[12:05] es/debian-installer-utils>$ grep fset debconf-set-selections [12:05] db_fset "$var" seen true [12:08] Kamion: how big a job would it be to add an overall progress meter to d-i? [12:09] that'd rock [12:09] rawk [12:09] that too [12:09] ogra: see ubuntu-devel list :) [12:09] mdz: requires fixing the cdebconf progress protocol to allow multiple simultaneous progress meters [12:09] I've thought about it in the past; I think the advantage will be much greater with the gtk frontend than with newt though [12:10] jdub: yay, thanks :-D [12:10] thom: CPU or I/O? [12:10] ogra: but now i'm going to file bugs [12:10] jdub: go on :) [12:10] also requires working out where the hell you actually are - before you've retrieved udebs, you don't actually know [12:10] Kamion: it would do very good things for the user experience, even in newt I think [12:10] good point [12:10] ogra: kidding ;) [12:10] mdz: cpu [12:10] that's a relatively short way into the process, at least with cdrom [12:11] I suspect some kind of waypoint system like in base-installer would be useful [12:11] and we could fake it [12:11] mdz: would like not to be completely wrong with monolithic though :-) [12:11] jdub: :) [12:11] I'll have a look if you think it's really nice-to-have, can't bump away feature goals at the moment though [12:11] mdz: (hey! could abuse menu-item numbers ...) [12:11] Kamion: the path to the cloop image is now a debconf variable [12:12] Kamion: do you think we should set it in debian-cd land rather than using the hardcoded default? [12:12] I was just about to ask that [12:12] since debian-cd decides where to put it [12:12] yes, I think that would be a good idea; file a bug on debian-cd please? [12:12] I'll do that tomorrow [12:12] will do [12:14] mdz: how do you normally pass a password to mysqldump? [12:14] Kamion: hmm, it has /cdrom in it [12:14] elmo: interactively, I use -p (no arg) [12:14] non-interactively, I use this awful shell/perl construct [12:14] echo password | mysqldump -p [12:15] sladen: doesn't work [12:15] or mysqldump -p=password if you don't mind it being in the shell history [12:15] sladen: and more to the point, in ps output while it's running [12:15] which for non-trivial databases is a reasonable amount of time [12:15] mdz: wah [12:15] I can feel a case of expect coming on [12:16] elmo: you want my monstrosity? [12:16] bbl [12:16] night [12:16] I haven't needed it before [12:16] jdub: ping [12:16] elmo: oh, wait === Nafallo [~nafallo@h91n7c1o1027.bredband.skanova.com] has left #ubuntu-devel [] [12:16] elmo: you said mysqldump [12:16] elmo: this is for unattended backups, and not for a maintainer script [12:16] ? [12:16] yes [12:16] elmo: you can stash the password in a file [12:16] I tried that? [12:17] ~/.my.cnf by default [12:17] section [mysqldump] [12:17] oh, MAH [12:17] defeated by cut'n'waste [12:17] hmm, or not === kent [~kent@83.249.61.131] has joined #ubuntu-devel [12:17] mdz: mmk [12:18] what's the least horrific shell idiom for "iterate over the files matching this glob, but if the glob doesn't match anything, do nothing"? [12:18] jdub: just saw the thread from shlomil regarding the language pack dependecies, I already talked about this with pitti and you in MAtaro, what needs to be done for the culmus pakcage to get into main so pitti could depend on it in the language pack package? [12:18] approval from mdz/myself [12:18] i've already given mine [12:19] mdz : are you ok with jdub approving or would you like a TB meeting for this? [12:19] jdub: coool [12:19] oof, this stuff doesn't need TB meeting bureaucracy [12:19] sivang: it needs to meet the criteria in the wiki [12:20] TB/CC are meant to be last resorts [12:20] http://www.ubuntulinux.org/wiki/SeedManagement [12:20] under Supported [12:20] main/supported right? [12:20] "Supported" is the section heading [12:21] mdz: hmm, sorry how does the mysqldump thing in my.cnf help? and is there a man page for that file? [12:21] elmo: each tool looks up its config in a section named after it [12:21] so you can specify different parameters for different programs in the one file [12:22] /etc/mysql/my.cnf is an illustrative example [12:22] it's documented in the mysql manual, iirc [12:22] [mysqldump] [12:22] password = XXXXX [12:22] doesn't work? [12:22] (it did for me under woody, where I did this in production) [12:23] note that you don't specify the -p option if you do that [12:23] uh, where XXX is the password? [12:23] yes [12:23] it's a 644 file dude [12:23] ?? [12:23] then make it 600 [12:23] it's 644 by default [12:23] should we fix that? [12:24] oh, you're talking about /etc/mysql/my.cnf? don't put it there [12:24] oh GOD [12:24] Default options are read from the following files in the given order: [12:24] /etc/mysql/my.cnf /var/lib/mysql/my.cnf ~/.my.cnf [12:24] sorry, I get it now. I neeed more tea [12:24] or use --defaults-extra-file= [12:26] mdz: it already meets (1) , (2) I can try and check , am willing to port it to arch if this is a must, and I think it would pass asecurity review because basically it copies files from the debian package to some place on the host system, what else? [12:26] mdz: merely a repository of fonts.. [12:26] Kamion: sshd won't let users login remotely with blank passwords by default, right? [12:26] Kamion: do you think it's necessary to set a password on the live CD? [12:26] mdz: need to usermod -U [12:26] mdz: score - working now, thanks. [12:27] sivang: if it's just data (no programs), then (2) should be easy [12:27] sivang: and fonts don't generally need porting, either [12:27] mdz: yes very much, the rules files is even pretty standard and doesn't do nothing but a couple of dh_install stuff [12:27] mdz: true, it has one target for all archs. [12:27] I just installed gcursor mentioned here before. Why have the theme-manager not support for changing cursor-theme? [12:28] kent: because no one's written it yet [12:29] mjg59: here? [12:29] mdz: only one issue, is if you want the files to reside in some other place then the current, oh and it does install to some of the X11 fonts dirs. [12:30] daniels: alive? [12:30] sivang: fine with me, then [12:30] has anybody else weird font probs in evo (looks like gtkhtmls font rendering cant decide on the right fontsize) [12:30] mdz: ok, coool. btw, does it need a maintainer to go with? :) although I'm pretty sure it could go by itself, and if there's a problem with it I don't mind to be the address for that. === T-Bone calls it a nite. bye all [12:34] sivang: is this a package you made, or did someone else make it? === jaco [~jaco@host42-20.pool80117.interbusiness.it] has joined #ubuntu-devel === JanC [~JanC@dD5770426.access.telenet.be] has joined #ubuntu-devel [12:51] mdz: don't see a problem with /cdrom, enough of d-i assumes it that I think debian-cd can get away with doing so too [12:52] Kamion: should I prepend /cdrom in casper, and expect a relative path in debconf? [12:52] or is it ok for the full path to be in debian-cd? [12:53] mdz: PermitEmptyPasswords is off by default, yes, although I'd test carefully if I were you [12:53] mdz: no, I think it's better for the full path to be in debian-cd in order that you can change it to be off an NFS mount or whatever [12:54] Kamion: good call [01:00] mdz: apt 0.5.28.1 is supposed to be OK for sarge? [01:02] seems ok === Kamion contemplates switching the weekly DVD to be install+live [01:11] ... maybe not today [01:17] Kamion: yes [01:18] Kamion: did you look into that issue with the ipw2200 firmware? [01:19] not yet [01:19] mdz: I did mention that the firmware is there but with a different name than it used to have, didn't I? [01:19] my money's on something looking for the old name [01:19] mdz: it made by baruch even, a DD [01:19] Kamion: oh, so you think it's a kernel problem? [01:19] well it's certainly in the udeb ... [01:20] looking at it now [01:20] unless that's not being installed for some reason [01:20] it's trying to loadipw-2.2-boot.fw [01:20] /lib/hotplug/firmware/ipw-2.2-boot.fw- exists [01:20] -rw-r--r-- root/root 6472 2005-01-17 18:04:55 ./lib/hotplug/firmware/ipw-2.2-boot.fw-2.6.10-2-386 [01:21] maybe this is the timeout problem keybuk talked about [01:21] what problem was that? [01:21] he said that the driver didn't wait long enough for the firmware to be loaded by hotplug [01:21] and that this would become a problem when we switched to udevsend [01:23] odd [01:23] I just added some logging to firmware.agent and it doesn't show up [01:24] it clearly works in the installed system (including the live CD), though [01:25] any reason why the firmware agent might not be called? === Kamion hopes nothing's hardcoding /sbin/hotplug or something [01:26] I don't see how it could [01:26] the kernel invokes the hotplug helper [01:26] and it's set to /sbin/udevsend [01:26] I meant in the kernel :) [01:26] ah [01:26] any easy way to get udevsend to log everything it gets? [01:28] Kamion: it's hard-coded in the kernel but you can set it by echo/cat'ing /proc/sys/kernel/hotplug [01:28] + rc = request_firmware(fw, name, &priv->pci_dev->dev); [01:28] the driver doesn't seem to do anything weird [01:28] most likely all firmware loading is busted [01:29] sladen: Uh. PMInitialize? [01:29] mdz: Hi [01:29] mjg59: I wanted to know how to determine whether a swap device was holding swsusp data [01:29] mjg59: so that the live CD would not touch it [01:30] mdz: Ah, right. There's a header right at the start - the struct is defined in kernel/power/swsusp.c (IIRC) [01:30] swapon -a will refuse to mount it [01:30] mjg59: Protected-Mode Initialize ; you're supposed to call that to initial the Video BIOS before you make any 32-bit VBE calls. Is there a chance that when coming out of S3, or on bootup (that hang issue you had), that isn't happening? [01:31] oh [01:31] yeah [01:31] it seems to overwrite the swap space signature [01:31] so that's perfect [01:31] and the bug won't exist in the first place === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has joined #ubuntu-devel [01:31] sladen: We don't make any 32 bit VBE calls [01:31] mjg59: how the nack are you doing vbesave ? [01:31] 32 bit entry to VBE is optional - AFAIK, lots of cards don't do it [01:31] Real mode [01:32] lrmi does vm86 real-mode emulation [01:32] mdz: hrm, when's the next livecd build? lots of nice uploads i see ;) [01:33] tseng: ping [01:35] sladen: hotplug> yeah, that's what I understood certainly and d-i sets that sysctl, was just wondering if there could be a bug somewhere [01:35] jdub: 1 8 * * * /srv/cdimage.no-name-yet.com/bin/cron.daily-live [01:36] sladen: To the best of my knowledge, neiter vbetool nor the kernel ever makes any 32-bit VBE calls [01:36] although many of today's uploads require the debian-installer I uploaded tonight to become active, which requires elmo's byhand [01:37] ... which he's already done. rocks. [01:37] rock. [01:40] jdub: it's mostly under-the-hood stuff, though [01:40] jdub: much cleaner, but if I've done it right you won't notice any difference [01:41] Kamion: I haven't been able to test the passwd stuff properly, since it needs a new cloop image, but I think it should work [01:41] it creates the user OK, but the group and sudo stuff is missing, since that isn't in the old version [01:42] I need to find a clever way to unmount the root filesystem and eject the CD [01:42] the udeb retrieval stage should be shorter now, jdub was asking about that recently I think [01:43] s/asking/bitching/ :-P [01:43] jdub: *hug* [01:43] mdz++ [01:43] mdz: doesn't [01:43] eject work? [01:43] lamont: I bet it would [01:43] lamont: but it's hard to get to after unmounting root [01:43] works on the warty livecd, albeit fatally, and only as root... [01:43] the closest I can think of so far is to create a min-chroot and pivot into it [01:44] or modify init; amu pointed out that knoppix takes that approach [01:44] haha, WinXP fails to deal with my amd64's SATA drive [01:44] probably the easiest way would be to simply _not_ unmount the d-i initrd [01:44] and then pivot back into it [01:44] but it's huge and occupies a lot of memory just for that [01:44] ew! [01:44] remount root as readonly, then eject [01:45] I don't think that remounting r/o will release the device [01:45] nor should it? [01:45] is there an eject --i-really-mean-it [01:45] eject hasn't been seen to care if the device is in use or not, at least not by me.... === lamont tests [01:46] lamont: eject tries to unmount if it's mounted [01:46] otherwise it will fail [01:46] mdz: didn't mean to be bitching [01:46] jdub: just teasing [01:46] yeah - shows busy [01:46] bummer [01:46] jdub: your live CD feedback is R0X0R [01:46] so how many udebs were optimised out (roughly)? [01:47] were we able to skip much of the kernel module ones, if the kernel is in the main image? [01:47] we didn't try that [01:47] it would require some significant changes [01:48] there's no way to get to the image until casper is unpacked [01:48] jdub: deliberately didn't attempt to skip the kernel modules [01:48] and casper is unpacked at the same time as everything else [01:48] took out the ones that are there to be menu items, so base-installer, archive-copier, prebaseconfig, {grub,lilo,yaboot,...}-installer, all of partman [01:48] thom: you stole my stopwatch, you bastard [01:48] ahr [01:49] thom: you'd better bring it to sydney [01:49] maybe a dozen or so at a rough guess, although I didn't count [01:49] Kamion: elite [01:49] tomorrow's live CD improvements are all behind the scenes ;-) [01:49] if it works, we win [01:49] jdub: I tried skipping all standard udebs including kernel modules; didn't work, network configuration failed because it didn't have nic-* [01:50] Kamion: why is it that all of /etc/hotplug is world-writable under d-i? [01:50] Kamion: yah [01:50] does udpkg do that deliberately or something? [01:51] er [01:51] mdz: yah [01:51] mdz: uh, shouldn't [01:51] daniels: I think I was looking for an informed opinion on sivang's X fonts package [01:51] Kamion: is it just me? [01:52] Kamion: oh, that's on the initrd [01:52] maybe genext2fs is stupid about permissions? [01:52] /bin/busybox is 777 too [01:52] yes, apparently so [01:52] it's like that in warty too [01:52] (just happened to be booting a rescue CD^W^Wwarty amd64 CD) [01:53] mdz: oh === usual [~colin@alb-69-202-36-213.nycap.rr.com] has joined #ubuntu-devel [01:53] mdz: server-side fonts are pretty hard to get wrong [01:53] fabbione: There's a possibly important fix in the current ACPI patch - I'll get more details to you tomorrow [01:53] daniels: CRACK? [01:54] mjg59: uploaded === colin_ [~colin@alb-69-202-36-213.nycap.rr.com] has joined #ubuntu-devel [01:56] crap, it'd probably help if I actually filter @ubuntu.com somewhere other than spam wouldn't it [01:56] daniels: I would offer to fellate you, but I feel you might turn me down [01:57] jdub: hiya [01:58] tseng: dude, three people have already asked me about tomboy 0.3, and it was only released today ;) [01:58] tseng: you got yourself a popular package ;) [01:58] hah :P [01:58] tseng: perhaps it could be your first universe upload? :) [01:58] Kamion: not even /etc/hotplug.d/default/default.hotplug is called when the firmware is requested [01:58] mdz: that's a bit depressingly bad [01:59] Haha [01:59] jdub: sure, i can check out the update here in a few, having some dinner [01:59] I've shocked daniel into silence [01:59] tseng: rawk :_) [01:59] tseng: and mono is fixed too! [01:59] hell yes it is [01:59] fabbione: We need ACPI CA 20050114 [02:00] mdz: nothing seems to be obviously missing from hotplug-udeb [02:00] my money remains on the kernel [02:01] ubuntu.com now replaces canonical.com for employee distro people? [02:01] yes [02:03] oh [02:03] request_firmware is returning -ENOENT [02:05] sivang: partly to try to help erase the distinction between Canonical and non-Canonical distro hackers [02:05] mjg59: it's entirely possible [02:05] the configured hotplug handler is getting called [02:05] with a firmware add event [02:05] and reasonable arguments === mdz yearns for strace [02:06] do you have libc6-udeb installed yet? [02:06] I'm at netcfg [02:06] if so, udpkg -i /cdrom/pool/main/s/strace/strace_*.deb [02:06] is libc6-udeb something that gets installed automagically? [02:06] might need to udpkg -i /cdrom/pool/main/g/glibc/libc6-udeb_*.udeb too [02:06] mdz: yes [02:06] Kamion: that's what I though, ok. [02:07] strace isn't on the live CD, but I could copy a deb I suppose [02:07] it's depended upon by various bits that get pulled in by anna [02:07] USB storage? [02:07] yay [02:07] (I have wired network as well) [02:08] mdz: wired network is the best :) [02:10] woo, more video cards :) [02:10] GAH [02:10] busybox tar doesn't support creating tar files??/ [02:11] nope [02:11] Kamion: found it [02:11] /etc/hotplug.d/default/default.hotplug is #!/bin/bash [02:11] %^@#$%@#$ [02:12] haha [02:12] silly hotplug people [02:14] if [ -t ] ; then [02:14] wonder what that's meant to mean [02:14] probably tests if stdout is a terminal [02:15] Kamion: changing it to #!/bin/sh worked perfectly [02:15] a quick read didn't reveal any obvious bashisms [02:16] mdz: -t is only documented to take a file descriptor; none of SUSv3, bash(1), test(1) document what it might do when given no arguments [02:16] it comes as #!/bin/bash from upstream [02:16] mdz: yeah, I checked too, there are a few bits of test -a/-o junk in hotplug.functions but that's fine in busybox sh anyway [02:17] Kamion: if you're satisfied, I plan to patch it immediately [02:17] satisfied [02:17] mdz: I don't understand this mail about kdelibs - are you saying it's depending on something that should be pulled into main? [02:18] 'cos anastacia disagrees [02:19] in fact kdelibs isn't even in main, I'm entirely confused [02:19] elmo: I sent you mail about kdelibs? [02:20] . [ 24: Matt Zimmerman ] Re: oo.o2 FTBFS [02:20] blegh [02:20] elmo: oh, I was probably thinking oo.o2 was going into main [02:20] which it probably ought [02:21] for HOARY? [02:21] that's the plan [02:21] mdz: did you approve the fonts package then? (I think I saw you say you're reviewing it) === elmo runs away screaming === sivang notes it has some rocking hebrew support. woudl be cool to see it in hoary. nearly zero effort hebrew ena/loc distro :) [02:21] elmo: the old fglrx-driver diverts away libGL.so.1, but I'm moving that diversion into xorg-driver-fglrx. moving the diversion removal as an unconditional in postinst doesn't help, because fglrx-driver is unpacked but not configured before xorg-driver-fglrx, despite the conflcits. is there a pre-depends, only for conflicts? [02:21] elmo: it doesn't replace v1, so we can take our pick when we're ready [02:22] elmo: i.e. have xorg-driver-fglrx declare that it can't unpack while fglrx-driver << 8.8.25-0ubuntu2 is configured [02:23] daniels: that's "Conflicts" [02:23] only unversioned I think [02:23] mdz: can't [02:24] if you conflict with something then you can't exist on the system at all at the same time as it [02:24] hm? a versioned conflict would force the new fglrx-driver to be unpacked before xorg-driver-fglrx [02:24] mdz: i conflict fglrx-driver << 8.8.25-0ubuntu2 already, but it only gets *unpacked*, not configured [02:24] mdz: yeah ... [02:24] daniels: you should be removing the diversion in preinst [02:24] mdz: and that will dtrt? [02:25] daniels: preinst is called before unpacking [02:25] so it would go: fglrx new-preinst, fglrx unpack, xorg unpack [02:26] mdz: mmm, but if you're upgrading, wouldn't that arse around with diversions? [02:26] daniels: isn't that the point? [02:26] mdz: fglrx new-preinst removes diversion, fglrx new-unpack removes libGL.so.1, so your xlibmesa-gl copy of libGL.so.1 is gone ... [02:26] the diversion is gone in the new fglrx [02:26] so on upgrade it should be removed [02:26] correct [02:26] given fglrx-driver no longer ships libGL.so.1 [02:26] I don't see the upgrading/arse problem [02:27] the new acpi-support disables suspend-to-ram by default [02:27] that seems slightly silly [02:27] ok, I'll try to explain what I fear may happen [02:27] original: old fglrx-driver unpacked; scenario: dist-upgrade [02:27] first, new fglrx-driver preinst runs, since the old one had no pre* scripts, right [02:28] and removes the libGL diversion, so /usr/X11R6/libGL.so.1 is now xlibmesa-gl's [02:28] then new fglrx-driver gets unpacked, and as it's no longer providing libGL.so.1, libGL.so.1 gets removed [02:28] so xlibmesa-gl's copy is gone until you unpack that again [02:28] that doesn't happen [02:29] dpkg knows which package the file belongs to [02:30] it won't remove xlibmesa-gl's file just because it's not in fglrx-driver [02:30] ok [02:30] so no adverse affects from removing the diversion in preinst? [02:30] bob2: We'll probably switch to enabled by default, but it's known broken on some percentage of hardware [02:31] that is the correct way to handle this situation [02:32] if I understand correctly [02:32] mdz: if it's not, I'll beat Keybuk up *shrug* [02:32] neither xorg-driver-fglrx