[00:03] <ogra> lamont, oh no .... i was looking forward to meet you
[00:04] <lamont> ogra: again???
[00:04] <ogra> heh
[00:05] <ogra> .oO(and i could swear elmo lost a lot of weight together with his hair, but probably is only the beard)
[00:05] <ogra> *it's
[00:05] <infinity> ogra: No, he definitely shrunk.
[00:05] <ogra> heh
[00:06] <ogra> infinity, will you at least be in prague ? or do you skip again ?
[00:07] <infinity> ogra: I'll be there.
[00:07] <ogra> yay
[00:07] <ogra> you wont regret
[00:07] <ogra> prague is the most beautiful city in .eu imho :)
[00:12]  * lamont would love to visit prague someitme
[00:13] <ogra> (beyond wimen and beer they also invented art deco)
[00:13] <ogra> ;)
[00:13] <ffm> cjwatson, in "European working hours", where in Europe?
[00:13] <cjwatson> Europe is not that big in timezone terms
[00:14] <ogra> ffm, that doent really vary much
[00:14] <cjwatson> as a general rule I don't give out ICBM coordinates :-)
[00:14] <slangasek> lamont: you're really missing out on the opportunity to have Patty be your tour guide to all the ice cream shops in Prague
[00:14] <ffm> cjwatson, darn, I wsa planning to wake him up.
[00:14] <ffm> *was
[00:14] <lamont> slangasek: maybe I'll take her with me when I go? :)
[00:14] <slangasek> heh
[00:15] <slangasek> that might be too late, she may have succeeded in retaining a second word in Czech by then
[00:16] <hwilde> you know europeans only work like 25hrs /week
[00:17] <cjwatson> damn, that reminds me, I meant to learn at least a few words of Czech
[00:17] <ffm> hwilde, its insane.
[00:17] <hwilde> the french govt tried to up it to 32 and they rioted
[00:18] <slangasek> cjwatson: "prosím jedno pivo"
[00:18] <ffm> personally, I wouldn't mind professionally programming or working for canonical.
[00:18] <LaserJock> I thought it was 35/week for France
[00:18] <ffm> but then again i'm just a student.
[00:18] <cjwatson> http://www.triplet.com/50-10_employment/50-20_workingtime.asp "The standard French working week is 35 hours"
[00:18] <ogra> hwilde, what do i have to pay you to convince my GF that this is true ?
[00:19] <LaserJock> hah
[00:19] <persia> cjwatson: The most important thing to remember is that "No" is the affirmative response.
[00:19] <slangasek> nebo "prosím jedno plzeňské pivo"
[00:19] <slangasek> persia: and /yes/ means "eat" ;)
[00:20] <cjwatson> I can see how pivo might be a useful word to remember
[00:20] <slangasek> :-)
[00:20] <ogra> heh
[00:20] <persia> /pivo/ /yes/ and /no/ will get one fairly far.
[00:21] <slangasek> well, "jest" -> /yes/ is a command, so it will likely get you funny looks more than anything
[00:21] <persia> Right.  Perhaps I should have eaten more last I was there.
[00:21] <hwilde> ogra, I could keep her occupied for the remainder 15hrs you are working for a very small fee
[00:21]  * persia heads off, thinking a dictionary might be handy
[00:23] <slangasek> persia: coming to UDS this round?
[00:24] <persia> slangasek: such is my intention
[00:24] <ogra> persia, cool
[00:25] <TheMuso> persia: It will be great to meet you in person.
[00:25]  * TheMuso gets intrepid chroots ready./
[00:25] <ogra> TheMuso, wow, you're fast
[00:25] <slangasek> persia: huzzah
[00:25] <TheMuso> ogra: Have you not seen the topic?
[00:26] <norsetto> themuso: are you succesfull?
[00:26] <cjwatson> I'm going to hold the autosync output for a bit anyway, since kees wants to get one last toolchain change in
[00:26] <ogra> TheMuso, if someone added it to the end i would have seen it :P
[00:27] <TheMuso> norsetto: Not sure yet.
[00:27] <ogra> xchat snips off the front of the topic (which is still far better than xchat-gnome though)
[00:27] <pochu> oh, so Intrepid is open! that means I need to start merging/syncing things... :)
[00:27] <cjwatson> ogra: while irssi doesn't show the end unless you have a terminal window the width of London
[00:28] <ogra> lol
[00:28] <TheMuso> cjwatson: Damn right. Even a 1680x1050 resolution doesn't even allow for that. :p
[00:28] <norsetto> themuso: I'm getting a failure because of bsdutils missing
[00:29] <cjwatson> util-linux got uploaded, might be mid-build
[00:29] <cjwatson> or mid-mirror
[00:29] <TheMuso> norsetto: The only thing I can think of is the mirror you are using is not up to date yet.
[00:29] <cjwatson>   bsdutils | 1:2.14~rc2-0ubuntu1 |      intrepid | amd64, hppa, i386, ia64, lpia, powerpc, sparc
[00:29] <cjwatson> so *should* be in place
[00:30] <norsetto> themuso, cjwatson: its http://archive.ubuntu.com
[00:30] <cjwatson> which is a mirror :-)
[00:30] <norsetto> ah ...
[00:30] <ogra> did scott wake up our big mama already ?
[00:30]  * ogra checks
[00:30] <soren> Yes.
[00:30] <ogra> geez
[00:31]  * TheMuso goes to try and reproduce a pulseaudio high CPU load bug thats been reported...
[00:32] <cjwatson> TheMuso: the hardy-proposed kernel upload today claimed to improve pulseaudio load
[00:32] <cjwatson> in case you didn't see it
[00:33] <TheMuso> cjwatson: Yeah I saw it, I want to try and reproduce without that kernel however.
[00:33] <cjwatson>    * Kernel should use CONFIG_FAIR_CGROUP_SCHED. Fixes high load issues
[00:33] <cjwatson>      with pulseaudio.
[00:33] <cjwatson>      - LP: #188226
[00:33] <cjwatson> ok
[00:33] <elmo> I like the sound of that
[00:34] <elmo> my impression of CFS so far has been entirely less than stellar
[00:34] <jdong> I certainly haven't seen it perk anything up in terms of responsiveness
[00:35] <slangasek> the other issue that was linked back to CFS was a total failure of synergy when running as non-root; is that referenced by the upload in hardy-proposed?
[00:35] <elmo> that line is much funnier if you don't know synergy is a package
[00:35] <Nafallo> hehe
[00:36] <ogra> heh
[00:36] <cjwatson> synergy is mentioned in bug 188226
[00:36] <cjwatson> though only in passing, so I don't know for sure if it's the same thing
[00:37] <infinity> elmo: I got the funny reading of that.  Didn't realise synergy was a package until you mentioned it.
[00:39] <crimsun> TheMuso: the bug will remain reproducible until we pull in glitch-free PA, which requires alsa-lib 1.0.16.
[00:39] <TheMuso> crimsun: great, however I have never had high load with pulse here.
[00:39] <TheMuso> At least not that I've noticed.
[00:42] <hwilde> TheMuso, is it stuttering ?
[00:42] <TheMuso> hwilde: let me find you the bug report. I haven't experienced anything personally.
[00:43] <TheMuso> hwilde: bug 190754
[00:43] <hwilde> is it this one ?  https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.22/+bug/131439
[00:43] <hwilde> the stutter! yeah
[00:43] <hwilde> do you have the AMD Geode audio CS5536 ?
[00:43] <TheMuso> hwilde: I am not the one who is experiencing it.
[00:44] <TheMuso> hwilde: I am trying to reproduce it.
[00:44] <hwilde> oh
[00:44] <hwilde> can you find out if they have that hardware ? :)
[00:44] <TheMuso> Well I'm going to mention that a new kernel is in hardy-proposed, and to test that to see if that helps them.
[00:44] <hwilde> ubuntu audio team and alsa-project are tracking bugs on the AMD Geode Companion CS5536
[00:44] <hwilde> same symptoms but it's pure hardware
[00:45] <hwilde> (atleast until someone fixes it)
[00:45] <TheMuso> Most people seem to be having problems with rhythmbox + pulseaudio, which to me sounds a little more like rhythmbox than anything else.
[00:45] <TheMuso> However, I don't use rhythmbox, so I need to run it through its paces to see if I get similar behavior.
[00:45] <hwilde> oh...my hardware stutters regardless of application, driver, file format, etc.
[00:45] <hwilde> it's freakin embarassing
[00:46] <elmo> TheMuso: I use amarok and pulse and have problems
[00:46] <elmo> since hardy
[00:46] <TheMuso> elmo: Amarok uses the xine backend afaik.
[00:46] <elmo> TheMuso: which uses pulse
[00:46] <TheMuso> Yeah I know.
[00:46] <crimsun> elmo: what sort of problems?
[00:46] <TheMuso> I'll install amarok later and try it out.
[00:47] <elmo> crimsun: stutter under load that gutsy was fine with
[00:47] <crimsun> right, linux.
[00:47] <elmo> crimsun: general lag and slowdowns when I'm using amarok
[00:47] <ffm> crimsun, http://www.alice.org/index.php?page=license is non-free, right?
[00:47] <elmo> crimsun: *shrug* gutsy was better is my hand-wavy, non-scientific impression
[00:48] <ffm> crimsun, they have the "this software includes foo" advert requirement
[00:48] <infinity> ffm: Advertising clauses aren't non-free, just annoying and GPL-incompatible.
[00:49] <ffm> infinity, could it be included in, say... universe?
[00:49] <infinity> ffm: The license is free enough for main/universe, yes.
[00:49] <cjwatson> that's just the 4-clause BSD licence
[00:49] <ffm> It is, isn't it...
[00:49]  * infinity nods.
[00:50] <cjwatson> http://www.gnu.org/licenses/license-list.html#OriginalBSD is the FSF's analysis of it
[00:51] <ffm> infinity, is it debian-ok?
[00:51] <ogra> crimsun, do you think its possible to get upstream to split the source of the alsa plugins so we can get the jack-alsa plugin back in without having to do the cut ourselves ?
[00:51] <TheMuso> ogra: There has actually been talk about getting jack back into main.
[00:51] <infinity> ffm: Of course.
[00:51] <ogra> TheMuso, jack back into main ?
[00:52] <ogra> TheMuso, i doubt that will ever happen
[00:52] <ffm> infinity, It can't be included in main, right?
[00:52] <TheMuso> ogra: Yes, it was in main way back when, and was demoted to universe.
[00:52] <ogra> i saw the dscussion though
[00:52] <cjwatson> ffm: it *can* be included in main, as infinity already said.
[00:52] <ogra> jack was surely never in main
[00:52] <cjwatson> ffm: the only significant thing you can't do with it is combine it with GPLed works.
[00:52] <ffm> cjwatson, infinity, because canonical would have to put a notice on all of their ads, wouldn't they?
[00:52] <ogra> the plugins moved to main which was the reason why i disabled buildng of the jack plugin
[00:52] <ffm> or am I misunderstanding the licence?
[00:52] <TheMuso> ogra: Hrm I thought I remember seeing it in main back in the warty/hoary days.
[00:53] <cjwatson> ffm: only if we specifically advertised use of that software
[00:53] <TheMuso> ogra: Yeah I know that.
[00:53] <ffm> cjwatson, Ahhh... like "this disk includes ALICE"
[00:53] <cjwatson> right
[00:53] <cjwatson> we only do that for a very small number of things
[00:53] <infinity> ffm: Yes, if we advertise "Edubuntu, now with ALICE!", then we need to include their license blurb too.
[00:53] <ffm> infinity, Which we are unlikely to do.
[00:54] <ffm> Wait, what _do_ we advertise?
[00:54] <infinity> ffm: Right.  Just as we're unlikely to advertise "Ubuntu, built on the cryptographic hooplah of OpenSSL!" and end up subject to their advertising clause.
