[00:04] Where can I find the script that ubiquity uses for install? === Lutin is now known as Guest87444 [00:16] unomi: apt-get source ubiquity and look at scripts/install.py? [00:18] thanks, jpds there wouldn't be a way to modify it from a usb-stick install medium? [00:19] unomi: ubiquity: /usr/share/ubiquity/install.py [00:19] cheers === Guest87444 is now known as Lutin [00:27] unfortunately I am not seeing any references to parted in there === tkamppeter_ is now known as tkamppeter === Amto`OFF is now known as XawaN === XawaN is now known as Amto`OFF [02:13] jdong: https://bugs.launchpad.net/ubuntu/+source/ia32-libs/+bug/601499 [02:13] Launchpad bug 601499 in ia32-libs (Ubuntu) "[lucid] ia32-libs stable release update" [Undecided,New] === dendro-afk is now known as dendrobates [04:08] what's the proper channel to ask about packages from the canonical DX team PPA? === Amaranth_ is now known as Amaranth [11:26] Any idea why schroot no longer works in maverick with encrypted homedirs? Used to work fine in lucid:-( === lifeless_ is now known as lifeless [13:55] Replacing bind with rbind in the line mounting /home in /etc/schroot/default/fstab fixes schroot for me:-) [14:13] debian-installer takes 1h+, whereas ubiquity takes ~20min, to install an equivalent standard desktop install. Is this how things are and should be? Or should this be brought up as a bug report? [14:26] Alternate version is extudied to work on pc with low ram [16:43] njin: re. ubiquity / d-i ... So the adaptation for low memory usage comes at the toll of d-i not using system resources even if they exist? I was under the impression that d-i was meant for far more than just lowmem installs..? [16:52] arand: i'm not an expert, the only things that i can see is D.I. is used in alternate versions, Ubiquity in others, and effectively on the same machine is slower than ubiquity [16:56] njin: Yep, I just did an equivalent desktop install on the same machine using both the liveCD and the alternate installer, at the diff in time was at least 40min. Which does seem a bit odd. [16:56] s/at/and/ [16:59] no, i'm an iso-tester and the difference isn't about 40 min, it can be about 10-15 [17:33] hmm [17:33] do other SRU team members have an opinion on the massive Banshee diff sitting in lucid unapproved? [17:52] bugfix uploads have generally been accepted so far === Lutin is now known as Guest23212 [18:40] does anyone know which part of ubuntu 10.04 is responsible for drawing/loading the desktop wallpaper by default? like is it nautilus? [18:42] Carb0n: execute this and then click on the desktop: xprop | grep CLASS [18:43] cool thanks [19:23] Is anyone here familiar with #ubuntu-toolchain and whether it is still used or needed? === Guest23212 is now known as Lutin [20:27] i want to know why evolution is preferd over thunderbird? I am just a regular user. want to know [20:35] abhi_nav: it's personal preference [20:36] micahg, no. I want to know reason why evolution is default. is it best? if yes then tell me so [20:36] abhi_nav: probably because it integrates better into the desktop [20:36] micahg, ok. thanks :) [20:36] abhi_nav: but you're free to use whatever client you like [20:37] micahg, yah I know. I just wanted to know whic is the best? its ok [20:38] abhi_nav: it depends on the person [20:38] micahg, you mean persons likes or you mean persons requirements? [20:38] abhi_nav: both :) [20:38] micahg, hmm [20:51] micahg, I am a comp engg student. I do little contributin using laucnpad etc. and no prefessional work still now. so which one you wll recommend for me? [20:52] abhi_nav: I can't be impartial as I'm one of the Mozilla maintainers :), try both if you don't have a preference and keep using the one you like more [20:52] micahg, :( ok. btw thanks. [20:52] :) [22:12] arand: that's how things are and should be. unpacking .debs is more flexible, but is intrinsically slower than copying a filesystem [22:14] arand: (although there is an existing open bug that we should be telling dpkg to use unsafe I/O during installation, which would make d-i faster) [22:14] cjwatson: has anyone looked into the dpkg safe IO implementation - reading the db doesn't seem like it should be writing, and reading shows up as slow [22:15] cjwatson: Okay, right, I wasn't aware of a major diff in the way they installed the system, but then it makes quite a lot of sense I guess, thanks for the info! [22:16] lifeless: that's orthogonal [22:16] arand: completely different installation method [22:16] cjwatson: ok, different cause for that slowdown? [22:17] lifeless: the unsafe I/O thing is specifically and only due to the recent sync changes - I don't want to scope-creep into speeding up db reading [22:17] lifeless: (which, if it were to be optimised, could be done generically and wouldn't need to be specific to installation) [22:17] cjwatson: for clarity, it seemed to me that db reading took a nosedive at the same time safe-IO arrived. [22:17] cjwatson: I had thought they were thus related [22:17] it can't be related [22:18] it actually should have speeded up around that time, though! [22:18] - Use FIEMAP when available (on Linux based systems) to sort the .list [22:18] files loading order. With a cold cache it improves up to a 70%. [22:18] Thanks to Morten Hustveit . LP: #442114 [22:18] interesting [22:18] (dpkg 1.15.5.6ubuntu2) [22:19] and it did improve things for me, IIRC, I seem to remember testing that [22:19] so maybe somewhat system-specific, though I don't know how - if it's reproducible in a Debian chroot, maybe take it upstream? [22:20] cjwatson: when I get some spare cycles I'll have a fiddle. No promises - more than slightly frenetic at the moment. [22:30] holaaas === RainCT_ is now known as RainCT [23:44] What part of ubuntu is responsible for the laptops Fn+F1-12 shortcuts ? === lifeless_ is now known as lifeless