=== mr_pouit [n=mrpouit@ubuntu/member/mrpouit] has joined #ubuntu-devel === Amaranth [n=travis@ubuntu/member/Amaranth] has joined #ubuntu-devel === blueyed [n=daniel@pdpc/supporter/active/blueyed] has joined #ubuntu-devel === zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-devel === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel === cr3 [n=marc@pdpc/supporter/bronze/cr3] has joined #ubuntu-devel === administrator_ [i=administ@nat/cisco/x-5adc03529a82d667] has joined #ubuntu-devel === ogra [n=ogra@ubuntu/member/ogra] has joined #ubuntu-devel === tokj [n=tokj@151.82.4.189] has joined #ubuntu-devel === wick2o [n=wick2o@72.25.57.229] has joined #ubuntu-devel === bddebian [n=bdefrees@c-71-224-172-103.hsd1.pa.comcast.net] has joined #ubuntu-devel === freeflying [i=root@gobstopper.dreamhost.com] has joined #ubuntu-devel === _human_blip_away [n=mike@202.83.118.27] has joined #ubuntu-devel === jack_wyt [n=jack@61.48.78.117] has joined #ubuntu-devel === jono [n=jono@ubuntu/member/jono] has joined #ubuntu-devel === jamesh [n=james@203-59-195-215.dyn.iinet.net.au] has joined #ubuntu-devel [02:04] is there a generic way to preseed the Disk psace to partition: ? [02:05] ive tried partman-auto/init_automatically_partition select Erase entire disk:* === j_ack [n=rudi@p508DB1E8.dip0.t-ipconnect.de] has joined #ubuntu-devel === j1mc [n=jim@adsl-75-22-225-51.dsl.chcgil.sbcglobal.net] has joined #ubuntu-devel === alex-weej [n=alex@halls-129-31-82-59.hor.ic.ac.uk] has joined #ubuntu-devel === jack_wyt [n=jack@61.48.78.117] has joined #ubuntu-devel === freeflying [i=root@ubuntu/member/freeflying] has joined #ubuntu-devel === slomo_ [n=slomo@ubuntu/member/slomo] has joined #ubuntu-devel === TomB_ [n=tomb@host217-44-151-84.range217-44.btcentralplus.com] has joined #ubuntu-devel === cypher1 [n=cypher1@59.92.143.246] has joined #ubuntu-devel === lukaswayne9 [n=lukas@c-68-84-69-12.hsd1.nj.comcast.net] has joined #ubuntu-devel === lukaswayne9_ [n=lukas@c-68-84-69-12.hsd1.nj.comcast.net] has joined #ubuntu-devel === _human_blip_away [n=mike@202.83.118.27] has joined #ubuntu-devel === _human_blip_away [n=mike@202.83.118.27] has joined #ubuntu-devel === jack_wyt [n=jack@61.48.78.117] has joined #ubuntu-devel === Amaranth [n=travis@ubuntu/member/Amaranth] has joined #ubuntu-devel === _human_blip_away [n=mike@202.83.118.27] has joined #ubuntu-devel === hunger_t [n=tobias@p54A73E7E.dip0.t-ipconnect.de] has joined #ubuntu-devel === robertj [n=robertj@66-168-215-105.dhcp.athn.ga.charter.com] has joined #ubuntu-devel === stub [n=stub@ppp-58.8.5.179.revip2.asianet.co.th] has joined #ubuntu-devel === Hobbsee [n=Hobbsee@ubuntu/member/hobbsee] has joined #ubuntu-devel === human_blip [n=mike@202.83.118.27] has joined #ubuntu-devel === TerminX [i=8a23c712@adsl-68-122-4-55.dsl.pltn13.pacbell.net] has joined #ubuntu-devel === asac_ [n=asac@debian/developer/asac] has joined #ubuntu-devel === LongPointyStick [n=user@ubuntu/member/hobbsee] has joined #ubuntu-devel === cables [n=cables@65-78-17-151.c3-0.nwt-ubr2.sbo-nwt.ma.cable.rcn.com] has joined #ubuntu-devel === cables [n=cables@65-78-17-151.c3-0.nwt-ubr2.sbo-nwt.ma.cable.rcn.com] has left #ubuntu-devel ["Leaving"] === freeflying [i=root@gobstopper.dreamhost.com] has joined #ubuntu-devel === nags [n=nags@125.16.129.16] has joined #ubuntu-devel === freeflying [i=root@ubuntu/member/freeflying] has joined #ubuntu-devel [06:00] morning [06:01] hi fabbione [06:02] hey fabbione! === _human_blip_away [n=mike@202.83.118.27] has joined #ubuntu-devel === poningru [n=poningru@pool-72-64-213-219.tampfl.fios.verizon.net] has joined #ubuntu-devel === arre [n=ac@1-1-5-33a.gfa.gbg.bostream.se] has joined #ubuntu-devel === arre [n=ac@1-1-5-33a.gfa.gbg.bostream.se] has left #ubuntu-devel ["Konversation] === keyes [n=keyes@easyubuntu/keyes] has joined #ubuntu-devel === giftnudel [n=mb@p54A932DA.dip0.t-ipconnect.de] has joined #ubuntu-devel === dholbach [n=daniel@i59F77881.versanet.de] has joined #ubuntu-devel [06:46] good morning === viviersf [n=cain@196.44.1.98] has joined #ubuntu-devel === somian [n=somian@unaffiliated/somian] has joined #ubuntu-devel === Gman [i=gman@nat/sun/x-9d1e0046e983441b] has joined #ubuntu-devel === administrator_ [i=administ@nat/cisco/x-998b676acde85424] has joined #ubuntu-devel === pitti [n=pitti@ubuntu/member/pitti] has joined #ubuntu-devel [07:14] Good morning [07:14] heya pitti! === human_blip [n=mike@202.83.118.27] has joined #ubuntu-devel [07:28] ogra: new dia === Hobbsee [n=Hobbsee@ubuntu/member/hobbsee] has joined #ubuntu-devel === giskard [n=giskard@213-140-22-74.fastres.net] has joined #ubuntu-devel === tfheen waves to Hobbsee, pitti and Daniel [07:40] hey tfheen === dfarning [n=dfarning@d32-63.rb.lax.centurytel.net] has joined #ubuntu-devel [07:41] morning tfheen [07:42] hey tfheen [07:43] hiya Fabio; nice vacation? [07:43] tfheen: it was ok overall, but as good as i expected [07:43] good morning all, I am working on my first spec. Does anyone have moment to take a look at it to make sure I am on the right track === infinity2 [n=adconrad@cerberus.0c3.net] has joined #ubuntu-devel [07:44] i didn't get around to finish my MythTV box, but i guess i will