[00:02] <mterry> cjwatson, thanks for quick review!  It was too quick.  I tried to put the merge in WIP mode, but you must have reviewed before I did that.  If you'll look at it again, it should be better
[00:02] <cjwatson> I saw the WIP but thought it might be worth saying anyway
[00:02] <xxiao> ok works
[00:03] <xxiao> had to /bin/sh in to passwd then reboot
[00:04] <mterry> cjwatson, well I had been an idiot and forgot about pkexec stripping DISPLAY.  So I had to redo it altogether after I thought to at least test it  :)  Thankfully synaptic ships a pkexec version now, and I don't think we need to stress about cddistupgrader, assuming the "Suggests" status is appropriate for it
[00:04] <mterry> In the sense that update-notifier-common merely Suggests gksu
[00:05] <cjwatson> re-reviewed
[00:06] <xxiao> hold on...i am inside ubuntu-core as the rootfs over nfs, the thing does not include ifconfig/dhcp then how can I get an ip then do apt-get? is this just for chroot then add stuff?
[00:06] <sarnold> xxiao: is 'ip' installed?
[00:06] <cjwatson> ubuntu-core is intended as a base chroot that you add more things to as needed
[00:07] <xxiao> sarnold: thanks forgot that :) it's there
[00:07] <xnox> xxiao: https://wiki.ubuntu.com/Core   bring your own bootloader, kernel, configs
[00:07] <xnox> xxiao: https://wiki.ubuntu.com/Core/InstallationExample
[00:18] <xxiao> cjwatson: ubuntu-core also works as a standalone rootfs, the only thing I need is to change root passwd for ttyS0 access, now i can apt-get
[00:20] <cjwatson> xxiao: well, sure, but it's not guaranteed to have things like useful networking :-)
[00:21] <cjwatson> if you can use it standalone that's a bonus
[00:21] <xxiao> so far so good, loading stuff to it now
[00:30] <bryce> @pilot out
[01:22] <xxiao> BenC: 13.04 rootfs is basically usable, until kvm is needed that is
[01:23] <BenC> xxiao: How do you mean?
[01:25] <xxiao> BenC: could not install qemu-kvm
[01:26] <xxiao> i am installing folsom which pulls in kvm
[01:29] <xxiao> dumb me, kvm is not enabled in the kernel yet
[03:43] <infinity> xnox: Both those Debian ports are unfficial for a reason.  They're largely unnecessary, but fun toys.
[03:43] <infinity> xnox: (Well, and spe is also usually mostly broken, so could never meet the release port criteria)
[05:57] <pitti> Good morning
[05:59] <pitti> infinity: want me to mass-copy all the precise langpacks to -updates?
[06:01] <infinity> pitti: If you think they're as validated as we can manage, go nuts.
[06:02] <pitti> infinity: they haven't gone through the usual validation cycle of per-language community testing in that short time, of course
[06:02] <pitti> infinity: we can do that, of course; but at least for the languages on the images we probably want to copy them to reduce size
[06:02] <infinity> pitti: The images are still building from -proposed currently, I believe.  Though, that should change.
[06:03] <infinity> Oh, maybe they're not anymore.
[07:03] <dholbach> good morning
[07:20] <dholbach> wow, xsensors tell me it's 95°C
[07:23] <mvo> dholbach: hey! does "cat /proc/acpi/ibm/thermal " tell that as well?
[07:23] <dholbach> mvo, file does not exist :)
[07:24] <dholbach> but ../fan says "speed: 4485" :)
[07:25] <mvo> dholbach: :)
[07:47] <ajmitch> dholbach: I'd say that was a bit warm, but my old laptop was topping out at 105 when it was spamming syslog about throttling power :)
[07:47] <dholbach> ajmitch, I'm back to 58°C already :)
[07:47] <dholbach> it might have been the two builds which were running :)
[07:47] <ajmitch> possibly :)
[08:59] <pitti> "Fetched 14.1 MB in 213503982334601d 6h 0min 40s (0 B/s)" -- go apt!
[09:00] <pitti> looongest update evar
[09:00] <pitti> (it really just took some 5 seconds, FTR)
[09:07] <astraljava> It just felt like 5 seconds, it actually took longer than what the range for time data type allows, and started over from the negative end.
