[12:31] <SEJeff> OT: http://xkcd.com/ <-- pretty good ubuntu joke
[01:23] <infinity> Anyone feel like writing a self-hosting toolchain that builds in a tenth of the time that glibc/gcc does?  kthx.
[01:23] <_ion> Ok, will do.
[01:23] <zul> sure...ill get right on it
[01:23] <infinity> Excellent.  Champions, the lot of you.
[01:23] <infinity> I look forward to your work.
[01:24] <jorgp> lol
[01:25] <Mez> infinity, I would do, but i've got work
[01:25] <plugwash> out of interest why exactly is this bootstrap needed?
[01:25] <plugwash> why can't new versions of gcc etc just filter in from debian like everything else?
[01:26] <infinity> We use far more cutting edge versoins of glibc than Debian does.
[01:26] <infinity> And we need to boostrap the toolchain to make sure it all gets along Just So.
[01:26] <lifeless> infinity: s/cutting/slicing and dicing/
[01:27] <infinity> Add to that the ABI breakage that we just landed in GCC (go us), and we have all sorts of fun.
[01:27] <lifeless> what did we do ?
[01:27] <ajmitch> ABI breakage is only ppc/sparc?
[01:27] <infinity> ajmitch: Yeah.
[01:27] <infinity> lifeless: Backported the ldbl-128 changes from 4.2 (better to eat the breakage now than later)
[01:28] <infinity> Here's hoping I get it all done before I need to be on a plane in 24 hours.
[01:29] <ajmitch> shoddy & expensive
[01:29] <infinity> ajmitch: What's connectivity like at the Auckland airport?
[01:29] <ajmitch> how long do you have to wait in auckland?
[01:29] <infinity> Ahh.  Expensive.  Awesome.
[01:29] <ajmitch> probably poor, I haven't tried recently
[01:29] <infinity> I have a couple/few hours, nothing drastic.
[01:30] <infinity> Customs?
[01:30] <infinity> From NZ to NZ, they don't generally care about what you're  importing. :)
[01:30] <plugwash> any particular reason for using such cutting edge versions? does it improve performance or something?
[01:31] <infinity> plugwash: Gives us some shiny features, and it can be handy to track upstream more closely.
[01:31] <ajmitch> well I'm coming from a domestic flight
[01:31] <ajmitch> there's some minimal checking on the way out, iirc
[01:31] <infinity> plugwash: It also gives us warm fuzzy feelings to be able to mention it to our users (who, for the most part, don't care, but think they should)
[01:31] <ajmitch> 1 hour should be plenty
[01:32] <lifeless> ajmitch: domestic->international in nz will mean no checking on the way into aukland, but regular emmigration on the way to the international flight
[01:32] <ajmitch> yeah, that's what I expect
[01:32] <infinity> I wonder if I'm in Auckland long enough to clear security and come back...
[01:33] <infinity> Not that there's any valid reason to do so, but then I'd have an NZ stamp in my passport. :)
[01:33] <mjg59> infinity: You've been in .au how long and you don't have an NZ stamp?
[01:33] <mjg59> Hell, *I've* got one
[01:33] <infinity> mjg59: Three years.  I know, it's sad.
[01:33] <infinity> Zofia and I kept meaning to vacation in .nz, but never got around to it.
[01:35] <mjg59> It's a nice place
[01:35] <infinity> So I've heard.
[01:35] <infinity> Lovely scenery and friendly poeple with funny accents.
[01:35] <infinity> Sounds good.
[03:33] <dabaR> Why does the gnome panel not ask me for my password when I run an administrative applet from the panel in edgy?
[03:34] <dabaR> Do you think it is a bug?
[03:34] <dabaR> A security bug!
[03:35] <dabaR> Anyhow, I thought you guys might want to be the first ones to know about it...
[03:35] <grndslm> soo...lvm is included in dapper, but not in edgy?
[03:35] <lifeless> its in edgy
[03:35] <grndslm> umm....not mine
[03:35] <grndslm> i tried pvscan, and no go
[03:36] <infinity> dabaR: The first one you opened probably triggered gksudo, and then your credentials were cached for a while.
[03:36] <lifeless> what do you mean
[03:36] <dabaR> infinity: nope...
[03:36] <HrdwrBoB> grndslm: restart lvm
[03:36] <lifeless> (I'm typing from a machine using lvm/evms)
[03:36] <HrdwrBoB> /etc/init.d/lvm restart
[03:36] <dabaR> I just opened the panel. And there is another person on #ubuntu talking about the same thing, I thought it was cause I use openbox.
[03:36] <grndslm> i had to install lvm2!! ...and still can't use pvscan, but maybe the restart will work
[03:36] <grndslm> brb
[03:37] <Fujitsu> Isn't lvm2 depended on by some low-lying metapackage!?
[03:37] <dabaR> actually, I blamed it on being high last night, but I remembered it cause this guy in #ubuntu is asking about it.
[03:37] <infinity> dabaR: Well, it's pretty much impossible for it to be running as root without some sort of su/sudo.  So it's either running as you (and effectively useless), or you have cached credentials.
[03:37] <grndslm> ok...did need a restart...but i thought it shoulda been installed by default
[03:37] <infinity> grndslm: We don't install it by default anymore, no.
