[00:48] <ogra_cmpc> evand, eventually, yes ... i cant tell if that means hardy though, thats a RichEed decision, afaik he wants something to demo an edu desktop
[00:49] <evand> ogra_cmpc: ok, thanks
=== ogra_cmpc_ is now known as ogra_cmpc
[01:04] <ogra_cmpc> grmbl
[01:04]  * ogra_cmpc just learned that the lenovo page kills classmates 
[01:06] <LaserJock> ogra_cmpc: yucky
[01:08] <ogra_cmpc> and i didnt even get to see any pics of the new lappie ...
[01:48] <bobwhoops> join #gsoc
[01:49] <bobwhoops> whoops
=== fabbione is now known as thegodfather
[02:44] <superm1> slangasek, did you ever manage to get those changes merged into cdimage?
[03:22] <slangasek> superm1: now I have... :)
[03:23] <superm1> slangasek, yay :)
[03:23] <superm1> slangasek, so with space being a commodity, would it be feasible to only ever keep one image of ours at a given time?
[03:24] <superm1> slangasek, or will we have to wait until cdimage gains more hard drives?
[03:25] <slangasek> superm1: currently the etc/purge-days script has a single rule for all daily alternate images; I've made a policy change already that gives us a little more space though, so mythbuntu should fit in ok
[03:25] <superm1> slangasek, wonderful :)
[03:26] <superm1> slangasek, we're not going to explore doing the livecd via livecd-rootfs for some time, to that will work great
[03:34] <TheMuso> superm1: You guys are getting images done via cdimage for mythbuntu? Awesome!
[03:34] <superm1> TheMuso, at least for an alternate disk (which we have needed for ages) :)
[03:35] <TheMuso> superm1: Again, awesome!
[04:33] <Hobbsee> *drool*
[04:33] <Hobbsee> iwlwifi fixes!
[04:33] <TheMuso> Hobbsee: I can only guess that they benefit you.
[04:33] <RAOF> Hobbsee: As in: "please don't fill up /var/log with gigabytes of debug spam"!
[04:34] <RAOF> Things get really wierd when there's 0 bytes free on /, /tmp and /var :)
[04:34] <ScottK2> Hmmm. Spam.
=== _Apollo_2 is now known as _Apollo_
[04:35] <Hobbsee> RAOF: yeah, that too.  the led, in particular
[04:36] <RAOF> Hobbsee: My LED has been working?
[04:36] <Hobbsee> ipw or iwl?
[04:38] <RAOF> iwl
[04:38] <RAOF> ipw doesn't work anymore, right?  I don't have the non-free daemon anymore, certainly.
[04:38] <RAOF> Or whatever it was keeping ipw in -restricted.
[04:39] <Hobbsee> strange.
[04:39] <RAOF> I suppose it depends on what you mean by "working".  My LED is _on_.  It doesn't seem to turn off when I hit the killswitch, though :)
[04:40] <RAOF> I've just never noticed this, because I treat the killswitch as basically a way to accidentally break my networking.
[04:40] <StevenK> "accidentally" ? :-)
[04:40] <RAOF> It's in a place where I *can* accidentally brush against it.  Or removing the laptop from a bag, or...
[04:40] <Hobbsee> RAOF: ah right.  mine's always off, killswitch or not
[04:41] <RAOF> Hobbsee: Yay for crazy non-standard laptops.  Standards are for _pussies_!
[04:41] <Hobbsee> hehe
[04:49] <pwnguin> its a bit sad though
[04:49] <pwnguin> iwl is a much bigger powertop offender than ipw
[04:50] <RAOF> pwnguin: I haven't noticed that, but then the nvidia driver is a big powertop offender, too.
[04:51] <pwnguin> i turned off the nvidia driver
[04:51] <pwnguin> but check out the top offender
[04:52] <pwnguin> 59.1% (148.6)      <kernel IPI> : Rescheduling interrupts
[04:52] <RAOF> I like compiz, when I'm not using nouveau.
[04:52] <pwnguin> is nouveau usable for 2d?
[04:52] <pwnguin> i just use nv
[04:53] <RAOF> pwnguin: Nouveau is *extremely* usable for 2d.
[04:53] <RAOF> It's basically as fast or faster than the blob for 2d, at least on my chip.
[04:53] <pwnguin> neat
[04:53] <RAOF> (Also, has better dual head support).
[04:53] <pwnguin> double neat
[04:54] <pwnguin> the twenty thousand dollar question: xrandr rotations
[04:54] <RAOF> Accelerated.
[04:54] <pwnguin> 180?
[04:54] <RAOF> and 90, and 270.
[04:54] <RAOF> As far as I can tell.
[04:54] <pwnguin> have you tried 180?
[04:54] <RAOF> No, I don't think so.
[04:54] <RAOF> Oh, actually, yes I have.
[04:55] <RAOF> Probably.  I played around with "mirror x", "mirror y", "mirror xy" at one point.
[04:55] <RAOF> Somewhere in that playing I probably managed a 180 rotation :)
[04:56] <pwnguin> not quite the same
[04:56] <RAOF> Try it yourself, then :)
[04:56] <pwnguin> im considering it
[04:57] <RAOF> I can even play OpenArena on nouveau :)
[04:57] <pwnguin> heh
[04:57] <pwnguin> i'd be more interested in the crack attacks and gunroars
[04:57]  * StevenK remembers submitting debugging information about his card to the nouveau guys on RAOF's insistance
[04:58] <RAOF> :)
[04:58] <pwnguin> RAOF: does your ppa still have nouveau up for testing?
[04:58] <RAOF> Yup.
[04:58] <RAOF> I should work out the module-assistant magic required to make this useful for Debian and push it into experimental.
[05:01] <RAOF> StevenK: And in return, the randr12 code probably properly handles your chip :)
[05:02] <StevenK> RAOF: Heh. Come back to me when Nouveau is in Ubuntu and handles WoW, and I'll switch :-)
[05:02] <RAOF> StevenK: I've tried, it doesn't quite yet :P
[05:04] <StevenK> RAOF: It doesn't run, or it looks wierd?
[05:04] <RAOF> It won't be in Ubuntu until DRI2 is stable, probably.
[05:04] <RAOF> StevenK: Doesn't run.  There's still a bunch of unimplemented stuff in gallium, would you believe it? :)
[05:04] <StevenK> Heh
[05:06] <RAOF> Although I also had to fight with my additional 32bits, so it's possible (but unlikely) that an i386 install would have better luck.