survive [07:44] https://blueprints.beta.launchpad.net/ubuntu/+spec/derivativeteam [07:44] heya tfheen! === Fujitsu fends off clone-infinity. === MagnusR [n=magru@c83-252-237-96.bredband.comhem.se] has joined #ubuntu-devel [07:49] hey Tollef [07:51] dfarning: sounds like it should be an informal spec at least. The scope isn't well defined in the spec [07:52] dfarning: is there a mailing list for derivatives? [07:53] tfheen, how would you define the scope more correctly [07:53] LaserJock, i have submitted a request ticket [07:53] dfarning: I'd probably move a bit of the bullet list from the summary to scope. What is and what isn't covered by the spec? [07:55] tfheen, ok so who and what is affect defined more completely [07:55] does informal mean set is as drafting? [07:57] no, informal means there is no implementation [07:57] actually, sorry, I meant Informational, not informal. :-) [07:58] ok I remember reading about that [07:58] thanks === sabdfl [n=sabdfl@ubuntu/member/pdpc.silver.sabdfl] has joined #ubuntu-devel [07:58] I'll go work on it some more === shackan [n=shackan@85-18-14-13.fastres.net] has joined #ubuntu-devel === marilize [n=marilize@dumbledore.hbd.com] has joined #ubuntu-devel [08:10] woooo go seb! go dani! more crack more crack :) === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel === Fujitsu watches the stable crack roll in. === Lure [n=lure@external-7.hermes.si] has joined #ubuntu-devel [08:24] any archive-admin around? Somebody needs to get libkexiv2* through binary new in order to get digikam 0.9.1 built - would be great to have this for beta === AndrewB [n=andy@tuxhacker/pdpc.supporter.student.AndrewB] has joined #ubuntu-devel [08:28] Lure: looking === cjwatson [n=cjwatson@82-69-40-219.dsl.in-addr.zen.co.uk] has joined #ubuntu-devel === cypher1 [i=cypher1@nat/hp/x-91675f96a138399c] has joined #ubuntu-devel === cypher1 [i=cypher1@nat/hp/x-f04c086db0654a99] has joined #ubuntu-devel === human_blip [n=mike@202.83.118.27] has joined #ubuntu-devel === carlos [n=carlos@163.pool85-48-166.static.orange.es] has joined #ubuntu-devel [08:46] pitti: hi [08:46] hey carlos [08:46] pitti: I had to delay again the time when lang packs are generated [08:46] the process will start in 5 minutes (I think the delay was of two hours) === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel [08:47] carlos: ok, no problem [08:47] carlos: how does feisty look? [08:47] pitti: also, Feisty is more or less full imported now [08:47] yay! === pitti yearns for tarballs [08:47] pitti: there are some remaining things, but need manual approval, so I think we should start preparing base lang packs [08:47] pitti: you have already the ones from yesterday [08:48] if you want to start playing with them, otherwise, in a couple of hours, you should have the new ones for toay [08:48] today [08:48] ok, couple of hours is alright, I need to do some CD testing right now [08:48] ok [08:48] pitti: same url schema as before [08:48] http://people.ubuntu.com/~carlos/language-packs/feisty/ [08:49] once we do the release, it will become http://people.ubuntu.com/~carlos/language-packs/feisty-updates === sky_walkie [n=hrdlo@193.85.244.121] has joined #ubuntu-devel === pitti -> breakfast, bbl === arre [n=ac@alchemist.openend.se] has joined #ubuntu-devel === arre [n=ac@alchemist.openend.se] has left #ubuntu-devel ["Konversation] === Mithrandir [n=tfheen@vawad-xen1.err.no] has joined #ubuntu-devel === mdke [n=matt@212-139-78-17.dynamic.dsl.as9105.com] has joined #ubuntu-devel === Nuscly [n=nuscly@29-231.206-83.static-ip.oleane.fr] has joined #ubuntu-devel === hunger_t [n=tobias@pd95b0676.dip0.t-ipconnect.de] has joined #ubuntu-devel === tkjacobsen [n=tkjacobs@105.18.235.85.dk-boa.res.dyn.perspektivbredband.net] has joined #ubuntu-devel === thekorn [n=thekorn@a89-182-13-53.net-htp.de] has joined #ubuntu-devel === mvo_ [n=egon@p54A67F71.dip.t-dialin.net] has joined #ubuntu-devel [09:22] are people working on the network-manager problem, bug 82335? it has a massive number of duplicates, looks difficult to fix, and hasn't been assigned to anyone yet. [09:22] Malone bug 82335 in network-manager "network-manager should not set offline mode when it manages no device" [High,Confirmed] https://launchpad.net/bugs/82335 [09:23] mvo_: moin; can I bother you with upgrade-manager -d not working? [09:23] mdke: I don't know of anybody working on it, but I think it should be relativetly easy to solve. [09:24] Mithrandir: removing network manager? :) [09:24] mdke: That's what everybody seems to suggest. [09:24] Feisty simply can't be released with that bug... [09:25] agreed [09:25] Mithrandir: it is not that simple: n-m does not know if there are other interfaces at all, therefore it consideres it has all [09:25] (it would be best if NM had proper static IP support, but I don't think that's going to happen) [09:25] Mithrandir: backend would need to be changed to show all interfaces (also static) and just do not manage them [09:25] Fujitsu: not happening for feisty, at least. [09:25] Fujitsu: it's not just the static ip support, but also the failure to detect perfectly normal dhcp configured interfaces [09:25] Lure: no, it