[03:37] <lifeless> Fujitsu: the lvm core is, but not the lvm userspace
[03:37] <dabaR> infinity: look, I use Ubuntu since warty...
[03:37] <grndslm> ok...just checkin, thanks
[03:37] <infinity> grndslm: The general theory being that most people don't want or need LVM, and it just serves to confuse.
[03:37] <dabaR> infinity: it's not a joke...
[03:38] <grndslm> makes sense to me
[03:38] <lifeless> infinity: is the lvm2 userspace needed for lvm booting to work ?
[03:38] <Fujitsu> lifeless: Aha.
[03:38] <infinity> lifeless: Yes, it's where the initramfs hooks come from.
[03:38] <infinity> I think.
[03:38] <lifeless> so that his next reboot will work
[03:39] <dabaR> infinity: I just added a user to make sure.
[03:39] <infinity> (base)adconrad@cthulhu:~$ dpkg -S /usr/share/initramfs-tools/hooks/lvm 
[03:39] <infinity> lvm-common: /usr/share/initramfs-tools/hooks/lvm
[03:39] <infinity> However, that hook script won't do much of anything useful without lvm2 installed as well.
[03:39] <dabaR> infinity: and succesfully logged in as him
[03:39] <lifeless> righto
[03:40] <lifeless> so, I trusted that it would just work, given that the lvm2 packages were auto-removed during upgrade.
[03:40] <infinity> lifeless: If you have a working initramfs, you don't need the userspace tools, of course, but the very next time your initramfs is generated without the userspace installed, you lose. :)
[03:40] <lifeless> boy am I glad I haven't rebooted recently
[03:41] <infinity> You might want to "update-initramfs -u" after reinstalling lvm2, if you're paranoid. :)
[03:41] <dabaR> So you guys are sure it was not a change in pam settings or something like that?
[03:41] <dabaR> Or sudoers, or something...
[03:42] <lifeless> infinity: doing so ;)
[03:43] <dabaR> Im in #ubuntu often...
[03:43] <lifeless> however, as its a xen box, its more a case of 'sudo mkinitramfs -o /boot/xen0-linux-2.6.17-6-generic-xen0.initrd.img 2.6.17-6-generic-xen0'
[03:43] <infinity> lifeless: Hrm.  lvm2 being auto-removed due to the minimal->lvm2 dependency causing it to be "auto-installed" is a bit problematic, isn't it?
[03:43] <infinity> lifeless: You're the first person who's mentioned that...
[03:43] <lifeless> infinity: I would have thought so
[03:43] <lifeless> but like I say, I trust you guys not to hose my system :)
[03:44] <infinity> I don't. :P
[03:44] <Fujitsu> Another way to hose your system is to create an LVM snapshot immediately after installing.
[03:45] <Fujitsu> It sort of explodes, but then works after a reboot or two.
[03:45] <lifeless> I found one interesting thing
[03:47] <lifeless> which is that theres some geometry/translation thing
[03:48] <lifeless> makes my pv[2]  fail until I go into evms and out again
[03:48] <lifeless> very bizarre
[03:48] <lifeless> infinity: I'm guessing the devmapper uninstall is also problematic ?
[03:49] <lifeless> infinity: sorry, dmsetup
[03:50] <infinity> Don't know if anything you use would use dmsetup, but if so, "yes"?
[03:50] <lifeless> well, evms uses devmapper, dunno if it needs the dmsetup tools or not.
[03:51] <infinity> I see no calls to dmsetup in the evms initramfs scripts.
[03:52] <lifeless> easiest way is for me to do up a xen guest with evms, uninstall that shit and see what happens
[04:02] <infinity> Was this box a dapper->edgy upgrade, or an edgy->edgy upgrade?
[04:03] <infinity> Cause I didn't think we marked auto-installed packages in dapper (so the lvm2 autoremoval shouldn't happen in that case), now that I think about it.
[04:07] <lifeless> dapper->edgy
[04:07] <lifeless> its my home stable server
[04:08] <infinity> Hrm.
[04:08] <infinity> Fun.
[04:08] <lifeless> it started life a long long long time ago
[04:08] <grndslm> uh...anybody here got a treo 650 that they sync with ubuntu?
[05:00] <grndslm> soo...nobody here owns a palm treo, eh?
[05:00] <grndslm> i find that hard to believe
[05:01] <tritium> grndslm: support in #ubuntu please
[05:01] <grndslm> nobody there can support me
[05:02] <grndslm> i've tried to sync my 650 with ubuntu for the past 3 releases and it never works, no matter how many howtos i read
[05:43] <grndslm> why would i get a fsck.ext3 error at boot that says "unable to resolve UUID: 23fe9a9b02d-239....yada yada"?  even tho it says it's clean??
[05:52] <grndslm> ok....found why the fsck.ext3 error about UUID stops my boot process....but could you guys explain why the UUIDs are in my fstab to begin with?
[05:55] <Fujitsu> grndslm: Because Ubuntu uses UUIDs for all filesystems it can in Edgy.
[05:57] <grndslm> Fujitsu:  so...UUIDs should be in fstab?
[05:57] <Fujitsu> Correct.
