[00:37] <YokoZar> ArneGoetje: I'm worried about weird bugs from symlinking what was a .otf to a .ttf since .ttf is a subset.  Should I?
[04:16] <literal> Please direct me elsewhere if I'm asking in the wrong place, but why doesn't the lib32asound2 package include a pulseaudio plugin? The libasound2-plugins package has it, and the descriptions for both packages are identical (they both mendion pulse).
[04:17] <literal> This is causing skype to crash on login unless I remove pulseaudio from my system completely.
[04:17] <literal> Running on 64bit karmic.
[04:18] <TheMuso> literal: It has to do with the way the 32-bit packages like lib32asound2-plugins are built. A fix is coming, not for that package, but for another package that will provide the pulse plugin.
[04:18] <literal> Ok, good to know.
[04:18] <literal> Which package will provide it?
[04:19] <literal> In the future, that is.
[04:21] <TheMuso> literal: ia32-libs
[04:22] <literal> ok
[06:04] <TheMuso> c
[06:04] <TheMuso> woops wrong tab
[06:52] <dholbach> good morning
[07:12] <jussi01> good morning dholbach
[07:12] <dholbach> hiya jussi01
[07:24] <dholbach> zul: can you take a loook at bug 378240?
[07:29] <dholbach> can somebody please have a look at bug 414298?
[07:29] <dholbach> i gave it an initial review already, but I'll feel better with another pair of eyes
[07:30] <dholbach> it blocks bug 414471
[07:32] <dholbach> ogra: does bug 373100 look good to you?
[07:33] <dholbach> asac: how does bug 323752 look to you?
[07:35] <dholbach> bug 416933 too
[07:41] <Ryan52> geez dholbach, you get crazy amounts of stuff done :)
[07:41] <Ryan52> even if indirectly by pinging others.
[07:43] <dholbach> Ryan52: let's hope they get to it and do it :-)
[07:44] <dholbach> just trying to clean up the sponsoring overview a bit for the release
[07:56] <pitti> Good morning
[07:58] <TheMuso> pitti: If you could get me that symaptic command-line for use in installing packages, hat would be useful, thanks.
[07:58] <TheMuso> pitti: Or if you could tell me where to find it, thats fine too.
[08:00] <pitti> oh, I thought I had it in jockey, but seems I'm using python-apt now
[08:00]  * pitti grabs out of early history
[08:03] <pitti> TheMuso: ah, there we go: http://bazaar.launchpad.net/%7Eubuntu-branches/ubuntu/hardy/restricted-manager/hardy/annotate/head%3A/RestrictedManager/RestrictedManagerGtk.py#L73
[08:03] <pitti> TheMuso: the "self.gtk_idle()" is for showing a bouncing progress bar
[08:04] <pitti> TheMuso: that's why I eventually replaced it with python-apt, then you get proper numerical progress
[08:11] <TheMuso> pitti: hrm ok thanks.
[09:06] <asac> dholbach: 416933 looks good. confirmed
[09:10] <asac> dholbach: the other not. i commented and removed sponsors
[09:33] <ogra> help ! why is my inbox being flooded with rosetta mail ?
[09:41] <pecisk> is loosing icons of all main menu categories identional or could be a bug? (after upgrade) If first one, then it is worst decision ever made by gui guys :)
[09:54] <pochu> pecisk: it's a design decision, but I think it's been reverted
[09:55] <pecisk> I think is going from one extreme to another
[09:55] <pecisk> it's
[09:55] <pecisk> too much icons are burden for eyes, but also they help
[10:00] <loic-m> Is the removal of the possibility to change the number of workspaces also a design decision?
[10:07] <pecisk> loic-m, smells like bug
[10:08] <pecisk> it just doesn't change value
[10:08] <pecisk> it = dialog
[10:08] <loic-m> What package is workspace-switcher in?
[10:10] <pecisk> loic-m, gnome-applets?
[10:10] <pecisk> not sure
[10:12] <loic-m> nope, at least the files installed doesn't look like it
[10:12] <pecisk> loic-m, or it could be gnome-panel
[10:14] <seb128> loic-m, it's a known bug
[10:14] <loic-m> Idoesn't appear to be it
[10:14] <loic-m> seb128: started with Karmic Alpha 5 (different from the old bug)
[10:15] <seb128> pecisk, it's not a decision made by us but by GNOME
[10:15] <seb128> loic-m, what old bug?
[10:15] <pecisk> seb128, ohhhh
[10:15] <loic-m> Old bug was only the login at install could change the number of workspaces, no new user
[10:15] <pecisk> seb128, hopefully it will be reverted in Ubuntu?
[10:15] <seb128> pecisk, dunno yet, but I guess that will change before karmic
[10:15] <pecisk> sight
[10:16] <seb128> loic-m, dunno about this bug but the current one is a glade->gtkbuilder error
[10:16] <loic-m> In Karmic Alpha 5, even the login used at install can't, because the number of column/rows is stuck to 0
[10:16] <seb128> I expect it will be fixed today or tomorrow
[10:16] <seb128> I never had the "old bug"
[10:16] <loic-m> Wasn't a problem in Alpha 3
[10:16] <loic-m> but fresh Alpha 5 install it's like that
[10:17] <loic-m> ok, thanks
[10:17] <loic-m> I'll use gconf in the meantime
[10:18] <pecisk> Jesus Christ, all main menu icons are gone
[10:19] <loic-m> seb128: at least without compiz, users added after install can't get more workspaces than 2 when right-clicking on workspace switcher. And AFAIR it's also the case with compiz light effect enabled
[10:20] <seb128> loic-m, again that bug is well known and will be fixed today or tomorrow
[10:20] <seb128> loic-m, no need to keep talking about it or describing it
[10:21] <loic-m> s/light/normal
[10:23] <loic-m> I can't help, I'm but a bot
[10:29] <tseliot> pitti: is "nvidia-graphics-drivers-71" in karmic or has it been removed already? If it's still there, can you remove it, please?
[10:30] <pitti> tseliot: still there
[10:30] <pitti> tseliot: doesn't work with current x.org any more?