nor fglrx-driver contains the file, right? [02:32] they just both divert it [02:32] or do they divert it and also contain the file in the .deb? [02:33] if they contain it in the .deb, you need to be doing the diversion stuff in preinst anyway [02:34] mjg59: yeah, it's just that it was enabled until I upgraded to acpi-support 0.12 [02:34] so I was kinda confused when it stopped working ;) [02:35] mdz: would it be a bad thing if I added hppa to linux-meta? [02:36] lamont: bad? no. silly? maybe [02:36] as long as it doesn't break it on real architectures... [02:37] it's just the meta packages [02:37] mdz: divert and also contain [02:37] shouldn't break anything [02:38] (hppa has 783 of 822 packages in main built) [02:38] lamont: wooo [02:38] jdub: have a package here for latest and greatest tomboy [02:39] i hopefully fixed alot of retardation i caused on my first attempt at packaging it [02:39] that'll slow down the heat death of the universe :) [02:40] tseng: yay! [02:40] jdub: given the amount of heat an hppa puts out? :) === sivang LOLs [02:41] mdz: my system is now very, very, very confused [02:41] Kamion: cheques and balances ;) [02:43] mdz: Removing `diversion of /usr/X11R6/lib/libGL.so.1.2 to /usr/share/fglrx/diversions/libGL.so.1.2 by fglrx-driver' [02:43] hppa's aren't that bad [02:43] dpkg-divert: rename involves overwriting `/usr/X11R6/lib/libGL.so.1.2' with [02:43] different file `/usr/share/fglrx/diversions/libGL.so.1.2', not allowed [02:43] jdub: ia64 is much hotter than hppa [02:43] wow-wow-wacka-wacka [02:45] daniels: that looks like "you told me to undivert the file, but there's a file in the way [02:46] daniels: that's on dpkg-divert --remove? [02:46] I suppose that's a reasonable complaint [02:46] last hppa I was anywhere near sounded like nothing else than an aeroplane taking off when booting [02:46] mine is hot and noisy [02:46] and not hugely useful [02:46] and HPUX is horrific [02:47] sounds like a typical hppa machine [02:47] mdz: that's on --remove, yes [02:47] jdub: the one I care about is my firewall [02:47] mdz: that's fglrx-driver trying to remove its own diversion in the new preinst [02:47] daniels: while its file is still in the way [02:47] b180's aren't very noisy. The rack-mount machines are kinda bad though [02:48] mdz: any other suggestions? [02:48] daniels: rm -f ? :-P [02:48] hrm.. do we care that the version of linux-meta won't match the api version of the kernel any more if I do this upload? [02:48] daniels: if this is all for the sake of renaming the package, there will be trouble [02:48] mdz: not just renaming, splitting [02:49] mdz: xfree86 and xorg [02:49] hasn't before, so it shouldn't matter [02:49] mdz: (but it was a really stupid name anyway) === lamont uploads, goes to fire training [02:51] daniels: can't you get rid of fglrx-driver entirely? [02:51] linux-restricted-modules-2.6.9 produces a load of uninstallable binaries according to britney; should we just kill it? [02:51] daniels: and just conflict with it (all versions)? [02:51] Kamion: uninstallable how? [02:51] mdz: mmm, probably [02:51] Kamion: yes [02:51] Kamion: and the corresponding linux-source too [02:51] the corresponding linux-source is already gone [02:51] mdz: will apt dtrt and remove fglrx-driver if xorg-driver-fglrx/xfree86-driver-fglrx crp it? [02:51] which probably explains why that l-r-m no worky === daniels reads u-d and stares at sladen. [02:54] sladen: ... so much pain [02:54] daniels: you just need to conflict [02:54] daniels: though rp is good documentation [02:54] mdz: 'kay, thanks [02:54] elmo: could you kill linux-restricted-modules-2.6.9 and everything in it from main, please? [02:55] germinate doesn't seem to want it there any more [02:56] lamont: any idea why i386 hasn't tried to build efibootmgr? === lamont looks [02:56] wanna-build -bi386/build-db -dhoary --info efibootmgr [02:56] efibootmgr: not registered [02:57] PaSA [02:57] odd, Debian has it for i386+ia64 [02:57] efibootmgr: ia64 # ia64 specific [02:57] elilo: ia64 # ia64 specific boot-loader [02:57] so why do we have elilo/i386? === lamont shrugs [02:57] either we should have both elilo/i386 and efibootmgr/i386, or neither [02:58] having the first but not the second is broken. :) [02:58] does elio have any arch: all components? [02:58] no [02:58] it does list i386 in Architecture: [02:58] elilo: i386 ia64 # ia64 specific boot-loader [02:59] as does efibootmgr, mind [02:59] in any case, fixing hoary is an elmo edit [02:59] fixing debian I could do, if you tell me what you want it to say... [02:59] but I'm supposed to be at fire training in < 1 minute, must really run [03:00] lamont: bdale added i386 to the list of supported architectures in elilo 3.4-4; I assume he knew what he was doing, so I figure i386 ought to be added to both elilo and efibootmgr === `anthony [~anthony@220-253-16-58.VIC.netspace.net.au] has joined #ubuntu-devel [03:00] and efibootmgr 0.4.2-2 [03:00] there's an i386 floppy disk with EFI on it that you can download and play with [03:01] jdub: ping [03:01] sladen: if you're feeling nostalgic about MS-DOS? [03:02] :) [03:02] any chance we'll see beagle in universe/multiverse? [03:02] someone asked that about an hour ago [03:02] lol [03:02] is it beagle day? [03:02] sorry [03:03] mdz: beagle love day :) [03:03] it's mentioned on slashdot [03:03] eruin: probably not until it stops deping on cvs packages [03:03] + planetgnome [03:03] ah [03:03] plus they have a pretty, pretty screenshot [03:03] which always helps grab attention :P === kent [~kent@c83-249-61-131.bredband.comhem.se] has joined #ubuntu-devel [03:08] is the bootsplash still planned for hoary === sladen whistles and looks at ceiling [03:10] lol [03:11] anyone here in the know about gnomevfs packages? [03:11] robertj: it's unlikely that the 'bootsplash' package with ever make it into Ubuntu. It's too evil. [03:11] err not bootsplash [03:12] usersplash === Nafallo [~nafallo@h91n7c1o1027.bredband.skanova.com] has joined #ubuntu-devel === tritium [~tritium@sjcc176x41.sjccnet.com] has joined #ubuntu-devel [03:16] robertj: hopefully [03:16] are there experimental packages being built somewhere for hoary usersplash? [03:16] I know it was supposed to come in along super phat laptop support [03:17] hi again. where can I read about a MOTU's dutys? :-) [03:17] usersplash sure sounds promising. :) [03:18] or just plain sexy [03:18] robertj: there should be a separate usplash respositary for people to test in the next week or two [03:19] robertj: with the various packages to enable it [03:19] that will be good [03:19] I noticed the other day I closed my laptop lid and opened it and it was very boring [03:19] even XP acts very interesting when I do tha [03:19] sladen, where will the usplash repo be announced? [03:19] so all in all a good sign [03:20] robertj: what sort of interesting thing? [03:20] tritium: on the mailing lits [03:20] sladen: well it locks up or sits there with no mouse input [03:20] but with hoary it just opens === Nafallo starts to wonder who he should ping ;-) * [03:21] robertj: is that just the lid turning the screen off, or is it actually going to S3? [03:21] I think it's actually sleeping [03:21] robertj: mjg59 is the person to thank for that [03:21] i need to probe around the logs [03:21] strange things happen when I suspend and the close the lid [03:22] tritium: try updating, there was a kernel rev last night and some more stuff went in which may fix the Interrupt routeing and non-waking [03:22] sladen, okay, will do. Thanks. [03:23] robertj: back to the question. What kind of ''interesting thing'' does XP do---just that it close and when re-opened it still works? [03:24] not it stops working in interesting ways === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has joined #ubuntu-devel [03:27] wifi stops working or pancs [03:27] sometimes no input is handled ,etc [03:28] how can I tell for sure if it's properly asleep [03:28] I close the lid, open it and get a black terminal for a second before the screensaver is activated [03:29] its not quite like that, its more like it stored the state from X before the lid close [03:29] and then updates to the current, which has a locked screen [03:29] youll notice the first image you see is way behind if something on the screen is updating, like your irc client [03:30] dunno, it happens pretty quick, maybe it aint [03:30] what should I be looking for [03:30] you could sleep 10 and echo something i guess [03:30] close the lid, wait 10, open [03:30] if it really sleeps you wont be able to ping it [03:30] the first thing you see should have no echo yet [03:31] doh [03:31] it's pinging === wasabi [~wasabi@c-24-1-67-127.client.comcast.net] has joined #ubuntu-devel === wasabi [~wasabi@c-24-1-67-127.client.comcast.net] has joined #ubuntu-devel [03:34] I can ditch this grub splash image I made once usplash is available: http://mip-lab4.ecn.purdue.edu/~rimbert/ubuntu-splash.xpm [03:35] i'm still feeling so-so about the default colorscheme [03:36] it has grown on me. I used to be a big fan of debblue [03:37] ThinGeramik with Crux & Gartoon are great [03:37] clean but fun [03:38] I think ThinGeramik is hands down the best widget set though [03:47] I like the ClearLooks / Industrial / Human combination [03:47] http://appelsinjuice.org/Skjermdump.jpg [03:48] robertj: is this is XP, or Warty, or Hoary? [03:50] What's that icon that looks like 3 shirts stacked on each other, slightly offset? [03:51] oh, it's the theme manager thing [03:51] thats the icon for theme-manager [03:51] oh, okay [03:51] tritium: usplash is more of an 'engine'. It still needs something to display and if you're got some funky ideas for you'd like to see or some out of this world-ideas that nobody has seen before, I'd love to hear them! [03:52] sladen, Okay, thanks for asking. If I have ideas, I'll let you know. [03:53] eruin, that gdesklet-thing to the right, what is it called? [03:53] tritium: that image might useful in the case of a 16-colour framebuffer [03:54] sladen, Okay, I'll definitely share it. I was planning on doing so anyway. I still need to set up my wiki page, etc. [03:54] kent: CornerRhythmlet-bottomright.display [03:54] it's in media/control if you install the gdesklets-data package [03:55] sladen, how should I get involved? [03:56] I'm definitely no artist, by the way. I'm not that creative. [04:02] whoooa === bronson [~bronson@node-40240852.sjc.onnet.us.uu.net] has joined #ubuntu-devel === tritium [~tritium@sjcc176x41.sjccnet.com] has left #ubuntu-devel ["Leaving"] === zenrox [~zenrox@wbar7.sea1-4-10-181-189.sea1.dsl-verizon.net] has joined #ubuntu-devel === erich [~erich@erich.developer.debian] has joined #ubuntu-devel === zenrox [~zenrox@wbar7.sea1-4-10-181-189.sea1.dsl-verizon.net] has joined #ubuntu-devel [04:48] tseng: pong [04:49] jdub: oh.. hi [04:49] jdub: the tomboy package i noticed is fairly botched, not sure how to best fix it [04:50] jdub: look at diff.gz, there is a bunch of stuff that doesnt seem to belong [04:51] sounds like you've built the package after building it before [04:51] and it didn't properly clean up after itself the first time [04:52] quite possibly, but its in the source package already. [04:54] yeah, so [04:54] untar tomboy [04:54] copy over your debian directory [04:54] then do the source-only build [04:54] yay [04:54] then do fakeroot debian/rules binary [04:55] to make sure the binary is ok [04:55] that won't futz with your source packages though === dasenjo [~dasenjo@200.21.83.173] has joined #ubuntu-devel === dasenjo [~dasenjo@200.21.83.173] has left #ubuntu-devel [] === alerios [~alerios@201.245.164.174] has joined #ubuntu-devel === alerios [~alerios@201.245.164.174] has left #ubuntu-devel ["Abandonando"] === jbailey [~jbailey@CPE000ded9d787c-CM014260028338.cpe.net.cable.rogers.com] has joined #ubuntu-devel === jdub_ [~jdub@home.waugh.id.au] has joined #ubuntu-devel === `anthony [~anthony@ekorp-203-63-137-225.eoff.ekorp.com] has joined #ubuntu-devel === wasabi [~wasabi@c-24-1-67-127.client.comcast.net] has joined #ubuntu-devel [05:42] mdz: was intending to do a sweep of artwork pages tonight, just watching your changes [05:42] mdz: if you can hook up cliff and i soon, that would be very helpful [05:51] jdub: would appreciate any clarifications that you can make on the pages [05:51] jdub: I don't know jack about themes [05:51] jdub: I sent cliff email tonight; going to set up a time to meet and talk [05:51] ok [05:51] jdub: will probably be in the evening local time [05:51] any times you're not available this week? [05:53] not really === wasabi [~wasabi@c-24-1-67-127.client.comcast.net] has joined #ubuntu-devel === AndyFitz [~andy@220-245-97-227-qld-pppoe.tpgi.com.au] has joined #ubuntu-devel [06:35] morning guys [06:36] jdub: http://sparc.ubuntu.com/ :-) [06:36] we still cannot bootstrap from it [06:36] but almost [06:36] still missing a few NEW love from elmo [06:36] like the kernel ;) === doko [~doko___@dsl-082-082-069-195.arcor-ip.net] has joined #ubuntu-devel [06:49] anyone know whats happening to hoary repositories right now ? [06:50] AndyFitz: how do you mean? [06:51] fglrx has broken and now cant be fixed/removed etc. also the security repo doesnt authenticate all the time === sbrath [~usrshb@sbrath.user] has joined #ubuntu-devel [07:03] fabbione: good point - my linux-meta upload is gonna be stuck in NEW too. [07:05] fabbione: how soon do you plan to upload -10, btw? [07:06] lamont: later today or tomorrow morning [07:06] ok. [07:07] i got flooded by bugs during the night [07:07] playing with -9lamont1 here. :-) [07:07] cool :-) [07:07] does it boot? [07:08] dunno yet, trying to actually build the source package... just learned about 00list- :-( [07:08] argh [07:08] and 00list-.hppa :-) [07:08] yeah - saw that one === lamont wonders if he should beat up daniels about xresprobe [07:10] you probably need to add the stub for hppa [07:11] like we did for sparc and the others [07:14] fabbione: nah - it has horribly b0rked #ifdef cruft in it [07:15] ah ok [07:15] lamont: which #ifdef cruft? [07:15] Installing new version of config file /etc/mozilla-firefox/pref/firefox.js ... [07:15] Updating mozilla-firefox chrome registry...E: Registration process existed with status: 1 [07:15] E: /usr/lib/mozilla-firefox/extensions/installed-extensions.txt still present. [07:15] +Registration might have gone wrong. [07:15] (i assume you mean ddcprobe) [07:15] hrm.. there's an english error there, and I wonder if I can reproduce that on i386... [07:16] daniels: xresprobe source package. ddcprobe sounds about right [07:21] lamont: i think today (later) we can give back ooo1 on i386 [07:21] the problem is e-d-s [07:21] that exported an extra symbol [07:21] it wasn't my fault at the end ;) [07:22] but i am not sure seb already fixed it with yesterday upload [07:22] just say when.. [07:23] lamont: sure [07:25] lamont: what's the situation with hppa in general? [07:25] (just curios) [07:25] 799 installed [07:26] sounds good :-) === lamont runs the script to slurp in uploads again. [07:27] mirror freshened around around 0130UTC, stage2 is pretty much current [07:28] pool/universe/z/zsh-beta/zsh-beta_4.2.0-dev-1+20040622-1_powerpc.deb [07:28] 1930298 100% 51.18kB/s 0:00:36 (3429, 99.9% of 91862) [07:28] whoooooooooo [07:28] come on, fortnight-long rsync [07:28] you can do it [07:29] ehhe [07:29] i am rsyncing sparc now :-) [07:29] so i can finally change some buildd setup to something sane [07:29] heh, nice [07:29] yup [07:29] i'm just uploading xine now btw [07:30] i think either today or tomorrow it will be possible to bootstrap from it [07:30] daniels: thanks [07:30] no worries [07:30] oh, nice! [07:30] so when will I be able to install my u5? :) [07:30] daniels: i am waiting elmo to bless the kernel (NEW) [07:30] and i think we will need ubuntu-meta seeded properly [07:31] + i need to check if all the uploads have been munged properly [07:31] (just run a quinn-diff on the archive) [07:31] because hounestly i didn't bother to read 1553 mails of REJECT/ACCEPTED [07:31] + i started building universe [07:32] oh my god, xine has a toolkit of its own [07:32] based on xt [07:32] main is pretty much in sync other than OOO [07:32] head -> desk [07:32] wow, that's awesome dude :) nice work [07:32] daniels: thanks :-) [07:32] on the otherside there is a bug (at least on my sparc) [07:32] that needs fixing [07:32] that is mostlikely related to the fact that i am headless [07:33] Total 9 package(s) in state Dep-Wait-Removed. [07:33] Total 8 package(s) in state Failed-Removed. [07:33] Total 799 package(s) in state Installed. [07:33] Total 816 package(s) === lamont smiles [07:33] and on serial console all the lsb crack manage to hang the console badly [07:33] but the machine is alive [07:33] hrm.. not sure that gdb and evolution should be 'removed' though... [07:35] RSYNC COMPLETED [07:50] grumble [07:55] daniels: did you also check the Xmvc in xine-lib? [07:55] because adding the b-d isn't enough [07:55] (+ you can close the bug) [07:55] ;) [07:56] Kamion: livecd rootfs seems to have an issue or 2 for me to fix, but sleep needs to happen first, so I can think straight. [07:56] good night lamont === lamont goes to fall over [07:58] fabbione: hmm? [07:59] fabbione: i installed xine, it started, that was good enough for me [07:59] daniels: i did open a bug for the xine-lib stuff [07:59] with all the details [07:59] ok [08:00] and it is assigned to you [08:00] starting is not enough.. it was starting before too [08:00] but fullscreen wasn't working [08:00] not with xinerama enabled and other stuff [08:00] but there is the Xmvc bits that need some autocrap love that i am not able to do [08:01] that's why i assigned the bug to you :) [08:02] blegh [08:02] 'k [08:11] jdub: i think we can get gfs in for hoary :-) [08:11] fabbione: ... ! [08:11] jdub: do you have a url for the patch? [08:12] it's a feature and we are not in feature freeze ;) [08:12] elvis.redhat.com? [08:12] might be there [08:12] haha [08:12] can you please dig it for me? [08:12] you've changed your tune ;) [08:12] yep, will do [08:12] i am test-building the new inofoty patch [08:12] inotify even [08:12] don't spend too much time on it if it sucks [08:12] i am not... [08:12] just test-building [08:13] updating patches is easy now [08:13] but i need to know if it builds at least [08:13] it might as well fix another bug [08:13] 5431 [08:15] fabbione: not the page you need, but check this out: http://www.redhat.com/software/rha/gfs/ [08:15] fabbione: http://sources.redhat.com/cluster/gfs/ [08:16] yeah i know what it is :-) [08:16] i planned to use it on my file server :-) === d3vic3 [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-devel [08:16] daniels, ping [08:16] http://people.redhat.com/cfeist/cluster/tgz/ [08:16] fabbione: the, um, price is what you should check out ;) [08:17] jdub: yes i know... [08:17] is the code GPL btw? [08:17] yes [08:17] can we actually include it? [08:17] ok [08:18] fabbione [08:18] morning [08:18] hi d3vic3 [08:18] my distro broke, need help [08:18] fabbione: golly, they're maintaining patches in cvs [08:20] when I run `apt-get dist-upgrade` I get errors about dependencies [08:20] and gnome will not start [08:20] jdub: i can see that there a bunch of patches that needs to be integrated... [08:20] ubuntu-desktop fails to load [08:20] fabbione: lots of stuff [08:20] jdub: and they are up to 2.6.9.. so we might need a cvs checkout [08:20] fabbione: might be a bit hairy for the supported kernel [08:20] + userland tools [08:20] d3v yo [08:20] er [08:20] fabbione: there's 2.6.10 in the patches area [08:20] d3vic3: yo [08:21] jdub: yes.. i agree.. specially because we don't have experience on it and not enough time to test it properly [08:21] punt! [08:21] i tought it was a bit smaller than that [08:21] daniels, ubuntu-desktop wont install [08:21] maybe we can make an unofficial module package ;) [08:22] jdub: that's an option.. good luck and thanks for offering volunteer [08:22] :) [08:22] wich kernel is hoary using ? [08:22] :) [08:22] d3vic3: calm down.. one thing at a time [08:22] first of all... what error do you get? [08:22] I'm calm [08:22] hm, u-d should be fine, i thought we fixed that the other day [08:23] dist-upgrade gives me errors about dependencies [08:23] lots of them [08:23] d3vic3: start with an apt-get update [08:23] I did [08:23] then apt-get dist-upgrade [08:23] and see what it says [08:23] I did that [08:23] same things, dependencies [08:24] ok.. let's check your sources.list [08:24] poinst to hoary [08:24] please paste it [08:24] either in pvt or to pastebin [08:24] not in the chan [08:24] ya defentaly not in chanel [08:25] ok [08:25] ok.. start removing hoary-security [08:26] ok [08:27] and ? [08:27] now do again the dance of apt-get update && apt-get dist-upgrade [08:30] gives same errors [08:30] ok [08:30] can you paste (as before) a few of these errors? [08:31] d3vic3: it says in the beginning what you have to do [08:32] apt-get -f install [08:32] wont -f