[12:00] lamont: trivial fix [12:00] kewl [12:00] so base should be installable now. [12:00] can you adjust PaS? [12:01] or is only elmo able to do that? [12:01] Mithrandir: grub test still needed on i386? [12:01] no PaS adjust [12:01] doko: no, not really. [12:01] lamont: ok. :/ [12:01] we kinda need grub-installer on amd64. [12:01] lamont: gcc-3.4 is not yet built? [12:01] ah, that'd be elmo work, since (I think) I already changed it. === Gman [~Glynn@amfea-proxy-1.sun.com] has joined #ubuntu [12:02] yo Gman [12:03] doko: ubuntu5 built on all 3 architectures [12:03] Mithrandir: PaS has been adjusted, may need to be sync';ed [12:04] mdz: will we build an amd64 kernel for i386 as well? [12:04] debian-installer/grub-installer_0.50ubuntu3: Installed [-:uncompiled] === aes [~andrew@dsl-212-23-23-154.zen.co.uk] has joined #ubuntu [12:04] that'd be installed, Mithrandir [12:04] lamont: hmm, only ubuntu4 is in the archives. [12:04] lamont: woo. :) [12:05] so tomorrow's daily might work. === Gman [~Glynn@amfea-proxy-1.sun.com] has joined #ubuntu [12:07] doko: any chance you added new binary packages? [12:07] or renamed? === lamont fears that gcc-3.4 needs some NEW lovin; [12:08] there's nothing in NEW [12:09] clues on where gcc-3.4 went? upload was sept 2 [12:10] what upload? it's in "Building" for i386 [12:10] the ubuntu5 upload. [12:10] yes, I know that, but it's not in the "Uploaded" state in w-b [12:10] elmo: there's a gcc-3.4...upload file on macaroni... [12:11] Uploaded happens when the anon queue daemon sends its mail. That or it's broken - I've never seen Uploaded for warty packages, except when I manually marked them thus [12:11] interestingly, macaroni, royal, and crested all think they uploaded it... [12:11] Rejected: libgcc1_3.3.4-9ubuntu1_i386.deb: old version (1:3.3.4-9ubuntu1) in warty >= new version (1:3.3.4-9ubuntu1) targeted at warty. [12:11] Rejected: lib64gcc1_3.3.4-9ubuntu1_i386.deb: old version (1:3.3.4-9ubuntu1) in warty >= new version (1:3.3.4-9ubuntu1) targeted at warty. [12:12] gcc-3.4_3.4.1ds1-7ubuntu5_amd64.changes [12:12] lamont: no it doesn't, it happens on ACCEPT mail and the buildd should be getting them [12:12] Rejected: libgcc1_3.3.4-9ubuntu1_amd64.deb: old version (1:3.3.4-9ubuntu1) in warty >= new version (1:3.3.4-9ubuntu1) targeted at warty. [12:12] Rejected: libgcc1_3.3.4-9ubuntu1_powerpc.deb: old version (1:3.3.4-9ubuntu1) in warty >= new version (1:3.3.4-9ubuntu1) targeted at warty. [12:13] elmo: OK. could just be the hyper-short hours... [12:13] doko: what elmo said... gcc-3.4 can't upload libgcc1. [12:13] or gcc-3.3 needs to stop, and 3.4 needs to upload something newer.... [12:14] while being conformant to whatever the freeze policy is, of course. [12:14] s/freeze policy/decision about gcc-3.4/ [12:15] ah, ok, that's the crap to have the libgcc1's version number at 3.3.4, although built from the 3.4.1 sources (because we wanted to have a chance to remove gcc-3.4 from warty, if it breaks something). I'll change this to 3.4.1 now. [12:16] elmo: any chance where to look for those failures? [12:20] doko: not at the moment no, sorry, not really === jdub swishes around the fixy-fixy wand [12:24] i should put on suspenders and get a feather duster [12:25] jdub: that would be a pretty good look for you :p [12:29] jdub: It would make for excellent pre-release artwork. [12:29] jdub: Someone would do an icon theme, no worries. === [Clint] [~c123p456@user-12hdtek.cable.mindspring.com] has joined #ubuntu === cef [~cef@c211-28-159-51.thoms1.vic.optusnet.com.au] has joined #ubuntu === HrdwrBoB [hrdwrbob@220-253-15-102.VIC.netspace.net.au] has joined #ubuntu === tvon [~tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu [12:47] Mithrandir: btw, newt is ftbfs on amd64: non-PIC in shlib === lamont trusts the process, apt-get dist-upgrades his warty desktop [12:49] hrm... questions from X [12:53] lamont: hm, ok, I'll look at it [12:54] lamont: ! [12:55] lamont: output of lspci? [12:55] oh [12:55] xresprobe has to be installed first ... is it? [12:55] daniels: dunno - this is upgrading my system [12:55] pretty sure xresprobve was installed already [12:56] lspci| grep VGA [12:56] 0000:01:00.0 VGA compatible controller: ATI Technologies Inc Radeon RV200 QW [Radeon 7500] === lamont will accept donations of more supportable hardware...\ [12:56] it asked for my PCI path to the card, and whether or not to use the kernel's fb interface [12:57] Setting up apache2-mpm-worker (2.0.50-12) ... [12:57] Forcing reload of web server: Apache2httpd (no pid file) not running [12:57] that looks ugly.. [12:58] lamont: what happens when you type 'xresprobe ati'? [01:00] xresprobe ati [01:00] id: [01:00] res: [01:00] freq: [01:00] that's with an X server running, of course.. === lamont brbr [01:01] brb,even === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu [01:04] "Ubuntu Debugging Artwork"?? === lamont finds a certain amount of irony that firefox, xchat, etc are moving towards _not_ focusing on new tabs, while metacity blithely strips focus away from an ssh-add prompt. [01:07] lamont: whoa, that really sucks. can you please send me the output of ddcprobe? [01:08] daniels: you didn't remind me to be root... :-( [01:08] xresprobe ati [01:08] id: SAMSUNG [01:08] res: 1280x1024 1024x768 800x600 640x480 1792x1344 [01:08] freq: 30-85 50-160 [01:08] and yes, I'm running 1792x1344 [01:08] lamont: that's been a per hate of mine for ages [01:08] lamont: sounds like it's missing "a bit" of error checking. [01:08] HrdwrBoB: I filed a bug in debian asking for an option to fix that... [01:09] Mithrandir: ?? [01:09] lamont: that xresprobe doesn't say "permission denied" or something [01:09] yeah [01:09] daniels: still want dccprobe output? [01:10] Mithrandir: epiphany-browser also died on amd64: compiler can't create executables [01:10] lamont: hm, weird [01:10] sure it does error checking [01:10] try xresprobe ati; echo $? [01:10] daniels: that was going from ubuntu9?? to current [01:10] lamont: huh?, that _really_ shouldn't happen. [01:11] lamont: care to mail me the build log? [01:11] or put it online somewhere [01:11] daniels: as mortal: xresprobe ati; echo $? [01:11] id: [01:11] res: [01:11] freq: [01:11] 0 [01:11] lamont: x shouldn't ask questions at that stage; my best guess is that xresprobe/et al aren't installed before x is configured [01:11] lamont: ! [01:11] Mithrandir: it points at config.log, which is gone... [01:11] *sigh*, ok. [01:11] I'll try without ia32-libs-dev installed [01:12] Mithrandir: if you want, I'll kick it again and not remove it.. [01:12] I have that and it might be using -m32 for some weird reason [01:12] please [01:12] ia32-libs is not a build-dep, if that helps... [01:12] hm === lamont bets so [01:12] I just upgraded X and it's decided that none of the modes are appropriate anymore.. and I get 640x480 [01:12] lamont: it shouldn't need that. If it needs ia32-libs-dev, something is fucked [01:13] Does anyone else use "screen" inside a gnome-terminal? [01:13] hypatia: as we speak [01:13] HrdwrBoB: which rev of X? [01:13] I have a problem where it regularly causes the terminal to stop rendering, and eat 100% CPU for about 30 seconds. [01:13] lamont: whatever the latest is, let me check [01:13] w [01:13] hypatia: i use screen in g-t [01:13] If nothing else, this is hell on my laptop's battery life. [01:14] hypatia: there is known weirdness between screen and vte, but i haven't seen that === lamont got annoyed at g-t, and switched to xterm... [01:14] By "regularly", I mean, "once a day, and then it happens intermittantly for half an hour." [01:14] hypatia: no idea, I haven't seen it. [01:14] hypatia: is screen running locally or remotely? [01:14] jdub: I've been getting it consistently for about a year. [01:14] Mithrandir: Remotely. === jdub wishes the vte maintainer didn't go AWOL, even though he's sitting next to other active gnome maintainers. [01:15] hypatia: same here.. I haven't seen it. [01:15] jdub: Within 15 minutes of spiv letting me use his laptop it started happening to him too... [01:15] maybe it's the app you run? === lamont giggles... [01:17] hypatia: though, my remote box is running debian unstable, not warty, no idea if that matters. [01:17] /build/buildd/epiphany-browser-1.3.8/./configure: line 1: gcc-3.4: command not found [01:17] Mithrandir: missing build-depes [01:17] lamont: sounds like seb128 forgot to add a build-dep.. [01:17] lamont: fixed in new upload [01:18] Version: 4.3.0.dfsg.1-6ubuntu14 [01:18] daniels: of X?? :-( === lamont hates X uploads almost as much as OO.o uploads [01:18] lamont: no, xresprobe [01:18] lamont: I'm not *that* harsh [01:18] kewl [01:19] '* This one goes out to LaMont -- fixing some really small bug for him so he'll have to do ten times the work to get it uploaded. Cheers.' === lamont contemplates having his rsync script copy the prior rev of _all.deb's into the new version before the rsync... But that'd be work [01:19] my laptop is a bit dead now, could somebody please upload a newt with http://debian-amd64.alioth.debian.org/patches/current/newt_0.51.6-11.0.0.1.amd64.patch applied (and a changelog entry saying "enabling -fPIC" or something along those lines)? [01:19] HrdwrBoB: output of xresprobe nv? [01:19] mga [01:19] daniels: this is the EPERM bitch? [01:19] id: DELL P1110 [01:19] res: 1600x1200 1152x864 1024x768 800x600 720x400 640x480 1800x1440 [01:19] freq: 30-121 48-160 [01:19] lamont: yah, $? wasn't getting properly passed down [01:19] I pissed off the gf2mx and put ye olde trustworthy g400 in it [01:20] Mithrandir: I'll grab it and do it. [01:20] or xresprobe mga, then [01:20] lamont: thanks. [01:21] is it supposed to force agp 1x? [01:21] HrdwrBoB: .. no [01:21] Mithrandir: It's one of those "uh... yeah, interesting bug, I don't believe you!" bugs that I'm always reluctant to file. [01:22] Mithrandir: Remote is debian testing. [01:22] so I put a CD in the drive, and my prefs are set to auto-play it. It doesn't. What'd I break? [01:22] HrdwrBoB: The application is irssi. [01:22] hypatia: oh.. heh.. a lot of other people also use that [01:22] I really need to run irssi remotely, otherwise you can see me rejoin every half an hour. [01:22] perhaps it's the way you use it? [01:22] hrm odd [01:23] Because every half an hour, I need to reboot on this to make the wireless card work. [01:23] Well yeah, but I don't think my usage of it is that unusuall. [01:23] hypatia: no idea.. that works just fine here, and has been for a long time. [01:23] hypatia: irssi also has this talented bug that makes the whole screen go blue [01:23] I press ^P and ^N to switch windows, I type in the box. [01:23] and the only way to fix that is to open a new terminal [01:23] hypatia: oh, rad. is this the blue thing, or different? [01:23] yeah [01:23] that really bits [01:23] only ever happens to me when i change themes ;) [01:24] but GAR! [01:24] daniels: The entire gnome-terminal stops rendering and chews 100% CPU for about 30 seconds. [01:24] jdub: no, detach + reattach fixes it as well [01:24] jdub: Well, detaching and reattaching screen wfm. [01:24] Mithrandir: :) [01:24] doesn't work for me ;) [01:24] although some C-a C-l action might [01:24] C-l alone doesn't help [01:24] daniels: And once it starts doing that, it will do it about every third time someone says something until I completely restart g-t. [01:24] not C-l, yeah [01:24] or, it helps, but fucks once somebody says something [01:25] On the wireless subject, is anyone else here using ipw2200? [01:25] I think I've also seen the irssi-blue-screen behaviour when opening a tab on a maximised gnome-terminal with no other tabs, i.e. causing the terminal to resize a little. [01:25] And if so, do you have to reboot once an hour too? === jdub is worried his bandwidth bills are going to be astronomical this month === lamont knows jdub's pain === Mithrandir thinks .au is silly bw-wise. [01:26] jdub: You can look it up on swift's site can't you? [01:27] jdub: bandwidth bills? [01:27] yeah [01:27] i'm probably below the line [01:27] Mithrandir: au is broken :( [01:27] hypowow-ee. [01:27] though i'm less distributed on-peak/off-peak wise now [01:27] Mithrandir: it b-d's gcc-3.4-base, but not the actual compiler, I guess.