[10:30] <tseliot> pitti: no, and Nvidia don't plan on fixing it any time soon
[10:31] <tseliot> pitti: therefore its removal is the only solution
[10:32] <pitti> tseliot: done
[10:32] <tseliot> pitti: thanks a lot
[10:39] <pitti> seb128: \o/ seems the retracers finally behave
[10:39] <seb128> pitti, good job!
[10:39] <pitti> weekend with wife away -> productive :)
[10:40] <pitti> (but boring to the point when I start doing nasty things, like cleaning the flat and clearing out my lockers and desk..)
[10:41] <seb128> lol
[10:46] <mr_pouit> does someone know if the new gdm supports changing its background picture? (I'm able to change the logo using gconf, but not the background)
[10:46] <seb128> mr_pouit, it uses the default system background image
[10:46] <seb128> ie the same that you have a default wallpaper
[10:46] <seb128> you can probably set the background image for the gdm user to change it
[10:47] <mr_pouit> is it /desktop/gnome/background/picture_filename?
[10:48] <mr_pouit> (because there is no background image at all here)
[10:48] <seb128> mr_pouit, correct
[10:48] <seb128> mr_pouit, weird, libgnome has a default value for that key
[10:50] <mr_pouit> mmh
[10:50] <mr_pouit> % gconftool-2 --owner gdm --get /desktop/gnome/background/picture_filename
[10:51] <mr_pouit> /usr/share/xfce4/backdrops/xubuntu-jaunty.png
[10:51] <mr_pouit> if this is correct, this is indeed weird, as I can't see it
[10:53] <seb128> mr_pouit, well this key has a value for you apparently no?
[10:54] <mr_pouit> seb128: yeah, but gdm doesn't display the image
[10:54] <seb128> mr_pouit, I don't know then
[10:54] <seb128> could be that it takes the default system value and not the gdm user one
[10:54] <seb128> I've not really played with that yet
[10:55] <james_w> anyone know if we are supposed to transition to the split zope.* packages from the zope3 package?
[10:57] <wgrant> james_w: You will have to eventually. Has Debian decided yet?
[10:57] <mr_pouit> ok, thanks anyway, I'll wait for an ubuntu theme that changes the background then :p
[10:57] <james_w> wgrant: they are in the process
[10:57] <james_w> wgrant: we have some packages synced, but they are still in universe
[10:58] <wgrant> james_w: Ah, great.
[10:58] <wgrant> james_w: Zope 3 has been split completely and renamed upstream, so we must follow.
[10:58] <james_w> wgrant: sure, but I want to know if we are doing it for karmic
[11:05] <geser> james_w: see bug #419262 and Debian already removed zope3 from unstable
[11:06] <james_w> geser: thanks
[11:06] <james_w> geser: teach me for working on bugs before AA duties :-)
[11:09] <al-maisan> wow! pulseaudio just had my CPU heat up to 93 degrees centigrade.
[11:10] <al-maisan> .. and it usually never exceeds 70 degrees.
[11:10] <pitti> !
[11:11] <al-maisan> I did a 'sudo apt-get upgrade' this morning, followed by a reboot. At some point pulseaudio crashed but I was working and too busy to notice until I saw the CPU temperature value :/
[11:16] <Laney> pulseaudio: Keeping you warm in Winter
[11:16] <al-maisan> yeah :)
[11:22] <soren> cjwatson: So, apparantly, I forgot to bzr push after my most recent Eucalyptus upload. Is there a trick to fix this (since both you and Dustin have pushed to the branch since then) or do I just merge your changes and then push?
[11:32] <cjwatson> soren: did you at least commit and tag locally?
[11:32] <soren> cjwatson: Alternatively, I could grab the branch currently on Launchpad and merge my commit into that one.
[11:32] <soren> cjwatson: I did, yes.
[11:32] <cjwatson> soren: then merge and push, it doesn't much matter which way round
[11:32] <cjwatson> soren: I strongly recommend using a bound branch ('bzr help bind') to avoid this problem in future
[11:33] <elmo> what's the process for requesting a change in component these days?  it was originally meant ot be part of the sync of rancid, but got missed
[11:34] <soren> cjwatson: It used to be bound, but that blew up in my face when the two of us where both committing a lot at the same time. I'll bind it again now that things are calmer.
[11:35] <cjwatson> soren: blew up how?
[11:35] <soren> cjwatson: I
[11:36] <cjwatson> mostly the explosions that happen with bound branches are explosions you want
[11:36] <cjwatson> i.e. "somebody else just committed, you need to update before you can commit"
[11:37] <soren> I just got annoyed by not being able to commit anything locally, because every time I tried, you had updated something, I'd fix the conflicts, commit again, and you had made another commit. I just couldn't keep up. I unbound the branch so that I could get something done and then I dealt with it afterwards.
[11:43] <pitti> soren: bzr rebase FTW :)
[11:43] <soren> pitti: I
[11:43] <soren> m not proud of it, but that's essentially what I ended up doing by hand.
[11:43] <soren> :)
[11:44] <pitti> rebase works quite well for such situations IMHO (but merging should just do as well, but leaves some litter in the history)
[11:45] <soren> rebase isn't that horrible if you're only dealing with your own patches. I don't like it if other people's patches are involved.
[11:46] <soren> Merges clearly show the context in which someone committed and tested their code. Reapplying their patches in a completely different point in history may have unwanted sideeffects, and it looks like the original committer's fault.
[11:53] <pitti> elmo: change component> if it's universe->main, subscribe ubuntu-mir to the bug, otherwise ubuntu-archive
[11:53] <cjwatson> mvo: did you have a chance to look at bug 387112 and figure out how we might resolve this? I've committed the debconf change but as mentioned in a comment it won't actually help *this* upgrade
[11:53] <elmo> pitti: reopen the old sync bug?
[11:54] <elmo> pitti: or open a new one?
[11:55] <pitti> elmo: reopening will do
[11:57] <al-maisan> Oh no, there we go again, pulseaudio trying to convert my laptop in a space heater :/
[11:57] <mvo> cjwatson: no, sorry. I have a look now
[11:57] <al-maisan> How does one debug a problem like this?
