[12:03] Riddell: just add a 'c2' at an appropriate place [12:03] libkdecorec2 how hard can it be [12:03] Riddell: I'm open for better ideas, but we used that in Debian for the last transition as well (c102) === ironwolf [~ironwolf@c-24-6-169-124.hsd1.ca.comcast.net] has joined #ubuntu-devel [12:04] doko: the package name is libqt3c102 but the library is still libqt-mt.so [12:04] why's that? === herzi [~herzi@d006098.adsl.hansenet.de] has joined #ubuntu-devel [12:05] Riddell: look at the wiki [12:05] Why don't we just change the sonames? [12:05] Because upstream chooses the soname to match their API. If we change the soname then we render ourselves binary-incompatible with other distros and vendor-supplied binaries. This is important because the LSB3 intends to standardise the GCC 4.0 ABI; for Ubuntu/Debian to become binary-incompatible at this point would be the height of perversity. [12:05] Of course, when your upstream does bump the soname, you can drop the c2' from the package name, just like very few libs still have a g' on the end. [12:05] doko: any reason not to have c1002, apart from slight ugliness? [12:06] Kamion: good point. [12:06] as a compiler option, you write -fabi-version=[12] , the macros is defined as 102/1002. [12:06] just worried that at some point GXX_ABI will actually be "2" and we'll be screwed; or is that not possible? [12:06] Kamion: I prefer c1002, since it is very unlikely to be chosen by someone for a package suffix. [12:07] if GXX_ABI won't ever actually be "2", then I don't have a real objection [12:07] yes, we should address this and choose something that can be choosen by Debian as well. [12:07] mm, that's an interesting point [12:08] can we pre-clear it with Debian? [12:08] if I want to compile my own kernel (from kernel.org) to test some powerpc things... what inotify version should I grab? [12:08] because if there's a similar c1002 vs. c2 debate, and Debian chooses the other one, we're in for really serious pain [12:08] Kamion: y'mean ask the release manager and gcc maintainer? ;) [12:08] Kamion: no, I think the multiplier was introduced to get rid off the confusion (ABI 2 = 1002) [12:08] jbailey: +1 funny [12:08] jbailey: last time round, it was neuro who did it; I guess that was with a gcc hat on though [12:09] doko: which multiplier? [12:09] 10 [12:09] ok [12:10] I think going for a name that's picked for exactly the same reason as c102 was picked might leave us with less chance of being bikeshedded later [12:11] I'm tossing around the idea of implementing some sort of changes summary for the update-manager. Something that would list the changelog for all the packages at once. It seems however that update-manager is quite a bit more complex than I anticipated. Could someone give me a summary of the logic by which it works? [12:11] mvirkkil: I can do this, but I would rather like to do it tomorrow [12:12] because I want to go to bed soon :) [12:12] mvo: Ok. I'll ask again then :) Thanks. [12:13] Kamion: so what is the most politically correct way to address this change? On Debian-release people were asked for transitions and transition plans, so I can make at least a proposal. === koke [~koke@adsl229-164.unizar.es] has joined #ubuntu-devel === gustavor [~gustavo@200-171-140-32.dsl.telesp.net.br] has left #ubuntu-devel ["Leaving"] [12:14] doko: that's definitely a good place to get a proposal semi-blessed [12:16] ok, will post something tomorrow. my preference still is: len('c2') << len('c1002') [12:16] elmo: ping [12:16] smurfix: ? [12:17] What's the procedure for removing packages from breezy (like libgcrypt7...) -- assing the relevant bug to you? [12:17] elmo: ^ [12:17] assign [12:17] smurfix: already removed from Debian, or ubuntu specific? [12:17] hey elmo [12:18] smurfix, thats in universe [12:18] hi fabbione [12:18] smurfix, put it on our MorgueCandidates page after bereezy opened [12:18] smurfix: in any event, if it's ubuntu-specific removal in universe coordinate with the MOTMOTU [12:19] smurfix: if it's removed-in-debian, that'll happen semi-automagically shortly-ish after breeze opens [12:19] ogra: It's on doko's FTBFS list, didn't know he also imported Universe packages [12:19] smurfix: doko's large list had mainly universe packages, iirc [12:19] elmo: OK, thx [12:20] Hi elmo [12:20] hi pitti [12:20] elmo: please could you install a hoary chroot on davis and install jbailey's glibc packages (p.u.c) [12:20] elmo: can you please make {warty,hoary}-security uploads work again? [12:20] doko, elmo: Thos epackages aren't ready. [12:20] pitti: warty broke? [12:20] elmo: I uploaded a couple of packages, and according to mdz they are stuck in unchecked/ [12:20] elmo: they aren't in accepted or in REPORT or accepted/REPORT [12:21] oh, eh, blah, sorry [12:21] elmo: and I did no get a REJECT mail either [12:21] ajmitch_: the bug reports for main are already imported in bugzilla [12:21] anyway, I'm workin gon breezy now, they'll go live again with that [12:21] elmo: thanks [12:21] doko: I haven't got all the langpack stuff done, I'm cracking through that now, and then I have a couple l-k-h fixes to do. It'll be done by tonight. [12:21] jbailey: they are ok for building GCC biarch. [12:21] smurfix, no idea why its there, but for MOTU we have http://www.ubuntulinux.org/wiki/MorgueCandidates to propose packages that should get dropped, its just not wiped yet... [12:21] doko: alright [12:21] jbailey, elmo: maybe we need another chroot for that kind of thing === jnc balks "Morgue!?" [12:22] might as well call it the DebCemetary [12:22] doko: True, but the resulting binaries are also nptl-only right now as well as needing a minimum glibc version bump (for symbol versioning) [12:22] doko: I can create another chroot in a bit [12:23] doko: As long as the result doesn't wind up in the archive somehow... [12:23] elmo: that would be nice, so we have a platform to polish ppc64 [12:24] night everybody [12:24] night pitti [12:24] g'n Martin. [12:25] doko/elmo/jbailey: is there interest in building all of main to see what is ftbfs with 4.0? [12:25] lamont: let me setup breezy first? [12:26] elmo: sure [12:26] (but yes, I definitely think we should do that) [12:26] i.e. with breezy-test as soon as it exists [12:26] elmo: either hoary-test or breezy-test is fine with me for the test build... :0) [12:27] lamont: yes, that would be fine. although we should dtermine the order of rebuilds first. [12:28] doko: order??? what order... we just turn it loose... :-) [12:29] Kamion: Around? [12:29] order of library rebuilds, or do you want not have the rebuilt packages available for installation? [12:29] lamont,elmo: just as importantly as building, I think we should ensure that the packages that we actually publish to the world are all built with 4.0 [12:29] lamont,elmo: so we know that they work [12:29] jbailey: yes? [12:29] Kamion: Is zoneinfo-udeb something that should be contrib'd back to Debian eventually? [12:29] mdz: ABI change? [12:30] elmo: that too, but I mean just in general [12:30] the test builds tell us whether the stuff compiles, but not whether it still works [12:30] [meh, god I wish irssi had per-channel/window currently-typing buffers] [12:30] xchat has that [12:30] jbailey: I'm not sure. It was for the first-stage-questions crack, but I don't think we actually use it at the moment [12:30] yah, I know, that's why I miss it [12:31] mdz: current gcc 4.0 RC is known to e.g. miscompile python [12:31] jbailey: however we might want to use it in the future since some people (*ahem*mark*ahem*) want me to lose the dependence on the base system being installed for questions like timezone [12:31] so while I agree that's critically important, until we have 4.0 final and are prepared to start the C++ transition, it seems like a good head start to get started on the FTBFSes [12:31] elmo: fixed in my package [12:31] doko: keeno [12:31] elmo: oh, I agree, I don't think we should delay anything for that [12:32] wouldn't a python miscompile be caught by build-time tests? [12:32] we just need to have some plan for getting everything rebuilt [12:32] Kamion: we hope [12:32] (dreaming) [12:32] Kamion: yeah, now just try to not think about the 99% of the archive which doesn't have a test-suite never mind a remotely comprehensive one [12:32] Kamion: I can see that, cool. [12:32] jbailey: so I reckoned you were going to ask me about that, but I honestly don't know whether it's just bad crack yet [12:32] given I made it up on the spot under a certain amount of goal-related pressure [12:33] Kamion: No worries, just trying to pick obvious places where merging could be made less hairy. [12:34] jbailey: if you want to kill it, go ahead; I've kept the patch here in case I need it again in future [12:35] Kamion: I'll keep that in mind for the next time I have to do this. [12:35] elmo: ping [12:35] lifeless: hi [12:35] I'm here to nag [12:35] jbailey: (it's the interdiff from 2.3.2.ds1-19ubuntu3 to 2.3.2.ds1-19ubuntu4 if you want to reverse it) [12:35] lifeless: we know you ;) [12:35] doko: I'm touched ;) [12:36] lifeless: did you mail me about this at all? [12:36] lifeless: 'cos unfortunately, steve talked to me while I was hugely asleep and I have no idea what he said [12:36] elmo: I talked with you last week on IRC [12:36] [beyond "authserver. kick."] [12:36] you did? score my memory [12:36] steve was just a opportunistic nag [12:37] jbailey: mm, apart from adding libnss-dns-udeb and libnss-files-udeb to control_deps in debian/rules.d/control.mk; if that hasn't gone back to Debian yet, it should [12:37] ok, gimme a minute I'll dig up details and mail you. then I'll nag again. [12:38] lifeless: don't worry, got it === jcole [~jcole@palwebproxy2.core.hp.com] has left #ubuntu-devel [] [12:38] elmo: cool [12:38] elmo: so I realised I was being dumb - i tonly needs to be the chinstrap chroot [12:39] it won't hurt the others though - so whatever is easiest for you [12:40] PINE [12:40] pine [12:40] hmm ;] [12:40] definitly wrong screen === zul [~chuck@CPE0006258ec6c2-CM000a73655d0e.cpe.net.cable.rogers.com] has joined #ubuntu-devel === lamont wanders off to convert another user [12:45] yeah [12:47] lamont: wander of to convert another mirror ;] === hackeron [~hackeron@hackeron.user.gentoo] has joined #ubuntu-devel === Zugot [~bryan@pcp0010860530pcs.longhl01.md.comcast.net] has joined #ubuntu-devel [12:57] doko/jbailey: what arch name are you guys using? [12:58] elmo: The package name right now is libc6-dev-ppc64 and libc6-ppc64, but using "powerpc" as their arch. [12:58] elmo: Same type of config as sparc/sparc64 [12:58] ok === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel [01:12] that'll be breezy open then === JanC [~janc@JanC.member.lugwv] has joined #ubuntu-devel === Arrogance [~aks@CPE0050ba556e4b-CM001225423850.cpe.net.cable.rogers.com] has joined #ubuntu-devel [01:12] indeed [01:14] YAY [01:16] adjust your .procmailrc accordingly :) [01:18] Riddell: but /dev/null hasn't changed.... [01:18] Riddell, breezy is open now? [01:18] ogra: dholbach will be so happy :) [01:18] yeah [01:18] kent: according to breezy-changes it has === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel === HrdwrBoB [~matt@bob.is.teh.admin.at.vicnet.net.au] has joined #ubuntu-devel === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel [01:26] <|QuaD-_> Riddell: breezy changes is empty, isn't it? [01:26] <|QuaD-_> no messages yet [01:27] |QuaD-_, 2 [01:27] <|QuaD-_> ah, let me check again :) [01:28] <|QuaD-_> just checked :) that means we should start seeing universe open up again [01:28] <\sh> breezy is open? [01:29] no, i just did a clean install on my desktop :( === mike_douglas [~mike@70.70.204.42] has joined #ubuntu-devel === robertj [~robertj@66-188-77-153.cpe.ga.charter.com] has joined #ubuntu-devel === lamont_r [~lamont@209.120.156.178] has joined #ubuntu-devel === zerokarmaleft [~zerokarma@ip68-12-45-133.ok.ok.cox.net] has joined #ubuntu-devel === blueyed [~daniel@iD4CC00C5.versanet.de] has joined #ubuntu-devel === lamont__r [~lamont@209.120.156.178] has joined #ubuntu-devel [02:01] Is it a known bug that the password for a new user added through the "Users administration tool" is wrong? There is a different value in the shadow file from using passwd or KUser.. [02:01] could not find anything in bugzilla.. [02:03] blueyed: different in that the password doesn't work when the user types it, or different in that it doesn't match exactly? [02:04] it does not work.. haven't tried if it's lowercase though.. mom.. [02:04] because I would expect it to not match exactly, 4095 out of 4096 times or so [02:04] k, but it also does not work. [02:06] <\sh> shlibs-control file==libfoo.la? === Keybuk [~scott@syndicate.netsplit.com] has joined #ubuntu-devel [02:07] lamont_r, it's neither lowercase nor uppercase (would be strange enough). [02:07] not reproducable? [02:08] passwd works fine. [02:08] blueyed: dunno - not really somewhere that I can conveniently check that [02:08] whats the best tool for maintaining a small apt repo on chinstrap? [02:08] lifeless: apt-ftparchive packages . > Packages [02:09] lamont_r: so if I accrue my builds in a dir, then run that [02:09] lifeless: and (of course) apt-ftparchive sources . > Sources [02:09] apt-ftparchive searches the given directory (.) for all debs (packages) or dsc (sources), and builds the relevant file [02:09] so yeah, put them all in a flat tree, and go for it. [02:09] and yes, that can be multi-arch [02:10] at the moment I use dpkg-scanpackages [02:10] apt-ftparchive obsoletes that [02:10] to get Packages and Sources. whats the advantage ? ok [02:10] what about a Release file ? [02:10] so I can sign it and make mdz happy ;) [02:10] the other advantage is that apt-ftparchive is installed, and dpkg-scanpackages probably isn't [02:10] *I use dpkg-scanpackages right now* :) [02:10] vi Release; gpg --detach-sign Release; mv Release.sig Release.gpg [02:11] an empty file ? [02:11] lifeless: ah, ok [02:11] you can give apt-ftparchive a nice config file (although I haven't figured it out yet... see mvo, mdz, kamion, or keybuk?) [02:11] apt-ftparchive release might do it [02:11] see the one in dists/hoary on your favorite mirror, clone from that... [02:11] yeah. [02:12] but that doesn't set the invariant fields, just builds the md5sums/sha1sums for you [02:12] the config file lets you specify those [02:12] or just grep people's home directories on rookery for their scripts that perform archive voodoo :p [02:12] Keybuk: any favorites? [02:12] I use apt-ftparchive === lamont_r figured invoking a string of names would be productive === lamont_r will snarf Keybuk's config later then [02:13] the docs are good [02:14] Keybuk: to a point... :-) [02:14] then again, I've tended to want to do strange things with it. [02:14] oh, I just do "a directory with some packages in it" things [02:15] lamont_r: I have voodoo scripts at people.debian.org/~robot101/make-archive === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel [02:16] slight inefficiency (multiply scans arch: all packages) but generally works... [02:16] although probably doesn't solve whatever problem you're encountering, so I'll go away now :P [02:16] assuming it's still intact in the smoking embers of gluck [02:16] main problem is that the bazaar crack repo isn't signed [02:16] and I'm told it should be [02:16] oh, I don't understand that stuff either [02:17] it is intact :) [02:18] lifeless: once you have a Releases file, you just --detach-sign it and rename the .sig to .gpg === lamont_r heads off to dinner and fire dept training. back in a few hours. === dredg [niall@binky.blacknight.ie] has joined #ubuntu-devel === marcin_ant [~marcin@www.e-dev.tele2.pl] has joined #ubuntu-devel === cc [~cc@c210-49-121-44.eburwd3.vic.optusnet.com.au] has joined #ubuntu-devel [02:45] who should I ping about getting zsh from sid into breezy? [02:45] it has working baz completion [02:47] open up a bug maybe [02:47] it'll get synced eventually, I'm sure [02:48] it's forked! === bob2 forks Clint in the face === blueyed [~daniel@iD4CC04DD.versanet.de] has joined #ubuntu-devel [02:53] bob2: when breezy opens, we'll be merging the latest stuff from sid as a matter of course; no ping necessary === alex234 [new@pD9E1FCF0.dip.t-dialin.net] has joined #UBUNTU-DEVEL [02:54] ah, excellent, thanks === alex234 [new@pD9E1FCF0.dip.t-dialin.net] has left #ubuntu-devel [] === jcole [~jcole@palwebproxy2.core.hp.com] has joined #ubuntu-devel [03:06] mdz: fc3 kernel + ubuntu livecd is too hard [03:07] mdz: harder than michael jackson at disneyland === robitaille [~robitaill@d154-5-117-228.bchsia.telus.net] has joined #ubuntu-devel === fgx [~fgx@host134-170.pool8252.interbusiness.it] has joined #ubuntu-devel [03:10] <\sh> guys, for a shared library package e.g. libfoo_major.minor and libfoo-dev_major.minor what is the contents of shlibs.local? === Zugot [~bryan@pcp0010860530pcs.longhl01.md.comcast.net] has joined #ubuntu-devel === LeeJunFan [junfan@64-186-37-120.skycon.net] has joined #ubuntu-devel === bradb [~bradb@modemcable087.14-130-66.mc.videotron.ca] has joined #ubuntu-devel === ikuyaLoqu [~ikuya@openblocks.good-day.net] has joined #ubuntu-devel [03:31] mjg59: http://www.linuxsymposium.org/2005/view_abstract.php?content_key=95 [03:31] mjg59: given by len brown === marcin_ant [~marcin@www.e-dev.tele2.pl] has joined #ubuntu-devel === niran [~niran@lucianus.Stanford.EDU] has joined #ubuntu-devel === mdz boggles at #9122 [03:48] mdz: yeah ... [03:48] mdz: and while you're at it, i find that i