wouldn't. [09:26] Mithrandir: suse and fedora are presumably doing this proprely [09:26] mdke: True... It needs to be able to detect it better. [09:26] Mithrandir: why not? === cassidy [n=cassidy@16.132-240-81.adsl-dyn.isp.belgacom.be] has joined #ubuntu-devel === Fujitsu looks at debian-backend, to see what can be done rather than just ignoring anything other than `inet dhcp' [09:30] pitti: what exactly is not working? [09:31] mvo_: gpg verification fails [09:31] it's not very verbose unfortunately [09:31] and the /tmp/tmpUU.. doesn't exist any more [09:31] pitti: please update update-manager to the version in -proposed, its a known (and fixed) bug that waits for sru-verification [09:32] pitti: I will nag brian about it again so that it can move to -updates ASAP [09:32] mvo_: ah, will try that [09:33] mvo_: btw, I did apt-cdrom -m add before that to use the alternate CD as source; could this be the reason? [09:33] Mithrandir: so will you maybe give someone a poke to look at that bug? [09:33] mdke: yes [09:34] thanks a lot [09:34] mvo_: (still no 'upgrade from CD' option in the u-n CD insertion dialog) [09:34] pitti: no. but if you have the altenravtive cd, please try sudo sh /cdrom/cdromupgrade" === seb128 [n=seb128@ubuntu/member/seb128] has joined #ubuntu-devel [09:34] mvo_: ah, will do that [09:34] pitti: oh, no option from the dialog? [09:34] pitti: could you please also install the u-n from -proposed? [09:35] mvo_: well, just 'open package manager' and another one (unrelated) [09:35] mvo_: yup [09:35] pitti: thanks [09:35] pitti: I check the status of the two bugs now === cerbero_626 [n=jga@p213.54.172.243.tisdip.tiscali.de] has joined #ubuntu-devel === raphink [n=raphink@ubuntu/member/raphink] has joined #ubuntu-devel === chand| [n=rsamson@linagoraberri.pck.nerim.net] has joined #ubuntu-devel === Gman_ [i=gman@nat/sun/x-0fb923453678e824] has joined #ubuntu-devel [09:44] mvo_: with -proposed, the gpg error doesn't happen [09:44] pitti: great, thanks [09:44] mvo_: so I'll do the test with both variants now ('apt-cdrom add, update-manager -d' and '/cdrom/upgrade') [09:45] Can somebody look at bug #91012? There are a heap of duplicates, and it only appeared 36 hours ago. [09:45] Malone bug 91012 in hal "[apport] hal-device-manager crashed with ImportError in () App crashed running "Hardware Information"" [High,Confirmed] https://launchpad.net/bugs/91012 [09:45] pitti: great, thanks. and please try also if update-notifier gives you a correct dialog with the version from -proposed [09:46] mvo_: 'correct dialog'? [09:48] pitti:sorry. u-n should present a "upgrade from this volume" dialog when a upgradable CD is inserted. I will search for the bugnumber for this now [09:50] mvo_: ah, when running the script on the CD? [09:50] mvo_: I'm currently testing the u-m -d approach === jono [n=jono@ubuntu/member/jono] has joined #ubuntu-devel === herzi [n=herzi@p548FC116.dip.t-dialin.net] has joined #ubuntu-devel [10:14] I know I keep nagging, but a german translator please fix bug 60527 , since xchat-gnome is annoying with this bug [10:14] Malone bug 60527 in language-pack-gnome-de "xchat-gnome /me misbehaviour" [Undecided,Confirmed] https://launchpad.net/bugs/60527 [10:14] there are still strings wrong in both edgy and feisty === Keybuk [n=scott@quest.netsplit.com] has joined #ubuntu-devel === freeflying [i=root@gobstopper.dreamhost.com] has joined #ubuntu-devel [10:16] giftnudel: these translations bug should be fixed with last upstream release. The problem is Ubuntu using its own translations... [10:16] yes, the upstream ones are correct === Zdra [n=zdra@181.188-247-81.adsl-dyn.isp.belgacom.be] has joined #ubuntu-devel === Kagou [n=Kagou@81.185.184.59] has joined #ubuntu-devel === infinity [n=adconrad@cerberus.0c3.net] has joined #ubuntu-devel === markvandenborre [n=mark@d54C44CCF.access.telenet.be] has joined #ubuntu-devel [10:34] whom should I talk to on ubuntu installer iso localisation? === herzi [n=herzi@p548FC116.dip.t-dialin.net] has joined #ubuntu-devel [10:35] any ideas, ladies and gentlemen? === heno [n=henrik@ubuntu/member/heno] has joined #ubuntu-devel [10:36] markvandenborre: just wait until people wake up ;) [10:36] Treenaks, grapjas! [10:37] happy to see you out here, long time no see [10:37] I've been here for ages ;) === markvandenborre not... [10:38] Treenaks: Ages? I'd say ever. [10:38] Fujitsu: something like that :) [10:39] let me clarify my question, just to make sure that I'm in the right channel... [10:40] I'd like to see auto generated iso's for my combination of language and region [10:40] (and for others obviously) [10:42] so an iso with a sensible default i18n, l10n, dictionaries, keyboard layout and if possible location specific packages (electronic id card reader for example for Belgium) [10:42] markvandenborre: we don't have the manpower to QA that well. [10:43] Mithrandir, perfectly sensible answer [10:43] is there a way locoteams could help with the QA process? [10:43] so as to lower the QA load? [10:43] markvandenborre, I am getting together a new team to work on just such issues [10:43] it would also require a massive increase in the storage space, etc on the ISO building