[12:01] <pitti> al-maisan: it takes 100% CPU?
[12:01] <pitti> al-maisan: usually, I'd run it in foreground mode with --debug (or --verbose or something like this) and see if it's looping on something
[12:01] <al-maisan> pitti: yes, I attached strace to it but it just sits there..
[12:01] <pitti> al-maisan: if that doesn't help, attach strace to it and see what it's failing on
[12:01] <pitti> al-maisan: ah, so it uses 100% CPU, but strace doesn't show anything?
[12:02] <al-maisan> pitti: yes
[12:02] <pitti> that shows that it's busy wit calculations
[12:02] <al-maisan> apparently
[12:02] <pitti> al-maisan: so foreground/debugging would be my next step, and then installing -dbgsym and attaching gdb, to see what it's looping on
[12:02] <al-maisan> pitti: OK .. thanks!
[12:02] <pitti> speaking of debugging, /me dives deeper into gdm
[12:03] <pitti> gdm is a hilarious pile of indirected and abstracted indirection abstractions *sigh*
[12:03] <al-maisan> pitti: I don't envy you :P
[12:07] <pitti> al-maisan: likewise, though :)
[12:07] <al-maisan> :)
[12:12] <mvo> cjwatson: we could patch the old debconf to remove the use statement when the upgrade starts - what do you think?
[12:12] <cjwatson> mvo: sounds workable
[12:12] <cjwatson> (horrible, but workable)
[12:12] <mvo> ok, I assign it to me then
[12:13]  * drhorrible mvo
[12:14] <cjwatson> michael vogt's sing-a-long upgrader
[12:14] <mvo> lol
[12:26] <TheMuso> If a ports kernel with a new ABI number FTBFS, plesae don't NBS the older kernel out of the archive till the newer one is available.
[13:22] <TomaszD> slangasek, hey, thanks for helping out, here's another curious case https://bugs.launchpad.net/ubuntu/+source/nautilus-share/+bug/425677
[13:22] <hyperair> hmmm? nautilus-share?
[13:22] <hyperair> not using translations eh..
[13:22] <TomaszD> yep
[13:22] <hyperair> this is a curious case.
[13:22] <hyperair> are you sure about this?
[13:22] <TomaszD> well, I have fully updated karmic in vbox with all the langpacks applied
[13:22] <TomaszD> and it's in English
[13:23] <hyperair> find /usr/share/locale -name 'nautilus-share*'
[13:23] <hyperair> try running that
[13:23] <TomaszD> no result
[13:23] <hyperair> then you don't have it installed
[13:24] <TomaszD> how come I can see the window then, in English
[13:24] <TomaszD> when I try to share a folder
[13:24] <hyperair> i meant that you don't have the nautilus-share translations installed
[13:25] <TomaszD> oh, well, yes.
[13:25] <TomaszD> that's a bug then.
[13:25] <hyperair> so it's not nautilus-share's bug
[13:25] <Laney> http://packages.ubuntu.com/search?searchon=contents&keywords=nautilus-share&mode=filename&suite=karmic&arch=any
[13:25] <hyperair> i'm not sure which package's involved though
[13:26] <hyperair> ah.
[13:26] <hyperair> the langpack-gnome-XX-base packages.
[13:27] <hyperair> TomaszD: please try installing the said package and trying again.
[13:28] <TomaszD> it's already installed
[13:28] <seb128> $ dlocate nautilus-share.mo
[13:28] <seb128> language-pack-gnome-de-base: /usr/share/locale-langpack/de/LC_MESSAGES/nautilus-share.mo
[13:28] <seb128> ...
[13:28] <seb128> what locale do you use?
[13:29] <TomaszD> pl
[13:29] <TomaszD> (pl_PL)
[13:29] <hyperair> TomaszD: http://packages.ubuntu.com/karmic/language-pack-gnome-pl-base
[13:29] <seb128> in fact translations are installed
[13:29] <seb128> nautilus-share probably doesn't init the gettext domain correctly
[13:30] <TomaszD> yes they are, I'm suspecting a bug in n-s
[13:30] <TomaszD> right, something like that
[13:30] <hyperair> hmm
[13:30] <hyperair> wrong directory.
[13:30] <hyperair> i believe it looks in /usr/share/locale
[13:31] <seb128> no, that's the the upstream dir
[13:31] <seb128> we strip those and install langpacks in locale-langpack
[13:31] <seb128> to avoid conflicts
[13:31] <hyperair> doesn't the path need to be updated?
[13:31] <seb128> the translations are found since the context menu entry is translated
[13:31] <seb128> but the tab is all english
[13:32] <hyperair> hmm, is that so?
[13:32] <hyperair> wait
[13:32] <hyperair> tab?
[13:32] <TomaszD> the dialog window
[13:32] <seb128> the nautilus preferences tab
[13:32] <hyperair> okay which parts are not translated exactly?
[13:32] <seb128> the nautilus tab
[13:32] <hyperair> is that all?
[13:32] <hyperair> what about the stuff inside it
[13:32] <seb128> and the dialog
[13:32] <hyperair> hmm
[13:32] <seb128> only the context menu entry is translated
[13:32] <hyperair> i see.
[13:32] <hyperair> this probably came from the glade->gtkbuilder transition
[13:33] <seb128> when clicking "share this folder" the dialog is translated
[13:33] <hyperair> eh?
[13:33] <TomaszD> huh
[13:33] <hyperair> okay, that is strange.
[13:33] <hyperair> ery strange
[13:33] <hyperair> the gettext domain is bound when nautilus_module_initialize is called
[13:34] <hyperair> if the preferences dialog didn't get translated, the standalone dialog shouldn't either.
[13:34] <TomaszD> it doesn't get translated either way for me
[13:34] <seb128> it's not
[13:35] <seb128> only the context menu item is translated
[13:35] <seb128> and the dialog asking if you want to install samba
[13:35] <TomaszD> this dialog about samba is actually translated, but I don't it's a part of nautilus-share
[13:36] <hyperair> it is part of nautilus-share.