[05:57] <grndslm> hmm....interesting...well, i swapped out my second hard drive, so that's why i'm getting the error
[05:57] <grndslm> and i think i'm just gonna do a fresh re-install
[05:58] <grndslm> but while i'm gone...think about this:  to rephrase my suggestion from last night regarding moving the volume applet to the top-right corner....it would be the BETTER TRADEOFF, no?  if you don't agree, i don't think that you're practically thinking about the average desktop user and their usage of their desktop!   :D
[05:59] <grndslm> i feel like i'm gonna have to nag on that one issue 'til i'm dead...
[05:59] <grndslm> i'm out
[06:43] <StevenK> mpt: Are you able to push your glade changes for AboutWindow? Also, I'm unsure if I should just edit the spec's wiki page, since I've fixed the window closing bug, drag-n-drop now works, and Hardware Details now runs hal-device-manager. Does this make me a contributor? :-)
[06:59] <dholbach> good morning
[07:08] <mpt> StevenK, awesome
[07:08] <mpt> StevenK, I still get "Not a branch" from your branch
[07:09] <StevenK> Um. Wierd.
[07:10] <StevenK> Who's the bazaar.l.n person we can throw this too? :-)
[07:10] <mpt> ddaa
[07:14] <mpt> who should be around in five hours or so
[07:15] <StevenK> mpt: Hrm. It just worked for me on my Edgy box here.
[07:15] <StevenK> bzr pull http://..../~stevenk/about-window/dev gives 3 revision(s) pulled
[07:15] <mpt> Can you do "bzr log sftp://bazaar.launchpad.net/~stevenk/about-window/dev/" ?
[07:16] <mpt> oh, http works
[07:16] <StevenK> Neat.
[07:17] <StevenK> mpt: Rev 4 has been pushed, but hasn't hit http:// yet.
[07:17] <mpt> hmm
[07:17] <mpt> I think drag-and-drop should work without allowing selection
[07:17] <mpt> on the grounds that you never (or hardly ever) just want *part* of the version number
[07:17] <mpt> What do you think?
[07:18] <StevenK> Sounds reasonable, but how does one implement that? :-)
[07:18] <mpt> No idea :-)
[07:18] <StevenK> mpt: And should I just bend the wiki page to my will and remove some the bugs listed?
[07:18] <mpt> sure
[07:19] <StevenK> Kay. I'll do that when I get home in about an hour.
[07:19] <mpt> thank you
[07:20] <StevenK> Drag-n-drop without selecting text .... that's a nice tricky problem.
[07:21] <dholbach> we're past 70000 bugs!
[07:21] <dholbach> woah!
[07:31] <tarzeau> who knows about the cd mastering (glass) and production of ubuntu?
[07:37] <nixternal> is jdizzle the planet hacker around at all?  that is jdub ;)
[07:49] <mpt> StevenK, pushed the glade changes
[07:50] <mpt> tarzeau, nobody in this channel I think
[07:51] <mpt> tarzeau, actually, marilize might know :-)
[08:01] <pitti> Good morning
[08:01] <dholbach> heya pitti
[08:02] <ajmitch> morning pitti 
[08:08] <ajmitch> hm?
[08:08] <pitti> on -changes
[08:08] <ajmitch> which bugs?
[08:08] <pitti> ooh, they are from October 23
[08:08] <ajmitch> old mail, perhaps
[08:09] <pitti> why did I get them just now?
[08:09] <ajmitch> I got some mail come through, I assumed it was my ISP being annoying
[09:06] <tarzeau> marilize: ping
[09:06] <tarzeau> mpt: thank yyou
[09:11] <marilize> hi tarzeau
[09:11] <jdub> hi marilize!
[09:11] <marilize> jdub! hellooo
[09:22] <bronson> Is there any way to build a package from CVS directories?
[09:22] <bronson> Seems to work fine except that the CVS dirs end up in the source package.
[09:23] <bronson> Then lintian complains...
[09:23] <bronson> Is there any way to keep my files in CVS and yet ensure that the CVS dirs don't end up in the source package?
[09:24] <pitti> bronson: use dpkg-buildpackage's -i switch
[09:24] <pitti> bronson: (or debuild -i, whichever you use)
[09:25] <bronson> pitti: that makes sense.  Thanks!
[09:25] <Hobbsee> /aaway
[09:25] <StevenK> Heh
[09:25] <Hobbsee> ouch
[09:26] <pitti> speling iz hadr
[09:37] <pitti> hi lloydinho 
[09:38] <lloydinho> hi pitti!
[09:38] <lloydinho> Getting ready to go to west?
[09:39] <pitti> lloydinho: yeah, looking forward to it
[09:39] <tarzeau> marilize: do you know anything about cd glass mastering and production? real life example, w/ ubuntu?
[09:39] <tarzeau> marilize: like prices, business contact?
[09:39] <tarzeau> country, shipping etc
[09:40] <tarzeau> i have made GNUSTEPU that i'd like to press, version 1.0
[09:40] <lloydinho> pitti: I'm coming too, y'know.. so I'll see you on Sunday evening.. :-)
[09:40] <pitti> oh, nice!
[09:56] <tfheen> lloydinho: travels to .dk was fine?
