know the source package for them is just the .xpi file
[10:06] <pitti> carlos: it's been a while since I touched m-firefox-locale-all source, but everything should be in the xpi, yes
[10:06] <carlos> and those directories are filled based on the .xpi content
[10:06] <pitti> carlos: I'll talk to asac
[10:06] <pitti> door bell
[10:06] <carlos> ok
[10:06] <carlos> pitti: thanks
[10:08] <pitti> carlos: so, where would you ship them?
[10:08] <pitti> carlos: in a subdirectory 'firefox' in the tarball, or so?
[10:08] <carlos> pitti: that's an option, yes. It's up to you to choose it
[10:08] <pitti> carlos: or in the per-locale directories?
[10:08] <pitti> carlos: let me take a look at my importer and see what's better
[10:08] <carlos> pitti: whatever fits better for you to prepare the language pack
[10:09] <carlos> pitti: take into account that it's not just firefox
[10:09] <carlos> but any extension or other Mozilla application (thunderbird)
[10:10] <pitti> carlos: it would be nice to have something like non-gettext/<application name>/<locale>.ext
[10:10] <pitti> carlos: e. g. non-gettext/firefox/de.xpi or non-gettext/firefox/es_ES.xpi
[10:10] <carlos> ok
[10:10] <pitti> similarly with thunderbird, or OO.o's files
[10:10] <pitti> carlos: would that work?
[10:10] <carlos> yeah
[10:11] <infinity> pitti: hal should probably be made more resilient in the face of systems it doesn't like...
[10:11] <carlos> although OO.org will not be included there, we still will need manual work there
[10:11] <pitti> infinity: I don't know why it fails ATM; could be missing /sys, or an already running hal
[10:11] <pitti> infinity: or anything else; but making package installation not fail when startup fails is kind of violating policy
[10:12] <pitti> carlos: OOo> right, I just want to consider it dir structure wise
[10:13] <carlos> pitti: sure
[10:15] <StevenK> Hmph. Just because something failed to build, doesn't mean you need to e-mail me about it twice, Launchpad.
[10:15] <RAOF_> StevenK: Only twice?  Not once per arch?
[10:15] <StevenK> Actually, both were for ia64, but one per buildd.
[10:17] <infinity> pitti: I;m about to knock off for the day, shall we have a constructive argument about it tomorrow? :)
[10:17] <infinity> pitti: (including me figuring out why it fails in the first place)
[10:17] <StevenK> pitti: Oh yes, I forgot to mention - Debian bug #434704
[10:17] <StevenK> infinity: You should look at libnss-db. :-)
[10:18] <infinity> StevenK: No, I should upload coreutils, tar, and gzip. :P
[10:18] <StevenK> Aww
[10:18] <pitti> infinity: thanks
[10:18] <infinity> (Or we could just drop them from the distribution, they're not important)
[10:19] <StevenK> pitti: If we kill db3, it'll show in the NBS list, and if I recall correctly, there is a bunch of stuff that pulls in libdb3.
[10:23] <tuxcrafter> hello guys
[10:23] <tuxcrafter> how can i check the version of glib i am using
[10:23] <pitti> StevenK: not sure what you mean?
[10:24] <pitti> tuxcrafter: dpkg -l libglib2.0-0
[10:24] <tuxcrafter> pitti: thanks i mean 368 468 or 686
[10:25] <pitti> tuxcrafter: (#ubuntu, please)
[10:25] <StevenK> pitti: That's okay, I just read it and couldn't make sense of it either. :-)
[10:25] <pitti> tuxcrafter: we don't have packages for 486/686
[10:26] <tuxcrafter> pitti: i am debugging a serious bug with the via c7 cpu and glibs
[10:26] <tuxcrafter> i need to install and use glibc i386
[10:27] <tuxcrafter> i installed the 368 kernel
[10:27] <pitti> tuxcrafter: our _i386.deb packages are built with 486 instruction set and optimized command ordering for 686, AFAIR
[10:27] <tuxcrafter> ow than i have to install my own kernel
[10:27] <pitti> tuxcrafter: we do not have any other debs
[10:28] <pitti> tuxcrafter: we do have a special -i386 kernel, just not other packages
[10:28] <tuxcrafter> pitti: ok but do you know a way to check what glib there is uses 468 or 686
[10:30] <pitti> tuxcrafter: as I said, if you use the Ubuntu packages, it's 486 instruction set
[10:31] <infinity> (Not counting possible toolchain bugs that may slip in other instructions for fun)
[10:31] <infinity> Is that the sort of bug you're hunting right now?
[10:31] <tuxcrafter> pitti: ok, and can you tell me how to compile the glib with 368 instructions only
[10:31] <Mithrandir> tuxcrafter: first, it's glibc, unless you are talking about glib which is another library.
[10:32] <tuxcrafter> glibc
[10:32] <Mithrandir> tuxcrafter: secondly, changing to i386 will change the c++ abi, something you probably don't want to do.
[10:33] <asac> pitti: carlos: good to hear that we have xpi support now ... whats the next step?
[10:33] <pitti> asac: preferably I'd make mozilla-firefox-locale-all obsolete and ship the xpis right in the language-pack-XX
[10:33] <pitti> asac: WDYT?
[10:33] <pitti> asac: is the rather complicated packaging structure of m-f-locale-XX still necessary, or is it enough nowadays to just drop an xpi somewhere?
[10:34] <carlos> pitti: maybe that + a script to install the xpi files would be enough...
[10:34] <tuxcrafter> there is a via c7 instruction used in the glibc i686/486 lib that makes the system freeze completely, so i want replace the glibc with an i368 version that uses i368 instructions
[10:34] <asac> pitti: unfortunately its not possible to just drop xpi somewhere
[10:35] <pitti> carlos: right
[10:35] <carlos> asac: well, it's not fully implemented, still working on exports, but release date for that is end of August
[10:35] <pitti> asac: ok, so let's do the mangling in langpack-o-matic then, to avoid spreading copies of that code
[10:36] <asac> pitti: pitti what do yo mean with mangling in langpack-o-matic? langpack-o-matic producing debs instead of xpis?
[10:37] <pitti> asac: I mean, disassembling the xpi and ship the right bits in langpack-o-matic deb
[10:37] <pitti> erm
[10:37] <pitti> s/langpack-o-matic deb/language-pack-XX deb/
[10:39] <soren> tuxcrafter: Just remove libc6-i686?
[10:39] <asac> carlos: do you need info to do that?
[10:39] <carlos> asac: pitti is the one doing it
[10:39] <carlos> I just provide the .xpi files
[10:40] <asac> carlos: ah. fine.
[10:40] <carlos> the script that prepares the .deb file is still in pitti's hands
[10:40] <pitti> asac: don't worry, I'll pry it out of the m-f-l-a source
[10:40] <pitti> carlos: another question, what's the status of that full dapper export?
[10:41] <carlos> pitti: being executed right now
[10:41] <pitti> splendid, thanks
[10:41] <carlos> s/executed/exported/
[10:42] <tuxcrafter> soren: thanks
[10:42] <tuxcrafter> i am going to reboot to a other kernel
[10:44] <soren> pitti: About the eBox acl and schemas mangling madness: I've created a patch that creates /etc/ldap/acl.d/ and /etc/ldap/schemas-enabled/ and makes slapd read from those directories. That way I don't have to touch slapd.conf directly at all. The Debian dudes say it'll be included in the next openldap2.3 upload. \o/
[10:47] <ogra> seb128, ping
[10:47] <seb128> ogra: pong, new gnome-power-manager available
[10:47] <ogra> thanks
[10:47] <ogra> seb128, why cant i log in on my thi clients anymore ?
[10:47] <ogra> *thin
[10:47] <carlos> pitti: btw, you should also add .xpi extraction code from build packages
[10:48] <seb128> ogra: because you don't remember your password? ;)
[10:48] <ogra> seems it hangs during session startup since the last upgarde
[10:48] <seb128> no idea
[10:48] <carlos> pitti: so we get them also like any other .po and .pot files
[10:48] <ogra> (i didnt upgrae since before ubuntulive though)_
[10:48] <ogra> no traces in .xsessin-errors .... new user doesnt help either
[10:49] <cjwatson> infinity: I'm taking care of openssh/lpia, btw, in case you'd already got to that
[10:49] <carlos> pitti: oh, and, given there is no .xpi for en_US, I guess you should special case en_US.jar file and extract that one too so we get the English strings to be translated
[10:51] <ogra> seb128, does consolekit expect any kind of variable to be set to let me in ?