[00:54] <ffm> infinity, but doesn't that restrict our freedom?
[00:55]  * ogra wants to meet this alice first before he advertises her with edubuntu
[00:55] <cjwatson> a little, but it's not a practical problem
[00:55] <LaserJock> lol
[00:55] <infinity> ffm: Freedom to use, modify, and distribute, not so much freedom to write silly ads.
[00:55] <infinity> ffm: We pick our fights carefully. :)
[00:55] <cjwatson> it's an inconvenience, not an impediment
[00:55] <ffm> ogra, lol.
[00:56] <infinity> ffm: If we were against licenses that restricted ANY freedom, we couldn't carry GPL software either.
[00:56] <infinity> ffm: We specifically state (both in Ubuntu and Debian) which freedoms we hold dear.
[00:56] <ffm> We'd be... wait, that excluded New BSD, doesn't it?
[00:57] <cjwatson> http://www.ubuntu.com/community/ubuntustory/licensing
[00:57] <cjwatson> that lists the requirements
[00:57] <infinity> ffm: The very act of copyrighting and licensing reduces freedoms, really.  The only "completely free" software would be public domain, but that's not terribly practical in the current world.
[00:58] <infinity> ffm: So, yes, the link cjwatson posted is what we care about.
[00:58] <ffm> infinity, public domain doesn't exist in the USA, or is legally dubious.
[00:58] <ffm> infinity, I think cc-3
[00:58] <infinity> ffm: And for Debian, http://www.debian.org/social_contract#guidelines
[00:58] <ffm> infinity, I think cc-3 is probabbly the most free.
[00:58] <LaserJock> ffm: it does in the US I believe, not in some European countries such as Germany
[00:58] <infinity> ffm: No, any license restricts *some* freedom, even if it's the freedom to redistrubute without the license. :)
[00:59] <infinity> ffm: So, yes.  The point is just to know which freedoms the distributions do care about, and evaluate licenses based on that.
[00:59] <cjwatson> public domain certainly does exist in the US. Any software written by employees of the US Government is public domain.
[01:00] <cjwatson> (which covers a non-trivial amount of interesting free software)
[01:00] <ffm> If I want to get a package in debain + ubuntu, should I read both packaging guides?
[01:00] <LaserJock> you can,  but you don't have to
[01:01] <LaserJock> the packaging guide material in Ubuntu is largely based on Debian packaging material
[01:01] <ffm> "It is commonly believed by non-lawyers that it is impossible to put a work into the public domain. Although copyright law generally does not provide any statutory means to "abandon" copyright so that a work can enter the public domain, this does not mean that it is impossible or even difficult, only that the law is somewhat unclear."
[01:02] <ffm> Ah.
[01:02] <ffm> (Wikipedia, btw)
 Any software written by employees of the US Government is public domain.
[01:09] <BenB> how nice!
[01:09] <BenB> the geo data is also free. I wish they'd mandate that by law in europe, too.
[01:10] <elmo> someone mentioned that at the FSFE/FTF law conference
[01:10] <BenB> (I hope it only applies to paid work time, though)
[01:10] <elmo> in the US, goverments tend to force stuff into the public domain
[01:10] <elmo> in Europe, goverments tend to force stuff to be copyleft (if they do anything at all)
[01:11] <BenB> same should be true for universities. there, situation is vice-versa: US unis are very closed, european ones very open, by culture not law.
[01:11] <elmo> I'm not sure if the latter is 100% accurate, but it's an interestin gobservation
[01:11] <slangasek> well, the federal government in the US forces stuff into the public domain
[01:11] <slangasek> the state governments try to enforce copyright on their own laws
[01:12] <elmo> ah, interesting
[01:13] <BenB> elmo: europe is very heterogen. in germany, it even varies from department to department. the office for foreign affairs is very pro-Free software, others are very closed and MSy
[01:14] <BenB> some use Linux across the board, some MS everywhere.
[01:14] <ogra> well, thats a joschka inheritance i guess
[01:15] <BenB> heh, never made that connection.
[01:15] <TheMuso> Ok, on one machine, my CPU doesn't even break a sweat with rhythmbox/pulseaudio/firefox with no compiz... Time for the notebook test.
[01:15] <ogra> well it happened after he got in
[01:15] <BenB> ogra: the tech guys there are very clued, though. one of them is a debian maintainer.
[01:16] <BenB> ogra: I don't think you can credit Munich with the Greens, though :)
[01:17] <BenB> eh, other way around
[01:17] <ogra> heh, no
[01:17] <ogra> thats ude as lonesome rider rather
[01:17] <Lightkey> Munich happened thanks to Ede from the reds ;p
[01:17] <Lightkey> yeah, Ude
[01:18] <BenB> full name?
[01:18] <Lightkey> the mayor
[01:18] <ogra> tha mayor
[01:18] <Lightkey> haha
[01:18] <Lightkey> 2d1g
[01:18] <BenB> that's not a full name :)
[01:18]  * ogra never can remember the first name
[01:18] <Lightkey> but easily gooogable
[01:19] <ogra> SPD at least :)
[01:19] <BenB> Christian Ude ?
[01:19] <Lightkey> probably
[01:19] <ogra> yeah
[01:19] <BenB> i didn't know the SPD rules Munich?
[01:19] <ogra> right
[01:19] <ogra> it does
[01:19] <Lightkey> it does
[01:19] <ogra> the socialistic island in bavaria :)
[01:19] <Lightkey> the only part in bavaria
[01:19] <ogra> snap
[01:19] <ogra> :)
[01:20] <BenB> yeah, already noticed that Munich is so... not like CSU :)
[01:20] <cjwatson> probably not a surprise for cities to be more left-wing
[01:20] <cjwatson> compare e.g. Austin, Texas :-0
[01:20] <cjwatson> :-)
[01:20] <Lightkey> right
[01:20] <BenB> cjwatson: heh
[01:20] <Lightkey> or the blue/red election map ;p
[01:20] <ffm> What package should I mark as a dependancy of my package if my package needs JML? (bundled with it currently...)
[01:21] <ffm> Is it a bad idea to bundle software that isn't in the repos with my package? (jogl lib, etc)
[01:21] <ogra> BenB, btw i eard a lot of the office for foreign affairs stuff was set up by credativ (rumous though)
[01:21] <ogra> *heard
[01:21] <ogra> *rumours
[01:22] <BenB> I don't know credativ.... as far as I know, they did most themselves.
[01:22] <cjwatson> ffm: generally better to package up dependencies separately if there's any chance that something else might want them
[01:22] <ffm> cjwatson, unlikely, some of these things are esoteric.
[01:22] <BenB> they actually contracted me, too.
[01:22] <ogra> you dont know credativ but are a german debian user ?
[01:22] <BenB> but only to fix some mozilla bugs.
[01:23] <cjwatson> ffm: it may well not be a problem to bundle, then
[01:23] <BenB> ogra: I'm not a debian user.
[01:23] <ffm> cjwatson, I'd be looking at at least 20 subpackages then...
[01:23] <ogra> oh, i thought so
[01:23] <TheMuso> yay, I think I've reproduced this high load with pulse on my notebook. Now to try the new kernel.
[01:23] <cjwatson> though in the case of C libraries you'll need them to be a separate binary package in case of soname transitions
[01:23] <cjwatson> ffm: err, too many negatives. "it may well be OK to bundle, then"
[01:27] <ffm> cjwatson, What if one of them is a package that is in ubuntu, but not of that version?
[01:28] <cjwatson> then it should be upgraded to the proper version (assuming this is against intrepid)
[01:28] <cjwatson> the security team tend to get upset by duplication
[01:28] <ffm> argh....
[01:28] <ffm> This package is java, and its currently annoying.
[01:29] <ffm> I think I'll just try something a bit simpler, like a textbook or something.
[01:31] <ffm> cjwatson, Would it be bad form to package up a MIT-licenced windows app that runs seamlessly under WINE and just have it run under that layer of abstraction?
[01:32] <jdong> ffm: does it build with wine-dev or is it a binary?
[01:32] <ffm> jdong, binary iirc. I'll see if I can get the source...
[01:32] <jdong> ffm: if it's binary then i doubt the archive admins will be very thrilled.
[01:35] <mjg59> ffm: If we can't build it in Ubuntu, then it's multiverse at best
[01:36] <ffm> jdong, if I was able to get the source, and it was MIT/GPL?
[01:36] <jdong> ffm: if it can build on Ubuntu's build servers, then I don't see why the fact that it runs under WINE makes any difference
[01:36] <mjg59> ffm: The general assumption is that everything in main/universe should be buildable from source using an Ubuntu system
[01:37] <mjg59> ffm: So, if you can cross-build it against wine, then that's great - but then you could probably also build it against libwine and end up with a native binary
[01:37] <ffm> WINE isn't in main.
[01:37] <jdong> it's in universe
[01:37] <ffm> And I don't see winelib in the repos.
[01:37] <jdong> if you can build it with wine, it can be in universe too
[01:37]  * ogra scratches head about debootstrap
[01:38] <ogra> W: Failure trying to run: chroot /home/ogra/intrepid mount -t proc proc /proc
[01:38] <ogra> :(
[01:38] <ffm> jdong, Do all main packages have to be build independent of universe packages?
[01:38] <jdong> ffm: correct. main must only build against things in main.
[01:38] <mjg59> ffm: wine-dev
[01:38] <slangasek> why is main even a concern here?  mingw32 sure isn't in main, either
[01:39]  * jdong wonders the same
[01:39] <jdong> ok, Firefox is making me lose my mind.
[01:39] <ffm> slangasek, I'd like to see this package in main some day (like, intrepid+3 or so) along with a host of educational programming utilities I'm packaging for edubuntu.
[01:40] <ogra> ffm, oh ?
[01:40]  * ogra listens up hearing edubuntu
[01:40] <mjg59> ffm: If there's interesting and well-integrated software that falls into that category, then that's fine - wine can be pulled into main as a build-dependency. But we *must* be able to build it under Ubuntu, not just ship the binaries.
[01:40] <ffm> mjg59, I understand.
[01:41] <slangasek> ffm: well, if it's binary-only then that's not going to happen.  If it requires wine as a runtime (either via the wine package or via libwine), then any evaluation of it for main needs to take wine into account anyway.
[01:41] <ffm> slangasek, Again, I understand.
[01:41] <mjg59> ffm: That could either be by building it with mingw32 (and ending up with a native Windows binary) or, preferably, building it against libwine
[01:41]  * ogra found the root cause for his debootstrap breakage :/
[01:41] <ogra> W: http://archive.ubuntu.com/ubuntu/dists/intrepid/main/binary-i386/Packages.bz2 was corrupt
[01:42] <slangasek> ffm: oh, are you still talking about Alice then?
[01:42] <ogra> hmm
[01:42] <slangasek> or something else? Alice at least is available in tarball...
[01:42]  * sladen doesn't see wine as being much different to a python, perl, or jvm interpreter.
[01:43] <slangasek> sladen: we don't have any jvm interpreters in main either right now. :-P
[01:43] <sladen> it's an execution environment that allows execution of programs written is a particular style
[01:43] <jdong> slangasek: does gij not count?
[01:43] <ogra> slangasek, huh ?
[01:43] <slangasek> jdong: no ;)
[01:43] <cjwatson> ogra: I doubt that'll be persistent, it's probably just a transient mirroring problem
[01:43] <jdong> sladen: well true, but we don't allow Java bytecode or .pyc files either as "source"
[01:43] <ogra> cjwatson, yeah, i'll wait, fuse can wait a night as well :)
[01:44] <jdong> slangasek: lol ;-)
[01:44] <jdong> slangasek: if you wait long enough, it kinda acts like a JVM
[01:47] <TheMuso> I've just realized I can't test the new kernel unless I have lum... Which is in binary new...