[09:56] <lloydinho> tfheen: Yes, thank you. Only slight delight due to the ice on the ground.
[09:57] <lloydinho> s/delight/delay
[09:57] <lloydinho> weird typo..
[10:02] <pitti> ooh, new libc is in the archive
[10:02] <pitti> infinity: ^ does that mean the bootstrap is complete or is there still something missing?
[10:02] <infinity> I'm dangerously close to unfreezing feisty.
[10:02] <infinity> pitti: ldbl128 rebuilds happening now, THEN I thaw it.
[10:03] <infinity> Hoping to finish up before I have to catch my flight in the morning. :)
[10:03] <infinity> (should make it)
[10:03] <pitti> infinity: sounds like 'long double 128' -> is that a particular lib or so?
[10:04] <infinity> Slight gcc ABI change on powerpc and sparc, doing mass rebuilds of everything affected.
[10:04] <hunger> which gcc version will feisty be build with?
[10:04] <infinity> Or, rather, everything *possibly* affected, since measuring actual breakage looked a lot like effort.
[10:05] <infinity> hunger: 4.1... Ish.
[10:05] <hunger> infinity: So same as edgy?
[10:05] <infinity> Note the "ish".
[10:06] <infinity> But, other than the above-mentioned ABI break, it's pretty much the same as edgy, yes.
[10:06] <infinity> Shiny new binutils and glibc, though.
[10:06] <hunger> infinity: ish is something like post-but-pre-the-next-version?
[10:06] <hunger> infinity: with all those linker speedups?
[10:07] <Burgundavia> infinity plans on freezing feisty until feisty+1 opens, less buildd work :)
[10:07] <infinity> Not sure about "all", but there should be some bling in glibc 2.5, yes.
[10:07] <hunger> nice:-)
[10:07] <Hobbsee> Burgundavia: haha 
[10:07] <infinity> Burgundavia: More work when it's frozen, trust me.
[10:07] <infinity> Burgundavia: My job is 10-fold worse when the archive is frozen.
[10:08] <Burgundavia> hunger: the urge is getting stronger
[10:09] <hunger> Burgundavia: it definitly is.
[10:09] <Hobbsee> hunger: not long.  i'm thinking of adding a feisty partition tonight
[10:09] <infinity> It's nothing but a new toolchain so far.
[10:09] <hunger> maybe I should install xen and have a virtual feisty box instead.
[10:09] <infinity> Well, and a kernel, if you're painfully brave/stupid.
[10:10] <pitti> 2.6.19 runs happily on my iBook
[10:10] <pitti> I installed it to play with the new 'support pipes in core_pattern' stuff
[10:10] <hunger> infinity: Yes, I am painfully stupid when it comes to updating my linux distri:-(
[10:10] <fabbione> pitti: freezes on mind once in a while
[10:33] <whadar> is it possible for casper to persist over an NFS mount? i want home persistence over the network... i know this might be problematic cause nfs rw on unionfs doesnt play well
[12:28] <cbx33> If I wanted to identify a file, as to whether it was audio/document/movie/image is file the best way to go about it....I don;t want to rely on file extensions
[12:28] <Mez> mdz / cjwatson ping
[01:05] <mirak> hi
[01:06] <fernando> morning all
[01:06] <mirak> fglrx-kernel-source seems broken, or the name doesn't match the name that module assistant knows
[01:12] <kagou> how can i use evolution-dbg package ? Because i have some crash with evolution and i know that evolution-dbg is better to report crash logs
[01:19] <mirak> I filled a bug report
[03:32] <bddebian> Howdy
[03:33] <jonh_wendell> pitti: do you confirm lang pack updates for edgy-updates?
[03:33] <pitti> jonh_wendell: I prepare them and ask for testing, yes
[03:34] <jonh_wendell> pitti: when (do you think) will they available to people?
[03:35] <pitti> jonh_wendell: you can use the dailies for testing (feedback appreciated)
[03:35] <pitti> jonh_wendell: we just released edgy, so I won't update them again now
[03:36] <jonh_wendell> pitti: yes, i know that. i'm asking because last time we chatted you talked something about 1 month after edgy release. is it correct?
[03:37] <pitti> jonh_wendell: yes, start of December, when we'll also update the dapper packs
[03:38] <jonh_wendell> pitti: can you put here the line for edgy's sources.list?
[03:38] <pitti> deb http://people.ubuntu.com/~pitti/langpacks/daily/edgy-updates/ ./
[03:39] <jonh_wendell> pitti: thanks!
[03:43] <BenC> does "Breaks" work?
[03:44] <BenC> The kernel for feisty "breaks" the edgy wireless-tools package, and I don't want to add a conflict to the old version if Breaks will work
[03:48] <Keybuk> BenC: how do you feel about adding Depends: udev, initramfs-tools, etc. to the kernel?
[03:49] <BenC> Keybuk: I don't feel bad about it at all :)
[03:57] <BenC> Keybuk: Anything besides initramfs-tools and udev that are actual depends?
[03:57] <Keybuk> upstart?
[03:58] <zMott> where would I report a major bug in ubuntu networking?
[03:58] <Keybuk> Linux does kinda depend on /sbin/init
[03:59] <BenC> Keybuk: upstart isn't essentil?
