=== low [~low@84.100.98.28] has joined #ubuntu-devel [12:05] hi there === trondaso [~trondaso@dsl-160-196.oeke.tiscali.no] has joined #ubuntu-devel [12:05] high low [12:05] :) [12:05] hah [12:06] there's a little bug in lilo.conf when you use xfs (hoary-preview, amd64) [12:07] kernel is /vmlinuz instead of /boot/vmlinuz [12:07] links should be made to vmlinuz in / [12:08] (they are on my system) [12:08] at least lilo complains and returns with an error code during install, doing it by hand makes it complaining too, until i put in /target/etc/lilo.conf /boot/vmlinuz === winkle [~winkle@lgh3814234.vittran.norrnod.se] has joined #ubuntu-devel [12:09] hmmm how to use install cd as rescue so i can solve last problems with lilo without reinstalling ? [12:10] you can also manually install grub to the mbr... the scripts won't work, but you can use the grub shell to embed it yourself === schweeb did this on his laptop [12:10] schweeb: grub does work on amd64 ? === trondaso [~trondaso@dsl-160-196.oeke.tiscali.no] has left #ubuntu-devel ["Konversation] [12:10] I dunno, I would assume so [12:11] schweeb: i'm coming for a source distro and i'm working a bit on amd64, and we were unable to make it work === lamont__r [~lamont@phantom.acmeps.com] has joined #ubuntu-devel [12:12] yes, it works on amd64 [12:12] great :) [12:13] i'll have to check how you did it [12:14] google for "grub embed" [12:14] I think it's in the manual as well [12:16] hmmm looks like lilo borked my first sector of / (xfs) [12:17] low, why did you use lilo ? [12:17] (i am on amd64 ... with the default, which is grub) [12:17] ogra: errr it was what it proposed to me [12:18] ogra: grub can't handle all situations; the installer switches to lilo sometimes [12:18] thats why i put your name in front of the sentence, yes :) [12:18] ogra: xfs is one of those cases [12:18] Kamion, oh, didnt know that [12:18] Kamion, it wasnt the case on warty .... [12:18] low: if you're using the hoary install CD, type 'rescue' at the boot prompt [12:19] ogra: yes, warty just broke with XFS :P [12:19] Kamion: thx. [12:19] (randomly, for some people, not everyone) [12:19] Kamion, nope, worked fine for me [12:19] heh [12:19] ogra: it's a race condition [12:19] thus "works for me" isn't enough, sadly [12:19] Kamion: any new daily builds since this am? [12:19] Kamion: i've already run grub with xfs as / without problem [12:19] Kamion, i know, i saw the redhat fix of running grub twice :) [12:20] low: for some people, it's fine. for others, it breaks. [12:20] Kamion: damn :( maybe sis+sata is the cursed couple ;) [12:21] *shrug* I have a cheapo laptop where it fails [12:22] well, anyway, i'm off to bed, need to wake up in 5 hours [12:22] thx for tips and help [12:22] see ya [12:23] % acpi [12:23] % [12:23] hrm... something's fishy.... [12:23] mjg59: you around? [12:33] Kamion: I've found using the grub shell and manually installing grub gets it to work... grub-install almost always fails [12:33] (w/ XFS> [12:33] s/>/)/ [12:34] although that's more work than most people should have to do... [12:39] why dont you create a small boot partiion with someting like ext3 and then the rest xfs? [12:40] zul, he is gone already [12:41] heh im so on holiday today ;) === LeeJunFan [junfan@64-186-37-120.skycon.net] has joined #ubuntu-devel [12:42] seb128: you around? === lamont__r heads home === kain [~kain@ip-72-109.sn2.eutelia.it] has joined #ubuntu-devel [12:49] lamont-away: ? [12:49] hmm.. ubuntu-artwork 0.2.20-2 on hoary doesn't fix the human icon issue [12:49] mjg59: he just left for home [12:49] Ah [12:51] mjg59: hell be back soon === tritium [~rimbert@12-202-89-11.client.insightBB.com] has joined #ubuntu-devel === goedson [~goedson@200.150.28.17] has joined #ubuntu-devel [01:02] mjg59: wondering why acpi isn't happy on my vaio.... what does pcc_acpi do, and should it be loaded as well as sony_acpi? [01:03] hrm.. was it something I said? :-) === narb777 [~narb@narghoul.skycon.net] has joined #ubuntu-devel [01:15] yes...yes it was [01:15] hey zul [01:15] hey Treenaks [01:15] er..trulux [01:16] how is it going? [01:17] evening