[05:07] <pwnguin> wow is overrated
[05:07] <TheMuso> pwnguin: Watch out!
[05:07] <TheMuso> :p
[05:09]  * RAOF hopes DRI2 will make Ibex.  DRI2 not only brings back kittens killed by god, it also procs world peace (5% chance).
[05:12] <pwnguin> so did ati actually release 3d docs?
[05:13] <toresbe> Ibex?
[05:14] <toresbe> Is that the one after Heron?
[05:14] <TheMuso> toresbe: Yes.
[05:14] <toresbe> neat
[05:14] <toresbe> I think there's a painkiller in Norway named Ibex
[05:15] <toresbe> no, that's Ibux, probably.
[05:15] <toresbe> http://www.ibextours.com/ though.
[05:16] <StevenK> toresbe: https://lists.ubuntu.com/archives/ubuntu-devel-announce/2008-February/000383.html
=== lando_ is now known as lando
[07:03] <dholbach> good morning
[07:04] <asac> morning dholbach
[07:05] <dholbach> hey asac
[07:08] <warp10> Good morning!
[07:09] <pitti> Good morning
[07:22] <thekorn> hi dholbach, where can a file bugs/suggestions related to this 5-a-day tool?
[07:23] <thekorn> https://edge.launchpad.net/five-a-day/+filebug seems to be disabled
[07:25] <dholbach> thekorn: should work now - thanks for letting me know
[07:25] <thekorn> dholbach, yes, works, thanks
[07:35] <dholbach> asac: http://daniel.holba.ch/ubuntu/ic/ does not load up in hardy, but does in gutsy
[07:37] <pitti> lool: WDYT about the updated packages in bug 186647 ?
[07:37] <ubotu> Launchpad bug 186647 in pigment "promote to main" [Undecided,New] https://launchpad.net/bugs/186647
[07:38] <asac> dholbach: didn't i look at that once?
[07:38] <asac> dholbach: i think it was illegal html ;)
[07:39] <dholbach> "illegal html"
[07:40] <dholbach> EXPECT THE POLICE! :)
[07:40] <asac> hehe
[07:44] <asac> dholbach: close the script elements properly
[07:44] <asac> or use XHTML DTD :)
[07:46] <asac> e.g <script type='text/javascript'></script>
[07:48] <dholbach> asac: OK, seems to be the script tag :)
[07:50] <Mithrandir> asac: it doesn't handle <script type='text/javascript' /> correct either, even with doctype = XHTML
[07:50] <asac> Mithrandir: is |'| legal in xml?
[07:51] <Mithrandir> asac: well, it was "", iirc.  I don't have that bit of code in front of me here.
[07:51] <asac> hmm
[07:51] <Mithrandir> asac: I ran it through a validator and it didn't complain, at least.
[07:52] <asac> Mithrandir: scary. you didn't file a bug, did you?
[07:52] <Mithrandir> asac: no, but I could.
[07:52] <pitti> slangasek: WDYT about bug 193265 ?
[07:52] <ubotu> Launchpad bug 193265 in sane-backends "FF exception request: update sane-backends to 1.0.19 final" [Undecided,New] https://launchpad.net/bugs/193265
[07:53] <asac> Mithrandir: please do you if you can reproduce with latest ffox 3 ... and attach the testcase
[07:53] <Mithrandir> asac: 'k
[07:54] <asac> Mithrandir: we are still pretty good at forwarding bugs upstream and getting things fixed
[07:54] <asac>  ... though time runs low
[07:54] <asac> ;)
[07:54] <slangasek> pitti: wanted to finish reading through the upstream changelog, which I don't think I'll get through tonight
[07:55] <asac> Mithrandir: package "xulrunner-1.9" most likely
[07:56] <pitti> slangasek: ok, no problem; just wanted to check if it's on your list, thanks
[07:57] <slangasek> pitti: yep, definitely; I should be able to catch up on the list of FFe requests tomorrow (Tuesday)... and sooner if other members of ubuntu-release were to help ;)
[07:57] <pitti> slangasek: I obviously can't approve my own request, but if you need help I can look at others
[07:58] <slangasek> I'd appreciate it if you had the time
[07:59] <pitti> ok, noted
[08:01]  * Hobbsee headbashes quietly
[08:01]  * Mithrandir puts a pillow between Hobbsee's head and the wall
[08:03] <Hobbsee> hey there Mithrandir!
[08:30] <cjwatson> toresbe: FWIW, if we're going to use a single word to describe the release, we normally use the adjective, not the noun
[08:31] <cjwatson> RAOF: ^-- err, I guess that was really for you
=== rraphink is now known as raphink
[08:38] <pwnguin> what was the i? idyllic?
[08:39] <cjwatson> pwnguin: please see ubuntu-devel-announce
[08:39] <pwnguin> intrepid
=== Amaranth_ is now known as Amaranth
[09:11] <lool> pitti: I've poked philn to please commit his elisa/pigment packaging changes for the new upstream release/svn snapshot he mentions; if it takes too long, I'll do them; if you like, I can do a quick upload with only the python 2.5 packaging fixes, but it wont solve the issue you were seeing
[09:23] <ArneGoetje> Riddell: If you use "System Default" as antialiasing setting in KDE, it will probably fall back to fontconfig. However, hinting and antialiasing is not enabled by default in fontconfig.
[09:25] <ArneGoetje> Riddell: the Gnome desktop uses Greyscale (rgba none) and medium hinting as default settings. Question is if we should enable this setting in fontconfig globally... then KDE should get the same settings...
=== hunger_t_ is now known as hunger
[09:27] <pitti> lool: right; when we do an Ubutu upload, it shuold make the package actually working
[09:28] <pitti> lool: we can wait a few days, but for the MIR it would be good to get it working ASAP
=== Amaranth_ is now known as Amaranth
[09:58] <pitti> Mithrandir: since you seem to work on bluetooth, do you have an idea about bug 72033? what would need to happen to make this work OOTB?
[09:58] <ubotu> Launchpad bug 72033 in ubuntu-meta "Install gnome-bluetooth by default to enable receiving files via bluetooth" [Undecided,Confirmed] https://launchpad.net/bugs/72033
[10:01] <lool> pitti: Not sure we want to have an obex server listening on bluetooth 100% of the time
[10:01] <lool> (No open port policy?)
