[12:10] <tepsipakki> Seveaz: well, the 2x.com version doesn't seem much better ;) the 110MB tarball contains not only the X tree (twice) but also many other static source-tarballs
[12:12] <tepsipakki> (like cups, openssl, zlib...)
[12:13] <tepsipakki> but maybe the are there only because of GPL
[12:19] <tepsipakki> I'll be damned, the nx build went through.. time to test it tomorrow
[12:24] <lifeless> doko: where are you ?
[12:25] <doko> lifeless: in the auditorium
[12:29] <lifeless> theres an auditorium ?
[12:30] <doko> lifeless: the big room, near the food
[01:39] <lifeless> Riddell: your branch is out of dat e with edgy :(
[01:46] <seb128> Keybuk: http://weblogs.mozillazine.org/ben/archives/009210.html
[01:53] <Burgwork> seb128: what was the final conclusion from that tab discussion?
[01:54] <Keybuk> Burgwork: we're eliminating all tabs from all applications and using multiple windows
[01:54] <Burgwork> Keybuk: right
[01:54] <seb128> tabs are evil
[01:54] <Burgwork> and who is going to carry all those patches?
[01:55] <seb128> the packages
[01:55] <Burgwork> I can't wait to see Mozilla react to this one :)
[01:56] <apokryphos> wow, very surprising
[02:06] <_ion> I'm okay with eliminating tabs from applications as long as the window manager is modified to have tabs or changed to one that has.
[02:07] <_ion> One issue with Firefox is that opening a new window takes *much* longer than opening a new tab.
[02:07] <Riddell> lifeless: mm, I certainly thought it was up to date
[02:10] <plugwash> yeah, i really can't see mozilla standing for ubuntu shipping a patch to eliminate tabs in a branded firefox
[02:16] <thom> plugwash: why would you patch? it's just an about:configure option
[02:22] <lifeless> Riddell: archive is ubuntu16, your branch is ubuntu15
[02:23] <lifeless> ogra: find me today (after wrap up) please to talk hwdb.
[02:24] <Riddell> lifeless: yeah, just read your e-mail, I'll fix that when I have a moment (which could well be next week)
[02:24] <lifeless> Riddell: thats fine, I'm not blocked. But the longer the larger the merge skew :)
[03:09] <jdong> doko: ping regarding Azureus bug 42269; I heard you have a fix ready?
[03:09] <Ubugtu> Malone bug 42269 in azureus "Does not create a tray icon" [Medium,Confirmed]  http://launchpad.net/bugs/42269
[03:11] <doko> jdong: did I say so?
[03:12] <jdong> doko: crimsun told me that you had a fixed orig.tar.gz ready to go :D
[03:12] <jdong> I'm going by what I've heard
[03:12] <jdong> either way, I have fixed source packages available if you can spare the time to flex your upload muscles :D
[03:13] <jdong> or whatever the new-fangled SRU policy dictates
[03:14] <Fujitsu> jdong: Azureus is universe, is it not?
[03:14] <bluefoxicy> someone is telling me upgrading to edgy isn't supported, you have to do a full reinstall?   o.o
[03:14] <jdong> Fujitsu: yeah
[03:14] <doko> jdong: it's universe, so oyu have to talk to dholbach first
[03:14] <bluefoxicy> I thought update-manager -c worked
[03:14] <Fujitsu> doko: Not quite.
[03:15] <jdong> all I'm saying is that the bug has been there for a year, the fix has been ther for a month.... I'd like to prod it forward towards a fix!
[03:15] <Fujitsu> jdong: You need to get the update approved by a MOTU, uploaded to {distro}-proposed, tested by 5 people for full workiness, wait for a week, then get somebody to upload it to -updates.
[03:15] <Hobbsee> bluefoxicy: from what?
[03:15] <jdong> Fujitsu: ok, hmm
[03:15] <Fujitsu> Hey Hobbsee.
[03:16] <Hobbsee> hey Fujitsu 
[03:16] <jdong> I guess I should head to -motu for this then
[03:16] <Fujitsu> jdong: Certainly.
[03:16] <bluefoxicy> Hobbsee:  Dapper
[03:17] <doko> jdong, Fujitsu: dholbach will update the process/wiki
[03:17] <bluefoxicy> speaking of unsupported activity, when is edgy+1 opening?
[03:17] <Hobbsee> bluefoxicy: maybe you'd better tell the person to put down the crack pipe.  mind you, unofficial stuff like automatix and beryl tend to bring breakage.
[03:17] <Fujitsu> doko: Are you sure he is doing that?
[03:17] <Fujitsu> bluefoxicy: Feisty will open when it opens.
[03:17] <bluefoxicy> I am starting to get tired of not having to wonder if X will break the next time I reboot
[03:17] <Hobbsee> hah
[03:17] <jdong> Hobbsee: yeah, it seems like unofficial stuff has really taken its toll this time around
[03:18] <bluefoxicy> Feisty ... ?
[03:18] <doko> Fujitsu: you are allowed to kick him if he doesn't ;)
[03:18] <bluefoxicy> Fox?  Folf?
[03:18] <Fujitsu> bluefoxicy: Feisty Fawn.
[03:18] <bluefoxicy> wtf is a fawn
[03:18] <Hobbsee> jdong: they backported mesa and xorg to a higher version than what was in edgy, according to quinn.  enough said.
[03:19] <Fujitsu> bluefoxicy: A baby deer.
[03:20] <jdong> Hobbsee: why don't we, when upgrading, pin official repositories to priority 9999 ?
[03:20] <jdong> or higher :LD
[03:20] <bluefoxicy> Fujitsu:  oh ok
[03:20] <Hobbsee> jdong: i've got no idea.  maybe more effective would be to just reinstall dapper's versions of everything, then dist-upgrade.  but what's the point, you lose the config?