[13:36] <TomaszD> ohh kay :)
[13:36] <hyperair> this sounds like the other translation issue i had with the standalone window
[13:37] <TomaszD> alright guys I'm way over my head with this, getting back to translating GNOME, the deadline is approaching fast
[13:38] <hyperair> mm have fun
[13:39] <TomaszD> by the way, policykit-gnome has an outdated translation template in launchpad, thus making it impossible to translate properly, I'm gonna report this in a moment
[13:40] <hyperair> seb128: do i need to call gtk_builder_set_translation_domain?
[13:40] <hyperair> hmm i'll try that and see what happens. TomaszD: could you test a patch when i upload it to a PPA later?
[13:41] <TomaszD> hyperair, sure, as long as info lands in my mailbox or someone calls me out here, I'll make some time to test it
[13:41] <TomaszD> I'll be here all day, GNOME is a huge project...
[13:41] <hyperair> TomaszD: thanks
[13:41] <hyperair> =)
[13:41] <hyperair> i'll be done in less than half
[13:42] <seb128> hyperair, you don't if it uses the standard domain for the software but it might be required in the nautilus case to set it again
[13:43] <hyperair> alright, i'll do that then
[13:43] <hyperair> seb128: basically just after gtk_builder_add_from_file right?
[13:44] <seb128> hyperair, yes
[13:44] <seb128> trying
[13:44] <hyperair> i have no useful locales to test
[14:13] <hyperair> my, my.. what's eating the buildds?
[14:13] <TomaszD> well that's scary, my karmic install just died.
[14:13] <hyperair> O_O
[14:13] <hyperair> sounds fun
[14:14] <pitti> hyperair: https://launchpad.net/builders/
[14:14] <hyperair> that's a lot of waiting builds
[14:14] <TomaszD> http://imgur.com/OoLh9.png
[14:15] <TomaszD> I just rebooted normally :(
[14:15] <hyperair> sounds like you screwed over your system clock nicely =)
[14:16] <hyperair> set it right and you're good to go =D
[14:16] <TomaszD> I have touched nothing
[14:16] <hyperair> or wait half an hour
[14:16] <TomaszD> fsck fixed it, but I doubt it was my fault there
[14:16] <hyperair> well something changed your system clock
[14:16] <davmor2> TomaszD: It's known bug
[14:16] <hyperair> oh O_o
[14:16] <TomaszD> ah, good to know.
[14:17] <hyperair> say, will there be a gnome-shell ppa?
[14:17] <hyperair> or at least something with the latest unstable?
[14:17] <seb128> hyperair, universe
[14:17] <hyperair> seb128: that's got .0. now is .2
[14:18] <hyperair> and of course, we can't get anything new in since featurefreeze is on
[14:18] <hyperair> unless gnome-shell gets an exception?
[14:18] <seb128> hyperair, right, some people have been on vac in august
[14:18] <hyperair> ah
[14:18] <seb128> hyperair, and required build-depends just got updated this weekened
[14:19] <seb128> hyperair, GNOME has a standing exception
[14:19] <hyperair> ah i see
[14:19] <hyperair> what are the conditions?
[14:19] <seb128> ?
[14:20] <hyperair> nevermind
[14:20] <seb128> GNOME has an exception that's it ...
[14:20] <hyperair> i see
[14:20] <seb128> that's this way since warty
[14:20] <hyperair> how come, though?
[14:20] <hyperair> out of curiosity
[14:21] <seb128> we trust GNOME schedule to enforce freezes, respect timelines, etc
[14:21] <hyperair> aah
[14:21] <seb128> gnome-shell is a bit of a special case but that's a technology preview and we said we would track GNOME3 components too
[14:23] <hyperair> special case meaning frozen or not frozen?
[14:24] <seb128> we will do the updates
[14:24] <hyperair> ah
[14:24] <seb128> it's a special case for GNOME since they don't follow freezes etc
[14:24] <seb128> so technically we could not grant exception for that one
[14:24] <seb128> but as we said we would track GNOME3 updates
[14:26] <seb128> geser, thanks for the empathy fix, don't forget to subscribe the sponsor team though when you have something that needs to be uploaded
[14:28] <hyperair> ah
[14:28] <geser> seb128: I usually do that but wasn't sure if subscribing ubuntu-main-sponsors is right here as the sponsor should ideally also be able to commit the changes to the ubuntu-desktop bzr (or can all core-dev commit there?)
[14:28] <seb128> geser, anybody who can upload can commit there
[14:29] <geser> thanks, will remember that
[14:30] <TomaszD> any gnome-settings-daemon wizards around? What is this "Share Folder" string in the template, where in the UI is that used? isn't it redundant, with nautilus-share?
[14:31] <TomaszD> sorry, that was meant for #ubuntu-translations
[14:31] <TomaszD> *translators
[14:32] <seb128> TomaszD, there is no such string in the source
[14:33] <TomaszD> oh crap, I meant gnome-system-tools, sorry!
[15:07] <fagan> Hey, im trying to run gnome-zeitgeist in karmic and I get this error http://pastebin.ubuntu.com/266684/  it works fine in jaunty.
[15:07] <fagan> Anyone have a clue what needs to be changed to fix it ?
[15:12] <fagan> bzr branch lp:~gnome-zeitgeist/gnome-zeitgeist/new-interface is the code thats failing
[15:16] <fagan> Hmmm this question is kinda borderline for this channel I suppose but all im really asking is what changed in python or maybe pygtk since jaunty
[15:41] <Turl> fta: hi, may I PM you?
[16:01] <hyperair> TomaszD: https://launchpad.net/~hyperair/+archive/bugfixes <-- grab the nautilus-share deb from here and install it.
[16:02] <TomaszD> hyperair, right, one moment
[16:05] <TomaszD> hyperair, no change, logged out and back again to be sure, but no change
[16:05] <hyperair> hrm damnit
[16:09] <hyperair> seb128: is this some bug in gtkbuilder?
[16:10] <seb128> I doubt it
[16:10] <seb128> it could be nautilus which reset the domain or something
[16:10] <hyperair> hmm
[16:10] <hyperair> that's painful
[16:10] <seb128> it's a small bug I would not say that
[16:14] <hyperair> which means digging needs to be done in nautilus code
[16:14] <hyperair> how painful
[16:14] <hyperair> actually i'm very tempted to just rewrite the entire thing to not use gtkbuilder at all
[16:14] <seb128> I doubt it's a nautilus bug, the totem tab is correctly translated
[16:14] <hyperair> hmm seriously?