[10:01] <pitti> ah, right
[10:02] <pitti> so the bug is merely about making it much easier, not making it totally automatic?
[10:02] <lool> I guess so
[10:02] <mjj29> installed, but not enabled by default is probably reasonable
[10:02] <mjj29> so enableable from the gnome menus
[10:02] <lool> pitti: The bug seems to be about having it installed by default
[10:03] <lool> Someone proposes installing it when a bluetooth device is present/inserted
[10:03] <lool> I'd say we should install on mobile devices like laptops; I'm not sure it's easy to see the bluetooth adapter if it's off
[10:03] <lool> (ISTR it's an USB device appearing on the USB bus on my thinkpad)
[10:05] <lool> pitti: Also, I don't think there's an easy way to launch the obex server from the bluetooth adapter icon when it's visible
[10:09] <geser> pitti: Hi, please give-back: libwnck. Thanks.
[10:11] <pitti> hi geser; kicked
[10:18] <Riddell> ArneGoet1e: why would we ever not want antialiasing by default?
[10:29] <Keybuk> pitti: are ddebs not being generated automatically any more?
[10:29] <seb128> Keybuk: they are, what package is that about?
[10:29] <pitti> Keybuk: they are supposed to
[10:29] <Keybuk> pitti: there's nothing in gutsy-security
[10:29] <seb128> ah, pockets is an another story
[10:30] <pitti> Keybuk: right, it seems that the sbuild hack is not applied to the security buildds
[10:30] <seb128> I think it's a buildd thing
[10:36] <pitti> seb128: for bug 186569, TBH I'd prefer to fix nautilus for conforming to the SuSv3 spec rather than changing ntfs3g; do you object?
[10:36] <ubotu> Launchpad bug 186569 in ntfs-3g "cannot delete files off of an Fuse mounted NTFS partition in nautilus" [High,In progress] https://launchpad.net/bugs/186569
[10:36] <pitti> seb128: I'll do it and forward it to upstream, just asking for your opinion
[10:36] <seb128> pitti: let me ask alex, we discuss bugs on IRC usually
[10:36]  * pitti hugs seb128, merci
[10:37] <seb128> pitti: but are you sure it's not a ntfs3g issue?
[10:37] <seb128> pitti: man rmdir on hardy doesn't list EEXIST as a valid codecase
[10:37] <pitti> seb128: well, you can argue that ntfs3g should throw the errno as described in the Linux manpage of rmdir
[10:37] <pitti> seb128: but since the unix spec says otherwise, I see the point of ntfs3g doing it otherwise
=== ArneGoet1e is now known as ArneGoetje
[10:38] <seb128> why should we trust the unix specs rather than the ubuntu rmdir one?
[10:39] <pitti> "NTFS-3G uses EEXIST because more software handle the relevant error scenario better that way. Seemingly that doesn't include Nautilus." -> I can't really validate this, but it should be considered at least
[10:40] <pitti> seb128: well, rmdir() on ext3 just returns ENOTEMPTY exclusively, no doubt
[10:40] <pitti> but it doesn't actively say "rmdir must not return anything else"?
[10:40] <ArneGoetje> Riddell: in that case, the fontconfig package needs to be patched.
[10:41] <Riddell> ArneGoetje: doesn't sound too hard
[10:41] <seb128> pitti: well, if you can't trust the manpage ...
[10:41] <pitti> seb128: I don't have a strong opinion on doing either, but in the principle of being tolerant with inputs and strict with outputs I tend to fix nautilus
[10:42] <seb128> well, I'll discuss it with alex
[10:42] <seb128> but I'm not fan to use workarounds for broken code
[10:42] <pitti> seb128: I'll file a bug against the rmdir manpage in Debian
[10:45] <ArneGoetje> Riddell: yep, will do.
=== thom_ is now known as thom
[10:47] <pitti> seb128: done
[10:47] <seb128> pitti: thanks
[10:49] <Hobbsee> pitti: i didn't know MIR's could get done without going thru universe.  that's kinda neat.
[10:49] <pitti> Hobbsee: how do you mean? they should usually go through universe first
[10:50] <Hobbsee> pitti: the hwdb stuff
[10:50] <pitti> oh, right, that should get ~motu-release subscribed
[10:51] <Hobbsee> pitti: no point now - youv'e already given the ack...
[10:51] <pitti> we should consider it, though, hwdb-client is horrible
[10:51] <pitti> Hobbsee: no, I didn't?
[10:51] <ogra_cmpc> pitti, i was told hwdb2 is ready for hardy
[10:51] <Hobbsee> well, put it on the MIR list is the same as "you have an exception for this, if it's OK", isn't it?
[10:51] <Hobbsee> it's a good way for getting important stuff thru REVU
[10:52] <Hobbsee> er, with avoiding REVU
[10:53] <pitti> Hobbsee: no, not at all; I didn't approve anything yet, I asked to get it into universe first
[10:53] <pitti> ogra_cmpc: hwdb2?
[10:53] <Hobbsee> pitti: i thought your comment after reverted that, saying it was in a ppa, so that was OK
[10:53] <cjwatson> seb128: if SUSv3 documents things that aren't mentioned in the Ubuntu rmdir manual page, you should trust SUSv3 rather than the manual page because many more people were involved in reviewing SUSv3 and it applies to more operating systems. If the manual page explicitly mentioned a deviation from SUSv3 then that might be a different matter, but it doesn't
[10:54] <ogra_cmpc> pitti, the thing liw and cr3 work on
[10:54] <ogra_cmpc> (they didnt change the gui much though)
[10:54] <cjwatson> seb128: besides, the Linux manpages package only really documents what the Linux kernel does
[10:54] <Hobbsee> well, for MIR review
[10:54] <seb128> cjwatson: ok, I don't know what SUSv3 is and didn't know that was a reference on the topic
[10:54] <pitti> ogra_cmpc: right, isn't that what hwtest is?