[01:47] <sladen> jdong: no, but we do have lots of software written in C that we compile in bytecode for distribution
[01:47] <bd_> binary new? there's a binary new as well as a source new?
[01:48] <TheMuso> bd_: Yes.
[01:48] <sladen> jdong: we actually compile it into several types of bytecode, optimised for running on different types of computers
[01:48] <bd_> TheMuso: why? o_O
[01:48] <slangasek> TheMuso: your unfocused IRC comment is important to us.  Please hold for the next available archive admin. ;-)
[01:48] <sladen> jdong: so as long as this program has 100% source and compiles as a .deb like anything else, I don't see why it'd be a problem
[01:48] <bd_> TheMuso: and, if it's in binary new, can't you grab the source and build locally then? :)
[01:49] <TheMuso> slangasek: I know, I was not going to bother any of you, as I have other things to go on with, and I can wait.
[01:49] <cjwatson> slangasek: does that mean you're dealing with it and I can go to bed? :)
[01:49] <slangasek> cjwatson: yes. :)
[01:49] <cjwatson> rock on
[01:49] <ogra> sleep tight
[01:49] <TheMuso> bd_: Yes, but a kernel package does take a little while to build, and I'd need a build environment on amd64 set up, which I don't have yet.
[01:49] <ffm> For a book, what "Type of package: single binary, multiple binary, library, kernel module or cdbs?"
[01:49] <bd_> TheMuso: PPA :D
[01:49] <cjwatson> I do love it when the autosync gets to x and fails
[01:49] <cjwatson> ffm: single binary
[01:50] <cjwatson> binary there means binary package, not binary executable
[01:50] <jdong> sladen: I have no problem with something building against libwine-dev and being packaged, just like something building with gcj and being packaged...
[01:50] <ffm> cjwatson, it isn't to be compiled... (exept from lore sources...)
[01:50] <infinity> cjwatson: You missed the golden opportunity to say "kernel module" there...
[01:50] <cjwatson> ffm: binary packages are not necessarily compiled
[01:50] <jdong> sladen: it just sounded like you were proposing if I build it with visual c++ 7 and tar it up into a binary-wrapper deb it'd be fine
[01:50] <cjwatson> ffm: that's why I drew the distinction I drew just now
[01:50] <infinity> ffm: A "binary package" is any package that's not a "source package"... ie: anything ending in ".deb"
[01:51] <cjwatson> ffm: a binary package is a synonym for a .deb (or a .udeb in the case of the installer)
[01:51] <cjwatson> snap
[01:51] <ogra> ffm, that question is about the finally resulting package
[01:51] <ogra> after you have built it (in some hours o so)
[01:51] <sladen> jdong: no, you need to make it buildable under Ubuntu on a buildd.  Which might mean first ensuring that the builddeps exist and themselves compile  (eg. mingw32)
[01:51] <ogra> *or
[01:51] <jdong> sladen: ok, then we are on the same page :) *whew*
[01:52] <cjwatson> the only particularly noticeable differences between something that's compiled and something that isn't is that you don't run 'make' in debian/rules, and that you would usually make the latter Architecture: all and do stuff in binary-indep rather than binary-arch
[01:52] <ffm> ogra, It's a book, it isn't built, it's written.
[01:52] <cjwatson> ffm: we all know what you mean
[01:52] <cjwatson> ffm: you are missing the jargon terms we're using, though :-)
[01:52] <ogra> ffm, it will still need the *package building process* not compiling :)
[01:52] <cjwatson> a package is built even if there is no code to compile
[01:53] <cjwatson> (and documentation is in fact often built from source forms like TeX or DocBook XML, even if that isn't the case for you)
[01:53] <ogra> and the question was about the resulting package that comes out eventually (not matter how)
[01:54] <ogra> s/not/no/
[01:56] <ogra> ffm, imagine the package build process like compressing a dir into an archive ... its independent of the fact if the files in the dir get compiled or not
[01:56] <slangasek> TheMuso: lum accepted
[01:56] <ogra> s/of/from/
[01:56] <ffm> ogasawara_, kk.
[01:56] <ffm> *ogra
[01:58] <ogra> :)
[01:59] <bd_> nominatedarchindep is not present in legal_archseries <-- got this in a reject notice from a PPA upload to intrepid, is it a bug or am I doing it wrong?
[02:00] <ffm> Should my documentation depend on the packages it uses in its text? (It uses python, for example, as it is a python book)
[02:00] <bd_> it's an arch:all package
[02:00] <bd_> ffm: Suggests:
[02:00] <ffm> Oh, this is so much easier than RPM!
[02:00]  * ffm rejoices.
[02:00]  * slangasek laughs :)
[02:00] <bd_> ffm: the new debhelper 7 thing makes it even easier, I hear :)
[02:01] <ogra> *giggle*
[02:01] <ffm> Where should the suggests line go? before or after build-requires?
[02:01] <bd_> ffm: Suggests: goes somewhere in the binary package stanza
[02:01] <ffm> And what's the difference between suggest and recommend?
[02:02] <bd_> ffm: Basically, Depends: is an absolute nearly-unbreakable (unless you go to extremes) dependency; Recommends: auto-installs but lets you remove it without too much effory, and Suggests: just adds a line to the UI saying "You might want to install these as well" essentially
[02:03] <ffm> bd_, Ok. diveintopython (main) uses recommends, so I'll use that too.
[02:03] <bd_> there's more into in the debian policy document, but the idea is depends are things that are /vital/ for it to work properly, recommends are things that you'll usually want to have installed in all but unusual circumstances, and suggests: just enhances the package in some way but is far from required
[02:03] <ffm> bd_, Well, what's the use of a python manual w/o python?
[02:03] <bd_> selecting what goes into recommends and what goes into suggests does require a bit of a judgement call... :)
[02:04] <slangasek> ffm: you can read a python manual on a machine without a python interpreter
[02:04] <bd_> ffm: You might be using python on a remote machine... but I guess if you have to stretch to imagine such a situation then recommends is right :)
[02:04] <bd_> I suppose I should possibly ask about that message in #launchpad
[02:04] <slangasek> but this fits the definition of "Recommends", which is that in the /usual/ case you would want python installed
[02:04] <ffm> slangasek, in which case you remove python.
[02:05] <slangasek> ffm: er, yes.  You asked "what's the use of it", I was responding to that
[02:06] <ffm> What's "${shlibs:Depends}, ${misc:Depends}" mean?
[02:07] <bd_> ffm: it's part of the automatic shared libraries dependency thing. It's not needed in your case, but doesn't hurt either. I'd leave in misc:Depends as some debhelper scripts can add things to it, and I don't know which offhand :)
[02:08] <bd_> actually I'd just leave in both, it doesn't hurt
[02:08] <ogra> ffm, if you would compils stuff it would automatically add libs the binary depends on
[02:08] <ogra> (for ${shlibs:Depends})
[02:09] <ffm> In "Depends: " and releated stanzas, do you use whitespace or commas to deliminate?
[02:10] <bd_> ffm: commas
[02:17] <ogra> oh, sweet, l-u-m already has the fix for bug 224754
[02:22] <ogra> funny, someone just invited ubuntu-usuers@lists,ubuntu.com to linkedin
[02:24] <BenB> it will have many friends, eh, connections
[02:25] <ogra> hehe
[02:27]  * ogra thinks 3:30 is a good time to close the lid ...
[02:30] <BenB> ogra: night!
[02:30] <ogra> night :)
[02:30] <ogra> get some sleep as well ;)
[02:30] <BenB> heh, yes. thanks :)
[02:30]  * BenB goes
[02:32] <ffm> Hey, I'm having issues signing my package.
[02:33] <ffm> I get "gpg: skipped "ffm__ <ffm@cluenet.org>": secret key not available", but my key has a comment "Package Signature" on the end. How do I spesifiy that in the changelog?
[02:34] <StevenK> ffm__ (Package Signature) <ffm@cluenet.org>
[02:34] <ffm> StevenK, Ok, then.
[02:48] <ffm> Hm...
[02:48] <ffm> for some reason my package doesn't include any actual files...
[02:48] <ffm> *content files
[02:48] <LaserJock> that would be a problem :-)
[02:51] <jdong> LaserJock: well sometimes I wish that to be the case with certain packages....
[02:51] <jdong> LaserJock: like emacs for instance.
[02:51] <LaserJock> boooo
[02:51] <jdong> :P
[02:53] <ffm> LaserJock, Yeah, it's an issue.
[02:53] <ffm> No idea why that's happening, it's odd..
[02:55] <LaserJock> I'd guess you aren't installing files to the right place
[03:30] <ffm> #ubuntu+1 redirects to ubuntu.
[03:30] <ffm> Now that intrepid is open, shouldn-t it be its own channel again?
[03:32] <wgrant> ffm: I think Intrepid is radioactive enough to not be suitable for users.
[03:33]  * bd_ quitely slips intrepid into his sources.list
[03:33] <ffm> wgrant, Darn, I was planning on upgrading my grandmother yesterday. You mean it isn't for production use? Bah!
[03:33]  * bd_ still has battle scars from the debian C++ transition in unstable, mind. Was that before aptitude's resolver was in place? :)
[03:35] <ffm> Where in my debian package do I spesify where I want files to be installed?
[05:52] <YokoZar> Anyone else seriously regretting pulseaudio?
[06:00] <TheMuso> YokoZar: Yes, I am, only because its an LTS, but the decision was made before I got involved at a greater level.
[06:17] <jdong> YokoZar: it's been interesting to see how we can't win em all with regards to Pulse
[06:18] <jdong> YokoZar: when FC8 came out, Ubuntu started getting harsh criticism for not being faster with adopting pulseaudio to address the limitations of ALSA, etc
[06:18] <jdong> yet when we did it finally in Hardy, the criticism instantly reversed
[06:18] <jdong> I think a lot of the pulse headaches I'm having now are due to scheduler issues. the CFS config that ships with hardy is simply awful IMO
[06:26] <LaserJock> jdong: well obviously we're supposed to have a perfect implementation of PA
[06:26] <StevenK> jdong: So the CFS isn't completly fair? :-P
[06:27] <jdong> StevenK: oh it's fair. To OS X and ?
[06:27] <jdong> Windows.
[06:27] <jdong> users' jealousies of Linux's previously-good interactivity
[06:27] <jdong> compiling causes compiz performance to lag now, games are unplayable with music in the background
[06:27] <jdong> *grumble*
[06:37] <pwnguin> someone else was claiming that they were having terrible performance problems as well
[06:37] <pwnguin> i wonder if they'll find debian in the same place
[06:38] <jdong> pwnguin: apparently an upcoming -proposed kernel is to "fix" this.
[06:38] <ScottK> jdong: I got my Intrepid pbuilder set up.
[06:38] <jdong> lol gettin a head start
[06:39] <ScottK> I needed to test the new deboostrap before asking for a backport.
[06:43] <ScottK> jdong: deboostrap is requested now.
[06:45] <TheMuso> jdong: A few of Pulse's problems is that it addresses sound cards directly, and not all sound cards have the same mixer channel names.
[06:45] <TheMuso> Hell even jack has code for specific sound cards, to handle their different quirks.
[06:46] <jdong> TheMuso: I see
[06:46] <jdong> TheMuso: my major problems with pulse have been primarily with responsiveness and concurrent sounds from legacy (i.e. alsa) apps
[06:46]  * TheMuso nods.
[06:46] <jdong> TheMuso: responsiveness being the current major headache
[06:46] <jdong> for some reason Hardy's kernel seems to really suck at interactivity
[06:46] <TheMuso> jdong: crimsun and I are trying to work a solution that sees pulse go through dmix, at least for hardy.
[06:46] <jdong> TheMuso: I look forward to that
[06:47] <jdong> TheMuso: personally I think pulse is a good investment for Ubuntu but I'm not too happy about the way it finalized in Hardy
[06:47] <TheMuso> I msut admit, I turn pulse off, unless I want to send audi oover the network.