[09:10] <soren> pitti: wtf? Did ntpdate run in the middle of the apt-get run?
[09:13] <soren> pitti: Yeah, that adds up. 15 seconds real runtime, a clock that goes an hour backwards and then 64 bit timestamp.
[09:14] <pitti> soren: perhaps; it was a freshly booted VM
[09:14] <soren> 213503982334601d 6h 0min 40s is 18446744073709548040 seconds. 2^64 is 18446744073709551616. -
[09:14] <soren> 18446744073709548040-18446744073709551616 = -3576
[09:15] <soren> An hour minus 14 seconds backwards.
[09:20] <pitti> hah, neat
[10:01] <cjwatson> infinity: I haven't changed them to point to -updates yet; if somebody else has it would be nice if they'd told me ...
[10:06] <Laney> hmm, I don't think the old gksu u-n caused a root browser to be spawned, did it?
[10:09] <cjwatson> Laney: a root *browser*?
[10:10] <Laney> root     10972  7.3  0.4 917900 139716 ?       Sl   10:05   0:01      |           \_ /usr/lib/firefox/firefox https://bugs.launchpad.net/ubuntu/+source/kmod/+filebug/f4e681b0-6c56-11e2-8e58-0025b3df357a?field.title=kmod+assert+failure%3A+modprobe%3A+..%2Ftools%2Fmodprobe.c%3A550%3A+print_action%3A+Assertion+%60kmod_module_get_initstate%28m%29+%3D%3D+KMOD_MODULE_BUILTIN%27+failed.
[10:11] <cjwatson> bless you
[10:11] <cjwatson> oh, right, so apport launches firefox ... hmm
[10:12] <Laney> I wonder why it worked properly when gksuing though
[10:12] <cjwatson> apport has a thing to look at SUDO_UID
[10:12] <Laney> I always just got an additional tab in my existing browser session
[10:12] <Laney> ah, so that's not passed along any more
[10:13]  * cjwatson wonders if pkexec sets anything usefuPKEXEC_UID=1000
[10:13] <cjwatson> er, yeah, mangled, but YKWIM
[10:13] <cjwatson> pitti: would it be reasonable for apport to check PKEXEC_UID as well as SUDO_UID?
[10:15] <pitti> cjwatson: yes, absolutely; I'll fix that now
[10:15] <cjwatson> thanks
[10:15] <cjwatson> Laney: thanks for the report
[10:16] <Laney> sure
[10:16] <pitti> yeah, thanks Laney
[10:16] <Laney> It was a "WTF, where's vimperator?" that tipped me off
[10:18] <pitti> hm, why don't I get a notification for that in the first place now
[10:24] <pitti> oh, I do; compiz put apport in the back
[10:24] <pitti> it seems to do that for new windows now
[10:25] <ogra> pitti, now ? i have that in precise on my desktop
[10:28] <pitti> cjwatson, Laney: uploaded, thanks for reporting
[10:29] <Laney> great, thank you
[13:55] <shadeslayer> cjwatson: hi, would it be possible to merge a new live-build from debian? alternatively, just apply the patch from here : http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=693250 ?
[13:55] <shadeslayer> I've tested the patch and it works for me
[13:56] <shadeslayer> ( using live-build from raring ofcourse )
[14:31] <cjwatson> shadeslayer: OK - I'd rather merge than cherry-pick, but not sure I'll quite get to it today
[14:31] <cjwatson> shadeslayer: but on my to-do list now
[14:31] <shadeslayer> cjwatson: sure, no problem :)
[14:31] <shadeslayer> and thanks!
[14:31] <cjwatson> (we're a bit further behind than I'd realised)
[14:32] <xxiao> interesting, lunbuntu has ppc version iso
[14:32] <shadeslayer> yeah, I noticed they released 4.0 O_O
[14:32] <xxiao> precise 12.04
[14:32] <shadeslayer> they're like ultra quick
[14:32] <cjwatson> well, I wouldn't merge from experimental
[14:33] <Quintasan> Laney: Did you upload maliit-framework to Debian?