[03:23] <jdong> Hobbsee: both would work. Your proposal is more sane and probably will work better
[03:23] <jdong> (restore a Dapper system before upgrading)
[03:23] <jdong> but either way, as long as it's automated, all's good
[03:23] <Hobbsee> jdong: true.  i believe mvo is trying to make it more sane
[03:23] <jdong> that's good to hear
[03:24] <jdong> I know it's easiest to blame 3rd party packagers for these problems....
[03:24] <jdong> but it'd be really nice if Ubuntu was more resilient to this
[03:24] <Hobbsee> well, when it's their problem....unfortunately, we cant really fix their stuff
[03:24] <jdong> we can't expect to control what users install, and how they go about doing it
[03:24] <Hobbsee> of course - but htey cant expect their stuff to all dist-upgrade properly either
[03:24] <jdong> right
[03:25] <jdong> but there are some feasible workarounds that we can implement to hopefully make it go more smoothly
[03:25] <Hobbsee> we can make it easier so that they wont need to use crack though.  to a degree
[03:25] <Hobbsee> true
[03:25] <plugwash> i heared a lot of people mention ubuntuguide with regard to stuff that is liable to break your next upgrade
[03:25] <jdong> Hobbsee: in addition, most of the broken upgrades I've had to deal with were easily fixable
[03:25] <Hobbsee> true
[03:25] <jdong> Hobbsee: with combinations of dpkg --configure -a, dpkg --force-overwrite $pkgname, apt-get -f install, apt-get install ubuntu-desktop, etc
[03:26] <Hobbsee> plugwash: true.  i didnt think ubuntuguide was that bad...
[03:26] <jdong> those are all things a script can try
[03:26] <Hobbsee> jdong: force overwrite is a scary option to do via a script - how do you know what you're overwriting?
[03:26] <jdong> Hobbsee: the guide does have some HOWTO's that instruct to install some guy's debs
[03:26] <Hobbsee> true
[03:26] <jdong> Hobbsee: judging that the upgrade is TO an official Ubuntu version, I would HOPE an overwrite is appropriate :D
[03:27] <Hobbsee> good point.  but then they lose their config?
[03:27] <jdong> I wouldn't say so
[03:27] <jdong> config files aren't affected by installing a new deb package
[03:27] <jdong> they are managed as config files
[03:27] <jdong> force-overwrite only applies to the non-config files....
[03:28] <Hobbsee> they do?
[03:28] <jdong> config files are handled in the Setting up $pkgname step....
[03:28] <jdong> and if there are differences, the user gets prompted about it
[03:29] <jdong> also, dist-upgrader should check for locally installed packages, and also members of the 'checkinstall' group, and warn users about these packages
[03:29] <jdong> especially if they are members of checkinstall :D
[03:29] <jdong> Fujitsu: can you acknowledge in the bug ticket that someone is doing something about it? :D
[03:30] <jdong> mv lastmessage #ubuntu-motu
[03:30] <Hobbsee> but the user doesnt know, and shouldnt have to
[03:31] <plugwash> imho anything not installed from official or known sane unofficial repositries should trigger a warning
[03:31] <jdong> plugwash: +1
[03:33] <leonel> no official ubuntu  position on the ms - novell agreement 
[03:33] <leonel> ?
[03:34] <plugwash> leonel what agreement?
[03:35] <Hobbsee> heya mpt 
[03:35] <Hobbsee> leonel: why would ubuntu have one?  and you wont get one at this time of day
[03:35] <jdong> leonel: does Ubuntu have to give a statement on everything that happens in the IT world?
[03:36] <jdong> I'd really love a Mark Shuttleworth statement on my recent purchase of a 12-inch Subway sub that had a free Coke with it
[03:56] <leonel> wow  nice  answers ...
[03:57] <Hobbsee> leonel: basically, the people you want arent around here - they're all together at a conference.
[03:59] <leonel> Hobbsee: what worries me  is that they say that  only suse is free from  patent problems   that's why I'd like to hear something from Ubuntu  about it  as redhat did   that's all
[06:59] <tepsipakki> does the "don't show tabs on default" mean that I'm still able to open a new one with ctrl-t (or similar) on firefox et al? the spec-wiki doesn't mention that
[07:00] <Treenaks> yes..
[07:00] <tepsipakki> phew..
[07:00] <Treenaks> I don't see a tab bar when I only have one tab, basically
[07:00] <tepsipakki> makes sense
[07:00] <Treenaks> which has been the default since forever, afaik
[07:01] <tepsipakki> but not on every app that uses them
[07:01] <tepsipakki> like gedit
[07:01] <Treenaks> gedit should abandon them :)
[07:01] <Treenaks> imho
[07:01] <tepsipakki> altogether? noo :)
[07:02] <Treenaks> Except for web browsers, I'm a one-window one-file kind of person
[07:03] <tepsipakki> but.. that's not consistent :)
[07:03] <Treenaks> I never claimed to be consistent ;)
[07:03] <tepsipakki> heh
[07:03] <tepsipakki> I never use gnome-terminal tabs.. so there
[07:03] <tfheen> tepsipakki: sure is, all applications using tabs will then behave in the same way.
[07:04] <Treenaks> tepsipakki: gnome-terminal tabs should map to screen screens.. THEN I'll be happy ;)
[07:05] <Fujitsu> Treenaks: Wouldn't that be just so nice!
[07:05] <Fujitsu> screeen FTW!
[07:05] <Ng> Treenaks: you can do that with screen -x, but obviously it's a manual setup process per-tab
[07:05] <tepsipakki> tfheen: yes, thanks for clarifying that.. I was close to getting upset about the change, until I saw what it really meant :)
[07:06] <Treenaks> Ng: it also breaks when I do Ctrl+A space ;)
[07:09] <Ng> Treenaks: well yeah, you'd use the g-t shortcuts for next/previous tab
[09:21] <mnepton> Hobbsee: Brandon keeps asking if i have nude photos of you. make it stop.