machine. [10:44] I'm not saying it's impossible, I'm just pointing out a couple of problems which would need to be solved. [10:44] ok... [10:44] https://wiki.ubuntu.com/DerivativeTeam [10:45] building an ISO takes about one hour, a bit. [10:45] There are several loco team working on localized distros [10:46] Mithrandir, in our case, I'm just trying to do some research to avoid duplication of work [10:47] we want a localised iso anyway for our locoteam [10:47] and there seem to be many more interested... [10:48] markvandenborre, what langugage [10:48] nl_be [10:49] can you send me an email at dfarning@gmail.com so I can follow up with you [10:49] will do [10:49] thanks [10:49] I think we have about 20 different localization distos out there right now [10:50] Mithrandir, can I help you convince decision makers to give this some priority by gathering support from locoteams around it? [10:52] this would prove demand, and guarantee you at least somewhat technically knowledgeable qa testers [10:53] markvandenborre, I doubt that Ubuntu will be able build and provide the localized distro for you any time soon [10:53] I think providing a simple way to create the image(s) yourself would help [10:54] 'make-live-image.sh nl_NL' [10:54] oh yes, I really don't need much hand holding [10:54] markvandenborre: (or nl_BE) [10:54] and I'm not alone in this [10:54] markvandenborre: if we have people who have shown a commitment to test the regular ISOs for a while expressing interest in testing custom ISOs, that would make it more likely to happen. [10:55] Treenaks, I am putting together some derivative spec so we can put together the tools to easily develop such a localization [10:55] ok [10:56] Mithrandir, dfarning thx for your help [10:57] will try to find some people to rally around this === gicmo [n=gicmo@p5491CDB9.dip.t-dialin.net] has joined #ubuntu-devel === freeflying [i=root@ubuntu/member/freeflying] has joined #ubuntu-devel === mpt [n=mpt@121-72-136-145.dsl.telstraclear.net] has joined #ubuntu-devel [11:17] mvo_: yay, u-manager dist-upgrade worked flawlessly [11:18] mvo_: do I need edgy-proposed for the cdrom script as well? [11:21] pitti: no, that should work [11:21] pitti: thanks for testing! === carlos [n=carlos@163.pool85-48-166.static.orange.es] has joined #ubuntu-devel [11:23] Riddell: can you please take a look at the kdegames ftbfs? [11:25] Riddell: same for kde-systemsettings === dholbach_ [n=daniel@i59F77881.versanet.de] has joined #ubuntu-devel === dholbach_ [n=daniel@i59F77881.versanet.de] has joined #ubuntu-devel [11:29] Mithrandir: ok [11:36] markvandenborre: my standpoint has definitely always been that I'm happy to help people build the customised images more easily (e.g. the various documents on help.ubuntu.com/community/) but I don't think the Ubuntu CD image / release teams should ever get into doing it ourselve [11:36] s === Yagisan [n=Yagisan@doomsday/developer/Yagisan] has joined #ubuntu-devel [11:40] carlos: so if I start the langpack cronjob at 0900 rookery time, will that be enough? === pax [n=pax@pdpc/supporter/student/pax] has joined #ubuntu-devel [11:43] pitti: for Breezy, Edgy and Dapper, yes [11:44] pitti: Feisty is a full export and is not ready until 16:00 [11:44] hmm wait [11:44] something is not correct here... === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel [11:46] pitti: For Feisty, I will tell you once it finish for today [11:46] carlos: alright [11:46] carlos: I build test packs right now, btw, just to see if there are any major problems [11:46] pitti: it used to finish around 12:00, but for some reason, this weekend it took around 4 hours more [11:47] pitti: ok [11:47] cjwatson, thx for your information [11:48] cjwatson: Does policy require that packages with an original-maintainer field have an @ubuntu.com address in maintainers? [11:49] mjg59: no [11:49] implementation might, which I would consider a bug [11:49] mjg59: Policy doesn't, but builds will fail otherwise. [11:49] Fujitsu: well, that should be fixed. it's wrong. [11:49] (pkgmaintainermangler fails) [11:49] cjwatson: There's a bug about it. [11:49] It's because it tries to add Original-Maintainer, but it already exists. [11:50] Ok. My acpid upload failed because of that. [11:51] What's the appropriate thing to do? Wait for it to be fixed and then ask for the package to be rebuilt? [11:52] pitti's on that bug, I believe... What you do depends on when it's likely to be fixed, I guess. [11:52] mjg59: urgh, sorry about giving you extra mail by giving it back again. [11:52] It is a one-character change. [11:53] Fujitsu: Got a number? [11:53] Looking. [11:53] Thanks [11:53] Bug #91086 [11:53] Malone bug 91086 in pkgbinarymangler "binary-mangler chokes on non-Ubuntu email addresses that contain a reference (string-wise) to MOTU" [Wishlist,Unconfirmed] https://launchpad.net/bugs/91086 [11:53] The description isn't particularly true. [11:54] Sorry, summary. [11:54] Wishlist? Hm. [11:54] It's easy to work around. [11:54] Not relaly [11:54] You just need to use a whitelisted address. [11:55] (or convince pitti that a one-character change is worth it) [11:55] mjg59: you could probably just fix pkgmaintainermangler