[03:59] <BenC> zMott: launchpad.net
[04:00] <zMott> what area
[04:00] <zMott> this is not crashing the system
[04:00] <zMott> but its stalling out the network
[04:01] <BenC> Keybuk: upstart is Priority: required...do I really need to depend on it?
[04:01] <Keybuk> BenC: APT doesn't pay attention to priorities, no?
[04:01] <Keybuk> it's not like dselect, which actually installs everything <= standard unless you explicitly tell it not to
[04:02] <BenC> I'm just trying to think of a case where we may not want to depend on upstart
[04:02] <Seveas> people who don't like upstart?
[04:03] <Seveas> (then again, those people probably use custom kernels)
[04:03] <BenC> Keybuk: BTW, upstart is ignoring this line in my /etc/inittab
[04:03] <BenC> X:23:respawn:/bin/su root /usr/bin/startx
[04:03] <BenC> Keybuk: initctl list doesn't even show it
[04:04] <Keybuk> upstart doesn't parse /etc/inittab
[04:04] <BenC> it's a mame machine where I start the UI via inittab
[04:04] <BenC> Keybuk: Isn't that part of compatibility?
[04:04] <Keybuk> cat > /etc/event.d/startx << EOF
[04:04] <Keybuk> start on runlevel-2
[04:04] <Keybuk> start on runlevel-3
[04:04] <Keybuk> stop on runlevel-4
[04:04] <Keybuk> stop on runlevel-5
[04:04] <Keybuk> stop on shutdown
[04:04] <Keybuk> respawn /bin/su root /usr/bin/startx
[04:04] <Keybuk> EOF
[04:04] <Keybuk> no
[04:05] <Keybuk> upstart is not backwards compatible with sysvinit
[04:05] <Keybuk> it was never intended to be
[04:05] <jonh_wendell> iwj: can you take a look at bug 68663?
[04:05] <Ubugtu> Malone bug 68663 in firefox "[Edgy]  Incompatible with Google Toolbar." [Undecided,Confirmed]  http://launchpad.net/bugs/68663
[04:05] <Keybuk> at least, not from the system point of view
[04:05] <Keybuk> e.g. it doesn't parse /etc/inittab, it doesn't listen on /dev/initctl
[04:05] <BenC> how do I get it to reread the events?
[04:05] <Keybuk> it watches the directory with inotify
[04:06] <cbx33> is inotify easy to interface with in python
[04:06] <Keybuk> cbx33: I imagine someone has written a module by now
[04:06] <BenC> Keybuk: Thanks, it's started now
[04:07] <Keybuk> BenC: when you install upstart, it looks for common things in /etc/inittab and converts them to /etc/event.d files
[04:07] <Keybuk> but inittab is so utterly free form, it's very hard to do arbitrary things
[04:08] <BenC> Keybuk: Maybe mv inittab out of the way (/etc/inittab.upstart-disabled) and leave a file there explaining all this
[04:08] <Keybuk> BenC: too late now :)
[04:08] <Keybuk> edgy is released
[04:08] <BenC> what?!?! when did this happen!?! :)
[04:09] <Treenaks> BenC: you didn't get the memo?
[04:23] <dholbach> doko_: a shame you don't get karma for uploads yet
[04:25] <doko_> dholbach: nah, these are auto-generated
[04:25] <dholbach> :-)
[04:25] <seb128> dholbach: you say that because you beat doko number of uploads during the edgy cycle? :p
[04:26] <dholbach> wait for the "feisty results"! ROAR!
[04:26] <doko_> seb128: that will become difficult for feisty ...
[04:26] <doko_> seb128: do you have numbers?
[04:26] <seb128> doko_: 444 for dholbach
[04:26] <seb128> doko_: 441 for you
[04:26] <dholbach> muhuhuahhahahaha
[04:27] <seb128> :)
[04:27] <dholbach> seb128: how many for you?
[04:27] <seb128> dholbach: not the topic :p
[04:27] <dholbach> or as his friends call him  dput-sb
[04:28] <dholbach> i'm out - see you guys in a bit
[04:28] <seb128> dholbach: and I'm still scoring for edgy, I've uploaded a couple of other edgy updates today :p
[04:28] <dholbach> hehe
[04:28] <seb128> dholbach: drive safe :)
[04:28] <dholbach> just picking up the car 
[04:28] <dholbach> brb
[04:28] <seb128> k
[04:28] <seb128> see you soon then
[04:42] <BenC> # kopt=root=UUID=880e2998-3bf0-4e8c-a437-89fac4b411d2 ro
[04:42] <BenC> # kopt_2_6=root=/dev/sda1 ro
[04:43] <BenC> anyone know why new installs still have that second line in grub's menu.lst?
[04:43] <BenC> it's sort of pointless to have this neat UUID stuff that's supposed to make pata upgrades to feisty work when the UUID isn't being used :)
[04:44] <doko_> pitti: ping
[04:48] <pitti> hi doko_ 
[04:49] <lastnode> pitti, got a sec?
[04:49] <pitti> lastnode: hello; what's up?