[09:22] <Hobbsee> mnepton: hah.  no one has any of them.
[09:22] <mnepton> Hobbsee: that you know of.
[09:22] <Hobbsee> mnepton: i'd know if any had been taken.
[09:23] <Mez> wait... of Hobbsee?
[09:23] <Burgundavia> mnepton: I can said him nude shots of me with her head pasted on? :D
[09:23] <Mez> wrong person ;)
[09:23] <Burgundavia> might be good for a laugh
[09:23] <Hobbsee> hahaha
[09:23] <Mez> Burgundavia, DO IT :P
[09:23] <Mez> hmmm
[09:23] <Hobbsee> i'm sure that'd work out *real* well among a dev conference.
[09:23] <Mez> my router sounds like a brothel
[09:24] <Mez> "multiple PVC operation"
[09:25] <Hobbsee> Burgundavia: that would raise interesting questons
[09:28] <mnepton> we could just blame Novell
[09:30] <Mez> since when has feisty been open ?
[09:32] <mnepton> 10/28
[09:32] <mnepton> errr, 26
[09:32] <mnepton> Burgundavia: are you not at UDS?
[09:32] <StevenK> I see that, does that mean the toolchain is sorted out?
[09:32] <Burgundavia> mnepton: I wish. Work and I disagreed about whther I should go. I lost
[09:32] <mnepton> grah. le suck.
[09:32] <Mez> hmm... /me doesnt know
[09:33] <Mez> I just got a bounce from someone uploading a new version of katapult
[09:33] <Mez> or maybe everythings still under manual revire ?>
[09:33] <Mez> review *
[09:33] <mnepton> Burgundavia: anything you need brought to anyone's attention?
[09:33] <StevenK> Feisty is set to DEVELOPMENT, so it should be okay.
[09:33] <Burgundavia> mnepton: nope
[09:34] <Burgundavia> I have my little minions doing my work for me
[09:35] <mnepton> heheheh
[09:36] <mnepton> is any of that work "give mnep cash?" and if not, why?
[09:36] <Hobbsee> mnepton: because its' going to me.  duh.
[09:36] <mnepton> *facepalm*
[09:36] <Burgundavia> mnepton: no, no it is not
[09:37] <mnepton> i love you, too, Corey :)
[09:37] <Burgundavia> I try, but I am broke currently
[09:37] <Burgundavia> end of the month, I am rich, but not today
[09:37] <CheGuevara> hi
[09:38] <ereminii> have everyone been able to get the latest feisty updates
[09:38] <ereminii> like new gnome etc
[09:38] <Hobbsee> most people arent trying to run feisty on a separate partition yet
[09:38] <Hobbsee> and no, that wont have been merged yet
[09:42] <ereminii> heh i run it on the one and only box as main os
[09:42] <ereminii> oh that explains it
[09:42] <ereminii> i just saw them all in commit log
[09:42] <ereminii> but in update manager its empty
[09:42] <Hobbsee> that's kinda....silly...to say the least
[09:44] <ereminii> after using gentoo unstable for 2 years, i am pretty sure it dont get more on the edge then that
[09:44] <Hobbsee> oh good, as long as you're aware
[09:44] <Hobbsee> and know how to use dpkg when apt breaks
[09:47] <ereminii> yeah
[09:47] <ereminii> i am probably am silly
[09:47] <ereminii> just like always having the latest versino numbers
[09:47] <ereminii> cant explain it heh
[09:50] <hunger> ereminii: gentoo was always behind debian with package versions (both unstable) with all the packages I cared about.
[09:53] <ereminii> gentoo ehind packages vesions?
[09:53] <ereminii> seems impossible heh
[09:53] <ereminii> filing bug reports usually works
[09:54] <ereminii> unless of course its some rare stuff with no maintainer
[09:54] <hunger> ereminii: Well, I tend to care about the rare stuff:-)
[09:56] <ereminii> oh heh
[09:56] <ereminii> fun starts when u downgrade glibc lol
[10:00] <ereminii> gotta take off battery almost died
[10:00] <ereminii> cya and thx for ur help
[12:28] <blujay> Can I get a dev's attention on a very widely-annoying Edgy bug that desperately needs to at least have a priority set in Launchpad?  bug#64841
[12:29] <Fujitsu> bug #64841
[12:29] <Ubugtu> Malone bug 64841 in wlassistant "wireless assisant does not connect in edgy" [Undecided,Confirmed]  http://launchpad.net/bugs/64841
[12:30] <blujay> (Hm, on a different note, perhaps Ubugtu could also recognize "bug#xxxx" format as well.  :)
[12:32] <Hobbsee> blujay: why not use knetworkmanager?
[12:33] <Hobbsee> and what's the solution there?  fiddling with the interfaces file?
[12:33] <blujay> Hobbsee: Well, I thought it was pretty widely-known that NetworkManager only works for a few cards/configurations right now.  And no, the problem seems to be something with either dhclient or the kernel; fiddling with /etc/network/interfaces doesn't fix it.
[12:34] <blujay> Hobbsee: One of the duplicate bugs (there have been about 4 now) had a high priority set, but this one (the first) has none.
[12:35] <Hobbsee> hmmm
[12:37] <Hobbsee> hmmm, i thought it looked familiar. i replied to one of those bugs
[12:38] <blujay> Hobbsee: This is one of the things that bugs me about Ubuntu.  In Debian anyone can adjust the priority of bugs, but in Ubuntu...this bug has been filed for a long time, and has had many dupes, but no devs are taking notice, but it's very important.