\ [01:27] jdub: phone or internet? [01:27] lamont: "it" being e-b? [01:28] lamont: internet [01:28] Mithrandir: yeah [01:28] jdub: that's why I have the throttled mirror... [01:28] lamont: uhm, ok. Sounds weird. [01:28] i'm using apt-proxy-v2, which helps a bit [01:28] newt uploaded [01:28] lamont: thanks a lot [01:29] np [01:29] lamont: got any more fun build failures for me? [01:29] :) === Mithrandir makes a mental note not to move to .au until it gets decent bandwidth stuff. [01:30] Mithrandir: the only open mail I have right now is e-b (known) and grub (FTBFS: powerpc) --> duh... === lamont should really PaS grub on ppc [01:30] well, I don't care too much about grub not working on ppc. ;) [01:30] Mithrandir: you and the rest of the world, including me [01:30] universe has some [01:31] I know, but universe isn't that crucial [01:31] right. [01:31] I could think about how to do OOo [01:31] OTOH, main is continuing it's BORING state wrt FTBFS issues. [01:31] and that's a _GOOD_ thing. [01:31] true [01:33] do we care at all about pytables, python-utmp, pyx, rdiff-backup, tla-load-dirs (and others)??? [01:33] I consider the fact that my desktop machine with no wireless connections has a wireless icon and a battery icon in the taskbar a bug - is this a generic thing or has it inadvertantly been seen as a laptop? [01:33] HrdwrBoB: it was a discussion item [01:33] lamont: FTBFS-wise? [01:33] Mithrandir: missing b-d: python [01:33] yes [01:33] they're arch: all? [01:34] hrm.. could be, but I don't think so. [01:34] ok [01:34] I could just give them all back and see what comes back,. :-) [01:34] I guess they should be fixed, but as my key is being memtested at the moment, I'm not going to do anything about them now.. [01:34] your key? [01:35] or rather, the laptop with my key on [01:35] so I can't sign uploads tonight [01:35] ah, ok, [01:35] newt ACCEPTE [01:35] D [01:35] woo === lamont moves the missing-bd-python stuff to 'Failed' [01:40] lamont: does the archive handle that source1 builds package A on arch A, B, C while source2 builds package A on arch D? [01:40] (as in, can I build ooo from a different source on AMD64 than on other arches?) [01:43] yes [01:43] ok, good. [01:43] I'll do the evil ia32-libs-hack for OOo, then. === lamont makes a note to decide whether to ^5 or smite Mithrandir for his clever solution [01:44] (no, I don't like it, but as I see it, it's the only way to get ooo available on amd64 without fixing ooo or implementing multiarch) [01:44] lamont: ;) [01:45] (on both of those are out-of-scope for warty) [01:45] oh. s/clever/"clever"/ [01:45] lamont: it's frigging ugly, I do so agree. I should bribe you and elmo with proper amounts of beer to accept it. ;) === lamont still wants to know why is dvd's don't auto-play [02:20] grumble. [02:20] doesn't find CF devices anymore either... [02:20] all I did was remove gnome-volume-manager and reinstall it... [02:21] lamont: is g-v-m actually running? [02:22] hrm. no [02:22] it sometimes dies for me, I haven't been able to track it down, yet. [02:23] how does one restart it? === Gman [~Glynn@amfea-proxy-2.sun.com] has joined #ubuntu [02:23] just run it [02:23] (from the Run option in the menu, or a terminal) === cef_work [~cef@fw-01.amc.com.au] has joined #ubuntu [02:33] hrm.. doesn't look like it's finding the CF device at all now... === lamont still doesn't know what his plans are for tomorrow === Mithrandir goes to bed [02:38] nite guys === cef_work just got to work.. fun fun! [02:39] cef_work: huzzah. where are amc these days? [02:39] hrm. gnome-volume-manager running, still no happiness [02:39] daniels: still cheltenham.. moving to moorabbin in 2 weeks *sigh* [02:39] elmo: cheers [02:39] cef_work: ahr [02:39] daniels: feel like talking about X.org again? *grin* [02:40] cef_work: given i'm moving to east ivanhoe tomorrow night, how about no? :P [02:40] daniels: bugger! [02:40] daniels: where in ivanhoe? I'm in alphington [02:41] cef_work: ormond road, between ivanhoe and eaglemont station (just east of both, a little closer to eaglemont than ivanhoe) [02:41] doko: I had not intended to build an amd64 kernel for i386...is there any reason to, since we provide a full amd64 distribution? [02:41] daniels: ahh ok.. I know the area but not the road.. cool stuff.. hrm.. possibly have LOS.. *grin* [02:42] cef_work: i'm still not sure how much bandwidth I have, so that might actually be nifty ;) [02:43] daniels: heh.. [02:44] cef_work: 31 j7 [02:44] between maltravers rd and the eyrie (which runs east away from eaglemont station) [02:44] cef_work: i'm apparently about halfway down the street [02:48] ahh ok.. don't have LOS.. there is a big hill in the way *8-(.. but you're like 200-300m from nailer [02:49] daniels: and I have a 24 dBi grid sitting next to me that I can probably loan you for a few weeks [02:50] cef_work: er, mike's been in st kilda for a bit over a year now, dude :) [02:50] thanks for the offer tho [02:50] daniels: doh! keep forgetting that [02:50] haha [02:50] yeah [02:51] where nailer used to live anyway :) [02:51] haven't heard from him in ages [02:51] he comes into #lgl moderately frequently [02:52] lot of that crowd seems to not talk to me anymore - *8-( [03:09] :( [03:09] daniels: when I remove the monitor settings (HorizSync and VertRefresh) from the XF86Config, it works. [03:12] AHA [03:12] fabbione: ^^ [03:12] fabbione: this is what i was afraid of [03:17] heh, it appears sg is not loaded with the scsi cdrom - thusly, sound-juicer doesn't work [03:21] lamont: around? [03:21] yo [03:22] lamont: could you keep an eye on i855-crt and make sure it gets built? [03:22] Mark will need it tomorrow [03:22] should be trivial, but just in case [03:22] terranova built it, and it was uploaded [03:23] however... [03:23] Sep 6 01:50:01 buildd-mail: i855-crt must be manually dinstall-ed -- delayed [03:23] elmo around? [03:23] yes [03:24] thanks === kfish [~conrad@balgowlah-ep.tip.csiro.au] has joined #ubuntu [03:27] mdz: sorry I couldn't push it all the way, eh? [03:27] yo ... nice work all :) [03:27] lamont: no problem, elmo's nudged it [03:31] yo kfish [03:32] this weather is really pissing me off [03:32] http://descent.netsplit.com/~scott/flies.jpg [03:32] ^ my kitchen ceiling right now [03:32] X worked on my 17" powerbook out of the box ... --> happy kfish :) [03:33] ick [03:33] hm, it appears as though nothing much works with SCSI in ubuntu :D [03:33] Keybuk: i'm reverting the wifi change; i don't think the bars are the right thing, but they're better than the current one. [03:33] Keybuk: ew [03:34] warm as hell and 95% humidity [03:34] jdub: them little semicircles? [03:34] though the drive loaded, and there's a disc in it, 'device manager' sees it [03:34] it was never added to /etc/fstab [03:34] etc [03:34] tvon|x31: yeah, that concept needs to be gnomeised [03:34] jdub: I dig the mac-esque wifi applet bits [03:34] its perdy though [03:34] :) [03:35] the original one stank, half the space was taken up by a graphic that just said 'wifi' [03:35] tvon|x31: yeah, we all hate the original one [03:35] HrdwrBoB: we don't mess with /etc/fstab [03:35] except at install time [03:35] the one before the semicircles was okay, I never got to see it working because of my wireless setup at the time though [03:35] oh, yeah, that's why I'm up.. it's too damn hot to sleep === tvon|x31 yawns === tvon|x31 pokes around for xorg debs [03:37] mdz: yeah but it wasn't detected at install [03:37] it's been there all along [03:37] HrdwrBoB: so you have a SCSI device present, but you installed to a non-SCSI device? [03:37] yes [03:38] I also installed off a non SCSI device [03:38] I believe entries are only added to fstab for partitions with a recognizable OS installed on them, or ones that you configure manually [03:38] ah [03:38] it does not attempt to add entries for everything which might be mountable [03:38] Keybuk: string a few rubber-bands together and go hunting... It's a great southern sport. [03:38] is there any reason that that can't be done? [03:39] (at least for known things like cd drives in the system [03:39] ) [03:39] it sounds technically feasible [03:39] is there a metapackage that installs all of desktop? or which task do I want? [03:39] lamont: aptitude install '~tubuntu-desktop' [03:40] I'll file it as a request under d-i [03:40] HrdwrBoB: file a Severity: enhancement bug against debian-installer [03:40] ok :-) [03:40] :) [03:40] lamont: is gcc-opt packaged? [03:40] jdub: certainly. But it's not in warty [03:41] and not going to be for the moment? [03:41] jdub: that's my expectation, given the lack of push to make it otherwise... [03:43] do I care if libesd-alsa0 gets removed? [03:43] absolutely not [03:43] in fact it should be celebrated [03:43] heh [03:43] heh === lamont leaves totem-xine, but otherwise upgrades [03:43] jdub: can you update the email address for bugzilla-daemon on warty-bugs so its posts don't wait for moderation anymore? [03:44] maybe that'll make gnome-volume-manager/totem happy enough to play dvd's automatically again [03:44] mdz: aren't you the admin? [03:44] nm, i have site passwd [03:44] dbus-1-utils was on the list of things to install.. [03:44] jdub: yeah, but it still seems to send a confirmation message [03:44] which I can't reply to [03:45] its mail probably goes to /dev/null anyway [03:45] the way I did it, anyway [03:45] ok, the best way to do it [03:45] lamont: heh, I just evicted ~35 greenfly, 2 dragonflies and 4 moths from my bathroom [03:46] "City of Lost Devices" - heh [03:46] I'm using the tries and tested "glass and a piece of paper" technique [03:46] is add an entry to List of non-member addresses whose postings should be automatically accepted. [03:46] i've just added one for nny [03:46] the previous one was warthogs [03:46] i'll unsub the macquarie one, too [03:47] and I'm left with that great problem, do I (a) shut all the windows and die from the heat or (b) open them, and let all the flying beasties back in [03:48] Keybuk: c) buy an airconditioner [03:49] 1083.43 GHz processor.. kewl. [03:49] jdub: since the ubuntu ppl are doing a talk next month at LUV, how much can I give away in telling people what's up for next months talk? I'd like to basically plug the talk but I don't want to give anything away [03:49] HrdwrBoB: it's very tempting, our summers are just getting cruel these days :-/ [03:50] silly, I remember as a kid praying for a warm day in July/August [03:50] your seasons are broken [03:50] july/august is when it rains :) [03:50] heh, this is the first summer for years that it's actually rained properly [03:50] it hasn't got any cooler though [03:50] HrdwrBoB: except in Melbourne, where it rains every month [03:51] heh [03:51] we've had pretty good weather recently though [03:51] even the aussies were complaining about the humidity [03:51] cef_work: see debian.slug.org.au, except you should probably also mention 'Canonical' (the company) and 'Ubuntu' (the distribution) [03:51] craige hasn't updated that entry yet [03:52] jdub: cool' [03:54] jdub: cool. thought that was about what I could get away with saying anyway, but just wanted to have it confirmed === lamont discovers plugdev, brb [03:57] mdz: how's that going? [04:00] tvon|x31: there are debs, but they are out of date with the current debian stuff, and they're also just of 6.7.0 [04:00] tvon|x31: look for post-warty movement === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu [04:01] plugdev. Figures [04:07] lamont: hey? [04:07] should the default user be in the plugdev group [04:08] I'm thinking yes [04:08] ye [04:08] that's my thinking too. [04:09] that would be another installer bug? [04:11] daniels: thanks [04:11] base-config, iirc. [04:11] yeah [04:11] ok filing [04:21] er, base-config already adds users to plugdev [04:22] addgroup --system plugdev >/dev/null 2>&1 || true [04:22] adduser "$RET" plugdev >/dev/null 2>&1 || true [04:22] base-config (2.44ubuntu3) warty; urgency=low [04:22] * lib/menu/passwd: Create new group 'plugdev' for accessing local removeable [04:22] devices (USB sticks, digital cameras, etc.) and put the created user into [04:22] it. [04:22] HrdwrBoB: is this a fresh install? if so, how fresh? [04:23] (sounder from 2nd of sep or later?) [04:23] daniels: sounder 3... :-( Nowhere close to current [04:23] lamont: ah [04:24] daniels: I installed it last night [04:24] though it was sounder cd7 [04:25] sounder 7 is not quite recent enough to get plugdev love [04:25] yeah [04:44] um [04:45] is X supposed to use 100% CPU when I connect via vnc? === lamont giggles [04:46] heh [04:47] HrdwrBoB: without DAMAGE, yes :-) [04:48] it seems to be dependant on resolution (1744 is too much for a poor duron 700) [04:48] 'don't do that, then' ;) [04:48] the irritating thing is that I'm not even viewing the screen [04:48] I'm using x2vnc [04:48] so it shouldn't be a problem [04:48] HrdwrBoB: that's kind of bad [04:49] when visuals are involved, it's scraping the entire screen, which is a pretty vicious load on X [04:49] yes, very [04:49] i don't know if vnc can let you send events only [04:49] i thought it did [04:49] so did I [04:49] perhaps it's the particular implementation? [04:50] hmm.. there's about another 10% load increase when I actually connect using a visual vnc client [04:51] hrm... I get russian spam in russian now, but still no japanese/korean love [04:54] does 'printscreen' actually take a screenshot for anyone? [04:55] fonts work in the browser, but xterm doesn't like them.... [04:58] lamont: try uxterm, or forcing use of Xft with '-fa mono' (or some other font) [04:58] gnome-terminal is happy though.. [04:58] uxterm is happy. [04:58] xterm -fa 'Andale Mono' -fs 10 [04:58] ? [04:58] how does uxterm differ from xterm? [04:58] hrm [04:58] tvon|x31: nope [04:58] are the mscorefonts in a package anywheres? [05:00] msttcorefonts [05:00] fancy that [05:00] lamont: utf-8 love [05:00] is that all? [05:00] hrm [05:01] lamont: yah [05:01] daniels: is it supposed to be in main/universe? [05:01] tvon|x31: should be in universe, but it might be contrib [05:01] ah, didnt know about contrib [05:01] (it isnt in universe) [05:01] ho hum [05:02] feh. mutt isn't happy with it, even though bash is. [05:02] hrm, where is contrib? [05:02] bleh, even at 800x600 vnc uses so much CPU it's barely useable [05:02] universe [05:02] only it's not there [05:03] cabextract is in univ, msttcore is in sid contrib [05:04] if anyone gives a hooey === hrdwrbob_ [~hrdwrbob@220-253-15-102.VIC.netspace.net.au] has joined #ubuntu [05:06] \242\273 [05:07] hrm.. wonder what utf-8 character that is [05:08] remote desktop preferences says localhost.localdomain because it's the first entry in /etc/hosts for localhost, who's wrong? base-config or vino [05:14] where can I find what font has a given 16-bit utf8 character? [05:19] mako about? === lamont wanders bedwards [05:22] lamont, still around? [05:22] yeah [05:23] wondering where else to find fonts... [05:23] where was that 'I can eat class' page you had up during your BOF? [05:23] http://www.unicode.org/Public/UNIDATA/NamesList.txt [05:24] http://mako.yukidoke.org/projects/utf8-tests/ [05:24] and is there a trivial place to go paste in a 16-bit hex value and find what it should be? (and what font...) [05:24] i need to put in an apache fragment to have that automatically use UTF-8 for those text files [05:25] good gravy - that first one is the complete character list, yes? [05:25] the first one has SOME pangrams [05:25] but it's not wholly pangrams [05:25] YI SYLLABLE CUO and that means? [05:25] what is the character code? [05:25] A2BB [05:27] Mind you - that's just a couple of octets from the front of some spam... [05:28] but my russian spam has been working for a bit..:-) [05:28] http://www.unicode.org/charts/PDF/UA000.pdf === lamont still needs to get japanese input methods figured out... [05:29] i have never seen that character before :) [05:29] lamont: as long as you're awake, if you could check in on linux-source-2.6.8.1_2.6.8.1-1.2, I'd appreciate it [05:29] I just uploaded it, so it's probably not going yet === lamont checks [05:31] lamont: i, also known as Lolo, is the