[06:47] <TheMuso> must
[06:47] <TheMuso> jdong: As I said earlier, I think I would be fine with it, but for an LTS.
[06:47] <mortal1> hello, I have a question regarding the partman crypto in hardy.  does the current alternate installer disk allow existing partitions to be accessed?
[06:48] <jdong> TheMuso: I'll hold off my criticism for a chance to SRU the major issues with it :)
[06:48] <TheMuso> mortal1: I don't think so.
[06:48] <mortal1> given that debain's todo list for the installer states: partman-crypto: Allow a user to re-use an existing encrypted filesystem without data loss (ex: /home, /srv, etc) [BenjaminSeidenberg]
[06:48] <TheMuso> jdong: As I said, I would rather we not be using it at all.
[06:48] <TheMuso> But we can't turn back now.
[06:49] <mortal1> TheMuso: if they do get that in the next version of the installer, does that mean that the next version of ubuntu will be able to use partitions created in 8.04?
[06:50] <TheMuso> mortal1: Yes I would think so.
[06:52] <mortal1> TheMuso: at any rate, simply doing an upgrade would not break my ability to access those partitions would it?
[06:58] <TheMuso> mortal1: I wouldn't think so, but I don't use encrypted partitions for any length of time to comment.
[07:01] <Mithrandir> mortal1: I have used the same encrypted volumes on my laptop since I created them back in 2005 or so.
[07:14] <mortal1> Mithrandir: really?  How do you go about setting up ubuntu to access them?
[07:15] <mortal1> is there a guide that you use?
[07:16] <mortal1> I'll bbl but i'm interested the process you use.  I'll check back later
[07:19] <Mithrandir> mortal1: I mount them when I log in by using libpam-mount
[07:19] <Mithrandir> search for cryptsetup and libpam-mount and you'll probably find something.
[07:19] <mortal1> thanks
[07:19] <mortal1> good night all
[09:25] <tkamppeter> doko, hi
[09:26] <doko> tkamppeter: good morning
[09:26] <tkamppeter> doko, is pitti on vacation?
[09:27] <doko> tkamppeter: I think so, long weekend
[09:28] <tkamppeter> doko, can you then upload s-c-p for me (see mail).
[09:28] <tkamppeter> (for me getting core-dev seems that only one vote is missing ...)
[09:29] <doko> tkamppeter: will do (but I'm not allowed to vote, afaics)
[09:30] <tkamppeter> doko, thanks, I know that you cannot vote, but perhaps someone else here on IRC.
[09:37] <hunger> What is that devscripts update in hardy? It pulls in 40 new packages at my system. Is that really necessary?
[09:41] <bd_> hunger: it has some bad recommends I expect, same as in debian
[09:41] <hunger> bd_: That is the very first update to a LTS release... a great start:-(
[09:42] <bd_> hunger: it's a backport actually
[09:43] <bd_> +devscripts (2.10.23) unstable; urgency=low
[09:43] <bd_> +
[09:43] <bd_> +  * Move the current Suggests: to Recommends: so that they are pulled in by
[09:43] <bd_> +    default but may be removed if desired (Closes: #474559)
[09:44] <hunger> bd_: Not a good idea to have a update/backport install a smtp daemon... even if it is just the nullmailer.
[09:45] <bd_> hunger: I would suggest filing a bug in intrepid and hardy asking for whatever recommends are causing problems (aptitude why -v devscripts bad-package to see where it's coming from)
[09:45] <bd_> and in the meantime, use aptitude -R install devscripts to skip recommends
[09:45] <wgrant> hunger: It's a backport.
[09:45] <wgrant> Backports aren't allowed to eat your hat, but they can eat disk space if they want.
[09:46] <hunger> wgrant: Installing a SMTP daemon where non was required before is eating my hat.
[09:46] <hunger> wgrant: And update/backport does not make too much of a difference to a end user anyway.
[09:46] <bd_> nullmailer isn't a daemon, it's an implementation of /usr/sbin/sendmail
[09:46] <wgrant> hunger: backports aren't enabled by default.
[09:46] <bd_> there's a difference, sort of
[09:46] <wgrant> And you should be watching what it wants to install.
[09:47] <bd_> but yes, file a bug, mention specific recommendations you'd like removed
[09:47] <bd_> and maybe file it in debian as well, they're not to happy about this either
[09:47] <wgrant> .... or just don't install the recommendations.
[09:47] <bd_> wgrant: or both file a bug, /and/ don't install the recommendations :)
[09:47] <bd_> that way it gets fixed for everyone :)
[09:47] <wgrant> FSVO of fixed.
[09:48] <bd_> downgraded to Suggests: I mean
[09:48] <bd_> I mean, some recommends are sensible - curl | wget - but is www-browser really needed, for example?
[09:49] <cjwatson> each one of those recommends has a justification in the package description, and is because one of the tools in devscripts won't work without it
[09:50] <cjwatson> (or won't work in some particular mode)
[09:50] <cjwatson> there's no need to complain about www-browser, though; w3m provides www-browser and is in ubuntu-standard
[09:51] <hunger> cjwatson: I don't mind the additional libs, but installing a mailer is a bit extravagant for a recommend. Ubuntu keapt priding itself for not needing one a while.
[09:51] <cjwatson> it's not a real mailer; see what you were told above
[09:52] <hunger> cjwatson: Ah, so it is fine to have nullmailer? Please install it by default then and don't make it get dragged in by one of the very first backports.
[09:52] <cjwatson> sigh, stop making a mountain out of a molehill, it's intrepid
[09:52] <cjwatson> (if you're using hardy-backports, that might as well be intrepid ...)
[09:53] <cjwatson> I'll downgrade it to a suggests
[09:54] <hunger> cjwatson: Oh, I had not realized that -backports count as origversion+1.
[09:54] <hunger> I'll better deactivate them then.
[09:54] <cjwatson> but, for the record, any "pride" there may have been (and it really wasn't pride) was in not installing an MTA *in the default installation*. devscripts is not in the default installation
[09:55] <wgrant> And devscripts isn't a normal user package...
[09:55] <cjwatson> actually, I'm not sure this is a reasonable thing to change at all
[09:56] <cjwatson> if you're installing devscripts then something like a trivial MTA should not be a problem
[09:56] <hunger> cjwatson: I'm not sure either. Not having a mailer will break all the tools that rely on it and that tend to be a couple in debian.
[09:56] <cjwatson> it's hardly just Debian, the ability to send mail is a useful facility
[09:56]  * hunger thinks there should be a mailer in the default install, but that does not matter.
[09:57] <cjwatson> there are tools in ubuntu-dev-tools that rely on a mailer
[09:57] <cjwatson> so, on reflection, I think devscripts should stay as it is
[09:57] <bd_> There probably should be a mailer in the default install - but a local-only mailer which delivers mail to root to some user-visible GUI thingy.
[09:57] <bd_> So if some cron job breaks, the user would know about it (is cron in the default install...?)
[09:58] <wgrant> bd_: I'm not sure users care if updatedb fails.
[09:59] <bd_> I suppose you have a point :)
[10:22] <tkamppeter> doko, can you also upload HPLIP 2.8.4 for me (see mail)?
[10:42] <emgent> hello
[11:45]  * ogra scratches his head about http://paste.ubuntu.com/9450/
[11:47] <ogra> since when do packages not restore files anymore if i reinstall them ?
[11:48] <james_w> ogra: they are conffiles?
[11:48] <ogra> well, if i wipe the whole /etc/pulse dir they should re appear, no ? at least that worked in gutsy
[11:49] <james_w> I'm not sure, but I assume that would have something to do with it.  If you purge the package first does it work?
[11:50] <ogra> lets see
[11:50] <ogra> james_w, thanks, that was it, i'm apparently not awake enough yet :P
[11:51] <james_w> it's only lunchtime, you shouldn't be expected to be :-)
[11:51] <ogra> heh
[11:52] <ogra> well, i was up until 5
[11:53] <cjwatson> ogra: conffiles only get restored if you use dpkg --force-confmiss
[11:53] <cjwatson> it's always been like that
[11:53] <ogra> yeah, i didnt think about conffiles at all :)
[12:08] <tkamppeter> doko, I have now also prepared new foo2zjs packages. Can you upload them, too (see your mail)? Thanks.
[12:09] <doko> tkamppeter: I did see these are all intrepid uploads? it's not yet open
[12:11] <tkamppeter> doko, and why does the subject of this IRC say "Archive: Intrepid open, go wild!"?
[12:12] <cjwatson> intrepid is open
[12:12] <tkamppeter> If this is not correct please someone change it to yesterday's state.
[12:12] <cjwatson> I meant to send mail about it yesterday but forgot
[12:33] <davmor2> Help please.  I'm having hardware issues in hardy and I'm trying to track down what is causing it exactly.  My dvd-rw and cd-rom lock up and become unusable.  This also causes elements of the system to lock up also.  I believe I am lowering it down slowly but just need some help to nail it completely.  I believe it to be either a driver issue or the kernel.  I have tried 32bit and 64bit to rule out it being a 64bit is
[12:40] <\sh> cjwatson, any ways to fix debootstrap intrepid for W: Failure trying to run: chroot /tmp/schroot-H11754 mount -t proc proc /proc ?? :)
[12:41] <\sh> cjwatson, means intrepids debootstrap running on hardy? :)
[12:41] <cjwatson> it's a bug, either fix it or wait for somebody else to fix it
[12:41] <cjwatson> effort for supporting people on intrepid is a bit low right now
[12:42] <\sh> cjwatson, it's more "getting build infra ready for working on intrepid" ;)
[12:42] <\sh> cjwatson, working on hardy that is
[12:42] <cjwatson> if you want to help, investigate it
[12:43] <\sh> shermann@hom-emt64-l:~$ sudo  chroot /tmp/schroot-H11754 mount -t proc proc /proc
[12:43] <\sh> chroot: cannot run command `mount': Exec format error
[12:43] <cjwatson> that's not an investigation
[12:43] <cjwatson> debootstrap leaves log files
[12:43] <\sh> cjwatson, I'm on it :)
[12:44] <\sh> cjwatson, but logfile inside the mentioned chroot tells the same error only...joy ;)
[13:06] <ScottK> pitti: If you have a moment, I wanted to discuss copying clamav from dapper-backports to dapper-updates/security again.  jdstrand has endorsed the idea in Bug #217256.  In hardy we had zero problems going from 0.92 to 0.92.1 and there have been no reported problems with the dapper-backport.
[13:07] <ScottK> That'd be one less clamav variant we don't have enough time to figure patches for.
[13:15] <cjwatson> \sh: 'sudo debootstrap intrepid /chroot/intrepid' works for me
[13:15] <\sh> cjwatson, yes...sudo debootstreap hardy hardy_chroot <-- works , sudo debootstrap --arch=amd64 hardy hardy2_chroot <- doesn't work.
[13:16] <\sh> cjwatson, whereas sudo debootstrap --arch=i386 hardy hardy3_chroot works again ;)
[13:16] <\sh> cjwatson, and sudo debootstrap --arch=amd64 intrepid intrepid_chroot doesn't work...(all done on amd64 arch box)
[13:17] <cjwatson> if it's really amd64, --arch=amd64 is the default
[13:17] <cjwatson> I suspect you are running a 32-bit kernel despite the system's hardware capabilities
[13:17] <\sh> cjwatson, nope
[13:17] <cjwatson> well, that's what the message indicates
[13:17] <\sh> fck the hell
[13:17] <cjwatson> or you're in linux32
[13:18] <Mithrandir> \sh: what does uname -a say?
[13:18] <\sh> I installed amd64 hardy on that box
[13:18] <\sh> at least I thought so...