[12:38] <StevenK> blujay: It's important to *you*.
[12:38] <Hobbsee> blujay: just because it's set to high doesnt mean that any devs necessarily see it.  
[12:38] <blujay> StevenK: This affects most wifi users on Edgy.  Isn't that important?
[12:39] <Hobbsee> blujay: excluding all the ones who arent using knetworkmanager or dhclient, yes
[12:39] <StevenK> And using wlassistant.
[12:39] <Hobbsee> StevenK: it's a kde thing.  i've never been able to get the bugger to work, but i'd presumed it was a config somewhere
[12:39] <StevenK> Ah.
[12:39] <blujay> wlassistant works fine in Dapper, perfectly.  It worked in Edgy Knot2.  But not since then.
[12:40] <StevenK> The bug said Knot 3. 
[12:40] <blujay> StevenK: ...ok so I was off by one.  The point remains.  :p
[12:40] <blujay> Hobbsee: thank you.
[12:40] <Hobbsee> blujay: not a problem, but that doesnt really change anything
[12:41] <blujay> Hobbsee: Well...do you have any suggestions for getting the devs to pay attention to this?
[12:41] <StevenK> blujay: If you have the skills/want to learn, debugging it out could really help.
[12:41] <Jozo-> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=375551 wlassistant: Wrong call to dhclient 
[12:41] <blujay> StevenK: I'll be glad to help the devs, but digging into it isn't really practical, because I have a very old, very slow laptop to run it on.
[12:41] <Ubugtu> Debian bug 375551 in wlassistant "wlassistant: Wrong call to dhclient" [Normal,Open]  
[12:41] <Hobbsee> blujay: they're at a conference this week.  i dont think they're going to do anything about it at the moment
[12:42] <StevenK> blujay: That isn't a problem.
[12:42] <blujay> Hobbsee: Well I realize that, but...after that?  I mean, this was broken in the Edgy process.  It's just a shame.  :/
[12:43] <blujay> Jozo-: thanks, I'll add that to the Launchpad report
[12:43] <Hobbsee> StevenK: do you know the debian maintainer?
[12:43] <StevenK> I know of him.
[12:43] <Hobbsee> StevenK: is he active?
[12:44] <Hobbsee> hehe
[12:44] <Hobbsee> dont wave them around too much.
[12:44] <StevenK> :-P
[12:46] <StevenK> Hobbsee: Unsure. 
[12:47] <Hobbsee> StevenK: what happens if you remove -r from the config?  does it work then?
[12:47] <StevenK> Hrm?
[12:47] <Hobbsee> StevenK: the debian bug
[12:47] <StevenK> How can I be lazy when you're asking me to a look at bug!
[12:48] <StevenK> s/\(bug\)/\1s/
[12:48] <Hobbsee> hehe
[12:49] <StevenK> Neat.
[12:49] <Hobbsee> hmmm?
[12:49] <Hobbsee> i seem to remember reproducing debian's bug before
[12:50] <StevenK> wlassistant is dragged in by kubuntu-desktop.
[12:50] <Hobbsee> yes, it is
[12:50] <Hobbsee> as we dont load knetworkmanager by default, and kwifimanager is a POS
[12:53] <Hobbsee> indeed
[12:53] <StevenK> Oh geez, I hope I don't have to install kubuntu-desktop on my laptop.
[12:55] <StevenK> I'll kill networkmanager and see if I can convince wlassistant to work
[12:55] <blujay> StevenK, Hobbsee, thank you for checking this out.
[12:55] <Hobbsee> okay
[12:55] <Hobbsee> blujay: that doesnt mean we'll be able to fix it
[12:56] <blujay> Hobbsee: I know, but I still appreciate your checking.
[12:59] <StevenK> Oh neat.
[12:59] <StevenK> wlassistant looks to be crap, too.
[12:59] <blujay> I had a feeling that, "Oh neat," was sarcasm.  :)
[01:00] <blujay> StevenK: It's not great, but I will say that it works fine in Dapper, and it's the only GUI WiFi tool that does, AFAIK.
[01:00] <StevenK> Yeah, who'd thunk it. :-P
[01:00] <blujay> StevenK: for KDE, that is.
[01:00] <Hobbsee> StevenK: yeah well.  breezy was worse
[01:01] <StevenK> KDE users should stick to wired. :-P
[01:01] <Hobbsee> hah
[01:01] <Hobbsee> yes, but what about gnome users?
[01:02] <Hobbsee> you mean you hadnt?  what were you looking for before?
[01:02] <StevenK> The debug log.
[01:02] <StevenK> That told me enough
[01:02] <StevenK> The amount of *crap* it spews to the console? I mean, geez
[01:04] <Hobbsee> NM is worse
[01:04] <Hobbsee> mind you, it's built wiht a very verbose wpa supplicant
[01:04] <vogelfaull> why
[01:05] <Hobbsee> haha
[01:05] <vogelfaull> why
[01:05] <Hobbsee> the last log it appears to work....right?
[01:05] <Hobbsee> vogelfaull: ?
[01:05] <vogelfaull> why
[01:06] <Hobbsee> dont you start here too, you rotter...
[01:06] <StevenK> I hope I'm reading this wrong.
[01:06] <Hobbsee> StevenK: which bit?
[01:06] <StevenK> "Let's set all of this stuff up. Okay, done. Now let's kill the DHCP client. Hrrrm, I don't seem to have a connection."
[01:07] <bhale> Hobbsee: why?
[01:07] <Hobbsee> bhale: because i said so!