[16:14] <seb128> yes
[16:15] <hyperair> question. is totem's done using gtkbuilder?
[16:15] <hyperair> or manually constructed
[16:17] <seb128> dunno
[16:19] <hyperair> hmm strings seems to show properties.ui and fullscreen.ui.
[16:19] <hyperair> but dpkg -L doesn't show
[16:19] <Keybuk> kees: around?
[16:21] <kees> Keybuk: a bit; on my android, on holiday. sup?
[16:21] <seb128> totem-common: /usr/share/totem/properties.ui
[16:21] <seb128> seems it's using gtkbuilder
[16:21] <Keybuk> kees: have a binary I wanted you to run, and supply the output of
[16:21] <Keybuk> in a "who do I know who loves to have a difficult filesystem setup" kind of way :p
[16:22] <kees> heh, sure. what is it?
[16:22] <elmo> a r00tkit
[16:22] <Keybuk> kees: it's the udev/upstarty mount thing
[16:22] <pitti> heh, it works for kees → it must work for everyone?
[16:22] <Keybuk> pitti: exactly ;)
[16:23] <pitti> I bet it breaks single-partition, no-raid, no-lvm, no-crypto default ext4 :-)
[16:23] <kees> if i don't have to reboot, i can do it now. otherwise i can run it when i get home tonight
[16:23] <Keybuk> kees: you don't have to reboot
[16:23] <Keybuk> kees: amd64 or i386 ?
[16:23] <kees> Keybuk: amd64
[16:24] <Keybuk> ok, two seconds
[16:24]  * kees waits for customized rootkit
[16:27] <cjwatson> "here's the source, but you need to use the gcc from this PPA"
[16:27] <Keybuk> http://people.canonical.com/~scott/mountall
[16:27] <Keybuk> run as
[16:27] <Keybuk> sudo ./mountall --debug --tmptime=infinity
[16:27] <Keybuk> and capture the output (tee would be good to use)
[16:28] <Keybuk> it'll almost certainly pause
[16:28] <Keybuk> at which point, run udev udevadm trigger --subsystem-match=block from another window
[16:28] <Keybuk> if you have NFS mounts, etc. it'll still pause, so run killall -USR1 mountall
[16:28] <Keybuk> it should eventually declare "finished"
[16:28] <kees> heh
[16:28] <Keybuk> (this should all be perfectly safe)
[16:28] <Keybuk> I'd like the debug output ;)
[16:29] <ion> keybuk: Mind sharing the sauce?
[16:29] <Keybuk> ion: it's in the Upstart package in the ubuntu-boot PPA
[16:29] <ion> Thanks
[16:29] <pitti> ion: what do you believe we are, free software or what?
[16:29] <kees> Keybuk: one moment...
[16:30] <pitti> ion: you have a complex setup, too?
[16:30] <ion> pitti: On my server, i have lvm over md. Nothing more complex than that.
[16:31] <pitti> ion: still, good test case probably :)
[16:32] <davmor2> Keybuk: how complicated would you like a setup and in what arch i'll let you have ssh access to a box you can break ;)
[16:32] <Keybuk> in theory, there should be no setup that this doesn't work on
[16:32] <Keybuk> I've tested it in the most arbitrarily complex setups that don't even work right now
[16:32] <RainCT> oops - can some archive admin please reject   gnome-voice-control - 0.4-0~ppa3?
[16:32] <Keybuk> so LVM-over-RAID-over-two-loop-devices-one-on-a-USB-key-one-on-NFS works
[16:33] <cjwatson> Keybuk: have you tried wubi?
[16:33] <Keybuk> and swap on that works too
[16:33] <pitti> RainCT: sorry, that doesn't work
[16:33] <cjwatson> that adds the extra constraint of one of the bits being implemented using FUSE
[16:33] <Keybuk> cjwatson: I'm not sure that wubi makes any difference, unless it needs special handling during boot (of which there is none in the current scripts ;)
[16:33] <pitti> RainCT: uploads are accepted straight into soyuz, they don't need to wait for a publisher in the "ACCEPTED" queue any more nowadays
[16:33] <RainCT> Oh. Well, it'll FTBFS anyway.
[16:33] <cjwatson> Keybuk: I thought there was code to shove the fuse process into omitpids
[16:33] <cjwatson> maybe that's later
[16:33] <cjwatson> Keybuk: there's definitely magic in the initramfs
[16:34] <Keybuk> cjwatson: right, this replaces the bits in rcS only
[16:34] <cjwatson> ah, ok
[16:34] <pitti> RainCT: do you plan to upload it to karmic?
[16:34] <cjwatson> you might have to care when you do umountall :-)
[16:34] <Keybuk> cjwatson: indeed ;)
[16:34] <pitti> RainCT: I stopped the upload cronjob for now, so we have some minutes to consider
[16:34] <cjwatson> pitti: in theory, if we get into the right five-minute window, we can rm it from ~lp_queue
[16:35] <RainCT> pitti: Yes, I plan to get a FFe for it within the next days (but it still needs some testing, I'm getting crashes here).
[16:35] <pitti> cjwatson: right, that's why I just stopped the process-upload.py cron
[16:36] <RainCT> (The "Accepted" mail was already send some minutes ago though. I didn't notice it until it closed the related bugs.)
[16:36] <pitti> ah, too late then, I think
[16:36]  * ion added the ubuntu-boot PPA to sources.list.d and now runs safe-upgrade. Let’s see whether i’ll still come online after the reboot. :-P
[16:36] <pitti> I see it in the queue, but I don't want to mess with that
[16:36] <Keybuk> ion: DON'T
[16:36] <Keybuk> STOP!
[16:36] <ion> Hah, ok
[16:36] <cjwatson> that good?