yourself [11:55] Yeah [11:55] erm, what's up? [11:55] might need to prod folks to upgrade the buildd chroots - I forget [11:56] I'm nervous about poking that sort of infrastructure, though :) [11:56] mjg59: I recently made it accept @kubuntu.org addresses, I can easily extend that [11:56] pitti: why does there need to be a whitelist in the first place? [11:56] ah, /me looks at that bug [11:56] pitti: Why not just make it accept any address if there's an original-maintainer field? [11:56] pitti: Why not just make it not kill the build when Original-Maintainer already exists? [11:56] pitti: Maintainer: Original-Maintainer: should never error [11:56] cjwatson, regarding localised iso's you said: my standpoint has definitely always been that I'm happy to help people build the customised images more easily (e.g. the various documents on help.ubuntu.com/community/) but I don't think the Ubuntu CD image / release teams should ever get into doing it ourselves [11:56] I don't know, infinity might have had his reasons [11:56] (that is, change `exit 2;' to `exit 0;' on line 63 or similar. === finalbeta [n=finalbet@d54C689F7.access.telenet.be] has joined #ubuntu-devel [11:57] cjwatson: it doesn't, it chokes on non-Ubuntuish maintainer fields, like MOTU Media Team [11:57] I wanted to make clear that I'm not talking so much about custom packages and stuff, rather about i18n, l10n and stuff... [11:58] pitti: It only chokes if both Original-Maintainer and a non-Ubuntu maintainer address are present. [11:58] Fujitsu, cjwatson, mjg59: yup, it seems that it's time to remove this check, I'll look into it [11:58] I'm curious about where you draw the line [11:58] pitti: right, that's a bug and should be fixed [11:58] Thanks pitti, it's pretty trivial. === pitti adjusted the bug [11:58] pitti: Thanks [11:58] markvandenborre: yeah, that's customisation. We can't afford combinatorial explosion on our own infrastructure; this is exactly the sort of thing where others should pitch in to build the customised images [11:59] but you would be able to, say, supply a dead easy customisation script and stuff? [11:59] markvandenborre: I've never had time to write one myself, but I believe there are such scripts out there, or at least the beginnings of them [11:59] like Treenaks joked : sh customise_cd.sh -l nl_be or whatever? [12:01] cjwatson, sorry if I'm taking too much of your time here, just trying to look at the limits of what Canonical people could be expected to be able to contribute to [12:01] markvandenborre: scripting and code fixes and stuff are reasonable enough, but in this case IIRC somebody else has already taken care of much of the scripting [12:01] would further development of clean high quality customisation code be an option? [12:02] ok, thx, that answered my question before I even asked it ;) [12:02] markvandenborre: it would certainly be an option, although it always works better when the people who actually have the greatest need for doing it put something together that works for them ... [12:03] I'd have no objection in principle to Canonical developers doing the sort of work you mention, but it would have to wait its turn behind other priorities so I'm just warning you that it might arrive later than you'd otherwise like. :-) [12:04] heh, thx, that's a very clear answer [12:04] so if I summarise you correctly, you say: just start fiddling with what's out there, and gather some interest behind it === StevenK_ is now known as StevenK [12:05] and if you manage to do that, that will raise priority for us, but don't expect us to ever run an entire customised build infrastructure === seb128 [n=seb128@ubuntu/member/seb128] has joined #ubuntu-devel [12:07] cjwatson, except for Mithrandir and this dfarning guy, is there anyone in particular that should be kept into the loop on this? [12:08] markvandenborre: I think that's a reasonable summary, yes. In general there should be few things (core infrastructure, basically) where Canonical participation is *required*, although there may be many situations where it's useful or the quickest way to get the job done [12:09] ok, perfect, we know where to start now [12:09] thx [12:09] markvandenborre: me, and possibly jono in his capacity of locoteams coordination [12:09] yes, of course... [12:09] I wouldn't forget jono and you [12:09] thx for your time [12:10] Fujitsu, mjg59: pkgbinarymangler fix uploaded [12:11] pitti: Will the buildds automatically update when it makes its way into the archives? [12:11] They won't, if I recall what infinity said previously. [12:12] Fujitsu: yes, they will [12:12] StevenK: Soyuz' do, katie's dont === StevenK sits corrected [12:12] Fujitsu: so this can be given-back as soon as archive.u.c has it === Tonio_ [n=tonio@205.207.103-84.rev.gaoland.net] has joined #ubuntu-devel === markvandenborre [n=mark@d54C44CCF.access.telenet.be] has left #ubuntu-devel ["Ik] [12:13] Isn't katie the autosyncer? [12:13] no [12:13] Fujitsu: katie => old archive management software [12:13] Fujitsu: sorry, katie is the old upload and archive mainteance software [12:13] Wasn't that dak? [12:13] OK.... [12:13] Fujitsu: the one Debian uses [12:13] katie is the