[08:07] <ScottK> lamont: Thanks.  I want to backport that.  I think we need the multiarch stuff for oneiric (or did you shove that in too - I'll look)
[08:10] <lamont> ScottK: multiarch is there
[08:10] <ScottK> OK.  Thanks.
[08:10] <lamont> ScottK: I've been informed that the multiarch patch is safe-for-debian, by he who knows best
[08:11] <ScottK> This is the same guy that did the patch that led to postfix being sru'ed in Natty already?
[08:11] <ScottK> ;-)
[08:12] <Daniel0108> hi MacSlow
[08:13] <MacSlow> Mahlzeit Daniel0108
[08:30] <alex6567> hello all! Did you know it posible use unicode on ncurses lib? http://ubuntuforums.org/showthread.php?t=1755327
[08:32] <mdke> I wonder if anyone could give me some guidance on who might be able/willing to help out with my query here: https://lists.ubuntu.com/archives/ubuntu-devel/2011-April/033095.html
[08:33] <slangasek> alex6567: note the topic's suggestion to use #ubuntu-app-devel for app development questions.  But for starters, if you want widechar support you need to link against libncursesw, not libncurses.
[08:36] <alex6567> thanks try to google
[08:47] <slangasek> bdmurray, bryceh_: do either of you have a good way to mass-mark a group of bugs as duplicates?
[08:47] <slangasek> (https://bugs.launchpad.net/~fadly87)
[08:47] <slangasek> (all duplicates of bug #780877)
[08:49] <sbeattie> slangasek: http://bazaar.launchpad.net/~ubuntu-bugcontrol/ubuntu-qa-tools/master/view/head:/responses/is-duplicate
[08:50] <cjwatson> slangasek: lp-set-dup?
[08:50] <cjwatson> (in ubuntu-dev-tools)
[08:51] <slangasek> cjwatson: ta :)
[08:53] <ogra_> cjwatson, for oneiric we want to drop the netbook seed, i was wondering if we have an easy way for doing that transition through the seed management or if i need to roll an empty dummy package to do that transition
[08:54] <ogra_> .oO(how did we handle it for UNR->desktop ?)
[10:01] <jelmer> pitti, hi
[10:01] <jelmer> pitti, did the bzr packaging session get cancelled?
[10:23] <cjwatson> could somebody in ubuntu-mir review bug 780591 reasonably soon?  it's blocking part of the perl transition (indirectly)
[10:33] <kees> sladen: regarding capabilities, this is one review of the isolation failure issues: http://forums.grsecurity.net/viewtopic.php?f=7&t=2522
[10:40] <sladen> kees: loooong read.  I only know from the vserver perserpective (and no several years out of date); but none of the first 19 are/were given to container ruid=0 by default (so empty /dev and no mknod())
[10:42] <sladen> kees: and CAP_NET_RAW only given explicitly (in this case to get ICMP ping to work)
[10:42] <kees> sladen: right, with a very restricted cap bounding set, you can feel much safer in a container
[10:43] <kees> sladen: I think the addition of syscall filtering will make that even stronger. (currently being discussed upstream, driven by the chromium folks)
[10:44] <sladen> kees: can apparmour already provide that level of syscall filtering (either complete denial, or explicit validation of syscall parameters)
[10:45] <kees> sladen: no, MACs (SELinux, AppArmor, etc) filter a higher level of access, but don't filter syscalls.
[10:45] <kees> sladen: for example, when new syscalls are added to the system, the MAC may not have any hooks into it, so an attacker has a certain subsets of the attack surface to use
[10:46] <sladen> kees: nod, ah
[10:46] <kees> it's all crazy :)
[10:47] <kees> SECCOMP (and soon SECCOMP mode 2) is what will do the syscall filtering. see "man prctl" for SECCOMP
[11:46] <cjwatson> ogra_: I'm not convinced we *did* handle it for unr->desktop (except perhaps in update-manager?)