[16:37] <Keybuk> there's one or two key bits missing ;)
[16:37] <Keybuk> if this works, then I'll put those in
[16:37] <Keybuk> *then* people can test
[16:37] <Keybuk> but right now, it won't boot
[16:37] <pitti> cjwatson: you removed it from ./accepted/upload-20090907-162446-001727/ ? or what do you mean
[16:38] <RainCT> pitti: Oh well, thanks anyway. I'll fix it within the next days (or worst case upload a "really" package with the old version)
[16:39] <cjwatson> pitti: no, I didn't touch anything
[16:39] <Keybuk> kees: any luck?
[16:39] <cjwatson> pitti: if it's been accepted, there's not much sane we can do - build records have already been created
[16:39] <pitti> cjwatson: ah, "that good" referred to Keybuk's OMGNOO! :)
[16:39] <cjwatson> yeah
[16:41] <kees> Keybuk: yup, ran fine
[16:41] <Keybuk> kees: got the output?
[16:41] <kees> Keybuk: yup, sending you email (typing is slow on this tiny keyboard :) )
[16:42] <Keybuk> np
[16:42] <Keybuk> great
[16:42] <ion> /* FIXME look up underlying device
[16:42] <ion> I could contribute some “look up underlying device” code, since i already have a sh prototype somewhere. :-P
[16:42] <kees> Keybuk: ok, sent
[16:42] <Keybuk> ion: I was hoping you would
[16:43] <Keybuk> I left that out precisely because you said you were working on it <g>
[16:43] <ion> Hehe
[16:45] <Keybuk> kees: How Many Disks ?!
[16:45] <kees> physically or logically?
[16:45] <Keybuk> see, this is why I use you as a test case ;-)
[16:46] <kees> hehe
[16:47] <kees> it seems like it did the right thing with the fuse mounts, the autofs mounts and the unmounted lv chroots
[16:47] <cjwatson> come on, show the rest of us
[16:47] <Keybuk> yeah it should
[16:47] <Keybuk> kees: did it mount the unmounted things?
[16:48] <Keybuk> to me it looks like it left your system in the right state
[16:49] <Keybuk> cjwatson: lol, debug output of a tool is very uninteresting ;)
[16:51] <kees> Keybuk: no, i think it didn't change any mounts
[16:51] <cjwatson> oh I dunno, d-i's debug output has been very interesting in the past, e.g. http://www.ubuntu.com/usn/usn-262-1 ;-)
[16:51] <cody-somerville> cjwatson, Keybuk: http://www.ubuntu.com/community/processes/newdev seems to say that two people from the community council or just folks who are administrators of the ubuntu-dev team can approve developer applications. I've never heard of that before... does that page need updating?
[16:51] <cjwatson> cody-somerville: mm, that whole page looks a bit out of date
[16:52] <cjwatson> it would probably be better to turn most of it into wiki references
[16:52] <ScottK> james_w: Since today is a holiday in the US (so I assume slangasek is off today) I'd appreciate it if you'd put your archive admin hat back on and look at 425775.
[16:52] <cjwatson> I don't recall that two-community-council-members rule
[16:52] <Keybuk> right, the CC can't approve ubuntu developer membership
[16:52] <Keybuk> (they don't even have the requisite LP powers)
[16:52] <cjwatson> administrators of the ubuntu-dev team => techboard (well, nowadays, developer-membership-board)
[16:53] <kees> cjwatson: people.canonical.com/~kees/mountall.debug if you really want to see it :)
[16:53] <cody-somerville> Is the only needing two people from the TB also incorrect?
[16:53] <ScottK> james_w: It should help with some armel build failures.
[16:53] <cjwatson> that's disappointingly fewer than I expected :)
[16:54] <cjwatson> cody-somerville: a work of fiction, as far as I can tell. https://wiki.ubuntu.com/UbuntuDevelopers is more accurate
[16:54] <elmo> Keybuk: wanna bet? ;-)
[16:55] <cjwatson> elmo: not all of them have the requisite LP powers. :-)
[16:55] <cody-somerville> lol
[16:55] <Keybuk> elmo: duckie abuse doesn't count ;)
[17:01] <ion> keybuk: This might be a stupid idea, but how about adding a char **locks parameter to spawn, and having it open the given /var/lock files and flock them with LOCK_EX after fork and before exec? When fsck is spawned, e.g. {"/var/lock/mountall-fsck-sda", "/var/lock/mountall-fsck-sdb", NULL} would be passed as the parameter if sda and sdb are determined to be the physical devices.
[17:02] <Keybuk> ion: you'd need to clean them up as well
[17:03] <ion> Ok, if a few zero-byte files in tmpfs is a big deal. :-)
[17:03] <Keybuk> things should be clean ;)
[17:04] <ion> Heh
[17:09] <ion> keybuk: Oh, btw, mountall requests an ecryptfs passphrase from the user. I have an ecryptfs mounted at ~/Private.
[17:10] <Keybuk> ion: ?
[17:10] <Keybuk> you mean mount does?
[17:10] <ion> Yeah
[17:10] <Keybuk> have you got --debug output, not sure why it would have run mount for you
[17:12] <ion> keybuk: http://pastebin.com/f7963f585
[17:15] <ScottK> james_w: Thanks.
[17:15] <Keybuk> ion: err, that looks like an ecryptfs bug!
[17:15] <ion> keybuk: How about creating the directory /var/lock/mountall on startup and recursively deleting it before exiting? An internal queue still a better idea?
[17:16] <Keybuk> ion: it's asking for a passphrase even when called with -f (fake. ie. add to mtab)
[17:16] <ion> Ok
[17:16] <Keybuk> ion: only if you can figure out the sequencing to make sure /var/lock is mounted before any disk is checked :p
[17:16] <ion> Ah, good point :-P
[17:16] <Keybuk> I suspect that dir-o-locks is not going to work
[17:32] <hyperair> TomaszD: could you check something -- does it matter whether you open the standalone window or the properties window first?
[17:33] <TomaszD> hyperair, doesn't matter
[17:33] <hyperair> hmm.
[17:33] <hyperair> what about if both are open?
[17:33] <TomaszD> tried that just now, no change
[17:34] <hyperair> basically the standalone window is translated but the properties window isn't eh..