old name for dak, which is the archive management software used in Debian and which we used. [12:13] Fujitsu: right, it's dak === Tonio_ [n=tonio@205.207.103-84.rev.gaoland.net] has joined #ubuntu-devel [12:14] And I suppose that still manages security. [12:14] right === cprov [n=cprov@monga.dorianet.com.br] has joined #ubuntu-devel === jsgotangco [n=jsg123@ubuntu/member/jsgotangco] has joined #ubuntu-devel === EtienneG [n=etienne@modemcable178.77-70-69.static.videotron.ca] has joined #ubuntu-devel === Gman_ is now known as GmanZZZ === dholbach [n=daniel@i59F77881.versanet.de] has joined #ubuntu-devel [12:30] Mithrandir: FYI I've started moving some bits and pieces from people.u.c/~cjwatson to ~ubuntu-archive [12:30] I've left redirects in place === TeTeT [n=spindler@modemcable178.77-70-69.static.videotron.ca] has joined #ubuntu-devel === jrib [n=jasonr@upstream/dev/jrib] has joined #ubuntu-devel [12:31] cjwatson: thanks; I was going to gently prod you about it sometime. === twilight [n=twilight@ubuntu/member/twilight] has joined #ubuntu-devel === stub [n=stub@ppp-58.8.5.179.revip2.asianet.co.th] has joined #ubuntu-devel [12:33] pitti: Thank you! === shackan [n=shackan@85-18-14-13.fastres.net] has joined #ubuntu-devel === apokryphos [n=apokryph@unaffiliated/apokryphos] has joined #ubuntu-devel === LeeJunFan [i=junfan@s64-186-37-120.skycon.net] has joined #ubuntu-devel === shiyee [n=Shiyee@homer.cs.aau.dk] has joined #ubuntu-devel [12:59] fabbione: CLK_TCK does not seem to be exported by linux-libc-dev any more; any idea what replaced it? [12:59] pitti: oh yeah that's old news.. i don't remember.... [12:59] cjwatson: CLOCKS_PER_SEC [12:59] Oops. [12:59] pitti: CLOCKS_PER_SEC [12:59] Sorry cjwatson. [12:59] Fujitsu: already tried that, doesn't work either [01:00] How strange. [01:00] it's in time.h's includes, but it does not get exposed for some reason === pitti currently fights with the syslinux FTBFS, which hasn't been rebuilt for ages [01:04] pitti: I just built something that uses CLOCKS_PER_SEC fine. [01:05] pitti: bet you've got -ansi or -pedantic in your CFLAGS? [01:05] gcc -m32 -mregparm=3 -DREGPARM=3 -D__COM32__ -W -Wall -march=i386 -Os -fomit-frame-pointer -fno-stack-protector [01:05] not that I can see [01:06] pitti: is sys/time.h included or time.h? [01:06] Keybuk: time.h and sys/times.h [01:07] but adding sys/time.h doesn't help either [01:07] it's impossible to sleep/resume a machine running the nvidia proprietary driver, right? === shiyee_ [n=Shiyee@homer.cs.aau.dk] has joined #ubuntu-devel [01:08] <_ion> pitti: Btw, have you had time to review my changes to restricted-manager yet? [01:08] _ion: not yet, sorry [01:08] <_ion> pitti: No hurry :-) [01:09] pitti: hmm, I can't replicate it -- I get a __stack_chk_fail error instead [01:09] Keybuk: right, I already fixed those [01:09] pitti: which is the file that's failing for you? [01:09] but since a trivial program with just #include works, it must be somehting in syslinux itself [01:10] Keybuk: com32/libutil/get_key.c [01:11] pitti: if I change CLK_TCK to CLOCKS_PER_SEC, it's fine === jwendell [n=wendell@ubuntu/member/wendell] has joined #ubuntu-devel [01:12] Keybuk: hm, not here; maybe it's different on amd64? [01:12] get_key.c: In function get_key: [01:12] get_key.c:140: error: CLOCKS_PER_SEC undeclared (first use in this function) [01:12] get_key.c:140: error: (Each undeclared identifier is reported only once [01:12] that's odd; you're compiling on amd64 for i386 ? [01:12] that's what I get [01:12] or on amd64 for amd64? [01:12] Keybuk: amd64 === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel === hggdh [n=hggdh@pool-71-170-94-19.dllstx.fios.verizon.net] has joined #ubuntu-devel [01:22] Keybuk: argh, there is an empty time.h in com32/include; I guess it fell over that one [01:22] could be yeah === rraphink [n=raphink@ubuntu/member/raphink] has joined #ubuntu-devel [01:24] ah yes [01:24] the build line is totally different [01:25] hmm, public_html/germinate-output/ is ever so slightly bigger than I thought === cjwatson waits for it to copy to ~ubuntu-archive/ [01:27] Keybuk: making libutil_lnx.a it just does -I./include, but libutil_com.a it does -I../include as well [01:27] s/Keybuk/pitti/ [01:28] right, and this makes sense if using the com32 includes for _com [01:28] but the empty time.h there doesn't [01:28] indeed [01:28] that'd be included instead of the one in /usr/include === pitti hardcodes it to 1000000l just for testing === rtg [n=timg@gn-216-166-171-58.mtwireless.net] has joined #ubuntu-devel === cr3 [n=marc@pdpc/supporter/bronze/cr3] has joined #ubuntu-devel === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel === seb128 [n=seb128@ubuntu/member/seb128] has joined #ubuntu-devel === tkamppeter [n=till@bl8-114-140.dsl.telepac.pt] has joined #ubuntu-devel [01:41] I need some help concerning the Debian packaging system. [01:41] tkamppeter: just ask === zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-devel [01:42] I have a file in /etc and I want that it is not marked as config file. See bug 36532 and bug 90988. [01:42] Malone bug 36532 in gs-esp ""Unsupported format" when trying to