[10:54] <ogra_cmpc> thats the replacement
[10:54] <cjwatson> seb128: SUSv3 == POSIX
[10:54] <cjwatson> http://www.opengroup.org/onlinepubs/009695399/nfindex.html
[10:55] <ogra_cmpc> pitti, once you see its gui you understand :)
[10:55] <ogra_cmpc> it looks pretty miuch identical
[10:55] <pitti> cjwatson: my thought as well, that's why I filed a bug against manpages-dev (not in the BTS yet)
[10:55] <ogra_cmpc> just the insane backend is gone
[10:55] <cjwatson> seb128: definitely the applicable standard :)
[10:55] <seb128> cjwatson: ok, good to know, thanks ;-)
[10:58] <seb128> speaking about fuse, could we add desktop users to the fuse group on hardy?
[10:58] <seb128> what would be the right place to discuss that? irc? mailing list? launchpad?
[10:59]  * ogra_cmpc wouldnt mind
[10:59] <ogra_cmpc> would make things easier in ltsp land as well ...
[10:59] <seb128> gvfs does fuse mounts transparently when accessing a network location which means applications not using vfs can access network shares transparently
[10:59] <ogra_cmpc> seb128, do you plan gvfs-fuse in the desktop as well ? i heard there are some massive probs with sabayon
[11:00] <seb128> ogra_cmpc: no idea about sabayon, we got no bug on launchpad and it's not in the default installation
[11:00] <ogra_cmpc> seb128, i was planning to pull it into e-dsktop
[11:01] <seb128> ogra_cmpc: I don't think it's a desktop user application
[11:01] <ogra_cmpc> (thats why we had so many sessions about it in boston :))
[11:01] <seb128> and yes having gvfs-fuse in the desktop installation would be nice
[11:01] <ogra_cmpc> seb128, well, then edubuntu-server in any case it will end up on the edubuntu cd
[11:01] <pitti> seb128, cjwatson: FYI, Debian bug 467552
[11:01] <ubotu> Debian bug 467552 in manpages-dev "rmdir(2) can return EEXISTS, too" [Normal,Open] http://bugs.debian.org/467552
[11:02] <ogra_cmpc> seb128, ok, thats all i wanted to know :)
[11:04] <cjwatson> seb128,ogra_cmpc: we discussed this 20 days ago in this channel and the agreed output was not to add all users by default, and for Oliver to add a notification to fuse-utils. However, I'll make the user-setup change now to add the default user to the group
=== pedro__ is now known as pedro_
[11:05] <cjwatson> seb128: see http://irclogs.ubuntu.com/2008/02/06/%23ubuntu-devel.txt for the log
[11:06] <seb128> cjwatson: right, I know we discussed it but I didn't open a bug or anything and no change was made that's why I was asking again
[11:06] <seb128> cjwatson: thanks ;-)
[11:07] <ogra_cmpc> i'll do the fuse change today ...
[11:10] <cjwatson> user-setup> will do it once we've fixed a bit of bzr desync
[11:20] <seb128> pitti: http://bugzilla.gnome.org/show_bug.cgi?id=518816
[11:20] <ubotu> Gnome bug 518816 in general "should handle rmdir returning EEXIST correctly" [Normal,Unconfirmed]
[11:20] <pitti> seb128: cool, thanks; did alex agree?
[11:21] <hackeron_> hey, it seems hardy is ignoring my xorg.conf -- I have Modes           "800x480" - yet hardy starts X with "800x600" -- Is there anyway to force hardy to use my required resolution?
[12:05] <pitti> Riddell: does anything in Kubuntu still use cupsys' enable_{browsing,sharing}?
[12:06] <Ng> is sun-java6-plugin supposed to be empty apart from directories?
[12:12] <Riddell> pitti: no
=== Yvonne_ is now known as Yvonne
[12:19] <seb128> pitti: the gvfs fix has been commited to svn and uploaded to hardy now
[12:21] <cjwatson> mjg59: are you planning to push that vt font restoration patch to the hardy kernel?
[12:21] <pitti> seb128: wow, that was fast
[12:33] <mjg59> cjwatson: Can do
[12:34] <cjwatson> thanks
=== ember_ is now known as ember
=== lionel_ is now known as lionel
[13:10] <seb128> carlos, pitti: any idea of why seahorse.mo is not in the hardy language packs?
[13:12] <pitti> hm, according to the build logs they were stripped
[13:13] <pitti> seb128: from what I can see it's probably stuck in the LP translations moderation queue?
[13:13] <seb128> pitti: no idea, that's why I was asking
[13:13] <seb128> is carlos on holidays this week?
[13:15] <seb128> right, he's not working today, ok will ask him tomorrow
[13:15] <pitti> sorry, I tried to find it out using the webui, but it's not really searchable enough
[13:16] <seb128> that's alright
[13:16] <ScottK> pitti: I was looking into recent Debian changes in cyrus-sasl and it looks like there are some fixes worth a merge to get.  I'm a little uncertain how to proceed since when Debian incorporated your lidbd4.6 migration patch they did it slightly differently.  Would you please take a look at it and advise me if their patch is appropriate?
[13:17] <pitti> ScottK: looking
[13:17] <ScottK> Thanks
[13:18] <pitti> ScottK: meh, our Ubuntun diff is horrible, 2.6 MB
[13:18] <pitti> and only because someone autoreconf'ed with a wildly different version
[13:19] <pitti> ScottK: can you please clean this up during the merge?
[13:19] <ScottK> pitti: I can try.
[13:19] <ScottK> pitti: I'm not certain this is one I'll be sufficiently confident to do on my own, but I'm going to give it a shot and see how it goes.
[13:22] <pitti> ScottK: confirmed that the Debian patch for db migration is fine
[13:23] <ScottK> pitti: Thanks.  I'll add that back to my TODO then.