[04:51] <lastnode> pitti, regarding your conversation with Martin Meredith on http://www.upstreamdev.org 
[04:51] <lastnode> i answered a lot of his concerns
[04:52] <pitti> great
[04:54] <_ion> I think the "pci, video, network, apt" checkboxes should be checked by default, and the user should be told to uncheck some of them *only* if she knows what she's doing.
[04:55] <lastnode> pitti, http://upstreamdev.org/wiki/Releases/0.1.0 , break a .deb or two if you like ;-)
[04:55] <lastnode> _ion, those are usability issues, read http://upstreamdev.org/wiki/Releases/0.1.0#Release_Notes, some of them are addressed there
[04:56] <lastnode> this is an early alpha, so we still have only basic functionality. no polish at all :-)
[05:20] <sivang> hi all
[05:20] <sivang> is fesity open already?
[05:21] <pitti> sivang: not yet
[05:21] <sivang> pitti: ah, okay, I just saw tons of uploads to feisty-changes
[05:21] <sivang> pitti: of various stuff in main
[05:21] <EvanCarroll> I've got a really odd problem and I'm going to file a bug report on it, but I can't figure out enough information to make it really helpful, so I'm looking for ideas
[05:21] <pitti> sivang: just doko's rebuilds, I assume?
[05:22] <EvanCarroll> Yesterday night I rebuilt my /etc/network/interfaces script and today i notice something very strange, when i boot up, if it fails to get an ip address the whole system runs slow, almost like its borking acpi and throttling by 90%
[05:23] <_ion> You could start by showing your /etc/network/interfaces.
[05:23] <EvanCarroll> I can't figure out any more information, but the bottom for gnome for isntance, is about 3minutes if i run `ifup ath0`, where as if i shut the interface down its like 5 seconds
[05:23] <_ion> Also, use ifup -v ath0
[05:24] <EvanCarroll> let me get that interfaces, up, and the out put for -v didn't even know there was a v
[05:24] <EvanCarroll> nothing is being recorded to syslog
[05:25] <EvanCarroll> it should also be noted i'm using madwifi
[05:28] <EvanCarroll> hrm
[05:29] <EvanCarroll> this is the most screwed up thing I've ever seen.
[05:30] <thom> i'd suggest that something is trying to resolve a hostname
[05:31] <EvanCarroll> Yes, I think your right.
[05:31] <EvanCarroll> to some extent.
[05:31] <EvanCarroll> I mean aparently someone has it set to resync clcok with ntp.ubuntu.com
[05:31] <EvanCarroll> the script using static address so it makes it that far.
[05:31] <EvanCarroll> but the static address corresponds to a non-existant network
[05:31] <EvanCarroll> because I'm at work
[05:32] <EvanCarroll> and then I'm kind of lost, becuase it would see as if the termination of ifup indicates that it has timed out, and yet the performance lag doesn't recede. And the kicker is, that my gkrellium graphs don't show abnormal utilization, just that everything runs slow
[05:32] <EvanCarroll> as if it was throttled
[05:33] <EvanCarroll> so I think the bug lies somewhere in the assumption that every ifup on a static address will work, which in the case of wireless amongst others is surely not true
[05:33] <pitti> $ ls uploads/feisty/*.changes|wc -l
[05:33] <pitti> 25
[05:33] <pitti> *nnnng* must ... upload .. crack ... *nnng*
[05:35] <EvanCarroll> hrm
[05:36] <sivang> pitti: hmm, I have some problem with my .procmailrc , you were right :)
[05:38] <EvanCarroll> i'm attaching the strace, and ltrace, I'm just going to post it on launchpad
[05:45] <EvanCarroll> hrm
[05:45] <EvanCarroll> what should i file the bug in, I know it will just get reassigned anyway
[05:45] <EvanCarroll> ntp vs interfaces?
[06:01] <EvanCarroll> I filed it under netbase
[06:01] <EvanCarroll> bug num 70108
[06:02] <EvanCarroll> strace/ltrace/ifup -v/and the actual interfaces file are provided
[06:38] <jonh_wendell> who is the firefox mantainer? pitti?
[06:38] <pitti> jonh_wendell: iwj
[06:38] <jonh_wendell> pitti: thanks
[06:38] <pitti> I did some security updates, though
[06:38] <jonh_wendell> iwj: ping
[07:11] <cjwatson> jonh_wendell: always helps to say what you want, otherwise iwj will see your ping in two or three days and have to ask you then, rather than just knowing up-front
[07:12] <jonh_wendell> :)
[07:15] <tfheen> it also means people don't ping, then get a pong ten hours later and don't remember what they pung about.
[07:15] <Spads> that's why people ping instead of asking a question
[07:15] <Spads> if you don't answer, it was easier to just do something else instead
[07:16] <Chipzz> tfheen: aha
[07:17] <tfheen> Spads: bah, silly people.  There's nothing wrong with long-winded conversations which takes a few days to be completed because the answers are eight hours apart. :-)
[07:17] <Spads> haha
[07:18] <Spads> tfheen: no, but sometimes it's not worth it
[07:18] <Chipzz> tfheen: or freenode kicking them off the server because of a 'DDoS'
[07:18] <Spads> but yeah, contentless pings remind me sometimes of people who enter a technical channel and say "can I ask a question?" and then exit in a huff when nobody says "yes"
[07:19] <tfheen> Spads: if it's not worth it, then you shouldn't ping either. :-)
[07:19] <tfheen> Chipzz: oh, that too.