[01:07] <Hobbsee> :P
[01:07] <bhale> :)
[01:07] <Hobbsee> StevenK: yes, i *thought* that might be what it said.   and then appears to want to kill it harder
[01:07] <lastnode_> Hobbsee, he had a very limited vocabulary :-)
[01:08] <Hobbsee> lastnode_: the idiot was just in #ubuntu too - which was why i banned him so quickly
[01:08] <lastnode> :-)
[01:08] <Hobbsee> right, that's *3* channels the rotter has now been banned in
[01:09] <Hobbsee> [23:08]  [Whois]  vogelfaull is a user on channels: ##php #emacs #nethack #wikipedia-de
[01:09] <blujay> Hobbsee: I'm curious..."rotter"?  Where did that originate?
[01:09] <StevenK> rotter
[01:09] <StevenK>        n : a person who is deemed to be despicable or contemptible;
[01:09] <apokryphos> blujay: just possibly a bot. Same "why" repeating or general other abusive messages
[01:09] <Hobbsee> blujay: no idea?  rotter being rotten person?
[01:09] <Hobbsee> apokryphos: or an annoying user
[01:09] <Hobbsee> apparently it's a java user, but...
[01:09] <blujay> Hobbsee: ah, just not used much in the US I guess
[01:09] <apokryphos> right
[01:09] <Hobbsee> blujay: musnt be
[01:09] <StevenK> blujay: But neither is English.
[01:10] <blujay> StevenK: hehe...where are you from then?
[01:10] <StevenK> .au
[01:10] <blujay> ah
[01:10] <lastnode> i hate you xchat win32 :\
[01:10] <blujay> at least most of us don't eat Vegemite...yech
[01:10] <lastnode> Hobbsee, had a chance to check out the Upstream .debs yet? :-)
[01:10] <StevenK> Right, I'm mistaken, it does run dhclient.
[01:10] <Hobbsee> lastnode: nope
[01:10] <StevenK> blujay: But Vegemite is great!
[01:10] <lastnode> marmite rocks btw
[01:11] <Hobbsee> StevenK: after killing all other copies of it, apparently.  or trying to
[01:11] <blujay> StevenK: I actually tried some once, brought to the US by a genuine Australian.  But...great?  ...?  :)
[01:11] <lastnode> Hobbsee, cool, just asked
[01:11] <Hobbsee> :)
[01:12] <StevenK> blujay: All I know is I like it on toast with butter.
[01:12] <StevenK> Occasionally by itself on a knife, too.
[01:12] <blujay> StevenK: with butter too?  wow
[01:12] <lastnode> Hobbsee, don't get them .debs, ever!
[01:12] <Hobbsee> haha
[01:12] <Hobbsee> lastnode: fix wlassistant, kthnksbye!
[01:13] <lastnode> lolz what is wlassistant kk?
[01:13] <lastnode> Hobbsee, bug number?
[01:13] <StevenK> Hobbsee: That's, 'ktnxbye'
[01:13] <lastnode> kthxbye
[01:13] <StevenK> Oh right.
[01:13] <StevenK> I was closer.
[01:14] <StevenK> AH HA!
[01:14] <Hobbsee> lastnode: bug 62156
[01:14] <Ubugtu> Malone bug 62156 in dhcp3 "no ip address after boot" [Undecided,Unconfirmed]  http://launchpad.net/bugs/62156
[01:14] <Hobbsee> oops
[01:14] <StevenK> I know what is going on.
[01:14] <Hobbsee> lastnode: https://launchpad.net/distros/ubuntu/+source/wlassistant/+bug/64841
[01:14] <Ubugtu> Malone bug 64841 in wlassistant "wireless assisant does not connect in edgy" [High,Confirmed]  
[01:14] <Hobbsee> StevenK: what?
[01:14] <lastnode> lolz wrong bug kthx bye
[01:16] <StevenK> wlassistant fires off dhclient, which complains about a zero length PID file on stderr. wlassistant takes any output on stderr as being a problem, and so kills dhclient.
[01:16] <StevenK> Stupid bloody things.
[01:16] <blujay> Yay, good job StevenK!
[01:16] <StevenK> So the bug is dhclient being a piece of crap. :-P
[01:17] <StevenK> The bug can be worked around in wlassistant, however.
[01:17] <StevenK> Which requires learning QT file handling functions.
[01:18] <StevenK> And coding C++.
[01:18] <Hobbsee> hahaha
[01:18] <StevenK> Neither really appeals.
[01:18] <Hobbsee> StevenK: or fix dhclient?  nah...
[01:19] <StevenK> Maybe.
[01:19] <StevenK> dhclient sucks a lot, though.
[01:19] <Hobbsee> am i being slow tonight, or do you just have to check if the output on the stderr is about the zero length PID, or nothing else fail?
[01:19] <blujay> StevenK: is there anything that doesn't suck? :)
[01:19] <Hobbsee> haha
[01:20] <Hobbsee> blujay: sure.  linda.
[01:20] <StevenK> Hah!
[01:20] <Hobbsee> or does that suck too?
[01:20] <blujay> Hobbsee: I know lintian  but I'm not familiar with linda
[01:20] <StevenK> Of course it does, I wrote it.
[01:20] <Hobbsee> blujay: its' another package checker
[01:20] <Hobbsee> hahaha
[01:20] <blujay> ah
[01:20] <StevenK> Hobbsee: The point is dhclient should remove it's PID file on exit.
[01:20] <Hobbsee> well, seeing as you tend to write sucky things, how about you fix it so it doesnt suck
[01:20] <Hobbsee> true that
[01:21] <StevenK> Hmph!
[01:21] <blujay> StevenK: So, how come Dapper and pre-Knot3 didn't have this problem?  :/
[01:21] <Hobbsee> heh
[01:21] <StevenK> blujay: dhclient could have been upgraded?
