16:02 <slangasek> #startmeeting
16:02 <meetingology> Meeting started Thu Dec 19 16:02:16 2013 UTC.  The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
16:02 <meetingology> 
16:02 <meetingology> Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired
16:03 <slangasek> [TOPIC] Lightning round
16:03 <slangasek> $ echo $(shuf -e barry doko stgraber jodh bdmurray slangasek cjwatson xnox stokachu)
16:03 <slangasek> doko stokachu xnox jodh slangasek stgraber barry bdmurray cjwatson
16:04 <slangasek> doko: you first :)
16:04 <doko> - configury fixes for a bootstrap build
16:04 <doko> - openjdk-7 disable hotspot on i386, alignment issue with glibc 2.18.
16:04 <doko> - port openjdk zero to ppc64el
16:04 <doko> - update llvm to 3.4, mir b-d's, build mesa with 3.4.
16:04 <doko> - build fixes for the indicator stack to build update-notifier
16:04 <doko> - gcc-snapshot update
16:05 <doko> - and fixing hardening-wrapper
16:05 <doko> - gdb updated to the trunk
16:05 <doko> /done
16:05 <slangasek> oh, gdb, great :)
16:06 <xnox> i assume stokachu is still out
16:06 <slangasek> yes
16:06 <xnox> * uploaded initial cmake cross-compilation support (positive response
16:06 <xnox> and constructive further plans discussions from that)
16:06 <xnox> * fixed-up some of the cross-compilations issues in packages using
16:06 <xnox> cmake (e.g. executing compiled binaries, when cross-compiling)
16:06 <xnox> * fixed-up broken toolchain-base packages for arm* (due to eglibc 2.17
16:06 <xnox> -> 2.18 bump) p*pc* are still to do
16:06 <xnox> * uploaded changes to android & autopilot to resolve executing
16:06 <xnox> autopilot tests under the emulator. At request from evan, tested
16:06 <slangasek> (and he hasn't designated a surrogate, so)
16:06 <xnox> that a single host can execute multiple emulators, still need to
16:06 <xnox> check if it can also be done in a VM, without proper graphics card.
16:06 <xnox> * still to do in the emulator - unlock higher RAM memory, currently
16:06 <xnox> capped at 760MB due to memory map issues (kernel panics, segfaults
16:06 <xnox> with higher values.)
16:06 <xnox> * Running autopilot tests documented in this part of the wiki
16:06 <xnox> https://wiki.ubuntu.com/Touch/Emulator#Using_the_emulator
16:06 <xnox> * merged boost, and split python-pyste to demote gccxml to universe.
16:06 <xnox> * still to do resolve device name of the "generic", for system-image
16:06 <xnox> updates in the emulator
16:06 <xnox> ..
16:07 <jodh`> * upstart: cgroup:
16:07 <jodh`> - Continued work on cgroup support. Improved the internal
16:07 <jodh`> representation of a cgroup (currently updating tests to reflect
16:07 <jodh`> this).
16:07 <jodh`> - cgmanager discussions with hallyn+stgraber.
16:07 <jodh`> * misc: short week - out yesterday.
16:07 <jodh`> ס֙
16:07 <slangasek> xnox: emulator \o/  but hmm, I guess the question isn't whether we can run them in a VM, but whether they will perform well enough to be worth anything
16:08 <barry> slangasek: seems like boot (maybe disk access?) is quite slow, but cli performance doesn't seem too bad once it's booted
16:09 <xnox> slangasek: it's both. using juju / openstack instances is preferred over having a physical bare-metal box contention.
16:10 <xnox> slangasek: and performance.... cpu emulation is still slow and there is no obvious way to crank up emulated cpu speed, unless we start using generic-x86 builds with accelerated kvm style emulation.
16:10 <xnox> slangasek: which will take us into the nested kvm bugs, once again ;-)
16:10 <xnox> jodh`: has nested kvm been resolved at all, in the case of upstart tests?
16:10 <slangasek> yes, yes, it would
16:10 <slangasek> xnox: not in the least
16:11 <slangasek> barry: cli performance is not what we are testing :)
16:11 <xnox> slangasek: excellent, the more i do the more there is to do.....
16:11 <slangasek> * recruiting for the Foundations Team opening
16:11 <slangasek> * bootstrapping/porting work
16:11 <slangasek> * release engineering for alpha 1
16:11 <slangasek> * fix libfreetype-dev dependencies so things using freetype-config don't FTBFS
16:11 <barry> slangasek: yeah, i know
16:11 <slangasek> (done)
16:12 <stgraber> Done:
16:12 <stgraber> - CGroup Manager
16:12 <stgraber> - Helped with design work.