[13:19]  * \sh kills his wife tonight
[13:19] <\sh> if on the cd amd64 is written, but the contents is i386 ... what do I have installed...
[13:20] <\sh> i hate my life i hate my life
[13:20] <geser> a 48-bit Ubuntu? :)
[13:21] <\sh> yes
[13:21] <\sh> she mixed up my media
[13:22]  * \sh grabs his grave...and says I'm sorry for messing up my life ,->
[13:23] <geser> \sh: check now the i386 cd if it contains amd64
[13:23] <\sh> yes
[13:23] <\sh> it sounds like a joke...but it isn't really...
[13:25]  * \sh is offline now...at least dettaching from ircproxy...and reinstalls....
[13:34] <ogra> asac, FF3 caching and wiki editing dont seems to actually like each other, i seem to have to shift-ctrl-R every page to see the edits afterwards
[13:58] <asac_> ogra: cannot really confirm this. for me it worked last time i used the wiki
[13:59] <ogra> asac_, generally if i have the notice at the top after editing it looks ok, as soon as i click on "remove message" i get the old content
[13:59] <ogra> and have to force reload
[13:59] <ogra> which then is ok as well ...
[14:01] <asac_> ogra: never used the wiki that way :)
[14:01] <asac_> ogra: did it ever work?
[14:02] <ogra> huh ?
[14:02] <ogra> i mean the grey box you get at the top after every edit
[14:03] <ogra> s/edit/save/
[14:23] <guja_nebeska> I want to be Ubuntu developer. I am relativly beginner, and I am asking from You to give me some advice and literature to read and learn Ubuntu so I can develop that OS.
[14:23] <guja_nebeska> I am C and C++ programmer, but don't know as much about Ubuntu as programming.
[14:23] <guja_nebeska> So, please, give me some ebook adivces, books, links, anything.
[14:24] <guja_nebeska> I'd be very grateful for any kind of help.
[14:24] <guja_nebeska> Thank You.
[14:24] <tkamppeter> doko, I will replace the foo2zjs packages, forgot to switch from "hardy" to "intrepid" in the changelog.
[14:24] <Amaranth> guja_nebeska: The folks in #ubuntu-motu can put you to work, I'm sure. :)
[14:25] <james_w> guja_nebeska: https://wiki.ubuntu.com/MOTU/GettingStarted is a good place to start
[14:26] <guja_nebeska> james_w, Amaranth thank you for interesting in my question. I can't start fixing bugs or stuff like that, cus I am not familiar too much with system.
[14:26] <guja_nebeska> I need to spend hours and hours of reading about Ubuntu, kernel, distros and stuff to understand what am I fixing.
[14:26] <guja_nebeska> I need for start good literature based on that themes.
[14:27] <tkamppeter> doko, I have replaced the foo2zjs packages now, can you download them again from my site and upload this version? Thanks.
[14:27] <tkamppeter> doko, is syetm-config-printer OK?
[14:28] <guja_nebeska> But I'll ask same question on ubuntu-motu, thanks for the advice.
[15:05] <bimberi> win 21
[15:08] <gnomefreak> bin/win :)
[15:08] <gnomefreak> oops
[15:08] <gnomefreak> bimberi: /win
[15:08] <gnomefreak> ;)
[15:09] <bimberi> heh :)
[15:51]  * Rabiddog slaps the ubuntu devs
[15:51] <cjwatson> thanks, that's nice of you
[15:51] <Rabiddog> cjwatson: they broke software raid again in the final release of hardy
[15:52] <cjwatson> the bug tracker is -> that way ... can you refer to a bug report?
[15:52] <cjwatson> "broke" is awfully general
[15:52] <Rabiddog> https://bugs.launchpad.net/ubuntu/+source/dmraid/+bug/102973
[15:52] <cjwatson> that's not software RAID in general, that's SATA RAID
[15:52] <Rabiddog> jorgen (author of kanotix) provided all the fixes too :)
[15:53] <Rabiddog> cjwatson: worse case scenario is I have build the kernel myself properly with the patches in it I guess. so much for a Long Term Edition for servers
[15:54] <jdong> nobody in their right mind uses dmraid on servers.
[15:54] <cjwatson> *cough* servers don't use SATA RAID as a general rule
[15:54] <Rabiddog> home servers do
[15:54] <jdong> they shouldn't.
[15:54] <Rabiddog> jdong: whats the alternative?
[15:54] <jdong> Rabiddog: linux md raid?
[15:55] <Rabiddog> jdong: I'm just a home server noob, all I want is for it to work
[15:55] <cjwatson> anyway, this is rather beside the point; a simple patch to the dmraid package to use raid456 etc. instead of raid4 etc. would likely be acceptable for 8.04.1
[15:55] <persia> Err.  Nothing wrong with SATA raid, except when not using a proper RAID microcontroller to abstract it from the system.
[15:55] <Rabiddog> cjwatson, ah
[15:55] <cjwatson> the "fixes" suggested are to stick symlinks in /lib/modules/, which isn't so great
[15:55] <Rabiddog> I c
[15:56] <jdong> Rabiddog: yeah I didn't see a proposed patch that's really a fix on the bug
[15:56] <jdong> Rabiddog: though regardless of what I feel about dmraid, this bug can and should still be fixed for Hardy
[15:56] <cjwatson> that said, I don't see where the dmraid package actually modprobes raid4 etc.,
[15:56] <Rabiddog> hmmm
[15:57] <cjwatson> it mentions raid45 as a dm-target id
[15:57] <Rabiddog> I'm using www.kanotix.com/files/gutsy/updates/dmraid/ to try getting it to work
[15:57] <Rabiddog> ponders
[15:58] <Rabiddog> we had to implement a delay during startup also to give it time to be ready
[15:58] <cjwatson> is it just a matter of loading dm-raid4-5 in the initramfs?
[15:59]  * ogra rubs his cheek
[15:59] <ogra> Rabiddog, ouch !
[15:59] <Rabiddog> cjwatson: not sure
[15:59] <Rabiddog> checking something
[16:00] <Rabiddog> dmraid -ay says raid45 not in kernel, dmraid -r shows all 3 raid drives
[16:01] <Rabiddog> cjwatson: how would I make sure what u said happens?
[16:02] <Rabiddog> ogra: o_O
[16:03] <Rabiddog> I'd like to using the ubuntu original kernel tomake upgrading easier
[16:05]  * ogra wonders if Rabiddog would in real life come into a room and slap 200 ppl of which 98% never have seen the package he complains about ....
[16:05] <cjwatson> Rabiddog: add 'modprobe -Q dm-raid4-5' after the other modprobes in /usr/share/initramfs-tools/scripts/local-top/dmraid, run 'sudo update-initramfs -u', reboot
[16:05] <cjwatson> I really am just guessing
[16:08] <Hobbsee> ogra: no.  they'd kill him first, CoC or not.
[16:08] <Hobbsee> ogra: or at least knock him out.
[16:08] <Hobbsee> ogra: so he'd only get to for the first few, before being comatose.
[16:08] <Hobbsee> s/for//
[16:09] <cjwatson> easy
[16:09] <ogra> Hobbsee, come on
[16:09] <Hobbsee> ogra: :)
[16:10] <ogra> Hobbsee, while i think an aplology would be nice (especially since he gets help here) i didnt intend to put up any agression or flaming with that
[16:11] <Hobbsee> ogra: i think i've missed the relevance between your statement and mine.  but it's probably too late at night.  mine was saying what would happen, if he did so in real life.
[16:11] <Hobbsee> (unsure how that's agression or flaming - isn't that simply saying what would happen?)
[16:12] <ScottK> cjwatson: Thanks for the latest devscripts upload.  It's always nice when the response to a bug report is Fix Released.
[16:13] <ogra> in the hypothetical case that someone would do that in RL, probably ...
[16:13] <ogra> Hobbsee, you take my pictures to seriously sometimes
[16:14] <Hobbsee> ogra: i knew it was hypothetical.  i didn't realise it was a hypothetical, not-intended-to-be-responded-to, picture.
[16:15] <ogra> i'll put a tag on it next time :)
[16:22] <Rabiddog> ogra: its was friendly slap of frustration that it got broken :)
[16:24] <ogra> Rabiddog, well lets just drop the topic ... doesnt seem to go anywhere ...
[16:41] <Rabiddog> sorry u misunderstood it :)
[16:42] <ogra> :)
[16:42]  * Rabiddog lightbulb goes on and looks at cjwatson
[16:43] <Rabiddog> I recall seeing something about /usr/share/initramfs-tools/scripts/local-top/dmraid not being just before shell login come sup
[16:43] <Rabiddog> comes up*
[16:43] <cjwatson> not being what?
[16:43] <Rabiddog> wonder if that script is missing
[16:43] <Rabiddog> found
[16:44] <Rabiddog> gimme a sec, I'll see if I can get the exact wording
[16:44] <pitti> Caesar: 225333> sure, I updated the bug; thanks
[16:44] <pitti> tkamppeter: I'm on vac, yesterday and today
[16:44] <cjwatson> that would be odd, it shouldn't be referring to the /usr/share/initramfs-tools/... path, should be just /scripts/local-top/dmraid in the initramfs
[16:46] <pitti> ScottK: clamav> that sounds sensible to me; so the current dapper-backports variant is good/should be copied, or the one you'll upload with that bug?
[16:46] <ScottK> pitti: Current one should be copied.
[16:47] <ScottK> pitti: The current one in dapper-backports already has that bug fixed.
[16:49] <pitti> ScottK: that's just the clamav source package, or some other sources, too?
[16:49] <ScottK> pitti: Just clamav.  No rdepends changes needed on this one.
[16:50] <pitti> ScottK: You have a wealth of experience with clamav, so your word is good enough for me :)
[16:50] <pitti> ScottK: ok, say the word, and I'll copy it :)
[16:50] <pitti> ScottK: do I need to change any bug after doing so? or will you do so?
[16:51] <pitti> ScottK: hmm; backports has 0.92.1~dfsg2-1.1~dapper1, -updates has 0.92~dfsg-2~dapper1ubuntu0.1
[16:51] <ScottK> pitti: Yes.  The ubuntu0.1 changes were all incorporated upstream in 0.92.1
[16:51] <pitti> ScottK: it isn't quite the huge version jump I anticipated? I faintly remember doing that copying some weeks/months ago already?
[16:51] <ScottK> Yes.
[16:52] <ScottK> This is to resync.
[16:52] <pitti> ok, done
[16:52] <ScottK> pitti: We have 0.92.1 in feisty and gutsy backports as of this week.  I want to age those a bit there and then do the same thing (with all the needed rdepends).
[16:52] <ScottK> pitti: Thanks.
[16:53] <pitti> ScottK: shall I remove the version from dapper-backports then? it's in -updates now
[16:53] <ScottK> pitti: Yes.  I think that's sensible.
[16:53] <pitti> ScottK: done
[16:54] <ScottK> pitti: So once we copy feisty/gutsy backports to updates maybe next week, all the supported releases will be on the same version.
[16:54] <pitti> ScottK: please let me know if there are any problems; I'll go offline again in a bit (on holiday today)
[16:54] <Rabiddog> cjwatson: /scripts/local-top/dmraid: /scripts/local-top/: 18: udevsettle not found is the error
[16:54] <pitti> ScottK: cool!
[16:54] <ScottK> pitti: Thanks (particularly on your holiday).
[16:54] <ScottK> pitti: Will do.
[16:54] <ScottK> pitti: Of course clamav 0.93 is sitting in Debian New just now and it breaks everything, so we'll get to do this again in a few months. ;-)
[16:54] <pitti> heh
[16:55] <ScottK> All the libclamav-dev rdpends FTBFS againsta 0.93 and need patching.
[16:55] <ScottK> againsta/against
[17:00] <davmor2> Why would nvidia drivers break a cd-rom and dvd-rw ?