skip the dependencies ? [08:32] apparently one upgrade didn't succeed properly [08:33] d3vic3: no.. there is an inconsistency on your system that needs to be fixed [08:33] and that command will do if it can [08:33] i need to go for 20 minutes or so [08:33] bbl [08:34] ok === Mithrandir grumbles at new postgresql-dev requiring kerberos4 libraries. [08:53] re === herzi [~herzi@d043111.adsl.hansenet.de] has joined #ubuntu-devel [08:55] wb === zenrox [~zenrox@wbar7.sea1-4-10-181-189.sea1.dsl-verizon.net] has joined #ubuntu-devel === `anthony [~anthony@220-253-16-58.VIC.netspace.net.au] has joined #ubuntu-devel [09:20] fabbione, ping === mvo_ [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel [09:21] d3vic3: yes i am here [09:21] managed to fix lots [09:21] but now I get a diff error [09:21] not about deps anymore [09:22] ok such as? [09:22] pasted ptv [09:22] s/ptv/pvt / [09:23] d3vic3: that's a bug in kdelibs-data [09:23] issit [09:23] is it filed ? [09:27] fabbione, whats that command for reconfiguring X ? [09:27] dpkg-reconfigure xserver-xorg <-- hoary [09:28] dpkg-reconfigure xserver-xfree86 <-- warty [09:28] ty [09:30] anyone fixing the kdelibs-data bug, coz now I can install X [09:30] SEB!!! [09:37] d3vic3: in general.. just ask [09:38] don't point to one person.. everybody in here is able to help you [09:38] should we direct this back to #ubuntu? [09:39] as well [09:40] the panel is broken to hell at the moment === daniels files a bug on GTK. [09:42] daniels: it doesn't load when you login? [09:42] daniels: or it looks like it "hangs"? === Keybuk [scott@descent.netsplit.com] has joined #ubuntu-devel [09:43] Treenaks: nah, it seems to have gained xinerama support, so I don't have it spanning my MergedFB now [09:43] Treenaks: and adding more seems to be broken [09:45] next ubuntu conference and LCA were "almost at the same time", right? [09:46] or is that not final yet? [09:46] its definate [09:47] lifeless: ok.. then I need help with intra-Australia travel ;) flights to Sydney/Melbourne are 200 euros cheaper than direct flights to Canberra [09:48] Treenaks: www.virginblue.com.au [09:48] Treenaks: get a direct to sydney, and connect later on in the day to canberra for around $au120 return (~EUR60) [09:48] Treenaks: or quantas for possibly the same rate [09:48] but Ubuntu Down Under might be in Sydney, not canberra [09:48] daniels: hmm.. [09:49] daniels: 5117 <- [09:49] daniels: when will the final decision on that be made? [09:49] Treenaks: dunno, sorry [09:49] daniels: WILL it be made? :P [09:50] Treenaks: i hope so :) [09:50] fabbione: bugger. === Nicolas__ [~Nico@ip-81-11-161-179.dsl.scarlet.be] has joined #ubuntu-devel === mpt [~mpt@210-246-37-72.paradise.net.nz] has joined #ubuntu-devel [09:51] Morning folks! [09:53] moin pitti [09:56] it almost certainly is in Sydney, from what I understand? [09:56] Hi crimsun [09:56] Keybuk: yeah === TerminX [~terminx@terminx.envision7.com] has joined #ubuntu-devel === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has joined #ubuntu-devel === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has left #ubuntu-devel ["Leaving"] === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has joined #ubuntu-devel === sabdfl [~mark@host217-37-231-28.in-addr.btopenworld.com] has joined #ubuntu-devel [10:06] morning all [10:07] hi sabdfl [10:07] Kamion: in debbugs, some of the messages are presented as control messages. are the rules to identify them and present them documented anywhere other than the code? [10:07] sabdfl: I was just asking people.. anything certain about Ubuntu Down Under yet (except for "it will happen") ? [10:07] yes, dates are set [10:07] locations too? [10:08] april 25-30 [10:08] sydney === Mabus [~folk@phobos.coretrek.no] has joined #ubuntu-devel [10:08] many of us will go to LCA in Canberra the week before, then UDU in Sydney [10:08] I'm trying to figure out if I'm going, and if so, how BEFORE early bird registration for lca ends :) [10:09] Sweet. [10:09] Treenaks: what's the loose/gain risk for booking LCA just in case? [10:09] i'm looking forward to LCA [10:11] sabdfl: me too ... but I'm also conscious of the fact I have to decide what it is that I'm talking about *and* write an abstract in the next week or two ;) [10:13] sladen: not much, really.. [10:15] Treenaks: Earlybird finishes 1st Feb [10:15] tuo2: that's <2 weeks from now [10:15] yup. [10:16] Oh, sorry. I though you were asking *when* it ended [10:17] well, I should probably have a flight to sydney, and get on the plane to Canberra from there.. [10:17] morning [10:17] hi ogra [10:17] sabdfl: only 5 days :-O [10:18] Treenaks: you could also do the train from SYD->CAN [10:18] tuo2: how long does that take? [10:19] 4 hours... [10:20] oh that's doable [10:20] justdave: ping ? [10:20] it's quite pretty as well. [10:20] Mabus: pong [10:20] justdave: do you have any control of the build machines ? [10:21] I though there was some talk about doing a big train ride down with slug [10:21] nope [10:21] justdave: it seems the openoffice.org2 build is failing solely because the machine is missing bzip2: http://people.no-name-yet.com/~lamont/buildLogs/o/openoffice.org2/1.9.66-0ubuntu5/openoffice.org2_1.9.66-0ubuntu5_20050118-1612-i386-failed [10:21] oh [10:21] tuo2: thanks for the info :) [10:21] justdave: know who I should poke? :) [10:21] lamont probably [10:21] Treenaks: np. Good to see people coming to lca! :) [10:21] thanks === carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #ubuntu-devel [10:22] Treenaks: is it sure that you fly ? [10:24] hehe...ooo2 becomes really british it seems: Unpacking OO.o build tree - [ go make some tea ] ... [10:26] ogra: I'm not going to walk to australia :P [10:26] Treenaks: g* === Nafallo [~nafallo@h91n7c1o1027.bredband.skanova.com] has joined #ubuntu-devel [10:27] Treenaks: i'm still not sure if i can pay the flight.....have to finish some extra work before :( [10:27] ogra: same for me [10:27] ogra: but it's not april yet! :P [10:28] Treenaks: nope, and its to important to miss :) [10:29] only 5 days looks a bit dissapointing though [10:29] ogra: it's followed by Ubuntu Down Under in Sydney [10:30] err, 25-30 isnt the ubuntu conference ? [10:30] ogra: 25-30 is, but 18-23 is lca [10:31] ah, ok.....so its probably clever to hit them both :) [10:32] I would [10:33] since i never was in au before (and probably will never ever go there) i think i'll too [10:38] ogra: Hi! Congrats to your first package [10:38] ogra: however, the version number (0.061-ubuntu3) looks a bit odd === Astharot [~isager@elite.hacker.eu.org] has joined #ubuntu-devel [10:40] pitti: i will fix it in the next upload.... (elmo had no objections....but was probably tired to point out even more errors) ;) [10:40] pitti: thanks === maskie [~maskie@196-30-108-25.uudial.uunet.co.za] has joined #ubuntu-devel === Mabus [~folk@phobos.coretrek.no] has left #ubuntu-devel ["Leaving"] [10:41] pitti: the original package was build with debmake....so i had to changea _lot_ ....that one just slipped through [10:41] i promise i'll do better in the future :) [10:41] ogra: no problem :-) [10:42] ogra: it's not a big deal, I just wanted to point out this [10:42] ogra: nice to see you in the boat now :-) [10:42] pitti: i'm going for main.....there such errors are not acceptable i think :) [10:42] fabbione: do you know what d3vic3's problem with kdelibs-data was? [10:42] thanks :) i'm on my way :-D [10:42] ogra: everybody makes such silly errors from time to time, I think === mpt [~mpt@210-246-37-169.paradise.net.nz] has joined #ubuntu-devel [10:44] Riddell: provides file from another package [10:44] Riddell, bug #5354 === Crushed_Cigar [~zinc@adsl-68-123-129-15.dsl.sntc01.pacbell.net] has joined #ubuntu-devel [10:50] d3vic3: didn't see you there [10:50] d3vic3: it's supposed to be fixed but I don't think it worked [10:51] i see === Astharot [isager@host252-27.pool62110.interbusiness.it] has joined #ubuntu-devel === trukulo [~mzarza@26.Red-81-45-239.pooles.rima-tde.net] has joined #ubuntu-devel === decko [decko@decko.user] has joined #ubuntu-devel === `anthony [~anthony@220-253-10-120.VIC.netspace.net.au] has joined #ubuntu-devel === sabdfl [~mark@host217-37-231-28.in-addr.btopenworld.com] has joined #ubuntu-devel [11:33] https://wiki.launchpad.canonical.com/RoadMap [11:33] oww [11:34] morning sabdfl [11:34] hi all [11:35] connection is very up and down today [11:35] sabdfl: we are almost ready to launch sparc :-) [11:35] whoot! [11:35] sparc.u.c is there [11:35] missing only a few details and a bit of testing my side [11:35] sabdfl: elmo has done a great job :-) [11:35] Hi people [11:36] thanks elmo, fabbione, can't wait to have it out there [11:36] sabdfl: soon.. soon.. i am burning for it [11:36] :) [11:43] sabdfl: I don't believe any of debbugs' internal log message representation is documented anywhere other than the code; I'm assuming you've already read the documentation in Debbugs::Log, but that doesn't really deal with control messages [11:44] fabbione: I'll whip you up a patch to get useful debug output out of swsusp === seb128 [~seb128@ANancy-151-1-14-38.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [11:45] sabdfl: in general a control message is represented as an html record with a description of the action (which was a terrible design decision - it'd be nice to transform all such records into something properly machine-readable some day) and an incoming-recv record with the text of the mail that did it [11:46] sabdfl: nothing in debbugs right now ever needs to parse the html, only display it, because it applies the necessary changes to .summary at the same time as writing .log [11:48] ogra_: when you awake tell me, i have graveman package for universe [11:49] mjg59: cool thanks [11:50] fabbione: did i tell you that all the videos are up? [11:50] trukulo: no you didn't or i didn't read :-) [11:50] what was the URL again? [11:50] so all of them are up [11:50] badopi.org/videos [11:51] thanks :-) [11:51] sabdfl: yours too [11:51] anyone know the deal with firewire automounting? I was expecting it to work like USB, but this user doesn't seem to be having that with his iPod [11:51] Kamion: ok, thanks [11:51] sladen: does manual mount with pmount /dev/sda2 (or whatever) work? [11:52] heh, review of d-i in Wil Wheaton's blog, and mention of Ubuntu in the comments: http://www.wilwheaton.net/mt/archives/001771.php#001771 [11:52] pitti: do you have to modprobe anything first? [11:53] sladen: actually not === kent [~kent@c83-249-61-131.bredband.comhem.se] has joined #ubuntu-devel [12:06] mdz: don't worry, i can't work out how to stop the stupid thing beeping, you'll get it back asap === ogra [~ogra@s217-115-139-139.colo.hosteurope.de] has joined #ubuntu-devel === d3vic3 [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-devel [12:14] elmo: imagemagick sync, please [12:17] elmo: when you have time can you bless the sparc kernel please? [12:23] lamont: you should really give klogd's dd a bs=128 ---it's very good at buffering messages otherwise. So good that it makes debugging a pain [12:24] thom: here? [12:27] so... where can I read about MOTU's dutys? :-) [12:34] *reading logs* [12:36] what's the plan surrounding ooo2? Is it going to be the OOo for hoary (which means I have to go around and be bloody bothered about it wrt amd64) or can I ignore it? [12:37] AIUI it may be the OOo for hoary if it seems to work well [12:40] fabbione: Has daniels sent you updated DRM stuff yet? === `anthony [~anthony@220-253-10-120.VIC.netspace.net.au] has joined #ubuntu-devel [12:40] mjg59: no, but i want to grab it in full from bk [12:40] Ok, cool [12:41] The DRI tree is on bk now? [12:41] Or have they already submitted the DRM code to Linus? [12:41] it has been merged a lot between 2.6.10 and 11-rc1 [12:42] Ah, ok [12:42] and i don't want any fancy code that hasn't been merged yet [12:42] There's been some extra code added to support DRI over suspend/resume on i810, so it'd be good to get that in [12:43] mjg59: do you prefer to send me a patch? [12:43] that would save me extra work [12:43] fabbione: I'll wait to talk to daniels about it === enrico [~enrico@enrico.developer.debian] has joined #ubuntu-devel [12:43] ok [12:47] pitti: hrm? [12:48] thom: when pitti hunts people, it's better to hide [12:48] heh [12:48] it's usually the start of a endless series of security bugs :- [12:48] ) [12:50] Hrm. Why are people getting the autohinter switched on if they have sub-pixel anti-aliasing? === sbrath [~usrshb@sbrath.user] has left #ubuntu-devel [] === rmt_ [~rmt@pD9E59896.dip.t-dialin.net] has left #ubuntu-devel ["Leaving"] === Nicolas__ [~Nico@ip-81-11-161-179.dsl.scarlet.be] has joined #ubuntu-devel [01:11] elmo: could you refresh Task: ubuntu-desktop? It's broken at the moment; I just uploaded ubuntu-meta to fix it. [01:11] the fixed binaries are in the archive now [01:12] daniels: xlibmesa-dri needs to define a tighter depends on xlibmesa-gl [01:13] Hrm, no, that's not it. [01:13] shouldn't g-v-m try to mount thumbdrives and such which are inserted prior to login? [01:13] Grah. [01:14] daniels: http://www.mail-archive.com/dri-devel@lists.sourceforge.net/msg21142.html - I'm seeing the same === mirak [~mirak@AAubervilliers-152-1-6-165.w82-121.abo.wanadoo.fr] has joined #ubuntu-devel === Nicolas__ [~Nico@ip-81-11-161-179.dsl.scarlet.be] has joined #ubuntu-devel === maskie [~maskie@196-30-108-133.uudial.uunet.co.za] has joined #ubuntu-devel [01:25] mjg59: yeah. i had stuff to do tonight, but before I started doing stuff, I did this: [01:25] * Resync Imakefiles for Mesa, so t_vb_cull gets built, and i915_dri is [01:25] loadable. [01:25] * Add extra guarding to xserver-xorg.postinst, so it will silently assume a [01:25] CRT if xresprobe doesn't feed us a display type (closes: Ubuntu#5639). [01:25] mjg59: and that was before I saw dri-devel. I AM A GOD AMONG MEN. === trukulo [~mzarza@26.Red-81-45-239.pooles.rima-tde.net] has joined #ubuntu-devel [01:26] daniels: tsk... i am your god... [01:27] fabbione: dude, mesa is hardcaw === daniels flashes gang signs at fabbione. [01:27] daniels: You are, truly, a god [01:27] Is that uploaded? [01:28] god: a comment on 5117 please [01:28] mjg59: not yet [01:28] but it built on i386 [01:28] fabbione: if that's trackpoint, throw it away [01:28] did I mention that I HATE PS/2 MICE? [01:29] daniels: and what about the wishlist for getting it in? [01:29] is there any new patch out? [01:31] fabbione: i haven't been able to find any new patch [01:31] ok === jbailey [~jbailey@CPE000ded9d787c-CM014260028338.cpe.net.cable.rogers.com] has joined #ubuntu-devel [01:31] mjg59: because I am so goddamn good and did what you wanted before you asked, you must solve this bug -- https://bugzilla.ubuntu.com/show_bug.cgi?id=5466 [01:31] fabbione: i have lots of cool cards now :) got an imstt, bidding on an ark, a firegl on the way, lots of random cards too [01:32] ahhaha [01:33] daniels: you are always my little punk :P [01:33] anyone have a clue what on earth #5637 is about? [01:33] GTK? [01:33] if nobody knows I'll just punt to the kernel :P [01:33] HA. HA. HA. [01:33] daniels: Choke on my fuck [01:35] Kamion: we should ask him to buy serious hardware [01:35] daniels: For what it's worth, my Sid system doesn't link against them [01:35] Kamion: but that's USB.. so guess what? GTK! [01:35] haha, unlucky, kernel bug it is === fabbione isn't suprised [01:36] oh, you beat me to it [01:36] ehehe [01:36] grah, dude, you don't get to change the package to linux and leave it assigned to me. I am not joining the kernel team :P [01:37] I did push the radio to ressign to QA or owner [01:37] weird === fabbione hits bugzilla with a washing machine === Keybuk wonders whether daniels was being lazy, or really can't save to launchpad :P [01:41] Keybuk: can't seem to log in [01:41] fabbione: haha :) [01:42] daniels: me neither [01:42] fabbione: it's probably getting detected as a Synaptics touchpad [01:42] ho-hum [01:42] Keybuk: awesome [01:43] mdz: i'd like permission to defer 1421 until after the feature freeze -- letting vbetool subsume ddcprobe [01:44] mdz: i've got a kick-arse amd64 motherboard at the moment, but no cpu/ram/case/psu yet [01:44] (and my video card will probably take a while; given it's a pcie+pci board, i'll probably have to run a shitty pci video card for a while, in an sli-capable pcie motherboard. life's little ironies.) === Mithrandir pats daniels on the head. [01:46] what is sli, btw? [01:46] single line interlace (IIRC) [01:46] thom: you remember voodoo 2, where you could use two video cards to drive one screen? [01:46] same stuff. [01:46] righto [01:47] cool [01:47] they render every second line and you can shoot double the number of badassess. [01:47] but only nvidia do it [01:47] or something. [01:47] i did wonder, i had a dual voodoo2s [01:47] apparently two 6800gts in sli is pimpfast [01:47] but the best single card is still x800 xt pe/x850 [01:47] especially as in single mode you get pcie 16x [01:47] daniels: what happens if you put two ATI boards in a SLI board? Can you drive them independently? [01:47] which is like, your entire game's worth of textures in a nanosecond [01:47] Mithrandir: i'm not sure tbh [01:47] my tnt2 works great with gnome ;) [01:48] Mithrandir: but ati did actually make an sli card [01:48] the rage fury maxx [01:48] one board, two gpus [01:48] essentially working in sli [01:48] daniels: it'd be a cheap way to get dual DVI. [01:48] daniels: you're making my 9800pro feel inadequate, stop it :P [01:48] except without having to run some horrific cable between the two cards [01:48] Mithrandir: er, screw that [01:48] Mithrandir: basically every r4xx ever has dual DVI [01:49] daniels: r4xx is what? [01:51] Mithrandir: aiui, x700 is r400, and x800 is r423 [01:51] r400 being agp, r423 being pcie [01:51] but r4xx-class cards are all natively pcie [01:51] so the r400 is actually a pcie chip with an agp bridge :) [01:51] as opposed to the reverse, which is crap [01:51] a pcie firegl would be frightening; those things have astounding amounts of memory [01:51] daniels: Uh, the rage fury maxx wasn't really sli [01:51] They interleaved frames, not lines [01:52] being able to fill that over 16x pcie would be stunning [01:52] mjg59: oh, right [01:52] Which was crack-tastic [01:52] (for decent performance, you needed at least a frame of buffering) [01:53] right [01:53] daniels: actually, it seems like the price of dual-dvi 6600's has dropped significantly [01:53] apparently there were two of them made [01:53] daniels: Xorg says I have version 1.2 of the 915 DRM, and then says that suspend won't work with <1.2 [01:53] I think it means <=1.2 [01:53] ajax has been trying to get docs off ati with no success [01:53] mjg59: probably === Simira [rpGirl@m146i.studby.ntnu.no] has joined #ubuntu-devel [01:54] hiya Simira [01:54] mornin [01:54] /home/daniels/x/xorg/remote/xc/extras/drm/shared/i915.h: * 1.1: Original. [01:54] /home/daniels/x/xorg/remote/xc/extras/drm/shared/i915.h: * 1.2: Add Power Management [01:55] go figure [01:55] hey Simira [01:55] hi fabbione [01:55] mjg59: yeah, X.Org HEAD has 1.2 === no0tic [~no0tic@host191-110.pool8250.interbusiness.it] has joined #ubuntu-devel [01:55] fabbione: have you really integrated new DRM crack recently? [01:55] hwo can I test standby and hybernate hoary features? [01:56] daniels: read above my chat with mjg59 [01:57] thom: you should get an x800 xt 2dhtv, i hear they're awesome :) [01:57] fabbione: ah ok, so not merged yet? [01:57] i'll need a PCIe mobo first :P [01:57] hm, n silbs [01:57] n/no [01:58] thom: you should get one of those, I hear they're cool [01:58] although they went a little bit overboard [01:58] heh [01:58] there are 8 SATA cables [01:58] fuck [01:58] that's quite a bit :-) [01:58] as well as an external plate with 2 SATA connectors (and power) [01:58] and multiple IDE/floppy cables, etc, etc [01:58] there's a staggering collection of shit in there [01:59] daniels: uhm, on a what kind of mobo? [01:59] (the motherboard requires the 24-pin ATX plug, the 4-pin 'P4 power' plug, and a single Molex connector) [01:59] Mithrandir: k8n-sli or whatever [01:59] nforce4 + pcie + sli [01:59] asus? [02:00] oh my god, that's frightening [02:00] 'DFI LANPARTY NF4SLI-D Socket 939, 2000MT/s, PCI Express (SLI mode), DDR400, SATAII, RAID, Dual Gigabit LAN, UV Reactive Slots & Cables' [02:00] yeah, asus [02:00] daniels: dfi's lanparty stuff _is_ overboard, yes. [02:00] uv reactive slots and cables?!? [02:00] Mithrandir: this is mine -- http://www.asus.com/products/mb/socket939/a8nsli-d/overview.htm [02:01] it's a *very* nice motherboard [02:01] and I'll never run out of SATA [02:01] I have six sata thingies on mine.. quite enough. [02:02] mjg59: [02:02] daniels@nanasawa:~xap/debian/xlibmesa-dri/usr/X11R6/lib/modules/dri% objdump -x i915_dri.so | grep _tnl_vertex_cull_stage [02:02] 0019bbc0 l O .data 0000002c _tnl_vertex_cull_stage [02:04] daniels: shame there's a mobo fan on it. [02:07] Mithrandir: *shrug* [02:07] is