[10:51] <seb128> ogra: consolekit is not used yet
[10:51] <ogra> oh, i only saw ian making changes to gnome screensaver wrt CK
[10:51] <seb128> yes, it's using it a runtime if available
[10:51] <asac> carlos: so you need a langpack as blueprint to get the "to translate" entities?
[10:51] <seb128> consolekit has not been promoted nor seeded though
[10:52] <seb128> and gdm doesn't build with it yet
[10:52] <asac> carlos: usually it should be enough to give you an application .xpi
[10:52] <pitti> soren: awesome!
[10:52] <ogra> ok
[10:52] <carlos> asac: yes, most of the time it's en_US
[10:52] <ogra> hmm
[10:52] <asac> carlos: well ... why that?
[10:52] <carlos> asac: that's because Mozilla uses IDs instead of english strings as identifiers
[10:52] <asac> and?
[10:53] <pitti> carlos: can do; can you please file a bug against pkgbinarymangler and say where you want to get those XPIs shipped?
[10:53] <carlos> so you have DIALOG_SOMETHING = 'SPANISH TRANSLATION'
[10:53] <carlos> instead of 'ENGLISH TRANSLATION' = 'SPANISH TRANSLATION'
[10:53] <carlos> and thus, without en_US translations, translators don't know what to translate exactly
[10:54] <asac> hmm
[10:54] <carlos> asac: but don't worry, you don't need to do anything there
[10:55] <asac> carlos: but i mean ... usually an application/extension ships at least one translation on its own ... so does it just work?
[10:55] <carlos> Launchpad + pitti's script will handle it automatically
[10:55] <ogra> hrm
[10:55] <carlos> asac: which is en_US, isn't it?
[10:55] <carlos> at least it is for Firefox and I think Thunderbird
[10:55] <ogra> seb128, creatig a ~/.xsession containing gnome-session seems to run fine :/ so its not gnome then
[10:55] <asac> carlos: doesn't matter imo ... can be any
[10:56] <carlos> asac: I guess, but we assume it's en_US. If it's not the case, we will need to do some extra work to see how to handle it (our main target was to handle Firefox)
[10:57] <carlos> asac: the one used as the original English will not be exported in the language pack, if that's what worries you
[10:58] <asac> carlos: no ... i am just worried that you created a special purpose thing that is only useful for the applications ubuntu ships
[10:58] <cjwatson> infinity: I've also changed lintian in svn to recognise lpia as a valid architecture
[10:58] <carlos> asac: how's that?
[10:58] <asac> carlos: but if that is not the case and e.g. translating ubufox or any other extension just works, then i am fine
[10:58] <mvo> pitti: should we do a special case for tzdata too in StableReleaseUpdates? Its cumbersome to try to find out how to reproduce the "bug". so I think we should we should let it through with just basic testing
[10:58] <carlos> asac: it should work too
[10:58] <carlos> but we didn't pay too much attention to it
[10:59] <carlos> so maybe needs some code tweaks to be able to handle it
[10:59] <carlos> which is fine, and easy to do
[10:59] <asac> well .. so for every package we need to do manual tweaking?
[11:00] <carlos> I don't think so
[11:00] <cjwatson> mvo: tzdata has been special-cased in the past, since the change is typically enforced by external factors
[11:01] <carlos> and even so, it's Launchpad which will get such changes to cope with all those packages
[11:01] <asac> carlos: what concerned me a bit and makes me ask these questions is that nobody asked me for input ... and suddenly there is a solution ;)
[11:01] <carlos> asac: well, I think you are confused...
[11:01] <carlos> asac: we just add native .xpi import/export support to Launchpad
[11:01] <mvo> cjwatson: right, we have not written it down yet in the StableReleaseUpdates page as a special case and I think we should (just like the kernel and app-install-data-commercial)
[11:01] <carlos> completely unrelated with your packages
[11:02] <carlos> now, we are seeing how to integrate it with Ubuntu
[11:02] <asac> ah ok
[11:02] <cjwatson> mvo: that would be OK with me if it's just talking about updates to new upstreams
[11:02] <asac> carlos: so i can import an ubufox .xpi right now?
[11:02] <mvo> cjwatson: thanks, I will add it to the wiki
[11:03] <carlos> kind of, although the export is not on production, so we don't officially support it yet :-P
[11:03] <asac> carlos: if it is really as simple as just uploading an extension .xpi to get things started then i am fine.
[11:03] <carlos> asac: yeah, that's the inde
[11:03] <carlos> s/inde/idea/
[11:03] <carlos> asac: although you need two xpi files
[11:03] <carlos> one with the English strings
[11:04] <carlos> well, you just need the English lang pack to start translating
[11:04] <asac> carlos: this is what i don't understand
[11:04] <carlos> two if you want to import an existing translation
[11:04] <pitti> mvo: I agree; it would be nice if you could describe the tests you do on the wiki page? (the tzdump thing)
[11:04] <carlos> asac: have you ever translated a .xpi file?
[11:05] <mvo> pitti: I will dig and write something up
[11:05] <asac> carlos: first you have to tell what .xpi you are talking about
[11:05] <asac> carlos: an extension .xpi ... or a langpack
[11:05] <carlos> asac: any .xpi file that follows the Mozilla's spec for .xpi translations
[11:06] <asac> carlos: in real world ... you usually have an extension xpi
[11:06] <asac> and there are already translations included
[11:07] <asac> why should authors need to manually extract an en_us langpack first?
[11:07] <carlos> do you mean that extension xpi files include several translations in a single .xpi file?
[11:07] <asac> carlos: in real-life you usually have one-or-many translation in the extension xpi ... yes
[11:07] <pitti> carlos: you need the en_US.xpi as 'C' translation?
[11:08] <carlos> hmm, that's not in the spec from Mozilla
[11:08] <pitti> asac: are you talking about translations of extensions or translations of ffox itself? (we do about the latter)
[11:08] <carlos> pitti: yes, although it usually doesn't exist, so en_US.jar (which exists in Firefox) would be enough
[11:08] <pitti> asac: extension translations won't be covered by that for now
[11:08] <asac> pitti: well then its not worth anything
[11:08] <carlos> asac: in that case, to support extension xpi we would need to do some changes in Launchpad to handle them
[11:09] <asac> pitti: or at least not much
[11:09] <pitti> asac: it's about automatically updating the translations from Rosetta without the need to touch m-f-l-a
[11:09] <asac> pitti: our applications already have translations ... launchpad should provide means to collaborate in translating all those that are not yet translated
[11:09] <carlos> asac: could you point me to an example of such .xpi file to take a look?
[11:09] <pitti> asac: right now we cannot use LP to translate ffox/tbird/etc/
[11:09] <carlos> asac: we will import existing translations
[11:10] <carlos> as we do for .po files
[11:10] <asac> pitti: we don't want to either ... translations are updated upstream
[11:10] <asac> pitti: we want translation of those applications that are not yet translated
[11:10] <asac> pitti: which are mostly extensions
[11:10] <pitti> asac: but new languages, too
[11:10] <asac> pitti: yeah ... i agree with that
[11:10] <carlos> asac: dude, we need to do one step at a time
[11:10] <pitti> asac: a lot of upstream xpi are missing translations of help, too
[11:10] <asac> carlos: well ok
[11:11] <asac> carlos: if its just the initial step then its ok
[11:11] <carlos> asac: yeah, first native support for .xpi, then, cover any special case that use .xpi file format
[11:24] <pitti> mvo: hm, these automatic package installation failure bugs are not that useful (see bug 129168)
[11:24] <ubotu> Launchpad bug 129168 in apport "package apport 0.93 failed to install/upgrade: dependency problems - leaving unconfigured" [Undecided,New]  https://launchpad.net/bugs/129168
[11:24] <pitti> mvo: any chance that apt could add some details?