[13:57] <cjwatson> tjaalton: ooh. I figured out the console-setup 'any' keysym bug
[13:57] <cjwatson> -               && ($symbol !~ /^nosymbol$/i
[13:57] <cjwatson> +               && ($symbol !~ /^(?:nosymbol|any)$/i
[13:57] <cjwatson> all that over a one-liner ;-)
[14:20] <\sh> slangasek: thx again for your help yesterday night :) I found the error actually...and it has nothing to do with ubuntu or anything from the system
=== rproenca_ is now known as cypherbios
[14:22] <ogra> seb128, hmm, with gvfs my ltspfs patches seem not to work anymore, i suddenly have unmount options for ltspfs devices in nautilus
[14:23] <seb128> ogra: sure, gvfs has no ltsp patch yet
[14:23] <ogra> right, seems i need to look into that
[14:23] <seb128> you are welcome to do the patch porting
[14:24] <ogra> i have to look at gvfs first :)
[14:24] <seb128> I will have a look otherwise but that was not a priority on my list before getting gvfs uploaded
[14:24] <ogra> right, understood
[14:24] <ogra> i just noticed it after upgrading my ltsp server
=== dholbach_ is now known as dholbach
[14:46] <tjaalton> cjwatson: yay :)
[14:51] <cjwatson> Riddell: please push your ubiquity 1.7.12 release to bzr
[14:51] <Mithrandir> pitti: 72033 is fixed with the new bluez-gnome
[14:51] <pitti> ah, neat
[14:51] <cjwatson> Riddell: also if you could try to avoid the tab characters in debian/changelog that make vim display them in red as errors, that'd be good ;)
[14:59] <Riddell> hmm, I'm sure I did
[15:00] <Riddell> right, I missed the final commit
[15:01] <cjwatson> 'bzr bind' is your friend :)
[15:02] <ogra> http://paste.ubuntu-nl.org/57470/ any comments from native english speakers ?
[15:03] <ogra> (thast supposed to become teh fuse-utils notification)
[15:04] <jdong> ogra: awfully wordy for a simple concept IMO
[15:04] <ogra> jdong, i'm open for everything, suggest a better text
[15:04] <ogra> it shouldnt scare my mom on upgrades please :)
[15:05] <jdong> ogra: "Please add users to the 'fuse' group to grant permission to use fuse filesystems" would be a good paraphrase
[15:06] <jdong> probably not optimal either, but I don't see a reason why it can't be said in a sentence
[15:06] <jdong> if you choose to go with the original, s/base/basis/
[15:06] <ogra> well, i thought some slightly more explaining text of teh why/how might be good
[15:09] <jdong> ogra: "For security reasons, only members of the 'fuse' group are permitted to use fuse filesystems. Please use the Users and Groups administration tool to grant users this priviledge"
[15:10] <ogra> now *that* sounds good :)
[15:10] <cjwatson> (it's spelled "privilege")
[15:10] <jdong> oops thanks cjwatson :)
[15:10] <cjwatson> can we also ban the text "For security reasons"?
[15:10] <cjwatson> we aren't an airport :)
[15:10] <ogra> lol
[15:10] <jdong> lol
[15:11] <ogra> dont leave your files unattnded :)
[15:11] <ogra> *unattended
[15:11] <jdong> Please keep your tarballs in 3kb or smaller containers in a quart sized ziploc bag
[15:11] <ogra> ok, i'll go with the above then, minus the security warning and typo
[15:12] <ogra> what about the title ?
[15:12] <ogra> Name: Filesystem in userspace access info
[15:12] <cjwatson> "Only members of the 'fuse' group are permitted to use userspace file systems, since this is a powerful facility that has not been completely audited for safety."
[15:12] <cjwatson> something like that?
[15:13] <cjwatson> I think the installer goes for "file system" rather than "filesystem" consistently
[15:13] <ogra> uh, that implies its insecure
[15:13] <cjwatson> so does "for security reasons"
[15:13] <ogra> yeah
[15:13] <pitti> BenC: are you fine with bug 188806?
[15:13] <ubotu> Launchpad bug 188806 in git-core "Please sync git-core 1:1.5.4.3-1 (main) from Debian unstable (main)" [Undecided,New] https://launchpad.net/bugs/188806
[15:13] <cjwatson> the discussion the other day was that we weren't confident enough in it to grant access to it to all users
[15:13] <cjwatson> which is pretty much what the above is saying
[15:13] <BenC> pitti: yeah
[15:13] <pitti> BenC: i. e. do you know whether the new version introduces any behaviour changes, and have you used it?
[15:13] <pitti> BenC: ok, thanks
[15:14] <cjwatson> ogra: title> "Using userspace file systems"
[15:14] <ogra> thanks ...
[15:14] <cjwatson> although "userspace" is still jargon; I'm not sure how far we can go given that the whole thing is fundamentally jargon
[15:14] <ogra> i'd really rather formulate it positive ... like "to gain you conntrol over access of ..." or so
[15:14] <ogra> *control
[15:15] <cjwatson> actually, the obvious flaw in all of this is that if you don't know what fuse is then you have no idea what it grants you access to
[15:15] <ogra> instead of implying its not safe
[15:15] <ogra> right
[15:15] <jdong> cjwatson: well presumably someone who's just installed fuse-utils should know what it is?
[15:15] <cjwatson> so it needs some examples, like NTFS, sshfs, etc.
[15:15] <jdong> ah, examples would be nice :)
[15:15] <cjwatson> jdong: not given that it's installed by default, no
[15:15] <jdong> oh I totally forgot about ntfs-3g by default
[15:15] <ogra> jdong, you will get the note on upgrades
[15:16] <cjwatson> ogra: perhaps drop the bit about auditing but still include "powerful" in the text, which gives users a hint why it's not available across the board
[15:16] <jdong> ogra: should we instead make the fuse frontends display this message?
[15:17] <ogra> jdong, every single one ? that would get annoying if you have multiple ones installed
[15:18] <ogra> i.e. i have ltspfs here, sshfs, through the default install i get ntfs3g ....
[15:18] <cjwatson> "FUSE is a powerful facility that allows users to create virtual file systems. It can be used for applications ranging from digital cameras through online photo-sharing to compatibility with Microsoft Windows file systems."
[15:18] <pitti> ScottK, mathiaz: could one of you please have a look at bug 79371 while you touch cyrus-sasl? seems to be an easy fix
[15:18] <cjwatson> something like that?
[15:18] <ubotu> Launchpad bug 79371 in cyrus-sasl2 "saslauthd init script does not allow movement of PID" [Low,In progress] https://launchpad.net/bugs/79371
[15:20] <ogra> "By default access to it is only granted to the first user of the system to gain the system administrator full control over access. Please add users to teh fuse group to your liking"
[15:20] <ogra> the repititive "access" isnt good though
[15:20] <cjwatson> it also needs more punctuation
[15:20] <ogra> yeah, i suck at that
[15:21] <ogra> but something in that spirit combined with your explanation
[15:21] <cjwatson> "By default, only the first user is allowed to work with virtual file systems. To grant access to other users, add them to the 'fuse' group."