[07:19] <tfheen> Chipzz: wonderful IRC network we're on.
[07:19] <Chipzz> tfheen: anyway, what I meant
[07:20] <Chipzz> for illustration purposes I moved /usr/lib/pkgconfig/sqlite3.pc out of the way and ran configure for some program:
[07:20] <Chipzz> checking for SQLITE3... configure: error: Package requirements (sqlite3 >= 3.0) were not met:
[07:20] <Chipzz> No package 'sqlite3' found
[07:20] <Chipzz> the last line is from pkg-config
[07:20] <Chipzz> line 484
[07:21] <tfheen> yes, it is.
[07:21] <jonh_wendell> cjwatson: i've typed something to iwj early, but got no response so far. Can i subscribe him to a firefox bug?
[07:23] <Chipzz> what I was suggesting is changing that to "No package 'sqlite3' found"
[07:23] <Chipzz> crap
[07:24] <Chipzz> what I was suggesting is changing that to "No package 'sqlite3' found". Check http://www.ubuntu.com/pkgconfig?search=sqlite3" for how to install that package"
[07:28] <cjwatson> jonh_wendell: that hardly seems to be up to me ...
[07:29] <cjwatson> jonh_wendell: note that there's about to be a developer summit and so many of us are travelling; in fact, I should be packing now
[07:29] <cjwatson> jonh_wendell: if you actually want an answer to something rather than chatting, use e-mail rather than IRC
[07:30] <jonh_wendell> cjwatson: ok, i'll subscribe him so that he'll be notified by email. I hope he doesn't get furious with me :)
[07:34] <fdoving> mdz: could you take a look at the new patch added to bug 69583 ? (kopete icq fix, SRU) Thanks in advance.
[07:34] <Ubugtu> Malone bug 69583 in kopete "SRU: kopete can't connect to ICQ. " [Low,Fix committed]  http://launchpad.net/bugs/69583
[09:09] <mirak> does anyone knows how to rebuild a kernel image package from the source package and a specific target ? 
[09:28] <codingmaster> join #launchpad
[10:18] <keescook> versioning question... I'm starting with 4.4.1-9.1build1, but going to 4.4.1-9.1ubuntu0.1 does not appear to evaluate as "newer".  Should this be 4.4.1-9.1ubuntu1 instead?
[10:20] <BenC> keescook: probably
[10:20] <keescook> BenC: okay.  I'm used to doing the .1 revs for security updates.  :)  thanks!
[10:22] <cjwatson> keescook: um
[10:22] <keescook> cjwatson: or should it be 1.1?  hmm
[10:22] <cjwatson> keescook: $ dpkg --compare-versions 4.4.1-9.1build1 lt 4.4.1-9.1ubuntu0.1; echo $?
[10:22] <cjwatson> 0
[10:23] <cjwatson> keescook: ubuntu0.1 is newer
[10:23] <cjwatson> that's just fine, "ubuntu" > "build"
[10:23] <keescook> cjwatson: that's weird... doing an upgrade of this fix for rpm didn't catch that it needed the new version of librpm4...
[10:23] <cjwatson> I think something else must be wrong
[10:24] <cjwatson> the version comparison algorithm alternates digit and non-digit chunks and compares each in turn numerically or lexicographically respectively
[10:24] <keescook> ew, yeah, the deps aren't strong...
[10:24] <keescook> librpm4 (3 4.4.2)
[10:24] <cjwatson> do they need to be? i.e. does ubuntu0.1 introduce new ABI?
[10:25] <keescook> no, I guess not.  librpm4 has the fix.  I was just surprised when it let me install the rpm_*.deb file without the matching librpm4_*.deb
[10:27] <cjwatson> sometimes binaries from the same package have (= ${binary:Version}) or similar deps
[10:27] <cjwatson> but it isn't required
[10:28] <keescook> cjwatson: yeah, I wasn't really thinking about ABI versions, which is what really matters for the Depends.  when this rolls into security, they'll both get installed due to build1 < ubuntu0.1, so no one will run into my think-o.
[11:13] <lifeless> ok, so ubuntu boot too fast :(
[11:14] <lifeless> edgy specifically tries to bring up lvm2 before all the discs are ready
[11:14] <lifeless> worth filing a bug ?
[11:30] <grndslm> are there issues with syncing a treo 650 in edgy??  because it doesn't create a device node, even after adding the visor module & creating the custom udev file telling it to create the dev node at ttyUSB*
[11:31] <Kaleo> hello
[11:41] <grndslm> are there issues with udev creating a device node for palm devices (specifically a treo 650) in edgy??
[11:42] <LaserJock> grndslm: you can check for bug reports on Launchpad
[11:42] <cjwatson> this sort of thing is rarely udev; it's more likely to be e.g. the kernel not exposing the right thing in /sys
[11:43] <cjwatson> I know nothing of the specifics here
[11:44] <grndslm> well, damn...i don't understand how the kernel can create the node in 2.6.12 & 2.6.15, but not 2.6.17...i could never get it to sync before, but now when people say jpilot works flawlessly, I can't even start to use jpilot
[11:46] <cjwatson> grndslm: exactly what device node is not being created that should be?