[11:25] <mvo> pitti: I was working recently on a dpkg-log patch for apt that will capture all of dpkgs output
[11:25] <asac> carlos: fwif ... look at http://people.ubuntu.com/~asac/funkymonkey.xpi (its just greasemonkey)
[11:25] <pitti> mvo: ah, good
[11:25] <asac> s/fwif/fwiw/
[11:25] <mvo> pitti: it seems to be working quite well for me, I will merge it soon I think. I need to explain the report generator to attach the log inside libapt too. you will have to explain what format is expected etc
[11:25] <carlos> asac: cool, thanks
[11:26] <carlos> asac: that should be enough to figure a way to support that specific layout
[11:26] <ogra> ogra@laptop:~$ grep logintest /etc/passwd
[11:26] <ogra> logintest:x:1004:1005:,,,:/home/logintest:/bin/bash
[11:26] <ogra> ogra@laptop:~$ ps aux |grep -c logintest
[11:26] <ogra> 5
[11:26] <ogra> ogra@laptop:~$ ps aux |grep -c 1004
[11:26] <ogra> 40
[11:26] <ogra> can anyone explin why my system seem not to be able to resolve usernames anymore ?
[11:26] <ogra> *explain
[11:27] <pitti> ogra: logintest is longer than 8 characters
[11:27] <pitti> ogra: ps usually displays numbers in that case
[11:27] <asac> carlos: ... there is no standard layout of xpi's ...its all generic
[11:27] <asac> carlos: you have to look at chrome.manifest file
[11:28] <carlos> asac: I see
[11:28] <asac> carlos: otherwise you will end up implementing zillions of special cases
[11:28] <ogra> pitti, ah, thanks, i was hoping i had found the cause for my login problem :/
[11:28] <carlos> hmm
[11:29] <carlos> interesting...
[11:29] <leftcase> Hi folks, I'm trying to get to the bottom of a problem with network manager and atheros WiFi cards. I think I've sussed out whats going on but just wondered if anyone could help me puzzle it through
[11:30] <carlos> asac: so just to confirm it to you... our current code will only allow Firefox and Thunderbird translations, we will need more code changes in Launchpad to handle the kind of file you just gave me
[11:30] <mvo> pitti: could you please check https://wiki.ubuntu.com/StableReleaseUpdates?action=diff ?
[11:30] <iwj> soren: re Debian #421344> My initial reaction is `DDTT'.
[11:30] <asac> carlos: i don't know your code :) so i can't tell
[11:30] <carlos> I'm not asking but telling you it :-)
[11:30] <carlos> asac: thanks for your input
[11:30] <asac> carlos: as long as you cannot deal with arbitrary .xpi extensions
[11:31] <asac> then its true
 Error: Could not parse data returned by Debian: <HTTPError 500 Internal Server Error>
[11:31] <infinity> iwj: Congrats on confusing the bug bot.
[11:31] <asac> carlos: remember that the only thing that is special case in regards to translations are firefox/thunderbirda applications on their own ... but even they have chrome.manifest files iirc
[11:31] <iwj> infinity: :-)
[11:32] <carlos> asac: yeah, I guess should be easy to change our code to be more generic so we don't assume things like we do right now
[11:32] <asac> carlos: i am always on #ubuntu-mozillateam if you need more input
[11:32] <carlos> asac: cool, thanks
[11:32] <asac> carlos: i would be open to discuss the different approaches if you want
[11:32] <asac> carlos: if not ... well :)
[11:32] <carlos> asac: I will need to write another spec so I will ask you for more input
[11:32] <carlos> don't worry
[11:33] <pitti> mvo: looks ok to me
[11:35] <soren> iwj: ddtt?
[11:35] <jetscreamer> monitor autodetection could use some work eh
[11:36] <mjg59> jetscreamer: Bug #?
[11:36] <leftcase> sod it, I'll got for it anyway!
[11:36] <infinity> soren: Don't Do That, Then.
[11:36] <jetscreamer> i dunno... i just worked around it with xvrefresh= & xhrefresh
[11:37] <soren> infinity: Doh...
[11:37] <wolfe> mmmm dogfood
[11:37] <mjg59> jetscreamer: The autoconfiguration is based on the values obtained from your monitor
[11:37] <mjg59> If the monitor lies, things become harder
[11:38] <soren> iwj: You don't find it reasonable to mark a symlink as a conffile?
[11:38] <Mithrandir> soren: it's never worked and it's well-known that it doesn't.
[11:38] <iwj> Marking a symlink as a conffile is not what conffiles are for; the conffileness is applied by dpkg to the target of the link and not to the link contents.
[11:39] <leftcase> I noticed whilst checking out a bug report in network manager that a dev had suggested that teh bug was in madwifi - Checking over on madwifi, they have now fixed the bug (wext compliance)
[11:39] <jetscreamer> i'm not trying to say anything but relay this data: it didn't like to kanotix, which is where i got the values to enter for the xv&xh :)
[11:39] <iwj> This is so that if the sysadmin (or someone) moves a conffile and leaves a symlink behind, the right things happen.
[11:39] <jetscreamer> err s/like/lie/
[11:39] <mjg59> jetscreamer: Then please file a bug
[11:39] <jetscreamer> but no idea, i just know what it did
[11:39] <mjg59> That way we can work out what the problem is
[11:39] <jetscreamer> damn my hd is clicking... just xp :)
[11:39] <leftcase> But the bug is still happenening :-S so, could anyone point me towards which version of madwifi is supplied in gutsy in the restrictued modules ? :-S
[11:40] <soren> Mithrandir: That (in itself) doesn't mean it's not a bug and shouldn't be fixed.
[11:41] <Mithrandir> soren: I think it should be fixed by policy saying "don't mark symlinks as conffiles, HTH, HAND"
[11:41] <soren> This seems to be use-a-lot-of-acronyms-soren-doesn't-know day.
[11:41] <soren> Mithrandir: HTH? HAND?
[11:41] <infinity> soren: I don't honestly see how marking a symlink as a conffile makes sense, really.
[11:42] <infinity> soren: Hope That Helps, Have A Nice Day.
[11:42] <leftcase> So that I can create a proper bug report with enough information included to make it just mabey, possible bloomin' useful to someone!
[11:43] <ogra> hmmm, weird, how could my /usr/bin/x-session-manager vanish ?
[11:44] <soren> infinity: In this particular case I'm working on, there are symlinks from a dir in /etc/ldap/ that points to schema's that slapd should load.  If an admin for some reason don't want this, we should be able to just remove the symlink and not have it recreated on the next upgrade.
[11:45] <soren> infinity: The fix - of course - is to provide a script that handles these symlinks, so that they're not part of the package..
[11:45] <soren> infinity: But that's got "workaround" written all over it, IMO.
[11:46] <infinity> soren: They can either be "managed", or only created on fresh installs.
[11:47] <infinity> soren: It's not rocket science to not create them over and over on upgrade.
[11:47] <soren> infinity: Of course not.
[11:47] <soren> infinity: The same holds true for configuration files, but marking things as conffile just makes it easier.
[11:49] <infinity> soren: The problem, of course, is that it fundamentally mucks with how dpkg thinks about these things.
[11:49] <mjg59> leftcase: If nobody answers your question, it's likely that they either don't know or are currently busy
[11:49] <infinity> soren: And allowing your case breaks the other case (moving a conffile, symlinking to it, and having dpkg follow the symlink)
[11:49] <leftcase> mjg59, thankyou for your assistance
[11:51] <pitti> leftcase: madwifi is currently at version 0.9.3.1
[11:51] <pitti> oh
[11:51] <infinity> soren: Given that the current behavior is well-known (though perhaps not documented), "fixing" it could opena can of worms in breaking other systems.
[11:51] <soren> infinity: Well, yes, I can see that that is a) somewhat useful (I've never needed it myself, though), b) what people are used to and probably rely on in some cases.
[11:51] <soren> infinity: Exactly.
[11:51] <soren> Oh, well. Back to the drawing board.
[11:53] <infinity> if [ -z "$2" ] ; then ln -s foo bar; fi ?
[11:53] <infinity> (In postinst)
[11:53] <infinity> Seems the simplest route.
[11:54] <infinity> soren: ^^
[11:54] <norsetto> pitti: Hi Martin, did you receive my email about mentoring?
[11:55] <pitti> norsetto: not that I can see
[11:55] <norsetto> pitti: ok, I'm resending now
[11:55] <soren> infinity: Sure, sure.
[11:55] <pitti> infinity: you probably want that on upgrades as well, though
[11:56] <pitti> soren: ^ so you probably want dpkg --compare-versions "$2" le "versionthatintroducesthis"
[11:56] <soren> pitti, infinity: Yeah, I'm doing it on upgrades from version older than the one where I'm introducing this, and on new installs.