[17:02] <cjwatson> Rabiddog: odd? that's udevadm settle in the current code
[17:02] <cjwatson> dmraid (1.0.0.rc13-2ubuntu6) hardy; urgency=low
[17:02] <cjwatson>   * debian/dmraid.initramfs-local: call udevadm instead of udevsettle
[17:02] <cjwatson>  -- Scott James Remnant <scott@ubuntu.com>  Fri, 14 Dec 2007 16:13:57 +0000
[17:03] <Keybuk> missing update-initramfs ?
[17:04] <cjwatson> Rabiddog: confirm that you ran 'sudo update-initramfs -u' and haven't done anything odd so that you might not be running the current initramfs?
[17:06] <Rabiddog> cjwatson I forgot I installed a older dmraid, so I just updated and am in the process of doing what u said
[17:08] <Rabiddog> cjwatson: ERROR: device-mapper target type "raid45" not in kernel
[17:08] <cjwatson> Rabiddog: having made the change to /usr/share/initramfs-tools/scripts/local-top/dmraid that I suggested, before running update-initramfs?
[17:08] <Rabiddog> yes
[17:08] <cjwatson> ok
[17:09] <Rabiddog> modprobe -Q dm-raid4-5 added
[17:09] <Rabiddog> after the other modprobes
[17:09] <Rabiddog> waits to see if cjwatson has other suggestions :)
[17:10] <cjwatson> I'm on the phone, so bear with me
[17:10] <Rabiddog> np I've got plenty of time
[17:11] <Rabiddog> AFK 5 mins or so
[17:11] <Rabiddog> brb
[17:25] <Rabiddog> back
[17:47] <Rabiddog> ponders
[18:15] <hwilde> anybody claim to be the authority on partition tables?   :)
[18:15] <Rabiddog> no
[18:15] <hwilde> I am getting conflicting output from df -ha,  fdisk -l, and cfdisk
[18:15] <hwilde> http://pastebin.com/m43ca3380
[18:16]  * Rabiddog wonders about cjwatson phone call :)
[18:16] <hwilde> cfdisk says:      sda1        Boot        Primary   Linux ext3       [Ubuntu]         1498.75 (MB)
[18:16] <hwilde> df -ha says:  /dev/sda1             3.9G  561M  3.2G  15% /media/Ubuntu
[18:17] <hwilde> so why in the world would df -ha return the wrong partition size
[18:24] <sistpoty> hm... has anyone seen my upoad of xmms-crossfade from a few hours ago? didn't get an accepted mail :/
[18:25] <ogra> sistpoty, intrepid ?
[18:25] <sistpoty> ogra: yes
[18:25] <ogra> i think there was something up with soyuz, ask in the channel
[18:26] <sistpoty> ogra: ah, #soyuz?
[18:26] <ogra> i think so
[18:26] <ogra> bug 225642 iirc
[18:27] <sistpoty> hm.. #soyuz is empty apart from chanserv... I'll try #launchpad
[18:27] <sistpoty> thanks ogra
[18:31] <StFS> Hello. I'm having some trouble with my firewire web camera. What I need to solve is being able to run viewer apps as a regular user, however, changing the udev permission scripts has no effect and further investigation shows that udev isn't even getting notified when I plug in my firewire camera... can anybody comment on this?
[18:47] <Rabiddog> cjwatson: Are you still on the phone?
[18:48] <Rabiddog> Does someone know how to overcome the following error "ERROR: device-mapper target type "raid45" not in kernel"
[18:51] <james_w> Rabiddog: this is in the installer?
[18:52] <Rabiddog> no its a result of a hdd install upgrade, I had a previous kernl I compiled
[18:52] <Rabiddog> https://bugs.launchpad.net/ubuntu/+source/linux/+bug/220493 <---think i found the reason
[18:52] <Rabiddog> gonna see if I can get the older kernel from the repositories
[18:52] <Rabiddog> 2.6.24-12
[18:54] <Rabiddog> errr
[18:54] <ogra> did you talk to any kernel guys ? did you check the changelog of the linux kernel ?
[18:54] <Rabiddog> ogra..... ubuntu kernel guys?
[18:55] <ogra> s/linux kernel/linux package/
[18:55] <ogra> yes, they have a channel
[18:55] <Rabiddog> do you know the name offhand?
[18:55] <Rabiddog> nm found ity
[19:11] <cjwatson> Rabiddog: the kernel guys were talking about 220493 just earlier today, so I think it may be on their list
[19:11] <cjwatson> Rabiddog: (it turned out that I had to leave immediately after my phone call)
[19:13] <Rabiddog> np talking to stefan bader right now, he might have something for me to test
[19:19] <StFS> sorry for repeating the question but does anyone have any idea how I should change the permission for the firewire system, I'm trying to view the capture from my firewire webcam but I'm only able to do so as root and udev doesn't seem to be getting any events when I plug it in so changing stuff there doesn't seem to help
[19:21] <slytherin> cjwatson: can you please take a look at bug 199116 and approve it for SRU? Ihave discussed it with seb128 and he said it was ok to introduce additional strings and get them translated before 8.04.1.
[19:21] <cjwatson> not today, I'm afraid
[19:22] <slytherin> cjwatson: Ok. No issue. Whenever you get time.
[19:23] <slytherin> StFS: how are you trying to capture images?
[19:23] <crimsun> ogra: Luke mentioned a sample caching bug when using my config changes (in current PulseAudio bzr).  If you have resources, do you mind testing the config on LTSP/Edubuntu 8.04, please?
[19:24] <StFS> slytherin: well that's not really the problem... I have an application that just displays the video from the camera on screen (I've used both coriander and a little example app that comes with the dc1394 library)
[19:25] <StFS> slytherin: but the problem is that both of them only work if I run them as root... if I run them as a regular user I get "no cameras found" or "permission denied"
[19:25] <ogra> crimsun, how urgent is that ? i could probably put some hozrs in on the weekend
[19:25] <ogra> *hours
[19:25] <crimsun> ogra: low to wishlist
[19:26] <ogra> oki, so knowing it by monday should suffice ?
[19:26] <slytherin> StFS: The problme may be the permission for the device node that gets created. I can't provide much help, Idon't have firewire webcam. You should ask on #ubuntu
[19:26] <crimsun> ogra: that would be great!
[19:26] <ogra> will do then :)
[19:26] <crimsun> ogra: awesome, thanks
[19:26] <ogra> thanks for pinging :)
[19:26] <StFS> slytherin: so basically I'm trying to figure out what it is exactly that creates the /dev/video1394 node so I can make that something create the node using different permissions
[19:26] <jdong> StFS: udev creates those
[19:27] <jdong> specifically /etc/udev/rules.d
[19:27] <StFS> jdong: ok... so it must do so then during bootup I'm guessing... because udev isn't getting any event notification when I plug in the camera
[19:27] <jdong> StFS: udev doesn't know about you plugging in the camera? :-/
[19:28] <StFS> jdong: nope... udevmonitor doesn't display anything when I plug in the camera
[19:28] <StFS> jdong: all I get are two lines in syslog
[19:28] <StFS> jdong: http://pastebin.com/daf1208a
[19:29] <StFS> jdong: and I've tried this on two different computers
[19:29] <slytherin> StFS: is there any device /dev/video0 (or video1)?
[19:30] <crimsun> also, udevsettle(8).
[19:30] <StFS> slytherin: just /dev/video1394/0
[19:30] <crimsun> what're the permissions on /dev/video1394/0?
[19:30] <StFS> slytherin: but unplugging the camera does not delete the /0
[19:31] <StFS> crimsun: crw-rw-rw- 1 root video
[19:31] <crimsun> StFS: err, a fresh boot?
[19:31] <StFS> crimsun: that's my next attempt
[19:32] <StFS> crimsun: so you suggest changing the /etc/udev/rules.d/40-permissions.rules file first and adding a MODE="0666" to the line with the video1394* stuff
[19:32] <StFS> ?
[19:32] <crimsun> StFS: no, I'm trying to see what you get at a fresh plug on a fresh boot
[19:33] <StFS> crimsun: ok... so leave the permission file, unplug the camera and reboot right?
[19:34] <crimsun> meaning, if it works after you chmod it, then it's worth chasing the MODE [if it's in fact correct - my reading of 40-permissions.rules leads me to believe you could need raw access, but I don't know your hardware]
[19:34] <crimsun> right, and it's probably best that we do this in #ubuntu
[19:35] <StFS> ok... will do...
[19:35] <StFS> thanks for everyones help
[19:49] <Keybuk> err
[19:50] <Keybuk> what's that program that let's you monitor HAL ?
[19:50] <Keybuk> ah, lshal
[19:50]  * Keybuk was looking for halinfo for some reason
[19:51] <Rabiddog> cjwatson: fyi fixed ERROR: device-mapper target type "raid45" not in kernel still occurs but https://launchpad.net/~stefan-bader-canonical/+archive has the fix in the package update to linux-ubuntu modules, just add those sources and upgrade
[19:52] <Rabiddog> Despite that error dmsetup tables and targets shows raid45 and the array loads up fine and mounts
[19:52] <Rabiddog> now to fix my video setting issue :)
[19:54] <Keybuk> AHA!!!!!!
[19:54] <Rabiddog> o_O?
[19:54]  * Rabiddog watches the lightbulb shine
[19:54] <Keybuk> Having purged everything Till ever added to make printing work ... my printer now works again
[19:55] <jcastro> Anyone seen xivulon? He's due for an openweek session in 5 minutes
[19:55] <Rabiddog> Keybuk: thats another issue, gutsy worked intermittantly with my printer I'm hoping hardy has a better time
[19:56] <james_w> jcastro: he apparently quit IRC just over an hour ago.
[19:56] <jcastro> hmph
[20:03] <jcastro> james_w: whoops, I meant an hour and 5 minutes
[20:05] <Caesar> pitti: please let me know if you need anything further for #225333
[20:24] <cjwatson> Rabiddog: great, glad to hear
[20:44] <Rabiddog> ty
[20:52] <jdstrand> cjwatson: hi!
[20:52] <jdstrand> cjwatson: so I was just looking at kirkland's ubuntu-virt package. lintian and some advice you gave are at odds, and I want to be sure to give the right advice in the future
[20:53] <jdstrand> E: ubuntu-virt source: debian-rules-missing-required-target binary-indep
[20:53] <cjwatson> jdstrand: oh, I didn't say to remove the target entirely
[20:53] <cjwatson> it should be there, but do nothing
[20:53] <kirkland> cjwatson: ahh....  sorry :-)
[20:53] <jdstrand> cjwatson: ok, that is what I thought
[20:53]  * kirkland took "remove binary-indep contents if you aren't using it" wrong....  emphasis on __contents__  ;-)
[20:54] <jdstrand> cjwatson: thanks
[20:58] <PeterFA> Who's in charge of the Canadian Repositories servers?
[20:59] <PeterFA> I would like to report a problem.
[20:59] <cjwatson> ca.archive.ubuntu.com is an alias for mirror.csclub.uwaterloo.ca.
[20:59] <cjwatson> looks like the UWaterloo computer club
[20:59] <cjwatson> PeterFA: what's the problem, though?
[21:00] <PeterFA> cjwatson, well, I'm not really sure, but I was using the repositories and was running into problem installing, downloading, and fetching headers. These problems instantly stopped when I switched to the Main Servers.
[21:00] <PeterFA> Since, I've had absolutely no problem.
[21:10] <james_w> PeterFA: perhaps #ubuntu-mirrors will have some more information.
[21:20] <lamont> hrm.. why is it that the gusty version of jpilot can import a large address list on hardy, but the hardy version can;t, and the sid version cant?
[21:20] <lamont> one more thing for my debugging pile, I guess
[21:26] <oliver_1> hello
[21:27] <oliver_1> is apport not enabled on the hardy live cd?