[15:21] <ogra> oki
[15:21] <cjwatson> or something along those lines
[15:22] <mjg59> I think "virtual file systems" is potentially confusing
[15:22] <cjwatson> aye
[15:22] <mjg59> As well as being inaccurate :)
[15:22] <cjwatson> as it happens, I was quoting 'apt-cache show fuse-utils'
[15:23] <mjg59> Heh
[15:26] <ogra> mjg59, userspace filesystems instead ?
[15:27] <mjg59> Yeah, though I'm not sure that's terribly meaningful for most people
[15:27] <mjg59> I can't think of anything better off-hand, though
[15:29] <ogra> thats my final text now: http://paste.ubuntu-nl.org/57475/+
[15:29] <ogra> err
[15:29] <ogra> http://paste.ubuntu-nl.org/57475/
[15:30] <cjwatson> "fuse-based"
[15:30] <ogra> ah
[15:30] <cjwatson> and I think the second and third paragraphs should be merged into one paragraph
[15:31] <ogra> ok, done http://paste.ubuntu-nl.org/57476/
[15:33] <cjwatson> good enough for me
[15:35] <CyberSnooP> "create userspace file systems" or "use userspace file systems"?
[15:40] <cjwatson> CyberSnooP: "mount", really, except that's more jargon. You can use a fuse filesystem mounted by another user without needing to be in the fuse group yourself.
[15:41] <CyberSnooP> yeah, the whole thing is leaning toward jargon anyway so I guess there is no easy way to get it right.
[15:46] <Keybuk> that's ok compiz ...
[15:46] <Keybuk> I didn't need window management anymore
[15:48] <geser> Keybuk: isn't a terminal on each side of the cube enough?
[16:15] <keescook> Keybuk: can you take a look at the most recent patch to 45842?  We were unable to reproduce the "does not mount" bug, but this fixes documentation, possible races, and the mtab bug for network fs's
=== \sh is now known as \sh_away
[16:22] <Keybuk> keescook: err, so it's not an ifup script anymore?
[16:22] <Keybuk> oh, sorry, I didn't see the + at the end
[16:23] <keescook> *ctrl-u* yawp
[16:23] <Keybuk> I'm not sure the extra call is needed, but otherwise I see where you're going
[16:23] <keescook> Keybuk: also, if it uses watershed, should we explicitly add udev as a Depend
[16:23] <Keybuk> yes
[16:23] <keescook> okay, we'll get that fixed
[16:24] <keescook> the extra is kind of a "it doesn't hurt and may fix future problems or issues that are hard to reproduce"
[16:25] <Keybuk> I have a natural distaste for "just in case" coding ;)
[16:25] <Keybuk> but that's just a personal issue <g>
[16:26] <keescook> Keybuk: yeah, I figured that'd be the oddest part for you.  I'm taking the angle of "for Hardy, this seems more stable".  I'd be happy to rip it out for intrepid
[16:27] <Keybuk> sure :)
[16:27] <Keybuk> imagine my displeasure at discovering that the reason we call sync before reboot is "just in case"
[16:27] <keescook> okay, so I'll take this as a +1 from you then, and get it fixed and upload.  :)
[16:27] <keescook> sync> hehe
[16:27] <Keybuk> I like to think that the kernel is not silly enough to discard writes when it feels like it
[16:28] <Keybuk> but I'm odd like that
[16:28] <keescook> I'd tend to agree.  :)
[16:30] <Keybuk> had an interesting experience the other day along those lines actually
[16:30] <Keybuk> had copied lots of stuff onto USB storage device
[16:30] <Keybuk> and shutdown the machine
[16:30] <Keybuk> and as it was going down thought "hey, I wonder whether it's actually going to finish writing and unmount that"
[16:30] <Keybuk> happily it stopped at "Will now halt..." sufficiently long that usplash timed out
[16:31] <Keybuk> (it amuses me that halt() appears to not actually stop processes)
[16:31] <Keybuk> and after a short while, powered off
[16:31] <Keybuk> so it must have been writing out to the disk still
[16:31] <keescook> so it correctly sync'd then?
[16:31] <mjg59> Yes, filesystems are synced before the machine is halted
[16:32] <keescook> kernel ftw
[16:37] <Keybuk> mjg59: it seems they mostly are
[16:37] <Keybuk> we think we found a couple of cases where they may not have been
[16:37] <Keybuk> but didn't follow them up yet
[16:37] <mjg59> device_shutdown() isn't called until after that, to the best of my knowledge
[16:37] <Keybuk> I'm pretty sure one of them involved the ide subsystem, so was irrelevant
[16:38] <mjg59> That was cache flushing, no?
[16:38] <Keybuk> err, can't remember; you could possibly be right
[16:39] <mjg59> The filesystem stuff is at the VFS and block device level
[16:39] <mjg59> Hardware subsystems then need to make sure that writes have actually hit the disk
[16:39]  * mjg59 goes out
[16:39] <Keybuk> ah, you're absolutely right
[16:42] <ogra> cjwatson, hmm, did you ever notice the " #!/bin/bash -e" at the top of the fuse-utils.postinst ?
[16:43]  * ogra doesnt see any bashisms in there that would justify it 
[16:47] <cjwatson> ogra: I don't see any bashisms either. I imagine it was there before I touched it and I never bothered changing it
[16:48] <ogra> i'll try with plain /bin/sh
[16:48] <ogra> if it works i'll change it
[16:49] <cjwatson> Riddell: ubiquity 1.7.12 final commit is still missing from bzr
[16:50] <cjwatson> Riddell: please push it so that we don't have to merge
[16:52] <Riddell> cjwatson: sorry, done that now
[16:52] <cjwatson> Riddell: thanks. FWIW I find 'debcommit --release' to be a useful habit (and it tags too, if your bzr branch format is new enough)
=== thekorn_ is now known as thekorn
[17:27] <netzmeister> pitti:  ping
[17:28] <netzmeister> crimsun_:  did you know how can i contact Colin Watson?
[17:29] <netzmeister> s/did/do
[17:29] <cjwatson> netzmeister: hello
[17:32] <netzmeister> cjwatson:  hello..
[17:32] <netzmeister> cjwatson:  i need some info about that bug...
[17:32] <netzmeister> https://bugs.launchpad.net/ubuntu/+source/grub/+bug/22220/comments/26
[17:32] <ubotu> Launchpad bug 22220 in grub "Correct modules for I2O-based raid are not loaded" [High,Fix released]
[17:32] <netzmeister> ? nice feature here.. :-)
[17:33] <cjwatson> netzmeister: should be fixed in 6.06.2
[17:33] <netzmeister> cjwatson:  the latest ISO's on ubuntu.com?