[11:56] <pitti> soren: "le" will do that ('empty version is later than any')
[11:57] <norsetto> mvo: Hi Michael, you also have not received my email?
[11:57] <pitti> soren: erk, wrong; that's le-nl
[11:57] <pitti> soren: so, le-nl should DTRT then
[11:57] <soren> pitti: Got it.
[11:57] <mvo> norsetto: I got it, sorry for not answering yet
[11:57] <norsetto> mvo: np, just checking, thanks
[11:57] <pitti> (if only I could invent a good expansion for 'nl' to finally memorize it)
[11:58] <cjwatson> pitti: "no-[version-is] -later
[11:58] <cjwatson> "
[11:58] <norsetto> pitti: nl=netherlands (they are your neighbours after all....)
[11:58] <pitti> aah, thanks; so 'None is Later'
[11:59] <cjwatson> or that, that's simpler
[01:00] <asac> pitti: iceowl+iceowl-extension took an incouraging 2 hours to slip through debian NEW yesterday ... can we do something similar for sunbird+lightning :) ?
[01:00] <asac> s/incou/encou/
[01:00] <pitti> iceowl?
[01:01] <geser> aka sunbird
[01:01] <pitti> yet another copy...?
[01:01] <bhale> pitti: for freedom!
[01:02] <asac> pitti: right
[01:02] <asac> pitti: should be the last copy for some time :)
[01:02] <asac> pitti: then we would be complete ... just let me know if I should upload or not :) ... if you don't like the non-free files we can sync from debian as well :)
[01:02] <geser> aren't there any mozilla projects left which Debian didn't "iced" yet?
[01:03] <pitti> asac: well, today is Mithrandir's archive day, tomorrow Riddell's; so choose your blackmail target :)
[01:03] <Kmos> :)
[01:03] <pitti> asac: I'm not that concerned about the icons etc, just a little nervous about other-licenses/ (but we had that discussion already)
[01:03] <asac> pitti: honesty, i don't want to swim against the tide
[01:04] <asac> pitti: the icons that are non-free are in other-licenses :)
[01:08] <asac> geser: xulrunner + nss + nspr
[01:08] <alesan> hi, any info on how to prepare a different initram image?
[01:09] <alesan> I mean, create a custom image from the current one
[01:43] <zul> morning
[01:43] <pitti> hey zul
[01:46] <zul> hi pitti how are things?
[01:46] <pitti> zul: pretty good! and for you?
[01:46] <zul> pitti: son was sick this weekend but ok
[01:53] <pedro_> morning
[01:55] <realist> evening :-)
[01:56] <norsetto> pitti: got the email?
[01:57] <pitti> norsetto: ah, I got it now; weird, I didn't get the previous one
[01:58] <norsetto> np, I think some spam filters don't like my ISP....
[02:03] <pitti> Riddell:  kdepimlibs (3.92.0-0ubuntu1~ppa1) gutsy; urgency=low
[02:04] <pitti> Riddell: was that actually meant to hit gutsy, or was it a mis-upload?
[02:04] <norsetto> pitti: btw, when you have time let me know if what I did for bug 128697 can help.
[02:04] <ubotu> Launchpad bug 128697 in restricted-manager "restricted-manager misses man page" [Undecided,Confirmed]  https://launchpad.net/bugs/128697
[02:05] <Riddell> pitti: it was a mis-upload
[02:05] <pitti> norsetto: oh, thank you!
[02:05] <pitti> norsetto: right, no patch etc. necessary, I'll just put it into the bzr tree
[02:06] <Riddell> pitti: it can sit there happily enough until 3.92 is actually released later this week
[02:06] <pitti> Riddell: it's gutsy, so beta versions aren't that evil yet :)
[02:27] <cjwatson> would somebody NEW partman-auto-loop, please? ideally review it for main at the same time ...
[02:27] <cjwatson> (installer-for-windows spec)
[02:31] <Mithrandir> doko: is it pycentral or pysupport we like?
[02:32] <Mithrandir> cjwatson: looking
[02:32] <doko> Mithrandir: the former
[02:32] <geser> Hi Hobbsee
[02:33] <Hobbsee> hey geser
[02:34] <Mithrandir> cjwatson: looks mainable to me, but that's really pitti's domain.
[02:34] <Mithrandir> pitti: any objection to letting partman-auto-loop just go directly to main?
[02:34] <seb128> Hobbsee: can you archive accerciser gnome-keyring vte passepartout glade-3 gtkmm2.4 glibmm2.4 on REVU?
[02:35] <Hobbsee> seb128: sure
[02:35] <seb128> thanks
[02:35] <Hobbsee> seb128: no problem.  consider it done
[02:35] <Mithrandir> pitti: I just accepted it; if you object, we'll change it later.
[02:35] <Mithrandir> cjwatson: accepted
[02:35] <Mithrandir> doko: thanks
[02:36] <cjwatson> Mithrandir: thanks!
[02:36] <Hobbsee> seb128: you couldnt just request that i archive the rest, could you?  *g*
[02:37] <seb128> no :p
[02:37] <Hobbsee> darn.
[02:37] <coNP> thanks Hobbsee :)
[02:37] <Hobbsee> coNP: :)
[02:40] <Hobbsee> because if you did, that'd cut our reviews needed a lot...
[02:43] <alesan> re
[02:44] <alesan> on https://help.ubuntu.com/community/USplashCustomizationHowto there is info on how to update the splash screen. it is updated up to edgy, I guess the same applies to feisty right?
[02:45] <norsetto> /motd
[02:45] <norsetto> doh :-O
[02:45] <Hobbsee> alesan: sounds like a #ubuntu type question
[02:46] <alesan> Hobbsee, well sorry I thought it was more appriopriate here. most people on #ubuntu don't know what a initram is :(
[02:46] <alesan> I'm building my own ubuntu iso file.
[02:46] <alesan> with customized stuff etc
[02:47] <Hobbsee> oh, thought that was -motu
[02:48] <Hobbsee> REVU is sending me insane, it looks like
[02:56] <StevenK> pitti: Ping me when you get back from lunch, please
[02:59] <pitti> Mithrandir: fine for me
[02:59] <pitti> StevenK: pong
[03:00] <StevenK> pitti: Right, I'm trying to fix postgresql-pljava. Nothing of the Java stuff is broken, but the build fails because it Build-Depends on postgresql-server-dev-8.1, but pg_config returns "VERSION = PostgreSQL 8.2.4"
[03:00] <pitti> StevenK: right, it should really use 8.2
[03:01] <StevenK> pitti: Fair enough, it looks to be a s/8.1/8.2/ on the control file
[03:01] <pitti> StevenK: I don't know if the current upstream version even works with 8.2, but Peter Eisentraut is usually catching up ASAP
[03:01] <pitti> StevenK: isn't that fixed in Debian yet?
[03:01] <StevenK> Nope.
[03:01] <pitti> StevenK: hm, then upstream might not support 8.2 yet
[03:02] <StevenK> pitti: I'm happy to fix0r it against 8.2 and then test it if you throw me some pointers.
[03:02] <StevenK> And if you reply back with 0xdeadbeef, I *will* get you. :-P
[03:02] <pitti> StevenK: hm, if you never touched a PL, then it might actually be faster if I try myself
[03:03] <StevenK> http://xkcd.com/138/ for the pointers joke
[03:03] <pitti> StevenK: Debian bug 419305 FYI
[03:03] <pitti> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=419305, since ubotu cannot read it ATM
[03:04] <StevenK> Yeah, I know BTS URLs. :-)
[03:04] <pitti> I didn't doubt, just for clickability's sake :)
[03:04] <StevenK> But no reply since 14th April doesn't fill me with confidence ...
[03:05] <pitti> wow, I sent those bugs on my bday? shame on me
[03:05] <StevenK> Haha
[03:05] <pitti> yeah, that package didn't see action recently
[03:06] <Hobbsee> bad pitti!
[03:06] <StevenK> pitti: I'm happy to throw a 8.2 built package at you, if you like.
[03:07] <pitti> StevenK: oh, sure; amd64?
[03:07] <pitti> StevenK: if it builds, that's already a gooood sign
[03:07] <StevenK> pitti: xkcd for the win. :-)
[03:07] <pitti> yeah, 1.3.0 is the latest upstream version
[03:10] <pitti> StevenK: source or debdiff should work fine, though
[03:12] <StevenK> pitti: I don't mind, either works.