[21:29] <cjwatson> oliver_1: we tend to turn it off for release to avoid being flooded
[21:30] <cjwatson> and enable it during development cycles
[21:30] <cjwatson> it's not a perfect scheme by any means, but that's what we've settled on
[21:30] <oliver_1> cjwatson: sounds reasonable :-)
[21:31] <oliver_1> is there a way to enable it manually in a LiveCD session? I promise to check LP before submitting a bug...
[21:31]  * Rabiddog searche launch pad for a ui windows bug where when u mouse over the top bar with the minimize and maximize and close window button it dissapears with just a outline of its borders
[21:34] <_MMA_> Rabiddog: While using Compiz?
[21:38] <cjwatson> oliver_1: edit /etc/default/apport as root, set enabled=1, 'sudo /etc/init.d/apport start'
[21:39] <oliver_1> cjwatson: cool, thanks!
[21:39] <infinity> Wow, would you look at that, intrepid builds base livefs images.
[21:39] <cjwatson> some kind of miracle
[21:40] <infinity> cjwatson: BTW, I've moved lpia/intrepid-live to concordia, though it's still using the ubuntu-lpia project name for now.
[21:40] <infinity> cjwatson: Will require some mangling of livecd-rootfs to make it DTRT with just "ubuntu", which I might do nowish.
[21:40] <cjwatson> infinity: thanks, I'll adjust cdimage
[21:41] <cjwatson> infinity: is hardy still on terranova?
[21:41] <infinity> cjwatson: In fact, yeah, I will do that now and upload, so make antimony believe that concordia is a "normal" livefs builder for lpia, and I'll make that assumption true. :)
[21:41] <infinity> cjwatson: hardy will remain on terranova as ubuntu-lpia, should we need to manually trigger hardy builds at some point.
[21:41] <infinity> cjwatson: Though, since we don't release lpia images, I can't see why we'd need to rebuild them.
[21:42] <infinity> cjwatson: I can move hardy, if that's problematic.
[21:42] <cjwatson> I believe that the mobile project has this little release thing coming up ...
[21:42] <cjwatson> infinity: oh, hang on, is the mobile-fs stuff staying where it is?
[21:42] <infinity> cjwatson: Oh, are they doing live images too?  Not just moblin builds?
[21:42] <cjwatson> I have no idea :)
[21:43] <cjwatson> leaving hardy where it is is fine, and probably preferable
[21:43] <infinity> cjwatson: Yeah, I will move the moblin stuff, but not today.  I'm just talking about livefs for now, which I believe they only use for in-house testing.
[21:43] <Rabiddog> _MMA_, yes
[21:43] <cjwatson> probably right
[21:43] <_MMA_> ﻿Rabiddog: It's been reported. Was fixed last I saw. Was a theme issue.
[21:44] <Rabiddog> ah ok
[21:44] <Rabiddog> I'll stop searchingthen
[21:47] <cjwatson> infinity: all done
[21:47] <infinity> cjwatson: Hrm, looks like no one moved Sparc to ports in livecd-rootfs either...
[21:47] <cjwatson> whoops
[21:47]  * infinity fixes.
[21:49] <cjwatson> ta
[21:54] <infinity> cjwatson: Oh, one thing I noticed on antimony...
[21:54] <infinity> cjwatson: antimony seems to assume that flavour=generic for lpia, when it should be lpia (linux-lpia).
[21:54] <cjwatson> mm, I think that maybe predates the lpia flavour
[21:55] <infinity> cjwatson: If that results in unbootable media, I assume that answers the "does anyone actually use this?" question.
[21:55] <cjwatson> originally, the idea was to be able to boot lpia live CDs on normal i386 hardware to see if it blew up
[21:55] <infinity> Yeah, I suspect that goes way back to the livecd-rootfs with the biarch apt hack.
[21:55] <cjwatson> do lpia kernels boot on normal i386 hardware?
[21:56] <infinity> Maaaaybe.
[21:56] <infinity> I wouldn't bank on it.
[21:56] <cjwatson> I think that might be a case of "file a bug, subscribe StevenK, see if he cares"
[21:56] <cjwatson> I'm not sure I want to change it unilaterally
[21:56] <infinity> But lpia doesn't have -generic x86 kernels anyway, so it's moot.
[21:56] <infinity> linux-generic | 2.6.24.16.18 | intrepid/restricted | amd64, i386
[21:56] <infinity> linux-lpia | 2.6.24.16.18 | intrepid/restricted | lpia
[22:00] <cjwatson> err. does it build successfully at the moment?
[22:00] <cjwatson> I guess we don't autobuild lpia live CDs so who knows
[22:01] <cjwatson> ok, you've convinced me, I'll switch
[22:03] <cjwatson> done
[22:12] <persia> Hardy LPIA kernels are booting OK on i386 hardware and qemu at least.
[22:16] <cjwatson> persia: ok, cool
[22:19] <infinity> cjwatson: You want to have a quick glance at my livecd-rootfs changes in bzr before I upload?
[22:22] <cjwatson> infinity: I haven't gone through all the ARCH users, but that looks OK, except that you don't assign LIST if TARGETARCH == i386
[22:22] <cjwatson>         i386)
[22:22] <cjwatson> -           case $FS in
[22:22] <cjwatson> -               ubuntu-lpia) LIST="$LIST linux-lpia";;
[22:22] <cjwatson> -               *)      LIST="$LIST linux-generic";;
[22:22] <cjwatson> -           esac;;
[22:23] <cjwatson> that bit
[22:23] <infinity> cjwatson: Err, oops. ;)
[22:23] <cjwatson> otherwise I think it's correct
[22:23] <infinity> cjwatson: Exhuberant leaning on the d key.
[22:23] <cjwatson> oh, not quite
[22:23] <cjwatson> you need to adjust the getopts arg as well
[22:23] <cjwatson> add a:
[22:24] <infinity> cjwatson: \o/
[22:25] <infinity> cjwatson: Commited those.
[22:26] <cjwatson> looks good
[22:26] <infinity> cjwatson: The arch/targetarch thing is so I can drop this in place on all the buildds and terranova will still happily build ubuntu-lpia as it did before, but overriding ARCH=lpia on concordia will get a full set of lpia builds.
[22:26] <cjwatson> ah, yeah
[22:27] <infinity> (Cause we all know how much the community has been clamouring for kubuntu/lpia)
[22:27] <infinity> *crickets*
[22:27]  * cjwatson grins
[22:30] <infinity> cjwatson: Uploaded.  Shall I just accept it myself? :0
[22:30] <cjwatson> intrepid isn't frozen, so it should auto-accept
[22:30] <infinity> Oh, that's "new".
[22:31] <infinity> I need a script that checks suite states for me and tells me when they change.  Or something equally nerdy.
[22:32]  * Rabiddog curses , cant get my windows box to connect to my smb share on my linux box
[22:32] <cjwatson> reminds me, I should mail ubuntu-devel-announce
[22:33]  * Rabiddog wonders if something changed with samba since gutsy
[22:33] <infinity> Hrm, there's one bug in there, but I always mangle BuildLiveCD by hand anyway, so I'll just commit and not upload a fresh one.
[22:33] <infinity> (Can't pass "-a $ARCH" to livecd.sh if suite << intrepid)
[22:38] <Rabiddog> hey cjwatson got a sec?
[22:40] <Rabiddog> Do you know by chance why am I getting the following error "sudo: unable to resolve host House" on my shell prompts?
[22:40] <cjwatson> Rabiddog: bug 32906, I imagine
[22:41] <Rabiddog> cjwatson, how do u find those things so quickly
[22:41] <cjwatson> Rabiddog: please test the fix in hardy-proposed
[22:41] <cjwatson> Rabiddog: because I knew about this one in advance :-)
[22:41] <Rabiddog> lol
[22:41] <cjwatson> it's one of the major known issues in hardy, that's on the list for 8.04.1
[22:42] <cjwatson> Rabiddog: (if you test the proposed update, please report success/failure in the bug report)
[22:42] <Rabiddog> smb wasn't sure if his fixed could be in 8.04.1 btw
[22:42] <Rabiddog> k
[22:42] <Rabiddog> for the dmraid issue
[22:43] <Rabiddog> err
[22:43] <Rabiddog> syanptic is refusing to start
[22:43] <Rabiddog> brb
[22:46] <infinity> cjwatson: A quick boo again at livecd-rootfs?
[22:47] <Rabiddog> cjwatson, I'm confused I have prereleased-updates  (hardy-proposed) checkmarked, is this hardy propose your talking about cjwatson
[22:47] <cjwatson> yes, that's what I'm talking about
[22:47] <infinity> cjwatson: Nevermind the unterminated cases, I'm half asleep.
[22:48] <cjwatson> infinity: looks fine aside from that
[22:49] <Rabiddog> hmm no update is showing up for sudo
[22:49] <cjwatson> it's version 1.6.9p10-1ubuntu3.1
[22:50] <cjwatson> but perhaps #ubuntu or #ubuntu-bugs or something for further discussion of that, please
[22:51] <Rabiddog> k
[23:01] <ffm> So, there's no way I can get my package into universe if I submit it psedunonymously? I have to have someone else sign it?
[23:02] <LaserJock> ffm: huh?
[23:02] <ffm> LaserJock, I have a package I want to get into universe (I packaged it).
[23:02] <Amaranth> He/She doesn't want to use a real name but does want to do packaging
[23:02] <LaserJock> fine
[23:03] <ffm> Amaranth, He.
[23:03] <ffm> LaserJock, Supposudly it won't be accepted unless my key is in the SWOT, and no one would sign a pseudononymous key.
[23:03] <LaserJock> there's nothing stopping you
[23:03] <LaserJock> no, that's not true
[23:04] <LaserJock> you probably can't become a developer
[23:04] <LaserJock> but you can certainly get your package uploaded
[23:04] <Rabiddog> cjwatson, fixed simple by correctly setting hostname and hosts file correctly :)
[23:05] <ScottK> ffm: No one has to sign your key.
[23:05] <LaserJock> ffm: a signed key is generally required to become a MOTU or Core Developer
[23:05] <ScottK> LaserJock: No.  It's not.
[23:05] <ffm> LaserJock, Why wouldn't I be able to become a developer?
[23:05] <LaserJock> ScottK: well ... that's up for debate
[23:05] <ffm> LaserJock, In any case, what about this...
[23:05] <ScottK> I key uploaded to LP is required, but no one needs to sign it.
[23:05] <LaserJock> ScottK: to become a MOTU?
[23:05] <ScottK> LaserJock: I was already a MOTU before anyone signed my key.
[23:06] <LaserJock> well, *I* was told that it was required and I know that's been enforced in the past
[23:06] <ScottK> The only reason I've bothered at all is because I decided to enter Debian NM.
[23:06] <cjwatson> I think pseudonymous development would be at best controversial, and the developer would have to be especially good, I suspect
[23:06] <slangasek> it's harder to get ICBM coordinates on abusers if they're also pseudonymous, and that's always a nice insurance policy to have
[23:07] <ffm> Email my key to a canonical employee, sign that key with my real name key (which is SWOT'd). The canonical employee then signes my pseudonymous key and publishes his signature. Thus, I am tieable to a real person.
[23:07] <ScottK> I don't see where we have any policies in place to require an actual real name.
[23:07] <LaserJock> ffm: if you just want to have your package uploaded it's no problem.
[23:07] <ScottK> I used my real name on my key, but no one ever verified it.
[23:07] <cjwatson> any Canonical employee who signs a key after only e-mail contact is likely to receive a stern talking-to about key management
[23:07] <slangasek> :)
[23:07] <LaserJock> ScottK: it should have been
[23:08] <LaserJock> when I had my key signed I had to show legal ID
[23:08] <ffm> LaserJock, I'd like to be a MOTU one day.