[17:34] <netzmeister> if it is 6.06.2 the bug isn't fixed..
[17:34] <cjwatson> http://releases.ubuntu.com/6.06.2/
[17:35] <cjwatson> netzmeister: if there's still a problem, please update the bug with all relevant detail
[17:35] <cjwatson> it is entirely possible it is a specific problem with your hardware rather than a recurrence of the general problem, for instance
[17:36] <netzmeister> cjwatson:  jep there is still a problem..
[17:36] <netzmeister> :(
[17:40] <netzmeister> i don't understand, because the installer works perfectly.
[17:42] <nosrednaekim> pitti: ping
=== XSource_ is now known as XSource
[18:12] <shaya> anyone use bzr on feisty?  the package seems broken to me
[18:13] <nosrednaekim> shaya: if you are try to use it with launchpad, I think you need to use the latest version
[18:15] <shaya> nosrednaekim: just trying to run it
[18:15] <shaya> errors out w/ python error
[18:15] <shaya> can't find bzrlib
[18:15] <shaya> very default feisty installation
[18:15] <nosrednaekim> shaya: oh... whats the error?
[18:18] <shaya> ImportError: No module named bzrlib
[18:18] <shaya> Please check bzrlib is on your PYTHONPATH....
[18:19] <shaya> could it be a python problem? (updating python now, but download is slow, only 120kB/s)
[18:19] <nosrednaekim> shaya: did the package install correctly? and did you install bzr-common (I think there is apckage called that)
[18:19] <shaya> no bzr-common in fesity
[18:19] <shaya> bzr, bzrtools
[18:21] <shaya> works on my debian testing box w/o a proble
[18:23] <pochu> is ubuntu-minimal installed in buildds?
[18:24] <pochu> or does anyone know why gnome-build fails to build in the buildds but builds in my pbuilder and in debian? http://launchpadlibrarian.net/12236635/buildlog_ubuntu-hardy-i386.gnome-build_0.2.3-1ubuntu1_FAILEDTOBUILD.txt.gz
[18:24] <pochu> this is the new configure check: http://pastebin.com/f65e1711a
[18:25] <cjwatson> the buildd chroot does not install ubuntu-minimal
[18:25] <cjwatson> start with 'debootstrap --variant=buildd' if you want to reproduce it
[18:26] <shaya> ok, seemed like a python bug
[18:26] <shaya> updated python fixed it
[18:26] <pochu> cjwatson: thanks. seb128 suggested it could be liblocale-gettext-perl, which is priority required in debian, but I've noticed ubuntu-minimal depends on it (indirectly)
[18:27] <cjwatson> it's actually priority required in Ubuntu too (due to adduser), so I'm not entirely sure why it isn't installed, but no matter, you should have a build-dependency on it anyway
=== heno_ is now known as heno
[18:56] <emgent> pitti, ping
[19:38] <Riddell> slangasek: how come you didn't remove xen-unstable in bug 195453 ?
[19:38] <ubotu> Launchpad bug 195453 in xen-source "Please remove xen-{source,unstable} from hardy" [Undecided,Fix released] https://launchpad.net/bugs/195453
[19:40] <slangasek> Riddell: I was checking with geser whether it needed blacklisting; it's blacklisted now, you can do the xen-unstable removal or I can
[19:40] <Riddell> slangasek: thanks, I'll do it
=== rproenca is now known as cypherbios
=== chuck_ is now known as zul
=== slangase` is now known as slangasek
[21:38] <okaratas> hello
[21:39] <tino> hello
[21:41] <luisbg> is there a problem with the bzr package in gutsy?
[21:41] <luisbg> two friends already asked they are having trouble and they are running an old version
[21:41] <luisbg> but up to date to the repo
[21:43] <TheMuso> luisbg: They could still be bitten by the python-central issues that were around a week or so ago./
[21:44] <luisbg> TheMuso, ouch! that's true
[21:44] <geser> TheMuso: they affect also gutsy? I thought they affected only hardy
[21:44] <TheMuso> oh sorry, right.
[22:56] <somerville32> Where are the global gtkrc-2.0 settings kept?
[22:57] <seb128> somerville32: what do you want to know exactly?
[22:57] <seb128> somerville32: there is no global configuration, each theme has its own gtkrc
[22:58] <somerville32> hmmm
[22:59] <somerville32> Ok, I think you've answered my question thanks.
[22:59] <seb128> you are welcome
=== mdomsch is now known as mdomsch_BOS
[23:09] <zyx386> hi
[23:11] <zyx386> we are as Kurdish users of ubuntu, have we ubuntu on our language however mussen we manuel after that installation kurdish fonts the package to add. it would better be if beside many fonts packages kurdish fonts also in usr/shre/fonts its.
[23:12] <zyx386> can anyone answer me!?
[23:12] <zyx386> :)
[23:13] <zyx386> that is very very important for us.
[23:14] <somerville32> zyx386: You should file a bug report.
[23:14] <zyx386> somerville32, that is not bug!
[23:14] <slangasek> zyx386: the solution is to identify the missing font packages and have them added to a language-support-fonts-ku package, depended on by language-support-ku
[23:14] <slangasek> zyx386: of course it's a bug.
[23:15] <zyx386> ok who can report bug?
[23:15] <somerville32> zyx386: You can :)
[23:15] <somerville32> zyx386: http://launchpad.net/ubuntu/+filebug
[23:16] <zyx386> yes thanx
[23:17] <slangasek> rather, http://launchpad.net/ubuntu/+source/language-support-ku/+filebug
[23:17] <ScottK> zyx386: I think I know of a font that might work reasonably well that's already packaged for Ubuntu.  I need a moment to check.
[23:18] <zyx386> ok
[23:19] <zyx386> her is some report about my Q
[23:19] <zyx386> https://bugs.launchpad.net/ubuntu/+source/djvulibre/+bug/33981
[23:19] <ubotu> Launchpad bug 33981 in djvulibre "No font for Kurdish in Arabic script" [Medium,Fix released]
[23:20] <zyx386> uboto no, is not fixed
[23:21] <ScottK> https://launchpad.net/ubuntu/+source/ttf-sil-scheherazade is a font that is said to support Kurdish.