there a way to figure out from /sys which kernel driver owns a given network device? [02:07] s/device/interface/ === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has left #ubuntu-devel [] [02:09] mjg59: so what was the conclusion for nvidia+acpi? [02:09] seb128: did you fix that e-d-s symbol problem? [02:10] the only way I can see to do it is by looking at the target of the /sys/class/net/*/driver symlink, which is arse [02:10] daniels: and small fans like that are often noisy. I want my machine to be silent. [02:10] <`anthony> mjg59: re daniels' message - on this laptop, suspend/resume works, except that on resume, the screen never gets powered back on. [02:11] fabbione: yep [02:11] seb128: did you tell lamont to give-back ooo1 on i386? [02:12] nop [02:12] ok [02:12] i will do it later :-) [02:12] thanks [02:13] daniels: Yeah, I just rebuilt mesa with the one-line fix [02:13] mjg59: sweet [02:13] daniels: We /could/ fix it, except, uh, we have no rights to modify nv.c [02:13] mjg59: d'oh [02:13] So fuck them [02:14] <`anthony> mjg59: Do you have a capsule description of the problem? I'm happy to send them "as a pissed off owner of one of your cards, please fucking fix this" emails [02:15] anyone here have an ipw2[12] 00? [02:15] Kamion: yeah, ipw2100 [02:15] Kamion: yeah, as thom [02:15] Kamion: ipw2200, why? [02:15] does /sys/class/net//device/rf_kill exist? [02:16] `anthony: Their power management code doesn't integrate with the kernel correctly [02:16] Kamion: yes [02:16] Kamion: not what I can see (warty kernel) [02:16] thom: good, I've got the path right then [02:16] <`anthony> mjg59: suprise! stupid fucking half-open-source piece of crap driver. [02:17] maswan: odd that it doesn't exist for you, but ... [02:17] (this is with hoary) [02:17] maswan: do you have a "hardware" rf switch/ [02:17] maswan: try 'find -follow /sys/class/net/ -name rf_kill'? [02:17] I might just be weird then. :) [02:17] Kamion: I just did that, no result [02:17] ok, no worries [02:19] Kamion: same with hoary + ipw2200 here. [02:19] daniels: Where can I get recent enough DRM to do i915 suspend/resume? [02:19] mjg59: err ... err ... [02:19] mjg59: i assume you're using -686? === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has left #ubuntu-devel ["Leaving"] [02:20] if so, i can build you xorg head's i915 [02:20] daniels: Yeah [02:20] Mithrandir: same as thom? [02:20] Rock [02:20] Kamion: same as thom, yes. [02:20] daniels: Is that in the code that's been pushed to the kernel, or do we need another patch? [02:20] thom, Mithrandir: oh, could you also tell me what /sys/class/net//driver points to? [02:20] fabbione was going to pull the stuff from 2.6.11 [02:21] mjg59: dunno [02:22] guys either way is ok for me [02:22] mjg59: in any case, http://amnesiac.heapspace.net/~daniel/i{8[13] 0,915}.ko [02:22] Kamion: lrwxrwxrwx 1 root root 0 2005-01-19 13:16 /sys/class/net/eth1//device -> ../../../devices/pci0000:00/0000:00:1e.0/0000:02:02.0 [02:22] but it would be better if we can run a test on it before uplaoding [02:22] fabbione: i'm almost certain the stuff that got pushed is what's in xorg head now [02:23] thom: driver rather than device [02:23] oh, sorry [02:23] lrwxrwxrwx 1 root root 0 2005-01-19 14:17 /sys/class/net/eth1/driver -> ../../../bus/pci/drivers/ipw2200/ [02:23] Kamion: ../../../bus/pci/drivers/ipw2100 [02:23] thanks [02:23] I wonder why Mithrandir's has the trailing / [02:24] because I've aliased ls. :P [02:24] oh :-) [02:24] goes away if I call /bin/ls instead of the alias [02:26] just a shame I have to use readlink(2), since that syscall officially sucks [02:27] oh, why? [02:27] you have to guess how long the symlink text is and repeat if you were wrong === Kamion cheats [02:29] oh, and it doesn't append NUL [02:29] yeah, looks like suckage. [02:30] argh...another libc6 upload === jbailey [~jbailey@CPE000ded9d787c-CM014260028338.cpe.net.cable.rogers.com] has joined #ubuntu-devel [02:31] sorry :P [02:31] Kamion: just use PATH_MAX? :) [02:31] hi jeff [02:31] Mithrandir: Jeff would kill me ;) [02:32] [context: readlink(2) sucks] [02:32] Kamion: heh, true. :) [02:32] *lol* [02:32] What have I walked into now? =) [02:33] daniels: Rock, working 3D across suspend/resume [02:33] But I get: [02:33] libGL warning: 3D driver claims to not support visual 0x22 [02:33] libGL warning: 3D driver claims to not support visual 0x23 [02:33] libGL warning: 3D driver claims to not support visual 0x26 [02:33] libGL warning: 3D driver claims to not support visual 0x27 [02:33] mjg59: errrrr [02:33] daniels: Yay! Most recent X update fixes my hang-on-start on ppc. =) [02:33] pitti: done [02:33] Kamion: done [02:33] thanks [02:33] elmo: thanks [02:33] mjg59: how can I test suspend/resume? [02:33] jbailey: really? sweet deal. [02:33] jbailey: let me guess -- you're using an r128 [02:34] no0tic: http://www.ubuntu.com/wiki/HoaryPM [02:35] mjg59: thanks [02:35] daniels: Radeon 9200SE [02:36] jbailey: thanks for that hint, so I won't upgrade my iBook for now :-) [02:36] daniels, mjg59: would it be ok drm in -11? [02:36] jbailey: oh well, close enough ;) [02:36] fabbione: yeah, that's fine for me [02:37] ok === lexhider [~lexhider@dip-220-235-87-133.vic.westnet.com.au] has joined #ubuntu-devel [02:38] mjg59: I have two swap partitions, can I use both in some way to resume? [02:38] fabbione: done [02:40] thom: around ? [02:41] fabbione: Sure [02:41] no0tic: Not currently, nope [02:41] Just choose the biggest [02:41] mjg59: hibernating will copy ram contents to it? [02:41] mjg59: or also something else? [02:42] seb128: kinda head down in firefox, but sup? [02:42] thom: http://bugzilla.gnome.org/show_bug.cgi?id=157435 [02:42] thom: could be interesting [02:43] i'll apply it and see [02:43] Mostly RAM contents [02:43] thom: BTW firefox build-dep on libkrb5-dev ... is there any way to use heimdal-dev instead ? Just wondering because evolution and friends build-dep on heimdal-dev which conflitcs with libkrb5-dev [02:44] mjg59: ok thanks [02:44] seb128: ugh [02:44] grrrrr [02:44] seb128: I think #5632 may still be a problem. [02:44] File "/usr/share/dput/ftp.py", line 59, in upload [02:45] if e.args and e.args[0] [:1] =='5': [02:45] TypeError: unsubscriptable object [02:45] second type that dput crashes in the middle of a gnumeric upload [02:45] that sucks [02:46] lexhider: I've read the comment yep and get the bug here [02:46] mvo_: ping ? [02:47] hm [02:47] seb128: cool, just checking if it was just my end, bye. [02:47] seb128: hmm, one sec === lexhider [~lexhider@dip-220-235-87-133.vic.westnet.com.au] has left #ubuntu-devel ["Leaving"] [02:47] so, ATI are all set up for lib/lib64, not lib32/lib [02:47] seb128: try it again? [02:47] so its 32-bit (binary-only) libGL.so.1.2 tries to load /usr/X11R6/lib/modules/dri/%s_dri.so, not /usr/X11R6/lib32/modules/dri/%s_dri.so [02:48] elmo: the upload ? sure, just a 20min job with my sucking upload link :p [02:48] would patching it with a vi script at l-r-m build time be wrong? [02:48] ** (gnome-cups-manager:10738): WARNING **: FOOBAR [02:48] how _useful_ [02:48] Mithrandir: when I was looking at synaptic source ~2 years ago, it had cerr << _("OMG DUNNO WTF IS GOING ON?!?\n"); [02:49] (it got translated into most languages) [02:49] elmo: thanks [02:49] seb128: ouch [02:49] daniels: *chuckle* [02:49] seb128: tell me about it ... uploading l-r-m orig is 45min [02:49] daniels: coolness [02:50] seb128: while you have TEN THOUSAND message windows with 'HEY DUDE DID YOU SEE ATI RELEASED FGLRX?!? DUDE! FGLRX! AMD64? XORG? DID YOU SEE?' [02:50] daniels: well, did you? [02:50] Treenaks: did I see? === Treenaks hides [02:51] daniels: yes [02:51] daniels: well, they're not in the archive yet, so you can't have noticed. === silbs [~sbsm0084@host217-37-231-28.in-addr.btopenworld.com] has joined #ubuntu-devel [02:51] Mithrandir: ?!? [02:51] giggle [02:52] daniels: ah ah [02:52] Mithrandir: http://archive.ubuntu.com/ubuntu/pool/restricted/l/linux-restricted-modules-2.6.10/xorg-driver-fglrx_6.8.0-8.8.25-0ubuntu2_amd64.deb [02:52] bleah [02:52] daniels: bah, ok, then. :P [02:52] Mithrandir: if thom wasn't such a fascist, I'd figlet kthxbye at you [02:53] Kamion: i think you should be able to start seeding from sparc.u.c [02:53] we still don't have a real sparc ubuntu-meta [02:53] even if it builds. it's empty [02:54] daniels: ---> Installing 'figlet-2.2.1' from a port (misc/figlet) [02:54] ---> Building '/usr/ports/misc/figlet' [02:54] <`anthony> mjg59: fwiw, if I can stomach it, my plan is to try and figure out why the nvidia driver is busted, fix it, and post an ed-format diff somewhere. nvidia can go fuck themselves if they want to complain about that, since it contains none of their code. [02:55] seb128: cool, you packaged nautilus-sendto :) [02:55] Simira: you were looking for me? [02:55] ogra: yep, that was the plan :p [02:55] thom: thanks [02:55] will this go into main ? [02:58] fabbione: ooo1 is ready to give back then? [02:58] lamont: good morning.. and yes [02:58] fabbione: ok. ubuntu-meta will need its source changed to look at sparc.ubuntu.com; best ask mdz if that's ok [02:59] Kamion: sure... [02:59] i will ask him later [02:59] fabbione: done [02:59] lamont: hi! [03:00] lamont: MLKJ+2 today? [03:00] nah - I got enough vacation in yesterday - although I my go back to bed for a couple hours this morning - bed happened late last night. [03:04] fabbione: that's just a give-back on sparc? === koke [~koke@rm-001-26.serve.com] has left #ubuntu-devel [] [03:05] fabbione: ugh, germinate doesn't support working off sparc.ubuntu.com yet; no Sources file ... [03:05] lamont: so I can tell you that apache security update does not build? [03:05] it'd need to be extended to allow getting Sources from a different place from Packages [03:05] lamont: I didn't get a failed build log either [03:05] pitti: either way, I'm going to reboot my machine in about 20 seconds... [03:05] brb [03:06] so I'll still have to do slightly scary hacks [03:07] Mithrandir: nah sparc will build fine and it is building right now. the problem was i386 specific [03:07] Kamion: there are no sources on sparc.u.c. It needs to use the same from archive. [03:08] Kamion: sparc has only the binaries. there is very litte point in mirroring them twice [03:08] yes, I know [03:08] fabbione: ok, so it'll be given-back on i386. That's fine. I just bloody don't want anybody to upload a new OOo unless they have to. It's painful for me. [03:08] it still needs a germinate extension to support that. :) === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu-devel [03:08] Mithrandir: it has been done already. === zul [~chuck@zul.developer.gentoo] has joined #ubuntu-devel [03:09] lamont: what's the status with the hppa kernel? [03:09] fabbione: did you get the files I tossed you? [03:10] lamont: via email? [03:10] i don't have irc scrollback or log [03:10] actually, I think it was a URL in /msg [03:10] so if you pasted here no [03:10] argh... [03:10] let me find it. [03:10] sorry no.. [03:10] thanks [03:10] people.ubuntu.com/~lamont/fabbione.pa.tar.gz [03:10] daniels: Hrm. Is the nvidia driver patched in linux-restricted-modules? [03:11] has modified control and control.stub, debian/config/hppa, and a modified pa_patch [03:11] lamont: ok. perfect [03:11] you'll want to migrate the build-depends from control*, since it's relative to -8 [03:11] it will be in my tree in a few minutes :-) [03:11] but the others just drop in. [03:11] mjg59: how do you run vbetool to post? vbetool post? [03:11] lamont: sure.. don't worry [03:11] mjg59: yeah, needs to be patched for .10. why? [03:11] pitti: checking on your build === fabbione -> coffee [03:12] daniels: There's no permission to do so [03:12] daniels: Yeah [03:13] daniels: Check the copyright headers on nv.h and nv.c [03:13] morning all [03:14] thom: http://live.gnome.org/Epiphany_2fMozillaPatches === lamont wonders WTH is sound went... [03:15] at full volume, I get popping [03:15] GRRRRR [03:15] Uploading via ftp gnumeric_1.4.2.orig.tar.gz: Error '(32, 'Broken pipe')' during ftp transfer of gnumeric_1.4.2.orig.tar.gz [03:15] HATE dput === seb128 scp the files on chinstrap this time [03:16] seb128: dude, it's not dput [03:16] sorry, I didn't realise it was dieing on the orig.tar.gz [03:16] our new upload daemon has timeout issues - stevea is working on it, but in the meantime, if you have a large orig.tar.gz, like gnumeric, on a slow link, 'rsync -e ssh' it to chinstrap first [03:16] elmo: [03:16] File "/usr/bin/dput", line 857, in main [03:16] progress=config.getint(host,'progress_indicator')) [03:16] File "/usr/share/dput/ftp.py", line 59, in upload [03:16] if e.args and e.args[0] [:1] =='5': === ogra_ [~ogra@p508EBB3D.dip.t-dialin.net] has joined #ubuntu-devel [03:16] elmo: oh ok ... thanks [03:17] I'm really surprised we haven't had this problem before now.. and now 3 people in as many days have hit it.. [03:17] elmo: I've had it for a while, but I usually get by by uploading to something which does a few MB/sec. [03:18] elmo: it's timing out because of lack of activity on the control socket?? [03:18] lamont: nah, the code's buggy.. it's timing out because the timer's not being updated ;-) === mirak [~mirak@AAubervilliers-152-1-6-165.w82-121.abo.wanadoo.fr] has joined #ubuntu-devel === sladen [paul@starsky.19inch.net] has joined #ubuntu-devel [03:18] Mithrandir: thanks for telling me dude ;-P [03:19] elmo: I didn't know if it was my end or dput being dodgy or not. [03:19] elmo: and I don't like to complain if I can work around issues relativetly easy and I don't know where the problem is [03:24] lamont: ia64's still building glibc, I take it? [03:25] ok, new upload daemon in place.. if this doesn't work, we get to tar'n'feathe the entire zope3 upstream ;-) [03:26] seb128: why doesn't shift-insert paste what I just button-3-cut? [03:27] seb128: nice! i'll look at and review [03:27] lamont: that's a xfree's question for daniels probably [03:27] thom: yep, thanks :) [03:28] shift-insert is an application thing not an X thing surely [03:28] in which app so ? :) [03:28] Kamion: make[3] : Leaving directory `/build/buildd/glibc-2.3.2.ds1/build-tree/glibc-2.3.2/login' [03:28] xterm, of course [03:28] elmo: ... so, if I were to tell you that we can't actually distribute l-r-m at all, and we were technically violating the licence of one of the modules in there ... [03:28] ah, so it is an xorg question then ;) [03:29] lamont: works for me (tm) ... [03:29] lamont: button-3-cut is the PRIMARY selection, shift-insert is CLIPBOARD [03:29] (that's how most apps implement it -- gtk bug) [03:29] elmo: (speaking of, can you demote l-r-m-2.6.9 to universe?) [03:29] daniels: wasn't that clear from the day you included the ipw2100 firmware? Or do you have a special license for that? [03:29] (there's PRIMARY, SECONDARY, and CLIPBOARD) [03:29] lamont: see, daniels knows about this :) [03:29] seb128: ok. [03:29] azeem: there's no ipw2100 firmware in lrm [03:29] (even if the evil guy blames gtk *again*) [03:29] oh :) [03:30] I wonder how my WLAN works then [03:30] daniels: so how do I copy into the clipboard? [03:30] lamont: edit->copy or whatever === netjoined: irc.freenode.net -> benford.freenode.net === sladen [paul@starsky.19inch.net] has joined #ubuntu-devel [03:30] you don't want CLIPBOARD; try to use PRIMARY (which is usually just selecting with the mouse) whereever you can [03:30] azeem: it's in linux-source, not l-r-m [03:31] I WANT MY SOUND BACK! === mirak [~mirak@AAubervilliers-152-1-6-165.w82-121.abo.wanadoo.fr] has joined #ubuntu-devel [03:31] ah === haggai [~halls@i-83-67-20-196.freedom2surf.net] has joined #ubuntu-devel [03:32] lamont: eh? [03:32] are the modules loaded? [03:33] fabbione: _I_ didn't change anything (other than an upgrade of the system :-) [03:33] lamont: i didn't say you did change something... [03:33] lsmod |grep snd [03:34] kamion: boggle, no? [03:34] echo $(lsmod | awk '/snd/ { print $1}' ) [03:34] snd_intel8x0 snd_ac97_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_timer snd soundcore snd_page_alloc [03:34] Kamion: multiverse maybe, or just remove it [03:34] but not universe [03:34] lamont: check the mixer [03:34] elmo: er, sorry, multiverse is what I meant [03:34] lamont: probably you hitted one of these alsa "i kill your mixer settings" obscure bugs [03:35] elmo: linux-source-2.6.9's in universe at the moment so it would make sense to match [03:36] fabbione: and the recovery would be? === netjoined: irc.freenode.net -> benford.freenode.net === mirak [~mirak@AAubervilliers-152-1-6-165.w82-121.abo.wanadoo.fr] has joined #ubuntu-devel [03:36] pitti: yes, having the langpack base package depend on the update packages seems sensible to me [03:37] Kamion: would the current structure work reasonably easy with the installer? [03:37] i hate how synaptic doesn't use my gtk theme. =/ [03:37] Kamion: i. e. do you already have the language code in d-i? [03:37] pitti: yes, I have the language code [03:38] Kamion: okay, so actually there is now nothing left to decide any more [03:38] just give me a set of rules for (language/country => package names) basically and I'll implement them [03:38] or locale => package names, either works [03:38] Kamion: basically, it's just "install language-support- and language-pack-" for language code [03:39] what about things like pt_BR? [03:39] does -pt get both? [03:39] Kamion: it was decided to merge all countries in a single per-language pack [03:39] jdub: hey. this new tomboy is using dbus mono bindings, i have them locally (source in hoary dbus-mono) but there is no bin in hoary yet.. no go for now [03:39] ok [03:39] Kamion: otherwise you get crazy with e. g. de-at and de-de === no0tic [~no0tic@host191-110.pool8250.interbusiness.it] has joined #ubuntu-devel [03:39] libc6 new update works? [03:40] pitti: there are obviously different cases; I think that should have been decided case by case [03:40] no0tic: was a one-line change [03:40] Kamion: we discussed a little about zh_CN and zh_tw [03:40] pt_BR is a totally different situation from de_AT [03:40] Kamion: but in the end there was an agreement to have only per-language [03:40] Kamion: this also saves us some 150 packages [03:40] Kamion: that messed up all ;) [03:40] no0tic: huh? [03:41] Kamion: one-line change... (a joke) [03:41] no0tic: if you have problems, please give details. === Treenaks reads the description of daniels on http://lca2005.linux.org.au/speakers.php and LOLs [03:42] Kamion: no, it was a joke, np [03:43] fabbione: someone decided to switch things to the oss mixer... [03:43] lamont: no recovery.. :( just set the mixer back again [03:43] (fixed) [03:43] eh? oss? [03:43] that's not me [03:43] lamont: gtk! [03:43] remember.. it's always gtk [03:43] fabbione: nah - let's just blame seb128 directly [03:43] :-) [03:43] haha [03:44] grrr :p === lamont makes a note to buy seb128 a beer in .au, before he goes postal or something [03:45] lamont: I would recommend some morphine or similar substances. [03:45] Mithrandir: some opiate, eh? === netjoined: irc.freenode.net -> benford.freenode.net === mirak [~mirak@AAubervilliers-152-1-6-165.w82-121.abo.wanadoo.fr] has joined #ubuntu-devel [03:46] lamont: can i upload 2.6.10-10 ? [03:46] or that would kill your mirror? [03:46] yeah [03:46] fabbione: whatever. === fabbione feels a tone of desperation in "whatever" [03:46] nah === lamont has other issues with his mirror atm [03:46] oh ok [03:47] wasabi: There? [03:47] yes === alerios [~alerios@201.245.164.174] has joined #ubuntu-devel [03:48] wasabi: When you're running eclipse in the various JVMs, are you including the standard gcj in that set? I had thought that eclipse3 needed gcc4's bits to make it work. [03:49] No. [03:49] Kaffe only at this point. [03:49] ANd Kaffe just barely === alerios [~alerios@201.245.164.174] has left #ubuntu-devel ["Abandonando"] [03:49] 'kay. I thought I saw a success on #classpath for it with jamvm and CVS classpath. [03:49] Might be. It all falls to classpath at this point. [03:49] All the VMs are fairly good. Sable, Kaffe, GCJ, whatever, etc. [03:50] I haven't tried it yet, though. I usually have either blackdown or sun's jvm installed. [03:50] But they all use classpath. [03:50] And classpath is a bit lacking in sid currently. [03:50] My packages work flawlessly on sun's. ;) [03:50] fabbione: so more dispair or apathy than desparation [03:50] Either way, it runs Good Enough on Kaffe for main. [03:50] In sid at least. [03:51] Seeing as our current packages are 1.5 years out of date and only work on Sun's period. ANything is an improvement. ;) [03:51] No kiddin'. [03:51] Did you wind up disabling all of the auto update features? [03:51] Nope. [03:51] It's off by default. [03:51] lamont: ehe don't worry.. i will probably release later today or tomorrow [03:51] Oh? Cool. [03:51] You have to specifically update it yourself... and to do so you have to update it into a second update site. [03:52] lamont: i need to check that the changes will not break anything [03:52] SUch as /usr/local/share/eclipse or ~/.eclipse [03:52] lamont: like the binary compatibility [03:52] And if a user choose to go through all that, let em. [03:52] pitti: ping? [03:52] wasabi: Yeah. [03:52] fabbione: pong [03:53] I'm pretty confident all the free VMs will run Eclipse. [03:53] Depending on how close to classpath CVS they are. [03:54] Eclipse gives us a free compiler in main too, a perfect compiler. ;0 THe Eclipse JDT. [03:55] Hmm, I'll give your packages a try. I usually use jamvm just because it uses cp directly. But the package hasn't been updated yet. [03:55] please do. I want any and all info/testing regarding them. ;) [03:55] How does the Eclipse JDT do vs. jikes? [03:55] Jikes is barely competitent. [03:56] I would never it near production code. [03:56] It can't handle inner classes at all. IT's import parsing orders are all out of wack. [03:56] Eclipse JDT is *perfect*. And I mean it. [03:56] IBM would not be able to sell WebSphere if it wasn't. [03:56] So... yeah. =) [03:57] It's CPL though. [03:57] It's good enough to compile stuff, but not GPL compatible. [03:58] is that a problem for a compiler? [03:58] Dpeending on your interpretation of the GPL. [03:58] heh [03:58] seb128: the e-d-s symbol is not fixed... [03:58] seb128: ooo1 failed again [03:59] and judging from the log size, exactly at the same point [03:59] wasabi: rather, depending on whether it claims that object code it produces is also CPL [03:59] It does not claim that. [03:59] Kamion: As long as there's an exemption allowing the code output to be whatever license you'd like, it ought to be fine. [03:59] if it doesn't then I don't think one's interpretation of the GPL matters [03:59] seb128: what version of e-d-s should fix that? [03:59] jbailey: indeed [03:59] There is a big ass d-l thread regarding Java compilers + GPL. [04:00] I can see both sides of it. [04:00] -legal ... *shrug* [04:00] fabbione: 1.1.3-0ubuntu8 [04:00] fabbione: but haggai said there was an OO.o issue too, are you sure that's fixed ? [04:01] seb128: that's what i understood from you [04:01] wasabi: Do you know if it's possible to drop their compiler into a separate package and invoke it as javac? [04:01] The package name is eclipse-javac [04:01] I made a wrapper script for it. [04:01] Nice. =) [04:01] THere is also an ant plugin [04:02] fabbione: are heimdal-dev and libnspr-dev installed for the build ? [04:02] In fact... I'm building Eclipse with it. The eclipse packages are two stage... use Kaffe to build the JDT... then use the JDT to rebuild everything. [04:02] kjc isn't good enough to compile ALL of eclipse without modifications. [04:03] fabbione: i should've taken your advice and run away [04:03] thom: at some day I had caught you anyway [04:03] MUHAHAHAHA [04:03] seb128: sorry i need to.. you can check in the log :-) [04:03] thom: ahhaha [04:03] i have a meeting with the kitchen company now [04:03] but you would have had to do this firefox release [04:04] bbl [04:04] sorry, thom! :-) [04:04] that patch is a SCREAMING NIGHTMARE [04:04] thom: the injection? right [04:04] thom: if it does not work, and you don't have the feeling that this is wartyable, let's forget about it [04:04] thom: there are enough other issues :-) [04:05] i don't see it being wartyable [04:05] it's blowing the ABI and API into tiny little shreds and stomping on them [04:05] thom: then I rather write a stanza about it in the USN [04:08] go mozilla, it's your birthday. one of the hunks is 400 lines of mixed style change and code change [04:09] I vote that security advisories give a rating out of 10 for quality of vendor provided patch === pitti tries to escape thom's axe [04:09] mjg59: this window injection thingy is not really a patch; it's a complete rewrite [04:10] thom: what about hoary? [04:11] what's the way to make baz shut up about it not knowing what kind of archive an archive is? [04:12] Hello. Someone in the docteam is taking care of making nice release notes for Hoary, and could use some details about it (where it will be published, how long it should be, what level of detail...). Who can I refer him to? === Nafallo [~nafallo@c-14ff70d5.038-1-65736b2.cust.bredbandsbolaget.se] has joined #ubuntu-devel === jinty [~jinty@212.145.79.86] has joined #ubuntu-devel [04:13] Mithrandir: beat lifeless until he shuts it the hell up [04:14] pitti: i'm still applying the patch for hoary, will see how it goes [04:15] thom: since the bug is fixed, is it already contained in a new upstream release? [04:15] thom: might be easier to package that === `anthony [~anthony@220-253-10-120.VIC.netspace.net.au] has joined #ubuntu-devel [04:16] <`anthony> mjg59: You are an a-grade legend. Your vbetool was the magic bit I needed to make this laptop unsuspend right [04:16] <`anthony> Thankyouthankyouthankyouthankyou. [04:16] `anthony: Rocking [04:17] <`anthony> mjg59: Well, that, and making sure i rmmod'd uhci_hcd first. [04:17] `anthony: The scripts in Hoary ought to do that [04:17] <`anthony> the wacky dell usb-to-ps/2 replication stuff plays, well, very very badly, with suspend. [04:17] pitti: nope - there's not been a firefox release since 1.0 [04:17] bad [04:17] they should [04:17] yep [04:18] mozilla? [04:18] <`anthony> mjg59: That would be awesome if it automatically did this. It doesn't _always_ work. One time in about 5, the video driver goes utterly batshit and you get a gray screen fading in from the sides, but that could've been something else I was fucking about with causing that. [04:18] they want to go straight for 1.1 i think, which means merging trunk on to the firefox branch [04:18] pitti: 1.8 is still alpha [04:19] 1.7 won't get the fix since it's too intrusive [04:19] hahaha === douglas [~douglas@suporte2.unilestemg.br] has joined #ubuntu-devel [04:19] `anthony: In Hoary, the default is to rmmod all the USB stuff and do vbe stuff on suspend/resume [04:20] pitti: (they say the same for a 1.0.1 of firefox if it was considered) [04:20] <`anthony> mjg59: Also, kill orinoco [04:20] thom: okay, so WONTFIX for warty [04:20] definitely [04:20] <`anthony> s/kill/rmmod [04:20] `anthony: We call cardctl eject, which ought to be enough for that [04:21] oh man, i so loathe C++ [04:21] <`anthony> mjg59: I needed to do vbetool post - vbetool vbestate save/restore didn't do the job. [04:21] `anthony: Hm. That's a shame. [04:21] vbetool post kills some machines :) === no0tic [~no0tic@host191-110.pool8250.interbusiness.it] has joined #ubuntu-devel [04:21] <`anthony> Yay! [04:22] it's useful for you to know my experiences with standby & hibernate? [04:22] no0tic: Yup, but I have to go out for a few minutes [04:22] Can you add them to www.ubuntu.org/wiki/HoaryPMResults ? [04:22] yes [04:23] <`anthony> mjg59: I can add stuff, although this box is still running FC until Hoary is out. === koke [~koke@rm-001-26.serve.com] has joined #ubuntu-devel === koke [~koke@rm-001-26.serve.com] has joined #ubuntu-devel [04:25] mjg59: how can I edit the wiki, once logged in? [04:32] there's an "Edit" tab around the top left [04:32] pitti: i'm bailing on this patch for hoary, too [04:32] pitti: i don't think it's reasonable or supportable [04:35] c [04:35] uh, this isn't mutt === thom trys that on the correct desktop === dasenjo [~dasenjo@200.21.83.173] has joined #ubuntu-devel === Nafallo [~nafallo@c-14ff70d5.038-1-65736b2.cust.bredbandsbolaget.se] has joined #ubuntu-devel [04:43] :( === no0tic [~no0tic@host191-110.pool8250.interbusiness.it] has joined #ubuntu-devel === douglas [~douglas@suporte2.unilestemg.br] has left #ubuntu-devel ["Fui] === pitti [~martin@box79162.elkhouse.de] has joined #ubuntu-devel === eruin [~eruin@eruin.user] has joined #ubuntu-devel === amu [~amu@amu.developer.debian] has joined #ubuntu-devel === wasabi_ [~jhaltom2@64.25.11.66] has joined #ubuntu-devel [05:09] Kamion: i assume /etc/profile is the best way to set global environment variables? [05:09] daniels: the whole notion of global environment variables is a horror show [05:09] Kamion: no shit [05:10] /etc/profile is global to all interactive login shells === ironwolf [~ironwolf@c-24-6-169-124.client.comcast.net] has joined #ubuntu-devel === netjoined: irc.freenode.net -> benford.freenode.net === mirak [~mirak@AAubervilliers-152-1-6-165.w82-121.abo.wanadoo.fr] has joined #ubuntu-devel === ironwolf [~ironwolf@c-24-6-169-124.client.comcast.net] has joined #ubuntu-devel === ironwolf [~ironwolf@c-24-6-169-124.client.comcast.net] has joined #ubuntu-devel [05:25] oooh. [05:26] jbailey: rumers about a free jvm by ibm or something. ;) [05:26] yeah, they happen every 6 months [05:26] wasabi_: I've read those rumours for a while, but haven't been able to track one down. [05:26] Frankly I think the best bets are going to be the funding from Brazil for classpath hackers. [05:27] ross burton has something on his blog [05:28] He's on planet.gnome, right? [05:28] yeah [05:28] Feh. Quoting esr. Must be vapourware. [05:28] haha [05:30] I've been looking at the jpackage stuff the last couple days trying to figure out how I feel about it. It looks so handy, but it's so... rpm based. =) [05:31] I had originally used the jpackages for Eclipse as a base for the .debs [05:31] But, they are horrid. [05:31] Unreadable spec files. [05:31] tons of pushd and popd, in no particular order. [05:31] So, I just looked at them and figured out what they did. [05:31] I am not excited about JPackage at all. [05:31] scripts that use pushd and popd generally need to be thrown away [05:31] It's a good resource for learning what needs to be done to make .debs. ;) [05:32] (as opposed to subshell, cd, exit subshell) [05:32] wasabi_: mjw pointed me at it as a way to consider doing things. === pepesan_ [~pepesan@40.Red-80-33-116.pooles.rima-tde.net] has joined #ubuntu-devel [05:34] morning [05:34] lamont: ping? [05:34] morning [05:34] Hi mdz! === lamont bets mdz is after livecd answer [05:34] s [05:35] lamont: curious about the rsyncable experiment [05:36] elmo: any idea why libxml2-python2.4 fell out of ubuntu-desktop? python2.4-libxslt1 still claims to need it ... [05:36] uh, huh? [05:36] seb128: can I change libxslt1 to make the python bit depend on the new python2.4-libxml2 package name? [05:36] hi all [05:37] lamont: sladen said you were going to take a shot at getting advfs to use the gzip --rsyncable stuff [05:37] Kamion: sure [05:37] mdz: I changed the langpacks so that l-support does not depend on l-pack any more [05:37] mdz: news to me [05:37] mdz: so you can install auxilliary packages and gettext data independently [05:38] but now I see that the most recent live CD build is 20050115, in which case I'm more interested in that [05:38] is germinate bug, I bet ... [05:38] mdz: as proposed on the ML [05:38] Kamion, lamont: what happened with today's build? [05:38] mdz: I know that there was some discussion about doing that [05:38] mdz: looking into what happened [05:38] pitti: sounds good [05:38] rootfs build barfed, at the very least === lamont disappears to the other workspace to figure things out [05:39] make: *** [/srv/cdimage.no-name-yet.com/scratch/tmp/hoary-amd64/packages-stamp] [05:39] mdz: I built another round of pristine source and binary packages now, I can upload afterwards if you are fine with that :-) [05:39] Error 1 [05:39] ERROR WHILE BUILDING OFFICIAL IMAGES !! [05:39] whatever that means [05:39] pitti: I am fine with that [05:40] sheesh, it's been doing that for days. sorry I didn't notice [05:41] lamont: king.warthogs.hbd.com is not answering HTTP queries [05:41] please fix, kthxbye [05:42] Kamion: it's part of the whole 'deal with the datacenter changes' cruft [05:42] I think it's fixed, testing now [05:42] cjwatson@little:~$ ping king.warthogs.hbd.com [05:42] PING king.warthogs.hbd.com (82.211.81.150) 56(84) bytes of data. [05:42] From little.warthogs.hbd.com (82.211.81.136) icmp_seq=1 Destination Host Unreachable [05:42] Kamion: btw, the name is 'king.buildd'. [05:42] same for all the other buildds [05:42] oh, I didn't know it had been changed [05:43] yeah - this past saturday [05:43] the fact that I broke the livecd rootfs build at the same time is, um, interesting [05:43] lamont: they answer from chinstrap, but not from little [05:44] see routing on chinstrap vs on little, then [05:44] and pester elmo [05:44] would also be nice if somebody would turn off the old names so that I got proper errors [05:44] there is some lagtime in making dns changes to w.h.c :-( [05:44] elmo: could you teach little how to route to the buildds so I can fetch live rootfs builds, please? [05:45] I've fixed debian-cd for the new names [05:46] actually, I didn't break the livecd rootfs until after the build on the 17th [05:53] Kamion: did you upload a new debootstrap with the libunwind7-dev fix [05:53] ? [05:53] in scripts/hoary? [05:53] lamont: is it fixed now? [05:53] (live images, not ia64) [05:54] lamont: I took libunwind7-dev out, yes [05:54] rootfs build appears to be fixed [05:54] debootstrap would have failed since libc6.1-dev isn't in required (nor should it be) [05:54] lamont, Kamion: please do a fresh build so that I can test yesterday's casper changes [05:54] Kamion: yeah - that's what I just ran into... [05:54] mdz: will do once I have routing [05:55] mdz: he needs routing [05:55] elmo? [05:55] it seems to have been broken since 20041115 :( [05:55] Kamion: yeah, that'd be right [05:55] lamont: but once he has routing, there are three new cloops waiting? [05:55] elmo: can you please sync dpkg-1.10.26? [05:55] well, in about 20 minutes or so [05:55] ok, thanks === Kamion is also trying to get a new Array CD done today, ideally ... [05:56] mdz: so that could be a 'yes' :-) [05:56] 10.211.37.0/24 via 82.211.81.157 dev eth0 [05:56] Kamion: I know I told elmo that the move required changes on the cd build side... should have told you then, too.. sorry [05:56] looks like what you need [05:57] thom: here? [05:57] mdz: yes [05:57] thom: can you do a quick fix for us on little until elmo returns? [05:58] just to fix the buildd routing? [05:58] sure [05:58] yeah [05:58] Kamion: I'll tell you when the rootfs's are ready [05:58] and someone email elmo so that he can fix it permanently [05:59] Kamion: you want that, or shall I? [05:59] I'll do it [05:59] rt@rt...? [06:02] yup [06:03] mdz: done [06:03] thom: thanks [06:03] mail sent === mdz mumbles about iproute not being installed [06:04] --- king.buildd ping statistics --- [06:04] 1 packets transmitted, 1 received, 0% packet loss, time 0ms [06:04] thanks, seems better === thom goes home to see how his brother is [06:11] had arthroscopy (however you spell it) yesterday [06:14] Kamion: images current on all 3./ === lamont still wants to know what we're gonna do about ooo/ia64 [06:17] mdz: how about making openoffice.org stuff be [amd64 i386 powerpc sparc] ? === smurfix [~smurf@run.smurf.noris.de] has joined #ubuntu-devel [06:17] how big are those _all.deb's I wonder... /me checks [06:18] new daily-live build running [06:18] Kamion: fine with me [06:18] lamont: thanks [06:18] at least the stuff in desktop [06:19] Kamion: that bloats the archive by about 54 MB currently [06:20] what does? [06:20] how about making openoffice.org stuff be [amd64 i386 powerpc sparc] ? [06:20] that [06:21] doing that wouldn't kick them out of main [06:21] it would merely kick them out of ubuntu-desktop [06:21] ?? [06:21] why would it kick them out? [06:21] of which? [06:22] we're talking at cross-purposes, I feel ... [06:22] oh -wait - where are you suggesting to change../ === x4m [~max@235-114.240.81.adsl.skynet.be] has joined #ubuntu-devel [06:22] lamont: I'm suggesting adding [amd64 i386 powerpc sparc] to all the seed entries in desktop for openoffice.org [06:22] Kamion: woot. agree [06:22] lamont: oh, you thought I meant make the actual .debs arch-specific? no, not that [06:22] right [06:22] same syntax and all that [06:23] done [06:23] next ubuntu-meta upload should take care of it [06:24] mvo_: did you receive my email about apt 0.6.30? [06:24] mdz: yes, I would like to do it all in one go and was asking about dpkg first [06:25] Keybuk gave me his ok to upload a patched dpkg as well, he just asked me to wait for the dpkg-1.10.26 sync [06:25] mdz: daily-live build up, seems to work this time [06:25] excellent [06:25] thanks, Kamion and lamont [06:25] and thom [06:26] did you ask for the sync? [06:26] Keybuk: yes [06:27] coolios [06:27] but only a couple of minutes ago :) [06:28] mvo_: done [06:28] elmo: great, thanks! [06:31] mvo_: about #5632 ... your patch was not supposed to fix that ? [06:32] seb128: that was what it intended, let me check [06:35] seb128: the problem seems to be in gnome-system-tools [06:36] modem-applet calls g-s-t [06:36] seb128: just assign the bug to me, I'll take care of it [06:36] mvo_: gni ? network-admin ask for the root password if it's not runned by root ... that's why we have "gksudo network-admin" as a menu entry :p [06:37] mvo_: there is a bug open about "gst should use sudo instead of su internaly" ... is that a dup ? [06:37] ie: do you want to fix gst or to workaround the applet ? [06:37] seb128: probably [06:37] seb128: probably a dup [06:37] ok [06:40] mdz: ok with supressing $distro-changes mails for language packs? [06:43] hi all!!!!!! [06:44] im lookin for someone what is developin the ppc port [06:51] mvo_: it is already using it IIRC? === T-Bone [varenet@T-Bone.developer.debian] has joined #ubuntu-devel [06:54] sivang: speaking about gst ? [06:54] seb128: yes [06:54] sivang: it uses su [06:54] seb128: ergh, I thought that was already patched... [06:54] sivang: we workaround by using "gksudo <...-admin>" in the menu entries [06:54] pepesan_: you're more likely to get an answer if you say what you want ... :) [06:55] seb128: oh, ok, so mvo is going to make it use libgksudo ? [06:55] sivang: you are in the cc of the bug, I've just commented on it [06:55] Kamion: Hi, i wanna help to the ppc port and help to the live ppc port [06:55] sivang: dunno ask him for the details. The bug has a patch to use sudo instead of su but doesn't build here, the idea is here BTW [06:55] pepesan_: download, send patches :) [06:56] pepesan_: bear in mind we try to minimise the amount of stuff that's architecture-specific; the new live powerpc CD deliberately uses pre-existing code almost all the way, so it took almost no explicit porting effort [06:56] Kamion: i represent a part of a spanish region waht develop a distro for educational pourpuses [06:57] if you want to work on the live powerpc CD, you might as well just work on the live CD in general, for most purposes [06:57] Kamion: we want to help the betatesting and development of the live ppc and ppc distro [06:58] what do you need to know? [06:58] there are many mac in the institutes of this region [06:58] Kamion: who help in the port [06:58] sorry [06:58] how help in the port === T-Bone sees cdimage.u.c throughput sucking: 180ko/s ;^P [06:59] pepesan_: find something you need to fix, fix it, repeat :-) [06:59] Kamion: gonna try last ia64 hoary. Seems that there are still issues with it [06:59] Kamion: we want to port many packages to ubuntu [06:59] pepesan_: seriously it's hard to give better advice than that ... [06:59] Kamion: i know [07:01] Kamion: there is a list of development of the ppc port? [07:01] pepesan_: once you add universe, ppc is pretty close to i386 in terms of package count [07:01] no, we don't have port lists [07:01] development is on ubuntu-devel@ until it accumulates enough traffic to annoy other developers enough to want it split off :) [07:02] the Mac-specific development that needs work is along the lines of support for very new hardware like the G5, support for power management, and figuring out how to make the installer work on oldworld powermacs [07:03] package builds take care of themselves in the vast majority of cases, under the able stewardship of our build daemon maintainer [07:04] yep [07:04] the prinpicipal problem is the way to manage pointers [07:04] bigendian a nd litle endian [07:04] ? [07:05] and the hardware support [07:05] that's only a problem with a tiny minority of very broken packages [07:05] most of those got fixed in Debian years ago, since powerpc is a mature port [07:05] beleave me there are many new packages that doesnt support that [07:06] especialy that kind of packs [07:06] not IME [07:06] there are very espeficic [07:06] occasionally you have to fix an endianness bug or an assumption of unsigned char, but it's relatively rare nowadays [07:06] ok [07:06] s/unsigned/signed/ [07:06] i beleave u :) [07:07] support for very old or very new hardware is more interesting and much harder [07:08] there are any road map for the ppc lice distro? [07:12] once again, the plans aren't powerpc-specific [07:12] ahh [07:12] ok [07:12] the live CD plans are at http://www.ubuntulinux.org/wiki/LiveCD, under "Ongoing development" [07:13] Kamion: did you upload a new ubuntu-meta to go with your seed changes? [07:13] lamont: no, not yet [07:13] shall I do so? [07:15] Kamion: does today's ia64 ISO