[14:36] <mpt> ev, https://wiki.ubuntu.com/ErrorTracker#settings-developer
[14:39] <Laney> Quintasan: sure did
[14:39] <Laney> you can fetch the packages from raring NEW
[14:39] <Quintasan> Laney: Do you have appropriate powers in Ubuntu to sync the package?
[14:39] <Laney> I have ALL the powers
[14:39] <Quintasan> :D
[14:40] <Quintasan> I'VE GOT THE POWER!
[14:40] <Quintasan> Laney: Can you sync it then?
[14:40]  * Laney gyrates
[14:40]  * Laney hands the mind bleach to anyone reading
[14:40] <Laney> Quintasan: yeah, when it gets accepted I can
[14:40] <Quintasan> Laney: Splendid. I'll throw plugins your way when I'm done with them
[14:41] <Laney> Quintasan: Oh, I uploaded that one too
[14:41] <Quintasan> Laney: You did that already?
[14:41] <Laney> but I can reupload any improvements over the top of what's already there
[14:41]  * Quintasan hides
[14:41] <Laney> yeah I said in the email
[14:41] <Quintasan> Well, I'll take a look but I doubt there is anything I can improve there
[14:42] <Laney> I'm sure there will be :-)
[14:42] <Quintasan> Ha, it seems like a challenge.
[14:42] <amitk> /quit/quit
[16:43] <zyga> cjwatson: do you know if it's possible to use pybuild to generate packages for 12.04+
[16:48] <cjwatson> zyga: You may have mistaken me for a pybuild expert
[16:49] <cjwatson> zyga: But, AFAIK, you won't be able to build *on* 12.04/12.10 unless somebody has backported pybuild; however the binary packages that pybuild produces don't in themselves require anything special
[16:49] <xnox> zyga: the binaries should be usable on 12.04, but pybuild itself is not available on precise, so it will not be able to build there.
[16:49] <zyga> ah
[16:49] <zyga> so if my desire is to release a new package of my own library I should be able to build it on raring (the source package), push to a ppa and just copy to other releases?
[16:50] <xnox> yes
[16:50] <zyga> will I need to manually modify the changelog per package to get it to co-exist in all systems?
[16:51] <xnox> only the top version wins, so you'll either have the same binary (and hence the same version number) across all - or you don't use pybuild and upload source package per release
[16:52] <xnox> (or e.g. use a recipe to do those)
[16:52] <zyga> hmm
[16:52] <xnox> zyga: packaging without pybuild is easy and has been done before.
[16:52] <zyga> I'm _probably_ fine with using the same binary
[16:52] <xnox> zyga: do you have both python2 and 3 or just one python series?
[16:52] <zyga> yeah, I know, I just feel annoyed each time I want to package a trivial pure python package
[16:52] <xnox> %:
[16:53] <xnox>        dh $@ --with python2
[16:53] <xnox> done.
[16:53] <zyga> that has a library (for 2k,3k pythons), maybe a few command line programs (this time none) and some shared data, and docs
[16:53] <zyga> well, I want py3k too
[16:53] <xnox> make a second source package
[16:53] <xnox> %:
[16:53] <xnox>          dh $@ --with python3
[16:53] <zyga> what?
[16:53] <zyga> why two source packages?
[16:53] <xnox> done and it works across all precise -> raring in PPA.
[16:53] <zyga> I mean, duplicate debian/ directory or just another entry in control
[16:53] <infinity> cjwatson: Oh, wait, PROPOSED moved from being an environment twiddle in cron to an etc/config file thing, didn't it?  I live in the past.
[16:54] <xnox> well sed the top line in changelog, source & package names in control, & python2 -> python3 in debian/rules.
[16:54] <zyga> xnox: could you show me an exampl?
[16:54] <infinity> cjwatson: So, yeah, no one switched it, I was just mistakenly looking at crontab. :P
[16:54] <xnox> zyga: hehe =) or I can just upload pybuild backport into precise & quantal ;-)
[16:55] <cjwatson> infinity: it's in a couple of places still, I think
[16:56] <xnox> zyga: I am yet to figure out if pybuild lets me choose what shebang the scripts have (e.g. python2 or 3)
[16:57] <zyga> xnox: normally I generate scripts with setuptools
[16:57] <zyga> xnox: IIRC there were parts in dh_python that sed them to be right
[16:59] <infinity> cjwatson: Well, the two usual suspects, cdimage/etc/config and debian-cd/CONF.sh ... Except, curiously, the latter is commented out.