[23:21] <ogra_cmpc> according to that bug above the kurdish characters are in the DejaVu Sans font
[23:22] <slangasek> uh.  how did a font bug get filed against djvulibre?
[23:22] <zyx386> ScottK, i mean kurdish in arabic script, if inot add the font paket manuel. can't see any website in kurdish (sorani)
[23:23] <ScottK> zyx386: ttf-sil-scheherazade is an arabic font.
[23:23] <zyx386> but not kurdish
[23:23] <zyx386> arabisch fonts are not utf-8
[23:24] <slangasek> that's not true
[23:24] <ScottK> zyx386: http://www.wazu.jp/gallery/views/View_Scheherazade.html is does have Kurdish
[23:24] <slangasek> whether a font is unicode has nothing to do with whether it's arabic
[23:24] <ScottK> is/it
[23:25] <zyx386> ScottK, that is kurdisch fonts Sorani
[23:25] <zyx386> http://www.wazu.jp/gallery/Fonts_KurdishA.html
[23:26] <ScottK> zyx386: Scheherazade is one of the fonts listed on that page.
[23:28] <zyx386> ScottK, this page are not complet correct kurdisch fonts, sschehrezade fonts is arabisch fonts but is edited to kurdisch and have arabic encoding
[23:28] <zyx386> i mean 44 kurdish fonts Unkurd Fonts(unicode fonts)
[23:29] <ScottK> zyx386: Would Scheherazade be better than what you have now?
[23:29] <zyx386> no schehrezade is not Unikurd fonts
[23:30] <zyx386> the unikode font begin with Uni###
[23:30] <zyx386> that is correct list from unikurd fonts
[23:30] <zyx386> http://kurditgroup.org/download/fonts
[23:32] <slangasek> zyx386: can you please explain again why the scheherazade font is not suitable for this purpose?  It's already in the archive, so if there are bugs with it that we can fix, that would be a shorter path to a solution
[23:32] <ogra_cmpc> so scheherazade wouldnt display your webpages correctly ? do you have a page that shows wrong font behavior with the fonts missing ?
[23:33] <zyx386> slangasek, i try to explain you, sorry for my bad englisch. i can german to
[23:34] <zyx386> also Sherhrezad fonts is arabic fonts under encoding arabich
[23:34] <slangasek> what do you mean by "encoding arabisch" (arabic encoding)?
[23:34] <slangasek> are the glyphs needed by Kurdish present in the font?
[23:35] <zyx386> but the Unkurd font under encoding UTF-8, and is standard in all kurdish webpage
[23:35] <zyx386> ok
[23:35] <zyx386> the 99% kurdish website used this fonts as standard
[23:35] <zyx386> Unikurd Web
[23:36] <slangasek> the encoding for a webpage, and the encoding for a font, do not have to be the same in order for them to be used compatibly
[23:36] <slangasek> it's only required that the system *know* the encoding of each, and know how to map between the encodings
[23:37] <ScottK> http://www.omniglot.com/writing/kurdish.htm seems to show that standard Arabic fonts should be adequate.
[23:37] <zyx386> schehrezad font is unknown, the standard fonts is Unikurd Web
[23:37] <slangasek> when I visit http://kurditgroup.org/download/fonts from Ubuntu, I see text in Kurdish.  I of course don't know if it's correct, or if there are characters missing, because I don't speak Kurdish; but the text displays
[23:37] <zyx386> for exapmle
[23:38] <slangasek> so I don't see any encoding problem
[23:38] <zyx386> slangasek, can you tke screen shot
[23:39] <zyx386> for expample show all CSS page
[23:39] <zyx386> http://www.webchin.org/style/style.css
[23:39] <zyx386> Unikurd Web is  standard
[23:39] <zyx386> or Tahoma
[23:40] <slangasek> zyx386: http://people.ubuntu.com/~vorlon/Screenshot-kurditgroup.png
[23:40] <zyx386> wow
[23:40] <zyx386> i test with ubuntu!!
[23:41] <zyx386> but have ubuntu or debian, because of debian is ok
[23:41] <ScottK> Works fine with Kubuntu too.
[23:41] <zyx386> ScottK, just 1min
[23:41] <somerville32> Works fine with Xubuntu too
[23:41] <slangasek> this screenshot is with Ubuntu; it's hardy, the upcoming release, so I don't know for sure if it works the same in Ubuntu 7.10
[23:42] <ScottK> slangasek: It displays fine on my Dapper Kubuntu in Firefox, so I suspect all the ones in between work too.
[23:42] <zyx386> slangasek, of 7.10 not worked fine
[23:42] <zyx386> in Faist fawn and dapper have no problem
[23:43] <zyx386> i remove the kurdish font paket and take screenshot ok
[23:43] <ScottK2> I'm on Gutsy Kubuntu now and it looks like slangasek's screen shot.
[23:45] <zyx386> i find it
[23:45] <zyx386> the problem is firefox
[23:45] <zyx386> because of galeon have no prblemo
[23:45] <ScottK2> Works fine for me with both Firefox and Konqueror.
[23:47] <zyx386> ScottK, test this website please and take screenshot
[23:47] <zyx386> i think the problem is CSS atribude
[23:47] <slangasek> zyx386: can you show us a screenshot of the wrong behavior, instead?
[23:47] <zyx386> ok
[23:48] <zyx386> http://img508.imageshack.us/img508/8038/screenshotnh0.png
[23:49] <cjwatson> that looks like lack of proper Arabic shaping to me ...
[23:49] <cjwatson> (by which I mean Arabic and related scripts)
[23:49]  * ScottK2 needs to go eat dinner.
[23:50] <cjwatson> seems to display much better in Firefox 3 in Hardy; it may simply be that Firefox 2's Arabic shaping support was poor
[23:50] <zyx386> cjwatson, correct
[23:51] <zyx386> ok i found my annswer, and thank you Ubuntu
[23:51] <zyx386> thanx all
[23:51] <cjwatson> (not that I would know for sure, I can just see that characters are joined up better; and I have to go to bed ;-) )
[23:52] <ogra_cmpc_> probably a good ised
[23:52] <ogra_cmpc_> *idea
=== ogra_cmpc_ is now known as ogra_cmpc
[23:53] <zyx386> have nice time all
[23:55] <TheMuso> Hrm. Are there any known issues with python and apps not being able to import module _bsddb?