[23:08] <LaserJock> ffm: then you may face some issue, *may*
[23:08] <ScottK> LaserJock: When I had my key signed I had to show ID, but that was after MOTU.
[23:08] <cjwatson> I'd have no problem signing a pseudonymous key if I'd met the person and had some way of tying their real-world identity to their pseudonym
[23:08] <LaserJock> ScottK: but that doesn't mean it's right :-)
[23:09] <ScottK> LaserJock: If we want to require a signed key, then we need to have that in our process.
[23:09] <ffm> cjwatson, thus the signature.
[23:09] <ScottK> We don't now.
[23:09] <LaserJock> I've brought this up before and it didn't seem like it went anywhere
[23:09] <cjwatson> ffm: you're not getting it just by e-mail though
[23:09] <LaserJock> ScottK: we *did*, somewhere along the lines it got silently dropped
[23:09] <ffm> cjwatson, Key A (real name) is in the WOT.
[23:09] <ffm> cjwatson, Key A signs key B (ffm), and emails signature to Canonical Employee A.
[23:10] <LaserJock> ffm: the point is nobody is going to sign a key over email
[23:10] <LaserJock> or shouldn't in any case
[23:10] <cjwatson> ffm: that's (IMO) a dangerous kind of transitive trust, and I'm not going to be socially-engineered into it, sorry
[23:10] <cjwatson> if you want a signature, you have to meet in person
[23:11] <ScottK> ffm: The other option is to come back with a pseudonym that looks like a real name and never mention it.
[23:11] <ffm> cjwatson, What about if I got a well connected Ubuntu member who I know IRL to sign my key?
[23:11] <LaserJock> ScottK: it shouldn't matter
[23:11] <ffm> or would that cause issues with privacy...
[23:11] <cjwatson> ffm: nobody should *ever* sign a key based on what somebody else has done. It should only ever be from personal experience.
[23:11] <LaserJock> ID should be checked so you'd know if ti was a pseudonym or not presumably
[23:12] <crimsun> ffm: just bring the forms of ID that I requested next time ;-)
[23:12] <ScottK> LaserJock: Maybe it should, but that's not part of our current process.
[23:12] <LaserJock> ScottK: umm, yes it is
[23:12] <cjwatson> ffm: there's no reason why the link from your pseudonymous key to the WOT has to be through a Canonical employee, BTW; you don't need to get hung up on that
[23:12] <ScottK> LaserJock: Where?
[23:12] <ScottK> Maybe that's what I'm missing.
[23:12] <ffm> cjwatson, Yeah, I know.
[23:12] <cjwatson> somebody else in the WOT who's willing to certify it would be fine
[23:12] <ffm> cjwatson, However, I trust them more than J. Random User.
[23:13] <cjwatson> but I still think that when giving root access to our systems to somebody that at least the development team ought to have some idea of whom they're dealing with
[23:13] <infinity> ffm: You clearly haven't met all of us, then.
[23:13] <ffm> crimsun, PM?
[23:13] <ffm> infinity, clearly.
[23:14] <LaserJock> ScottK: well, there's not a technical "how to sign a gpg key" standard that I know of, but all the documentation I've ever seen says that you must know the identity of the person who owns the key your'e signing
[23:14] <LaserJock> and hence you need to check legal ID
[23:14] <cjwatson> LaserJock: ScottK's talking about MOTU-joining process not key-signing process, as I read it
[23:14] <ffm> LaserJock, I don't think I can get a govt ID that says FFM.
[23:14] <ScottK> LaserJock: I agree with that.  I'm just saying no where in the "here's what you have to do to become a MOTU" process does it say get your key signed.
[23:14] <ScottK> cjwatson: Yes.
[23:14] <LaserJock> ScottK: oh, that was before it was ever written down
[23:15] <infinity> LaserJock: Note that there are cases where "know the identity" and "check legal ID" don't need to match up, it's just a good baseline.  I, for instance, wouldn't insist on seeing my mother's passport to verify her identity to me.
[23:15] <cjwatson> ffm: somebody has to have shared enough information with you in your ffm identity that they can confirm upon meeting you in real life that this is the same person who identifies as ffm
[23:15] <LaserJock> infinity: very true, but for random people I feel like it's appropriate
[23:15] <ScottK> LaserJock: OK.  There may have been such a requirement in the distant past.  Currently there is not.
[23:15] <crimsun> ffm: sure
[23:15] <LaserJock> ScottK: currently no, hence why I said sometime in the past it was silently dropped
[23:15] <ffm> cjwatson, I'm pretty sure I know people like that.
[23:16] <crimsun> ffm: also, I've met you, so anonymity is out the window.
[23:16] <ScottK> So getting his key signed is not required.
[23:16] <LaserJock> ScottK: or at least I believe it was silent. I don't remember a TB ruling on it
[23:16] <ScottK> Before my time.
[23:16] <LaserJock> ScottK: well ... I'm not sure
[23:16] <ffm> crimsun, omg, you just ruined my anonymity... :(
[23:16] <Nafallo> baah. easier to just change your name to your nick :-)
[23:16] <ffm> sabdfl, _the_ sabdfl ?
[23:16] <ScottK> LaserJock: Let's turn it around: Should we eject any MOTU who's key isn't in the WOT?
[23:16] <LaserJock> I think either way the TB ought to say something either way
[23:17] <crimsun> ffm: you're online.  Ain't much anonymous there.
[23:17] <ffm> crimsun, pretty pseudonymous.
[23:17] <emgent> i go to sleep
[23:18] <emgent> night people :)
[23:18] <ScottK> Good night emgent.
[23:18] <LaserJock> ScottK: I guess I wouldn't, but I would want them to get their key signed as soon as possible
[23:18] <persia> There's currently a bit of a mess with the MOTU WOT.  Likely good to do some graph analysis and try to get it to have fewer fragments before seeking to establish/reestablish a policy.
[23:19] <ScottK> So far I have exactly one signature on my key (some Debian guy who goes by vorlon on Debian IRC).  I haven't checked, but I doubt I'm alone.
[23:19] <LaserJock> I personally don't have much of a problem with somebody who doesn't have their key signed by an Ubuntu person
[23:19] <infinity> ffm: No, it's not _the_ sabdfl, it's just one of many.  He's a fictional character, and we take turns pretending to be him on IRC.
[23:19] <LaserJock> but I do feel it should be signed to *somebody*
[23:19] <ffm> infinity, lo.
[23:19] <jcastro> ScottK: I hear that guy is shady. :)
[23:19] <ffm> *lol.
[23:20] <ffm> ScottK, Only one? I have 3, and I'm just in secondary school.
[23:20] <LaserJock> my key was signed by a CS guy at my uni
[23:20] <infinity> ScottK: That was sarcasm, right? :)
[23:20] <infinity> ScottK: (The "some Debian guy who goes by vorlon" thing...)
[23:20] <slangasek> jcastro: I hear he's some kind of double agent
[23:20] <ScottK> infinity: Yes.
[23:20] <infinity> ScottK: Just checking.  My sarcasometer is off today.
[23:20] <ScottK> It was at UDS.  Understand.
[23:21] <LaserJock> in any case, do people think it's worth a TB agenda item?
[23:21] <ScottK> LaserJock: Under our current process I think there is no difference between a key belonging to ffm and John Doe.  The fact that one is more obviously a pseudonym doesn't change anything.
[23:21] <ffm> LaserJock, What, key signing being required?
[23:22] <LaserJock> I don't care about pseudonyms
[23:22] <LaserJock> that's not my issue
[23:22] <ffm> LaserJock, Fedora requires real names. (fedora was a pain, I had to go through a proxy)
[23:22] <ffm> As does debian AFAICT.
[23:22] <ScottK> Debian requires a key signed by a DD if you want to be a DD.
[23:23] <infinity> We also require real names.
[23:23] <ScottK> So yes, it means your identity verified.
[23:23] <infinity> s/We/Debian/
[23:23] <LaserJock> my point was about requiring a signed key
[23:23] <LaserJock> but well, pseudonyms might as well be thrown in at the same time ;-)
[23:24] <persia> Keys aren't currently important, although they'd be nice.  Names are more important, although even Debian allows pseudonyms if appropriately registered.
[23:24] <ScottK> I think we should be about consistent identity and not worry so much about correct.
[23:24] <ScottK> Unless we are going to have an enforced identity verification process, then there's really no point in insisting on a name.
[23:25] <LaserJock> but it would be nice to know what the rules are in any case, right/
[23:25] <ScottK> Agreed.
[23:25] <ffm> persia, Appropriately registered?
[23:25] <ScottK> I think that we know the defacto rules.
[23:25] <Amaranth> bug 225941 has what I think is needed for an SRU (haven't done one before) but as I cannot upload to main I'm not sure what to do next
[23:25] <LaserJock> ScottK: well, I didn't
[23:25] <LaserJock> ScottK: up until not long ago I assumed it was still a requirement
[23:25] <ScottK> OK.
[23:26] <ScottK> I think the defacto rules are known.
[23:26] <ScottK> Not necessarily to all.
[23:26] <LaserJock> now, yes
[23:26] <LaserJock> I guess
[23:26] <LaserJock> do we know if the MC/TB know the defacto rules?
[23:26] <persia> ffm: A registered pseudonym is one that can be uniquely tracked to a real identity, sometimes through a proxy.  Common cases are pen names, screen names, etc.
[23:26] <LaserJock> as opposed to just not paying attention
[23:27] <ffm> persia, How do you go about accomplishing that?
[23:27] <ScottK> persia: As a minor I doubt he'd be able to accomplish the registration without defeating his purpose in being pseudonymous.
[23:28] <persia> ScottK: In his jurisdiction, it's usually not an issue with parental consent.
[23:28] <ScottK> OK.
[23:28] <ScottK> I'm not familiar with the process.
[23:28] <calc> grr i saw a useful link a few days ago that showed the ide spec and where APM part is defined
[23:28] <ScottK> I thought the point was he was trying to avoid his parents googling his name.
[23:28] <calc> but i can't find it anymore
[23:30] <james_w> Amaranth: you want to propose an SRU for 225941 now?
[23:31] <Amaranth> james_w: yeah
[23:33] <james_w> Amaranth: you need to nominate for Hardy I think, and then subscribe motu-sru.
[23:33] <Amaranth> not motu, that's in main
[23:33] <james_w> sorry, ubuntu-sru
[23:38] <LaserJock> Amaranth: read wiki.ubuntu.com/StableReleaseUpdates
[23:38] <LaserJock> :-)
[23:50] <Amaranth> LaserJock: that says to upload it then subscribe those to get it accepted :P
[23:51] <LaserJock> what?
[23:52] <LaserJock> Amaranth: you subscribe ubuntu-sru first
[23:55] <jdong> Amaranth: you gotta get it ACKed before uploading it :)
[23:56] <Amaranth> ok, well now someone else can ACK it and upload it :P
[23:56] <LaserJock> jdong: though we dont' say that on the wiki page
[23:56] <jdong> LaserJock: you're joking?
[23:56] <LaserJock> nope
[23:56] <Amaranth> it's james_w's fix anyway :P
[23:56] <calc> from what hitachi has told me if you don't set the standby timer to something sane for their hard drives it parks _immediately_
[23:56] <LaserJock> it just says subscribe the SRU team
[23:56] <LaserJock> never says that you it has to be ack'd
[23:57] <LaserJock> we had a slight "accident" because of that the other day
[23:57] <calc> and from what it looks like the opcode that hdparm uses to set the standby timer is (perhaps?) old
[23:57] <LaserJock> jdong: I'm working on a fix
[23:57] <LaserJock> :-)
[23:57] <calc> in any case setting the timer with -S doesn't work
[23:57] <jdong> LaserJock: be sure to subscribe motu-sru and target it at the wiki
[23:57] <jdong> *ducks*