[16:59] <infinity> cjwatson: Have we only been half-proposed this whole time? :P
[16:59] <PrincessLuna> Are the multitouch patches for kernel 3.8 included in the rc5?
[17:01] <roadmr> ev: how to get access to stack traces on errors.ubuntu.com? (I get stuck in the infamous login loop)
[17:01] <cjwatson> infinity: oh, I fixed bug 1023991 and forgot about it
[17:01] <micahg> zyga: xnox: I'd be happy to process a backport request for pybuild if someone files the request with requestbackport and does the testing
[17:01] <cjwatson> stgraber fixed most of it, actually
[17:02] <cjwatson> infinity: I've cleaned up the bit in CONF.sh
[17:03] <ev> roadmr: you need to be in ~canonical-ubuntu-platform
[17:04] <cjwatson> infinity: so, I don't know.  my gut feel is that I'd still like a few more things in -updates to avoid impeding verification.
[17:04] <ev> roadmr: slangasek is working with legal on an NDA for this
[17:06] <roadmr> ev: I see.. so is it best to wait for that, or is there an interim process, like getting help from someone with access?
[17:06] <xnox> micahg: I'd like to first test and see that it does everything it says to do on the tin ;-)
[17:06] <roadmr> ev: I want to look at stack traces for auto-reported checkbox crashes, but I guess it's not urgent, after all there's still plenty of time to fix bugs
[17:07] <micahg> xnox: please do :)
[17:07] <infinity> cjwatson: Yeah, I'm in no massive rush to do switch flippery.  It just came up because pitti was chomping at the bit to copy langpacks, not realising we were still building against proposed and, thus, building with the new ones already.
[17:07] <ev> roadmr: there's no alternative process yet, but I can't imagine it taking much longer. slangasek will be able to say for sure. If it does look like it's some time off, we can add additional teams, but I'd like to run that by elmo if we're going to do it.
[17:08] <infinity> cjwatson: s/copy/promote/
[17:08] <roadmr> ev: ok, understood. I'll wait then :) thanks!
[17:09] <ev> sure thing
[17:09] <cjwatson> infinity: aye
[17:43] <slangasek> ev: it's still a ways off, so if there's a pressing need I would suggest working around it
[17:43] <ev> noted, thanks
[17:51]  * cjwatson gets down to fifteen remaining Ubuntu deltas in grub2 (and 4300+ lines of changelog diff due to all the "remaining changes" entries)
[17:51] <cjwatson> I wonder if I can get that to zero for 13.04
[18:04] <roaksoax> slangasek: howdy!! I ws wondering somthing about the MAAS stuff. So when we continue to include the yui3/raphael libraries within the source, wouldn't there be a conflict when sruing the same upstream version to both quantal and precise, because the source package differs from each other (quantal no yui3 libs, precise will have those libs), as what happens in PPAs?
[18:05] <roaksoax> s/source package/upstream tarball
[18:12] <slangasek> roaksoax: you could possibly add the yui3 and raphael bits into the debian packaging rather than the upstream tarball?  that might be easiest
[18:14] <roaksoax> slangasek: yeah that indeed seems to be the easiest
[18:28] <bdmurray> pitti: these nexus7 stacktraces don't seem very helpful bug 1112443
[18:30] <slangasek> that looks like the error I was seeing when first trying to get cross-retracing working
[18:53] <jtaylor> why does python-imaging have such a weird version number?
[18:54] <jtaylor> nevermind, says in the changelog
[19:51] <wjtaylor_> Is ubuntu associated with ubuntuupdates.org? Or is this a 3rd party collecting packages and forming ppas?
[19:54] <roadmr> wjtaylor_: read here: http://www.ubuntuupdates.org/about
[23:52] <infinity> ev: Any plans to upload a fixed whoopsie?