print" [Medium,Confirmed] https://launchpad.net/bugs/36532 [01:42] Malone bug 90988 in cupsys "printing any file give "Unsupported format 'the/format'" (dup-of: 36532)" [Undecided,Unconfirmed] https://launchpad.net/bugs/90988 [01:42] Malone bug 36532 in gs-esp ""Unsupported format" when trying to print" [Medium,Confirmed] https://launchpad.net/bugs/36532 [01:43] tkamppeter: you want to create a new config file? [01:43] Seveas: problem with the ubugtu timeout? [01:43] tkamppeter: or convert an already existing conffile to a config file? [01:44] apokryphos, no... first message had 2 bug numbers, second bug title had a dupe number [01:44] No, gs-esp has a file /etc/cups/pstoraster.convs This file shouldNOT be considered as a config file. [01:44] Mithrandir: for the record, I did my amd64 alternate and desktop tests (including autoresize this time), I didn't see anything that hasn't been reported before [01:44] tkamppeter: if it's shipped in the deb, it must be a conffile. If you don't want it to be, you should move it somewhere other than /etc [01:45] Mithrandir: edgy upgrades work fine with the /cdrom/upgrade script, but fail with edgy's update-manager; using the one from edgy-proposed works fine [01:45] elkbuntu: yeah, I would've thought the timeout wouldn't (or probably, shouldn't) permit that though; doesn't matter much though I guess. [01:45] So should I make the postinst generate it then [01:45] tkamppeter: you usually ship a template in /usr/share or generate it on the fly in postinst [01:45] tkamppeter: but NB that you must gracefully handle and respect already existing files on upgrades [01:46] tkamppeter: 'shouldNOT be considered as a config file' -> you mean conffile here, right? [01:46] tkamppeter: if you really mean 'config file', then the file should be in /var/ [01:47] Yes, I mean a conffile. [01:47] for files in /etc whose format changes gradually over time, it's generally expected that either (a) it's a conffile so the user gets prompted about the changes or (b) the maintainer scripts take care to upgrade the contents of the file === thekorn_ [n=thekorn@a89-182-26-242.net-htp.de] has joined #ubuntu-devel [01:47] /etc/cups/pstoraster.convs tells CUPS that there is the pstoraster fileter and how to use it. [01:47] tkamppeter: since this already is a conffile, you must not touch it on upgrades [01:48] tkamppeter: why do you want to change this, btw? [01:48] tkamppeter: it's usually much easier to fix conffiles with package uploads than fixing config files [01:48] pitti, but this file is not supposed to be modified by the user. So the user got a stone-old version yeasr ago and it will never get updated. [01:49] tkamppeter: incorrect: conffiles get automatically changed unless the user has changed them; otherwise the user will be prompted [01:49] tkamppeter: your answer has several wrong concepts [01:49] tkamppeter: everything in /etc can be modified by users [01:49] (assuming there's a controlling terminal) [01:49] tkamppeter: and "config files" will never be automatically updated; conffiles will [01:49] tkamppeter: that's why I said that conffiles are usually what you want === stu1 [n=stub@ppp-58.8.12.203.revip2.asianet.co.th] has joined #ubuntu-devel [01:51] So then how to fix bug 90988? The problem here is that the user got a file /etc/cups/pstoraster.convs from an old Ubuntu version and with the time the file was changed in the Ubuntu package gs-esp. Updates of gs-esp do not replace the file. [01:51] Malone bug 90988 in cupsys "printing any file give "Unsupported format 'the/format'" (dup-of: 36532)" [Undecided,Unconfirmed] https://launchpad.net/bugs/90988 [01:51] Malone bug 36532 in gs-esp ""Unsupported format" when trying to print" [Medium,Confirmed] https://launchpad.net/bugs/36532 [01:51] tkamppeter: one mistake that may be responsible is that sometimes developers attempt to both ship a file as a conffile (i.e. in the .deb) AND edit it in maintainer scripts; this is wrong wrong wrong and can often lead to users getting erroneously prompted. If your package has === moquist [n=moquist@pool-70-20-57-230.man.east.verizon.net] has joined #ubuntu-devel [01:52] cjwatson: your reply was clipped, btw [01:52] Here it seems that the user never got prompted. He was always left with his old /etc/cups/pstoraster.convs file. [01:52] tkamppeter: ... ever edited that file in a maintainer script as well as shipping it in a conffile, then that could be responsible for problems [01:52] tkamppeter: that doesn't normally happen. [01:53] tkamppeter: but if the file got edited and they upgraded using something which didn't present them with the standard conffile prompt, then it's possible [01:54] The current maintainer scripts do not touch pstoraster.convs. [01:54] tkamppeter: there are the following suitable responses: (a) the user edited the file by hand => reject bugs about it not getting upgraded, as it's their problem now; (b) a maintainer script edited the file by hand => fix the maintainer script not to do that; (c) the file should never ever ever be edited by users under any circumstances => move it out of /etc [01:54] why is all of the sudden linux-image-dbug-2.6.20-10-server going to get installed? I am