sometimes make typos cos the # key is right next to enter... [03:49] dredg: in that case, the problem is solved by getting a us-layout keyborad [03:49] also, keyboard [03:49] oh, and can we force everyone to use a proper UK keyboard [03:49] thom: oxymoron city [03:49] so i don't have to contend with vertically challenged return keys [03:49] thom: will they be listening to good alternative Avril Lavigne while they're at it? ;) [03:49] daniels: yeah but then the symbols don't match up and i cry :( [03:50] thom: you can borrow sideshow's x40 if you like [03:50] I find the ` key confusing because I don't know what it does. Please disable it in all keyboard layouts [03:50] dredg: hm? every key I press on my keyboard matches up with what's printed on the caps [03:50] daniels: ozone now wants a uk keyboard for the key :-) [03:50] thom: oh dear. [03:50] daniels: if i were to use a US layout [03:51] dredg: oh sure. you just need to replace your keycaps also. :) [03:51] well, the entire physical keyboard, since we don't need this weirdo extra key. [03:51] daniels: send your keyboard here. it obviously does what i want [03:51] my keyboard is attached to my laptop, and so am I. [03:51] this could get messy === maswan hands dredg a proper knife [03:53] maswan: proper? :) [03:53] dredg: sharp, not too small? :) === dredg laughs [03:53] not a dull tiny knife for making decorative things to fruit. ;) [03:54] ARGH === maswan ducks and covers [03:54] E: Failed getting release file http://ftp.debian.org/debian/dists/hoary/Release [03:54] that's so lame === daniels giggles. [03:54] elmo: ... fix it? :P [03:55] haha === Zugot [~bryan@pcp0010860530pcs.longhl01.md.comcast.net] has joined #ubuntu-devel [03:55] heh. that would be one way of solving mirroring infrastructure. dump it into ftp-master.debian.org in the appropriate directories. :) [03:56] I suspect some people might have issues with that though. [03:56] us.archive has breezy already, they certainly are quick [03:56] mdz: they're triggered [03:56] oh [03:56] maswan: meh, they're too busy figuring out what the insert key is used for [03:57] maswan: "we don't have space for your lame ass architectures, but 50Gb of Ubuntu??? bring it on!!" [03:57] elmo: indeed. might even lead to a heated discussion on debian-devel@lists [03:58] elmo: once you're happy with the breezy stuff, let me know before you head off and I'll announce it to -devel (unless you want to do it) [03:58] mdz: I'm happy with it [03:58] ok === infinity breathes a sign of relief at remaining a legal immigrant for a while longer. === zul [~chuck@CPE0006258ec6c2-CM000a73655d0e.cpe.net.cable.rogers.com] has joined #ubuntu-devel [04:00] infinity: huzzah. [04:00] infinity: how long have you got? [04:01] daniels : We just put in an application for permanent residence, and I'm currently on a bridging visa while that's being sorted. [04:02] The Australian government how has 1845 of my dollars, so they better say "yes". [04:03] Well, to be fair, they have 1845 of American Express's dollars, but AmEx is likely to ask for it back soon. [04:03] infinity: heh [04:04] infinity: ouch. === infinity looks at the size of his INBOX and winces. [04:05] infinity: is that AUD, CAD OR USD? [04:05] mdz: aud [04:06] AUD, but I paid with a CAD credit card, so a bit more after exchange rape. === infinity makes a mental note to apply for an AUD credit card soon. [04:07] can't you juts kick amex really hard? [04:07] although you really want a mastercard or visa, since people actually accept those [04:07] Yeah, I generally use a MasterCard, I used the AmEx cause it has a higher limit, lower interest rates, and the government accepts it. [04:08] I can also do a chargeback if they deny my app. ;) [04:08] haha [04:08] (Wonder how well that would go?) [04:08] the words 'federal pound me in the arse prison' come to mind at this juncture [04:08] infinity: they would probably let you come back and then kick you out again [04:09] (II) NV(0): Unable to detect which CRTCNumber... [04:09] (==) NV(0): ...Defaulting to CRTCNumber 1 [04:09] (II) NV(0): Using DFP on CRTC 1 [04:09] (--) NV(0): Panel size is 1 x 1 [04:09] or what daniels said [04:09] thank you, nvidia! [04:10] good size! [04:11] That's almost as big as my craptop's screen. === ogra suggests Xmag [04:12] 'i couldn't work out how the hell the screen was wired, so i set up something that couldn't possibly work anywhere. cheers!' [04:12] debootstrap has a default mirror? I didn't realize [04:12] I thought the third argument was mandatory [04:12] mdz: suite and location are the only mandatory arguments [04:12] daniels: I see that now [04:13] If they were reordered, even suite could default to something reasonable. [04:13] A bit late for that now, though. === jsgotangco [~jsg@202.57.71.235] has joined #ubuntu-devel [04:15] oh, nvidia. [04:15] pNv->FlatPanel ? (pNv->Television ? "TV" : "DFP") : "CRT", [04:38] Kamion: ping [04:38] time to rename firefox [04:39] thom: ?!? [04:39] oh, to debian-firefox or whatever, thanks to the trademark stuff? [04:41] just firefox [04:43] thoms-firefox ? [04:43] ogra: heh. more likely to be firefox-really-blows-use-epiphany-instead ;-) [04:46] heh === womble [~mpalmer@216.017.dsl.nsw.iprimus.net.au] has joined #ubuntu-devel === susus [~sz@p5089DA77.dip.t-dialin.net] has joined #ubuntu-devel [04:54] Riddell: do you guys care about kvim? it's been totally dropped from vim upstream because it's unmaintained [04:54] (i'm uploading a new vim that takes 'breezy' as a distribution keyword) [04:55] haha; i'm just building that here [04:55] thom: i'll fight you! [04:55] thom: more seriously -- are you basing it off the hoary sources, or the sid sources? [04:56] naw, i'm just doing it locally with hoary [04:56] if you have sid done go for it [04:58] yeah, the only question is what to do with kvim === nohar [~nohar@je.suis.t1r.net] has joined #ubuntu-devel [05:03] kvim was dropped upstream for good reasons. [05:04] daniels: I'm unsure what to do, I think it would be nice to have something but kvim is unmaintained and yzis is reported not to be ready for use [05:04] infinity: yeah, but just throwing it out seems a bit harsh to the kubuntu folk, since they explicitly reverted debian's last throwing-out of kvim === tritium [~mrimbert@12-202-90-180.client.insightBB.com] has joined #ubuntu-devel [05:04] Riddell: would you be terribly offended if I uploaded vim without kvim? [05:04] daniels: but I think that since upstrea don't maintain it and debian don't have it any more it's unrealistic to expect us to have it [05:04] cool [05:04] daniels: of course not, I'm an emacs user :) [05:04] haha [05:05] ooh...evil [05:06] hey hey hey crazy kids! [05:06] aaaaaand GOOOOOD MORNING FREEDOM LOVERS! [05:06] hey jdub [05:06] whattup jdizzle [05:08] my blood pressure! [05:09] jdub: try a chill pill [05:10] MORNIN JDUB ! [05:17] i drank a lava lamp [05:17] it wasn't lava [05:17] did it taste like happy? [05:20] good morning [05:31] mako: like the blog post, haha [05:31] Riddell: uh, dude. x86 ubuntu cds are on a *seperate* mirror on se.archive [05:32] hum. so gcc4 has a new c++ abi? [05:32] thom: they used to be, we actually removed the redirect yesterday since we seem only to have 30-50MB/s peak demand. :) [05:33] maswan: right, but the stats are still broken [05:33] thom: right. I now remember that I didn't make apache log in xferstats format, so I'll have to write a perl script to integrate it [05:34] well, and rewrite some paths and stuff too anyway, probably. as well as resolving IPs, but we already have scripts for that. [05:35] so anyway, there is ~4.7TB of x86 ubuntu isos that are missing from those stats, I'll just make a note of that in irc for now. :) [05:36] heh === LeeJunFan [junfan@64-186-37-120.skycon.net] has joined #ubuntu-devel === stub [~stub@203-206-239-200.dyn.iinet.net.au] has joined #ubuntu-devel === dholbach [~daniel@p54A6535E.dip.t-dialin.net] has joined #ubuntu-devel [05:50] morning === kentaur [~kentaur@adsl-065-006-206-159.sip.mia.bellsouth.net] has joined #ubuntu-devel === infinity [~adconrad@loki.0c3.net] has joined #ubuntu-devel === diamond_ [~diamond@194.46.72.1] has joined #ubuntu-devel === maswan reads [05:54] hmm.. now I have a purpose for fixing the stats soon, to prove Riddell's blog false. ;) === diamond_ is now known as diamond === moyogo [~moyogo@69.156.166.86] has joined #ubuntu-devel [06:04] maswan: what url? [06:05] ah i see it [06:06] mako ? what about CC meeting ? === jdthood [~jdthood@aglu.demon.nl] has joined #ubuntu-devel [06:13] somebody call mako, since sabdfl, elmo and kamion don't seem awake as well :-) [06:13] dholbach: yeah, maybe having 3/4 in GMT timezones isn't such a hot idea... :) [06:14] er, BST currently [06:14] maswan: url? [06:15] planet [06:15] dholbach: his phone number is available here: http://mako.yukidoke.org/contact.html [06:16] somebody could send him an SMS [06:16] err whats up? [06:17] jsgotangco: ubuntu community council meeting, due to start about 15 minutes ago ,-) [06:18] i'll send an sms if nobody else has. don't want to plague him [06:18] dredg: *nod* [06:20] bob2: could you join #ubuntu-meeting if you're here? [06:20] oooo [06:20] sent === ups [~ups@203.200.160.36] has joined #ubuntu-devel === astro76 [~james@astro76.user] has joined #ubuntu-devel [06:25] are those meetings public? [06:25] aj: yes [06:25] aj: yeah, #ubuntu-meeting. open to the public. === aj watches [06:25] dholbach: no reply. [06:25] of course it's probably best if the CC shows up [06:25] not a lot to watch atm... with no chair [06:26] crimsun: i thought we were an autonomous collective? [06:27] "debian/rules dopatch"... Yeah, that's an intuitive target. [06:27] who's chair? sabdfl? [06:27] aj: elmo, kamion, sabdfl or mako could... [06:27] aj: mako [06:29] hrm, they're all in pretty similar timezones, where it's all late too [06:29] aj: i prefer to consider it as early ;) [06:30] very early.... === ogra mumbles inhuman early [06:32] elmo: ping? [06:33] fabbione: seems to have went to bed 2h30m ago :-/ [06:33] i was hoping he didn't === crb_ [~crb@hermitage.hs.net.nz] has joined #ubuntu-devel === kentaur [~kentaur@adsl-065-006-206-159.sip.mia.bellsouth.net] has left #ubuntu-devel [] [06:41] BREEZY-CHANGES MAIL! [06:41] jdub: -) [06:41] someone make the CC meeting happen! [06:41] jdub: I guess that means I need to understand what the breezy chroot toolchain looks like, so we can start having binaries.... [06:42] lamont: i don't think we have such a beast yet :-) [06:42] lamont: binaries are good ;) [06:43] thom: that was a longwinded way of me saying that none of the buildd's are currently configured to build anything for breezy.... [06:43] lamont: so we are go for source uploads, but nobody is building, right? [06:43] are you going to wait on gcc4 for that? [06:43] fabbione: no clue on the source side, but there are no binaries [06:44] lamont: ok [06:44] Amaranth: I'm waiting for a decision on the gcc4 process to build the chroots [06:44] but right now breezy is made of hoary binaries? [06:44] and sparc would be well advised to do likewise, fabbione [06:44] fabbione: yes [06:44] ok [06:44] I guess... :-) [06:44] than i think i can flush my queue to breezy safely [06:44] if it has binaries, they come from hoary [06:45] yes there are binaries [06:45] and thanks god i kept all i built even before hoary was closed [06:45] i thought i didn't === fabbione hacks the changes files [06:46] ls -asl /mirrors/ubuntu/dists/breezy/main/binary-i386/Packages.gz [06:46] 628 -rw-r--r-- 1 root root 635865 Apr 13 01:33 /mirrors/ubuntu/dists/breezy/main/binary-i386/Packages.gz [06:46] so i guess there are binaries [06:46] From: Matt Zimmerman [06:46] Subject: Breezy suite now open for business [06:46] in both cases i can only build old packages [06:46] Those of you who like to stay on the bleeding edge, update your sources.list [06:46] and hang on tight. The next few weeks will be a rough ride. ;-) [06:46] [06:46] lamont: ^ ha ha ;) [06:46] jdub: without buildd's is interesting :) === akurashy [~akurashy@65.23.250.25] has joined #ubuntu-devel [06:51] wait [06:51] so are the servers up yet? [06:52] fabbione: we have magic package making fairies for this release! [06:52] or are they going up today? [06:52] OddAbe19: which servers? [06:52] breeezy [06:52] OddAbe19: breezy exists in the archive, and source can be uploaded === HiddenWolf [~hidden@136.121.dynamic.phpg.net] has joined #ubuntu-devel [06:52] check check, i just didn't recall seeing anything in the mailing list [06:53] and once a decision is reached (in about 12 hours, by my estimate...) on the gcc4 transition process, etc, then we'll have binaries building [06:53] mail was sent today from mdz - see above [06:53] ok, i have to glance around [06:53] thanks alot [06:54] i can't do much anyway [06:54] sparc.u.c is not updated [06:55] i need to get elmo to move it or fix it [06:55] fabbione: and going away, if I hear things correctly... (will become another name, with i386, hppa, and sparc binaries...) [06:55] lamont: yeah ports.u.c afaik [06:55] yeah [06:55] you mean ia64? [06:56] /ubuntu-ports, I think it was [06:56] er, yeah. [06:56] ia64, hppa, sparc === lamont hangs his head in shame at that typo [06:56] well that is going to be interesting, since there are people already mirroring sparc.u.c [06:56] that is not me === lamont ran out of livecd's tonight... have to burn another stack [06:56] trivial to write a redirect for it... [06:57] goody [06:57] katie did like my upload to breezy [06:57] time to mangle the changes and reupload the queue [06:58] lamont: so hoary-updates stuff isn't building yet either? [06:58] hrm... [06:58] a bunch of stuff hit h-c, but my mirror update isn't interesting [06:59] jdub: will be shortly [06:59] (it didn't used to exist yet...) [06:59] heh [06:59] thanks :-) [06:59] opinion: which is worse - 3 hrs sleep or no hrs sleep? [07:00] dredg, 3 hrs, because it's just a tease [07:00] tritium: yeah, that's what i'm thinking too :) [07:01] :) [07:02] 3 hrs might be good still... anything less than that is a tease for sure [07:02] dredg: no sleep. i find that even 20 minutes helps, tho you mightn't feel great for half an hour afterwards [07:04] diamond: i see. sounds nasty [07:06] the real problem is being woken up in the middle of a rem cycle [07:06] 5 minutes can mean the difference between feeling like crap and feeling good when you wake up [07:07] this is getting too complicated :) [07:07] heh [07:07] 3 hours is too much [07:07] get 45 minutes [07:08] 3 hours is good [07:08] Amaranth: i typically survive on 2-5hrs/night [07:08] i can live on 3-4 hours [07:08] cat naps, you don't enter rem, you feel refreshed for the next couple hours [07:08] i used to sleep like 3-4 hours for 3 months [07:08] fucked myself after that hah [07:08] couldn't lseep less than 8-10 hours fo rlike 2 years [07:08] ive only just gotten back into being able to sleep a few hours and live through the day [07:08] When I was in school I slept 2-4 hours a night. [07:09] Amaranth: heh thats when i did this [07:09] maswan: ping [07:09] year 1 i'd stay up till ike 4am [07:09] get up at 7 and goto school [07:09] jdub: pong [07:09] Now I sleep 8 hours, no matter what (slaps and tornados not included). [07:09] maswan: hey hey! [07:09] I'm almost done with converting the stats and putting them in for the next xferstats run btw. :) [07:09] jdub: greetings, Gdubly One [07:09] maswan: haha gogo :) [07:09] maswan: thom said there was something funny with the .se ftp stats-- oh [07:09] ehm. Jdubly [07:10] maswan: hmm, so what's the deal? [07:10] Gdubly sounds better, more gnomeish :) [07:10] yeah, once i get to sleep i'll sleep up to 12hrs. getting to sleep is the problem [07:10] We can create a weak clone called Kdubly. :) [07:10] jdub: I offloaded the x86 install/live cds to a temporary server. [07:10] or vice-versa [07:10] maswan: aha! [07:10] maswan: that'd bugger the stats a bit ;) [07:10] jdub: so there is 4.7TB of those downloads that are missing from the main stats. :) [07:11] that'd explain why kubuntu appears to beat ubuntu [07:11] maswan: ha ha [07:11] yep :) [07:11] \ [07:11] maswan: livecds too? [07:11] jdub: yeah [07:11] maswan: i thought you only did the livecd the second time? [07:12] Lathiat: well, there is a far bit of livecds among those downlaods [07:12] maswan: rightio [07:12] maswan@miffo-m:~$ grep -c live access.log [07:12] 25940 [07:12] maswan@miffo-m:~$ grep -c install access.log [07:12] 61754 [07:14] maswan: heh === Burgundavia [~corey@24.68.134.11] has joined #ubuntu-devel [07:16] jdub: at least one buildd of each of i386,amd64,ppc is now running hoary-updates === infinity applies a patch to mozilla's 'libpr0n' and decides that, in itself, is enough reason to hack on mozilla.. [07:16] lamont: thanks! [07:17] Ok, that should run a while, now I need to rush off to work. [07:17] lamont: hoary-security? [07:17] jdub: that was already there [07:18] infinity: There are talks about changing the name, but they've been going on for years. [07:20] Amaranth : Changing the name would be just plain wrong. Software development without the occasional bit of levity is terribly boring. [07:23] lamont: tops [07:23] jdub: generally speaking, it's best for me to wait for the archive to exist for the release before I turn the buildd's loose on it (since it bitches every 5 minutes if I don't wait. xN buildd's) [07:24] and actually, I should go kill the hoary chroots now === eruin [~eruin@eruin.user] has joined #ubuntu-devel === auxesis [~lindsay@107.24.233.220.exetel.com.au] has