Sino-Tibetan language used by aboriginal people of south-west China. [05:31] reverting patch alpha-epoch-comment from ./ ... ok. [05:31] cp -p debian/patches/00list-1.1 debian/patches/00list [05:31] cp: cannot stat `debian/patches/00list-1.1': No such file or directory [05:31] make[1] : *** [debian/monolith/patch-2.6.8.1-1.1] Error 1 [05:31] :-( [05:31] lamont: i don't have that font either [05:31] lamont: that'd be the -1.1 version [05:31] I just uploaded a -1.2 [05:31] ah, OK. [05:31] before :30:00? === lamont stalls for about 3 more minutes [05:32] lamont: just barely, yes [05:32] -rw-rw---- 1 mdz mdz 363 2004-09-05 20:25 linux-source-2.6.8.1_2.6.8.1-1.2_source.upload === jdub installs every desktop app in supported [05:32] mdz: ah, OK [05:32] jdub: glutton! [05:32] lamont: or sucker, depending on your POV :) [05:33] Date: Mon, 6 Sep 2004 04:30:27 +0100 (BST) [05:33] Subject: Accepted linux-source-2.6.8.1 2.6.8.1-1.2 (source) [05:33] mdz: ok. about 20 seconds ago, cron.daily started dealing with it [05:33] if it manages to apply the patches and starts actually compiling things, I'm happy [05:34] Databases locked for general maintenance by katie -- please wait [05:36] mdz: thoughts on pulling the fix for #267837 from debian (postfix, ipv6 relay issues) [05:36] lamont: if that patch is correct, it's fine with me [05:36] it doesn't get much more unintrusive than that :-) [05:36] [gcc line] [05:36] /tmp/ccYo6i3Q.s: Assembler messages: [05:36] /tmp/ccYo6i3Q.s:18678: Warning: end of file not at end of a line; newline inserted [05:36] /tmp/ccYo6i3Q.s:20052: Error: unknown pseudo-op: `.lc10' [05:36] make[4] : *** [gimpimage-new.o] Error 1 [05:36] rad! [05:37] yeah - I'll merge it over then [05:37] awesome [05:37] what's the stance on metacity's refusal to allow windows above the screen [05:37] could ubuntu have a patch to add this as an option? [05:37] HrdwrBoB: windows can't cover the panels [05:38] mako: Lalland Scots / Doric: Ah can eat gless, it disnae hurt us. (?) [05:38] no :) [05:38] jdub: i think he's talkin ga bout the alt-trag above the top border thing [05:38] yeah [05:38] mdz: powerpc is quick and painful: [05:38] dpkg-buildpackage: host architecture is powerpc [05:38] /usr/bin/fakeroot debian/rules clean [05:38] find: debian/config/powerpc: No such file or directory [05:38] lamont: that's OK; powerpc and amd64 aren't there yet [05:38] daniels: which is? [05:38] lamont: they should be within the next day or so, but I wanted to get i386 on the table [05:38] windows butting up against the panel, surely [05:38] jdub: if you alt-drag, it's not possible to put the title of the window outside the screen [05:38] jdub: hold down alt, try to drag a window up so far that the titlebar is above the top of the screen [05:39] jdub: you can do it for *below*, but not above [05:39] the theory is that if you put the titlebar outside the screen, it's Bad [05:39] lamont, i'm going to run out now, but tomorrow you should explain to me why you want/need aboriginal sino-tibetan script :) [05:39] mako: nfc why - was just noticing lots of square boxes in my spam subject lines... [05:39] which is fine, but you can't do it unless you know what you're doing anyway, and metacity refused to add it even as an option, despite lots of complaints === lamont doesn't get asian spam - just russian it seems. [05:40] i don't think that's worth a local patch, really [05:40] better to deal with it upstream [05:40] maybe it was actually a non-utf-8 character set? [05:40] mdz: probably [05:40] two characters, A2 and DD [05:41] ouch: sh: line 1: uudecode: command not found [05:41] lamont: b-d: sharutils? [05:41] mdz: want me to cause 1.3 results to go to you if they fail? [05:41] so the only INSANE menu after installing everything from supported is 'internet' [05:41] lamont: sure [05:41] mdz: yep [05:42] :0:c [05:42] * ^Subject: Log for .*build of linux-source-2.*dist=warty [05:42] ! mdz@alcor.net [05:42] thanks [05:42] only failures get that far [05:42] this is so much more fun than testing in a chroot :-P [05:43] s/failures/failures beyond the atomaton's ability/ [05:43] mdz: Ok, Camm. [05:44] :-) === lamont decides to augment his script in some manner to mail failures to the uploader. maybe. [05:46] also need to chat with elmo about a central hub for the automaton to live on. [05:46] 9+ copies of the scripts is rediculous [05:47] Subject: Buy V\343lium and X.anax now [05:47] That'd be a non-UTF8 font :-) [05:48] lamont: failures to uploader would be totally rad [05:48] anyway - bed time [05:48] jdub: pb is that the uploader's email isn't necessarily available at the point that I'm looking at the log. [05:48] :-( [05:49] I could just auto-file bugs in bugzilla :-) [05:49] but then I fear mdz would kill me. :-) [05:49] haha [05:49] something to ponder tomorrow while I'm vacating (or was that vacationing? laboring?) [05:50] but I have to check out USB sync'ing tomorrow. [05:50] night [05:56] it takes forever just to run dpkg-source -b on this thing [05:57] jdub: did you make that bugzilla-daemon change? there was still a whole load of stuff in the queue, but maybe it was all old [06:01] mdz: yeah, changed a while back === Gman is now known as GmanAFK === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu [07:00] morning guys [07:06] yo fabbione === stub [~zen@203.220.53.14] has joined #ubuntu [07:10] hey jdub === pitti [~pitti@195.227.105.180] has joined #ubuntu [07:33] Good morning everybody! [07:34] Good morning! [07:34] .. afternoon.. whatever your timezone is.. [07:43] hey pitti === pitti [~pitti@195.227.105.180] has left #ubuntu ["Verlassend"] === pitti [~pitti@195.227.105.180] has joined #ubuntu [07:45] hmmm [07:45] this gdm behaviour of automatic restarting on upgrades is really really annoying [07:47] that's... unusual [07:47] WTF! [07:47] i think there's a debconf setting for that, perhaps yours is borked? [07:47] gdm that starts as soon as possible? [07:48] jdub: mine is normal installation. i don't customize my desktop === jdub hasn't had gdm automagically restart on upgrade [07:48] no no this does not work [07:48] there were errors while booting [07:48] and now i can't see them because of gdm starting immediatly [07:49] there are no debconf settings for gdm === fabbione kills gdm in the hope to see the errors [07:56] 5/09/04 11:30 Cleared Cleared at Destination Customs [07:56] ^ my laptop :) [07:56] wish i could get an rss feed of the shipping status ;) === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu [08:00] lol === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu [08:26] jdub: nice :) === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu [09:06] hrm [09:08] http://freedesktop.org/cgi-bin/viewcvs.cgi/xserver/xserver/hw/kdrive/ephyr/README?view=markup [09:08] ^ rad [09:08] seriously rad [09:13] yeah, that's mallum's baby [09:13] bah [09:14] (just found it on his blog) === hrdwrbob_ is now known as UbuntuBoB === Riff [~davyd@cook.theducks.org] has joined #ubuntu === bdale [~bdale@rover.gag.com] has joined #ubuntu [09:52] daniels: recordext... not in freedesktop cvs? [09:54] mdz: gar, vicious bastard :) [09:55] mdz: no go for me. mail on the way === HrdwrBoB [hrdwrbob@220-253-15-102.VIC.netspace.net.au] has joined #ubuntu === GmanAFK is now known as Gman === pitti [~pitti@195.227.105.180] has joined #ubuntu === seb128 [~seb128@ANancy-111-1-6-198.w81-249.abo.wanadoo.fr] has joined #ubuntu [10:03] morning [10:03] yoseb [10:04] hey hey jdub :) [10:04] Hi seb! [10:04] jdub: any progress on the menus plan ? :) [10:04] hey pitti [10:05] haven't had time :| === ik5pvx [~Bus_Error@paperino.noc.seabone.net] has joined #ubuntu [10:05] i was going to explain ti all, but i'll just send you a layout instead [10:07] jdub: do we have sanity happening ? [10:07] we should be in the "string freeze" phase now :p [10:08] lifeless: not really [10:08] seb128: :-) [10:08] seb128: no string changes :0 [10:08] :) [10:09] menu stripe menu stripe menu stripe! ;) [10:09] heh === Gman reckons an ubuntu launch button would go down very well too [10:10] vroooms [10:10] with a soyuz module sound :} [10:11] ok [10:11] I want a NEEDINFO state in bugzilla [10:11] seriously, my list of bugs is full of bugs waiting for informations [10:12] I'm responsible for two of those I think. [10:12] rhythmbox crash is from you IIRC [10:12] seb128, isn't there one already?! [10:12] yeah [10:12] and an acpi bug. [10:12] Gman: for GNOME yes, for warty no [10:12] oh, the battery applet preview in the preferences is completely horked. [10:13] seb128, suck, didn't think gnome had patched that bit of bugzilla :/ [10:13] in that its not a preview of what you get anymore. [10:18] lifeless: no it's not [10:18] lifeless: it's previewing part of the applet we don't turn on === jdub does not like battstat's prefs [10:20] if there is no way to turn it on, the preview should go.. right? [10:20] we just have it off by default [10:20] (That siad I really like the bar :[) [10:20] you can turn the batter graph on === Gman doesn't like battstat [10:20] Gman: you're biased [10:20] Gman: (we should've replaced battstat with battfink) [10:20] jdub: how? [10:20] :) [10:20] lifeless: Appearance > Show battery status [10:21] i am sure if it was "buttslut" instead of "battstat".... [10:21] lifeless: those prefs are buggy, click a couple of times [10:21] I have it on. [10:21] oh, ok. [10:21] right, thats better. thanks. [10:22] seb128: sent [10:22] jdub: thanks === pitti [~pitti@195.227.105.180] has joined #ubuntu [10:38] hmmmm [10:38] gdm didn't start on fresh install [10:38] can anybody reproduce it onm daily crack? === pitti [~pitti@195.227.105.180] has joined #ubuntu [10:42] TOO MUCH MAIL [10:42] hey Kamion [10:43] morning [10:43] fabbione: thanks for sorting #1019 [10:43] jdub: thanks for resubscribing! [10:45] sun are shipping webmin in solaris [10:45] ... [10:45] *bong8 [10:45] wtf? [10:45] Kamion: well i just used Matt suggestion, but at least the system is installable [10:46] Kamion: i am not sure is the rigth fix, but it works here [10:46] that's why the bug is still open === fabbione enters the deep doungeon of / on lvm2 on raid1 test again [10:46] fabbione: yeah, sure [10:46] ok, guess I should make the installer use these fancy new kernels [10:47] no more local packages in the cdimage tree! [10:47] Kamion: will they be part of sounder 9 ?? [10:47] the next one is 8, not 9 [10:47] oh yeah well [10:47] the last +1 ;) [10:48] probably, yes; given that we have the crack we may as well smoke it === fabbione likes crack.. and smoking [11:02] a friend has made a function in his IRC bot that quotes me :/ [11:02] did you point out to them that they are on crack? [11:03] Kamion: I still try to fix #996 (remember our discussion about the 'disk' group?) [11:03] Kamion: do you think it is a problem if I change only the removable devices to root:plugdev? [11:03] in udev? that seems to make sense, but I really don't know much about this stuff [11:04] Kamion: yes, in udev. I have a solution ready [11:04] Kamion: I'm just askign about the principle [11:04] Kamion: since you told me that you did not want to break the 'disk' group [11:06] I was mostly thinking about in Debian [11:07] Kamion: by now I only want to change the Warty version [11:09] pitti: what would happen on people upgrading from debian? === rburton [~ross@82-44-126-41.cable.ubr03.croy.blueyonder.co.uk] has joined #ubuntu [11:10] fabbione: if they upgrade udev as well, they get the new behaviour automatically [11:10] fabbione: that is, if they did not change the config files and rejected upgrading them [11:10] I don't think it's a problem for removable devices [11:10] fabbione: if they don't upgrade udev, they just will get the old behavior [11:10] hm, no actual kernel binaries for me [11:12] mdz: so, what does the latest build failure look like? [11:20] yuppi... segfault is still reproducible :-) === fabbione needs more coffee now [11:36] mmm, coffee, good plan [11:40] bye bye #462! [11:46] Sep 6 11:45:59 trider-g7 rpc.mountd: getfh failed: Operation not permitted [11:46] does anybody know what that means? === baptiste [~baptiste@217.26.192.224] has joined #ubuntu [11:47] EPERM on something, I'd strace it [11:47] getfh must be a local function in rpc.mountd [11:47] mounting the same export from another machine (in the same lan) works [11:48] oh different kernels === fabbione wonders [11:50] wasn't 2.6.8 having prolems with nfs? [11:50] 2.6.7 works [11:51] morning [11:51] 2.6.8 has issues, 2.6.8.1 works [11:51] morning thom [11:51] rburton: i am using debian kernels [11:52] hello thom [11:52] they should have the fixes [11:56] odd... restarting the nfs server fixed the problem [11:57] fabbione: about #1045, you're sure the buffer is lost ? IIRC you can switch back to the console [11:58] seb128: try this: go to console N, do a ls -lR, go to console M and back to N. You cannot do shift+pg up anymore [11:58] of course given that ls -lR will print out more than one page [11:58] oh ok [12:30] jdub: can you remind me what the policy on firmware in the installer was? [12:30] jdub: i.e. should I be adding these new firmware-supporting modules to our d-i? [12:31] firmware == okay [12:32] okie [12:32] one wonders how well it'll actually work given that d-i doesn't have the real hotplug in it, just a fake one ... [12:32] I suspect "not at all" === doko [doko@dsl-082-083-145-166.arcor-ip.net] has joined #ubuntu [12:39] seb128: do you instantly know which package contains the "Computer" menu entries? I would like to translate them completely [12:39] which Computer ? [12:39] the desktop place, or the menu one ? [12:39] seb128: the menu one [12:39] gnome-panel [12:40] seb128: thanks! [12:40] I'm making changes according to jdub's mail [12:40] and I'll mail the list about new strings to translate after that [12:41] seb128: whoopy, did jdub send a mail about this recently? I look in the archives... [12:41] just to me [12:41] I'm doing the menu changes [12:42] seb128: shall I send you the updated po file? [12:42] seb128: or jdub? [12:43] wait for my change to get the po [12:43] okay [12:44] I'll put all the po files somewhere with a list of new string to translate and ask for diffs [12:49] jdub: er, sure it is [12:49] maybe not in freedesktop.modules [12:52] jdub: are you going to put the current evolution-exchange in Debian soon ? Some guys were asking yesterday ... === mvo_ [~egon@suprimo-225.ping.de] has joined #ubuntu [01:11] Kamion: mdz wa suggesting to add the module lp in /etc/modules for i386 [01:11] Kamion: would it be possible to add ipv6 on all arch too? [01:12] I sort of feel that should only be added if netcfg detects that you have an ipv6 network (avoid any issues unless you need to) [01:12] of course, that's kind of predicated on netcfg having ipv6 support [01:13] make sence.. let's move it to hoary.. [01:13] i don't think hacking netcfg now is good [01:14] ipv6 support in netcfg should definitely happen upstream [01:14] hm, I have an IPv6 network, just about; I could theoretically do that [01:15] i don't think you can detect an ipv6 network without loading the ipv6 module [01:16] that means netcfg should load the module and then? [01:16] either you wait sometime for autoconfig to grab an ipv6 address [01:16] or ask? [01:16] because you might want to add ipv6 manually [01:16] I have no idea, I haven't thought about it :-) [01:16] dhcp+ipv6 is still vaporware [01:17] ehheeh [01:17] I dunno, I can chuck it in if you think it's a good idea [01:17] is there a risk of ipv4 stuff breaking? [01:18] not that i am aware of. [01:18] atleast i run ipv6 here and i didn't notice any breakage [01:18] well but once you have the script part that stick modules in /etc/modules, make it so that it can accept a list.. we can still change it after mdz will bless it [01:19] probably this is the easiest... [01:19] otherwise i will have to do some work on ifconfig [01:19] that's not really what i like to do [01:22] fabbione: no need for a script, d-i already has a single command called register-module which does all the work [01:22] ah.. [01:22] i didn't know.. but it sounds very good :-) [01:22] fabbione: wanna get mdz to ack that addition? [01:23] yeah i will mail [01:25] done === Kamion moves d-i to 2.6.8.1 for i386 === fabbione gets ready to test 2.6.8.1 [01:31] Kamion: are there finally 2.6.8.1 images for powerpc? [01:33] pitti: apparently not yet [01:34] Kamion: thanks. I already thought I missed sth. So d-i will use different kernel versions for different arches? [01:38] Kamion: do you have any pending lock on base-config? [01:41] pitti: d-i's fully capable of doing that, and it already does; it hopefully won't for the final warty release though [01:41] fabbione: not right now === Kamion thanks $DEITY^Wjoeyh for new-kernel-checklist.txt [01:42] Kamion: ok i will start preparing the universe/restricted stuff for apt-setup [01:43] but feel free to change things around [01:43] it shouldn't be too hard to merge later [01:43] I'll be buried in kernel changes for some hours yet, I expect [01:44] Kamion: no problem, because in any case i can't upload. the changes will require the archive to be ready [01:44] otherwise people will get broken sources.list [01:44] (that i don't think it's nice) === SteveA [~steve@adsl-213-190-44-43.takas.lt] has joined #ubuntu [01:46] I updated / upgraded using synaptic... warty hung. [01:47] on a reboot, the text mode boot up messages had corrupted characters [01:47] then, gdm wouldn't load [01:47] "There was an error loading the theme ubuntu" "can't open file /usr/share/gdm/themes/ubuntu/ubuntu.xml" [01:48] the "drawer" in my gnome panel doesn't work any more [01:48] well, not properly... it opens in a different place to where the drawer is [01:48] my "sticky notes" are all on desktop 1, with no text in them, and no application. [01:49] I'm guessing that my system is in an awkward state, as it was part way through setting up software when X hung === SteveA [~steve@adsl-213-190-44-43.takas.lt] has joined #ubuntu [01:57] SteveA: make sure both gdm and ubuntu-artwork are up-to-date [01:59] things work better now, after I logged into a text mode console, ran dpkg --configure -a and let it do its stuff. === StoneTable [~stone@c-67-167-111-186.client.comcast.net] has joined #ubuntu === Mithrandir [~tfheen@vawad.raw.no] has joined #ubuntu === Kamion wonders if linux-kernel-di-i386-2.6 built ... [02:16] SteveA: sounds like the best strategy, yeah [02:17] queue/accepted/linux-kernel-di-i386-2.6_0.64ubuntu1_i386.changes [02:17] woo [02:17] I tested with kernel-wedge build-all but that's not 100% accurate [02:18] (was cross-building) [02:24] :q [02:24] sorry, wrong window focus === ploum [~ploum@95-234.243.81.adsl.skynet.be] has joined #ubuntu === pitti [~pitti@195.227.105.180] has joined #ubuntu === debianist [~debianist@212.199.219.190.forward.012.net.il] has joined #ubuntu === debianist [~debianist@212.199.219.190.forward.012.net.il] has joined #ubuntu [02:40] first router install of ubuntu I think :) [02:43] anybody knows if the no-name-com server has limited bandwidth? [02:43] not especially [02:43] am trying to download the daily build, but seems rather SLOW than 2 days ago [02:43] you're using rsync I hope? [02:44] hmm [02:44] oops [02:44] stop my download immediately and rsync it ? (angle icone next my nick) [02:46] whatever you want [02:46] rsync -av --progress --partial is nice [02:46] --partial? [02:46] *maning* [02:46] am using gnome download manager. Like it alot [02:46] debianist: the server doesn't have a bw policy at the moment. [02:46] so you should be able to get all-you-can-eat [02:46] ok lifeless thanks. [02:46] far more likely to be network limits on your end [02:47] or delay-product issues [02:47] lifeless: I usually only get around 500kB/sec, which is fairly slow, considering my pipe is a 100Mbit. [02:47] and as I'm in .no, the delay is fairly neglible. [02:48] HcE: --partial keeps partial downloads. Very nice if you run out of space or something. [02:48] I don't know how fat the pipe out of the LAN is [02:48] Kamion: 10Mbit ethernet at the moment. [02:48] Mithrandir: saw that, nice for backup-solutions [02:48] there is a 1Gb lead there, but AIUI its not plugged in yet :} [02:49] ok, I think all the pieces for 2.6.8.1 in d-i are there now [02:49] 100Mbit currently [02:49] just a matter of waiting for builds etc. [02:49] thom: oh you got it plugged in? cool. === lifeless is obviously out of date [02:49] lifeless: 100, not 1000 [02:49] thom: is the 1Gb still planned ? [02:50] we have a fibre drop [02:50] but since no-one confirmed it was fibre, nothing to plug it into [02:50] Kamion: can I remove the 2.6.7 d-i stuff? [02:50] oh, hahahaha. [02:50] that so doesn't surprise me. [02:51] that initial hardware purchase was chaotic. [02:51] kamion: the NBS udebs I mean [02:51] elmo: hm, is it a problem to leave it until the new debian-installer has built, just to make sure? [02:51] kamion: no that's fine [02:51] after that, no problem [02:54] Mithrandir : are you at work or something?? 500kB/sec ???!! The best we can get here is around 3MB download, that is mega BITS [02:54] debianist: university network. [02:54] Mithrandir : it's lucky if I get to download something in 92k bytes per second. [02:55] debianist: I have a 2.4-3Mbit at home [02:56] debianist: .il is isreal, isn't it? [02:56] yep [02:56] the most advanced third world country :) [02:56] heh [02:56] I thought you were fairly well-connected there? [02:56] we are, but the quality of connections here are is so low compared to the states [02:57] it's a pain. [02:57] oh, ok. [02:57] hmm yeah I seem to get 200kb/s [02:57] I mean we have million types of cellphone, companies that sell advance software and equipment to words biggest coprs and government. [02:57] from a 100mb link [02:57] yet, in house we kind'a dumb [02:57] :) [02:58] I am on the cable infrastructure for the internet access, and there are still region in which they just cannot overcome the technical difficulties which prevent the service from being relaient [02:59] sounds sucky.. [02:59] you think that's bad [02:59] come to australia [02:59] first world country [02:59] third world service [03:02] HrdwrBoB : it is? [03:03] HrdwrBoB : well, consider the VAST (and I mean that) distants you have there, you have an exucse..What about israel? a country the size of NYC ? what's our excuse ? :)) === HrdwrBoB [hrdwrbob@220-253-15-102.VIC.netspace.net.au] has joined #ubuntu [03:03] Hello === debianist is anxiuosly waiting for his warty daily to download... [03:04] Will the CD 8 be released today ? [03:04] I swear to god I am going to kill the person that made ctrl+x quit [03:04] ploum: probably :-) [03:04] any fixes from the one built on saturday ? [03:04] debianist: totally different kernel? :) [03:04] debianist: I got a flyer in my mailbox, $80/month for 2gb. [03:04] Kamion, so I will wait and hang around there before testing :-) [03:04] debianist, for start, you could complain to your ISP and tell them to buy more bandwidth from their upstream (which I happen to work for ;-) [03:04] ploum: might be quite late today though [03:05] debianist: also, vast distances, but almost the entirety of the population is fairly closely living in the capital cities [03:05] ik5pvx : Global ONe? [03:05] ik5pvx : AT&T ? :) [03:06] ik5pvx : funny enough, the ISP has nothing to do with it. They're the coolest ISP here in israel, however in order to give you cable broadband, they MUST relay on the cable company infrastructure, and those guys, they SUCK BIGTIME [03:06] ah, I see... it's the final piece of wire that sucks... oh well [03:08] BTW : do we need a ubuntu hebrew translator? [03:08] or is there someone already on this? [03:09] ik5pvx : which one do you work for? === debianist is wondering where is the place to find about changes from on daily to another. [03:15] seb128: I'm taking care of the German translation of gnome-panel [03:15] pitti: ok [03:18] debianist: we don't really have a translation infrastructure at the moment; we will eventually [03:19] Kamion : Well, I'll start the hebrew, when the infrastructre realized, it would be ready [03:20] what parts besdies gnome and d-i (which I believe have mostly been localized) need localization? [03:21] The d-i Ubuntu branding for Hebrew was pretty straightforward, IIRC; it's not one of the languages where proper nouns seem to get declined. [03:22] Kamion : please explain "branding" ? [03:22] debianist: s/Debian/Ubuntu/, basically [03:22] seb128: I just noticed an error in the English gnome-panel sources: there is no word "informations", it's information. [03:23] should be information or informations ? [03:23] seb128: maybe you want to send an updated mail? Or just do a sed -i 's/informations/information/g'? [03:23] seb128: just "information" [03:23] Mithrandir : oh [03:24] pitti: "Display informations about Ubuntu", you would remove the "s" to information in this string ? [03:24] does ubuntu allow for resize of a FAT32 or an already existing ext3 one for install? [03:24] yes [03:25] seb128: ^ [03:25] ok, so please reply to the mail on the list with the corrections [03:25] so everybody works on the right strings [03:25] I'll adapt the sources on the next upload === thom -> install latest daily i386 [03:26] Kamion : most of hebrew strings for ubuntu already exist? [03:27] seb128: Okay, I'll followup [03:28] debianist: should do [03:29] Kamion : you mean take them from sarge's d-i? something like that? [03:31] yes [03:31] exactly [03:32] ok [03:33] and I guess d-i's docs about localization apply all the way down :) [03:35] more or less, although we're trying to avoid diverging too much unnecessarily as it makes changes hard to merge [03:36] ofcourse [03:37] ubuntu is supposed to feed debian back with it's patches and improvments. [03:37] Kamion : where lies the list of changes between daily builds? === fabbione gets spammed by the BTS [03:39] debianist: there's none in that exact form; there's a warty-changes list (no idea if it's generally readable) [03:39] seb128: shall we add a complete po file header or are the translations itself enough? [03:39] Kamion : basically, there are locale strings for each language right? All I have to do is take the english translation and replace the strings [03:40] pitti: the translation of the string is enough, I'll just append it to the nn.po files [03:40] debianist: if you're not already familiar with gettext-based translation, look for documentation elsewhere on that; there's plenty in any number of free software projects [03:43] Kamion : ok, google is my best freind :) [03:43] Kamion : BTW any success with using gnome-volume-manager and dbus-1 under ubuntu? === rburton [~ross@82-44-126-41.cable.ubr03.croy.blueyonder.co.uk] has joined #ubuntu [03:47] debianist: dunno, I'm just the installer guy ... [03:48] Kamion : installer guy? you make the installer? [03:48] does anyone here have one of those ipw2100 or ipw2200 wireless network cards? [03:48] debianist: I lead the Ubuntu work on debian-installer and CD image production, yes. [03:50] doesn't leave me much spare time :) [03:52] Kamion : oh right. reagrding CD image production, I'm offering my help in making the live CD, I said that before right? :) [03:52] debianist: you'll have to talk to Alex de Landgraaf about that, he's the live CD guy ... [03:52] debianist : plus I'd be cool helping you whenever you feel you collapse under the pressure :) === ploum [~ploum@95-234.243.81.adsl.skynet.be] has joined #ubuntu === Riddell [jr@muse.19inch.net] has joined #ubuntu === thom [~thom@amnesiac.heapspace.net] has joined #ubuntu === lifeless [~robertc@dsl-156.23.240.220.rns02-kent-syd.dsl.comindico.com.au] has joined #ubuntu [03:54] the actual live CD build process (when it becomes really official) will require login access to our servers, so it'll probably have to be employed-people-only for now; hopefully eventually things will be in a form where we can take patches in some convenient way [03:57] Kamion : hmm, so ubuntu is targated at being an employed only distro? :( [03:58] Kamion : Alex de Landgraaf has an email address ? something of that sort? [04:01] debianist: not at all; just certain bits are still hard to do community cooperation on, because we haven't quite got ourselves set up for it yet [04:01] we're still setting a lot of community things up [04:02] hence "for now" :-) [04:02] hrm, gdm