16:12 <stgraber> - LXC
16:12 <stgraber> - Release 1.0~beta1 on Tuesday, updated our packaging and uploaded
16:12 <stgraber> - Finished implementing the container autostart specification (patched up for review now)
16:12 <slangasek> specifically: we're exercising autopilot tests, which run the gui
16:12 <stgraber> - Finished work on the configure-overview branch and landed upstream.
16:12 <stgraber> - Tried to debug and ended up working around an issue with clang-3.4 (reverted to 3.2 for now as it looked like a compiler/parser bug).
16:12 <stgraber> - The usual set of code reviews
16:12 <stgraber> - Ubuntu touch
16:12 <stgraber> - Enabled discard use by default (once initamfs-tools-ubuntu-touch manages to build...)
16:12 <stgraber> - Fixed a rather critical bug in the upgrader code (factory reset would also wipe the system)
16:12 <stgraber> - Other
16:12 <stgraber> - SRU and New reviews
16:12 <stgraber> - A couple of merges
16:12 <stgraber> - TODO tomorrow: OpenVPN packaging fixups, a couple more minor LXC changes to work better with unpriv containers and maybe some networking work if time allows
16:12 <stgraber> - Holidays
16:12 <stgraber> - I was off on yesterday
16:12 <stgraber> - I'll be working tomorrow, then off till the 2nd
16:12 <stgraber> (DONE)
16:13 <barry> system-image: LP: #1256229, LP: #1260712.  started looking at system-image and autopilot on emulator; filed some emulator and u-d-m bugs with issues found - with those fixed, we're probably not too far off from system-image working in emulator.
16:13 <ubottu> Launchpad bug 1256229 in Ubuntu system image ""system-image-cli -i" reports old version" [Undecided,Invalid] https://launchpad.net/bugs/1256229
16:13 <ubottu> Launchpad bug 1260712 in system-image (Ubuntu) "Apply update failed: Can't apply the current update (can't contact service)"" [Undecided,Fix released] https://launchpad.net/bugs/1260712
16:13 <barry> ubu/deb: genshi 0.7-1 (added py3 support), pytest 2.5.0-1 (blocked on dep), pecan 0.4.2-1 (in progress), debian bug 732038 (singledispatch), twisted-py3 13.2.0-0ubuntu1, alacart 3.10-1-1ubuntu1 (removed b-d and thus main-inclusion on python-support), claws-mail 3.9.3-1ubuntu1, debian bug 714896 (py3 reportbug - in progress).
16:13 <ubottu> Debian bug 732038 in wnpp "ITP: singledispatch -- single-dispatch generic functions for Python" [Wishlist,Open] http://bugs.debian.org/732038
16:13 <ubottu> Debian bug 714896 in reportbug "reportbug: port to python3" [Wishlist,Open] http://bugs.debian.org/714896
16:13 <barry> (done)
16:14 <bdmurray> bug triage of dpkg already installed and configured bugs
16:14 <bdmurray> investigation into using apt-cdrom ident information to confirm validity of installation media
16:14 <bdmurray> uploaded SRUs for usb-creator bug 915626
16:14 <ubottu> bug 915626 in usb-creator (Ubuntu Saucy) "usb-creator-gtk crashed with SIGSEGV in _dbus_watch_invalidate" [High,Fix committed] https://launchpad.net/bugs/915626
16:14 <bdmurray> SRU verification of bug 1250679, bug 1258639
16:14 <ubottu> bug 1250679 in apport (Ubuntu Saucy) "crash files from guest users considered system reports" [Medium,Fix committed] https://launchpad.net/bugs/1250679
16:14 <ubottu> bug 1258639 in ubuntu-release-upgrader (Ubuntu Saucy) "need to support upgrades from 12.10 to 13.10" [High,Fix committed] https://launchpad.net/bugs/1258639
16:14 <xnox> barry: yeah, IO is horribly, emulated SDCARD speeds. there is even code to emulated / delay it further. i need to make sure it is not delayed. But i think, the IO speed is lightning times faster than the emulated CPU speed, so actually unrealisticly performed IO is emulated for the given CPU, I think =)
16:14 <bdmurray> sorted out an issue with the release upgrader tarball building
16:14 <bdmurray> resolved an issue with the creation of demoted.cfg files in the ubuntu-release-upgrader dist-upgrade tarball
16:14 <bdmurray> uploaded new version of ubuntu-release-upgrader to the saucy proposed queue