joined #ubuntu-devel === snaggen [~snaggen@c-ae7070d5.034-37-73746f12.cust.bredbandsbolaget.se] has joined #ubuntu-devel === moyogo [~moyogo@69.156.166.86] has joined #ubuntu-devel === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel === rjo [~jordens@rjo.developer.debian] has joined #ubuntu-devel === milli [~milli@phantom.acmeps.com] has joined #ubuntu-devel === jsgotangco [~jsg@202.57.71.235] has joined #ubuntu-devel === pitti [~martin@box79162.elkhouse.de] has joined #ubuntu-devel [08:05] morning === ups [~ups@203.200.160.36] has joined #ubuntu-devel [08:06] pitti: morning! [08:06] hey pitti === zenwhen_ [zenwhen@h-67-102-63-103.phlapafg.covad.net] has joined #ubuntu-devel [08:10] morning pitti [08:10] hey thom [08:10] Hi thom, fabbione, dholbach === pitti whines about having to do an OO.o hog security update half a week after release [08:13] pitti: if the alternative was doing it the day before release, I'd pick 3 days after... :-) [08:13] lamont: why, day before CD building had saved lots of people lots of downloads... [08:13] lamont: it's an obvious two-line patch for 100-killme MB of software... [08:13] yeah, but it causes much more stress in that final few hours [08:13] shit. some idiot named Darryl Clarke deleted the LocoTeams page on the wiki. [08:14] Is anybody able to restore it? [08:14] smurfix, and the canadian page as well [08:14] i didn't know you could delete pages [08:14] sure you can :-/ [08:15] robitaille: I don't monitor that ... [08:15] He managed to first rename it to DarrylClarke. Stupid. [08:15] smurfix: maybe get the old version from the history? [08:15] pitti: It's *deleted*, I can't get at it any more. [08:16] pitti: otherwise that would be easy [08:16] D'oh; they aren't kept somewhere? [08:16] http://216.239.59.104/search?q=cache:4ERL0ReqpiwJ:www.ubuntulinux.org/wiki/LoCoTeams+ubuntu+LocoTeams&hl=de&client=firefox [08:16] he deleted it and added a DarryleClark page in it's place? [08:17] smurfix: ^ that's the google cache, at least the text is intact [08:17] Amaranth: you can rename pages [08:17] Amaranth: That also updates all the pages referred to [08:17] we hatses are wiki page even more noww.... [08:17] rename it back? :) [08:17] Amaranth: It Was Deleted. [08:18] Amaranth: ... suppose I could create a new empty one in its place [08:18] yeah [08:18] I wouldn't [08:18] he even made links to other pages referencing it [08:18] doh [08:18] ... though it'd be uch easier if an admin could rollback its history [08:18] much [08:18] You might wipe the history. [08:18] has anyone tried recycle_bin? === Amaranth checks [08:19] seems to have the frontpage? [08:19] that's an old frontpage there [08:20] ah, I get it [08:20] ohhh! === zerokarmaleft [~zerokarma@ip68-12-45-133.ok.ok.cox.net] has joined #ubuntu-devel [08:20] not in the recycle bin [08:21] nah, subpages of r_bin don't work the way the should [08:21] oh well [08:21] got it :) [08:21] doh [08:23] https://www.ubuntulinux.org/wiki/LoCoTeams [08:23] What else got wiped? [08:23] darryl clarke seems to be a BAD bot [08:23] I'm trying to get it back now [08:24] exchanged nearly everything with DarrylClarke [08:24] he changed every rerefence to LoCo Teams [08:24] Amaranth, http://www.ubuntulinux.org/wiki/CanadianTeam is one that seems gone [08:24] jsgotangco: You can do that easiy just pressing Rename [08:24] smurfix: you get the wiki-mails? [08:24] dholbach: No, he seems to have made an honest mistake [08:25] dholbach: still stupid though :-/ [08:25] smurfix: shall i forward you a bunch of wiki-changes-mails? [08:25] robitaille: back [08:26] I leave it up to you guys to fix anything else he did to the pages. :) [08:26] dholbach: I have the loco* ones [08:26] I think when you rename a page it renames all the references to it too. [08:26] goodness [08:26] Amaranth: exactly [08:26] Amaranth, thanks [08:26] smurfix: he also changed your page :-) [08:26] so, not a bot [08:26] dholbach: If you have the one where he deleted the locoteams page content, I'd appreciate it [08:26] smurfix: locoteams is back? [08:27] pitti: do you want to add the xhosa translation in the OOo build? [08:27] Amaranth: not exactly :-/ [08:27] oh, he screwed with it? [08:27] doko: not in a security update [08:27] doko: this belongs into -update === stub [~stub@203-217-40-144.perm.iinet.net.au] has joined #ubuntu-devel [08:28] doko: well, does it require substantial changes in the main OO.o package? [08:29] smurfix: doesnt look good :-/ [08:29] dholbach: getting there [08:30] smurfix: ha... got 124 wiki-changes mails from you [08:30] wow [08:31] and counting... [08:31] that's just from undoing all darryls changes? [08:31] dholbach: you're going to get a few more I'm afraid [08:31] im afraid so [08:31] Amaranth: Well, the thing sends change mails for each page it touches when renaming [08:31] dholbach: wiki-changes would have saved AptGet, wouldn't it? [08:31] smurfix: ah [08:31] so anything referring to LocoTeamList is shouting at us now [08:32] Amaranth: yeah... but i didnt keep them, once i read the changes, i delete them [08:32] ouchie [08:32] especially since I've managed to mistype the page name when getting it back :-/ [08:32] This is why I have a gmail account for reading mls. :) [08:32] Amaranth: i'm not sure 1GB is enough for this volatile stuff === elbi [~lbm@cpe.atm2-0-1071006.0x50a0824e.abnxx3.customer.tele.dk] has joined #ubuntu-devel [08:33] 2GB [08:33] eheh [08:33] dholbach: Up to 2080MB and counting [08:33] but it fills up pretty fast if you are subscribed to a lot of the lists [08:33] 2087MB now [08:33] my imap-server is fine with that much too :-) [08:33] sure, but your imap-server deletes things for real :) === elbi [~lbm@cpe.atm2-0-1071006.0x50a0824e.abnxx3.customer.tele.dk] has joined #ubuntu-devel === opi [~emil@217.153.156.1] has joined #ubuntu-devel [08:34] o_O 40 ubuntu-devel mails? === tritium [~mrimbert@12-202-90-180.client.insightBB.com] has left #ubuntu-devel ["Leaving"] [08:34] hmm lots of renaming going...and going.. [08:35] damn, too many changes, the wiki threw away the history :-( [08:35] and since I'm not subscribed to anything named DarrylClarke I didn't get the one where he deleted the list [08:36] we should really have better access control on the wiki :-( [08:36] dholbach: you don't have it do you? [08:36] jsgotangco: we really should have *a*better*Wiki*. [08:36] yessss....we hateses it a lot... [08:37] Moin can do real access controls, editable in the page. Me like. [08:37] MediaWiki sound good? [08:37] mediawiki is php [08:38] our wiki has so many good stuff but needs a lot of logical information flow === smurfix hateses PHP [08:39] Who cares about the language if it gets the job done? [08:39] well === lamont looks at the clock, decides that maybe it really is bed time [08:39] php/java taboo in our servers [08:40] Amaranth: I am actually amazed that people write stuff that's this good in that crappy a language [08:40] smurfix: dholbach is subscribed to EVERYTHING [08:40] jsgotangco, what's up with the wiki now? [08:41] Treenaks: If he deleted that change mail, it's hot helpful [08:41] something's been deleted? [08:41] mdke, Daryl Clarke [08:41] i'm subscribed, can i help? [08:41] smurfix: true, true [08:41] mdke: He renamed the team list page to his own insted of adding his, then he deleted the content [08:41] *very* helpful [08:42] right [08:42] darryl clarke, 61 edits [08:42] thats probably a bot [08:42] Anyway the whole wiki has 19 subscribers. *One* of them, please, send me the edit to restore the page [08:42] no its just links updating [08:42] i'll try [08:43] Otherwise I know what to spend the morning on. :-( [08:43] whoa, rms cc'ed ubuntu-devel for an email :) [08:44] Amaranth: not the first time, IIRC [08:46] smurfix, i don't have time to fix it, i have a class: but the subscriptions go to a gmail account, i'll give you the password? [08:47] smurfix: unfortunately dont have it anymore :-( [08:47] i've had a cursory look but can't find it [08:51] mdke: gimme === Zomb [~eb@linda.rhrk.uni-kl.de] has joined #ubuntu-devel [08:53] just fyi: http://www.ubuntulinux.org/community shows an almost empty page [08:53] smurfix, actually you might not need [08:54] smurfix, check pm === zenwhen [zenwhen@h-67-102-63-103.phlapafg.covad.net] has joined #ubuntu-devel [08:55] page restored *whew* [08:56] are there any ubuntu devs working with the two debian devs to package mono 1.1.6? [08:56] zerokarmaleft: i think tseng is working on mono stuff with the debian guys [08:57] k...i'd like to help out if i could getting mono figured out for breezy [08:58] zerokarmaleft: tseng has packages more or less done AIUI [08:58] when will the breezy branch be usable? [08:58] elbi: it's accepting uploads now, although no binaries are being built yet [08:59] doko: awake yet? [08:59] i keep getting these random spikees of like 100M traffic on my wireless interface graphs and i can't figure out wtf they are [08:59] wrong channel [08:59] thom, just the core framework + gtk#, gecko# etc. i assume? or third-party applications as well [08:59] lamont: okay, any idea when this process will begin? [09:00] zerokarmaleft: i assume it will all be rebuilt [09:00] Ok, running a new xferstats [09:00] elbi: waiting for a decision (due within a few? hours) of what the build tree should look like.. That will be finalized tomorrow US time [09:00] once that's there, we'll have a plan for what happens how [09:00] lamont: sure :) [09:01] doko: just figured I'd remind you about toolchain decisions, etc, etc... :-) [09:01] Lathiat, right, i'm just wondering where i could throw some effort [09:02] amu: ping [09:02] zerokarmaleft: ask tseng? [09:02] thom: can I please have the openoffice.org build deps in hoary-i386? [09:02] lamont: what was that about the "etc, etc"? [09:03] lamont: okay. sounds reasonable [09:04] lamont: last question; are these things discussed in a closed community? [09:04] doko: probably the question of toolchain ordering for the bootstrap type stuff [09:04] elbi: all of the discussion I've seen has been in this channel [09:05] lamont: oh, could be true, but some things isn't happening either here or the public mailinglists, is this in some way correct? [09:06] lamont: ppc64? [09:07] doko: I have two semil-immediate concerns: (1) what toolchain do I put in chroot-breezy? (2) what toolchain do I put in chroot-breezy-test [09:07] pitti: done [09:08] and then what (if any) manually ordered builds must I do to make it really ready to just open the gates and let it go [09:08] thom: thanks :-) === sivang [~sivang@box79162.elkhouse.de] has joined #ubuntu-devel [09:09] lamont: in both chroots the packages I'm currently preparing. in breezy-test, a new gcc-defaults as well [09:09] doko: for breezy test, I'm happy with manually forcing the symlinks... we don't need to upload a new gcc-defaults pacakge. [09:10] (since breezy-test doesn't actually exist in the real archive, and therefore can't take source uploads...) [09:11] ** (process:26931): CRITICAL **: egg_desktop_entries_add_group: assertion `egg_desktop_entries_lookup_group (entries, group_name) == NULL' failed [09:12] during Setting up xpdf-common (3.00-11ubuntu3) ... [09:21] updated se.releses stats for the last week: http://www.acc.umu.se/technical/statistics/ftp/index.html.en === hsprang [~henning@c174040.adsl.hansenet.de] has joined #ubuntu-devel [09:22] 20 TB of ubuntu-5.04-install-i386.iso love :) [09:23] WOW [09:23] nice [09:24] smurfix, ping [09:24] opi: === GheRivero [~ghe@hiscpdprx01.upsa.es] has joined #ubuntu-devel === mvo [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel === lamont sleeps [09:30] 4TB of the same on bittorrent === Pizbit wonders if that includes the mirrors. [09:31] thom: total, or only the seeds/peers you can see? [09:31] Pizbit: maswan is giving stats for _a_ mirror [09:31] Treenaks: this is the stats from the tracker [09:31] thom: (I'm not familiar with trackers, do they give stats?) [09:31] Treenaks: torrent.u.c:6969 [09:31] ah ok :) [09:33] such a classy port number === rubenv [~lambda1@kotnet-148.kulnet.kuleuven.be] has joined #ubuntu-devel [09:35] quite === Pizbit wonders if that includes the mirrors. [09:36] oops, wrong window heh === Pizbit had his glasses of and cleaning them while hitting uparrow+enter :) [09:38] daniels: d'oh, purging dbus-1 in favor of libdbus-1-1 breaks tons of stuff... [09:38] pitti: not entirely surprising [09:39] hey mvo [09:39] thom: of course not, but a clean upgrade path will get funny :-) [09:39] yeah [09:39] this is why it's not uploaded [09:40] hey dholbach === martink [~martin@pD955E1AE.dip0.t-ipconnect.de] has joined #ubuntu-devel [09:42] bbl [09:44] res [09:44] hm, I'm getting "I tried to upgrade but all I got was a brown background when I tried to log in" errors on the -nl list [09:44] 4 already === SuperLag [~aaron@SuperLag.developer.gentoo] has joined #ubuntu-devel === koke [~koke@155.210.13.152] has joined #ubuntu-devel [09:48] Treenaks : Oddly enough, I had that problem on warty, but not on hoary. [09:49] Treenaks : Never bothered to figure out why. :) [09:55] morning [09:55] morning mdz [09:56] Hi mdz === carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #ubuntu-devel [09:56] mdz: "morning"? you aren't in California any more? :-) [09:56] how was the CC meeting? [09:56] pitti: 0100 is morning ;-) [09:56] morning mdke [09:57] morning mdz [09:57] actually I just say "morning" regardless of the time of day [09:57] it's simpler === mvo grumbles about his tab completion [09:57] mdz: ah, ok :-) [09:57] hey mdz [09:57] mvo: completion_amount = 1 [09:58] mvo: set completion_amount = 0 [09:58] mdz: basically there was almost no meeting... not enough CC members around [09:58] :) [09:58] mako was there [09:58] a bad time for the UK [09:58] thanks mdz, pitti [09:58] mdz: well it is always a bad time for somebody [09:59] s/bad/bed :P [09:59] both bad and bed [09:59] yay... last 3 external drivers to update and 2.6.12rc2 is almost ready [09:59] new ipw2200? === fabbione loves these insane marathons [10:00] mdz: new everything [10:00] fabulous [10:00] mdz: updated everything to the last bleeding edge [10:00] break it early.. [10:01] mdz: we also got a couple of patches upstream in like 10 minutes :) [10:01] our precompile on 6 arches tests are very good === d3vic3 [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-devel [10:02] mdz: + i updated all the external drivers documentation.. it was missing a bunch of drivers... [10:02] that will make it simpler in future === Seveas [~seveas@ksl403-uva-142.wireless.uva.nl] has joined #ubuntu-devel [10:02] have you written some tools to make it easy to convert the drivers into patch form? [10:03] mdz: no.. because some of them come as patches, others as tarballs, others need a precompilation beofre becoming a patch.... [10:03] it's way to complex to write a generic tool [10:03] it needs to be done case by case.. [10:03] what a pain [10:03] mdz: don't tell me :( [10:04] mdz: specially the 2 or 3 that needs precompilation are a mess [10:04] because they generate some files at build time [10:04] and some are even arch specific [10:04] (reason why we still don't have ndis for amd64) [10:05] ndiswrapper for one arch is too many already ;-) [10:05] mdz: heheheh but it has been reported to be working on amd64 so i expect people will ask for it pretty soon [10:06] and the really nice thing is that for each driver we update.. it is another rebuild [10:06] just to be sure the update is correct [10:06] that still doesn't involve that the code is portable [10:06] that comes at a later stage :P [10:08] so now you have documentation for how to update each driver? [10:08] mdz: partially yes [10:08] mdz: it changes all the time.. so it is kinda a bet on each update [10:09] the best practise is to compare the interdiff between the old and the new patch [10:10] it will catch 99% of the cases [10:11] smurfix: I guess I can see why moin disables page renaming by default [10:11] well, in zwiki it's far to easy to make a mistake that way [10:11] plus the rename feature conveniently forgets to take the pake subscriptions into account [10:11] I have never had that problem; how does it happen? [10:13] s/pake/page [10:15] I got an apology email from the perpetrator ;-) apparently he just hit the wrong button. Dunno how he managed that, they aren't exactly next to each other [10:16] smurfix: and you get a big fat confirmation dialog in addition... [10:16] at least we know it was no bot that did that but somehow he did make links [10:16] rename doesn't have a confirmation dialog [10:16] pitti : 25 out of 27 patchsets applied for mozilla. Just sorting out the last two, as they contain several patches each, in a none-too-organised fashion. [10:16] jsgotangco: He fudged around, trying to undo the nonsense, but only made it worse === pitti hugs and kisses infinity [10:17] "Undo" not working across renames apparently didn't help either [10:17] infinity: so it will become one big update? [10:17] infinity: that's not too bad from an user's POV, just one download :) [10:17] pitti : Yeah, looks like it. Stalling on the "hard" patch was pointless once I went past it. :) [10:18] anyway, off to teach a client how not to screw up his Asterisk server again ;-) [10:18] pitti : I'm hoping to finish up all the patches tonight, and we can spend some time begging people to test that nothing (much) broke. [10:19] pitti : Probably won't want to upload until we've had several people test for several days, in light of the massive amount of code