didn't get started when base-config finished [04:03] and X still ends up one res too low [04:03] alextreme@xs4all.nl [04:03] (current daily) [04:03] thom: any errors during base-config? [04:03] Kamion: no [04:04] Kamion : oh i see. say kamion, what would you think would be the criteria for future hiring by Canonical? [04:05] I wouldn't like to presume [04:05] @all: if I do 'gksudo something' from the console, gksudo segfaults at the end, regardless of what command I'm executing. Does this occurr only at my box? [04:05] pitti: are you sure that's an utf-8 file ? I've weird chars in utf-8 and iso modes ... [04:06] seb128: I'm looking again at it [04:06] thom: ah, it happened to you too? (the gdm start)... [04:06] thom: for the resolution too low, we need to change xresprobe. [04:07] thom: X doesn't guess or scale anything. It just accept xresprobe output as-is [04:07] seb128: maybe the mta/mua messed it up. I send it again gzipped [04:07] ok, thanks [04:07] fabbione: it's the old thing - i get 1280x1024 added to the end of the line, but this is an lcd monitor, and that's the native res [04:08] thom: yes. i remember that... we need to change xresprobe to not sort resolutions [04:08] tho it won't work on my machine... [04:08] but i guess the others have higher priority [04:11] seb128: does gksudo crash at the end on your box? === fabbione needs to learn how to use lvm snapshots [04:14] pitti: at the end of what ? === Kamion uploads may-work-if-you're-really-really-lucky-no-I-can't-test-this firmware loading support for d-i [04:16] seb128: see above. whenever I execute 'gksudo whatever' I get a segfault at the end [04:16] seb128: just type 'gksudo true' in a terminal [04:17] pitti: no, doesn't crash [04:19] pitti: the translation is ok this time [04:19] seb128: odd. thanks for trying. [04:20] Kamion : this is PowerPC firmware support? [04:21] architecture-independent; it's for drivers (mostly for network cards) that need binary-only firmware [04:23] pitti: gksudo works fine here, too, fwiw [04:23] has anyone uploaded anything to restricted yet, btw? if not, please shout when you do, I'm not sure it'll work ;) [04:23] elmo: what happens if we do a foobar upload to test? [04:23] elmo: remind me what the policy there is? firmware seems to be headed into main ... [04:24] kamion: non-free software, e.g. binary only drivers [04:25] fabbione: might as well wait until there's something to actually upload to there [04:25] kamion: (that we want to ship and support (as far as the license allows)) [04:25] kamion: and under the liberal, "non-free software" means "software programs running on the host CPU" definition [04:26] elmo: we can start with NVIDIA crap :-) [04:30] thom: thanks for testing. This is odd... Maybe a powerpc specific thing. I'll debug this, I don't like crashing root programs [04:32] elmo: aha, ok [04:33] elmo: did the most recent debian-installer upload build, or did it crap out hideously? [04:33] katie@jackass:/srv/ftp.no-name-yet.com$ m debian-installer -S [04:33] debian-installer-manual | 20040801ubuntu8 | warty | amd64, i386, powerpc [04:34] ah, just not on the mirrors [04:35] ... there it goes [04:45] can anyone suggest where would help needed currently in ubuntu? besides testing? [04:45] Kamion: testing now === ploum [~ploum@185-233.243.81.adsl.skynet.be] has joined #ubuntu [04:46] fabbione: testing which? [04:46] Kamion: d-i 2.6.8 [04:46] Kamion: from ubuntu8 that it just hitted acukland [04:46] that was quick, I only just built the CDs [04:46] Kamion: netinstall :) [04:47] aha [04:47] well it's installing base now [04:49] just testing? [04:49] debianist: any kind of testing is important [04:50] fabbione : ok, I have today's daily downloading and I am going to install a simple router system..Guess I am going to test the "Custom" seed [04:50] custom isn't a seed :-) [04:50] oops a d-i task [04:51] damn, forgot to document custom in the latest d-i upload, oh well [04:51] I'm still pretty confused about those [04:51] Kamion: well first d-i stage is completed.. rebooting now [04:51] seeds are really for internal distribution management, they aren't supposed to be exposed to users [04:51] Kamion : are you talking about code documentation or task ? [04:52] I don't know what you mean ... [04:52] seeds are used to generate the list of base packages for debootstrap, the list of what gets installed by default, the list of what goes on the CD, and the list of what goes in the archive [04:52] that's all [04:53] apart from that you can pretend they don't exist :) [04:53] ok === pitti [~pitti@195.227.105.180] has joined #ubuntu === debianist is feeling like a warthog on hold :) === Mithrandir sends debianist a SIGCONT === debianist goes out of hold, restarts and resume on hold state :) === debianist wish his warty iso image would finish downloading already. [05:00] the new init stuff is ugly as sin if you're not on the console :( [05:00] elmo : you don't get to see nothing if you're not at the console. only some ubuntu artwork and a progress bar [05:01] sorry, init.d stuff [05:02] debianist: the splash screen changes actually didn't make it into warty; they've been postponed until the next release [05:03] Kamion : oh, having seen this on the ubuntu live cd made me think this how the boot looks [05:04] the live CD is strange in a lot of ways still [05:07] Kamion: sounder8 is available? [05:08] lamont: not quite === lamont goes back to lurking [05:08] ping ping [05:08] actually, today is work on the van day, since it's putatively a holiday. === Kamion is burning an i386 candidate at the moment [05:10] [Lag 65] <- something is not really working [05:10] Kamion: 2.6.8 is a go for me [05:11] cool [05:11] hey lamont [05:11] Kamion: have you had any reports about successful amd64 installs? [05:11] fabbione: did base-installer do the right thing, namely install linux-image-2.6-386? [05:11] Kamion: i was checking right now :-) [05:12] Mithrandir: mdz said something about grub working now; I killed the broken local grub package from the cdimage tree this morning, so if any image will work then it'll be the current one [05:12] linux-image-2.6-686 [05:12] Kamion: ok, cool. [05:12] Kamion: i would say yes [05:12] I'll give it a shot when it releases, then [05:12] Mithrandir: there's a suitable daily up now [05:12] Kamion: and of course as dependency linux-image-2.6.8.1-686 [05:13] fabbione: excellent [05:13] Kamion: good job :) [05:13] now.. let see what happens upgrading a normal box [05:13] Kamion: ok, but my amd64 system is home ATM, and I'm not there until tonight [05:13] Kamion: my only showstopper now is that grub-installer doesn't add noapic to menu.lst :/ [05:14] (given that i booted the installer with noapic [05:14] ) [05:14] thom: there've been fixes for that upstream, I can import them, probably not for this release though [05:16] ok. cool [05:16] damn, all this ipw2[12] 00 stuff is BIG === Kamion is tempted to make a separate nic-firmware-modules udeb or something === debianist [~debianist@212.199.219.190.forward.012.net.il] has joined #ubuntu [05:18] ... except nic-extra-modules already depends on firmware-modules, oh well [05:18] what's npmccallum-bootsplash/current ? [05:18] an old experiment [05:18] I noticed it's slightly lighter than the daily [05:19] that's because it's based on a very old daily that didn't contain a number of the packages that the current one does [05:20] hmm. I tought I was asking one question too many, thanks for the exposure :) [05:20] thom: the same goes for one of my laptops and acpi=off [05:21] seb128: bah, my post to sounder got stuck because I didn't use the correct from. [05:21] who's listadmin? [05:22] jdub I think [05:22] currently there's only one mailing list for the canonical team? [05:22] nice === Oskuro resends again [05:23] ok [05:23] we can build nvidia kernel module directly on the buildd [05:23] ugh, why is my orinoco_cs card not being detected? [05:29] Kamion: i can't test more than this.. but it looks good to me [05:29] what should I have in /etc/sudoers in a default warty install ? [05:30] SteveA: [05:30] # Added by Ubuntu installer [05:30] Defaults: timestamp_timeout=0 [05:30] ALL=(ALL) ALL [05:31] thanks. [05:31] I don't have that [05:31] I guess because I installed before it was so [05:31] SteveA: good catch ;) [05:32] thom: did you report gdm not starting on a fresh install? === ploum_ [~ploum@73-243.242.81.adsl.skynet.be] has joined #ubuntu [05:34] fabbione: no, will do [05:37] thom: i can do it.. don't worry [05:38] seb128: dude === seb128 hides [05:38] i am not sure it's a gdm or base-config problem here [05:39] thom: gdm? [05:39] thom: what ? === savs [~savs@jicbioinc1.ext.uea.ac.uk] has joined #ubuntu [05:39] seb128: fresh install. gedit appears to be http handler again [05:40] fabbione: base-config, i guess [05:41] thom: same problem, gconf key for enable off ? [05:42] thom: no. it's gdm init script [05:43] seb128: yeah [05:43] seb128: both http and https have command and enabled set to [05:44] thom: base-config calls gdm restart, but the init script dies on stop (probably because gdm is not running and the lsd^Hb stuff doesn't like it) [05:44] ARGH [05:44] oh, not MORE initscript breakage === thom beats npmccallum with a big stick [05:44] yeps [05:44] confirmed [05:44] if i add a || true to start-stop-daemon call it works [05:45] baby jesus is weeping now [05:45] that doesn't sound like the right fix [05:45] --oknodo on the --stop maybe? [05:46] thom: LC_ALL=C HOME=/root GCONF_CONFIG_SOURCE=`gconftool-2 --get-default-source` gconftool-2 --makefile-install-rule /usr/share/gconf/schemas/desktop_gnome_url_handlers.schemas && killall gconfd-2 and then look at the key again helps ? === StoneTable [~stone@c-67-173-130-84.client.comcast.net] has joined #ubuntu [05:46] Kamion: i just checked the reason why it was failing [05:47] Kamion: i know it's not the proper fix [05:47] thom: oups, s#/usr/share/#/etc/# [05:47] fabbione: sure [05:47] but generally this lsb thingy is on crack [05:47] yeah [05:49] seb128: same [05:49] thom: very weird === Kamion bangs his head against several brick walls and decides to fall back to Sounder 7 to make sure the hardware's actually still working [05:50] thom: could you check the schema file, the true value ? [05:50] Kamion: please don't destroy your head, it's a very nice and useful head to have around [05:51] if I knew which of ddetect, discover1, discover1-data, the kernel, or my hardware to blame, it would help [05:51] still the orinoco problem? [05:54] seb128: i don't have a .gconf/desktop/gnome/url-handlers/http/ [05:54] yeah; I can't even figure out how discover finds orinoco_cs, or if it's entirely hotplugged, or what [05:54] thom: if you have not changed any value that's normal [05:54] thom: ~/.gconf is for user settings, ie: if you change from default values [05:54] Kamion: for me, orinoco_cs is loaded my the pcmcia layer [05:54] yup [05:54] Kamion: I think it's handled by hotplug or pcmcia_cs [05:55] /etc/pcmcia/config [05:55] so, where should i be looking? /etc/gconf/schemas/desktop_gnome_url_handlers.schemas ? [05:55] Mithrandir: that sounds right, but that makes it kernel breakage :( [05:55] thom: for the schema value yes [05:56] Kamion: what kernel version? [05:56] Mithrandir: 2.6.8.1, new kernels in warty [05:56] this sucks [05:56] /schemas/desktop/gnome/url-handlers/http/command [05:56] /desktop/gnome/url-handlers/http/command [05:56] gnome [05:56] string [05:56] mozilla-firefox %s [05:56] with all the changes we did to X build system, S390 almost managed to install... [05:56] it failed at like 99.5% [05:56] thom: and the enable one ? [05:56] as usual.. not enough space [05:57] default is true [05:57] hum, ok [05:57] and do you have a /etc/gconf/gconf.xml.defaults/desktop/gnome/url-handlers/http/%gconf.xml ? [05:57] [05:57] [05:58] all seems fine :( [05:58] Kamion: hm :/ [05:58] in gconf-editor, if you browse the schemas tree, the key is in it ? [05:58] i will create a test user and see if it happens there [05:59] seb128: yeah === Keybuk [scott@62.49.129.40] has joined #ubuntu [06:05] test@fandango ~ $ gconftool-2 -R /desktop/gnome/url-handlers/http [06:05] needs_terminal = false [06:05] command = (no value set) [06:05] enabled = (no value set) [06:06] brand new user [06:08] AAAAAAAAAAAAAAAAAAAAAARRRRRRRGH [06:08] lsb initscripts broke PCMCIA in d-i [06:10] thom: thinking about it ... [06:10] thom: any way to strace -e open ? === Kamion is not happy about these changes happening at this point === kagou [~kagou@AMontpellier-251-1-7-239.w83-113.abo.wanadoo.fr] has joined #ubuntu [06:11] hi [06:11] Kamion: you and me and everyone else,too [06:11] i think === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu [06:12] hey kagou [06:12] kagou: please reply to the comment on bug reports [06:13] kagou: I've added 2 comments in 4 days on the samba bug .... [06:13] ok seb128 [06:13] kagou: you don't get the mails from bugzilla ? [06:13] Kamion: yeah, we've been making far too many "feature" changes too close to the release [06:14] I think next time having an open grumpy branch/repository will help eliminate that a little [06:14] Kamion: i think we should rediscuss this LSD thing and revert it completely [06:14] seb128, i was not here since 1 day [06:14] all these initscript changes are extremely hard to review and have so far broken the installer in two variously subtle way [06:14] s [06:14] I fully expect more breakage [06:15] fabbione: it's got Mark's total backing, we're doomed [06:15] Kamion: i don't think so [06:15] we can still propose to revert the changes with the facts that they break A LOT [06:16] well, I'm going to fix this one up somehow, since it's either that or no Sounder 8 [06:16] I can't release with broken PCMCIA [06:16] well gdm needs love too [06:17] and probably many others we didn't notice [06:17] anyway i need to go for a while now [06:17] a compatibility shim in rootskel is probably easiest for PCMCIA [06:17] bbl [06:20] - /sbin/cardmgr $CARDMGR_OPTS [06:20] + /sbin/cardmgr $CARDMGR_OPTS 2>/dev/null [06:20] I'm also unhappy about rubbish like this [06:20] prettiness is not a justification for throwing away errors [06:21] Kamion: but you