contain all the fixes for the various problem we've been tracking down? [07:15] we might get an ia64 livecd if we do... [07:15] lamont: I was just thinking about that, not daring to suggest it ;^) [07:16] T-Bone: it's got the elilo fix and the others from earlier [07:16] Kamion: ok fine [07:16] should be ok [07:16] updating ubuntu-meta now [07:16] T-Bone: there are a couple of other uninstallables on ia64, but those should be cleaned up by the time we get u-m inb [07:17] lamont: heh ok === no0tic [~no0tic@host191-110.pool8250.interbusiness.it] has joined #ubuntu-devel [07:19] are there mirror for the daily ISO snapshots? === x4m [~max@228-229.244.81.adsl.skynet.be] has joined #ubuntu-devel [07:19] pepesan_: testing and bug-fixes for the powerpc live CD would be very welcome, though [07:20] Kamion: how hard would it be to put up an (unusable) iso of the livecd that was missing the cloop fs? [07:21] that would be a less painful rsync, and then we could just rsync the fsimage, or something... [07:21] lamont: please kick nautilus-sendto build [07:21] lamont: probably tolerable, but I don't see how it would make life any easier; rsyncing the ISO should be about as hard as rsyncing the filesystem image === justdave [~dave@66.227.241.236.gha.mi.chartermi.net] has joined #ubuntu-devel [07:22] since it's basically just a few bytes around the image [07:23] fabbione, lamont: Accepted pkgstriptranslations 4 (source) [07:23] fabbione, lamont: believe me, you _want_ this version on your buildds [07:23] *cough* [07:23] pitti: do you know if the culmus package got into main finally? I talked with mdz about it last night, don't know the outcome. [07:24] seb128: done [07:24] the system doesn't recognize correctly my cpu freq [07:24] lamont: thanks [07:24] what can I do? [07:24] Kamion: I meant the filesystem, not the image. [07:25] pitti: yeah, and the new sbuild and buildd-config, and.... [07:25] sivang: me neither, I had to leave after the meeting [07:25] pitti: ah ok. [07:25] lamont: oh, I see [07:25] Kamion: that one's highly rsyncable... [07:25] lamont: hm, yeah, could do [07:25] sivang: from an update a few minutes ago, it looks like it's still in 'universe', though I don't have access to the machines themselves to check. [07:26] crimsun: ok, then I have to talk to mdz then :) [07:26] culmus | 0.93-1 | warty/universe | all, source [07:26] culmus | 0.101-1 | hoary/universe | all, source [07:26] thanks, Kamion. [07:26] seb128: you know nautilus-sendto is universe, yes? [07:26] it certainly hasn't been seeded, which would be a requirement [07:26] Kamion: k [07:27] pitti: *grin* [07:28] lamont: yep [07:29] Kamion: Yes, I know, that was dumb... === ogra [~ogra@p508EBB3D.dip.t-dialin.net] has joined #ubuntu-devel [07:29] hey, it happens [07:29] who stole my sound ? [07:29] I managed to kill all the logic to add the initial user to groups, sudoers, and aliases the other day [07:31] ogra: open volume control, change it to the alsa mixer [07:32] and file a bug about it so seb128 will fix it. :-) [07:32] DOH [07:32] hmm, i got no alsa mixer....damned.... [07:32] ogra: I had the same problem [07:32] ah, nm, fount the hidden option [07:33] ooohh.....suddenly i got 4 volume controls in the panel *g* [07:33] Kamion: ok i have a ibook g4 for testing and the region admin have a lot of g5 [07:33] thanks lamont [07:34] seb128: need a bug as reminder ? [07:34] seb128: is that really in your arena, or where is it coming from? [07:34] havoc pennington ? [07:35] lamont: I have no idea on this bug, first time I heard about it [07:36] need details === silbs [~sbsm0084@host217-37-231-28.in-addr.btopenworld.com] has left #ubuntu-devel [] [07:36] ogra: reminder of what ? [07:36] seb128: somewhere in the last couple of weeks or so, upgrading hoary causes things to switch to the oss mixer, and sound goes *poof* [07:36] seb128: ogra's talking about the same bug [07:36] hum [07:36] seb128: to fix it if its your domain [07:36] no idea of that's due to the system or GNOME, fill a bug [07:36] seb128: looks like an upstream chane [07:37] change === maskie [~maskie@196-30-108-36.uudial.uunet.co.za] has joined #ubuntu-devel [07:37] seb128 [07:37] ogra: GNOME side ? do you know how to get it from scratch ? could you try to downgrade ? [07:37] ogra: mostly I was trolling seb128, but it might actually be his to fix (gnome) === lamont has no clue where the bug actually lives. [07:38] seb128: it could also be alsa in todays kernel upgrade [07:38] pitti: ping [07:38] ogra: right, that's why I need details on when it happened (just after a GNOME update ?) etc === ogra looks which upgrades came during the day.... [07:38] pitti: ping [07:38] trulux: pong [07:38] :) [07:38] lamont: ubuntu-meta uploaded, with any luck ubuntu-desktop should be installable on ia64 now [07:39] trulux: pong png [07:39] ogra: no GNOME change today if you upgrade every day [07:39] pitti: ok, time to do the gcc packages [07:39] if it isn't, the problem is not OOo ... :) [07:39] pitti: so, finally what you take is gcc-ssp [07:39] seb128: i do, so it might be a kernel issue [07:39] right? [07:39] Kamion: yeah, checking on the other languishing children now. [07:39] or gcc-hardened? [07:39] lamont: contact-lookup-applet probably still a problem [07:39] oh, which just built [07:39] Kamion: ssshhhh! [07:40] trulux: -ssp [07:40] seb128: must have happened during the day....i rebooted this morning and it worked.....i rebooted 1/2h ago and it changed... [07:40] ia64 had a few give-back's land on the floor [07:40] pitti: ok [07:40] pitti: is that the final decision? [07:40] ogra: ok, not GNOME so === pappy- [~pappy@fsai9.fh-trier.de] has joined #ubuntu-devel [07:40] nope [07:40] ogra: what did you update ? [07:41] i run update-manager .... everything that came since tonight..... [07:41] so it could also be hotplug which got updated .... [07:41] ogra: ok, good luck :p [07:41] I've to go for dinner [07:41] heh [07:41] later [07:42] good morning [07:42] pappy-: hey [07:42] pappy-: pitti is the guy who is working together with us in the deployment of a hardened toolchain === Hwolf [~hidde@136.37.dynamic.phpg.net] has joined #ubuntu-devel [07:43] hi trulux, pappy- [07:43] hey ajmitch [07:44] Hi. Sebastian told me to come here since after the recent upgrades I've got issue's with oss/alsa. [07:44] pitti: nice to meet you, Sir. [07:44] pitti: me and ajmitch are working on SELinux stuff, we are trying to form a work team [07:44] woot. jabber came back [07:45] pitti: pappy- is the man who you should ask for toolchain (gcc and glibc) issues [07:45] pitti: he is one of the best guys that can help you with the work [07:46] pitti: I will take the kernel work, so, ask me all about kernel-related issues [07:46] i will start filling the bugzilla with my junk ;) [07:46] lamont: ok, in that case I think you'll be good to go in an hour or so [07:47] Is there anyone here knowledgeable about alsa/oss? === Kamion wonders how many people will actually *use* an ia64 live CD ... [07:47] lamont: hi [07:47] Hwolf: what do you need? [07:48] Kamion: 1: lamont :) [07:48] Kamion: give me the HW, then i'll use it ;p [07:48] T-Bone: true :) [07:48] hehe [07:48] T-Bone: you better use it too... [07:49] lamont: gaah. Nah. I do install my boxes you know. Especially when they are _poweful_ ones ;) [07:49] actually, once the livecd works, I plan to drop a handful of them on the linux gang in hp ftc [07:49] crimsun, After the recent updates my audigy card doesn't show up in applications / sound & video / Volume control anymore. However, it does work. === Kamion goes for dinner with parents [07:49] crimsun: I think it is using oss rather than alsa [07:49] lamont: actually i think that a 'server-oriented' ia64 liveCD would prove useful [07:50] Hwolf: the contents of /proc/asound/cards should list your audigy [07:50] (from a technical and 'marketing' PoV, that is) [07:50] Hwolf: look in File->Change Device (german here, might be called different) [07:50] Hwolf: next, check lsof /dev/dsp* ; lsof /dev/snd/* [07:50] Hwolf: in the volume control.... [07:51] T-Bone: yeah - in it's raw form, I don't expect many to use it. But as a proof of concept for what you can do with ia64 and a bootable CD, .... [07:51] pitti: have a time? [07:51] lamont: yeah, exactly [07:52] crimsun: root@system:/home/hidde # lsof /dev/dsp [07:52] COMMAND PID USER FD TYPE DEVICE SIZE NODE NAME [07:52] esd 9488 hidde 5w CHR 14,3 7474 /dev/dsp [07:52] and it's a good thing to have when all you can find are hp-ux ia64 boxen... [07:52] Hwolf: that seems fine. Followed ogra's tip? [07:53] lamont: heh ;) [07:53] pitti: hey, are you there? === lamont will go into town tomorrow and download a couple of livecd's. [07:53] pappy-: nice to meet you; trulux, thanks for the introduction [07:53] lamont: as a matter of fact, I wouldn't have many friends owners of ia64 to make switch to Ubuntu :) [07:53] crimsun: there is a bug in the selection of the default mixer it seems, i just try to find out which package it introduced in todays update... [07:53] T-Bone: do we know anything about scsi vs ide? [07:54] lamont: not really. Thierry installed hoary on his zx2000 SCSI box [07:54] pappy-: no problemo, now it's your time, sure pappy will help all of us to make this going in the right way (tm) [07:54] and didn't hit the anna issue either [07:54] hi pitti [07:54] lamont: i have to test it on my zx6000. Will do that quickly in a few minutes [07:55] (burning the cd atm) [07:55] crimsun: it shows up now [07:55] T-Bone: kewl [07:55] t-bone, lamont [07:55] this sounds like a debian-hppa reunion here [07:55] pappy-: heh, always the same faces everywhere ;) [07:56] Hwolf: right, then we need to pinpoint which gnome package introduced it. I'll check aptitude logs in a sec. [07:56] crimsun: There is something quite wrong with that. [07:56] pappy: [07:56] tail -2 quinn-diff/list.stage2.hppa [07:56] Total 779 package(s) in state Installed. [07:56] Total 828 package(s) [07:56] crimsun: no gnome update today.... === T-Bone curses lamont for doing all the stuff by himself ;) [07:56] I was running rythmbox / audigy / alsa, opened gnome volume-manager and now I've got no sound [07:56] crimsun: i suspect linux-image or hotplug [07:56] lamont: :-) [07:56] Kamion: you still wondering about that xslt thing you asked about earlier? [07:56] T-Bone: was christmas break, and you were walk-about or some such... [07:56] you guys rock [07:57] lamont: fair enough. But I had warty ready before that, remember ? :) [07:57] pappy: apt-get source linux-meta [07:57] T-Bone: yeah, well.. [07:57] trulux: sure, I'm here. Just a bit busy... [07:57] and 2.6.10-10 kernel has hppa in it as well. === lamont is saving d-i for t-bone [07:57] lamont: lol, damn you! ;) === T-Bone is always doing the dirty parts, and only gets a rice-bowl a day for it ;) [07:58] T-Bone: just don't try to use a gscps2 mouse with that kernel, apparently [07:58] T-Bone: but it's good sticky rice, not that bara-bara crap. [07:58] lamont: doh? A kernel bug? Have to fix that :P [07:59] T-Bone: fixed in cvs [07:59] lamont: lol true ;) [07:59] lamont: ah ok [07:59] crimsun: opening volume control for me stops rythmbox from playing. I need to switch volume-control to audigy before it starts playing [07:59] in 2.6.11-rc1, which is why we don't have it... [07:59] roger that [07:59] ajmitch: Hi! [08:01] crimsun: I can't get my line-in working (from tv-card to audigy) But tv-card is happily outputting the sound to my line-out, and audigy line-in is open [08:01] geez these 4X RW are slow shit [08:01] pitti: why hoary uses PAM 0.76? [08:03] Hwolf: so you have two cards? [08:03] pitti: do you employ the default-disabled SSP patch in your default gcc packages [08:03] crimsun: tv-card and sb-audigy [08:03] pitti: afaik it has no negative side impacts i know of. [08:03] Hwolf: both of which use alsa modules according to /proc/asound/cards? [08:03] Hwolf: we should remove this to #ubuntu [08:03] (other than that it does not work on -hppa that is) [08:04] mdz: ping [08:04] trulux: hmm, what else? [08:04] pitti: sid and Hoary are using PAM 0.76 [08:05] should use at least 0.77 [08:05] pappy-: it's in the source package, but not activated === decko is going home... [08:07] pitti: that did not answer my question. [08:07] pitti: pong [08:08] pitti: when i apt-get install gcc, can i do gcc -fstack-protector-all, yes or no? [08:08] as a user? [08:08] pappy-: pappy-: it's in the source package, but not activated [08:08] mdz: culmus is still in universe, did you approve moving it to main and it awaits seeding? [08:08] pappy-: -> no [08:08] pitti: aha. [08:08] pappy-: it is not built [08:08] pitti: same as Debian then. [08:08] sivang: yes [08:08] pitti: intention to change it? [08:08] mdz: ok, thank you! [08:08] sivang: I'll add it now [08:08] pappy-: that's the reason why we want completely separated source packages in universe [08:08] pappy-: my patches are the one included [08:08] pitti: sorry for misunderstanding you, but with "activated" i refer to "on by default building SSP" [08:09] we have still the ability f changing the ssp code by the libssp [08:09] trulux: your libssp patches? [08:09] mdz: we are ready to upload, we only need a final ack about suppressing "Accepted" mails to *-changes for language-packs [08:09] pappy-: no, I meant that the patch is not applied; sorry [08:09] elmo: mdz is here, final ack :-) [08:09] pitti: okay, thanks anyway for making it clear to me. [08:10] pappy-: the problem is that we cannot just enable it and use it [08:10] pappy-: the libssp support patches [08:10] pitti: i do can. [08:10] pappy-: if it breaks anything, our release would be screwed [08:10] pitti: yes, i understand [08:10] pappy-: so we want some independent test packages in universe to play around with [08:10] pitti: Debian Hardened that is. [08:11] pitti: we will try to steal as much hoary as we can and test-drive our work on your stuff, if you like. [08:11] pappy-: theoretically it would even possible to automatically build a parallel ubuntu-ssp archive [08:11] heh, pappy-. [08:11] ha tseng [08:11] pitti: yes. [08:11] pitti: If this stuff went into main, it'd be for the release post-hoary? [08:11] pitti: trulux will coordinate that. [08:11] elmo: I meant, "to give final ack", not "he gave" [08:11] pappy- is right [08:11] ajmitch: yes, post-hoary in any case [08:12] :) [08:12] ajmitch: but it is important to get it working in the hoary timeframe [08:12] ajmitch: so that activating ssp would be one of the very first things we do [08:12] ok [08:12] ajmitch: (for hoary+1) [08:13] since the release process for hoary+1 will start straight away, and you'd need to recompile everything with the -ssp compiler [08:13] true? [08:13] pitti: Debian Hardened will build default PIE and SSP at as much packages as possible, being independent from RSBAC (which distinguishes it from Adamantix at the moment) [08:13] ajmitch: yes [08:13] pitti: i think you know the tricks i am playing on toolchains since some years. [08:13] ajmitch: but primarily to follow the "early breakage" principle [08:13] pitti: I'm doing SELinux work for Hoary, so, it could be added in the first release [08:14] just packages updates and fixes === ajmitch checks... I think libselinux is in main already === Nafallo [~nafallo@c-14ff70d5.038-1-65736b2.cust.bredbandsbolaget.se] has joined #ubuntu-devel [08:16] pitti: would you propose me for being an official Ubuntu dev? ;D [08:16] ajmitch: yes, but not the last revision [08:17] lamont: booting the zx6000 [08:17] trulux: that doesn't matter, getting it in there in the first place is what's needed :) [08:17] trulux: if you prepare and maintain packages, yes, why not [08:17] trulux, ajmitch: I'm off for a bit (about an hour) [08:18] pitti: ok, i'm going to make my first "stable" ones for Hoary :) [08:19] ok pitti, see you later [08:19] Kamion: something is screwed with the dating script. The iso claims to be "built on 20041227ubuntu6" === ajmitch should ask for inclusion as well soon :) [08:19] pitti: suppressing emails? [08:19] pitti: if you think it is appropriate, I don't see a problem [08:20] do you plan to update the packs every day? [08:21] ajmitch: have you sent the bug to the bugzilla? [08:21] trulux: nope, only just installed latest kernel to check [08:21] lamont: i'm up to partitionning HD, no anna-bug [08:21] lamont: but the internal SCSI controller isn't detected, it can't find any disk [08:22] mdz: well, would you like to get 246 accepted emails initially and then a bunch of them every day? [08:22] mdz: daily update is not really possible right now [08:22] mdz: since Rosetta does not yet support this [08:22] mdz: but eventually yes [08:22] mdz: I still get REJECTED and NEW messages, so I know about problems [08:22] Kamion: gah, casper is missing a depnedency on md-modules [08:23] pitti: whatever you and elmo agree on is fine with me; don't let me be a bottleneck === T-Bone lacks lspci on the installer to find out which driver to use ;P [08:25] trulux: bug filed [08:25] T-Bone: /proc/bus/pci? === martink [~martin@pD9EB20D7.dip0.t-ipconnect.de] has joined #ubuntu-devel [08:26] mdz: i walking through, but it's far less practical [08:26] ugh I dislike bugzilla :) [08:26] I thought there was a udeb with it now === mxpxpod [~bryan@208.252.112.24] has joined #ubuntu-devel [08:26] hey mdz [08:27] ajmitch: great === T-Bone can't find relevant info in /proc/bus/pci, looks in /sys === ogra wonders if either todays hotplug or the linux-source upgrade killed his sound setup [08:27] ajmitch: bug number? [08:27] trulux: almost filed, I should say.. [08:27] fabbione: good morning [08:27] it complained, I have to go back [08:27] ajmitch: okay dockey [08:27] fabbione: hey [08:27] mdz: if you have time could you take a look at ubuntu-meta for sparc? i know several bits need modification [08:27] hey trulux [08:28] T-Bone: i fixed that USB keyboard config thingy on ia64 [08:28] hi fabbione [08:28] mdz: the repo is at sparc.u.c === fabbione waves [08:28] fabbione: you have a patch that you want me to review, or you need me to do the changes? [08:28] fabbione: the former is much faster ;-) [08:28] mdz: sorry i don't have a patch [08:28] but Kamion was talking about germinate/seeds ubuntu-meta [08:28] fabbione: has it been integrated into todays iso? [08:29] T-Bone: no. it is in 2.6.10-10 that i uploaded a few hours ago [08:29] mdz: there's no way I can't find what PCI device i have, and i can't even look into 'load installer component from the CD', because d-i won't let me load anything, though the failing step comes _after_ loading components [08:29] so i believe it will be in tomorrows cd [08:29] fabbione: ok [08:29] mdz: that was actually the last outstanding question. Now I'm ready to upload :-) [08:30] s/I can't/I can/ [08:30] pitti: fantastic [08:30] Can anyone here help me find out why I since a recent upgrade I can't use my line-in anymore? [08:30] ok.. i really need a shower and cook some dinner [08:30] only 3 hours to pain the living room is a record :) [08:31] mdz: anyway there is NO rush for it [08:31] mdz: i still lack a few packages in main (outdated) to be 100% synced === fabbione & [08:32] fabbione, mdz: who of you broke my sound ? [08:32] not me [08:32] I don't break things [08:32] there were only two sound related updates today for me.... [08:33] ogra: talk with seb128 :-) [08:33] hotplug and linux-image [08:33] ogra: blame hotplug [08:33] heh, fabbione i already did [08:33] lamont was talkign about a similar problem [08:33] isnt it policy that he is always the first to bug ? [08:33] that for some reason the default in the mixer was changes to oss [08:33] yup, i know [08:33] ok [08:34] ogra: you can blame everything other than the kernel... [08:34] and before the kernel your crappy hardware [08:34] ;) [08:34] ok i need to go really [08:34] i am starving [08:34] mdz: are these okay for main? [08:34] i think thats ok....but the oss mixer doesnt affect anything anymore, so it also could be a module thing [08:34] fabbione: missing driver in ia64 conf, as it seems [08:34] cya tomorrow guys [08:34] ogra: do you have 2 cards? [08:34] fabbione: you're gonna have to roll up new kernels again ;) [08:34] elmo: language packs? yes [08:34] T-Bone: send me updated configs [08:34] pitti: please seed them [08:34] fabbione: will do [08:34] crimsun: nope [08:34] T-Bone: cool [08:34] byr [08:35] crimsun: laptop with via chip [08:35] ciao fabbione [08:35] fabbione: filed bug for the selinux & audit stuff for you [08:35] elmo: speaking of main, #4750 === bob2_ [rob@bob2.user] has joined #ubuntu-devel [08:35] elmo: only the metapackages are seeded, so I think you just need to sync with germinate [08:36] crimsun: it seems like the oss mixer stuff stopped working completely [08:36] lamont: could you retry nautilus-sendto again on ppc/ia64 ? It didn't pick libebook1.2-dev 1.1.3-0ubuntu9 previous time [08:36] mdz: oh, before uploading them? or after? does it matter? [08:36] pitti: before elmo processes them [08:36] crimsun: I've got my system working, however if I open