touched. [10:19] infinity: yeah, that's sane === infinity takes a coffee break before diving into the last two evil-looking sets. === ikuyaLoqu [~ikuya@openblocks.good-day.net] has joined #ubuntu-devel === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has joined #ubuntu-devel [10:26] ?Also in this spurt is the Worple Guide which was worpled from the Ubuntu Worple to Worple and is now a permanent feature of the Ubuntu worple worple.? === elbi [~elbi@cpe.atm2-0-1071006.0x50a0824e.abnxx3.customer.tele.dk] has joined #ubuntu-devel [10:27] interesting review ;) http://mpt.net.nz/archive/2005/04/11/ubuntu [10:27] yes we have all seen it :) [10:28] i agree, disagree and he is wrong on a few points but its definately an interesting read and brings up some stuff that needs attention [10:28] bah [10:29] doko: the current Hoary package FTBFS'es in hoary ... *sigh* [10:29] pitti: FTBFS? [10:29] Fails to build from source [10:29] ah right [10:30] where can i look for a TODO list? [10:30] for ubuntu next release in general or specifics if possible? [10:31] http://udu.wiki.ubuntu.com/UbuntuDownUnder/BreezyGoals is too general [10:31] pitti: which package? [10:31] doko: openoffice.org [10:31] where's the log? [10:31] OOo builds? :) [10:31] doko: I /msg'ed you === seb128 [~seb128@ANancy-151-1-61-150.w83-196.abo.wanadoo.fr] has joined #ubuntu-devel [10:32] Morning seb128 [10:32] hey hey pitti :) [10:34] morning seb [10:34] all this gcc-4 mess iz a gtk bug [10:35] fabbione: gcc links against gtk now? [10:36] Treenaks: yeah... [10:36] wow.. === fabbione hides [10:36] must be the VisualStudio like GUI they're working on [10:36] to make it better [10:36] fabbione: hi fabbione :) [10:36] Treenaks: yeah [10:36] ehhe [10:36] fabbione: gtk is the love :p [10:37] seb128: like gamin? === hno73 [~Henrik@henrik.gotadsl.co.uk] has joined #ubuntu-devel [10:38] fabbione: no no, gamin is a piece of crap :/ [10:38] seb128: i think it will be easier to rewrite than to fix it.. really [10:38] fabbione: some gnomevfs guys are thinking to use directly inotify instead of gamin [10:38] the idea behind it is very simple [10:38] rewrite it.. again? [10:38] apparently inotify is really easy to use === astharot [~isager@6a614fb7c5442de0.node.tor] has joined #ubuntu-devel [10:38] seb128: indeed it is [10:39] seb128: that'd rock, one piece of software less in the stack [10:39] does inotify work on non-local filesystems? [10:39] pitti: that's what the gnomevfs guys said, just drop the IPC stack [10:39] (and does gamin work on those?) [10:39] Treenaks: let me check.... [10:39] Treenaks: I don't think gamin does that [10:40] ciao [10:40] Treenaks: do you really want it to work on non-local filesystems? [10:40] Treenaks: in theory it can do that [10:40] I think nautilus currently does polling, but in a busy-wait loop [10:41] torkel: I can imagine some situations [10:41] Mithrandir: it does with dnotify, yes [10:41] torkel: though polling rates would have to be reduced drastically (if it's polling..) [10:41] Mithrandir: the dnotify backend is a mix with poll [10:41] not with inotify [10:42] fabbione: I'm just saying that I see ~70% CPU usage if I have a nautilus window open over ssh. :P [10:42] Mithrandir: ahaha [10:42] Mithrandir: ssh -X nautilus or nautilus ssh:// [10:42] what's going on with breezy and the buildchain changes ? [10:43] on both my laptop (which is a 1.4GHz P-M) and my desktop (which is an a64 4000+) [10:43] seb128: you can upload again :-) [10:43] Treenaks: for some remote/global filesystems (read AFS) it would be a nightmare, unless you can restrict it somehow [10:43] pitti: breezy is open? [10:43] seb128: I'm sure you are sitting on 281 packages to be uploaded :-) [10:43] Mithrandir: breezy is GO [10:43] yay! [10:43] CRACK!!! [10:43] :) [10:43] dump your crack, folks! === fwiffo [~b@cpe.atm2-0-1101155.0x503f8eca.bynxx8.customer.tele.dk] has joined #ubuntu-devel [10:44] is mom and such running too? [10:44] yeah but there are no binaries yet... are they? [10:44] fabbione: it's probably a copy of hoary. [10:44] right now yes [10:45] but afaik buildd's are not building breezy yet [10:46] oh right [10:46] no pmount-0.8 [10:46] pitti: right, I've to upload ... but have we switched to gcc4 ? [10:46] dunno [10:46] pitti: seb128 is 7bit only, so 281 looks a bit much ;) [10:46] Mithrandir: no mom yet [10:47] doko: certainly 7 bits of packages every day? :) [10:47] and he has a gross lag now [10:47] he's been steaming off by uploading packages to Debian, though [10:48] really, if Debian was smart they would bribe Mark into cutting off seb128 upload priv one day out of the week [10:48] lol [10:49] hah [10:51] Burgundavia: ? === zorglub [~zorglub@wahe.diwi.org] has joined #ubuntu-devel [10:51] get sarge out in about 2 weeks [10:51] mouarf :p [10:52] yeah, because what sarge needs is *more* package churn :P === zorglub [~zorglub@wahe.diwi.org] has joined #ubuntu-devel === jgotangco [~jsg@202.57.71.235] has joined #ubuntu-devel === maskie [~maskie@196-30-108-163.uudial.uunet.co.za] has joined #ubuntu-devel [11:01] Kamion: does the installer try to use unicode over a serial console too? [11:17] elmo: glib2.0 sync from debian please [11:21] seb128: i think he is going to push the bid red "import * fron Debian" button quite soon [11:22] oh right :) [11:24] hey seb128 [11:25] Morning jdub :) === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has left #ubuntu-devel ["Leaving"] === Seveas [~seveas@ksl403-uva-142.wireless.uva.nl] has joined #ubuntu-devel === Alessio [~Alessio@host147-51.pool80116.interbusiness.it] has joined #ubuntu-devel === Bnonn [~bnonn@60-234-131-84.bitstream.orcon.net.nz] has left #ubuntu-devel ["Leaving"] [11:54] zerokarmaleft: mono will be in breezy by the end of this month [11:55] \o/ [11:55] tseng: does that include beagle? :) [11:55] beagle runs like a champ [11:55] beagle's already there, yo [11:57] might manage to get tomboy into sid too [11:57] dajobe took my package and added the jimmac icons [11:57] tseng: should i keep updating beagle, or do you want to track it? [11:57] jdub: you may certainly manage it [11:57] jdub: its less of an issue now sans policy [11:59] heh [11:59] i'm not sure i want to keep tracking it ;) [12:00] jdub: YOU WIN [12:00] k, ill take it [12:00] mwah :) === jdub goes for more beer [12:01] daniels: at JSBH :) [12:01] jdub: bah, I'll drag thom to the better (and original, mofo) JSBH tomorrow night [12:03] have we really had 14 USNs for warty's kernel? [12:04] daniels: kernel bugs in the last 6 months have been at a crazy pace [12:04] alot of those usn's cover multiple cans [12:05] yeah [12:05] i've proofread most of them ;) i just didn't realise it was 14 [12:05] utterly batshit insane [12:06] those kernel people should learn to code! [12:06] I assume they took lessons from the Mozilla people. [12:06] Or vice versa. === infinity gives up for the evening before he completely loses it. [12:07] nah, the people to get lessons from are the ones who write apps like ohh i dunno, phpbb2 or phpmyadmin [12:07] infinity: omg [12:07] dredg: php4 itself [12:07] pitti : One patchset left, unless you have new CANs since I started. :) [12:07] pitti : Also, this last patchset contains a small patch we'll need to update hoary's Mozilla (but not firefox) with. [12:07] infinity: actually, four === astharot [~isager@8defad310e78e40e.node.tor] has joined #ubuntu-devel [12:09] infinity: which patch? [12:09] pitti : https://bugzilla.mozilla.org/show_bug.cgi?id=285438 [12:09] ciao [12:09] pitti : Last patch, check the last few comments. [12:09] pitti : Was supposed to be fixed in 1.7.6, but someone missed a file. [12:10] d'oh [12:10] infinity: okay, great to hear [12:11] pitti : In the future, someone should track these as they happen. Backporting one patch(set) at a time is a bit easier than 30ish. [12:11] Not to mention easier to test. [12:11] Obviously. :) [12:12] infinity: by any means === infinity will be mildly surprised if his builds of moz/tbird/ffox tomorrow don't just crash on load.... Or make the CPU explode or something. [12:14] pitti : Anyhow, your one-line patch for 1.7.6 should be simple enough. Lucky bastard. :) [12:15] pitti : That's CAN-2005-401, if you're too lazy to read all the comments. [12:15] (And who'd blame you?) [12:16] Anyhow, I'm going to head home and do something to make me stop thinking about C++ and XUL. === infinity -> gone. [12:16] infinity: I already marked this CAN as fixed in hoary, since Mozilla claimed that it was fixed in 1.7.6 :-( [12:16] pitti : Then unmark it. ;) [12:16] infinity: already done :-) [12:17] Heh. [12:17] 'Night. [12:17] night infinity [12:17] infinity: but this is fixed in ffox? === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-devel === stub [~stub@203-217-40-144.perm.iinet.net.au] has joined #ubuntu-devel [12:21] pitti, so there will be security fixes for Hoary soon? [12:21] kent: yeah, mozilla and OO.o [12:24] pitti, ok. I used to track Hoary during the last period of development, so I got used to the update-icon showing every day in the panel. Its been so strange not having to update for some time now.. :) [12:24] kent: that's actually the ideal state for a stable release :) [12:26] pitti, yeah, i guess so. But this is the first update since Hoary went final, right? [12:26] yes [12:33] https://www.ubuntulinux.org/support/documentation/usn/errorreferencefolder_view. "security notices that affect the current supported releases." The first one of the list is a bout png, and if you click on its link, it says that it affects only ubuntu wart. But the page with CAN-notices says that it affects current supported releases, and is not Hoary also supported, as is Warty (18 month support, or something right [12:33] ?) [12:34] Would it not be better if the page with security-notices had also a notice about which distrubution if affects? Right now it sort of looks like a new install of Hoary directly triggers some security-problems. [12:34] Sorry for the spam. You might shoot me at will.. [12:34] kent: hmm, right, that wasn't necessary during warty... [12:34] kent: right now all of them only affect wart [12:34] y [12:35] kent: this is mentioned in the texts, though [12:36] pitti, yes, for those who looks at the reports, it says which distribution it affects. But some might not do that, and just have a look at the page and draw some conclusions from that. If it was possible, it would be better if the first list could state also which version of ubuntu.. === odiX [~manfred@62.116.73.122] has joined #ubuntu-devel === theine [~theine@x1-6-00-11-24-09-2f-60.k324.webspeed.dk] has joined #ubuntu-devel === GheRivero [~ghe@hiscpdprx01.upsa.es] has joined #ubuntu-devel === CarlK [~ck2@c-24-13-52-189.hsd1.il.comcast.net] has joined #ubuntu-devel [12:39] daniels: i have an update of dbus-mono when breezy opens [12:40] its open [12:40] it is? [12:40] Yes [12:40] YEBO ! [12:40] is it broken yet :P [12:40] thats fast =) [12:41] if not, I can help [12:41] tseng: what are you doing to dbus-mono? [12:41] tseng: i have experimental 0.32 stuff at p.u.c/~daniels/dbus/ [12:41] daniels: unhack [12:41] cartman, no buildd yet, AFAIK [12:41] daniels: /usr/lib/mono 4 life [12:41] tseng: don't we have to wait for the mono packages to do that? [12:41] d3vic3: argh repos. up yet? ie can I point sources.list to it? [12:41] daniels: ok we can work on it later then [12:41] yes [12:41] daniels: yeah i have the mono packages also. [12:41] breezy is open afaik [12:41] tseng: oh, rad === herzi [~herzi@c208154.adsl.hansenet.de] has joined #ubuntu-devel [12:42] uhm yeah [12:42] tseng: ping me when you upload mono and we'll sort it then [12:42] Beagle's in breezy? [12:42] daniels: great [12:42] mjg59: not atm [12:42] mjg59: beagle is in NEW until we have a new mono === mpt [mpt@203.184.36.140] has joined #ubuntu-devel [12:42] ahh [12:42] Ah, got you [12:42] have we got a new kernel with new inotify yet? [12:42] Fabbione is working on the crack [12:43] we're all working on craaaaaaack [12:43] daniels: Except you, who's working on his lca paper? [12:43] yeah [12:43] i am almost there :) [12:43] crack is good [12:43] mjg59: *ahem* [12:43] when is buildd for breeze supposed to be alive? [12:44] hoary is not unstable enough for me ;P [12:44] cartman: within a day probably [12:44] fabbione: cool, thanks [12:46] Are there any compilers for linux that can output a windows executable of basic C programs [12:46] *of a basic === koke [~koke@155.210.13.152] has joined #ubuntu-devel === nullaresnata [~maia@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel [12:51] daniels: now hal 0.5 actually starts up :-) had to fix an upstream bug and lots of changes, but now it works :-) [12:52] pitti: sweet [12:52] pitti: CAN I UPLOAD YET? [12:52] YES YOU CAN [12:52] :) [12:52] daniels: however, please don't upload your current dbus-1 package yet :-) it's still pretty crackful [12:53] pitti: which parts are broken? [12:53] daniels: at first build, it linked against the old libdbus-glib in /usr/lib, not in the build dir [12:53] daniels: "it" -> python-dbus [12:53] UGH [12:54] that seriously sucks [12:54] daniels: and of course it's lacking all sorts of conflicts, provides, etc. for transition [12:54] daniels: LD_LIBRARY_PATH :-) [12:54] daniels: then, are you really sure that you want to rename /etc/init.d/dbus-1? [12:54] er, isn't LD_LIBRARY_PATH used for runtime? [12:54] surely we'd need -L../dbus or whatever [12:54] daniels: erm, sorry, -L probably [12:54] yeah [12:55] and, uhm, renaming /etc/init.d/dbus-1 kinda bites [12:55] i might fake it so it gets kept [12:55] daniels: why not just keep /etc/init.d/dbus-1? it's a conffile, and if you rename it, you have to do a proper migration [12:55] elmo: ping [12:55] pitti: because it requires more work than mv {dbus-1,libdbus-1-1}.init :P [12:55] daniels: exactly [12:55] pitti: dbus-1.init won't ever start anyway [12:56] pitti: 'cause it'll test for $DAEMON and bail. and dbus-daemon-1 got renamed to dbus-daemon. [12:56] daniels: why did you rename dbus-1 to libdbus-1? actually there should be two packages (one with the daemon, one with the library) [12:56] thom: around ? [12:56] pitti: libdbus-1-1 [12:56] pitti: well, I needed to rename it for the soversion change anyway [12:57] libdbus-1.so.0 -> libdbus-1.so.1 [12:57] daniels: yeah, but why not have an additional dbus-1 package for the daemon? [12:57] because then stuff that depends on dbus-1 ends up not dragging in ... oh blah [12:57] dbus-1 depending on libdbus-1-1 [12:57] that is *so* *nasty*. [12:57] daniels: I don't particularly mind that, but having a new dbus-1 package might ease transition and upgrades [12:58] yeah [12:58] daniels: the dependency should be automatic in ${shlib-depends}, shouldn't it? [12:58] hmph, I'll do that then [12:58] pitti: yeah [12:58] daniels: however, these were just some thoughts to ease transition, please just ignore them if you like [12:59] nah, we need a smooth transition [12:59] and daemon/lib should be split anyway [12:59] daniels: btw, libdbus-1.so.1 is for clients, too, right? [12:59] daniels: if so, then it should be separate, otherwise you would start the daemon if you build a client package === jdthood [jdthood@x126.decis.nl] has joined #ubuntu-devel [12:59] daniels: (since it's pulled in as a build-dep) [01:00] pitti: actually [01:00] dbus_bindings_la_LIBADD = $(top_builddir)/dbus/libdbus-1.la $(top_builddir)/glib/libdbus-glib-1.la [01:00] yeah, libdbus-1.so.1 is for clients [01:01] pitti: i don't see that -ldbus-1 is ever used? [01:01] daniels: yay, Makefile.am, that's nice [01:01] pitti: that's what's already there :P [01:01] daniels: ldd /usr/lib/python2.4/site-packages/dbus_bindings.so [01:01] daniels: this linked against libdbus-glib-1.so.0 after first build [01:02] bong [01:02] and thus did not work [01:02] daniels: I installed the new dbus-glib, built again, then it worked [01:02] daniels: this really looks like if it used /usr/lib/libdbus-glib [01:04] daniels: btw, I think we should collect dbus, hal, gnome-vfs, pmount, and g-volume-manager on p.u.c. and upload them all at once when it finally works [01:04] daniels: what do you think? [01:07] amu, Riddell: ping === cartman [foobar@cartman.developer.konversation] has left #ubuntu-devel ["Ich] === cartman [foobar@cartman.developer.konversation] has joined #ubuntu-devel [01:08] pitti: hi [01:08] Riddell: kubuntu needs your love: CAN-2005-1046 === xuesheng [~xuesheng@ALamentin-101-1-10-103.w81-248.abo.wanadoo.fr] has joined #ubuntu-devel [01:10] pitti: i'm just looking at it now: [01:10] cc -shared .libs/dbus_bindings.o -Wl,--rpath -Wl,/home/daniels/canonical/dbus/dbus-0.32/dbus/.libs -Wl,--rpath -Wl,/home/daniels/canonical/dbus/dbus-0.32/glib/.libs -L/home/daniels/canonical/dbus/dbus-0.32/dbus/.libs -L/usr/lib -L/usr/lib/gcc-lib/i486-linux/3.3.5/../../../ ../dbus/.libs/libdbus-1.so ../glib/.libs/libdbus-glib-1.so -lnsl -Wl,-soname -Wl,dbus_bindings.so -Wl,-version-script -Wl,.libs/dbus_bindings.ver -o .libs/dbus_bin [01:10] hey, seb128 now uploads with his @debian.org address :) [01:10] pitti: yeah, I think we should have a common staging area [01:10] pitti: doh, thanks for noticing [01:11] seb128: well, it shouldn't really matter [01:11] pitti: need to update my ubuntu script :) [01:11] <\sh> I'm tired guys [01:11] daniels: hmm, this actually looks right... [01:11] pitti: pebcak kthxbye [01:13] daniels: what went wrong? :-) [01:13] pitti: you broke the build === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel [01:14] it built fine, it should FTBFS if