can't colour stderr! [06:21] I guess chat to mdz, and get him to chat to Mark/Nathaniel is the right way to do that [06:21] thom: why can't you? [06:22] Mithrandir: i was being sarcastic [06:22] thom: :P [06:24] the trouble with this stuff is that usplash is the right way to do it, entirely hide stdout/stderr [06:24] and when you've got usplash, you *want* as verbose a start-up screen as possible, so don't want all this pretty [06:25] looking at the API, if this is what the LSB recommends then colour me extremely unimpressed [06:25] it's practically designed to encourage non-robust init scripts [06:26] aye, I wasn't impressed with the start/stop changes [06:27] which reminds me, thom: found a bug in the apache2 init script I think; when it does invoke-rc.d apache2 force-reload in the postinst it force-starts apache2 if it wasn't running, and as force-reload doesn't check /etc/rcX.d it means you end up with apache2 running when you don't want it === sabdfl [~mark@wblv-251-236.telkomadsl.co.za] has joined #ubuntu [06:30] hi all [06:31] Keybuk: hrm, yeah. that's a bug [06:32] hi Mark [06:33] hey sabdfl [06:33] seb128: to add to the fun, the preffered-applications app segfaults [06:33] good to be back in the bandwidth game! [06:34] thom: have you straced it ? [06:35] sabdfl: still in .za or back in .uk again? [06:35] still in .za, here till Sunday [06:35] seb128: strace _what_? [06:35] thom: gconftool-2 -R ... [06:37] open("/usr/share/locale/en_GB/LC_MESSAGES/GConf2.mo", O_RDONLY) = 12 [06:38] only thing that looks remotely useful [06:38] thom: could you try to register the schema in C locale ? [06:40] haha [06:40] Failed to write "/etc/gconf/gconf.xml.defaults/schemas/desktop/gnome/url-handlers/http/%gconf.xml": Failed to open "/etc/gconf/gconf.xml.defaults/schemas/desktop/gnome/url-handlers/http/%gconf.xml.new": Permission denied [06:40] are you root ? [06:41] b...b...bu...bu...but I'm ROOT! [06:44] LC_ALL=C gconftool-2 -R /desktop/gnome/url-handlers/http needs_terminal = false [06:44] command = mozilla-firefox %s [06:44] enabled = true [06:45] ok [06:45] problem with your locale [06:46] care to fix it? :P [06:46] I've a gconf hacker in query :) [06:46] tell mark to fix it faster then ;-) === tvon|x31 [tvon@dsl093-119-225.blt1.dsl.speakeasy.net] has joined #ubuntu [06:48] that's not mark, I try to not bother the same people all the time :p [06:48] heh [06:50] thom: what's the deal with sound on the X40? [06:52] sabdfl: you should have snd_8x0m loaded [06:55] and thus, it should all work [06:55] (and hotplug does that for me) [06:55] thom: snd_intel8x0? [06:55] yes that's loaded [06:56] but aplay foo.wav doesn't do anything [06:56] alsamixer show you have things loaded? [06:56] (and do you have volumes turned up) [06:57] hmm... turned out to be PCM volume turned down. I've never edited that, why is the volume not a sane default? [06:59] because alsa isn't vaguely sane [07:00] also, what's a sane default? === lamont gets dragged out the door [07:06] thom: and... what's the plan to make it *just work*? [07:06] for example, if there are no stored settings, can we detect PCM and master volumes and set them to 70%? [07:07] and, unmute them (master was muted on all computers I encountered so far) [07:08] yes! [07:09] sane defaults relate to the soundcard, the speakers, and a stack of other stuff [07:09] if you turn a master up to 70% and the person has headphones on, you could blow their ears off [07:10] *shrug* === ploum__ [~ploum@211-238.243.81.adsl.skynet.be] has joined #ubuntu [07:11] thom: we could set PCM to 70%, unmute master and leave master at 0. Many users will be able to find the loudspeaker icon in the gnome panel and drag it up [07:11] pitti: that would seem saner [07:11] thom: but many will not try to look for a separate mixer application to control the other channels [07:12] pitti: right click on the loudspeaker, open volume control, same as windows [07:12] thom: oh, okay. I don't know Windows :-) [07:13] thom: so far I even have to get to know GNOME [07:13] thom: (I'm an old text console freak and it's hard to change such habits) [07:20] setting master to 0 is no better than mute. [07:20] at least set it to 50 === fabbione can't use 50% with his card/speaker.. it would break walls and windows! [07:21] hey sabdfl :-) [07:21] that way the user thinks "oh, the sound is too low", then immediately finds the volume control and turns it up [07:21] fabbione: yo! [07:21] sabdfl: interesting that that doesn't apply for 0% though ... [07:21] I guess the slightly audible noise makes them turn it up, but no noise makes them assume it's broken [07:22] sabdfl: i suggest a mosedt 20 to 30% [07:22] no more than that [07:22] people can perceive "noise" === fabbione heads to the kitchen [07:24] fabbione, sabdfl: if I leave my mixers below 50%, I cannot hear *anything*, not even noise [07:25] if sound cards have such massively diverging amplification characteristics, we should keep the default low to be on the safe side [07:26] pitti: very :-/ my pc downstairs is plugged in to the DTS amp ... ESD beeps at anything greater than about 5-10% break windows [07:26] morning [07:27] hi mdz === debianist overviews the material written for ubuntu know how and start up , and suggests starting to work on a documentation repository. first doc in it would be "About Ubuntu" , "First Steps With Ubuntu" all targeted at the non technical user [07:27] Hmm. This means that the safest default is really 0% for master and sth reasonable for PCM [07:27] mdz: Good morning! [07:27] pitti: or 2% to usually make _some_ noise. [07:28] Mithrandir: All computers I know of need at least 60% to produce sth audible, so 2% is not much different from 0... [07:28] I'm not entirely sure we can change that easily [07:28] I think it may be set by the driver [07:28] Mithrandir: what the hell did you connect your sound card with? A 1000 W amplifier? :-) [07:28] yeey! my warty iso finished downloading. [07:28] pitti: my laptop needs that, but my home box is a lot happier with ~10-20% [07:28] off to my CDRW drive [07:28] pitti: nah, a 35W amp and a 100W amp [07:28] Mithrandir: just kidding. [07:29] and I like to adjust the volume with my computer, so I turn the amps up === pitti thinks about a clever speaker-microphone feedback volume autodetection thingy [07:29] pitti: you have -2 weeks ;-) [07:30] or FAQ "Why can I hear no sound?" "Did you turn the volume up? Did you? Bet you didn't? Bet the volume's turned down, or even muted? You FOOL! PLUG IT IN!" [07:30] thom: should be more than enough :-) [07:30] though that's just reminded me that the old SB manuals did used to have "Is the computer turned on?" under the "I hear no sound" troubleshooting [07:31] thom: no, seriously, if we break a single window with the default, somebody will kill us, or do sth even worse [07:31] fabbionne - how loud is it if everything is at 70%? [07:31] Keybuk: "Did you have turned on your ears?" [07:31] sabdfl: 70% on my desktop will just about bounce a monitor off my desk [07:32] thom: could make a nice startup animation :-) [07:32] yeah, 70% on desktop upstairs would probably quite effectively annoy people down the road; 70% downstairs would quite effectively annoy elmo, 100 miles away [07:32] Keybuk: do you have a weapon certificate for this thing? [07:33] keybuk: that's just insane, what if some cutie using your pc turns the volume up and then plays something? [07:33] sabdfl: they go deaf. [07:33] should always have it so 100% won't break things [07:33] heh, the computer upstairs is plugged into my old pro-logic amp [07:33] the computer downstairs is plugged into a THX-certified DTS amp [07:34] doesn't matter what it's plugged into. [07:34] sabdfl: BTW, do your hotplug USB drives finally work automatically? Last time we did not come to debug it. [07:34] point being, there's a reason debian ships the mixer muted [07:34] set the volume control on the amp so that a 100% soundcard output still won't blow it [07:34] oh, it won't blow the amp [07:34] the amp has no problem with 100% input ... it outputs it pretty good too [07:34] pitti: they are getting mounted correctly now, but nautilus isn't popping up to display the files [07:34] it tends to blow nearby eardrums though [07:35] keybuk then that's your bug :-) [07:35] sabdfl: does the device appear in the Computer window? [07:35] though, taking a different POV for a minute [07:35] I suspect mute is actually a really *good* default anyway [07:36] thinking of laptop-users-on-the-train kind of scenario [07:36] oh! I just plugged in a device and it popped up perfectly! [07:36] cool - must be this afternoon's update - thanks pitti! [07:36] I'm surprised there aren't more killings from people opening their Macs at 7.45am when everyone else on the train is trying to sleep === pitti jumps for joy [07:36] Keybuk: *g* [07:37] keybuk: no it's not, people expect sound to work, and they can turn it off if they want to [07:37] sabdfl: or turn it on if they want to? [07:37] Keybuk: there is a nice MacOS X application to disable this annoying startup sound [07:37] I suspect the real problem is that it's not immediately obvious the sound is working, and just muted [07:38] I get a red slash through the speaker, which does somewhat suggest "sound not working" [07:38] sabdfl: it *could* be today's update, depending on the type of device you have. But it's good to hear that it works now [07:38] "red slash through the speaker" is "YOUR SOUND CARD IS BROKEN" [07:38] there needs to be a slightly more gentle mute indicator :-) [07:38] pitti: works now, didn't this morning :-) [07:39] and then it wouldn't matter what the default was ... if it was obvious when you glanced at the speaker than it was just muted, you'd instinctively raise the volume [07:39] a gag in the speaker's mouth? [07:39] keybuk: yes, it could work that way, but the normal person expects sound to work [07:39] Kamion: file a bug about that 2>/dev/null, please [07:39] how else can you tell the difference between a PC where sound is *not* working, and one where it is simply still muted? [07:40] sabdfl: playing a sound and checking the volume [07:40] um, after wasting time surfing the net with a gnawing sensation that this Ubuntu thing was a bad idea..., [07:40] Kamion: i386 kernel built fine; powerpc and amd64 weren't implemented until this morning [07:40] I'll be doing a new upload soon [07:41] please, let's just set it to quiet defaults. 99.9% of people do NOT have their laptop or computer hooked to the local cinema sound system. [07:41] heh, I'd be happy with a 25%-50% default to be fair :p [07:41] people for who that's loud (desktop users, mainly) will turn it down on the first beep, people for who it's quiet (laptop users) will turn it up if they want [07:42] mdz: are you making sure that the amd64 kernel has the bd_claim patch applied? [07:42] the issue is how easy that is to do ... I suspect it's embedded deep in the kernel/device layer for each different soundcard :-( [07:42] fabbione: this is for first-time install [07:43] Mithrandir: all arches have it [07:43] I have *always* found that a sane default is 70% on PCM, 70% on Master [07:44] Linux ubuntu 2.6.8.1-1-amd64-k8 #1 Sat Sep 4 04:05:46 PDT 2004 x86_64 GNU/Linux [07:44] Linux warty 2.6.8.1-1-powerpc #1 Sat Sep 4 06:33:43 PDT 2004 ppc GNU/Linux [07:44] mdz: ok, cool. [07:44] mdz: 2.6.7 in ubuntu does not have it, IIRC [07:44] Mithrandir: right, this is Herbert's 2.6.8.1 kernel [07:44] which includes bd-claim [07:45] goodie, goodie. [07:45] it's uploaded now? === debianist installing warty under QEMU [07:45] Mithrandir: no, but it will be shortly [07:45] ok [07:45] I assume it will need some love from lamont and elmo [07:45] true [07:45] having about 800 NEW binary packages [07:45] sabdfl: yeah, but then you find someone who it's not only a good default but destroys their system ... and you find a nice lawsuit on your desk the next morning :-/ [07:46] what's "detecting hardware" ? :) discover working with hotplug something? [07:46] ok, think about this [07:47] debianist: discover, actually [07:47] it could only destroy their system if they've hooked it up to something through line-out [07:47] and then hooked that up to speakers that cannot handle it [07:47] mdz : ok, the installer looks nice, works damn fast under QEMU [07:47] debianist: you should see it on real hardware :-) [07:47] 100% output from the soundcard is the same as 100% from a CD line-out [07:47] yes, it should be no different than plugging in any other audio device [07:47] so then, if they plugged in a CD to the line-out, and played it, it would blow their system [07:47] you need to be careful with the level if you're plugging it into an amp [07:48] mdz : i'd like to install it on my pIV 2.6c GHZ D865PERL machine :) [07:48] mdz : I need to release some space off the sid's parition [07:48] sabdfl: one difference is that some PCs have a line out which doubles as an amplified output [07:48] you have to be careful *both* ways ... underpowering an input can be just as bad to an amp/device as over-powering it [07:48] now, i don't think anyone would believe it was a reasonable claim if you setup your amp so stupidly, then plugged in a CD [07:48] still, if someone's system is about to be blown, they'd hear a loud HUMMM and CLICK when they connect it [07:48] why can underpowering be bad? [07:49] and a CD in this case is no different to Warty [07:49] mdz: only if they hotplug it. [07:49] sabdfl: CD players have well-defined output levels. Sound cards don't. [07:49] Mithrandir: they should still get a rather loud hum regardless [07:49] mdz: not if you have a good sound card. :) [07:50] computer audio outputs are _noisy_ [07:50] mdz: not to mention dirty [07:50] yes [07:50] you'd expect random white noise as you moved the mouse if you got it wrong [07:50] some of mine play an enchanting little tune when the CPU is busy [07:50] I guess I'm weird, then. Having a ~150USD sound card with RCA connectors and all. [07:51] I think they might be gold-plated, even, as well. [07:51] on this ppc system, I've added therm-windtunnel to /etc/modules [07:51] but it _never_ gets loaded at boot [07:51] but when I log in and modprobe it, it