zul [01:17] evening lamont how goes the battle? [01:17] rather relaxing day [01:17] same here [01:17] is it a public holiday in the us too ? === lamont checks and finds only 2 udev-screwed chroots. [01:18] sigh [01:18] canada as well [01:21] ogra: not entirely 100% sure, but I've been making it a slow day [01:21] ah, yeah [01:21] like fetching isos, and such [01:21] speaking of which, time to burn some [01:21] not a public holiday, but many people take off for Good Friday [01:22] zyga, what have you heard about powerbook g5? [01:26] ogra, here's more goodness: Mar 25 13:21:34 localhost kernel: EXT3-fs warning (device hda1): ext3_unlink: Deleting nonexistent file (1130787), 0 [01:28] no school today :) [01:28] i feel old now [01:28] mako: ping === HiddenWolf [~hidden@136.96.dynamic.phpg.net] has joined #ubuntu-devel [01:28] zul: grey haired? === witiza [~witiza@84.232.56.34] has joined #ubuntu-devel [01:29] adobbie: a bit [01:30] zul: ok, you probably are old then :) [01:30] zul, you shouldnt say that while lamont and me are here ;) [01:30] adobbie, only 29 [01:30] tseng: hey ddue [01:30] heh, young guy [01:31] mako: manoj signed my key! === ogra doas the tseng dance [01:31] tseng: killer! [01:31] mako: do i need to resend you my coc, or you still have it [01:31] finally [01:31] zul: that makes you 7 years old than I am [01:31] bleah [01:31] and six younger then me :) [01:32] ogra: old foggie [01:32] yeah, my beard is already graying out [01:32] ogra: dude.. unix beard time [01:33] i can't wait to pull off a unix beard [01:33] hehe, you mean like maddog ? [01:33] totally [01:33] even rms now is greying out [01:33] maddog has a real unix beard [01:33] ah, i think i'll wait 10 years for that [01:33] yeah [01:33] http://www.urbandictionary.com/define.php?term=Unix+beard [01:34] heh i have a beard it just needs to be grey [01:34] I shaved this morning :) [01:34] mako: I have not heard that term in general circulation, but it should be [01:34] mdz: dude, several years ago at LCA, there was a unix beard BOF (!!) [01:35] mako: coc? [01:35] mdz: to discuss beard care, etc, i guess [01:36] once, three people with unix beards told me that i "didn't really have a beard" [01:36] mako, ever had a beard ? [01:37] ogra: dude! i have a beard [01:37] ogra: you met me! === jonathaN [jonathan@4-16-ftth.onsnet.nu] has joined #ubuntu-devel === Arrogance [~aks@CPE0050ba556e4b-CM001225423850.cpe.net.cable.rogers.com] has joined #ubuntu-devel [01:37] mako, ah, come on, i mean a beard with some more expertise ;) [01:37] mako, its itching like hell.... [01:38] mdz: there must be some serious unix beards at nanog [01:39] mako: and that's just the women. :-) [01:40] hrm.. forget I said that. [01:40] you know that cheers episode where they have that beard competition? === ggi [~ggi@ggi.base.supporter.pdpc] has joined #ubuntu-devel [01:43] mako: there are _seas_ of them [01:44] mdz: please merge colin.watson@canonical.com--2005/casper--translations--0 again, up to patch-6 === sgtshatta [~omar@24-193-228-145.nyc.rr.com] has joined #ubuntu-devel === emanuelq [~modyo@201.255.46.109] has joined #ubuntu-devel [01:52] Kamion: done, uploaded === emanuelq [~modyo@201.255.46.109] has left #ubuntu-devel [] === wasabi [~wasabi@207.55.180.100] has joined #ubuntu-devel [02:00] Not sure how I ended up in the kind of position Christian Perrier fills in Debian ;) === maskie [~maskie@196-30-109-18.uudial.uunet.co.za] has joined #ubuntu-devel === kain [~kain@ip-72-109.sn2.eutelia.it] has joined #ubuntu-devel [02:09] Kamion: maybe we should recruit him [02:09] any news on fixing ubuntu-artwork in hoary? I really like human icons but they're broken [02:10] they work for me.. are you tracking this on bugzilla? [02:11] no, not yet, since I saw the new package, but it doesn't fix the problem [02:11] ogra also confirmed this problem [02:11] tseng, are you up to date ? and did restart nautilus since he update ? [02:11] the even [02:11] i killed nautilus [02:11] it still works. [02:11] kain: what's broken? [02:12] jdub, there is only the default icon [02:12] jdub, ubuntu human icon theme, there's