[03:12] <StevenK> I'm just checking the thing builds first
[03:14] <StevenK> /tmp/buildd/postgresql-pljava-1.3.0/src/C/pljava/XactListener.c:107: error: 'xactCB' undeclared (first use in this function)
[03:15] <StevenK> Also implicit declaration of functions RegisterXactCallback and UnregisterXactCallback
[03:16] <soren>  /win 5
[03:16] <pitti> StevenK: don't bother, I'd say; that's something for upstream
[03:16] <pitti> StevenK: other PLs like the ones for Ruby generally needed new upstream versions for new major postgresql versions
[03:16] <StevenK> pitti: Actually, the error looks to be a code problem
[03:16] <pitti> StevenK: but why not just leave it building the 8.1 plugin?
[03:17] <StevenK> Well, then I need a pg_config that says it's version 8.1
[03:18] <StevenK> And there doesn't seem to be a libpq-8.1-dev
[03:18] <pitti> StevenK: it should be in postgresql-server-dev-8.1
[03:18] <pitti> StevenK: /usr/lib/postgresql/8.1/bin/pg_config
[03:19] <pitti> StevenK: /usr/bin/ just provides the version that client side packages should use (always the latest libpq)
[03:19] <StevenK> Ahhhh
[03:25] <StevenK> /usr/bin/ld: cannot find -lgcj
[03:26] <pitti> StevenK: is it not in libgcj8-dev?
[03:27] <StevenK> /usr/lib/gcc/x86_64-linux-gnu/4.2/libgcj.so is in libgcj8-dev
[03:28] <StevenK> If ld can cope with that
[03:28] <pitti> StevenK: hm, that's doko's realm, but I'd think it should be /usr/lib/libgcj.sdo
[03:28] <pitti> .so even
[03:28] <pitti> bah, my keyboard has just broken, producing 'd's all over the place
[03:29] <StevenK> /usr/lib/libgcj.so.80 exists, but no plain libgcj.so
[03:29] <Hobbsee> ...then stop hitting the d key?
[03:29] <pitti> Hobbsee: it happens by pressdingd any other keys
[03:29] <Mithrandir> dingding
[03:29] <StevenK> Shouldn't that be 'grrddrrs'? :-P
[03:29] <pitti> oh
[03:29] <Hobbsee> hehe :)
[03:30] <StevenK> Haha
[03:30] <pitti> brg
[03:30] <pitti> brb even
[03:30] <Hobbsee> the d's and the g keys are taking over!
[03:30] <StevenK> Keyboard drawers are a good thing...
[03:30] <StevenK> Considering I usually type without looking, with the drawer fully retracted.
[03:31] <StevenK> Which causes my small cousins issues like, "How do you know where the keys are?" and then getting dirty looks when I say, "They're under my fingers of course."
[03:32] <infinity> "I can tell where the keys are, based on the whitespace surrounding them.  If you were a Python freak like me, you'd understand."
[03:33] <infinity> StevenK: Speaking of python freakiness, does linda still get love, or did you get bored with it?
[03:36] <StevenK> The answer is somewhat complicated.
[03:36] <StevenK> The incredibly simple answer is "No, since that equates to Debian work."
[03:36] <StevenK> infinity: libnss-db! :-P
[03:38] <doko> StevenK: yes, know issue (different gcc and gcj versions); maybe I should add this link in /usr/lib/jvm/java-gcj-compat/lib ...
[03:38] <doko> StevenK: or use gcc-4.2 for linking
[03:38] <jdstrand> are old security updated packages archived and if so where?  (eg, I am looking for bind9_9.3.2-2ubuntu1.2.diff.gz)
[03:41] <cjwatson> jdstrand: as far back as dapper, yes
[03:42] <geser> jdstrand: yes, on LP: http://launchpadlibrarian.net/6260265/bind9_9.3.2-2ubuntu1.2.diff.gz
[03:42] <cjwatson> jdstrand: https://launchpad.net/ubuntu/+source/bind9 -> 9.3.2-2ubuntu1.2 -> dapper i386 (or whatever) -> Resulting binaries -> pick one
[03:42] <cjwatson> or what geser said for the source
[03:42] <jdstrand> cjwatson, geser: thanks
[03:43] <pitti> one key dismounting - vacuum - clean - key remounting cycle later, /me sets no-more-errant-d-keystrokes to implemented
[03:44] <Hobbsee> hurrah!
[03:45] <StevenK> doko: Actually, debian/rules contains CPPFLAGS += -I/usr/lib/jvm/java-gcj/include/, should I change that?
[03:45] <geser> pitti: luckily it wasn't a stuck 'y' key: "Are you sure to proceed?" :)
[03:47] <pitti> StevenK: you should rather be concerned about -L flags
[03:49] <StevenK> And fix the pg_config hack properly
[03:49] <StevenK> Aww, it seems infinity is ignoring me.
[03:50] <doko> StevenK: no, this is correct
[03:50] <pitti> StevenK: the three magic letters 'nss' terminate his IRC link
[03:50] <Hobbsee> a new exploit!
[03:50] <StevenK> pitti: Or at least, cause his screen to detach automatically
[03:51] <Fujitsu> Peril-sensitive screen sessions. I like it.
[03:51] <StevenK> pitti: I'm so tempted to beg the archive admin to paste the command-line and result when I get yada demoted like the pox that it is.
[03:52] <pitti> StevenK: it's not yet in anastacia
[03:52] <pitti> StevenK: ah, right, the only remaining rdepends is ... *cough* libnss-db
[03:53] <StevenK> Exactly ...
[03:53] <StevenK> Which is why I keep hitting up infinity.
[03:53] <pitti> sorry, infini*bzzzzt*
[03:56] <Mithrandir> > pkg-config --variable pyexecdir pygtk-2.0
[03:56] <Mithrandir> /usr/lib/python2.4/site-packages
[03:56] <Mithrandir> hmm, that doesn't look right, does it?
[04:01] <sladen> Mithrandir: I was trying just that the other day.  It didn't work
[04:02] <sladen> (trying to get pyclutter to work)
[04:02] <Mithrandir> pyclutter?  This is something else entirely
[04:02] <sladen> it depended on testing `pkg-config pygtk-2.0`
[04:12] <StevenK> GAH
[04:12] <StevenK> You stupid build system! *WHO* came up with using $(COMPILER), I'll kill them myself.
[04:12] <Mithrandir> StevenK: having fun with libnss-db?
[04:13] <StevenK> Nope, postgresql-pljava
[04:13] <StevenK> Mithrandir: I'm done with libnss-db, but infinity doesn't love me enough to sign off on it.
[04:13] <thom> StevenK: kill dexter
[04:14] <thom> on principle, relevancy is irrelevant
[04:14] <StevenK> I'm sorely tempted.
[04:16] <pitti> hi Keybuk
[04:16] <Keybuk> \o/  finally emptied my INBOX
[04:16] <Hobbsee> heya Keybuk!
[04:17] <StevenK> Keybuk: Select everything ... delete, it shouldn't be ardous. :-P
[04:17] <StevenK> arduous, even
[04:18] <mvo> hey Keybuk
[04:19] <Mithrandir> Keybuk: you've failed to answer my suggestion for call time tomorrow, it seems?
[04:19] <alesan> anyway, can somebody explain me how to modify the usplash default logo? I want to create my own version of ubuntu with custom packages drivers etc
[04:19] <pitti> Keybuk: what Mithrandir said, me2
[04:19] <Keybuk> those are new mails :p
[04:19] <Keybuk> actually, I just replied to them all
[04:19] <pitti> ah, cool
[04:20] <StevenK> I am so going to break postgresql-pljava's build system and its author into little pieces.
[04:21] <StevenK> What's the point of using make vars if I can't export (and actually override them) in debian/rules
[04:21] <Riddell> alesan: look at the source of any *-artwork-usplash packages
[04:28] <cjwatson> Keybuk: any strong opinions on the best way to add support to initramfs for having the root filesystem be loop-mounted from an image on another filesystem?
[04:28] <cjwatson> Keybuk: (for installer-for-windows, but I'd like it to be generic and controlled by a boot parameter)
[04:29] <Keybuk> ROOT=LOOP= or something?