[01:21] <Hobbsee> blujay: i think i saw one there for knot 2.  but a dhclient update could do ti
[01:21] <StevenK> Hobbsee: Why don't you fix it? :-P
[01:21] <blujay> StevenK: so, a new bug in dhclient, or a regression...?
[01:21] <Hobbsee> StevenK: because you like fixing things
[01:22] <StevenK> blujay: Oh, who knows. dhclient is getting that bad it could have damn well updated itself
[01:22] <blujay> hehe
[01:22] <StevenK> Anyway, gone now.
[01:22] <blujay> StevenK: thanks for looking into it, I'll update the bug report
[01:23] <Hobbsee> why is he going shopping for anything at this time of night?
[01:23] <Hobbsee> oh yeah, i see the problem now
[01:38] <StevenK> Hobbsee: Because we ran out of milk.
[01:38] <Hobbsee> StevenK: ahh
[01:39] <StevenK> Right, this bug can fixed in one (or more) of three ways.
[01:39] <StevenK> 1. wlassistant shouldn't assume stderr means an error.
[01:39] <StevenK> 2. dhclient cleans up its pid file on exit.
[01:39] <StevenK> 3. dhclient ignores zero-length pid files when it starts.
[01:40] <StevenK> I think it should be fixed in dhclient, since this might be biting other things.
[01:41] <Hobbsee> so will you take 2 or 3?
[01:41] <StevenK> Ahh, that I'm not sure about. :-)
[01:41] <Hobbsee> and will they accept it as a SRU?
[01:41] <StevenK> That's up to Kamion and/or mdz.
[01:41] <StevenK> I suspect so.
[01:42] <StevenK> When was knot 3 released?
[01:44] <Hobbsee> sometime before 19/10
[01:44] <StevenK> I hope so, that being the beta date
[01:45] <Hobbsee> maybe a week before, or something?
[01:45] <Hobbsee> yeah
[01:45] <StevenK> Hum.
[01:45] <StevenK> I think silencing it is the easiest option.
[01:47] <StevenK> Submitted by pitti, even
[01:50] <admin123> ok oem-config doesn't work correctly with nl_NL@euro it crashes!
[01:51] <admin123> i tried this on several machines, and they all work fine with oem-config until a missing locale is generated nl_NL@euro
[01:52] <Hobbsee> abattoir: ^
[01:52] <Hobbsee> abattoir: or is that not your domain?
[01:53] <Hobbsee> StevenK: fix it harder?
[01:54] <StevenK> dhcp3 is building now.
[01:54] <Hobbsee> :)
[01:54] <StevenK> But I will talk to pitti about it.
[01:58] <abattoir> Hobbsee: i think Kamion would be in a better position to answer it ;)
[01:58] <abattoir> admin123: was the installation in a
[01:58] <abattoir> *in English?
[01:58] <Hobbsee> abattoir: ah
[01:58] <Hobbsee> StevenK: heh.  clever
[01:59] <StevenK> Hobbsee: Hush.
[02:00] <abattoir> bug 68030
[02:00] <Ubugtu> Malone bug 68030 in oem-config "configures wrong locale" [Undecided,Unconfirmed]  http://launchpad.net/bugs/68030
[02:00] <abattoir> admin123: is it similar to that^^^ ?
[02:01] <StevenK> Oh DAMN
[02:02] <Hobbsee> haha
[02:02] <Hobbsee> StevenK: *very* clever.
[02:02] <StevenK> Hmph.
[02:02] <Hobbsee> :
[02:02] <Hobbsee> :)
[02:07] <StevenK> I have this feeling the way dhclient handles its pid files leaves a lot to be desired.
[02:14] <Hobbsee> hehe
[02:14] <Hobbsee> you can do it!
[02:18] <StevenK> If you don't initialise the variable, it's very hard to see if it's been set because it's full of uninitialised memory.
[02:18] <StevenK> Yes, pitti this means you!
[02:18] <Hobbsee> haha
[02:18] <Hobbsee> what's it in?  python?
[02:18] <StevenK> C, if you please.
[02:19] <StevenK> Python doesn't suffer from this problem.
[02:20] <Hobbsee> ah
[02:20] <Hobbsee> c++ it at least generates a warning.  then again, if there are a lot of them...
[02:21] <StevenK> That's a point.
[02:21] <pepsiman> python throws an exception
[02:21] <StevenK> Oh yeah
[02:22] <StevenK> Hobbsee: Yes, if you compile with -Wall... which dhcp3 doesn't. :-)
[02:22] <Hobbsee> ah
[02:23] <StevenK> Well, -W<stuff>
[02:26] <Hobbsee> heh
[02:27] <neuralis> pepsiman: no, python doesn't have the concept of uninitialized variables. don't confound this with undeclared variables.
[02:27] <Hobbsee> haha
[02:28] <StevenK> infinity: Oh good, wanna help me debug this bloody thing? :-P
[02:28] <StevenK> "ifconfig_dhcp: /sbin/dhclient -q eth1"
[02:28] <StevenK> *pause*
[02:28] <infinity> Wasn't making feisty's kernel build on i386 and doing a few more hours of hppa bootstrapping enough good deeds for the night? :)
[02:29] <StevenK> "kill_dhcp: /sbin/dhclient -r eth1"
[02:29] <neuralis> infinity: i've been unable to sleep as well.
[02:29] <Hobbsee> infinity: nope
[02:29] <infinity> Bah, it's only 12:30 back home.
[02:29] <Hobbsee> you're losing that battele
[02:29] <Hobbsee> *battle
[02:29] <infinity> Be a man.
[02:30] <StevenK> My wife has already asked me to come to bed twice.
[02:30] <infinity> Oh, then be a man and go to bed with your wife. :)
[02:30] <StevenK> Bwahaha
[02:31] <highvoltage> heh
[02:31] <StevenK> The conclusion I'm coming to is that wlassistant is POS.