works fine [07:51] elmo: ping? [07:51] mdz: ? [07:52] elmo: uploading a new linux-source-2.6.8.1 which produces about 3x as many binary packages [07:52] elmo: could you see that it gets the love it needs? [07:52] mdz: yeah [07:52] thanks [07:52] Choosing the "Logical volume manager" says it needs to write changes to disk, when I allow it to, and choose finish and continue I get a red screen telling "no root fileystem detected, please go back and fix thsi" [07:53] it's in the queue on jackass for the next */30 [07:53] after that, when choosing "manually edit paritions" I get nothing and it goes back to "Partition Disks[!] " [07:54] manual edit is no longer supoprted? [07:55] configuring logical volume manager only worked after choosing "erase entire pasrition" [07:55] under the "Guided partitioning" [07:57] Kamion: so the current amd64 daily should be a good one? [07:58] herbert's finished with the box for now, so I can do a reinstall [07:58] mdz: I would be very interested if anything breaks. :) [07:59] Mithrandir: you tested it already? [07:59] no, I'm at the student society getting dirty with PHP coding [07:59] I intend to test it when I get home, if it's not too late === mdz hoses off Mithrandir [08:00] it's never too late for more testing :-) [08:00] more that it's too late for me; I got home at ~04 last wednesday. [08:01] i'll test tomorrow's daily [08:08] gah, this is going to get confusing [08:08] there's now -ac branches of 2.6 [08:09] pitti: my card reader still comes up as root:disk 0660 [08:09] oh, I need 2.6.8 [08:09] fortunately, it's already in the archive ready to go :-) [08:10] sabdfl: i understand it's for first-time install.. if i had my speakers on at 70% i would have issues with the neighbours ;) [08:11] sabdfl: possibly.. they will start commenting on my work: "oh no.. he is installing ubuntu again" [08:11] fabbione: have you tested the 2.6.8.1. kernel? [08:11] mdz: yes, both from d-i and upgrading from kernel-image. on i386 is a GO for me [08:12] fabbione: from d-i? I didn't realize that was possible [08:12] oh, you did a netboot? [08:12] mdz: also.. i found a way to build the nvidia kernel without the running kernel. i might package all the stuff and test restricted tomorrow or so [08:12] mdz: yes. netboot [08:12] fabbione: oh, excellent. thanks for testing it [08:12] it's working great for me on all 3 architectures [08:12] mdz: no problem :-) [08:13] i only have i386 atm and it will be so for a while... [08:13] there are many people asking for powerpc to test [08:13] it should be built shortly [08:13] http://www.loria.fr/~thome/d600/radeonfb.patch.gz [08:13] amd64 should also, but there are very few amd64 sounders I tihnk [08:13] ^ oh, wow... I'm going to try that [08:14] Keybuk: fixing suspend with radeonfb? [08:14] mdz: allegedly so [08:16] pitti: still here? [08:16] mdz: I am. [08:16] mdz: sorry, xchat makes no noise on ppc [08:16] Keybuk: is that a kernel patch? [08:16] pitti: it's looking quite good with the current hal and udev! [08:16] mdz: and my workstation is broken [08:16] (and 2.6.8.1) [08:16] mdz: thanks! [08:16] pitti: I connect my card reader, and the devices are group plugdev [08:16] I insert a card, and nautilus opens [08:16] fabbione: yeah, adds proper reset/restart support to the radeonfb driver [08:16] it's just a bit strange, though [08:16] mdz: yes, 2.6.8 is a requirement for udev to detect removeable devices [08:16] mdz: what is strange? [08:16] this time two nautilus windows opened, one with the card, and one with / [08:17] pitti: the two windows [08:17] mdz: I also encountered this. Immediately after bootup, I see /, /home and the sda1 window [08:17] Keybuk: ok... [08:17] pitti: after closing the windows, unmountind, and reinserting, I get only one window [08:17] mdz: funnily enough it works correctly all subsequent times [08:17] (the correct one) [08:17] mdz: exactly. [08:17] how weird [08:17] mdz: I think this should be a gvm issue [08:17] pitti: yes, it sounds that way [08:18] I have also sometimes seen a window with / come up alone [08:18] I think when the mount of the real volume was failing [08:18] mdz: hmm. but the extra windows also appear when the mount was successful. [08:18] pitti: please file a bug about it, assigned to npmccallum and CCing me [08:18] mdz: I'll try to reproduce this exactly tomorrow [08:18] pitti: or if you have time to look into it yourself, that is of course fine too :-) [08:19] mdz: I'll do. Good to hear that your card reader works, though, so I can close that bug [08:19] nathaniel probably won't be here until tomorrow [08:19] mdz: I will do. My bug count came pretty close to zero today and I think it is not the best time to work on new projects [08:19] mdz: I will finish for today, I still need some workout after 11 hours of hacking... [08:20] mdz: if 2.6.8.1 will be the default kernel, everything should work fine [08:20] pitti: another thing that is strange [08:20] pitti: when I double-click an item which is on the card [08:20] it doesn't open [08:20] it works with files on the hard disk [08:20] mdz: hmm. That works for my USB stick. [08:21] pitti: vfat filesystem? [08:21] mdz: does it only happen with the "extra" volumes of your card reader? [08:21] mdz: yes [08:21] pitti: "extra"? [08:21] fabbione: maybe we can have a sound-levels test during installation? start low, and rise? [08:21] mdz: extra in the sense that they are hotpluggable in the hotpluggable device [08:21] for hoary, of course :-) [08:21] mdz: can you access the files manually on the shell? [08:22] mdz: if not, it might be a root mount point permission issue [08:22] mdz: what are the mount point permissions? [08:22] pitti: it works fine if I launch eog from the command line [08:22] so the permissions are OK [08:22] mdz: hmm. [08:22] (nautilus:5489): Eel-WARNING **: Error starting command ''/media/sdc1/Weston/AThenium-with-matt.jpg'': Failed to execute child process "/media/sdc1/Weston/AThenium-with-matt.jpg" (Permission denied) [08:22] that's what I see in ~/.xsession-errors [08:22] it is trying to exec the jpeg [08:22] mdz: Ah, I remember [08:23] seb128: any idea? [08:23] I cannot reproduce this, unfortunately, so I cannot debug it [08:23] sabdfl: if we switch to a GUI installer for hoary, i think we can easily integrate it. [08:23] Kamion: thanks for the d-i changes for the kernel [08:23] sabdfl: or otherwise doing it at a later stage.. but yes i think it's doable [08:25] sabdfl: you were seeing that problem with opening files on a removable device also, weren't you? [08:27] mdz: I'm just back from dinner [08:27] reading the log [08:27] mdz: where did you get that log message from? [08:27] mdz: I believe I can reproduce it [08:27] pitti: ~/.xsession-errors [08:28] seb128: I can summarize for you [08:28] mdz: can you open the files with right-click and choose an app manually? This works for me [08:28] yes please [08:28] seb128: I insert a card in my flash reader, and it is mounted by gvm with pmount on /media/sdc1 and nautilus opens. great! [08:28] seb128: however, when I double-click on a file, it doesn't open [08:28] mdz: I can reproduce it! [08:28] seb128: and an error is logged in ~/.xsession-errors [08:28] pitti: yes, that works [08:28] mdz: odd, this worked before [08:28] seb128: (nautilus:5489): Eel-WARNING **: Error starting command ''/media/sdc1/Weston/AThenium-with-matt.jpg'': Failed to execute child process "/media/sdc1/Weston/AThenium-with-matt.jpg" (Permission denied) [08:28] seb128: as if it is trying to exec() the file [08:29] weird [08:29] rather than launching eog or something [08:29] mdz: could be tomorrow's job for me [08:29] are you able to open other file (text, media file, ...) ? [08:29] I wonder if it might have something to do with the mount options [08:29] seb128: unless you want to do that yourself :-) [08:29] hmm [08:29] I think I know the problem [08:29] all of the files are executable [08:29] oh [08:29] so nautilus run them as first choice [08:29] when I try to open a text file, I get a dialog saying that it is an executable text file, asking if I want to display it or run it [08:29] apparently so [08:30] and doesn't fall back [08:30] right click on a file and look the list [08:30] mdz: but if you suppress the exec, you will not be able to enter directories any more [08:30] seb128: Open, Open with "The GIMP", Open with "Image Viewer", Open with Other Application [08:30] pitti: with vfat at least, you can set them separately [08:30] mdz: open with gimp/image viewer works ? [08:30] mdz: oh? I did not know [08:31] pitti: dmask and fmask, instead of umask [08:31] seb128: yes [08:31] it can access the file [08:31] ok, so that's the +x ... [08:31] the default is to run it [08:31] that is not entirely unreasonable, I suppose [08:31] mdz: hmm, this is getting tricky. We don't want to limit execution for other file systems, I suppose [08:31] but if it cannot run it and it has an application it can use, it seems like it should fall back [08:32] mdz: this sounds saner. [08:32] pitti: I think those options only apply to fat and possibly fat-like filesystems [08:32] mdz: there actually may be files on a vfat that the user wants to execute [08:32] but they might produce an error or warning on others [08:32] pitti: hmm, true [08:32] but it is very unlikely [08:32] mdz: If it works correctly for other file systems, I will implement that [08:33] mdz: let's regard it as a trojan horse protection feature :-) [08:33] mdz: can we put a file invocation into nautilus? [08:33] and only execute if it is actually an ELF or hashbang file? [08:35] pitti: what about binfmt_misc? :-) [08:35] I think attempting an exec is the only reliable way to test whether it will work [08:36] pitti: what about unmounting? [08:36] mdz: what do you mean? [08:36] pitti: how will these automatically mounted devices be unmounted? is that still an open question? [08:36] mdz: trying the exec is not the worst option, I think [08:36] seb128: nautilus --browser doesn't have a desktop icon anymore? [08:36] had an error with installing bsd-utils [08:36] mdz: you open the computer window and umount it by right-clicking on it and choose "unmount" [08:37] base install got hung by it [08:37] retrying from menu [08:37] "Couldn't download bsdutils" [08:37] again [08:37] doko: panel icon you mean ? [08:37] mdz: it would be nice if it were umountedd automatically when closing the window, but other processes might still access it [08:37] debianist: installing from CD? [08:37] debianist: if so, it's probably a bad burn [08:37] mdz: however, we could try and pop up a warning if unmounting did not work [08:38] mdz : is it supposed to download anything from the the internet? [08:38] debianist: no, not in a CD install [08:38] debianist: "download" there means "copy" [08:38] mdz : i see. well, it's a warty iso I downloaded today, and I'm faking a cd with the installer image under QEMU [08:39] seb128: if it's called so ... [08:39] debianist: hmm. you checked the md5sum of the CD? [08:39] mdz : darn. does that mean I need to redownload the whole installer image? [08:39] debianist: nah, you can use rsync [08:39] mdz : hmm, I checked the previous ones. so I passed this time :( [08:40] doko: we decided to only put a webbrowser/mailer/help here [08:42] seb128: so it's not possible to open the "file system browser" (the thing with the directory tree to the left) with the mouse/menus?with [08:43] just open a nautilus dir [08:43] and in the file menu [08:44] browse option [08:45] BTW that's just the default config, if you don't like it you can add a launcher for this, or change the default behaviour in nautilus' properties [08:45] mdz: removable devices had been bad for me, but it's looking much better [08:46] mdz : rsync would reconstruct the only missing / defective part of the image? [08:46] debianist: correct [08:51] mdz : cool [08:58] mdz : how would I instruct rsync to repair the image? [08:58] seb128: hmm, I can open it with "File/Open Browser", I cannot find anything in the preferences, and I have to know about the command to add it via "custom application launcher". why remove it? windows does have it, macosx has it. [08:58] mdz: hmm... yes, double-clicking a JPG in a /media/ popped-up nautilus window (USB pen storage) doesn't fire up any app [08:59] debianist: rsync rsync://ftp.no-name-yet.com/cdimage/.../warty-i386-1.iso [08:59] sabdfl: I think we nailed the issue (executable permissions) [08:59] doko: the command: "nautilus --browser" [08:59] seb128: what do you think is the best way to fix that issue? [09:00] seb128: yes, _I_ know now, but it non abvious. [09:00] s/it/it was/ [09:00] mdz: the +x file issue ? Not sure, would like to get nautilus' guys advice, but that's a bit late now [09:00] seb128: yes [09:00] I will check tomorrow with them [09:00] seb128: ok, should I file a bug as a reminder? [09:00] can't hurt, yes please [09:00] ok, will do, thanks [09:01] doko: preferences -> behaviour -> always open in browser windows [09:03] mdz: can you please cc me on the bug report? [09:03] pitti: already done [09:05] seb128: my mistake, logged out, and now I see the option. thanks! [09:05] np === kagou [~kagou@AMontpellier-251-1-39-242.w81-251.abo.wanadoo.fr] has joined #ubuntu [09:08] mdz, do i use pumount on my dongle usb or can i unplug it directly ? [09:12] mdz: yes, current amd64 daily should be ok ... [09:13] mdz: if you're uploading linux-source-2.6.8.1 again, I can hold the sounder release until we've got all the d-i ducks in a row for all architectures; shouldn't take too long to do the remaining two [09:13] (and quicker to do two at once) [09:14] kagou: please unmount it before by right-clicking on the device in the Computer window [09:15] Kamion: can we shoot at d-i ducks? [09:15] kagou: of course you can also pumount it by hand [09:15] thom: *bang* [09:15] ok pitti [09:25] I do so hate having to manually fix kernel drivers :-/ [09:25] hmm I see that most of the hebrew translation works are already present in ubuntu from d-i.. [09:27] hm, acx100 support [09:27] I wonder if it works on big-endian CPUs these days [09:27] it used to assume host and device endianness were the same [09:27] Kamion: doesn't work on amd64 at least :( [09:28] probably