using the -generic kernel... [01:54] but don't move files out of /etc just because users edited the file by hand and then the usual conffile resolution process did the wrong thing for them [01:55] cjwatson: (d) use ucf, and open up a whole other raft of problems ? [01:55] The file really should not be edited by hand, so (c) should be the right solution, but problem is that CUPS only searches /etc for such files. [01:55] tkamppeter: use a symlink then [01:55] CUPS is free software and therefore patchable ... [01:55] symlinks from /etc to elsewhere are a bit grotty [01:56] So the bug is an upstream bug of CUPS, so I should file a feature request for a second search place. [01:56] that's the sort of thing it's perfectly reasonable to patch in locally [01:57] tkamppeter: shuold be easy to fix ourselves [01:57] tkamppeter: of course that patch should be proposed upstream as well === Monk-e [n=guido@c529dd229.cable.wanadoo.nl] has joined #ubuntu-devel === sky_walkie [n=hrdlo@193.85.244.121] has joined #ubuntu-devel [01:58] If I use the symlink from /etc to my own place, what would an update do. Would it replace the existing file by a symlink? === sahin_h [n=ezaz@dsl5400DBF2.pool.t-online.hu] has joined #ubuntu-devel [01:58] tkamppeter: that would be wrong, so I hope dpkg won't [01:59] pitti, how would you solve it? In the CUPS package by patching CUPS to provide a second search location for .types and .convs files not editable by the user? [01:59] tkamppeter: that definitively sounds cleaner [02:00] pitti, or do you have an idea for a fix inside the gs-esp package? [02:00] fixes in one package aren't necessarily better than fixes in multiple packages [02:00] tkamppeter: the gs-esp package should then stop shipping the conffile, rm it in the postinst on upgrades, and generate it on the fly in /var in postinst === dholbach_ [n=daniel@i59F77B08.versanet.de] has joined #ubuntu-devel [02:01] pitti: sounds like it should be in /usr, not /var? after all it's shipped at the moment, not generated dynamically [02:01] pitt, but this will only work if CUPS is changed to search in /var. [02:01] cjwatson: I don't know [02:01] tkamppeter: correct. Fixing cups sounds useful === cypherbios [n=cypher@ubuntu/member/cypherbios] has joined #ubuntu-devel [02:01] tkamppeter: will a static file version work for everyone, or does it need to be dynamically generated? [02:02] I think, /usr is correct, as pstoraster.convs ships as a ready-made file with constant content. [02:02] tkamppeter: if the former, then it should be in some .d directory in /usr/share [02:02] pitti, the static file works for everyone. [02:02] oh, not a .d directory [02:02] cups should look in /etc first though, just in case somebody does need to generate it by hand [02:03] and the thing that removes it in the postinst on upgrades should (a) only do so once, when upgrading to the version that removes the conffile (b) check against various known md5sums for the conffile before nuking itt [02:03] it [02:04] So bug 90988 is a design bug of CUPS. So we must make CUPS searching in /etc plus the new directory and report an upstream bug (not a feature request) with this patch. [02:05] Malone bug 90988 in cupsys "printing any file give "Unsupported format 'the/format'" (dup-of: 36532)" [Undecided,Unconfirmed] https://launchpad.net/bugs/90988 [02:05] Malone bug 36532 in gs-esp ""Unsupported format" when trying to print" [Medium,Confirmed] https://launchpad.net/bugs/36532 [02:06] Then we must search all Ubunru packages for /etc/cups/*.types and etc/cups/*.convs files and do the same fix on them as we do on gs-esp. === giftnudel [n=mb@frnk-590c4952.pool.einsundeins.de] has joined #ubuntu-devel [02:07] mjg59: hal FTBFSed on powerpc and sparc due to the new macbookpro patch; do you have a minute to take a look at it? [02:09] pitti, this looks all like a higher-impact change. Should this be done for Feisty? === rraphink [n=raphink@ubuntu/member/raphink] has joined #ubuntu-devel [02:10] mvo_: I've committed a fix to DistUpgradeViewKDE in update-manager, should it work if I just upload in the normal way? [02:11] tkamppeter: adding another search path to cupsys doesn't sound particularly scary to me === jack_at_home [n=jack@222.131.250.230] has joined #ubuntu-devel [02:14] So pitt, will you do it? [02:14] s/pitt/pitti/ [02:15] tkamppeter: I have some other high-urgency stuff on my plate, so I'd prefer not to [02:16] pitti: ok, thanks. We should make sure to releasenote that bit, then [02:18] seb128: http://people.ubuntu.com/~pitti/langpacks/daily/feisty/ !! [02:18] pitti: waouh ;) [02:19] seb128: I'll use tomorrow's rosetta tarball for beta; the current ones are just for testing and sorting out major issues with carlos today [02:20] k, I'll give them a try [02:20] pitti, how much is that in total ? [02:20] ogra: how much? [02:20] i have pleanty of space on my add-on CD ... [02:20] Mithrandir, I think this is the best way then, as this file is usually really not manually modified or removed by users, the problemn in the bug does not happen. So we should tell in the release notes that when the user gets this problem he should do [02:20] dpkg -P --force-depends gs-esp [02:20]