16:14 <bdmurray> tested whoopsie changes for detecting .crash files and writing to them
16:14 <bdmurray> discussed whoopsie permissions issue with slangasek
16:14 <bdmurray> tested changes to apport-noui upstart job
16:14 <bdmurray> wrote some code to make it easier to associate packages to teams
16:14 <bdmurray> subscribed teams to packages
16:14 <bdmurray> fixed an issue with team bug notifier
16:14 <bdmurray> wrote a tool to make reviewing pending SRUs easier
16:14 <xnox> s/performed/performant/
16:14 <bdmurray> investigation into retracing issues with daisy and missing oopses
16:14 <bdmurray> submitted merge proposal requesting an increase to thrift sizes
16:14 <bdmurray> test and modification of back population for counter for release:src_pkg:src_pkg_version
16:14 <bdmurray> done
16:16 <stgraber> doko: if you're interested by what may be a llvm bug when building LXC under clang I can poke you about it afterwards. (It looks to me like it's wrongly reporting unused functions)
16:16 <doko> stgraber, clang is in universe =)
16:17 <stgraber> doko: sure and we still use and recommend gcc by default, but I like to have our QA setup build on more than one compiler just to make sure we're not doing anything too gcc-specific in there
16:18 <doko> stgraber, yes, but ENOTIME
16:18 <slangasek> and no cjwatson today, so that's all of us
16:18 <slangasek> any questions?
16:18 <xnox> slangasek: when is the next meeting?
16:18 <slangasek> xnox: January!
16:19 <slangasek> January 8?
16:19 <stgraber> doko: ok :) For now I've just hardcoded clang-3.2 in our test suite and that's done the trick, hopefully someone who actually cares enough will debug and fix that problem at some point ;)
16:19 <slangasek> no, January 2
16:19 <doko> 2015?
16:19 <doko> stgraber, did you check 3.4?
16:19 <slangasek> doko: I hope you don't count array indices the way you count years ;)
16:20 <xnox> slangasek: is there much point in 2nd jan meeting? i guess it will be a rather quick one anyway.
16:20 <slangasek> xnox: hmmmm
16:20 <bdmurray> doko: could you have a look at bug 1261604?
16:20 <ubottu> bug 1261604 in gcc-4.8 (Ubuntu) "static binaries compiled with gccgo can fail to set up their address space" [Undecided,New] https://launchpad.net/bugs/1261604
16:20 <slangasek> xnox: good point, we can skip it
16:21 <xnox> slangasek: endianess is ok, so it's almost the right count =)
16:21 <slangasek> xnox: let's still do the mumble chat though :)
16:21 <xnox> ack.
16:21 <stgraber> doko: yep, 3.2 and 3.3 work, 3.4 doesn't
16:21 <doko> bdmurray, yes, next year
16:21 <slangasek> [TOPIC] AOB
16:22 <slangasek> anything else?  Christmas shopping recommendations?  Eggnog recipes?
16:22 <barry> slangasek: for all the pythonistas out there: http://www.vat19.com/dvds/26-pound-edible-gummy-python-candy.cfm
16:23 <slangasek> oh my
16:23 <jodh`> love that calorie count!
16:23 * barry feels his breathing constricted just thinking about it
16:23 <slangasek> it doesn't list calories per serving?
16:24 <slangasek> haha, http://images1.vat19.com/party-python/gummy-python-8-feet-long.jpg
16:24 <xnox> barry: 7 foot, i would have thought a 3.4 foot long variant is needed
16:24 <barry> xnox: metric
16:24 <barry> :)
16:24 <jodh`> slangasek: only per snake :)
16:24 * barry wonders how many pig hooves go into something like that
16:25 <xnox> bloke with a steak knife eating it is hallarious pic!
16:25 <xnox> "Thankfully, the gigantic gummy Party Python has a shelf life of a year."
16:26 <barry> they also have the world largest gummy bears
16:26 <barry> and gummy bears on a stick (everything is better when it's on a stick)
16:26 <xnox> Question: Don't you think this is a bit absurd for a piece of candy?
16:26 <xnox> Answer: You must be new to our website. You need to check out this, this, this, this, this, and this. Oh, and this. While we're at it, check out this. It's not candy, but it is absurdly large.
16:26 <xnox> with "this" all URLs =)
16:27 <slangasek> #endmeeting