[04:29] <Keybuk> though you would need to specify the filesystem with the loop on it, and the path to the loop
[04:29] <cjwatson> mm, I was contemplating a different argument for the hosted filesystem so that they can both be UUID-style
[04:30] <Keybuk> it might make more sense to still keep root= the same, but if loop= is set, mount that after mounting the root
[04:30] <cjwatson> e.g. host=UUID=... ROOT=UUID=... and have it mount host= first
[04:30] <Keybuk> root=UUID=...
[04:30] <Keybuk> loop=/path/to/fs/on/root
[04:30] <cjwatson> that might be more straightforward to implement, yes
[04:30] <cjwatson> I guess UUIDs won't help for locating the image, thinking about it :)
[04:31] <cjwatson> and mount -o move ${rootmnt} /host or something
[04:31] <Keybuk> yeah
[04:31] <Keybuk> if [ -n "$LOOP" ]  ...
[04:31] <Keybuk> after mounting $ROOT to move it all out the way again and mount the image
[04:31] <Keybuk> that way you can use the existing device-exists-and-vol_id-works loop
[04:31] <cjwatson> yeah, that was the thing I was concerned about not duplicating
[04:32] <cjwatson> should I use a local-bottom script, do you think, or just bang it all into scripts/local:mountroot?
[04:32] <cjwatson> I guess the problem with a local-bottom script is that everything else would have to depend on it to ensure correct behaviour
[04:32] <Keybuk> put it in mountroot if I were you
[04:33] <cjwatson> I think I also need to expose /host in the real root filesystem so that /boot can be bind-mounted onto it
[04:33] <cjwatson> little maze of twisty filesystems, all alike
[04:39] <cjwatson> I wonder if I can mount /root; mount -o move /root /host; mount -o loop /host/blah /root; if [ -d /root/host ] ; then mount -o move /host /root/host; fi
[04:41] <Keybuk> yes you can
[04:52] <StevenK> pitti: Right, I finally convince postgresql-pljava to actually build.
[04:52] <StevenK> convinced, too
[04:52] <pitti> StevenK: wow
[04:53] <StevenK> The build system in it sucks hard.
[04:54] <StevenK>  4 files changed, 32 insertions(+), 1 deletion(-)
[05:08] <asac> stgraber: yt?
[05:27] <kylem> anyone running gutsy care to test a new intel driver release?
[05:28] <kylem> deb http://people.ubuntu.com/~kyle/testing/xf86-video-intel-RC/ ./
[05:47] <Tonio_> seb128: ping ?
[05:48] <seb128> Tonio_: sort of pong, I was about to do a break ;)
[05:48] <Tonio_> oki
[05:48] <seb128> ask your question
[05:48] <seb128> I'm not in a hurry ;)
[05:48] <Tonio_> seb128: can you drop the kdebluetooth upload I just commited ?
[05:48] <Tonio_> seb128: bad call in my bash-history, should have build and not upload....
[05:48] <Tonio_> seb128: it is just a testing package, not intended to be publish right now....
[05:49] <seb128> Tonio_: no such package in queue, is that a new package or a normal upload?
[05:49] <Tonio_> seb128: normal upload
[05:49] <Tonio_> seb128: I was wondering who can discard this
[05:49] <seb128> Tonio_: dunno how to reject those and if that can be done
[05:50] <Tonio_> seb128: argh...
[05:50] <seb128> the best way is to do a new upload reverting the changes
[05:50] <Tonio_> seb128: I'll do that then....
[05:50] <Tonio_> bye
[05:51] <ogra> seb128, enjoy
[05:59] <iwj_> doko: Thanks for disposing of that silly bug 126994.
[05:59] <ubotu> Launchpad bug 126994 in python-defaults "update-alternatives does not include python or gcc" [Undecided,Won't fix]  https://launchpad.net/bugs/126994
[06:00] <allee> Tonio_: hi
[06:00] <allee> Tonio_: were can I get the new debs (for testing and alioth merge)?
[06:02] <cjwatson> mvo: why does command-not-found emit its messages to stdout?
[06:02] <cjwatson> mvo: I think this is a bug and they should go to stderr; I have a somewhat screwed system at the moment that goes like this after a failed fsck:
[06:02] <cjwatson> bash: groups: command not found
[06:02] <cjwatson> bash: lesspipe: command not found:
[06:02] <cjwatson> bash: The: command not found
[06:02] <cjwatson> The program 'apt-get' is currently not installed. [...] 
[06:02] <cjwatson> (I have no idea why it's calling apt-get there ...)
[06:03] <Tonio_> allee: http://ubuntu.tonio.homelinux.org/
[06:03] <allee> Tonio_: thx
[06:03] <ogra> cjwatson, do you know if there is a non harmful way to replace a debconf note message in an udeb without turning it into an error ?
[06:04] <ogra> i heard notes should go away
[06:04] <cjwatson> ogra: replace it with what?
[06:04] <Tonio_> allee: the only thing I changed in the debian packaging is a patch to hide a few desktop files, which don't need to be in the kmenu as they are all accessible via the systray
[06:04] <cjwatson> ogra: oh, notes are still fine in udebs
[06:05] <ogra> cjwatson, ah, ok, the debian bug i got for ltsp sounded like it would be cancelled there as well
[06:05] <allee> Tonio_: makes sense.  I've check and ping upstream for 'commit okay'
[06:05] <cjwatson> ogra: it's possible somebody said that notes are deprecated in udebs too, but they're mistaken :-)
[06:06] <carlos> pedro_: hi
[06:06] <Tonio_> allee: btw I don't know if such a patch should be merged upstream.... I'd say yes, but you decide
[06:06] <cjwatson> mvo: ah, in fact apt-get is called because .bashrc evaluates the output of lesspipe which includes command-not-found suggesting an apt-get command to run
[06:06] <allee> Tonio_: I'm not upstream :)
[06:06] <pedro_> hello carlos!
[06:06] <ogra> cjwatson, calming :) i wouldnt want a red background just to tell the user something :) they want to make it an error
[06:07] <Tonio_> allee: hehe
[06:08] <cjwatson> mvo: do you mind if I file a bug and just make the relevant changes on bazaar.launchpad.net directly?
[06:08] <mvo> cjwatson: not at all, that is very welcome
[06:11] <Tonio_> allee: fyi, just checked, it crashes here :)
[06:11] <allee> Tonio_: :)
[06:12] <allee> Tonio_: why this strange version and not 1.0~beta4 ?
[06:15] <cjwatson> mvo: committed; would you upload it? it seems to have a weird source package build process I can't quite manage to reproduce accurately
[06:17] <Tonio_> allee: can you remind me of the alioth url for kdebluetooth please ?
[06:17] <asac> calc: do you have a minute testing a ipw3945 workaround?
[06:18] <Tonio_> allee: this strange version because it is not the one that should have been uploaded, it was just for personal test :)
[06:18] <mvo> cjwatson: sure, happy to do so (does bzr-buildpackage --native not work?)
[06:19] <cjwatson> mvo: I normally use debuild -S
[06:20] <cjwatson> mvo: it has a debian/arch-build target, and .bzrignore and a scan.data file were missing from the resulting source package when I built it
[06:20] <mvo> cjwatson: I will check it, the missing scan.data  is quite possible  :/
[06:21] <lamont> debuild -S with the right stuff in ~/.devscripts is love
[06:36] <calc> asac: sure, i'll have to grab my laptop
[06:39] <calc> asac: booting into gutsy cd now
[06:40] <Riddell> cjwatson: I copied out the seed diagram from the sprint into a nicer format http://kubuntu.org/~jriddell/tmp/seeds.png
[06:40] <Riddell> cjwatson: is there a seeds wiki page I can link to there from?  I can't seem to find one just now
[06:41] <calc> asac: ping
[06:43] <asac> calc: ok
[06:43] <asac> calc: do you have a process ipw3945d* running?
[06:43] <calc> asac: i had to reboot it failed to start X for some reason, grr
[06:43] <asac> hmm
[06:43] <calc> asac: i'll look for that as soon as it finishes booting
[06:44] <calc> ok it worked this time, loading gnome now
[06:47] <calc> grr now its hanging before the gnome splash comes up
[06:47] <asac> well ... i think we need to test when i return then :)
[06:47] <asac> calc: http://ubuntuforums.org/showthread.php?t=492386&page=3
[06:47] <asac> there is a potential workaround at the end
[06:47] <asac> maybe test that
[06:47] <calc> ok looking
[06:47] <calc> maybe it will boot all the way this time :)
[06:48] <asac> e.g. restart the " regulatory daemon"
[06:48] <calc> asac: interesting ok i'll test that as soon as i get it up
[06:54] <kylem> didn't we have tribe3 troubles at the sprint?