[17:34] <TomaszD> no here, I can give you a screenshot if you don't believe me ;)
[17:34] <TomaszD> *not here
[17:35] <hyperair> hmm a screenshot might be nice, but i believe you =)
[17:46] <TomaszD> hyperair, http://imgur.com/1bBf2.png
[17:46] <hyperair> ooh gnome shell
[17:46] <hyperair> wait a sec
[17:46] <hyperair> the entire standalone window isn't translated either!
[17:46] <hyperair> seb128: i thought you said the standalone window was translated!
[17:47] <TomaszD> don't mind the tacky theme, I'm just messing, this is a test install ;)
[17:47] <seb128> no, I said that nothing out of the context menu item and the "do you want to install samba" was translated
[17:48] <hyperair> ah.
[17:48] <hyperair> whoops
[17:48] <hyperair> TomaszD: could you try downgrading to nautilus-share 0.7.2-9 and seeing if you see the issue?
[17:48] <hyperair> that's pre-glade-to-gtkbuilder transition
[17:49] <TomaszD> hyperair, where do I get the old deb package?
[17:49] <hyperair> TomaszD: which architecture are you using?
[17:50] <TomaszD> i386
[17:50] <dpm> hyperair: I can also reproduce this. Did the translation domain remain the same in the transition to gtkbuilder?
[17:51] <hyperair> dpm: yes it did. but i didn't touch the .po files while converting the .glade to .ui
[17:51] <hyperair> TomaszD: https://launchpad.net/ubuntu/+source/nautilus-share/0.7.2-9/+build/1089292
[17:52] <TomaszD> oh crap
[17:52] <TomaszD> gdebi-gtk has the same exact issue
[17:52] <hyperair> ?
[17:52] <TomaszD> I just noticed that now while installing nautilus-share
[17:52] <TomaszD> it's not translated
[17:52] <TomaszD> I mean, partially it is
[17:52] <hyperair> hmm
[17:52] <TomaszD> I'll go check if that's my lang's problem, but nothing in the UI changed
[17:55] <TomaszD> hyperair, I can confirm that the downgraded version has no issues with the translation whatsoever
[17:56] <hyperair> TomaszD: aha. okay, thanks for confirming. i'll work on fixing it then
[17:56] <TomaszD> yep, I checked, gdebi's translation is also broken
[17:58]  * hyperair doesn't handle gdebi
[18:02] <mdz> The following NEW packages will be installed:
[18:02] <mdz>   grub
[18:02] <mdz> that doesn't seem right
[18:02] <ScottK> Would some archive admin with shell acess (maybe james_w) please move mgltools-gle from Universe to Multiverse?  It's got a not-for-commercial-use license so it got accepted in the wrong pocket by mistake.
[18:03] <james_w> ScottK: would you file a bug for reference please?
[18:03] <ScottK> james_w: Sure.
[18:06] <james_w> ScottK: what's the not-for-commercial-use clause?
[18:06] <ScottK> james_w:   1. Grant Of Limited License; Software Use Restrictions. The programs received by you will be used only for NON COMMERCIAL purposes. This license is issued to you as an individual.
[18:07] <ScottK> james_w: Bug #425818
[18:07] <james_w> I don't see that in the copyright file
[18:07] <ScottK> It's in the LICENSE file in the source package.
[18:08] <james_w> ok
[18:08] <hyperair> TomaszD: i noticed you've got commit access to nautilus-share's git repository. do me a favour and change the line "interfaces/share-dialog.glade.in" to "[type: gettext/glade]interfaces/share-dialog.glade.in"
[18:08] <m3F> About sound in Karmic: when i try to play sound with Nautilus putting my mouse over the sound file i succeed. But when i try to pley music with Listen/Rythmbox i always fail and players brake out Pulse in the process. An automatic bug report was sent from my Ubuntu Karmic instalation. My sound card is Creative Audigy SE. I used to have problems with this card in Ubuntu Hardy and before it. But with Ubnut Jaunty all was good. Now problems are back wit
[18:08] <m3F> h Karmic (i know it is in alpha, i am just reporting the bug). Even the annoying sounds are back (annoying noises when sounds start and end).
[18:08] <james_w> moved in karmic
[18:08] <james_w> can we move in released versions?
[18:08] <ScottK> james_w: Thanks.  We can.  Let me double check that version.
[18:08] <hyperair> TomaszD: that should fix it upstream. now i need to get the patch downstream =)
[18:08] <TomaszD> hyperair, are you sure? :>
[18:09] <TomaszD> about this fix
[18:09] <hyperair> TomaszD: almost completely sure. test build it and check? =)
[18:09] <TomaszD> oh boy
[18:09] <hyperair> haha
[18:09] <TomaszD> I'm gonna include you as the author of the commit :P
[18:09] <TomaszD> one moment
[18:09] <hyperair> no problem
[18:10] <hyperair> oh yeah, you might like to refresh all the .po files as well
[18:10] <hyperair> remove the nautilus-share.pot, and run make update-po inside the po directory
[18:10] <ScottK> james_w: It applies to Jaunty too.
[18:10] <james_w> ok, thanks
[18:11] <TomaszD> hyperair, and commit all those as well or just for my local build?
[18:11] <hyperair> TomaszD: commit *.po, and POTFILES.in but not nautilus-share.pot or POTFILES
[18:11] <TomaszD> alright
[18:12] <james_w> ScottK: did you file in Debian?
[18:12] <ScottK> james_w: Not yet.
[18:12] <james_w> ok, thanks for catching this
[18:13] <TomaszD> wait a moment hyperair, where should I find this line?
[18:13] <hyperair> TomaszD: po/POTFILES.in
[18:13] <TomaszD> ah, of course..
[18:14] <hyperair> TomaszD: intltool doesn't detect *.ui files as translatable, so if you run make update-po, all the strings get commented out
[18:14] <hyperair> TomaszD: that's why the [type: gettext/glade] needs to be set
[18:15] <TomaszD> hyperair, the line you talk about reads [type: gettext/glade]interfaces/share-dialog.ui
[18:15] <hyperair> TomaszD: it does? already?