[11:46] <cjwatson> ogra_: there's no way to do it in seed management - better and clearer to build an empty package by hand
[11:47] <cjwatson> ogra_: but if you can make update-manager help out, so much the better
[12:12] <alex6567> i have idea what libncursesw5-dev have problems. in src #include <ncursesw.h> say file not exist
[13:24] <highvoltage> if Martin Pitt is pitti, then why isn't Martin Pool called "pooli"?
[13:26] <Amaranth> highvoltage: you could be cartej :P
[13:27] <Amaranth> oh, I added an extra character
[13:27]  * Amaranth needs a nap
[14:20] <broder> cjwatson: if i run backport-helper against staging, where i've "approved" a backport of a single package to multiple releases, that bug comes up once for each series/backports project: http://paste.ubuntu.com/606144/
[14:20] <broder> does that not work for you?
[14:21] <cjwatson> broder: yes - that's not the point at which stuff is awkward for multiple target series
[14:24] <broder> oh, hmm, i guess i misunderstood the issue, then. what's the current problem?
[14:30] <cjwatson> broder: it's that the backports are all generated into a single directory, and the script doesn't like it when a file already exists
[14:30] <cjwatson> so .orig.tar.gz tends to clash
[14:31] <broder> ah, ok. so this is actually totally not my problem at all. got it :)
[14:34] <cjwatson> it's not a major problem, it just means us doing stuff in a couple of passes
[15:36] <hallyn> jdstrand: libvirt needs to be rebuilt for oneiric to find the new brctl (in /sbin/ not /usr/sbin now).  It doesn't need any changes, configure will find it.  How do I kick off the new build?
[15:37] <hallyn> i will take lxc integration, in cooperation with dlezcano
[15:37] <hallyn> we'll add it to the agenda for the lxc sprint in august
[15:38] <hallyn> doh
[15:38] <micahg> hallyn: someone just needs to upload a no change rebuild, dch -R to create the changelog
[15:39] <hallyn> micahg: ok, thanks
[16:01] <TerminX_> heh, I was bored so I decided to try upgrading a 32-bit install to a full 64-bit kernel/userland in-place without a reinstall
[16:02] <TerminX_> I actually made it work ;D busybox-static, 10 years of dpkg experience and some nasty hacks with sed on the contents of /var/lib/dpkg for the win
[16:10] <ogra_> infinity, poke
[17:08] <Cas07> is there a specific name for the special mode dash (Alt+F2) in unity?
[17:22] <tseliot> tjaalton: are you there?
[17:23] <tjaalton> tseliot: at my room
[17:23] <tseliot> tjaalton: do you already know what kind of food we're aiming for?
[17:23] <tjaalton> nope
[17:24] <tseliot> ok
[17:24] <tseliot> tjaalton: ok, let's talk downstairs then
[17:24] <tjaalton> yeah, meet you there
[17:29] <tseliot> bryceh: are you there?
[17:30] <bryceh> tseliot, yep, on my way down
[17:31] <bryceh> didn't find sarvatt
[17:32] <tseliot> bryceh, tjaalton: can you both join #dinner please?
[18:14] <hallyn> is htere a way to force lp to import from the package?  libvirt is two versions behind...
[18:34] <geser> hallyn: import to where in LP?
[18:43] <hallyn> geser: lp:ubuntu/oneiric/libvirt
[18:43] <hallyn> for now i'm just doing the sync by hand
[18:45] <geser> hallyn: this happens when the importer hits a bug, it stops importing any future version till this gets fixed (see http://package-import.ubuntu.com/status/libvirt.html#2011-03-29%2015:05:13.895974)
[19:26] <hallyn> geser: so how does one fix it?  do I bzr push a clean version cloned from the package?
[19:29] <geser> hallyn: ask james_w ; but probably pushing yourself won't improve the situation (you would need to take care of importing Ubuntu revisions and Debian revisions in the right order to have a proper history)
[19:59] <hallyn> geser: thanks
[19:59] <hallyn> james_w: ^
[20:09] <odla> will gnome3 be officially supported in 11.10 and beyond?
[20:29] <hallyn> jdstrand: do you happen to know why X-Python-Version in libvirt's debian/control is listed as 2.6?
[20:45] <infinity> ogra_: You poked?  I've been out all day.