[02:31] <StevenK> is a, even
[02:31] <StevenK> Use knetworkmanager, => Rejected
[02:31] <StevenK> :-P
[02:31] <infinity> NM does the job for me.
[02:31] <infinity> No perfectly, but "well enough".
[02:32] <StevenK> Which ever bastard though writing a C++ KDE GUI wrapper around dhclient, ifconfig and iwconfig needs to be SHOT.
[02:32] <StevenK> s/iwconfig/iwconfig was a good idea/
[02:33] <Hobbsee> haha
[02:33] <StevenK> infinity: Can I at least bounce ideas of you? I know what the problem is, even if I haven't fixed it.
[02:33] <infinity> Bounce away.
[02:33] <infinity> I may put on some pants, sneak out of the room (without waking thom, fun!) and go have a smoke.
[02:34] <StevenK> infinity: The problem is that wlassistant fires off dhclient -q eth1, notes there's output from dhclient saying that it has a stale pid file on stderr, declares the sky is falling and kills dhclient.
[02:35] <pepsiman> neuralis: technically you're right, but in practice it's the same thing
[02:35] <StevenK> So we can either fix wlassistant to ignore that erorr, or have dhclient not complain.
[02:35] <infinity> StevenK: And doing those steps by hand, you're confirming that wlassistant is correct about dhclient's crackful returns?
[02:35] <StevenK> That was reading the code by hand.
[02:36] <StevenK> There seems to be something else going on, though.
[02:36] <StevenK> I've rebuilt dhclient to quieten it, and it still doesn't work.
[02:36] <StevenK> Ahh, no. It's because my rebuild causes dhclient to segv.
[02:37] <StevenK> Riiiiiight.
[02:39] <Hobbsee> heh
[02:39] <Hobbsee|Remote> they're good for setting on fire
[02:39] <fabbione> morning
[02:39] <StevenK> Really now?
[02:39] <Hobbsee> hey fabbione (here too!)
[02:39] <Hobbsee> StevenK: of course!
[02:40] <StevenK> That's curious, since I still found one.
[02:40] <StevenK> So nyah
[02:40] <infinity> Hotel wireless go boom.  Yay.
[02:40] <Hobbsee> hehe
[02:40] <StevenK> Heh
[02:40] <infinity> StevenK: Hah.  Fixing the segv fixed it? :)
[02:41] <StevenK> I tell you, killing NetworkManager every 3 minutes is therapeutic.
[02:41] <Hobbsee> haha
[02:41] <Hobbsee> sure it is :)
[02:43] <StevenK> Right. SEGV fixed.
[02:44] <infinity> Was the segv from your "insult pitti and initialise his variables for him" fix?
[02:44] <StevenK> Yup
[02:44] <infinity> Always classy to introduce a worse bug when fixing a lesser one. :)
[02:44] <StevenK> Well, I try. :-)
[02:44] <StevenK> long temp = 0; fscanf(..., &temp); made it blow up.
[02:45] <infinity> Gee, I can't imagine why.
[02:45] <infinity> (Also, hooray for useless variable names, pitti)
[02:45] <StevenK> Heh
[02:46] <StevenK> scanf() and friends just seem to be entirely counter-intuitive.
[02:46] <infinity> Well, I won't disagree there.
[02:47] <StevenK> infinity: What bonehead did I make, so I can file it away?
[02:47] <StevenK> bonehead error, even
[02:49] <StevenK> Okay, wlassistant, you suck.
[02:49] <StevenK> DHCP works, and it goes on to test the connection by some nefarious means, and fails.
[02:51] <StevenK> Ahh, more Aussies who can't/won't sleep.
[02:53] <Yagisan> StevenK, its can't and only because I recently threw out a crackful, Win32 centric threading system and replaced it will SDL threads, and apparently I'm supposed to test it
[02:54] <Yagisan> StevenK, I mean, isn't that what users are for ;)
[02:54] <StevenK> Indeed.
[02:54] <StevenK> Just applications are the test suite for the kernel. :-P
[02:54] <StevenK> Just like
[02:55] <Yagisan> StevenK, at least you know "temp" is a temp variable, code I just fixed had "i" as a temp counter O_o
[02:56] <StevenK> Heh.
[02:56] <StevenK> I like using i as a counter for short for loops.
[02:56] <Yagisan> StevenK, and for reasons unknown to me, assumes pointers = ints everywhere. :'(
[02:56] <StevenK> Yagisan: Because that holds true on i386? :-(
[02:57] <Yagisan> StevenK, and 32bit powerpc
[02:57] <Yagisan> StevenK, if you want to chat more about it, we can go to pm
[02:57] <StevenK> Yagisan: Heh, I'll cope. :-)
[02:58] <StevenK> bhale: Ah, but is your amd64 a quiet SFF PC? :-)
[02:59] <bhale> StevenK: no its a big loud dell 2850
[02:59] <StevenK> Heh
[02:59] <bhale> actually a dozen or more of them
[02:59] <Yagisan> my poor abuse amd64 is in a big ugly green tower that looks like it houses an old i386 server
[02:59] <Yagisan> s/abuse/abused
[02:59] <StevenK> Heh
[03:00] <Yagisan> my p2 is in the shiny new case
[03:00] <bhale> that gives me an idea to mod an SFF into my sgi o2 chassis
[03:00] <bhale> not that I would do it
[03:00] <Yagisan> so if it gets pinched, someone gets a rude shock :)
[03:30] <neuralis> pepsiman: no, it's really not the same thing at all.
[04:17] <bddebian> Howdy
[05:50] <sladen> iwj: is there an equivalent way to (/updates) of updating dpkg/available without having to rewrite it each time?