something missing in package [02:12] morning btw [02:12] Keybuk: he's certainly a competent installer / localisation guy [02:12] hrm.. current world likes my vaio. acpi happiness. [02:12] jdub, mime types that is [02:12] ah, i must've broken failover [02:13] s/failover/fallback/ [02:13] also,as said, I remember four directories under /usr/share/icons/Human/scalable [02:13] now they're three [02:13] yeah, not really related [02:13] ok [02:13] Kamion: yeah, he does l10n for dpkg and apt as well === lamont pesters Kamion about the installers [02:14] dhcp client [02:14] where is that? [02:15] lamont: dhcp-client-udeb === lamont prepares a patch [02:15] dhcp2??? [02:15] dhcp3 so didn't even remotely start to fit [02:15] ah, yes. [02:15] well. === lamont makes dhcp2 also ask for interface-mtu [02:17] # install the udeb binary (no need to install dhclient-script [02:17] # since it is provided by another udeb package. See #249373) [02:17] hrm. [02:18] netcfg has its own dhclient-script [02:18] yeah [02:19] note I have a pending netcfg upload === trulux|working is now known as trulux|sleep [02:19] might be that the fix actually goes in dhcp-client-udeb... [02:20] netcfg doesn't deliver dhclient.conf, yes? [02:20] lamont: it writes it itself [02:20] at run-time [02:21] ah. [02:21] fprintf(dc, "send host-name \"%s\";\n", dhostname); [02:21] can you add 'request interface-mtu' in there somewhere? === trulux|sleep is now known as trulux [02:21] that is, add ",interface-mtu" to the request line? [02:22] definitely in "please send a patch" territory here :) [02:22] I don't feel confident munging that myself [02:22] source name for netcfg? [02:22] netcfg [02:22] oh. sure. who would think of that??? [02:22] :-) [02:22] testing it's a pig, though [02:22] is that in the ramdisk? [02:22] initrd [02:22] transferring an updated binary over on a USB stick is one option [02:23] lamont: it's in the netboot initrd, not the cdrom initrd [02:23] so you could bung a .deb on a CD [02:24] kewlness. === lamont has that technology. [02:24] can you rebuild him, though? [02:25] fprintf(dc, "send host-name \"%s\";\n", dhostname); [02:25] fprintf(dc, "request subnet-mask, broadcast-address, time-offset, routers,\n"); [02:25] fprintf(dc, " domain-name, domain-name-servers, host-name, interface-mtu;\n") [02:25] I _think_ that's the defaults + mtu... [02:26] does d-i care at all about time-offset? [02:27] you have plans to test a new netcfg? [02:27] and an in-house dhcp server? [02:28] well, so far mine's just YA translation upload, but sure, can do [02:28] and yes, have DHCP server [02:28] time-offset> no idea [02:28] dhcp3 or something else? [02:28] whatever woody has, I guess [02:29] ii dhcp 2.0pl5-11woody DHCP server for automatic IP address assignm === lamont double checks syntax [02:29] won't you need to do something with interface-mtu, too? [02:29] and I guess s/ /\t/ for neatness :) [02:30] uh, yes. But I think mtu was in dhclient-script in the system [02:30] oh yes [02:31] $new_interface_mtu [02:31] so it may just work [02:32] in the subnet declaration in question, add 'option interface-mtu 1496;' [02:32] then verify that the network is configured as such [02:33] meanwhile, I'll root around in the server code some more and see if I can find the 'always send this option' option [02:33] is 1496 generally safe? other people are using this network === mjg59 [mjg59@cavan.codon.org.uk] has joined #ubuntu-devel [02:34] well, it should really match across the entire network... [02:34] default here is 1492 [02:34] so you may want to create a bootp entry for the MAC in question [02:34] If i remember correctly [02:35] this is kind of not what I want to be doing at 1:30am :) [02:35] Kamion: so if your lease time is long enough, and no one asks during the test... [02:35] change it back to 1500 and all is magically suddenly good. [02:35] I use 1492 for my MTU [02:35] and you only lockup when you try to send a large packet from the 1500-machine to the 1496 machine. === lamont will test [02:36] I could set 1500 and set -x / otherwise trace the dhclient-script [02:36] Kamion: how soon you