[18:15] <TomaszD> yes, no "glade" part though
[18:15] <TomaszD> in the name
[18:15] <hyperair> ah.
[18:15] <hyperair> let's see..
[18:16] <hyperair> it is gettext/glade isn't it?
[18:16] <hyperair> -interfaces/share-dialog.ui
[18:16] <hyperair> +[type: gettext/glade]interfaces/share-dialog.ui
[18:16] <hyperair> well that's done then
[18:16] <hyperair> hah
[18:17] <hyperair> ..i'm beginning to really regret patching configure and Makefile.in
[18:18] <TomaszD> hyperair, so I don't have to touch anything then?
[18:18] <hyperair> nothing =)
[18:18] <TomaszD> *phew*
[18:18] <hyperair> haha =)
[18:18] <hyperair> i should probably poke federico for a new upstream release
[18:18] <hyperair> there's been loads of translation changes
[18:19] <hyperair> it'll also allow me to drop my crazy long series of patches
[18:20] <TomaszD> good idea
[18:20] <hyperair> there's also 10 odd bugfixes i submitted when i took over this package =\
[18:23] <ScottK> james_w: I'll also upload a fixed debian/copyright for Karmic (and file a bug with Debian).
[18:43] <blue0488> how is this alpha?
[18:44] <arand> alpha-ish
[18:52] <hyperair> well well. i thought 1 hour was bad for waiting time, but 8 hours?
[18:52]  * hyperair sighs
[18:53] <hyperair> TomaszD: https://launchpad.net/~hyperair/+archive/bugfixes/+build/1207117 <-- if you're around when this finishes, please test =)
[18:53] <TomaszD> ETA: 9hrs
[18:53] <TomaszD> if I'm awake in 9hrs, I'll be a dead man
[18:54] <hyperair> ;-)
[18:54] <hyperair> another time then.
[19:04] <hyperair> TomaszD: would you trust me enough to use a deb i compiled on my machine? (i'm lazy to wait for the buildds to finish)
[19:05] <TomaszD> hyperair, I have no trust issues... not in an isolated testing virtual machine :)
[19:05] <hyperair> ahaha
[19:05] <hyperair> okay then =)
[19:06]  * hyperair starts pbuilder-karmic-i386
[19:18] <Laney> It takes about 20 seconds from "GRUB loading" to the menu appearing
[19:18] <Laney> is that normal?
[19:20] <hyperair> TomaszD: http://dl.getdropbox.com/u/169656/nautilus-share_0.7.2-11ubuntu1%7Ehyper2_i386.deb
[19:21] <hyperair> Laney: no
[19:21] <sebner> Laney: not really but I didn't install the latest update from today yet
[19:21] <Laney> not today
[19:21] <Laney> it's been like that for ages
[19:21] <Laney> since grub 2
[19:21] <sebner> Laney: takes 2-3 seconds here
[19:21] <hyperair> same
[19:22] <TomaszD> hyperair, back to square one :(
[19:24] <hyperair> =(
[19:25] <hyperair> TomaszD: could you temporarily move the nautilus-share.mo files from /usr/share/locale-langpack elsewhere?
[19:25] <Laney> hyperair: can't you test by launching with LANG set to something else?
[19:25] <hyperair> hmm
[19:25] <hyperair> good point eh
[19:26] <hyperair> let's see if that works..
[19:26] <TomaszD> I thought you didn't want to have to install new langpacks
[19:26] <TomaszD> otherwise, just use LANG, yes
[19:27] <hyperair> well.. the package i've created isn't stripped of .mo files
[19:27] <hyperair> unlike the nautilus-share package in ubuntu
[19:27] <TomaszD> ah, true
[19:28] <hyperair> now then, what was your LANG set to again?
[19:42] <cjwatson> Laney: you're not the only person who's reported it, but neither I nor anyone else has managed to track it down yet
[19:42] <cjwatson> Laney: try 'set debug=disk' at the top of /boot/grub/grub.cfg and you might be able to see which disk it's probing (or 'set debug=all' if you want everything, but that might scroll by rather fast)
[19:47] <hyperair> how does one decompile a .mo?
[19:55] <seb128> you can run msgunfmt on it
[20:05] <hyperair> hmm it seems that gtk_builder_set_translation_domain needs to be set *before* add_blah_blahh..
[20:05] <hyperair> but still it's pretty strange why it didn't work before..
[20:05] <hyperair> hmm
[20:06] <hyperair> maybe it's --as-needed?
[20:06] <hyperair> no wait, can't be either
[20:15] <hyperair> ah totem also uses the set_translation_domain trick
[20:22] <hyperair> TomaszD: http://dl.getdropbox.com/u/169656/nautilus-share_0.7.2-11ubuntu1%7Ehyper3_i386.deb
[20:22] <hyperair> this one works on my system using pl_PL.UTF-8
[20:27] <TomaszD> hyperair, almost! the "Create Share" button is left untranslated
[20:28] <hyperair> !! WHAT
[20:28] <c_korn> lol :)
[20:29] <hyperair> TomaszD: that's another bug. but thanks for pointing it out =O
[20:30] <hyperair> it's supposed to be Create _Share, but i patched it into Create Share
[20:30] <hyperair> whoops..
[20:42] <hyperair> TomaszD: http://dl.getdropbox.com/u/169656/nautilus-share_0.7.2-11ubuntu1%7Ehyper4_i386.deb
[20:42] <TomaszD> take 4...
[20:44] <hyperair> hehe
[20:44] <TomaszD> aaand cut
[20:45] <TomaszD> good job everyone, great acting hyperair, 100% correct
[20:45] <hyperair> woo
[20:45] <TomaszD> and now I finally see where do I have to fix those accelerator conflicts
[20:48] <hyperair> =O
[20:48]  * hyperair falls asleep on keyboard
[21:00] <ulaas> exit
[21:00] <ulaas> quit
[22:18] <ulaas__> help
[22:19] <ulaas__> learning irssi :) sorry
[23:52] <RainCT> seb128: Hey, can you please also ack bug 303628?
[23:53] <seb128> RainCT, seems fine to me feel free to upload
[23:54] <RainCT> seb128: Thanks :)