gnome-volume-control the playing audio stops [08:36] mdz: ok [08:39] Kamion: why is there no 'british english' layout on powerpc or amd64 in d-i? === pepesan [~pepesan@40.Red-80-33-116.pooles.rima-tde.net] has joined #ubuntu-devel [08:39] Kamion: it breaks my finger macros [08:39] (at least, if it's there, it isn't at the same point in the list) [08:40] ogra: does linux-image-$(uname -r) 2.6.10-8 work correctly? [08:41] mdz: seed to supported for now? (and ship later, depending on how many languages we support?) === T-Bone test-installs zx2000 now [08:41] crimsun: hmm, let me try, i have only -5 installed ... [08:41] ack [08:42] seb128: done [08:42] I assumed he was running -9 [08:43] lamont: thanks [08:44] pitti: yes [08:46] pitti: will it be possible to include another package as a depend of the language pack pacage, pulling in conffiles for GNOME kbd chooser to be already configured with the respective pack's language? [08:48] lamont, Kamion: more breakage: unable to initrd-tools [08:48] because of libunwind7 [08:49] elmo: please rm -rf ~mdz/public_html/ubuntu-live on little (it has some files owned by alex) [08:49] it's trying to install libunwind7-dev which depends on libc6.1-dev [08:49] Kamion fixed that [08:49] after today's cd image? [08:49] Jan 19 08:55:22 lamont: I took libunwind7-dev out, yes [08:49] Jan 19 08:55:40 debootstrap would have failed since libc6.1-dev [08:49] isn't in required (nor should it be) [08:50] looks like it, yes [08:50] lamont: yup [08:50] lamont: or so i hope. It's labelled 'built on 20041227ubuntu6' === lamont thinks he was confused by the fact that he was messing with hoary.buildd, which does require libunwind7-dev (since it's essential + build-essential) [08:50] lamont: same player shot again, heh ;) === ogra [~ogra@p508EBB3D.dip.t-dialin.net] has joined #ubuntu-devel [08:52] ogra: sorry, I presumed you were using -9 [08:52] hmm...its not the kernel [08:52] mdz: done [08:53] ogra: needed you to back up to -8 if you were using -9. Are you using snd-via82xx? [08:53] crimsun: i have only -5 for testing [08:53] lamont: to make things better, the kernel in panicing on halt [08:53] yup === jaco [~jaco@host42-20.pool80117.interbusiness.it] has joined #ubuntu-devel [08:54] crimsun: my currently running kernel is -9 ... [08:54] T-Bone: it halted. :-) [08:54] lamont: no [08:55] had to unplug [08:55] 0 upgraded, 611 newly installed, 0 to remove and 0 not upgraded. [08:55] Need to get 430MB of archives. [08:55] that's a "go" for ubuntu-desktop/ubuntu-base installability on ia64. [08:55] thanks Kamion [08:55] lamont: oh btw, don't make ubuntu-dekstop depend on ooffice on ia64, or else it will never install [08:56] T-Bone: that was what Kamion fixed. [08:56] ogra: so volume control only now began defaulting to oss? [08:56] lamont: how long before I can expect casper 0.19 binaries (arch all) in the archive so that I can do a new live CD build? [08:56] and ooo2 should fix that the other way [08:56] lamont: ah ok cool [08:56] mdz: when did you upload? [08:57] -rw-rw---- 1 mdz mdz 302 2005-01-19 11:24 casper_0.19_source.upload [08:57] PST [08:57] ~35 minutes ago [08:57] crimsun: no, it keeps the setting you select....my defaut mixer was oss al the time, since i didnt touch the mixer since the warty upgrade of this lappie [08:57] so the sources should have snapped at 11:33, and binaries should be there at 12:03 [08:58] sources aren't there (on archive.u.c anyway) [08:58] crimsun: so it looks like the oss mixer doesnt work anymore but the unselected alsa works fine.... [08:58] mdz: and it's uploaded, so it built just fine for you [08:58] lamont: ok, thanks [08:58] crimsun: i'll downgrade hotplug for a test.... [08:58] lamont: can you document the time schedule on a wiki page and link it from DeveloperResources? [08:59] lamont: also I think the build logs byDate view (and any others you might have besides the raw index view) should probably be documented there [08:59] mdz: we'll do [09:00] right - BuildDaemonInfo or some such, eh? [09:00] Is there any reason for synaptic holding off on upgrading libslt1-python2.4? === Nafallo [~nafallo@c-14ff70d5.038-1-65736b2.cust.bredbandsbolaget.se] has left #ubuntu-devel [] [09:02] lamont: or ArchiveProcessingSchedule? [09:02] libxslt1-python2.4? [09:03] mdz: that doesn't quite describe the byDate stuff though... [09:03] Hwolf: might be because it was renamed to python2.4-libxslt1 [09:04] lamont: whatever you think is best, then [09:05] mvo_ Can I safely replace it with the new version? === T-Bone is exhausted, gives up for tonight [09:05] see yall [09:06] Hwolf: see the description of the libxslt1-python2.4 package :) === T-Bone [varenet@T-Bone.developer.debian] has left #ubuntu-devel ["enough] [09:07] Hm. There is openoffice2-debian in universe, but openoffice2 itself isn't [09:07] Hwolf: I can't reproduce the sound halt while opening gnome-volume-control. [09:08] crimsun: Anything I can do? [09:08] Hwolf: which device is gnome-volume-control set to manipulate by default (when you open it)? [09:08] Audigy-alsa [09:09] I get a soundhalt. Closing the program and opening it again causes sound to resume. [09:10] -2 [09:10] 2.6.10-2 is the latest I can see [09:11] kernel 2.6.10-2-386 === mvo__ [~egon@suprimo-162.ping.de] has joined #ubuntu-devel [09:12] Hwolf: alsa-base is 1.0.8-1, and the latest for 386 is 2.6.10-9 [09:12] crimsun: not according to my hoary :-S [09:12] Just ran an upgrade [09:12] did you update first? [09:13] Ofcourse [09:13] Got it from my grub menu, so I'm running -2 [09:13] are you using apt-proxy? [09:13] No [09:13] paste me `apt-cache policy linux-image-2.6.10-2-386' [09:14] hidde@system:~ $ apt-cache policy linux-image-2.6.10-2-386 [09:14] linux-image-2.6.10-2-386: [09:14] Installed: 2.6.10-9 [09:14] Candidate: 2.6.10-9 [09:14] Version Table: [09:14] *** 2.6.10-9 0 [09:14] 500 http://archive.ubuntu.com hoary/main Packages [09:14] 100 /var/lib/dpkg/status [09:14] hidde@system:~ $ [09:14] Hwolf: well, that was supposed to go to me in query, since this isn't directly related to ubuntu development. [09:15] excusez-moi === whiprush [~jorge@arslinux.com] has joined #ubuntu-devel === ogra [~ogra@p508EBB3D.dip.t-dialin.net] has joined #ubuntu-devel [09:18] lamont: hmm, :18 now and still no casper [09:18] crimsun, nothing... no idea what it is ( [09:19] ogra: troubleshooting w/ Hwolf, too === Jamie_Bennett [~jamie@81-179-207-85.dsl.pipex.com] has joined #ubuntu-devel [09:20] daniels: erm.... [09:21] 8879 root 5 -10 836m 583m 6056 S 0.3 57.6 20:04.41 Xorg [09:22] lifeless: there's a Debian bug about X sucking up huge amounts of memory [09:22] lifeless: what sort of apps are you running? [09:23] overnight, 2* gvim, evo 2.2, 1 uxterm running screen, and xscreensaver was running the lightning screen saver === lamont points mdz at the other window [09:23] mdz: it was ok before I went to bed. [09:24] lifeless: it would be nice if you could confirm that #5412 (aptitude wants to remove the kernel) is solved with my latest aptitude upload (if possible) [09:25] haggai: here ? [09:25] mvo__: erm, next kernel upgrade, sure. [09:26] Kamion: little mirrors from mirnyy, right? and updates its mirror as part of the CD build process? [09:27] lamont: ran anonftpsync on little, it's there now [09:32] ogra: I'll need to continue in 4-5 hours, meeting now. I'll forward you anything I uncover. [09:32] thanks :) === jaco [~jaco@host42-20.pool80117.interbusiness.it] has joined #ubuntu-devel === mvo_ [~egon@ip181.135.1511I-CUD12K-01.ish.de] has joined #ubuntu-devel [09:39] mdz: anything else you need before I nuke this sucker ? === Jamie_Bennett [~jamie@81-179-207-85.dsl.pipex.com] has left #ubuntu-devel [] [09:49] err, if I drag a folder into CD/DVD creator it is _copying_ it right? [09:49] and not symlinking it or something equally retarded [09:50] elmo: i think its something like symlinking until it runs the mkisofs in the background....copying would consume a lot of diskspace === Hwolf [~hidde@136.37.dynamic.phpg.net] has joined #ubuntu-devel [09:51] lifeless: daniels is the one to talk to about it [09:51] elmo: I think it's just remembering the name [09:51] elmo:(and probably a lot of time too if you copy big files between partitions) [09:51] elmo: there's a feature request to have it optionally copy, e.g. to enable copying CDs with a single drive simply === lamont realizes that he needs to run to town for a bit... back in a while [10:02] pitti: libselinux 1.20 for Ubuntu is almost ready [10:09] ajmitch: ping === smurfix_ [~smurf@run.smurf.noris.de] has joined #ubuntu-devel === smurfix grumbles. Bad power supply, keeps shutting off :-( [10:16] smurfix, have you checked the temperature of the cpu/motherboard? my amd system shuts of time to time due to heat :( Bad cpu-fan i guess.. === OddAbe19 [~OddAbe19@pcp02542642pcs.lncstr01.pa.comcast.net] has joined #ubuntu-devel [10:20] trulux: back [10:20] Kamion: can I remove any of the casper symlinks yet? [10:21] trulux: what changes did you have to make from the sid version? I just recompiled 1.20.. [10:21] elmo: I'm still curious as to why it broke, but I think it's some complicated germinate bug to do with versioned dependencies on virtual packages [10:21] elmo: I just uploaded libxslt to avoid the problem, since I needed it to work fairly urgently [10:21] Kamion: I need to do another upload anyway [10:21] ajmitch: me too, added extra .ubunut banner to version [10:21] mdz: yes, go ahead [10:21] trulux: hmm, why? :) [10:21] we would need to add our specifical dependencies on it [10:21] ok [10:21] mdz: little mirrors from mirnyy, yes, and anonftpsync is the first thing it does [10:22] ajmitch: to follow standard Ubuntu versioning style [10:22] excellent, thanks [10:22] trulux: it's not .ubuntu though, but ubuntu, AFAIK [10:22] check with others here - I thought ubuntu tags on versions was when stuff was patched, not just recompiled [10:22] mdz: I always use British English keyboard layouts, it's there [10:22] hey azeem [10:22] ajmitch: dude [10:22] azeem: ok, rebuilding [10:23] Kamion: on i386 for me, it goes us, british, dvorak [10:23] Kamion: but on amd64 at least it's us, dvorak [10:23] weirdness, I'd certainly notice immediately if it were missing [10:23] trulux: uhm, I'm no pro on this. That was just a suggestion to check back with the doku [10:23] might be something to do with keyboard type [10:23] (at/usb/etc. differ) [10:23] azeem: NP, this doesn't take a life to rebuild [10:23] there's only one keyboard type on powerpc though, it's all considered to be usb === ajmitch wonders if he should leave a stack of ubuntu cds in the CS lab at uni [10:24] mdz: hmm probably late 20050119.1 casper cant load ext2, dm_mod [10:24] ajmitch: :) [10:24] yeah, it takes about 30 seconds.. [10:24] ajmitch: when we make a source modification to 1.0-1, it becomes 1.0-1ubuntu1 [10:24] Kamion: I bet it's only missing when you select location 'united states' [10:24] Kamion: great, that's what I thought [10:24] trulux: so it should be safe to leave it at 1.20-1 [10:25] ajmitch: you never just recompile without changing the version number (ahem, *shush* those who remember the great archive flush), so it'd have to be something like that [10:25] Kamion: which, come to think of it, it should do consistently on all architectures [10:25] ajmitch: NOOOOO [10:25] Kamion: hmmph [10:25] mdz: I find that implausible, knowing the kbd-chooser code [10:25] ajmitch: dont make the same error i did ;) [10:25] Kamion: source change, or recompile, which is it? :) [10:25] Kamion: I have an i386 to my left, and an amd64 to my right, and the keyboard chooser menu is different [10:25] Kamion: I can take a photo if you like :-) [10:25] ajmitch: you must not just recompile and leave the version number as it is. It will confuse the world. [10:25] ajmitch: change is it [10:25] mdz: that might be useful actually [10:25] ogra: don't worry, I don't have any upload rights yet :) [10:25] ok, coming up [10:26] ajmitch: shouldn't you be changing build-dependencies anyway? that's a source modification [10:26] ajmitch: this might be interesting: http://www.ubuntulinux.org/wiki/Uploads/ [10:26] ajmitch: i'm only three in advance of you ;) [10:26] Kamion: it has no build-depends, really.. [10:26] ajmitch: also for the real archive I don't think we should do a big rebuild effort; instead, just change the buildds and let things gradually get recompiled as they're moved over anyway [10:26] apart from the usual build-essential [10:27] ajmitch: oh, you mean this is a Debian->Ubuntu port? [10:27] gah [10:27] yes [10:27] passwd is asking all the questions [10:27] ajmitch: does it build from scratch in Ubuntu with no modifications? [10:27] of a package that we would love to have in base [10:27] full name, username, password, password-again at least [10:27] yes, 1.18 is in ubuntu [10:27] 1.20 is in sid [10:27] ajmitch: so why not just request a sync? [10:28] that would be easiest, we're still waiting for manoj's new version to get into sid, iirc [10:28] ajmitch: from Debian->Ubuntu that's two separate archives so you're allowed to recompile without changing the version number; I thought you were talking about recompiling a package already in the Ubuntu archive [10:28] daniels: ping [10:29] we've got a few that will definitely need version number changes [10:30] since they'll need patches, which possibly won't be in hoary [10:30] depends on how much of a window to get the userland patches for selinux in [10:30] you'd want to ask mdz/jdub/some-master-of-the-universe for permission to sync newer versions [10:31] and for packages that are in main? [10:31] we're in upstream version freeze so that would be on a case-by-case basis and had better have a good reason [10:31] ajmitch: UVF ..... you will need a real good reason [10:32] that's what I thought.. [10:32] not to say that exceptions can't be made if there is a good reason, but that isn't my call [10:32] most of the packages we have to patch are in main, so we'll probably work on a separate archive of patched/upgraded packages until post-hoary [10:32] trulux: what do you think? [10:33] ajmitch: I think we could keep ubuntu banner on version strings, as it doesn't harm and it allows you to track the package [10:34] not for network daemons [10:34] that hurts security issues [10:34] trulux: I mean having a separate apt repository for now [10:34] pappy-: right [10:35] have it looking as vanilla as it can be [10:35] prevents information leaking through the network sockets [10:35] makes it harder for attacks [10:35] ajmitch: what are you working on? [10:35] locally i don't care personally [10:35] ajmitch: Hardened Debian separates reps, we have one for sid, sarge, woody and hoary [10:35] pappy-: I think trulux means just the debian revision, right? [10:35] because i can do fingerprinting on all of your binaries when i am a user on a local machina. [10:35] ajmitch: i hope so. [10:36] right [10:36] azeem: selinux integration with trulux & others :) === pappy- is talking about network service identification. [10:36] ah [10:36] yep [10:36] pappy- is completely right, we can't call hardened daemons "-hardened": Oh, this is running apache-1.3.26-hardened [10:36] pappy-: I'm totally unconvinced by that argument, always have been [10:37] best to leave the kids playing around and breaking their heads before they suddenly discover that it's a hardened package [10:37] "OMG, they run -hardened, off to the next IP" [10:37] Kamion: i don't care if you are unconvinced. [10:37] azeem: right [10:37] pappy-: in practice people just scatterbomb attacks without bothering to check the version strings first, and including version strings allows network admins to do auditing of machines [10:37] pappy-: that's nice of you [10:37] azeem: what are you up to? [10:37] ajmitch: lurking [10:37] :) === ajmitch is good at that === twisted_steel [~twisted_s@pool-129-44-6-15.alb.east.verizon.net] has joined #ubuntu-devel [10:37] Kamion: so you know about "practice" and "scatterbombing" [10:37] hehe [10:38] Kamion: show me figures. [10:38] and show me best before dates of zeroday exploits. [10:38] which make it through the internet "scatterbombing" === ajmitch can see this discussion going nowhere [10:38] Kamion: show me the money [10:38] show me a case where people bothered to check the version strings? come on === trulux too [10:38] script kiddies never do === pappy- gives in to your superior train of argumentation [10:39] Kamion: but their tools do [10:39] I don't think yours is any better [10:39] Kamion: aha, in the process of taking the photos I noticed the _other_ difference :-) [10:39] Kamion: i386 is using a PS/2 keyboard, amd64 and powerpc are using USB [10:39] Kamion: i could care less. i just leave it up to my users to decide. [10:39] so no british english layout for USB? [10:39] Kamion: i was a script kiddie many time ago, so, maybe i could get some fresh information on this thread [10:39] trulux: ok, so can we upload some of this modified stuff to the hoary repository on debian-hardened then? :) === trulux reveals his obscure past [10:39] quick hoary question: am I the only one who can't enter text in the address book applet? [10:39] ajmitch: right [10:39] trulux: l33t! [10:39] face it, if you have a network service with a security hole, it'll get exploited whether the version number exposes the fact or not [10:39] pappy-: w00t! [10:39] trulux: can I get access? :) [10:39] Kamion: we are talking about exposing -hardened being a feature on the machine or not. [10:40] Kamion: it is silly to "show off" with it. [10:40] Kamion: maybe i don't get it though. [10:40] mdz: sounds possible if strange; that would certainly be a more relevant distinction [10:40] ajmitch: It's not my box, but i could try to open access for you [10:40] Kamion: maybe i am too stupid. === pappy- baits Kamion [10:40] ok [10:40] *g* [10:40] ajmitch: we need to get mcp for install decent repositories [10:41] because I'm using the boxes where I'm a sysadmin to host some stuff [10:41] they are all using hardened debian stuff [10:41] pappy-: here's a use case: the network admins at the university I used to attend routinely do friendly probing of services throughout the network. Being able to detect vulnerable machines by banner probing allows them to notify local admins of problems often before the kiddies get around to it, which reduces the exposure of the network as a hole. [10:41] er, whole. [10:41] guys did you consider using -honeypot instead of -hardened (i think it has the same effect) [10:41] Kamion: i agree with you [10:42] ogra: good point [10:42] Kamion: but we were up to discussing "showing off l33t security improvements" via changing banners here. [10:42] ogra: -rootable sounds better [10:42] so, if your version string lies about what the binary is, this compromises that goal. [10:42] heh [10:42] Kamion: i often also deliberately change my banners to weak strings for honeypots [10:42] Kamion: compromising the goals of lazy admins is what i like. instead of doing it properly. [10:42] centralized revision control on the servers. [10:43] personally I'm all up for having complete package version strings in the banners, but I'm sure I lost that argument a while back [10:43] openssh does this though, and it's useful [10:43] Kamion: have all the banners you like [10:43] well, in an abbreviated form due to crap ssh clients [10:43] Kamion: i will not support it and i will suppress it whereever i can and force it back to default looking style. [10:43] i want to make it look like a virgin daemon. [10:43] Kamion: is there any way to override the cdebconf locks so that I can run two instances? [10:44] its not like some kind of obfuscation. [10:44] it is just not granting attackers an unwanted (and preventable) advantage. [10:44] nothing less. [10:44] pappy-: I'm sure; however I'll do what I think is best for my packages, such as openssh ... [10:45] Kamion: i don't think you will add -hardened to an openssh package that gets compiled by a hardened compiler. [10:45] no but the version string will reveal it to one who knows the history of the package [10:45] And if you did, i would be able to reverse it for my openssh packages. [10:45] So if we all keep our packages nice and handy. [10:46] the user base will decide in the end. [10:46] for somebody coming into an existing community and trying to get them to accept changes, you're taking an awfully aggressive tone, mister [10:46] Kamion: do you think so, Mister? [10:47] Kamion: i did not come here, i was invited. [10:47] you won't have to accept my changes. [10:47] (somebody now tries to be quiet) [10:47] we just take the best out of your development works and put our stuff into it. [10:47] whether you like it or not, i am better than you, Kamion. [10:47] Kamion: pappy- is a friend of mine, often he is acid, but he knows pretty well what he talks about [10:48] pappy-: how old are you? [10:48] anyway, don't fight please [10:48] azeem: not old enough. [10:48] haha, this is classic [10:48] trulux: we are just getting warm with