assumes stuff about sizeof(long) too [09:28] yup [09:29] I remember thinking it wasn't a very well-written driver, and I'm not even a kernel hacker :) [09:29] I should give the author an account on ravel and ask him to fix it. [09:29] never had time to fix its endianness problems though, I got an orinoco-based card instead [09:29] time > money [09:29] fabbione: would an earlier time work better for you tomorrow? [09:30] Kamion: I've uploaded it again [09:30] Kamion: it should build all of i386, amd64 and powerpc now [09:30] none of the binaries seem to be in the archive yet, though [09:30] mdz: will wait for builds, then do l-k-di-*, then build other stuff, then ... [09:30] I assume the build logs haven't arrived either since LaMont arranged for them to be mailed to you [09:31] he arranged for me to get copies of failures only === Kamion has a handy printed checklist next to him, snarfed from installer/doc/new-kernel-checklist.txt from d-i svn [09:31] Subject: Log for failed build of linux-source-2.6.8.1_2.6.8.1-2 (dist=warty) [09:32] /usr/share/kernel-package/rules:1298: *** Error. I do not know where the kernel image goes to [kimagedest undefined] The usual case for this is that I could not determine which arch or subarch tihs machine belongs to. Please specify a subarch, and try again.. Stop. [09:32] odd, apparently it built for him on my powerpc [09:33] Kamion: any clue? [09:34] nothing from amd64 yet [09:34] lamont: around? [09:34] sec [09:35] elmo, thom: could you /msg me /proc/cpuinfo from one of the buildds? [09:35] Kamion: which arch? [09:35] mdz: amd64's still building [09:36] elmo: that's promising [09:36] elmo, thom: er, sorry, powerpc [09:36] bingo, it's a kernel-package bug [09:37] does it not have 'machine'? [09:37] hm, I think arch=powerpc plus "Mac" in machine line of /proc/cpuinfo ought to indicate pmac pretty conclusively [09:37] I don't see why it cares what it happens to be running on [09:37] yes, it does, but the line is RackMac [09:37] I'll fix it [09:37] it's building all the subarches [09:37] I guess it's the make-kpkg clean === Kamion is not going to attempt to work that out tonight :) [09:38] sounds good to me [09:38] don't know why this didn't affect us before, I suspect differences in linux-source [09:39] ah, and it looks for Power and iMac [09:39] indeed [09:40] fix uploading now [09:40] I'll go file a Debian bug about that [09:41] odd, kernel-patch-powerpc doesn't seem to do it much differently [09:41] oh [09:41] export KPKG_SUBARCH=pmac [09:41] linux-source doesn't do that [09:41] that would do it [09:42] mdz : I started rsyncing my image some time ago, still it's working.. [09:42] mdz : I wonder how bad my image is damanged [09:43] I'll have herbert merge that in [09:43] debianist: are you sure that you're syncing the same image, and you pointed it at the correct file? [09:43] if they did not match, or the local file is missing, it would download the whole thing [09:43] mdz : I guess so. unless the "current" directory image has changed.. [09:44] mdz : what about -u update only mode? should I choose that? === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has left #ubuntu ["Client] === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu [09:44] debianist: 'current' could have changed, depending on when you downloaded it [09:45] Program received signal SIGSEGV, Segmentation fault. [09:45] [Switching to Thread 1084139584 (LWP 1670)] [09:45] 0x40bea748 in NSGetModule () [09:45] from /usr/lib/mozilla-firefox/components/libnecko.so [09:45] oh, rapture [09:45] mdz : downloaded it eariler today (about 4 hours ago) [09:45] segfaults everytime i open a new tab [09:46] thom: could be a lot worse; at least it's reproducible [09:46] :-) [09:46] heh [09:47] even without a .firefox or a .mozilla [09:47] anyone else reproduce? [09:48] mdz: would be better now to wait for a new kernel-package to build and just rebuild with that, I think [09:48] elmo: will you be able to kick off a rebuild? [09:48] thom: I browsed a bit, but didn't try opening a tab [09:48] I'll test it [09:49] kamion; sure [09:50] thom: opening new tabs works fine for me [09:53] hrm [09:54] I could've sworn I'd added amd64 support to base-installer; evidently not [09:54] oh, I did it in an uncommitted change upstream, yay me === kagou [~kagou@AMontpellier-251-1-24-230.w81-251.abo.wanadoo.fr] has joined #ubuntu === pitti [~pitti@195.227.105.180] has joined #ubuntu === seb128_ [~seb128@ANancy-111-1-15-76.w80-14.abo.wanadoo.fr] has joined #ubuntu [10:35] good night, boys! === seb128_ is now known as seb128 [10:41] elmo: new kernel-package seems to have built [10:44] seb128: i've sent jbailey some updates [10:44] ok [10:45] Kamion: what sort of base-installer support for amd64? [10:45] it seemed to work well enough for me [10:48] luck :-) [10:48] base-installer tries guesswork if you haven't included explicit selection of which kernel packages to install [10:48] Kamion: should I test the current amd64 daily, or wait for the next one? [10:48] it usually happens to get it right, but best not to release that way [10:48] current may be interesting since it uses warty's grub [10:48] I tested warty's current grub on its own; that works [10:48] I haven't tested grub-installer though [10:50] mdz: you know that annoying 'do you want to restart pcmcia on upgrade' dialogue? [10:50] mdz: i've been getting that on my ibook and test box, neither of which have pcmcia [10:50] jdub: yes, it happens anywhere that pcmcia-cs is installed [10:50] which is, well, everywhere [10:50] mmm [10:51] Kamion: was that an explicit change from boot-floppies? [10:51] is it fixable? [10:51] one of my sounders encountered a machine where pcmcia-cs hung the boot process (desktop system) [10:51] mdz: pcmcia-cs isn't particularly related to boot-floppies? [10:51] jdub: see #1002 [10:51] Kamion: base-config maybe? anyway, in woody, pcmcia-cs was removed on systems without pcmcia [10:52] mdz: oh, gotcha [10:52] hw-detect in Debian only installs pcmcia-cs if you have PCMCIA hardware, but you said very explicitly that you didn't want it to do anything like that :-P [10:53] so, it doesn't ... [10:53] heh [10:53] yeah, my only concern is this one machine that hangs [10:53] the new bottom bar in bugzilla looks like babyspew :) [10:53] I haven't been able to reproduce it, so I don't know if it can be fixed for warty [10:54] if you want to change that, move pcmcia-cs to Ship [10:54] I don't care about the question; as far as I'm concerned that shouldn't be asked on a desktop or a laptop [10:54] (by default) [10:54] gah, we ship shorewall? (eyeing RC bug) [10:56] jdub: how many more packages remain in #943? [10:57] uploading linux-kernel-di-amd64-2.6 ... [10:57] mdz: just checking gnome-cups-manager [10:57] Kamion: great, so the next daily should be 2.6.8.1 across the board? [10:57] mdz: i think, as i said, we should just change pcmcia-cs to 'always' and have done with it [10:58] that would let me close about 5 bugs [10:58] thom: I agree; please go ahead [10:58] still waiting for powerpc, but that's the general plan [10:58] thom: I think 'medium' is appropriate, as I commented in bugzilla [10:59] yep [10:59] i wish bugzilla had a "change owner to me and assign" button [11:00] it must be such a common operation [11:00] thom: OH GOD YES, I've wanted that for at least three years [11:00] everyone at Zeus was annoyed by that [11:02] it seems so weird that the mozilla guys have never wanted that enough [11:03] thom: what's different about your amd64 that breaks firefox tabs? [11:03] thom: do you have tab extensions installed or anything? [11:05] mdz: this is i386 [11:06] oh, never mind [11:06] hooray [11:06] mdz: brand new install this morning [11:06] from context, I thought you were talking about amd64 [11:07] i'm sure i canh make it break on amd64 too, but :-) [11:10] Kamion, thom: regarding bugzilla, either of you want to file a wishlist bug to ask justdave about it? [11:10] it might be easy [11:10] thom: please don't break it on amd64, since that would mean I would have to investigate. ;) [11:14] FUCKING TABS. [11:14] grrr [11:15] (yes, firef*&! just blew up again) [11:17] thom: does your X40 power off correctly? [11:18] ok, I'm off for a while, while all the kernel stuff builds [11:18] Kamion: do you have a new linux-kernel-di for powerpc ready for when the kernel itself is built? [11:18] hm, no, I'll do that now [11:19] wow, sven is a numpty [11:19] hm, or not just on this occasion [11:19] no need to stay around for it; presumably you'll have a chance to do it before the CD build [11:19] mdz: yes [11:19] thom: damn you [11:20] " I've just imported to GNOME CVS the code for the GNOME Pedia application, which is a GNOME frontend to the Wikipedia." - Rodrigo Moya [11:20] mdz: *g* [11:20] oh, sure, it's more or less the same as the others though [11:21] I'm not sure whether the current dep-wait magic is up to the challenge of it being uploaded before the kernel is built [11:21] it seems to get stuck sometimes [11:21] I wasn't planning to try [11:22] actually I need to wait to see what the linux-image packages look like on powerpc; there's one module I'm unsure about [11:22] mdz: your t42 doesn't? [11:22] thom: no, and neither does nathaniel's [11:22] and someone on the sounder list had the same problem [11:22] thom: Bug#1055 === justdave [~dave@24.247.63.44.gha.mi.chartermi.net] has joined #ubuntu [11:24] gah, #1050 needs to be fixed for the CD build [11:24] yep [11:24] I'll look at it [11:24] jdub: is there going to be a 2.[78] gdm? [11:24] none of my installs lately have got that far, due to other initscript breakage :P [11:25] mdz: no [11:25] mdz: not for gnome 2.8 anyway ;) [11:25] jdub: so gnome 2.8 will include gdm 2.6? [11:25] I guess the next one would be 2.9 then :) [11:25] yes [11:25] mdz: vicious had no time to hack on it for 2.8 [11:25] and it's already perfect anyway [11:26] apart from not having vnc transactional terminal support [11:26] dpkg-shlibdeps: failure: ldd on `debian/libgail-dbg/usr/lib/usr/lib/libgailutil.so.17.0.0' gave error exit status 1 [11:26] dh_shlibdeps: command returned error code 256 [11:26] make: *** [binary-makedeb-IMPL/libgail-dbg] Erreur 1 [11:26] WTF [11:26] Hmm, Yet ANother Electric storm, and I haven't bought the UPS. [11:27] Time to poweroff. [11:27] good night [11:27] 'night Oskuro [11:27] seb128: was the po file ok? [11:27] yes, I've included it [11:27] k [11:27] thanks === Oskuro waves. [11:28] #1916 in the other bugzilla [11:29] mdz/Kamion: ^ [11:30] ta [11:31] hrm, people really don't get that they can use dhcp to assign a static address, do they? [11:33] it's not a very common thing to do and the implementation of it in ISC dhcpd has.. issues in large, routed networks [11:34] seb128: current gtk seems to have an 'O' in the version number rather than a '0' [11:34] I don't know whether that has the potential to break anything or not [11:34] ISC dhcpd has issues, full stop [11:35] elmo: well, yes [11:35] mdz: *shrug* it works for the most part === debianist is booting under QEMU the fixed image [11:36] elmo: it doesn't strictly *suck*, and there certainly isn't a better alternative, but it has a lot of things which are awkword or borderline-broken [11:36] mdz: oups, yes. I don't think it's a problem === Gman [~Glynn@amfea-proxy-2.sun.com] has joined #ubuntu [11:37] it does seem that dhcp is something to get wrong, whatever the platfrom [11:37] loads of seb's packages did that, I thought it was by choice :) [11:37] it's probably the french keyboard, he can't tell [11:37] rohh [11:37] *argh* [11:37] mdz : is there a tool for our use to resize linux ext3 paritions? [11:37] dhcp triggers my highlight :P [11:38] HcE: howso? [11:40] debianist: partman can do that [11:40] hebrew text is is straightend to the left, that sould be fixed [11:40] select manual partitioning in d-i [11:40] fribidi support is in d-i upstream, but I suspect we're too close to our own release to attempt to sync it [11:40] Kamion : meaning we might include such an option for user for instance, wanting to resize their ntfs or fat32 paritions for ubuntu install? [11:40] it could perhaps be tried [11:40] debianist: not until parted supports those [11:41] that should really be sorted out in d-i upstream, not so much here [11:41] I was pretty sure parted could resize fat32 [11:41] just not ntfs yet [11:41] fat32, possibly; haven't tried [11:42] ok Kamion. so if I want to resize my already existing debian ext3 partition, that can be achived through ubuntu's partman? [11:42] yes [11:42] I want to make room for a real hardware install [11:42] select the partition, reduce size [11:42] back up your data first, though; we can't take responsibility for loss ... [11:42] Kamion: hmm, I didn't think udebs had init scripts [11:43] (re: #1052) [11:43] mdz: pcmcia is a special case [11:43] i know you wo't [11:43] Kamion: is it the only such? [11:43] Mithrandir: my name is HC, so I highlite hc :P [11:43] mdz: the only one I know of [11:43] HcE: bah, silly person. [11:43] oh, I've worked around that bug, I can downgrade it [11:43] if you come across any others, please file bugs straight away so that nathaniel can get them fixed up [11:44] this is a phenomenally poor day for a US holiday [11:44] d-i *should* tolerate them now [11:44] wouldn't like to swear to it, though [11:45] Kamion : what about my already installed grub on the system? would it be overwritten with ubuntu's version? [11:45] debianist: yes; the installer will pick up entries from the menu.lst in your already-installed grub, though [11:45] (that's a standard d-i feature) [11:47] boy that's nice [11:47] didn't realize it already does so much useful stuff :) [11:47] Kamion : are you implying that grub stores the meun.lst somewhere to boot block region? [11:48] Kamion : or how would d-i from ubuntu know to take the .lst file from my other installment? [11:48] mount other partition, read file [11:49] ohh that missing root filesystem is annoying [11:49] Kamion : it does so automagically ? ( mount other partition, read file) [11:49] yes [11:49] look at os-prober in the d-i repository [11:50] woowa, i had a feeling d-i is much much more than it might look :)