plan on netcfg upload? [02:36] Kamion: that'd work [02:36] lamont: have a few others to do first and some base-config stuff to test [02:36] lamont: so no vast hurry === macewan [~macewan@ip68-101-19-222.nc.hr.cox.net] has joined #ubuntu-devel [02:37] ok, will give it a go in a bit [02:37] Kamion: btw, ia64 server install fails trying to install elilo on the disk... you remember what the fix was for that? [02:38] lamont: all the fixes I remember, I've applied ... [02:39] /var/log/syslog and/or /var/log/messages ought to be enlightening === jdub growls at mipsel [02:41] i can't do big uploads through my firewall [02:41] big being, like 4M (ubuntu-artwork) [02:41] luckily there are lots of APs around ;) [02:41] elilo: unrecognized option /dev/sda2 :-( === lamont rolls it back to the top of the hill to manually partition [02:46] i'll never be a unix expert, i can't deal w/ the beard :/ [02:46] lamont: erm - any idea where that might be coming from? [02:47] chroot /target /usr/sbin/elilo --autoconf --boot $bootpart \ [02:47] --root $rootfs --efiboot [02:47] oh, gar [02:47] I bet I know [02:49] lamont: could you stick 'set -x' near the top of /var/lib/dpkg/info/elilo-installer.postinst? [02:49] lamont: and make sure you only have one EFI partition (which I'm betting is your current state) [02:49] uh, when I get back there... [02:49] yep. fdisk print just showed 1 partition [02:50] lamont: elmo complained about being asked a question with only one choice, so I lowered the priority in that case; only I suspect now that cdebconf doesn't set a sane default if that happens [02:50] lol [02:51] there are 2 drives on the machine, not sure what drive 2 looks like [02:51] sda2 is the rootfs [02:51] 1 = efi, 2=ext3, 3=swap === Kamion compares with yaboot-installer [02:53] elilo-installer instrumented, base install proceeding (again)( [02:55] patch written in the event it goes how I expect [02:56] namely, if you see 'bootpart=' in there [02:56] with no value [02:57] cool.. I want to get my new mirror machine into production, you see... [02:58] hrm.. it picked itanium-smp to install... that could be bad, can't remember [02:58] IIRC we talked about that and agreed [02:58] ok [02:58] but hmm === lamont was more introducer and/or conduit in that discussion [02:59] that is actually kind of odd [02:59] /var/log/syslog and /proc/cpuinfo would be useful to me [02:59] machine is 2-way mckinley [02:59] then indeed I think it should have picked linux-mckinley-smp [02:59] is there an ssh-client and/or netcat on the install disk? [02:59] openssh-client-udeb [03:00] nc is in busybox [03:00] nc works === lamont grumbles at archive-copier, watches it churn === lamont remembers to fix the mtu before he does anything else [03:01] lamont: what, in a server install? [03:01] yeagh [03:02] is b0rken then [03:02] this AM's ia64 install, fwiw [03:02] lamont: oh. er. so how exactly are you doing this server install? [03:03] vendor : GenuineIntel / arch : IA-64 / family : Itanium 2 [03:03] 'cos elilo.conf kind of, er, doesn't have an entry for server [03:03] at boot, adding 'server', otherwise picking the vga non-expert mode [03:03] lamont: that does precisely nothing ;) [03:03] feh\ [03:03] what must one do? [03:04] are there some of these elilo.conf entries that I could sensibly trim? [03:04] one must boot with preseed/file=/cdrom/preseed/server.seed as kernel argument [03:04] vga, serial clearly need to be there. expert mode makes sense for both as well. [03:04] ok [03:04] so I just have to double up all the entries? [03:05] I'm ambivalent about expert mode [03:05] well, you could drop server-expert... :-) [03:05] people overuse it; I almost think we could drop it and tell people to boot with DEBCONF_PRIORITY=low if they care [03:06] Kamion: I agree completely [03:06] I'm too cowardly to do that for !ia64 at this point, but I think I'll do it for ia64 [03:06] hey, language-support-* is installable on ia64 now, so I can remove that hack ... === wasabi [~wasabi@207.55.180.100] has joined #ubuntu-devel === bluefoxicy [~bluefox@pcp484971pcs.whtmrs01.md.comcast.net] has joined #ubuntu-devel [03:08] lamont: what do the 'processor' lines in /proc/cpuinfo look like? [03:10] Kamion: http://ia.mmjgroup.com/~lamont/ia64/cpuinfo [03:10] Mar 25 19:02:46 main-menu[5770] : (process:28216): bootpart= [03:10] lamont: bingo; willfix [03:11] should be /dev/sda1 for me> [03:11] ? [03:11] I guess, whatever your EFI partition is anyway [03:12] cute part is that after the failure, it prompts you to answer the question... :-) [03:12] lamont: yeah, your priority will get dropped - and then it should work [03:14] well, not 100% sure -- I edited the script and hardcoded /dev/sda1 :0- [03:14] hah [03:14] lamont: /var/log/syslog should have some wittering from base-installer about what it was thinking when selecting the kernel [03:15] ROTFL [03:15] Mar 25 18:55:05 base-installer: info: Found kernels 'linux-itanium-smp,linux-mckinley-smp,linux-image-itanium-smp,linux-image-mckinley-smp,linux-image-2.6.10-5-mckinley-smp,linux-image-2.6.10-5-itanium-smp' [03:15] Mar 25 18:55:05 base-installer: info: arch_kernel: linux-mckinley (absent) [03:15] Mar 25 18:55:05 base-installer: info: Using kernel 'linux-itanium-smp' [03:15] btw, syslog in the same directory === lamont copied all of /var/log [03:16] oh, your cpuinfo is only reporting one processor [03:16] what's the Right Way to find out the number of processors? [03:16] hrm... maybe I am UP... [03:17] checking [03:17] it should be running an SMP kernel, even === lamont grumbles. is uniprocessor machine. :-( [03:18] gonna have to pester the nice men [03:20] lamont: elilo-installer fix uploading [03:21] and yeah, mckinley-UP should pick mckinley-SMP (since no UP on CD) === lamont really does a server install this time.. and will verify the question thing from elilo-installer [03:23] lamont: ok. I can only think of one way to fix that without rather more serious upstream churn in base-installer, and it's a grievous hack [03:23] basically [03:23] - chroot /target apt-cache search ^linux- | grep '^linux-$amd64\|386\|686\|k7\|power\|itanium\|mckinley\|sparc\|hppa$'; [03:23] + chroot /target apt-cache search ^linux- | grep '^linux-mckinley'; [03:23] oh, that's vile. [03:23] + chroot /target apt-cache search ^linux- | grep '^linux-$amd64\|386\|686\|k7\|power\|itanium\|sparc\|hppa$'; [03:24] or something along those lines [03:24] is that to get the order correct/ [03:24] I think that makes me just too ill though [03:24] ? [03:24] yeah -not worth doing [03:24] since really, you're going to want to load a tuned kernel next anyway, ... [03:24] lamont: it's more because it's damned hard to detect which ones are the metapackages [03:24] the other solution would be to include a linux-mckinley (not smp) kernel [03:24] without just hardcoding a stupid list like the above [03:25] that'd suck a lot of CD space [03:25] Kamion: not true. anything with a size less than 2MB is a meta package. :-) === lamont vomits [03:25] haha [03:25] well.... 'tis truth, it is... :-) [03:26] you know, that's actually the least bad option I've heard, horrible though it is [03:26] but I think it should be in the "dirty hacks and how we can purge them" BOF at UDU [03:27] yes [03:28] I prefer to think of it as an elegant solution, not a gross hack. :-() [03:29] why don't you just tag the metapackages in the Packages file? [03:29] aj: was just going to suggest adding something to the packages file.. [03:29] Task: linux-meta [03:29] I think we thought about that, but apt-cache search wouldn't show them up, and we don't have many cleverer tools at that point [03:29] i mean, apt-ftparchive does extraoverrides these days [03:29] or even kernel-meta [03:29] you could have "Meta: yes" easily enough [03:30] also we only seem to remember to look at this issue ten days or so before release ;) it was the same last time round, it was a last-minute thing ... [03:30] Kamion: eh? [03:30] aj: can you search on random headers with apt-cache? [03:31] Kamion: yep. question works, even without modifying the script. ;-) [03:31] lamont: with new elilo-installer? that was quick [03:31] no [03:32] old elilo-installer, hit return a couple times after the failure [03:32] (prompted, and then it works0 [03:32] Kamion: you know, given the state of ia64/desktop, and the direction everyone is going with ia64, you could just make server the default... :) [03:32] hm, I suppose I could use apt-cache search --full and then clever, er, something [03:33] lamont: it is tempting [03:33] Kamion: I bet mdz would be OK with it. :-) [03:33] Kamion: pipe --full through grep-dctrl? [03:34] aj: don't have it there; we've only just installed the base system [03:34] Kamion: what're you trying to do? [03:34] aj: this is base-installer's logic to select the right kernel to install [03:34] Kamion: well, awk'll suffice anyway [03:34] no awk [03:35] oh, I do [03:35] I could chroot [03:35] i thought you installed the base system? [03:35] yeah, was in automatic "no awk in busybox" mode [03:35] oh, hrm [03:35] i was going to add something fairly similar to debootstrap's little C helper [03:35] have chroot /target perl, even [03:36] C is OK, just typically more effort and larger; but base-installer already has a bit of C code in it to interact with debootstrap [03:36] and b-i doesn't really have to be ultra-small [03:37] well, it was more going to be "pkgdetails list Base: yes /target/var/lib/apt/lists/debootstrap_Packages" or so [03:37] archive-copier has something very similar [03:38] but that's waiting on some apt-ftparchive changes i haven't gotten around to cleaning up for mdz yet [03:38] Ubuntu CDs have Task: ubuntu-base now, if that'd be helpful for testing debootstrap stuff [03:39] although it probably includes kernel packages and bootloaders and such that aren't appropriate in debootstrap, so blah [03:39] nah, the apt-ftparchive stuff just needs to be the per-arch-extraoverrides [03:39] then it's just moving debootstrap's idea of base into the packages file [03:39] not much testing required [03:39] hah, yeah, half the reason I have Task: ubuntu-base on the CD is that I needed to hack around the lack of per-arch extraoverrides [03:40] Task: ubuntu-desktop needed to be different on different architectures, in some cases [03:40] and the rest fell out in the wash [03:49] hrmpf. could have sworn I uploaded the new dhcp3 yesterday === tritium [~rimbert@12-202-89-11.client.insightBB.com] has left #ubuntu-devel ["Leaving"] [03:52] daniels around? [04:02] lamont: so, what should I do about netcfg? [04:02] I could just upload what I've got here to get the translation stats right, and then you can fix up what you need ... [04:02] I think I'm too tired to do sensible testing myself === LeeJunFan [junfan@64-186-37-120.skycon.net] has joined #ubuntu-devel [04:11] are debian-cd utils the best way to go to make a DVD/RW or 2 with main,universe? I have a local mirror with debmirror. [04:12] Is Ubuntu gonna use a dynamic theme when cairo comes into gtk+ main for ubuntu? :) [04:13] lamont: pcc_acpi is for Panasonics - it'll often end up loaded even when it's not strictly necessary (acpi has no infrastructure for hotplug yet) [04:16] luminosity as the default wm for bendy/breezy. that's what i'm gonna push for ;) [04:16] LeeJunFan: haven't quite finished exporting my debian-cd branch to the world; it's nearly there but not quite [04:16] LeeJunFan: come back in a few days and I should have it sorted :) === mjg59 heads to bed [04:18] Kamion: thanks. I have a load of friends who are gonna want to install ubuntu/kubuntu that have dial-up :) [04:19] LeeJunFan: we already produce a DVD of all the supported packages [04:19] if they're utterly and unredeemably desperate for something from universe, we'd like to know so that we can decide if it should be supported ... [04:20] lamont: I've gone ahead and uploaded netcfg 1.08ubuntu5 with just the translation fixes, since I need to sleep soon [04:20] lamont: go ahead and do whatever you need to do to its DHCP code if you've had a chance to test it [04:22] Kamion: no, mostly newbies - they don't know what they want yet. But I know a few of them want kde default, and the DVD's currently seem to be ubuntu flavored only. [04:23] granted they could apt kde, but... [04:25] LeeJunFan: there's a Kubuntu DVD too [04:25]