each other, thats the usual practice. [10:48] pappy-: there is a code of conduct lying on this channel, did you read it? [10:49] azeem: should i? [10:49] yup [10:49] if you want to participate and stay, I'd say yes [10:49] azeem: does it state that i have to shut up? [10:49] pappy-: RTFM [10:49] azeem: i do not recall starting this "one" here. [10:49] you may call it "issue". [10:49] http://www.ubuntulinux.org/community/conduct [10:49] and i really tend to be bored by stuff like that. [10:49] pappy-: stop please, it's late, no point for make an ass kicking party [10:49] trulux: am i? [10:50] trulux: you know me and you know when i reached temperature. so far i am very very cool. [10:50] still, daniels is the coolest guy on the block [10:50] trulux: what if this guy treats me like a jackass and thinks he can make something out of it? putting it bluntly: i won't go down that road in here. [10:50] pappy-: ... don't start such type of conversation again, you know i disliked it when it happened in hard-tch list.... === azeem wanders off [10:51] pappy-: if you don't like Kamion or anyone else, ignore him === pappy- seen enough ubuntu for tonight. === pappy- [~pappy@fsai9.fh-trier.de] has left #ubuntu-devel ["have] [10:51] oops [10:51] fh-trier.de ? [10:52] ogra: fh-trier.de? === mirak [~mirak@AAubervilliers-152-1-19-192.w83-114.abo.wanadoo.fr] has joined #ubuntu-devel === Kamion suggests not inviting people with known bad attitudes; it just makes life hard for everyone [10:52] sorry, but don't misunderstand pappy, when you work together with him you know how to enjoy it [10:52] that was fun.. [10:52] he is just a bit "acid" [10:53] ajmitch: not really but i'm surpirsed [10:53] haven't thought that pappy would start such a flame here ;D [10:53] tseng: hey [10:54] Kamion: there? [10:54] yes [10:56] Kamion: I just noticed that firmware loading is still broken because the initrd has the old version of hotplug [10:56] Kamion: anything wrong after the little fight? I'm sorry about it, don't be offended, it was just a stupid conversdation typical pof guys working harder in the wrong hours ;) [10:56] Kamion: has the initrd not been rebuilt? [10:56] mdz: hm, I'll check, what's the fixed version? [10:56] if not, I can't remove the casper symlinks yet, no? [10:56] yes, it's been rebuilt ... [10:56] Kamion: ubuntu8 or later [10:57] trulux: not particularly, have just had the same conversation a number of times before and am not particularly impressed with the line of argument, having thought about it in some depth; but he's entitled to his point of view [10:57] I'm just saddened that people take such hard-line approaches to security [10:57] mdz: you should have an initrd.list file on the CD [10:58] in /install [10:58] Kamion: it says ubuntu7 [10:58] hotplug-udeb 0.0.20040329-16ubuntu7 [10:58] hmm, bah [10:58] ubuntu8 was uploaded at ~0100 UTC [10:58] maybe just missed it [10:59] Kamion: i can imagine [10:59] 1763 F Jan 18 To hoary-change ( 46) Accepted debian-installer 20041227ubuntu6 [10:59] 1766 Jan 19 Matt Zimmerman ( 47) Accepted hotplug 0.0.20040329-16ubuntu8 ( [10:59] mdz: did you have time to read my paper? [10:59] mdz: ping lamont for a daily build? [10:59] trulux: URL? [10:59] lamont: ping ^^^ [10:59] mdz: it is in the wiki, i sent it to you i think [11:00] anyway, i will grab for it again , NP [11:00] mdz: I waited for all the symlink changes to be complete and built before uploading the new initrd [11:00] Kamion: ah, ok, so that bit at least is safe [11:01] mdz: https://www.ubuntulinux.org/wiki/HardeningNeededResources [11:02] mdz: if you give me a chance i can try to do half of the work for SELinux deployment in two weeks [11:02] trulux: I can't offer you human resources; we have none to spare [11:02] it just depends on how and when Hoary will get released [11:02] mdz: NP, i will recruit them ASAP [11:02] trulux: you want me to do the other half then? :) [11:03] trulux: so excluding that, you are saying that all you need are machines? [11:03] trulux: I'm sorry if he felt offended, but he did come back extremely dismissively to my initial remark, before I'd even finished typing the explanation [11:03] trulux: you will take care of autobuild infrastructure and whatever else is necessary? [11:03] mdz: there are a few of us working on it [11:03] ajmitch: you will be the one working behind the scenes ;P [11:03] selinux deployment shouldn't need much in the way of changes to packages.. [11:04] trulux: why do I get stuck behind the curtain? :) [11:04] I understood that there was a good bit of work needed in the packaging toolchain [11:04] mdz: right, we need building machines, and we could take care of the rest of things [11:04] ajmitch: lah, just joking [11:04] Keybuk works for us is the dpkg maintainer, he's talked to people before about this in case you weren't aware [11:04] s/is/and is/ === stub [~stub@dsl-246.248.240.220.dsl.comindico.com.au] has joined #ubuntu-devel [11:05] Kamion: yeah, he is not offended, it's difficult to make him feeling like that, don't worry ;) [11:05] Kamion: yes, and manoj is working on hacking dpkg at the moment, iirc [11:05] trulux: heh, should imagine :) [11:05] he's been quite active in selinux work lately [11:05] yeah [11:05] yes, I need to get back to Manoj about openssh [11:06] i must go to bed in 10 mins [11:06] :( [11:06] rjc is helping advise on policy, I think [11:07] trulux: resources> hm, why alpha? we don't support that architecture at the moment [11:07] rjc is one of the best guys to help with that, he has wroten near the 30% of the SELinux policies === Kamion fixes the resources page to be ReST, it looks like it was meant to be that way [11:07] tseng: libdbus-cil [11:07] Kamion: it's not the important one, as seems to be in the last position [11:07] trulux: I don't think it is necessary to try to bootstrap on 5 architectures at once. I think it might be possible to offer an x86 build machine; I will check into it [11:07] tseng: not sure if daniels has fixed it yet or not [11:07] morning all [11:07] mdz: ok, great [11:08] jdub: ubuntu .au meetup going to be in sydney? [11:08] mornign jdub [11:08] seb128: back [11:08] the next conf will be in sydney or canberra [11:08] i will try to talk tomorrow night [11:08] jdub: aha [11:08] now it's bed time [11:08] good night! [11:08] night trulux [11:09] night trulux [11:09] jdub: just got to see if I can scrounge up the $ from the govt :) [11:09] haggai: is /usr/share/applnk needed for KDE ? tvtime has 2 desktop files, one in /usr/share/applications/ and one in /usr/share/applnk/ for KDE according to the debian/rules [11:09] Kamion: hey, db_register seems to have done the trick [11:09] I wonder why that wasn't necessary before [11:09] mdz: you had initial-passwd-udeb loaded by anna, probably [11:10] mdz: that would have loaded the questions [11:10] ahh [11:10] but now nothing was registering the questions so there was nothing for db_set/db_fset to work on [11:10] makes sense [11:11] Kamion: last bit is to change sudoers to use NOPASSWD [11:11] not something that had occurred to me in advance, but debconf-set-selections does a similar trick for preseed [11:11] is there a way to specify a sed address range which means "the line _after_ the one matching this regex?" [11:11] mdz: perhaps a never-asked debconf question in shadow would be better? [11:11] "Require password for sudo?" [11:12] mvo_: update-notifier is segfaulting for me on the live CD [11:12] Kamion: if you're ok with that, sure [11:12] Kamion: casper also currently forces the user password to empty, but I'd like to find a better way for that too [11:13] yeah, not sure how to work that [11:13] group setup seems to work fine [11:13] ajmitch: http://apt.debian-hardened.org/hoary/ [11:14] ajmitch: there you can find the new packages, just finished the uploading and repository updating [11:14] now, bye! [11:14] great, thanks :) [11:14] ajmitch: err, tell pitti about them, so he could make the upload [11:14] bonna noite! [11:14] :) [11:14] gah, so hot here [11:14] 30C today [11:15] seb128: not any more - that was the older KDE standard but current KDE can also use /usr/share/applications [11:15] trulux: we *have* to work on a better method of installing policy - answering 1001 questions is not a good way :) === ogra hands over his -2C to mdz [11:15] mdz: segfaulting :( ? [11:15] mdz: send some over here! [11:16] haggai: ok, so I can drop the /usr/share/applnk one (that makes a double entry in the GNOME menu) ? [11:16] mdz: has California moved to the southern hemisphere recently or something? === Simira gives mdz a -6C [11:16] mvo_: yes, i can send a stack trace if you need one [11:16] mdz: reproduceable? [11:16] Kamion: it stopped raining [11:16] seb128: yup, please do [11:16] mdz: please do [11:16] haggai: what was the change that you made to KDE's menus? [11:16] and now we have record heat instead [11:16] which, mind you, is preferable [11:16] mvo_: 100% [11:17] haggai: nice, thanks [11:17] Riddell: using the Gnome template (looking for filename) [11:17] mdz: I think you lied and you don't live in California at all; you really live somewhere in the sixth circle of Hell [11:17] mdz: it's good that they let you out for holidays though [11:17] mdz: is it on a downloadable image? if so I would like to grab it and have a look [11:17] http://weather.yahoo.com/forecast/USCA1107_c.html?force_units=1 [11:17] Riddell: /etc/menu-methods/stuff [11:17] mvo_: the downloadable image requires a few fixes in order to work [11:17] need to wait for casper 0.21 to build, and then do another live CD build [11:18] mdz: gar, apt still thinks the cache is not authenticated [11:18] average high for january is usually 19C [11:18] mdz: /var/lib/apt/lists/_cdrom_dists_hoary_Release.gpg exists [11:18] haggai: what is the advantage in your change? [11:18] mdz: then just send me the stacktrace please. and ping me when a live-cd is ready [11:18] haggai: lots of OnlyDisplayIn=KDE stuff to be done [11:18] haggai: want me to file bugs for you? [11:19] mdz: however apt-cdrom doesn't appear to have added a _Release.gpg file [11:19] haggai: i might just mail the kubuntu lists when i set them up [11:19] haggai: you definitely want kubuntu and kubuntu-devel? [11:19] hmm, still no oo.o2 build? :-( [11:19] mdz: can I make it do so? [11:19] jdub: kubuntu-user and kubuntu-devel I'd say [11:19] Kamion: hmm, apt-cdrom is weird [11:20] Kamion: I don't think anyone has considered the possibilty of signed CDs as yet [11:20] can somebody consider it real fast? :) [11:20] copying _Release.gpg into place will override it and have it consider the source authenticated [11:20] but it really ought to verify the signature before doing that [11:20] it doesn't seem to copy the Release file even === robtaylor__ [~robtaylor@nat2.camtrial.co.uk] has joined #ubuntu-devel [11:21] weird, I say [11:21] fabbione: ping? [11:21] mdz: so I could work around it in archive-copier? it's blocking Array CD 3 so a quick fix would be ideal [11:22] elmo: around? [11:22] haggai: what's the next blocker for oo.o2? [11:23] Riddell: 1. the files conflict 2. attempt to follow ubuntu menu structure [11:24] mdz: moz problems, the same as are blocking OOo1 [11:24] mdz: copying Release and Release.gpg seems to work; how evil is that? [11:24] assuming I do a gpgv first [11:24] mdz: ubunutu's libs now have extra symbols in them which breaks OOo's internal moz libs [11:24] haggai: who's working on it? [11:25] mdz: me [11:25] haggai: oh, good :-) [11:25] is it a mozilla problem or an oo.o problem? [11:25] Kamion: it's less evil if you do a gpgv first [11:25] OOo really, it's their b0rken internal patched libs that cause the problem [11:25] Kamion: at that point, it's only as evil as messing with /var/lib/dpkg [11:25] but because of their patches it makes it hard to fix quickly [11:26] jdub: re OnlyDisplayIn=KDE, why is that needed? [11:26] haggai: things like kcontrol, etc. [11:26] mdz: I can have a look at the stacktrace today, otherwise I go to sleep :) [11:26] haggai: logged into gnome with kde installed recently? :) [11:26] jdub: surely if they are installed on the box, they should have menu entries? [11:26] jdub: yeah :) === robtaylor_ [~robtaylor@nat2.camtrial.co.uk] has joined #ubuntu-devel [11:26] haggai: not everything should [11:27] kcontrol (particularly where it sits in the menus, but even disregarding that) really shouldn't [11:27] mdz: OK, I think I'll do that, mark it TODO, and file a bug [11:28] jdub: ok, take e.g. you install a kde proggy on your mainly gnome system and don't use KDE. The prog has a kcontrol module - now what? [11:28] mvo_: go to sleep; I will mail you something [11:29] haggai: in general, i'd hope that the app would point to it too [11:29] mdz: thanks [11:29] haggai: see nautilus file manager properties, etc. [11:29] haggai: my only worry with splitting off kubuntu-* is that we'll miss out on lots of integration discussion [11:30] haggai: i'll almost certainly subscribe to kubuntu-devel, but won't be tracking it closely [11:30] hey. i've just been playing with getting jackd to work on ubuntu :) [11:30] jdub: yeah, I understand that, but the ubuntu lists are already too much traffic [11:30] jdub: they are not a friendly place to nuture new devs who aren't into Gnome ;) [11:31] 1 problem from the off, apparently the realtime-lsm module needs security configured as a module. any chance of this happening? [11:31] haggai: that's definitely true [11:31] haggai: but i'm already convinced [11:32] haggai: i'm wondering how to help with the integration problems it introduces [11:32] jdub: we should maybe crosspost stuff that is applicible to both or something, but I think much traffic won't be really interesting such as whether to include the menu shadow patch or not etc [11:32] jdub: maybe we need a place to talk about derived distributions in general? oh no another list.. [11:32] fear :) [11:33] haggai: and debian-custom wont suffice? [11:33] this is about ubuntu derivatives [11:33] daniels: ping [11:34] robtaylor_: I guess debian-custom would only be suitable for a subset of the discussion, since much will talk about deriving from ubuntu [11:34] haggai: yep, i'm seeing the issue now === kent_ [~kent@83.249.61.131] has joined #ubuntu-devel [11:35] whats kde doing on the sound server front, out of interest? [11:36] dieing slowly :( [11:36] :( [11:36] the arts maintainer has just thrown in the towel [11:36] cos this is something it'd be nice to solve for both desktops.. [11:36] well,i can see why.. [11:36] yup, maybe polypaudio is the way to go. I'm not following the KDE discussion [11:36] haggai: polypaudio... :) [11:37] is lennart on the kde lists? [11:37] jdub: dude I was at the BOF :) [11:37] pushing for it? [11:37] no idea sorry [11:38] jdub: what do you reckon to the curredn discussion on desktop-devel-list? i think i'm just pretty much reiterating what we decide at the bof, no? [11:38] nothing's going to happen in 2.10 timeframe [11:38] obviously :/ [11:39] for 2.12, gnome will switch to gstreamer at the library level and deprecate esound [11:39] i'm slighty confuesed about this, i thought gstreamer only defined audio chains *within* a process? [11:40] mdz: so udebs are not at all suited at creating live cd targets? maybe casper could be used to create installer iso then :) [11:40] surely you need to specify that apps should use gstreamer-polypaudio as the final sink/ [11:41] or is this something runtime configurable, and i've missed something completely? ;) [11:46] sivang: udebs are used only for the bootstrap process [11:46] sivang: you don't need to change the bootstrap process at all [11:46] jdub: how many more gnome 2.x releases are planned? [11:46] sivang: casper is very specific to the live CD task ... === Kamion hacks archive-copier egregiously [11:46] Kamion,mdz : ok :) I will try to keep on track :) [11:47] mdz: what is the casper applet? what python modules are needed to run it? === thully [~knoppix@wuser102-league.umnet.umich.edu] has joined #ubuntu-devel [11:47] ajmitch: NaN [11:47] jdub: *laugh* [11:48] Hi - which snapshot of Hoary works best for installation? I tried the last few days and they didn't work well - also, I tried installing Warty and then dist-upgrading from a snapshot CD (as I already had one burned and didn't want to download again) and I got a bunch of dependency problems. [11:48] ajmitch: http://www.gnome.org/~jdub/blog/projects/gnome/1095703226 [11:49] mdz: could you tell me how do I extract the cloop image so I can loop mount it? [11:49] heh [11:50] mdz: thanks for the backtrace === mvo_ goes to bed now [11:50] sivang: you use the tools in cloop-utils [11:50] sivang: the casper applet is hypothetical [11:50] I haven't written it yet [11:50] sivang: extract_compressed_fs filename.cloop >bla.chroot [11:51] jdub: I guess the gnome people are wanting to commit to ABI/API stability for awhile then? [11:51] amu,mdz : thanks alot :) [11:51] ajmitch: see live.gnome.org/ThreePointZero [11:52] sivang: my thanks will be good documentation for the customisation process ;-) [11:52] thully: 20050119.2 appears to work except that it has problems with apt authentication so you have to say 'y' to aptitude [11:52] ajmitch: the major interest in 3.0 is deprecating the crap [11:52] ajmitch: well, sorry, removing the deprecated crap [11:52] thully: this turns out to be because apt-cdrom doesn't support authentication, so I'm testing a workaround right now [11:52] jdub: that's what I understood, that there's a fair bit of accumulated crap since 2.0 :) [11:52] that has been deprecated [11:53] oh - cool, the last few releases have had some significant breakage - does this image fix the issue where the created user isn't in the sudoers group? [11:53] sadly I've only done a little coding with gnome, mostly with python [11:53] not group - file I mean [11:53] thully: yes, I fixed that a day or two ago; nobody reported it :P [11:53] not much has actually been deprecated yet [11:53] but there are lots of things to be purged [11:53] I happened to notice it myself :) [11:53] Kamion: lamont has done a d-i build for us; do i need to get elmo to BYHAND it? [11:54] mdz: yeah [11:54] elmo: pretty please? [11:54] I'm still seeing the date problems that I've seen since the warty pre-release [11:54] timezone, I mean [11:54] thully: there was also a good deal of breakage in ubuntu-desktop installability which I only managed to sort out today [11:54] mdz: Ok, on the TODO list, I don't forget promising this :) [11:54] thully: gah, how can that have persisted?! === carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #ubuntu-devel [11:55] well, I wonder how it slipped into the Warty release even! evidently I'm the only Ubuntu user who uses dial-up a significant amount [11:55] I don't understand this at all, that code is now in an utterly different place [11:55] Well, it doesn't happen on Debian - maybe check their code [11:55] thully: we were already using their code [11:55] yes - but with changes [11:56] yes, I know, I made most of those changes === zenrox [~zenrox@wbar7.sea1-4-10-181-189.sea1.dsl-verizon.net] has joined #ubuntu-devel [11:56] which is why I'm so baffled [11:57] I dunno - I'll jive 0119.2 a shot and see if it by any chance disappears [11:57] give [11:57] thully: you've tried an image that has the timezone question in the first stage, I take it? [11:57] yeser [11:57] yes [11:58] ok [11:58] thanks for your eternal patience :-) [11:58] I'd like to know what /etc/timezone, the /etc/localtime symlink, and 'date' look like when chrooted into /target before the first reboot [11:59] So, does anyone know why autohinter is disabled by default in recent Hoary snapshots? I liked having it on by default (it's subpixel which causes issues on CRT, not autohinter) [11:59] hwclock.sh is not now run during the timezone question, which I'd thought might be what was causing your problem [11:59] so i looked at the ubuntu backports repos yesterday [11:59] Kamion: I'll check that out [11:59] they are *insane* [12:00] hwclockfirst.sh *should* take care of it [12:00] there are going to be a bunch of very unhappy users trying to upgrade to hoary soon === buga [burjang@pandora.inf.elte.hu] has joined #ubuntu-devel [12:00] jdub: because of the version madness? [12:00] yeah [12:00] jdub: brokenness? [12:00] what have they done? [12:00] serious, serious brokenness [12:00] jdub: those repositories sprung up like a week after warty released [12:00] Kamion: you kinda have to look at the version numbers to truly understand [12:00] jdub: *six months* ffs [12:00] mdz: uh huh. [12:01] people are so afraid to upgrade to hoary that they are willing to subject themselves to worse breakage because of backports [12:01] oh how I need a decent monitor for my hoary box, to see gnome in all its glory [12:01] uh.. another d-i daily? [12:01] the 403 on warty-backports/main/binary-i386/Packages.gz is pretty special [12:02] elmo: mdz was crying for it for live CD testing :-) [12:02] elmo: the hotplug fix I did yesterday didn't make it, apparently