[05:51] <sladen> iwj: see the Issues section at the end of https://wiki.ubuntu.com/LiveCDStackedFileSystem
[05:55] <psusi> what's wrong with the file being replaced a few times?
[05:57] <tfheen> sladen: unionfs doesn't give you diffs, it gives you the file multiple times anyway.  I've discussed evil ways of working around this already.
[05:58] <sladen> tfheen: could you update the page if you already have a solution figured out
[05:58] <tfheen> sladen: no, we have not decided on what approach we'll use.
[05:59] <sladen> tfheen: if there are some ideas for a solution that I didn't write up, could you add them
[06:00] <tfheen> sladen: the spec is not "we might want to pursue this approach, or this approach or this approach".  It's "we'll solve the problem this way"
[06:04] <iwj> sladen: No, there isn't.
[06:04] <sladen> tfheen: so, are there any other ideas ("I've discussed evil ways of working around this already") that are not one of the three listed at the bottom
[06:04] <sladen> iwj: thanks
[06:06] <tfheen> sladen: correct.
[06:06] <iwj> sladen: You could supply an empty available file and download it as needed ?
[06:08] <sladen> tfheen: what are those other ideas, it would be useful to have them recorded somewhere
[07:04] <fabbione> jdub: ping?
[07:08] <Riddell> keescook, tfheen: I added KDE bits to network-roaming, please let me know if that's good or no
[07:40] <tfheen> oh, this is special.
[07:40] <tfheen> >>> isinstance("blah", str)
[07:40] <tfheen> True
[07:40] <tfheen> >>> isinstance(u"blah", str)
[07:40] <tfheen> False
[07:40] <tfheen> (yes, it's true, but it makes mailman unhappy)
[07:41] <robertj> tfheen: I thought isinstance was eebil?
[07:42] <tfheen> it's used in the email package
[07:42] <tfheen> not my code
[08:29] <thom> infinity: quit slacking and get lrm in for feisty
[08:34] <chrisj> How do I submit a bugfix for a package thats in main?
[08:34] <chrisj> a patch, it's just a one liner
[08:34] <bddebian> chrisj: Attach it to a bug report
[08:34] <Lure> chrisj: attach patch to bug
[08:35] <chrisj> bddebian, Lure: thanks
[08:36] <bddebian> NP
[08:42] <keescook> Riddell: thanks for the kde bits; I think it looks fine.  We'll see what smurf thinks now.  :)
[08:45] <keescook> doko: what do you think of turning on RELRO by default for the feisty toolchain?  gentoo seems to have tested this and not seen any problems with it.
[08:46] <fabbione> keescook: too late
[08:47] <fabbione> keescook: plus glibc detects it automatically
[08:47] <fabbione> i don't think you need more than that
[08:48] <cr3> fabbione: quick question about preseeding: how can I write or generate a preseed file?
[08:48] <keescook> fabbione: yeah, figured it was too late, but I think I'd ask.  I thought the toolchain needed to mark up the ELF sections for glibc to actually allow for the differing segment permissions?
[08:48] <desrt> seb128; poke
[08:48] <seb128> outch
[08:48] <desrt> got a sec?
[08:48] <fabbione> keescook: not sure, but it's for feisty+1
[08:48] <seb128> on IRC? yep
[08:48] <fabbione> cr3: cjwatson.. i am not the installer god
[08:48] <desrt> please make https://features.launchpad.net/distros/ubuntu/+spec/easy-codec-installation depend on https://features.launchpad.net/distros/ubuntu/+spec/gnome-app-install-codecs
[08:50] <fabbione> pitti: ping?
[08:51] <desrt> alternatively, please flip whatever switch will allow me to do so :)
[08:54] <gnomefreak> is it safe to reboot when linux-libc-dev doesnt match kernel version?
[08:54] <seb128> desrt: looking at doing the change, launchpad UI doesn't make it easy
[09:01] <seb128_> desrt: done
[09:02] <desrt> seb128_; thanks
[09:02] <seb128_> np
[09:04] <pitti> fabbione: busy pong
[09:04] <fabbione> pitti: no rush.. libvirt is out of NEW
[09:04] <pitti> ah, nice
[09:42] <ulaas> hi where can i get good technical info on "upstart"
[09:44] <doko> keescook: you missed the feisty toolchain session
[09:44] <Ng> ulaas: upstart.ubuntu.com
[11:02] <mario_> sivang, ping?
[11:27] <cjwatson> Keybuk: can I just dump override-Ubuntu-changes syncs I want to make into ~lp_archive/syncs/ and have you punt them into the sync queue when you're done with the rest?
[11:28] <mario_> hey cjwatson 
[11:28] <cjwatson> hi
[11:59] <niktaris> hi, I am trying to pass some preseed options to casper but it seems no to work. I am giving xserver-xorg/config/inputdevice/keyboard/variant=us,gr on the boot prompt. Anyone has an idea why ?
[11:59] <cjwatson> niktaris: I'd recommend using console-setup/variant=us,gr instead - xorg should grab its defaults from there
[12:00] <cjwatson> assuming edgy
[12:01] <niktaris> hi cjwatson. I am trying to pass 2 options. 1 is the variant option to xorg.conf. and 2 the "Xkboptions" 
[12:02] <niktaris> cjwatson, so basicly xorg.conf should have Xkblayout "us,gr" and Xkboptions "ctrl:alt_shift_toggle". Any tips ?
[12:02] <niktaris> yes on edgy
[12:04] <cjwatson> do you mean grp:alt_shift_toggle?
[12:04] <niktaris> cjwatson, yes
[12:04] <cjwatson> niktaris: just say console-setup/layout=gr and console-setup should sort out the rest
[12:05] <cjwatson> it knows that gr is a non-Latin keymap and automatically does the us, variant and alt_shift_toggle business
[12:06] <niktaris> lets see