[11:46] <grndslm> ttyUSB0 or ttyUSB1
[11:47] <grndslm> i figured it was because i didn't load the visor module, so i did that....and it loads properly, according to /var/log/messages....so i tried creating a custom udev file telling it to create the dev node at ttyUSB*, and still nothing
[11:47] <cjwatson> udev doesn't touch those at all
[11:48] <cjwatson> it just passes the kernel device name straight through
[11:48] <grndslm> according to ubuntuguide.org, the udev file has to be created
[11:48] <cjwatson> URL?
[11:49] <grndslm> it's on ubuntuguide.org....here's a quick jump, tho:  http://ubuntuguide.org/wiki/Ubuntu_Edgy#How_to_configure_PalmOS_Devices
[11:49] <leonel> Hello    I'd like to know  what Ubuntu Developers  and leaders think about  the  novell Ms  agreement 
[11:50] <leonel> any link ?
[11:50] <cjwatson> grndslm: looks pretty bogus to me
[11:50] <grndslm> ok...well, there's still no device node either way, with or without that file
[11:50] <bhale> leonel: i imagine mark will blog it soon enough
[11:51] <bhale> leonel: watch planet.ubuntu.com for all kinds of opinions
[11:51] <bhale> dont expect them all to reflect ubuntu as a whole, though
[11:51] <cjwatson> grndslm: unless I'm much mistaken (and I concede I don't know udev as well as some), adding that file will break the rest of udev
[11:51] <bhale> go quoting it out of context etc
[11:51] <cjwatson> grndslm: because it uses = for matching things rather than ==
[11:52] <cjwatson> grndslm: that will cause all udev events to be changed to claim they're on the usb bus, are Palm Handhelds, etc., regardless of device
[11:52] <grndslm> so, you don't have any advice then?  other than to file a bugreport or wait it out?
[11:53] <cjwatson> and NAME{...} isn't documented in the udev man page so is probably wrong too
[11:53] <cjwatson> I'd look in /sys for ttyUSB* nodes
[11:53] <grndslm> ok
[11:53] <grndslm> but /var/log/messages should still say where the node is created, no?  it doesn't
[11:53] <jdong> out of pure curiousity, is it a bad idea (tm) to use the network inside a build script?
[11:54] <tfheen> jdong: yes, it won't work.
[11:54] <cjwatson> grndslm: I don't see why it would
[11:54] <grndslm> hmm....ok...i'll look in /sys
[11:54] <jdong> tfheen: well, it won't work on Ubuntu/Debian build servers :D
[11:54] <cjwatson> grndslm: /var/log/udev would be a better bet
[11:55] <grndslm> ok
[11:55] <jdong> never meant to see the light of official (tm) day
[11:55] <tfheen> jdong: it's a violation of policy and is frowned upon, but if it's just your own package nobody else cares much, I guess.
[11:55] <Burgwork> grndslm: from a member of the doc team, I can say that ubuntuguide is seriously crack and shoudl be avoided
[11:55] <grndslm> Burgwork, will keep that in mind
[11:55] <jdong> tfheen: ok... just trying to conserve some space/time on my end with building certain svn snapshot debs
[11:55] <jdong> the end output is the same regardless of how the source is obtained
[11:56] <jdong> so I guess not too much harm done :D
[11:57] <grndslm> cjwatson, definitely no ttyUSB anywhere in that log
[11:58] <grndslm> it's not gonna be a problem if i restart udev, is it?
[11:58] <leonel> bhale: Ok  thanks
[12:01] <grndslm> is it ok to restart udev?  don't wanna do it if it's gonna make my computer crap out
[12:02] <cjwatson> grndslm: remove that broken rule first
[12:02] <grndslm> did it already
[12:02] <cjwatson> then it should be fine
[12:02] <cjwatson> assuming you haven't applied any other weird crap since last starting udev :)
[12:03] <tfheen> udev watches the rules directory, so restarting it makes less sense
[12:04] <grndslm> ubuntuguide suggesting adding a rules file, which i did, but cj tells me that the suggestion itself made even lesser sense so i should remove it
[12:04] <grndslm> suggested
[12:04] <cjwatson> yeah, there's no particular need to restart it
[12:05] <grndslm> well, you're right 'cause there's still no dev node created
[12:07] <cjwatson> if you're in a hurry, just (a) file a bug and (b) create the device node manually
[12:07] <grndslm> wow...how do i create it myself
[12:07] <cjwatson> (you'd have to do that on every boot)
[12:07] <grndslm> i'm willing to learn
[12:08] <cjwatson> it's not guaranteed that this will work, if the visor module isn't actually working properly; but anyway
[12:08] <cjwatson> sudo mknod /dev/ttyUSB0 c 188 0 && sudo ln -s ttyUSB0 /dev/visor
[12:09] <grndslm> i could change the link to /dev/pilot, right?
[12:09] <cjwatson> character major 188 minor 0 are the proper magic numbers for /dev/ttyUSB0
[12:09] <cjwatson> I misspoke, I meant /dev/pilot
[12:09] <grndslm> cool