[06:55] <calc> kylem: not sure, i didn't try tribe3 at that time :\
[06:55] <calc> i heard there might be some kind of issue with it
[06:55] <calc> but not sure what
[06:56] <asac> tribe3 livecd has fs issues iirc
[06:56] <calc> asac: ah that might explain what I am seeing
[06:59] <cjwatson> Riddell: SeedManagement
[07:00] <calc> asac: restarting the daemon doesn't seem to help
[07:00] <calc> asac: at least in getting NM to connect to a network
[07:00] <cjwatson> Riddell: thanks!
[07:01] <asac> calc: well i would like to see if time needed wpasupplicant alone is improving
[07:01] <asac> e.g. just give wpasupplicant essid + driver ... in ap_scan mode 1
[07:01] <asac> then see how long it takes until it associates
[07:02] <calc> ok
[07:02] <asac> it took stgraber about 2-3 minutes ... which is too long
[07:02] <keescook> bryce: can you take a look at http://revu.tauware.de/details.py?upid=6210 ?
[07:03] <bryce> ok
[07:03] <asac> ok out ... bbl
[07:04] <bryce> ah, openchrome's been on my todo list for some time
[07:07] <bryce> keescook: the diff looks good and clean
[07:08] <bryce> keescook: I'm curious what "svn357" means
[07:08] <keescook> I imagine that's subversion snapshot from rev id 357
[07:10] <calc> asac: its taking more than 3min on an open AP
[07:11] <calc> asac: i'll try the wpa one
[07:11] <calc> asac: the timeout times may be faster though from the looks of it
[07:12] <bryce> ahh, good call
[07:12] <calc> asac: i gave it 4.5min and it still didn't connect to the open AO
[07:12] <bryce> keescook: yes I think it'd be good to upload
[07:14] <AngryElf_> how would I get in touch with the team/person that packages FF?
[07:18] <cjwatson> AngryElf_: #ubuntu-mozillateam; the lead maintainer just left for the evening though
[07:18] <cjwatson> there's a corresponding mailing list too
[07:22] <calc> asac: after playing around with associating with the open ap when i tried the wpa ap it did it immediately, so i'm rebooting now to see how long it takes on a fresh boot
[07:37] <afflux> Is it safe/recommended to use the reboot script provided by upstart-compat-sysv or sysvinit from an application like that one: http://cs.unm.edu/~dmohr/grub.php (I'm working with upstream to get this packaged for ubuntu)
[07:37] <calc> asac: ap_scan can't even find my wpa AP before trying to attach to the open ap from the look of it
[07:39] <calc> asac: interesting... :)
[07:39] <calc> asac: when i killed and restart daemon it could see both, doing further testing
[07:42] <calc> asac: it saw it at first but checked later and it was no longer in the list
[07:50] <superm1> Hi guys, when doing usplash artwork, how is the pallete limited on the progress bar?  Is it by the colors used in the full screen image?  I made both indexed images, but the color of the progress bar appears wrong during splash
[08:01] <mjg59> superm1: The progress bar has to use the same palette as the rest of the image
[08:02] <superm1> that would likely be the issue then
[08:02] <superm1> k
[08:03] <cjwatson> afflux: shouldn't that GUI just call grub-set-default?
[08:06] <mr_pouit> could an archive-admin remove avifile-win32-plugin for amd64 from the buildds? It is not built anymore in gutsy (only for i386), and avifile FTBFS on amd64 because it still tries to build it :/ (cf. http://launchpadlibrarian.net/8598963/buildlog_ubuntu-gutsy-amd64.avifile_1%3A0.7.47.20070718-1ubuntu1_FAILEDTOBUILD.txt.gz)
[08:06] <afflux> cjwatson: err, yes. I'll tell upstream, thank you :)
[08:20] <geser> mr_pouit: I'd say you need to patch avifile to call dh_gencontrol in debian/rules with the right package set on amd64 (and similar archs)
[08:29] <wereHamster> what's Ubuntu's policy on RPATH/RUNPATH?
[08:35] <calc> wereHamster: i'm pretty sure its the same as debian's which is don't do it
[08:35] <wereHamster> even RUNPATH is discouraged?
[08:36] <calc> RPATH is, i'm not sure about runpath, never actually heard of it
[08:36] <wereHamster> calc, do you have wine installed right now?
[08:37] <calc> no
[08:39] <wereHamster> marcel@Shuttle:~$ objdump -x `which wine` | grep PATH
[08:39] <wereHamster>   RPATH       $ORIGIN/../lib
[08:39] <wereHamster> argh.. why?
[08:55] <calc> grr gateway released a really nice new laptop less than a month after i bought mine :\
[08:56] <calc> 250gb hd, 2gb ram, n wireless, etc :\
[08:58] <kopfgeldjaeger> hi
[08:58] <wereHamster> what is needed to compile 32bit apps on ubuntu?
[08:58] <kopfgeldjaeger> build-essential
[08:58] <kopfgeldjaeger> if youre on a 32-bit ubuntu
[08:59] <kopfgeldjaeger> if youre on a 64 bit one, try chroot
[08:59] <kopfgeldjaeger> i have made a bash script, that downloads videos from video sharing sites (youtube, google video, etc.). it uses a webpage and filters the link... any chance to get such little things in universe? cause theres nothing about youtube and so...
[09:01] <wereHamster> chroot? Don't you have special 32bit packages that [provide 32bit libc.so etc?
[09:02] <calc> wereHamster: not for every 32 library, but for libc yes
[09:02] <kopfgeldjaeger> hm.. im quite sure that this exists, but if this is easier than a chroot...
[09:03] <superm1> well wouldn't it be possible to have a 32 bit pbuilder base to just use instead of maintaining a chroot?
[09:03] <calc> superm1: yea probably so
[09:04] <superm1> well i've yet to try sbuild, so pbuilder has done the trick for me normally
[09:05] <Chipzz> kopfgeldjaeger: that already existed ;P
[09:05] <mathiaz> superm1: The Pbuilder how-to explains to setup a 32bits pbuilder on a amd64.
[09:05] <geser> kopfgeldjaeger: something like youtube-dl or clive?
[09:05] <mathiaz> superm1: https://wiki.ubuntu.com/PbuilderHowto
[09:05] <kopfgeldjaeger> yeah
[09:06] <superm1> mathiaz, ah good, so i was right :)
[09:06] <kopfgeldjaeger> but NOT in the repos and not for so many sites... well, im sure there is something for more than youtube, but i dont think that theres so much support as the download sites give
[09:07] <geser> !info clive gutsy
[09:07] <ubotu> clive: Video extraction utility for YouTube and Google Video. In component universe, is optional. Version 0.2.0-1 (gutsy), package size 30 kB, installed size 172 kB
[09:07] <mathiaz> superm1: I had to modify my pbuilder script to support arch base files.
[09:07] <geser> !info youtube-dl gutsy
[09:07] <ubotu> youtube-dl: download videos from youtube.com. In component universe, is extra. Version 2007.06.22-1 (gutsy), package size 7 kB, installed size 64 kB
[09:07] <wereHamster> calc, do you have any idea how the 32bit glibc pakcage is named?
[09:08] <geser> !info ia32-libs
[09:08] <ubotu> Package ia32-libs does not exist in feisty, feisty-seveas
[09:08] <kopfgeldjaeger> well, apt-cache search sux :d
[09:08] <calc> libc6-something
[09:08] <calc> its only available on 64bit platforms though
[09:08] <kopfgeldjaeger> ...