something is wrong.. ? what did I do wrong? [01:14] pitti: i dunno [01:14] pitti: we could blame thom if you like [01:15] daniels: I'd like that much better than blaming me :-) [01:15] daniels: anyway, we can hope that this doesn't happen at the buildd [01:17] yeah [01:17] but i dunno [01:17] i'll see if I can reproduce it [01:17] totally shouldn't happen tho [01:18] daniels: the really bad and h4ck1sh solution is a build-conflicts: to itself [01:18] daniels: but there must be a better way [01:18] build-conflicts so suck... [01:18] b-c'ing on dbus-1 (<< 0.32-1) is shit [01:19] daniels: so shall I convert hal to use /etc/init.d/libdbus-1 now? or will you revert to /e/i/dbus-1? [01:19] pitti: i'll revert to dbus-1 [01:19] okay, thanks [01:23] Kamion: Around? === xuesheng [~xuesheng@ALamentin-101-1-10-103.w81-248.abo.wanadoo.fr] has left #ubuntu-devel ["Leaving"] === cartman [foobar@cartman.developer.konversation] has left #ubuntu-devel ["Ich] [01:36] seb128: did you already happen to package g-v-m 1.3? [01:36] seb128: I like to start with it since I have the new hal running now [01:36] seb128: I just want to avoid duplicate work [01:37] pitti: nop, I consider g-v-m to be yours :) [01:37] seb128: thanks :-) === pitti starts to package [01:37] thank you :) [01:38] seb128: I want this new crack to play around with device encryption [01:38] you are packaging hal 0.5 ? [01:38] seb128: yeah, initial version is ready [01:39] seb128: it's a PITA to upgrade currently (needs some dbus coordination, hold your breath), but once it's installed, it works reasonably [01:39] cool [01:39] seb128: I just had to drop ogra's patches for now :-( [01:39] oh ? [01:39] seb128: well, and the thing does not recognize my hd partitions at all - let's see what breaks with this [01:40] seb128: yeah, hal has a completely new architecture, we need to redesign ogra's stuff [01:40] hum, k [01:40] seb128: we have 6 months to break it furth^W^W^Wfix it === nullaresnata [~henrique@cb-217-129-175-184.netvisao.pt] has joined #ubuntu-devel [01:50] jbailey: ping? [01:51] pitti: here! === trukulo [~mzarza@26.Red-81-45-239.pooles.rima-tde.net] has joined #ubuntu-devel [01:51] it's me or ubuntu-devel list is full of bugs insteadof threads of development? [01:51] jbailey: would it be possible for cdbs, "debclean", to just remove the build-tree instead of trying to unapply all patches, make clean, etc.? [01:52] trukulo: unfortunately this is the case :-( [01:52] trukulo: not just you [01:52] daniels, pitti: so that's a real problem [01:52] pitti: Yes, I've just never done it. Can you file a bug? [01:53] jbailey: debian or ubuntu? [01:53] trukulo: yeah [01:53] hello [01:53] Hi zyga [01:53] daniels, perhaps closed devel-list ? [01:53] is there any # with gpl-savvy lawyers? [01:53] pitti: Sadly, ubuntu if you want it looked at quicker. the debbugs for cdbs is a bit of a maze at the moment. [01:53] I'm currently in the process of releasing the project I'm working on under GPL [01:54] jbailey: okay :-) as long as you push it into Debian, too? :-) [01:54] and I have a few questions [01:54] trukulo: that would suck :( [01:54] pitti: sb would kill me if I didn't. =) [01:54] trukulo: maybe closed posting, but definitely not closed reading [01:54] pitti: hi :-) [01:54] zyga, ask me if you want, if i can ask, i'll tell you [01:54] daniels, i know, i know, but it's a problem with that list [01:54] yeah [01:54] daniels, perhaps closed posting would be good [01:54] as you say [01:54] trukulo: generally there are several issues [01:55] jbailey: was the intentional? [01:55] zyga, tell me in private if you want [01:55] trukulo: the project was created by me and later on by my friend [01:55] zyga, then you need the permission of ALL the creators [01:55] trukulo: (it's not top secret ;-) I can speak here if it's okay with the others [01:55] daniels: e? I don't do e... [01:56] trukulo: the problem is that while I was working for the institution that sponsored the project [01:56] jbailey: #9152, TIA [01:56] trukulo: my friend was doing it as his major in CS [01:56] trukulo: we are wondering what to write in each file [01:57] pitti: No worries. I've promised sb some serious cdbs love for UDU, perhaps if our slave drive^W^Wmdz gives us some time I can drag you off to a corner as well to look at what I'm doing with cdbs. =) [01:57] trukulo: currently we decided to list the institution as the copyright holder [01:57] zyga, who is the intellectual property owner? you and your friend? oy did you do that work for a company/institution? [01:57] trukulo: and each/both authors as.. .authors [01:57] trukulo: that's complicated [01:57] so you need permision of everyone related in the project [01:57] trukulo: part belongs to the institue [01:57] trukulo: other part to me [01:57] jbailey: that'd rock! (we will bribe you to release cdbs2 soon :-) ) [01:57] so you need agreement of the institute [01:57] and yours and your friend [01:58] trukulo: and other to the university of my friend [01:58] pitti: The needed bribe is spare time. =) [01:58] jbailey: or, rather, make you drunk and make you sign the "I will do it in two days" contract :-P [01:58] trukulo: the university will cooperate most probably [01:58] zyga, that's ok, your friend too [01:58] and the institute? did you do that work on work-time? [01:58] trukulo: but we still don't know if we should list everyone or just the institue [01:58] or in free time? [01:58] jbailey: make it so that debian/rules contains nothing more than "#include /usr/share/cdbs/doitright.mk" :-) [01:58] zyga, if it's released in GPL [01:58] trukulo: in the beginning I worked there [01:59] you don't HAVE TO mention anyone [01:59] trukulo: then the funding stopped and techically it was in my free timne [01:59] because it's easy, if it's GPL you have no credit obligation [01:59] pitti: Nope, always has to be at least 3 lines. [01:59] just kidding :-) [01:59] BUT, you can say in README or ABOUT, thanks to Insitute and University [01:59] trukulo: so as long as the institute, the university and us agree it should be okay? [01:59] zyga, yes [01:59] pitti: #!/usr/bin/make -f , include /usr/share/cdbs2/cdbs.mk, CDBS_MODULES = debhelper autotools =) [02:00] trukulo: well there is one more issue but I'm not so sure about it [02:00] zyga, tell me [02:00] trukulo: the project was funded by the state [02:00] jbailey: ah, that's the new boilerplate? nice [02:00] trukulo: I'm not sure how does that look from the legal side [02:00] pitti: Yeah. Module ordering is handled internally. [02:00] jbailey: what about "#!/usr/bin/cdbs-make\nCDBS_MODULES= ..." then? [02:00] zyga, if university use the founds, and university let you publish it GPL, then there's no rpoblem [02:01] because university decides what to do with founds, it's not state responsability [02:01] trukulo: the university did not fund this at all, but it explicitly reserves all rights to the work published by it's students [02:01] zyga, so rights are from university, no problem at all [02:01] trukulo: the funds came from the state to the institute [02:02] rights is what matters here, intelectual property (copyrights) [02:02] pitti: Debian policy sais that it must be a Makefile. [02:02] zyga, ah, that's another history [02:02] jbailey: oh, ok [02:02] zyga, it's institute owner of the rights or state? [02:02] trukulo: we are still trying find a lawyer with sufficient knowledge [02:02] trukulo: that's still the fuzzy issue [02:03] trukulo: we think that rights belong to the institute [02:03] zyga, i'm spanish, and VERY probably, laws are different here [02:03] trukulo: but we are still not sure [02:03] zyga: Do they even know?:) [02:03] trukulo: I know but we hope that since everyone cooperates it's not a matter of if but how this gets GPLd [02:03] Pizbit: ? [02:03] but anyway, it's very UNUSUAL that state told you not to publish that work in GPL [02:04] zyga: Does the institute know if they have the rights or not? [02:04] Pizbit: the institute is composed of scientists ;-) [02:04] Pizbit: we simply don't know for sure ;-) [02:04] Surely someone to be in charge heh [02:04] zyga, probably the state don't neither [02:04] :) so try to release as gpl [02:05] Pizbit: most people are old and while very inteligent they have very little legal knowledge [02:05] and if there is problem (very rare) contact a lawyer [02:05] we are trying to do that ATM [02:05] zyga, that's my point of view, but i'm talking from spanish legislation [02:05] I hope this is over soon though [02:06] trukulo: thanks [02:06] zyga, you're wellcome === eruin [~eruin@eruin.user] has joined #ubuntu-devel [02:08] http://www.suxx.pl/poliqarp/license.txt [02:08] we plan to prepend this to each file [02:10] zyga, what is for ? [02:10] you don't explain it.. [02:12] trukulo: as I said we plan to prepend that to each source file [02:12] trukulo: the project details can be found at www.korpus.pl (there's an english version) [02:12] thanks === mvo is away for a couple of minutes to go to the bank [02:15] you can get to the bank and back in 2 minutes? [02:16] Lathiat: must be one of those brodband folks ;) [02:16] broadband even [02:18] sjoerd: ping === Amaranth goes to bed === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel [02:30] pitti: i'm going to bed now, but i think we should have a reasonably smooth transition with the packages I just put up on p.u.c/~daniels/dbus/. didn't change the version number. [02:30] daniels: cool, will try them out. [02:30] daniels: good night! [02:31] night dude [02:31] daniels: yeah, good time to go to bed :-) sleep well [02:31] so you are 8 hours ahead of us [02:32] <\sh> http://www.ubuntulinux.org/wiki/LocalAptGetRepositories my experiences from last night to this morning...summarized on the wiki === doko [~doko___@dsl-084-059-052-103.arcor-ip.net] has joined #ubuntu-devel [02:45] mvo: ping === gustavor [~gustavo@200-171-140-32.dsl.telesp.net.br] has joined #ubuntu-devel === tle [~tle@220-244-212-78-vic.tpgi.com.au] has joined #ubuntu-devel [02:48] long time no see ppl === Safari_Al [~tr@ppp141-226.lns1.adl2.internode.on.net] has joined #ubuntu-devel [02:49] I found something very interesting with bug #5917 [02:49] hey Safari_Al [02:49] Ubuntu encounter low res 640x480 with all PC using i810 card [02:50] jdthood, Hi! [02:50] erm, jdub, hi! [02:50] thanks for the email today. [02:51] it'd be nice if u can have a look on it, danield, [02:53] jdub, gonna miss the 1st half of gnome.conf.au :/ [02:53] Safari_Al: d'oh! [02:53] jdub, I'm talking at the edulinux miniconf, so that will be good exposure anyway and good fun too I hope === moyogo [~moyogo@69.156.166.86] has joined #ubuntu-devel [02:57] zyga: pong === mxpxpod [~bryan@wuw-ojr3gmca.dybb.com] has joined #ubuntu-devel [02:58] daniels: pong [02:58] Mithrandir: I think it avoids Unicode for the installation itself, not sure about the second stage, that may be buggy [02:58] mjg59: pong [02:59] Kamion: the installer? it uses utf8 on serial console with the slang frontend [03:00] Mithrandir: probably a bug then, feel free to diagnose :) [03:01] Kamion: right now I'm trying to undo some of the damage I've unleashed with new pkg-config versions [03:03] right now I'm unfreezing my upload queue :) === fgx [~fgx@barlach.spin.it] has joined #ubuntu-devel [03:06] mvo: have a look at www.suxx.pl/update-manger/patches [03:07] zyga: nice, thanks [03:09] mvo: I'll remove the old ones === bradb [~bradb@MTL-ppp-144647.qc.sympatico.ca] has joined #ubuntu-devel [03:10] mvo: some thing got mixed in update-manager.in [03:11] mvo: package version stuff especially [03:11] mvo: but as long as you like them they're safe IMHO [03:15] zyga: I'm looking over it now === zul [~chuck@198.62.158.205] has joined #ubuntu-devel [03:17] yo [03:20] seb128: ha ha, clearlooks added to gtk-engines ;) [03:20] what is a.u.o/bigfile? === msturm [~msturm@t-20-214.athome.tue.nl] has joined #ubuntu-devel [03:21] Kamion: I'm just about go get lunch, but can I talk to you about custom CD images at some point? [03:22] jdub: please update gtk-engines :p [03:22] oh fun [03:22] seb128: evil man :) [03:22] mjg59: yeah [03:22] :-P [03:23] I've managed to make a GSList pointing to itself [03:23] Mithrandir: is that hard? [03:24] Treenaks: not really, but it consumes about 250MB/sec of memory when you try to copy it. [03:24] Mithrandir: *headdesk* [03:24] I guess anybody trying to build epiphany with pkg-config 0.17.1 is up for a surprise. [03:27] hmm [03:27] where did our models go? [03:27] april calendar is one big blue goo ;-) [03:27] zyga: with hands! === bradb_ [~bradb@MTL-ppp-144647.qc.sympatico.ca] has joined #ubuntu-devel [03:28] Treenaks: are you suggesting that next month they'll make it green with feet instead? [03:28] zyga: who knows === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel === UbuntuGet [~gustav@gw5.web.thalamus.se] has joined #ubuntu-devel === lamont discovers a lovely timebomb waiting to embrace him in the livecd rootfs build... [03:59] boom [03:59] every 3 weeks or so, we build a livecdfs that wants to fsck at boot time... every time... Gonna have to fix that... [04:00] why every three weeks? [04:00] well, was watching things scroll by, and it said every 21 mounts [04:00] lamont: why don't you just tune2fs -c 0 -- if its ext2? [04:00] since we reuse the image, etc.. [04:00] Lathiat: that'd be the fix, yes. [04:01] ah === Kamion wonders just how enormous the debian-installer merge is going to be === lamont thinks more, realizes just what an fsck of the cloop would do to the ramdisk, giggles more. [04:02] lamont: eh? [04:02] rsyncability death? === Lathiat wonders where to find dbus hackers [04:02] Lathiat: #dbus ? [04:03] Treenaks: no #dbus [04:03] Kamion: nah - I expect it'd touch at least something pretty frequently, resulting in a things moving over [04:03] they must have caught d-bus and buggered off :) [04:03] Lathiat: you could poke sjoerd [04:03] Lathiat: or pitti [04:03] i just want to know if theyve done implementing objects in the glib api yet [04:03] then again, it shouldn't actually need to change anything - just read the whole cloop... depends on how smart the cloop code is on not exhausting memory [04:04] Lathiat: ross bloged about that recently IIRC [04:04] Lathiat: no idea -> daniels [04:04] more to the point, in 0.23 [04:04] mvo: burton? === ogra [~ogra@p5089C0BB.dip.t-dialin.net] has joined #ubuntu-devel [04:05] ah, so its in 0.32 [04:05] oh well === decko [~decko@decko.user] has joined #ubuntu-devel === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel === Baby [~nena@baby.kavi.silver.supporter.pdpc] has joined #ubuntu-devel === wasabi_ [~wasabi@207.55.180.100] has joined #ubuntu-devel [04:20] thom: please restart the torrent tracker once it gets the Kubuntu DVD torrents that are currently syncing === thoreauputic [~debianarc@wolax7-001.dialup.optusnet.com.au] has joined #ubuntu-devel [04:25] sjoerd: ? === jsgotangco [DaWorm@dialup-222-126-78-215.infocom.ph] has joined #ubuntu-devel [04:28] maswan: hmm, could we clean out a bit on ravel? [04:33] aaaw...http://www.about-linux.com/ubuntu_1-5.html [04:33] a video tutorial on "how do i remove all nice ubuntu advantages in warty" [04:35] warty video tutorials online..that is nice [04:35] Mithrandir: yes. [04:36] maswan: I'm cleaning out about 800M of packages from pure64 at least [04:36] Mithrandir: I'll go clean some of my stuff up. [04:36] some of the -mw$n stuff isn't needed. [04:36] maswan: great [04:37] hmm.. actually, none of it is, because the only reason I kept many aroudn was that they had different state (mostly wrt multiarch), but since I have forgotten what states were involved... [04:38] heh [04:40] I think the same goes for the pure64-mw variants, unless you know of anything different there? === maswan also remembers to umount the bindmounted /home before rm -rf:ing ;) [04:40] haha [04:41] maswan: no idea :) [04:41] yo === jamin [~jamin@cust-24.241.99.54.cbnstl.net] has joined #ubuntu-devel [04:42] Lathiat: I've done that before on that machine. Luckily my $HOME was first, and I had a bunch of unpacked kernel trees, so I managed to break it before I clobbered Mithrandir's stuff etc. :) [04:46] elmo: ping? [04:47] fabbione: ? [04:47] elmo: would it be possible to get sparc.u.c going? [04:47] a 4 times/day pulse would rock [04:47] otherwise i cannot even start breezy [04:48] fabbione: I'll do ports when I can, but atm getting breezy syncing is a higher priority [04:49] elmo: yes i really understand that, but i can't even prepare the buildd otherwise.. can you give me at least one pulse? [04:49] fabbione: done [04:49] elmo: thanks [04:54] <|QuaD-_> ls === lamont back in a while - errands in town === Simira [~simira@56.80-202-210.nextgentel.com] has joined #ubuntu-devel === ironwolf [~ironwolf@c-24-6-169-124.hsd1.ca.comcast.net] has joined #ubuntu-devel [05:07] maswan: oops :) [05:09] Lathiat: the lesson learnt is to always keep half a dozen kernel trees in your $HOME, it gives you time to think and break that rm -rf before it takes out the whole filesystem :) [05:10] rm needs one of those flags not to traverse file systemws [05:10] i thought gnu rm had one === ross [~ross@82.133.103.84] has joined #ubuntu-devel [05:11] daniels: ping? [05:12] so the hoary hardware device database tool is broken for me [05:12] ross ? [05:12] it tried to send to the server, failed to connect. it then said it would save a file to disk, and didn't [05:12] whoops, is it up to date ? [05:12] a fresh hoary install [05:13] should be 0.6-ubuntu2 iirc [05:13] no upgrade from debian or such ? [05:13] nope, i wiped my disk [05:13] hrm [05:13] i did purge a load of python modules i don't use so it may be incomplete deps [05:13] <\sh> hey ogra, finally u awak ;) [05:13] could you run hwdb-gui from the commandline ? [05:13] <\sh> +e [05:14] ross: does that mean you have hoary runing on one of your machines? *congrats* then :) [05:14] mvo: yeah, nice to see g-a-i in its native environment :) [05:14] heh [05:14] mvo: and this time i used lvm so i can make space for a new install later [05:14] looks cool now (since it has some apps in the tree) [05:15] i had a few install issues, i'll have to hit bugzilla later [05:16] mvo: hey [05:17] zyga: hey [05:17] hi \sh [05:17] mvo: I think that the easiest approach is to simply add debian [05:17] Mithrandir: better? [05:18] maswan: a lot, thanks :) [05:19] ross: Are you planning on doing work on g-a-i? Wondering because I'm working on splitting the different parts (ie apt/synaptic handling) in to separate classes and files. [05:19] mvirkkil: how do you determine current icon theme? [05:19] morning [05:19] hi mdke [05:20] hi mdz [05:20] grr [05:20] hey mdkz [05:20] zyga: I'll send you a mail then [05:20] grr [05:20] mvo: great, thanks [05:20] mvirkkil: would make sense, in the future we may go with python-apts interface [05:21] mvirkkil: yes, i don't plan to drop it [05:21] zyga: I haven't done any changes to that part. 