[09:09] <geser> !info libc6-i386
[09:09] <ubotu> Package libc6-i386 does not exist in feisty, feisty-seveas
[09:09] <geser> !info libc6-i386 gutsy
[09:09] <ubotu> Package libc6-i386 does not exist in gutsy
[09:09] <calc> hmm not sure what it is named
[09:10] <calc> search for libc6* in apt-cache if you are on amd64
[09:10] <geser> gah, ubotu doesn't know about package on other archs
[09:10] <calc> or apt-cache search 32
[09:10] <geser> libc6-i386 - GNU C Library: 32bit shared libraries for AMD64
[09:10] <kopfgeldjaeger> !info network-manager-gnome gutsy
[09:10] <ubotu> network-manager-gnome: network management framework (GNOME frontend). In component main, is optional. Version 0.6.5-0ubuntu8 (gutsy), package size 138 kB, installed size 1752 kB
[09:10] <calc> geser: ok
[09:10] <superm1> http://packages.ubuntu.com/cgi-bin/search_packages.pl?keywords=libc6&searchon=names&subword=1&version=gutsy&release=all will show them all
[09:26] <wereHamster> what version of binutils does ubuntu use? I've heard that the nwer versions have --enable-new-dtags enabled by default, but ubuntu's binutils don't
[09:29] <superm1> !info binutils gutsy
[09:29] <ubotu> binutils: The GNU assembler, linker and binary utilities. In component main, is optional. Version 2.17.20070718cvs-0ubuntu2 (gutsy), package size 1414 kB, installed size 7376 kB
[09:33] <cjwatson> wereHamster: IIRC RUNPATH is ok
[09:33] <cjwatson> but not RPATH; use of it has burned us (well, Debian) in the past and is thus discouraged
[09:33] <wereHamster> superm1, thanks (doesn't your nick have one letter too much? :P )
[09:34] <wereHamster> cjwatson, ubuntu's wine uses RPATH but not RUNPATH which would override RPATH
[09:35] <cjwatson> it probably gets a lintian warning for that then ...
[09:35] <wereHamster> binutils-2.17 seems recent enough, I'm wondering why it doesn't use --enable-new-dtags by default
[09:36] <cjwatson> I don't recall exactly though, it's been years
[09:36] <wereHamster> because here on gentoo I don't have to pass anything extra to get the wine binary compiled with RUNPATH
[09:39] <tarzeau> i have a bug in ubuntu, and i have a fix, can i get karma when i close it?
[09:41] <asisak> tarzeau: you get some karma points, but if you are not familiar with all the details of bug triage, you should ask on #ubuntu-bugs
[09:46] <cjwatson> you can only close it if you actually do the upload, in any case
[09:47] <kopfgeldjaeger> I have a working ubuntu repository (test). How to create the Releases-file? Packages.gz and Packages.gpg already exist. i want to sign my packages
[09:47] <cjwatson> contributed fixes would normally just be an attachment with the patch checkbox checked
[09:47] <kopfgeldjaeger> or do i have to sign the packages themselves?
[09:47] <cjwatson> kopfgeldjaeger: apt-ftparchive release /path/to/repository
[09:47] <kopfgeldjaeger> thanx
[09:47] <cjwatson> and then create a detached signature for the resulting Release file with gpg
[09:51] <kopfgeldjaeger> ok. what does "unknown error while executing gpgv" mean? (translated from german)
[09:51] <kopfgeldjaeger> that comes when apt-get-updating
[09:52] <kopfgeldjaeger> apt-ftparchive release /home/andy/repo/ > Releases      was ok, or?
[09:52] <cjwatson> it's Release not Releases, and the error probably means that the format of Release.gpg is wrong
[09:52] <kopfgeldjaeger> i mean, instead of Releases "Release"
[09:52] <kopfgeldjaeger> no ascii?
[09:53] <kopfgeldjaeger> still same problem with non-ascii Release.gpg
[10:08] <wereHamster> superm1, is your binutils version patches somehow? if yes, can I see the patches you're using?
[10:08] <superm1> wereHamster, That is the binutils version in gutsy
[10:08] <superm1> so not just the one i'm using
[10:09] <superm1> the entire package is available here: http://packages.ubuntu.com/gutsy/devel/binutils
[10:15] <wereHamster> superm1, I'm asking because for example gentoo patches binutils to enable the new dtags by default. Ubuntu doesn't do it AFAICS becuause packages are still compiled with only RPATH
[10:19] <superm1> wereHamster, perhaps you will want to try a local build with adding --enable-new-dtags and see if that is the case
[10:19] <superm1> ah i see
[10:20] <superm1> well if you set yourself up a gutsy chroot, you can at least experiment
[10:20] <superm1> gentoo has an ebulid for debootsrap i believe
[10:21] <wereHamster> http://www.winehq.org/pipermail/wine-patches/2007-July/042051.html
[10:21] <wereHamster> I hope that will do the trick
[10:22] <alesan> do you have an idea if it's possible to go back from the "lss" isolinux image format to the original image in png or similar formats?
[10:22] <wereHamster> I finally found out why yukon/wine wouldn't work on ubuntu.. a least I hope I did :)
[10:54] <alesan> I would really like to understand the isolinux bootlogo... what is the source image, how was it prepared?
[11:05] <CydeWeys_> Can some Ubuntu person please check this URL and see why it isn't properly gzipped on the server?  http://us.archive.ubuntu.com/ubuntu/dists/feisty/main/binary-i386/Packages.gz
[11:05] <CydeWeys_> It's causing apt-get update to fail.
[11:05] <CydeWeys_> Because apt-get expects it to be gzipped, tries to gunzip a text file, and dies.
[11:07] <elmo> CydeWeys_: works for me
[11:07] <elmo> CydeWeys_: that file hasn't been modified in a long time, it's more likely you have a broken web cache between you and the server
[11:08] <CydeWeys_> Hrmm, so what do I do?
[11:08] <elmo> try 'apt-get -o Acquire::http::No-Cache=1 update'
[11:10] <CydeWeys_> elmo: See PM
[11:10] <elmo> CydeWeys_: I don't have any - are you registered with freenode?
[11:10] <elmo> CydeWeys_: (if not, I suggest a pastebin)
[11:11] <CydeWeys_> elmo: http://pastebin.ca/639777
[11:12] <elmo> CydeWeys_: that's with the -o Acquire?
[11:12] <CydeWeys_> elmo: Yes
[11:12] <elmo> CydeWeys_: rm /var/lib/apt/lists/us.archive* and try again
[11:14] <kopfgeldjaeger> wooooowo
[11:15] <kopfgeldjaeger> i just made an apt repository, with gpg verification!
[11:15] <kopfgeldjaeger> *g
[11:15] <CydeWeys_> elmo: Still the same error.
[11:15] <kopfgeldjaeger> only took 5 hours
[11:15] <CydeWeys_> elmo: Although it does go through and re-download all of the other ones.
[11:15] <elmo> CydeWeys_: ok, in that case, you definitely have a broken web cache between you and the intarwebs
[11:16] <elmo> CydeWeys_: you could change from us.archive to pony.archive in your sources.list, that would work around it for now
[11:16] <elmo> CydeWeys_: but I suggest you bug your ISP
[11:16] <CydeWeys_> elmo: How many ISPs keep web caches?
[11:17] <elmo> CydeWeys_: I don't know any that don't
[11:23] <CydeWeys_> elmo: pony worked, thanks
[11:24] <CydeWeys_> This ISP has been nothing but problems.
[11:24] <CydeWeys_> It's some kind of business T1.
[11:25] <CydeWeys_> elmo: Is there any harm in sticking with pony in the long run?
[11:25] <Nafallo> lol, pony :-)
[11:26] <CydeWeys_> Hey, don't make fun of me.  As far as I'm concerned, pony is the best server, because it's the only one that's working for me.
[11:26] <elmo> CydeWeys_: right now, pony.archive == us.archive == archive.ubuntu.com.  in the future, us.archive might become an actual host in the US.  so while it won't cause you any problems, it may make things slower for you one day
[11:26] <elmo> CydeWeys_: whether that matters is up to you
[11:26] <CydeWeys_> Where is it all hosted now?
[11:26] <elmo> the UK
[11:27] <Nafallo> CydeWeys_: I didn't. I realised pony.archive was archive :-)
[11:27] <CydeWeys_> It kind of sounds like one of the dev's daughters named the server.
[11:27] <Nafallo> CydeWeys_: cydeweys.archive works as well...
[11:28] <CydeWeys_> Oh, so any wildcard subdomain will work?
[11:29] <Nafallo> except the ones pointing to other places. like se.archive, which I think would also have good speed to the US.
[11:29] <CydeWeys_> So now where did the pony idea come from, elmo?
[11:29] <Nafallo> but yea, I'll go back to dishes now ;-)
[11:58] <Chipzz> CydeWeys_: never seen a presentation by Jeff Waugh?
[11:59] <Chipzz> "No you can't have a pony!"
[12:00] <Chipzz> http://www.hadess.net/blog/images/cant-have-a-pony.png
[12:01] <superm1> Chipzz, i saw my first presentation by him at Ulive - quite something else.