't uses: self.icons = gtk.icon_theme_get_default() [05:21] mvirkkil: thanks, I'm not familiar with gtk :) [05:21] <\sh> 2 days off from work, and more work in my sparetime...I'm ill [05:21] ross: I mean that it will take a few days and will include large and invasive changes (again) [05:22] zyga: Neither am I :-) [05:22] next week i'll probably review all of the patches i've been sent and merge any hoary changes [05:22] ross: So merging it with other changes will be painful. [05:23] either wait a bit or be prepared for conflicts [05:24] ross: I guess I'll be prepared for conflicts. The changes aren't that large compared to the ones I've already made (the large patch in bugzilla). === dholbach [~daniel@p54A634FE.dip.t-dialin.net] has joined #ubuntu-devel [05:25] zyga: What are you woking on if you don't mind me asking? [05:25] hi [05:25] dholbach: hi [05:25] hi dholbach [05:25] hey :-) [05:25] hi dholbach [05:25] <\sh> dholbach: I set myself on the list for wannabe new member of ubuntu [05:26] \sh: What's that? [05:26] who does admin-type stuff at Canonical these days? (like if I need to get an employment verification letter showing they employed me last year) Is that still Jane? [05:26] \sh: yeah... already saw :-) [05:26] mvirkkil: I wan to add pretty user friendly icons to u-m based on desktop files of existing applications [05:26] mvirkkil: and learn python + gtk at the same time [05:26] <\sh> dholbach: finally, i could bash ogra all the day, just because he forced me to play with ubuntu ;) [05:27] :-) [05:27] \sh, nice wikipage [05:27] zuga: Cool. Check out the icon function in http://cvs.gnome.org/viewcvs/gnome-app-install/src/AppInstall.py?rev=1.2&view=markup (should prbably be called something like getIcon and not just icon) [05:27] <\sh> mvirkkil: http://www.ubuntulinux.org/community/processes/newmember [05:27] <\sh> ogra: the howto? [05:27] yep [05:27] mvirkkil: cool, thanks [05:27] justdave: Jane or Claire, I think; probably just send it to Jane and she'll forward it. [05:27] <\sh> ogra: check for wrong spelling ;) [05:28] \sh, thats better done by a native english speaker ;) [05:28] Mithrandir: thanks [05:28] ross: How many pathces do you have for g-a-i? [05:29] mvirkkil: just a few pending [05:29] <\sh> ogra: hehe for sure ;) but i tell you...my eyes are small, my brain is dead, but I want to finish my vnc2swf...if this is working...we can do a lot of stuff for ubuntu, showing desktop sessions via flash movie...this will be the hell :) === martink [~martin@pD9EB3C6D.dip0.t-ipconnect.de] has joined #ubuntu-devel [05:29] \sh, doesnt really convince the amd64 user here :) [05:30] \sh, (no flash for amd64) [05:30] ogra: you are very right :) [05:30] <\sh> ogra: well...for promotion is x86 enough ;) [05:30] ross: Do you do your work in the gnome cvs? (ie how do I keep up to date?) [05:30] <\sh> ogra: even not with ming? [05:30] <\sh> oh ming i prepare just here ;) [05:30] mvirkkil: yeah, gnome cvs [05:30] Kamion: ping, re: breezy germinate [05:31] \sh, dunno, i guess with a lot of fiddeling i'd get it to work.... but i dont think flash is worth that [05:32] <\sh> ogra: u have a x86 box at home? check http://www.unixuser.org/~euske/vnc2swf/ [05:32] <\sh> ogra: complete sessions with live sound explanations etc. [05:33] ogra: You can swf2avi, avi2gif. [05:33] \sh, http://www.about-linux.com/ubuntu_1-5.html same like this.... [05:33] (Which is how I did http://www.inference.phy.cam.ac.uk/dasher/images/newdasher.gif .) [05:33] \sh, which makes me sad .... [05:33] <\sh> ogra: try with gplflash it's amd64 compatible [05:33] \sh, since it only shows how to rip out the nice ubuntu imrovements out of your system [05:34] \sh, i'm really not interested in flash [05:34] ross: I'd be interested in knowing what the patches are for. Mind sending me them? [05:35] <\sh> ogra: no..much nicer...u see a direct session from your desktop :) vnc to your box and record it ;) [05:35] What we *really* need in an X server plugin that uses damage to record changes to sessions in a really bandwidth-efficient way and allows us to play them back and export them and all kinds of useful stuff. [05:36] cjb: doesn't need to be a plugin, a normal app could do that [05:36] that was the point of damage :) [05:37] ross: Well, a bit of both. I guess the perfect way to do this is to have a compositing manager do the recordings. [05:38] (Since it speaks Damage, will know when to take snapshots to end up with a consistent state, can avoid repainting as much as possible.) === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel [05:42] MITHRANDIR [05:43] what have I done now? [05:43] except breaking pkg-config, but that's being fixed now [05:44] maswan: run dmesg on ravel [05:45] elmo: pong? [05:45] Mithrandir: you uploaded mozilla-thunderbird without an ubuntu version number and with a different .orig.tar.gz from debian's [05:45] elmo: argh :( [05:46] elmo: what can I do to fix the damage? (except not doing it again) [05:46] (if anything) [05:47] WOW! http://www.ubuntulinux.org/wiki/WhyYouShouldntUseUbuntu Someone is spamming/trolling. === astharot [~isager@50e67c856007b2ee.node.tor] has joined #ubuntu-devel [05:47] Mithrandir: I'm not sure atm [05:48] mvirkkil: at least somebody with a broken caps lock key [05:48] mdz: pong [05:49] Mithrandir: And someone who could use a good beating with the clue-stick :) [05:49] mdz: what do you need? I switched the default RELEASE/DIST earlier today [05:49] <\sh> hmmm. [05:49] <\sh> well. [05:49] <\sh> anti-white? [05:49] <\sh> mark is not a coloured [05:49] <\sh> so...anti-white? === nice-19 [~stefan@www.wintermeyer.de] has joined #ubuntu-devel [05:49] <\sh> finally anti-buur [05:49] <\sh> but not anti-white [05:50] deleted [05:50] mdz: oh, I guess you want output on people.u.c/~cjwatson/germinate-output/ too ... doing [05:50] <\sh> thx... === zenrox [~zenrox@wbar7.sea1-4-4-043-090.sea1.dsl-verizon.net] has joined #ubuntu-devel [05:50] dholbach: Good. Did you block the ip? Also notice the link on the front page. [05:50] oh didnt :-/ hrm [05:51] apt-cache show whiptail - Whiptail is a "dialog" replacement. but it does not provide "dialog" - I think it should include ln -s whiptail dialog - yes? should I bugzilla this? [05:51] dholbach: Hmm... I can still see the page. [05:52] dholbach: Nevermind, the page is gone now [05:52] i cant, but the search returns taces.... [05:52] traces [05:52] mvirkkil: i'm no wiki admin, i can't block ips [05:56] CarlK: no, it should not provide /usr/bin/dialog; that would unnecessarily cause the packages to conflict, and I'm not even sure that they're 100% command-line compatible [05:56] ciao [05:57] mdz - should 'something' provide dialog? [05:57] CarlK: the dialog package does [05:58] Kamion: I'm trying to diagnose why anastacia wants to move a few hundred packages to universe [05:58] Morning mdz [05:58] mdz: it appears in Warty at least, the dialog package doesn't exist [05:59] mdz: I diffed hoary/breezy germinate outputs, they were identical apart from kvim [05:59] thoreauputic: it's in universe [05:59] CarlK: whiptail is not 100% command-line compatible with dialog [05:59] mdz: not here it isn't [05:59] debconf has code to do slightly different things depending on which one it finds [05:59] mdz: I got cron.sync minimally functional, it's not really live yet [05:59] thoreauputic: in the Ubuntu archive it is [05:59] please leave it alone :) [06:00] elmo: ah [06:00] mdz: oops, I beg your pardon - it is here after all, sorry [06:00] for example whiptail has --scrolltext, dialog doesn't [06:00] the whiptail idea was just because i couldn't find dialog, and the description said it would work ;) [06:01] ok, us lusers will go back to #ubuntu ;) [06:01] CarlK: mdz is right - I must have done a typo [06:01] no prob [06:05] jbailey: did you remove the ppc64 glibc binaries from p.u.c.? [06:05] doko: I did - I realised that they didn't have the libc.preinst hook to make sure that there was a 2.6 kernel running. [06:05] doko: Want another set? I can push them up now. [06:06] I was just about to tweak the l-k-h dependancy to 2.6.11.2-0ubuntu1 instead of to -1 [06:06] (Having watched elmo lart someone for that mistake just moments ago... *g*) [06:11] meh 2.6.11 ?? === trulux starts Breezy business [06:11] fabbione: Will that cause you grief? [06:11] fabbione: could we talk on the Breezy goals for the kernel packages? [06:11] fabbione: after it I will say you rcok :D [06:11] fabbione: Generally the defines are the only interesting bit. glibc always rolls back to previous syscalls. [06:12] fabbione: l-k-h isn't part of your packages though ... [06:12] jbailey, Kamion: ah ok [06:12] fabbione: Upstream recommends that you always use the recent kernel headers even if you're running an older kernel. [06:12] trulux: why don't you subscribe to kernel-team and discuss what has been asked teice already ? ;) [06:12] jbailey: make sense [06:12] s/teice/twice [06:13] fabbione: ok [06:13] jbailey: could we just install the current packages in the ppc64 chroot, and work from this state on? [06:13] trulux: there is a team behind.. i want people to discuss all together [06:13] doko: also, could we talk on the gcc packages? [06:13] ok, sorry [06:13] I like team work, so, it's better even [06:13] :) [06:14] i have the feeling that elmo just imported sid... === thoreauputic_ [~debianarc@wolax9-007.dialup.optusnet.com.au] has joined #ubuntu-devel [06:14] I've been trying all day [06:14] doko: Better to give me 20 minutes. The version of lkh that I'll be uploading will be a lower version that I was using before (-0ubuntu1 instead of -1), so it would never get rolled back. [06:15] elmo: I don't need to run the testsuite on it anymore, though. [06:16] elmo: well you succeed to some extents i gues... [06:16] who is in charge of managing the meetings? [06:17] trulux: you can announce one yourself as long as it doesn't clash with an existing one [06:17] bbl [06:18] jbailey: sounds fine [06:18] pitti: ok, great [06:18] trulux: did you want to talk about the SSP patch for gcc-HEAD ? ;) [06:19] trulux: please look in #ubuntu-meeting for the already scheduled ones and add yours [06:19] trulux: and announce the meeting on u-devel@l.u.c [06:19] doko: about *many* stuff, I have created the breezy chroots and testing stuff [06:19] ok [06:19] trulux: SSP for gcc-HEAD would really *rock* :-) [06:20] pitti: I'm porting it to 4.0, as I haven't got response from Etoh [06:20] nor Yoda === decko [~decko@decko.user] has joined #ubuntu-devel [06:20] mdz: cron.sync is fixed [06:20] so, following Yoda's saying... "don't try to do it, do it or just don't do it" [06:20] ;) [06:20] elmo: thanks [06:21] hey mdke [06:21] erm [06:21] mdz [06:21] :) [06:21] mako: nothing from Amaya (yet) [06:22] elmo: I seeded openoffice.org2 yesterday, but anastacia doesn't want to promote it [06:24] that'd be a mirroring problem, I'll investigate [06:24] mirroring of the seeds? ah [06:24] why the hell is baz update checking every signature ever. grr. [06:24] my fault, incomplete script [06:25] was only baz updating warty hoary kubuntu-hoary [06:25] mdz: try again [06:26] trulux: ok [06:26] trulux: send her another ping [06:26] mako: ok === trukulo [~trukulo@176.red-62-57-69.user.auna.net] has joined #ubuntu-devel [06:28] doko: so what is the toolchain bootstrap sequence for the buildd? [06:28] ok.. so i'm hearing that ubuntu was discussed on us television last night [06:28] national television [06:29] Neat. In a positive light? [06:29] 18:26 <@micah> mako: apparantly last night on the TV show "Veronica Mars" two of the main characters had an argument about which is better, OS X or Ubuntu [06:29] 18:27 <@mako> micah: WHAT? [06:29] 18:27 <@micah> the argument was as if you and I had the argument, but on a mainstream TV show ("But Ubuntu is Free Software" "Yeah, but it always breaks!") [06:29] mako, any video avalaible on internet? [06:29] 18:27 <@micah> mako: seriously... I was just told that [06:29] i just heard about this [06:29] Ahah [06:29] fabbione: do we need one, as long as we don't change the g++ default? [06:29] can someone verify this [06:30] doko: well, what does need to be builded and installed in the chroot first, before start batching the builds? [06:30] doko: there are already upload queued.. including gcc-4. [06:30] doko: so i guess we need that in as first... === tritium [~mrimbert@12-202-90-180.client.insightBB.com] has joined #ubuntu-devel [06:31] yes, this one and glibc first [06:32] I've just uploaded linux-kernel-headers, since it doesn't need any furhter love AFAICT. === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel === tortoise_ [~tortoise@81-86-196-118.dsl.pipex.com] has joined #ubuntu-devel [06:42] $ debian/rules `pwd`/stamp-dir/patch-stamp [06:42] meh [06:42] jbailey: is there an easier way than the above to just unpack/patch the glibc source tree? [06:43] Kamion: debian/rules patch [06:45] what's the release number for Breezy? [06:45] trulux, 5.10 [06:46] trukulo: many thanks, sir [06:46] :) [06:46] trulux, it's easy: year.month (5.10) [06:47] jbailey: tried that [06:47] yep, I'm just about to test a game that a friend gave to me [06:47] jbailey: oh, bah, maybe I didn't, OK. :) [06:47] if something doesn't stop me to have fun for a while [06:48] I hate that particular guessing game [06:48] Kamion: Oh good. I was fairly certain I had used it this morning. =) [06:49] I tried 'setup', 'source.make', 'patched-source' [06:49] then gave up and started trying to read the rules files [06:49] YEah. setup used to be in there, I'm not sure why it's not now, but I haven't hunted it down. === mvo is away, bb in 2h === Frafra [~frafra@d83-176-65-202.cust.tele2.it] has joined #ubuntu-devel [06:53] hi [06:54] exist ubuntu for sparc and/or ultrasparc? [06:54] Frafra, no [06:54] Frafra: yes [06:54] heh [06:55] it will be available during breezy release cycle [06:55] fabbione, i think not, but you know better than me :P [06:55] trukulo: i am doing the port.. there is already an archive of packages [06:55] fabbione, ah! then it doesn't exist !! it will exist [06:55] 5.10 for ultrasparc? [06:55] don't confuse me, man [06:55] but it's not good enough yet to be sent around [06:55] Frafra: i hope that will be a reality.. yes [06:56] :D [06:56] good [06:57] can i download the unstable version? [06:57] Frafra: there are no iso.. only netinstall.. if you can wait a few days.. yes [06:57] :D [06:57] intresting... :) [06:58] we are moving it to a new (and slightly more powerful) server [06:58] before making the port available === srbaker [~srbaker@blk-137-77-155.eastlink.ca] has joined #ubuntu-devel === doko [~doko___@dsl-084-059-052-103.arcor-ip.net] has joined #ubuntu-devel === doko [~doko___@dsl-084-059-052-103.arcor-ip.net] has joined #ubuntu-devel === tritium [~mrimbert@12-202-90-180.client.insightBB.com] has left #ubuntu-devel ["Leaving"] === HiddenWolf [~hidden@136.121.dynamic.phpg.net] has joined #ubuntu-devel [07:19] doko: The new glibc is in http://people.ubuntu.com/~jbailey/glibc/ sorry about the lag. [07:20] elmo: ROCKING! === fabbione hugs elmo a lot === theine [~theine@fw2.nbi.dk] has joined #ubuntu-devel [07:22] elmo: how is it going with the syncs? [07:24] so, breezy development has started? === astharot [~isager@a4a929587ef21b99.node.tor] has joined #ubuntu-devel === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel [07:29] jbailey, elmo: so we can set up the chroot? [07:30] doko: I'm fine with it now. lkh hasn't made it through the buildd yet, so if you want I can upload a ppc deb for that too. [07:31] jbailey: that would be nice ... [07:31] jbailey: hey === blueyed [~daniel@83.135.29.255] has joined #ubuntu-devel [07:34] mxpxpod: Hullo [07:34] doko: http://people.ubuntu.com/~jbailey/linux-kernel-headers/ has it now. === trygvebw [~trygvebw@217-131-213.adsl.tele2.no] has joined #ubuntu-devel === ska-fan [~ska-fan@dsl-082-082-198-136.arcor-ip.net] has joined #ubuntu-devel [07:50] jbailey: will the linux-headers-2.6.11 stuff from breezy work for hoary? [07:50] also, are there going to be minor updates to hoary for like metacity 2.10.1 and such? === Rickdangerous [~semaphor@dsl81-214-61756.adsl.ttnet.net.tr] has joined #ubuntu-devel [07:56] re [07:58] mxpxpod: Define 'work for'? [07:58] mxpxpod: And no, probably not. [07:58] jbailey: will I be able to compile a kernel-image [07:58] jbailey: I'm trying to make sense of the huge pile of Java packages that want to move from universe to main; can you take a look at it? [07:58] mxpxpod: With a short release cycle, there's not enough time to be maintaining a backports setup. [07:59] jbailey: ah, ok [07:59] so, my best bet would be to switch to breezy ;) [07:59] mxpxpod: l-k-h is completely unrelated to the kernel builds itself. The package will probably be renamed to linux-libc-headers at some point to avoid confusion. [07:59] jbailey: ah, ok [08:02] jbailey: is there a "simple" way to take a vanilla kernel and put the ubuntu patches (like wlan-ng) into it? [08:02] maswan? [08:03] mdz: Are you looking at JavaPackagingProgress: [08:03] ? [08:04] mxpxpod: Dunno. I try to stay on this side of the syscall layer. =) IIRC, patches are maintained as .dpatch or something like that - There ought to be some way to easily replace the upstream version. [08:04] jbailey: no, see the email I sent === Baby [~nena@baby.kavi.silver.supporter.pdpc] has joined #ubuntu-devel [08:04] jbailey: according to germinate, those packages ought to be in main, and not universe, but there's far more than I expect to see [08:05] and some of it we definitely don't want in main === lionel_ [~lionel@ip-128.net-82-216-65.rev.numericable.fr] has joined #ubuntu-devel [08:08] mdz: This is my first encounter with germinate. How does it produce this list, missing build-deps and depends for packages that are in main now? === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel [08:11] jbailey: germinate takes as input the seeds and the Packages files [08:11] jbailey: it walks the depends and build-depends trees, and outputs lists which incorporate dependencies === milli [~milli@phantom.acmeps.com] has joined #ubuntu-devel [08:11] jbailey: so the output of germinate for the 'desktop' list is all seeded packages, plus all their dependencies (recursively) [08:11] Also build-deps? [08:11] jbailey: all build-dependencies are added to the 'supported' output\ [08:12] Ah, cool. [08:12] jbailey: elmo has a tool which basically diffs that against the state of the archive [08:12] I mailed you part of the output from that tool === fabbione plays around [08:13] I assume that some of it is related to openoffice.org2, but we had openoffice.org2 in main for part of the Hoary cycle without it [08:13] mdz == sabdfl? [08:13] and it seems like a huge amount of stuff (kaffe AND ecj, 4 different jikes packages, jython, etc.) [08:14] ska-fan: no, sabdfl=sabdfl [08:14] or mdz=mdz [08:14] yeah [08:14] s/or/and/ [08:14] ==, depending which language your brain parses :) [08:14] ok, thanks. No idea how I got that impression :) [08:14] mdz: The source packages is basically what's in Hoary at the moment, or is this from the sid merge? [08:15] mdz: 2.6.12rc2 is basically ready to go in as update from hoary. We are working on parsing bugzilla for the extra drivers before any upload to see what is sane to pull in before UDU [08:15] fabbione: can i give it a test run? [08:15] Lathiat: there are no binaries around sorry [08:15] jbailey: this is breezy [08:16] jbailey: (which is basically hoary, plus a few uploads over the past few days) === decko [~decko@decko.user] has joined #ubuntu-devel [08:16] ska-fan: sabdfl has more hair [08:16] fabbione: trivial to build from source isn't it? (i assume kernels are no different?) [08:16] Lathiat: + the kernel can still change from what will hit breezy and the aBI too [08:16] Lathiat: it's not trivial without the first .diff.gz and the .dsc and the .orig around :) [08:17] heh ok [08:17] Lathiat: i will ping you as soon as i have something builded [08:17] fabbione: ok :) [08:17] well actually more than you... but you get the idea [08:18] jbailey: the 'all' file has a column which shows which package pulled in each package [08:18] fabbione: yuh :) [08:18] jbailey: or if it was listed in a seed explicitly [08:18] Lathiat: it also depends what flavour you need [08:18] mdz: Ooo, thanks. [08:18] jbailey: the rdepends/ directory contains a tree showing all the reverse depends for each package [08:18] Lathiat: i am not going to get too crazy uploading outside the archive [08:18] 686 (or 386) [08:18] Lathiat: ok [08:18] not hassling, jsut thought i'd could give it a shot and see how it goes :) [08:19] jbailey: e.g., python2.1 is being pulled in by jython, via libbsf-java via ant [08:19] Lathiat: inotify works, if that's what you want to know :) [08:19] jbailey: we won't support python 2.1, so we need to break that chain somewhere [08:19] fabbione: i was trying to avoid asking :) haha [08:20] mdz: Yup. Like I'm sure xerces-j doesn't actually need a build-dep on ash. [08:20] fabbione: but i was just more generally interested in testing it and seeing if theres any problems === fabbione has some kind of mental control [08:20] Lathiat: i am building a 686 image right now.. not sure if i will manage to upload it before my wife is back [08:20] ok [08:20] and it has some experimental stuff that might not go in breezy at all === Seveas [~seveas@seveas.demon.nl] has joined #ubuntu-devel [08:21] so don't rely on them [08:21] whatever, when you have time, i'm not fussed :) [08:21] testing is good.. but i like to set people mind first [08:21] nps [08:22] it's a 12rc2 + extra junk + extra royally untested crack [08:22] it can hurt [08:23] yeah, lets wipe our disks ... and burn the chipsets ;) [08:24] fabbione: i bet it can :) [08:24] i'll keep both pieces, don't worry :) === alexandrrr [~alexandrr@201009180092.user.veloxzone.com.br] has joined #ubuntu-devel === maskie [~maskie@196-30-109-220.uudial.uunet.co.za] has joined #ubuntu-devel [08:31] doko: still here? [08:31] yep, my flight is tomorrow evening ;) [08:32] doko: okay, so things keep changing and I've lost track - what do you want installed in the chroot? === [EVO] [tankabbot@host217-42-137-224.range217-42.btcentralplus.com] has joined #ubuntu-devel [08:33] davis:~doko/powerpc64/install/* + jbailey's glibc packages [08:33] where are jbailey's pkgs? [08:34] geez, we were almost 2Gb out-of-sync with sid, source-wise [08:34] elmo: http://people.ubuntu.com/~jbailey/glibc/ [08:34] and that's the unmodified packages [08:34] elmo: I can move the specific ones to a server if you'd like. [08:34] (just need to know where) === theine [~theine@fw2.nbi.dk] has joined #ubuntu-devel [08:35] elmo: copied in the same dir [08:36] plus glibc & gcc-4.0 build-deps [08:37] doko: is that for ppc64 kernel? [08:38] fabbione: This is to get ppc64 capable toolchain into the archive. [08:38] yes, you can start kernel baking, but please let me build gcc-4.0 biarch first ... === [EVO] [tankabbot@host217-42-137-224.range217-42.btcentralplus.com] has left #ubuntu-devel [] [08:40] mdz, jbailey: I hope jython isn't a problem for breezy anymore. there's at least some work done to implent something equivalent to python2.3 [08:41] argh [08:41] seb128,doko: as part of moving to oo.o2 in desktop, should we add oo.o2-evolution to desktop? [08:41] someone pretty pretty please fix findutils to not trawl bind mounts [08:41] doko: i won't start anything until you will tell me that gcc & co is go [08:41] davis is crawling so many copies of /home it's not finishing in 24 hours [08:42] mdz: no real opinion on it, I don't know what this package does [08:42] Description: Evolution 2 Addressbook support for OpenOffice.org [08:42] mdz: yes, seb128: it accesss the addressbook [08:43] oh, it uses eds [08:43] right === LinuxJones [~LinuxJone@blk-222-206-208.eastlink.ca] has joined #ubuntu-devel === jbailey grabs food. [08:49] elmo: does dev/ino match for the top-level directory? [08:49] mdz: actually, sorry, it's checksecurity which is causing the problem [08:50] not findutils === chris38-home2 [~Christian@82.127.81.96] has joined #ubuntu-devel [08:51] mdz: GCC, did talk with dan@debian.org about g++-3.3/gcc-4.0 compatibility, it looks like there are no problems (except hppa and maybe sparc). But with this combination we add a third variant to the existing and our target toolchain, so the tests/builds have limited value. So either stick with the current defaults until UDU and do the gcc/g++ switch at the same time, or do it now (which is sub-optimal, because many people are in vaca [08:51] tions next week). [08:52] a majority of our packages are C programs; there is definitely value in getting them built with gcc-4.0 [08:53] if we switch after the sync, we'll need to do source uploads of every C program to get them rebuilt [08:53] mdz: note that Andreas Jochens has been building the gcc-3.4 archive for amd64 in Debian with gcc4 for quite a while, so bugs should be filed in Debian for most issues already. [08:53] mdz: why? [08:53] elmo: er, unless you want to do thousands of binary-NMUs, or flush the archive? [08:53] mdz: why do we need to do either? [08:54] gcc-3.3 and gcc-4 built binaries/libraries (excluding c++) can happily intermix? [08:54] elmo: they can [08:54] elmo: to avoid releasing with packages built by gcc-3.3 which gcc-4.0 miscompiles [08:54] Mithrandir: (I know it was a rhetorical ? ;) [08:54] so that they would blow up in our face during a security update [08:54] elmo: I didn't know if it was. :) [08:54] elmo: but considering you maintain the packages you do, I should learn to shut up [08:54] bah; *off* [08:54] or a last-minute change before the release, for that matter [08:54] mdz: well - there are going to be packages that haven't changed since hoary in sid either [08:55] not that we ever do those [08:55] but *shrug* ok [08:55] elmo: yeah, but like two orders of magnitude less [08:55] in any event, there's still time, as breezy building seems to be down [08:56] doko: in what way do the tests have limited value? for which packages? [08:57] Mithrandir: these bugs are already imported. you can have a look at my assigned bug reports, if you find some of your packages ... [08:58] elmo: wow, checksecurity is a fucking rat's nest [08:58] yes :( [09:00] for mixed C/C++ binaries, we have a new possible source of errors, and may see/debug things that we won't release with breezy. we did a test rebuild for haory as well some weeks before the release, so that should be doable for breezy as well. [09:02] all of the mixed C/C++ stuff will be rebuilt as part of the C++ transition, no? [09:02] yes, we have to. [09:03] let's go with gcc-4.0 and g++-3.3 then, before we start syncing sid to breezy. I'd rather have an uncomfortable situation with a few packages, than an uncomfortable situation with many packages [09:04] if strange things happen, test builds with gcc-3.3 are easy to try [09:04] sorry, why are we not just doing g++-4 now? [09:04] so we go with gcc-4.0, gobjc-4.0, g++-3.3, gcj-3.3, gpc-3.4 and g77-3.4 as default ? [09:04] is it possible that ubuntu will become incompatible with debian? (as heard on slashdot) [09:05] elmo: man power during the next week, and maybe no final 4.0 release yet [09:06] mmk [09:07] zyga, what do you meen "incompatible"? [09:07] E: Build-Depends-Indep dependency for gcc-4.0 cannot be satisfied because no available versions of package doxygen can satisfy version requirements === elmo looks at doko [09:08] hmm, IWBNI if apt-get had a "do the best you can" for 'build-dep' [09:08] we do have doxygen 1.4.0 ... [09:09] it's 1.4.1 [09:09] doesn't matter, I'll do it manually as best as breezy can [09:09] (it's 1.4.1 in the build-deps, I mean) [09:09] kent: not sure what exactly really, something like many rpm based distros are not compatible and packages built for one will fail on other [09:11] ahh, ok, I remember, yes there are some formatting fixes for the docs, but that should not hurt now ... I'll downgrade the build-dep. === pitti can already smell the support for encrypted storage in Breezy [09:16] zyga, I would not worry if I was you. If you are not even sure what you are worrying about, then you should just relax. And as for slashdot, dont belive everything ;) There is always a bit of problem using packages built for another distro, and if problems arive, they will be solved :) And I think its more a problem for rpm-based distros since they dont have their own universe like ubuntu. === tseng|work [~ccc27a01@sls-eb20p9.dca2.superb.net] has joined #ubuntu-devel [09:17] kent: I don't believe everything I read on slashdot but I thought it would be wise to ask anyway ;-) [09:17] boggle - an interactive prompt [09:17] during an install. how very... debian. [09:19] Installing new version of config file /etc/locale.alias ... [09:19] /var/lib/dpkg/info/locales.config: line 435: syntax error: unexpected end of file [09:19] jbailey: ^--- [09:19] Preparing to replace locales 2.3.2.ds1-20ubuntu13 (using locales_2.3.5-0ubuntu1_all.deb) ... [09:20] libc6-dev depends on linux-kernel-headers (<= 2.6.11.2-0); however: [09:20] Version of linux-kernel-headers on system is 2.6.11.2-0ubuntu1. [09:20] doko: ^-- [09:20] elmo: /me checks. =( [09:23] does anybody know how to update a thinkpad's bios without windows? [09:23] elmo: Ugh, that doesn't fail the install. I must've missed it before. [09:23] zyga, Im realy not the right one to answere, but I dont think there is a need to worry. If for example, som package wont be able to be used on both Debian and ubuntu, people will make a package for ubuntu and one for debian. Its nothing that cant be solved. [09:24] kent: the last statement is flawed === jdthood [~jdthood@aglu.demon.nl] has joined #ubuntu-devel [09:25] kent: this is exactly why you have problems with rpm distros, one package for fedora 2, one for mandrake 10, and no packages for everything else [09:25] kent: I don't want to exagerate threat to ubuntu (I'm just asking) but I certainly do not agree with you on that === stratus [~stratus@200.198.184.97] has joined #ubuntu-devel [09:26] <\sh> can I setup a pbuilder enviroment for two distributions? 1. hoary and 2. breezy? [09:27] yes === mvo [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel [09:27] \sh, yes but you can go with debootstrap, dchroot and family too. [09:28] zyga, the only problem i can see is the people who starts to whine becaus some one else have not done their job for them. And i dont meen that that person is you in any way, i just meen that it should not be a big problem to solve, and there for its not a *problem* at all. [09:28] <\sh> in this moment I thing dchroot enviroment is better? [09:28] <\sh> -g+k [09:29] kent: I agree on the whining part ;-) [09:31] zyga, but ok, you might have a point about rpm-based distros. But personally I have had no problems this fare, so i see now point in worrying until i do. :) [09:35] <\sh> stratus: well....u have the actual breezy files for deboostrap? [09:36] kent: I came to ubuntu exactly because of this === tortoise_ [~tortoise@81-86-196-118.dsl.pipex.com] has left #ubuntu-devel ["Leaving"] [09:37] \sh, no sorry. === tortoise_ [~tortoise@81-86-196-118.dsl.pipex.com] has joined #ubuntu-devel [09:38] wow, debootstrap is nice! [09:45] <\sh> ok..then from hoary chroot to breezy ;) === Folletto [~folletto@host197-116.pool8248.interbusiness.it] has joined #ubuntu-devel === Folletto [~folletto@host197-116.pool8248.interbusiness.it] has left #ubuntu-devel ["Sto] [09:51] anyone who maintains g-a-i around? [09:52] elmo: in theory, if I would upload a security update with a new package, this would land in NEW. are there any problems that could happen? [09:53] zyga: well, I hacked on it :) [09:53] mvo: why does it maintain separate .desktop files? [09:53] mvo: (and outdated too) [09:53] zyga: because it needs the information from packages that are not installed (and therefore have no dekstop file on the installed system) [09:54] pitti: should be fine - will require me to NEW it, but other than that, no [09:54] elmo: okay, thanks [09:55] mvo: any chance to pull up-to-date .desktop files from the web? [09:56] zyga: yes, a planed feature. we need a tool that can extract them automatically [09:56] mvo: well it is pretty easy to write a shell script to do that [09:57] mvo: run it nightly on some server but where? === azeem [~mbanck@ppp-82-135-13-128.mnet-online.de] has joined #ubuntu-devel [09:58] zyga: what we plan now is to extract them and put them into a deb package (g-a-i-data or something). I don't think it's very importend to have them in a seperate index file [09:58] mvo: if you want I can write you one but I'm sure that would be nothing you cannot do :) [09:59] thom: I need to edit fields.html for bugzilla on macquarie; can you find out where it is and grant me permission? [10:00] mvo: are you not afraid of one bloated binary package that will be changing over and over just because someone changed/added an icon? [10:02] zyga, that doesnt happen this easy in the supported set [10:02] zyga: I don't think it's a big issue. we don't want too many desktop files in that package, otherwise g-a-i will become to crowded. and getting the desktop files from the web will lead to new problems. we either need a index file with all packages or we need a tarball (and then we can go with a package as well) [10:02] (changing apps or icons) [10:03] ogra: good point about the supoorted set === decko [~decko@143.107.220.163] has joined #ubuntu-devel [10:04] I'm wondering wether I should use some pre-packaged icons or really extract all info from .desktop files and track their locations [10:04] there's a patch here: http://primates.ximian.com/~sandino/python-glade/#main that gives glade the abilty to create python applications, would it be possible for this to be merged into ubuntu since it is python centric. [10:05] (Still trying to add icons to u-m) === dholbach [~daniel@p54A63D81.dip.t-dialin.net] has joined #ubuntu-devel [10:05] hey [10:05] hey dholbach [10:05] hey ajmitch [10:06] tortoise_: what advantage does it have over using pythons libglade? [10:06] doko/jbailey/elmo: do I have a cookbook for bootstrapping toolchains yet? === tsume_ [~tsume@physics.code-exec.net] has joined #ubuntu-devel [10:06] mvo: it's much easier! [10:07] zyga: we probably need a common package with the desktop files for g-a-i and u-m [10:07] lamont: I'm pausing for food atm. elmo found one typo and a merge error. The typo was easy to fix, the merge error I don't see yet. [10:07] jbailey: ah, ok. [10:07] mvo: that will keep all unsupported apps from being pretty in u-m [10:07] mvo: I was thinking about adding a few tabs [10:08] so y'all will give me a 'build these things and install them, in this order' list? [10:08] one would list applications (from user's point of view) - with icons and descriptions [10:08] lamont: We have a glibc and a gcc to go in for ppc so that ppc64 can be built. The rest of it doesn't need manual help. [10:08] other would list libraries (and a list of applications that use them in the details tab perhaps) [10:09] and the last tab would contain all other updates [10:09] 99% users will only look/care at the application tab IMHO [10:09] jbailey: given that _nothing_ is building on breezy now, if I just kick everything free, then bunches of stuff will be built using the hoary gcc... you're saying that's OK? === lamont expects to at least want to build/install gcc-4.0 and glibc before opening the flood gates. [10:09] lamont: Ah, I thought that mdz's message last night meant things were building. [10:10] lamont: do you need a cookbook? we only need it for ppc64, if at all [10:10] lamont: But that would explain why I haven't seen the new linux-kernel-headers show up in your build logs yet. =) [10:10] source uploads are piling up, pending me getting told how to get the toolchain into the right state for said floodgate opening [10:10] <\sh> nice my chroot is working [10:10] doko: so gcc-3.3 is fine for i386???? [10:10] zyga: having icons for most applications in universe will probably get a bit big :) [10:10] lamont: Do you need to be handed the source code, or can I ask you to start with the linux-kernel-headers that I've already uploaded? [10:11] lamont: just build 4.0, then I'll have to upload a new gcc-defaults [10:11] mvo: yes but since u-m cannot install new packages users will most probably already have an icon already [10:11] that's the sequence of steps I want... when does linux-kernel-headers build? before or after gcc-4.0, etc, etc. [10:11] zyga: very good point [10:11] or is it really just 'build gcc-4.0, install that, force the symlinks to happiness, and open the gates'? [10:12] (except on ppc, where more handholding is needed) [10:12] lamont: It's a prerequisite to the new glibc, it should be built now. [10:12] mvo: and that bring us back to the point of choosing correct icon, bleh === zyga goes back to api reference [10:12] jbailey: does it care what compiler it gets built with? [10:12] lamont: glibc's GCC version is hardcoded in the glibc sources, so it does not matter ... [10:12] lamont: Nope. === lamont goes to make sure that (a) it won't autolaunch, and (b) build some breezy chroots [10:13] doko: After that, should we install glibc and gcc-3.4 into the ppc chroot, and build glibc? [10:13] zyga: we cheat in g-a-i, we have a x-package-name or something to make the mapping work [10:14] doko: question regarding BreezyToolchainTransition: In the "Time Line" section, you mention " Upload build-essential version 11, depending on g++ (>= 4:4.0)". Is that g++ supposed to be gcc, or are we truly going gcc-4.0 and g++-4.0 ? [10:14] jbailey: hmm, I'd like to start a gcc-4.0 build first, tomorrows is my last work day before UDU [10:14] mvo: I've noticed that (vimdiff) [10:15] doko: That works for me. [10:15] I haven't read my gcc-ml yet. Did they do rc2 already? [10:15] mvo: I don't like that solution really [10:15] ok really back to api now === hno73 [~Henrik@henrik.gotadsl.co.uk] has left #ubuntu-devel [] [10:16] jbailey: so breezy should get linux-kernel-headers_2.6.11.2-0ubuntu1 built against hoary toolchain. objections from you, elmo, or doko? [10:16] building then [10:16] crimsun: good point, maybe we'll have two the b-e version bump not before g++-4.0 becomes the default [10:17] doko/jbailey: l-k-h on ppc too? [10:18] lamont: does l-k-h even use a compiler during the build? [10:18] lamont: yes, we should [10:18] only for the test suite apparently [10:18] Kamion: dunno - just being pedantic [10:18] building on all 3 === Goshawk [~Goshawk@host163-106.pool8251.interbusiness.it] has joined #ubuntu-devel [10:22] and uploaded on all 3. [10:23] jbailey: you will have lkh at :33 [10:23] lamont: did you get the toolchain build order? [10:23] fabbione: they're pretending to give it to me here and now [10:23] fabbione: so far it's: build/upload linux-kernel-headers [10:24] lamont: thanks === lamont waits for the doko/jbailey illuminati to tell him what to build/upload next [10:24] is that an arch: all package? === dholbach giggles at lamont [10:24] fabbione: no [10:24] lamont: ok [10:25] fabbione: in fact, there is no arch: all deb in the build [10:25] nah i was wondering of the resulting deb was an arch all [10:27] lamont: gcc-4.0 [10:28] then let jbailey return from lunch [10:28] mvo: hmm this stinks pretty much [10:29] lamont: Thanks. [10:29] doko: curiosity: does -mtune=pentium4 -march=i486 do the right thing, or does the -march override the -mtune? [10:29] mvo: IconTheme.load_icon needs icon name, first package I've tried has filename (which is probably not the same thing) instead [10:30] doko: Are you doing gcc-4 rc1? [10:30] doko: ppc as well, or !ppc? [10:31] zyga: wrt the slashdot incompatibility thing... ubuntu has always maintained that you should not mix-n-match binaries from more than one distro: to do so is "fraught with peril" - that's why we have universe. [10:32] night everybody [10:32] night pitti [10:32] bah [10:32] l-k-h is FTBFS [10:32] that's a good start [10:32] lamont: I was more worrying that ubuntu chooses some deeply incompatible (or simply different) way to do something fundamental [10:32] lamont: the curiosity does the right thing [10:32] I think the problem is rather that independent developers will have to start building packages for both Debian and Ubuntu. Or worse, keep two source packages around. [10:33] doko: coolness [10:33] doko: so.. doxygen... [10:33] fabbione: On Sparc? [10:33] doko: l-k-h bugs to you or jbaley? [10:33] jbailey: yes [10:33] to I just build doxygen_1.4.2-1 with 3.3, or what? [10:33] fabbione: To me. [10:33] (gcc-4.0 is dep-wait) [10:33] yes, I need to change it ... [10:33] fabbione: I need access to a sparc to do the cleanups (although I can guess what they might be) [10:34] you need to change it, or we should just build the current doxygen and get on with life? [10:34] or sync doxygen first ... [10:34] doxygen is sync'ed, but we've not built it yet [10:34] any issue with building it? [10:34] no [10:34] ok === astharot [~isager@b2768fa9cfc13eaf.node.tor] has joined #ubuntu-devel [10:36] and gcc-4.0 build on ppc, or no? [10:36] jbailey: can you give me an ETA to get it fixed? [10:36] jbailey: take into account that more time it passes less hopes there are for sparc to make breezy [10:36] fabbione: Any possibility of getting a chroot? [10:36] jbailey: yes, send me a ssh pub key [10:36] lamont: yes, ppc is ok, we don't need biarch on the first upload. [10:37] doxygen building on all 3 [10:39] we lost hoary thanks to doko fucking up gcc-4 build-dep for 2 days [10:40] jbailey: ^^ [10:40] fabbione: ? [10:40] doko: never mind.. steaming down [10:40] :) [10:41] doko: you forgot to bump a versioned build-dep in gcc-4 for hoary [10:41] making basically going sparc out of sync for 2 days [10:41] (2 rebuils of gcc-4) [10:41] = no sparc for hoary [10:42] fabbione: sorry, which one? [10:42] it was a libcairo something [10:42] oops, I remember ... [10:42] good night === dholbach [~daniel@p54A63D81.dip.t-dialin.net] has left #ubuntu-devel ["Verlassend"] [10:43] doko: yes.. i told you a couple of times.. :( [10:44] Mithrandir: yes? === lamont waits for :03 so he can kick gcc-4.0 again [10:47] lamont: are you taking notes of what needs to be done? :) [10:47] lamont: sparc is going to lag :) [10:49] elmo: breezy-chroot-ppc64, dpkg-checkbuilddeps: Unmet build dependencies: libc0.3-dev (>= 2.3.2.ds1-19) | libc0.1-dev | libc12-dev (>= 2.3.2.ds1-19) | libc6-dev (>= 2.3.2.ds1-19) libc6-dev-ppc64 doxygen (>= 1.4) graphviz (>= 2.2) [10:49] dpkg-buildpackage: Build dependencies/conflicts unsatisfied; aborting. [10:49] dpkg-buildpackage: (Use -d flag to override.) [10:50] doko: You're waiting on a fixed glibc. [10:50] doxygen and graphviz, fine, but libc6-dev and libc6-dev-ppc64 ? [10:50] I do? [10:50] doko: I've tracked down the bug that keeps you from being able to install it, I haven't tracked down the locales.config bug yet. [10:50] fabbione: yes. hppa is also very laggy right now [10:50] doko: Yes. I made a typeo and depended on linux-kernel-headers (<= 2.6.11.2-0) [10:51] lamont: ok thanks.. [10:51] lamont: jb is taking a look at l-k-h.. i guess without that one we can't do anything [10:52] jbailey: should I expect hppa b0rkage in lkh as well? [10:53] lamont: probably.. [10:53] give it a shot.. it's the only way [10:53] /build/sparcbuildd/linux-kernel-headers-2.6.11.2/debian/linux-kernel-headers/usr/include/asm-sparc/posix_types.h:38: warning: ISO C90 does not support `long long' [10:54] lamont: Yes, probably. I'll take a quick pass and guess where the culprits will be. [10:54] lamont: these are the kind of error i am seeing [10:54] fabbione: It's the testsuite we have. [10:54] jbailey: yeps.. i could see that [10:58] doko: after gcc-4.0, then what? (not that I'm there yet, mind you...) [10:58] Bah, found the pasto. [10:58] gcc-defaults, have to downgrade g++ to 3.3 ... [11:00] doko: that source is there, just needs to be built? [11:00] note that we never built a gcc-defaults that uses 4.0, to date... :-) [11:00] gcc-defaults 1.21? [11:01] maswan: read dmesg on ravel [11:01] Need a little bit of time for ppc glibc to build to hand you, afk while it's doing that. [11:02] jdub: ping? [11:02] jbailey: who was that last to? === gustavor [~gustavo@200-171-140-32.dsl.telesp.net.br] has left #ubuntu-devel ["Leaving"] === motaboy [~motaboy@host204-41.pool80182.interbusiness.it] has joined #ubuntu-devel === zyga wants win32 style icons embedded in binary files for a moment === Mitario [~michiel@sikkes.xs4all.nl] has joined #ubuntu-devel [11:10] hello everyone === zerokarmaleft [~zerokarma@ip68-12-45-133.ok.ok.cox.net] has joined #ubuntu-devel === motaboy_ [~motaboy@host204-41.pool80182.interbusiness.it] has joined #ubuntu-devel [11:12] should a library name be: libNAME-VERSION-SONAME ? (for the policy) [11:13] Total 4360 package(s) in state Installed. [11:13] Total 2469 package(s) in state Needs-Build. === lamont cries [11:13] lamont: that's about right :( === MOTU comforts lamont [11:14] grep ': Needs-Build' zz.i386 | grep -v iverse/ | wc [11:14] 285 855 17276 === lamont giggles at MOTU [11:14] Goshawk: http://www.debian.org/doc/debian-policy/ch-sharedlibs.html [11:14] :-P [11:14] Goshawk: depends [11:14] sometimes it needs to be libnameVERSION-SONAME [11:14] depends on the lib [11:15] ok thanks [11:17] <\sh> oh shared libs...yeah its my special ;) [11:18] \sh: really? I thought it was compatibility shell programming ;] [11:19] \sh: recovered from your ordeal then? :) [11:19] <\sh> hehe [11:20] <\sh> after playing 18 around with one shared lib ;) [11:20] <\sh> i tried everything even to make an autotools package out of it ;) [11:23] <\sh> but right now, i will go to bed...have to do some serious things tomorrow morning [11:23] <\sh> g'night folks...have fun :) [11:24] night \sh [11:24] night === thoreauputic_ [~debianarc@wolax8-068.dialup.optusnet.com.au] has joined #ubuntu-devel === blahrus [~blahrus@12-223-50-121.client.insightbb.com] has joined #ubuntu-devel [11:35] might be posted some where, but has breazy devel been started? [11:37] blahrus: seems so but you might want to hear the same thing from someone more knowledgable [11:38] zyga: not really, just wanted to know i I could update my sources and start fallowing the build and watch for bugs === seb128 [~seb128@ANancy-151-1-61-150.w83-196.abo.wanadoo.fr] has joined #ubuntu-devel [11:39] blahrus: what sources are you reffering to? [11:40] blahrus: yes, it has been started, and see ubuntu-devel@; no, it's not useful yet (still bootstrapping the kernel) [11:40] er, not the kernel, the toolchain [11:40] blahrus: it will be VERY VERY UNSTABLE for a bit [11:41] Kamion: cool, just wondering didn't want to rush the guys :) [11:41] did the toolchain stuff get figured out? [11:42] Amaranth: looks like it's in progress [11:42] me, I'm busy resolving the enormous installer merge :P [11:42] d-i changed that much? [11:44] lots of mostly small changes on both sides [11:44] plus some changes that were made identically on both sides and then intermixed with nearby different changes on both sides so that patch can't tell that it's just an already-applied diff [11:44] this is only in the build system and manual, mind you === KaiL [KaiL@pD958B40C.dip.t-dialin.net] has joined #ubuntu-devel [11:46] ..here :) [11:47] any developer around? I have 2 small ideas to make ubuntu a lot better..... [11:47] doko? [11:47] /bin/sh ../../src/gcc/../move-if-change tmp-attrtab.c insn-attrtab.c [11:47] cmp: tmp-attrtab.c: No such file or directory [11:47] mv: cannot stat `tmp-attrtab.c': No such file or directory [11:47] make[6] : *** [s-attrtab] Error 1 [11:47] make[6] : Leaving directory `/build/buildd/gcc-4.0-4.0ds11/build/gcc' [11:47] make[5] : *** [extra32] Error 2 [11:47] (amd64) [11:47] yay [11:48] ..it's about packaging :) [11:48] 1. add the kernel-headers as default [11:48] KaiL: i think it's better you mail ubuntu-devel [11:48] 2. create binary packages for ndiswrapper [11:48] KaiL: 1. has been discussed a number of times and decided against [11:48] btw, home come i saw some where today that debian released som RC of its installer, is it not the same ubuntu is using? And is then ubuntu using an installer which debian dont treat as final? [11:48] it's on the CD [11:48] Kamion: working on modutils - just need to see what else is b0rken [11:48] kent: it's rather more complicated than that [11:49] Kamion: why against? [11:49] KaiL: please see the archives [11:49] KaiL: the headers are on the CD as a compromise [11:49] compromise..? [11:49] kent: yes, it's the same; the reason that it's not final is (nowadays) not to do with the installer but to do with the rest of Debian [11:50] for around 99% of all wlan users not having them on the CD would make ubuntu totally useless! [11:50] KaiL: we don't want to install it by default (bloat, lowest-common-factor arguments, etc., all in the archives), but we include them on the CD so that people who need them have them [11:50] Kamion, becaus debian wants it to be known to work for all the platforms in the world before they treat it as "final", right? [11:51] kent: no, you're missing the point [11:51] kent: the Debian installer's perfectly fine, it's just that the rest of Debian still needs more work before going stable [11:51] and there we come to the binary ndiswrapper.. that would shorten the list of people requireing headers a lot [11:51] Kamion, ok. thanks [11:51] kent: the installer has worked on all architectures for something like a year now. [11:52] (to varying degrees of workingness over time) [11:52] lamont: pong [11:52] KaiL: we already have the ndiswrapper-utils binary package, which is on the CD. [11:52] Kamion: that doesn't help to get it working :) [11:53] KaiL: we can't ship the binary blobs [11:53] the support is there [11:53] that's all we can do [11:53] doko: you saw the gcc-4.0 amd64 issue? [11:53] fabbione, binary blobs? whats a blob? :) [11:53] KaiL: people with such cards will usually have CDs shipped by the manufacturer containing the driver they need [11:53] kent: ehehe [11:53] lamont: no, only fabio's comment ... [11:54] fabbione: if you do a module-assistant on ndiswrapper, you get a .deb [11:54] KaiL: dude, ndiswrapper.ko is in our default kernel package [11:54] KaiL: and why do you need to do so when the default ubuntu kernel ships ndiswrapper? [11:54] it is? [11:54] -rw-r--r-- root/root 132844 2005-04-05 14:40:39 ./lib/modules/2.6.10-5-386/kernel/drivers/net/ndiswrapper/ndiswrapper.ko [11:54] aha... [11:54] (true in warty as well) [11:55] and it will be true in breezy too [11:55] didn't know - so forget my lines.... [11:55] only one point left: isn't 1.1 already released? :) [11:55] KaiL: it will in a few days. it's already merged in the ubuntu kernel tree [11:56] cool [11:56] crimsun: you around? [11:57] this really reduces the requirement for linux-headers [11:58] oh, and a new line for the FAQ: "where is package XYZ?" "enable universe in /etc/apt/sources.list" :) === mkde [~matt@mdke.user] has joined #ubuntu-devel [12:00] elmo, thanks for that docteam thing === KaiL [KaiL@pD958B40C.dip.t-dialin.net] has left #ubuntu-devel ["Glaubt]