16:00:59 #startmeeting 16:00:59 Meeting started Wed Dec 19 16:00:59 2012 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 16:00:59 16:00:59 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 16:01:14 * stgraber waves 16:01:40 [TOPIC] Lightning round 16:01:47 $ echo $(shuf -e barry doko stgraber jodh ev bdmurray slangasek ogra cjwatson xnox stokachu) 16:01:50 ogra stgraber barry cjwatson xnox jodh doko stokachu slangasek bdmurray ev 16:01:55 win! 16:02:00 ogra's busy sunning himself on a ski slope somewhere, so stgraber 16:02:06 - Upstart 16:02:06 - 3 days sprint (Monday -> Wednesday) 16:02:06 - Got prctl branch merged. 16:02:06 - libnih 'as' type fix merged and uploaded. 16:02:06 - Various fixes to the tests to allow for reliable daily builds on virt and non-virt builders. 16:02:09 - Got upstart-dbus-events merged. 16:02:12 - Wrote new upstart-event-bridge, currently waiting for review/merge. 16:02:14 - Container 16:02:17 - Usual code reviews. 16:02:19 - Release 0.9~alpha1 (and 0.9~alpha2 because of some communication problems ;)) 16:02:22 - Built a custom kernel to test for the uevent fix when network devices are moved to other namespaces. 16:02:25 - Sent fix for LXC on nexus7 to the kernel team. 16:02:28 - Started working on some changes to LXC upstream to allow building with alternate libc implementation (testing with bionic). 16:02:31 - Other 16:02:33 - Spent some time debugging a rather odd biosdevname issue, looks like it was a firmware issue on the specific machine. 16:02:36 - Did some UEFI SecureBoot tests, running a locally built and signed shim. 16:02:39 - Patch pilot on Friday. 16:02:42 - TODO 16:02:44 - Implement user socket for upstart session (to be used before dbus is started). 16:02:47 - Cleanup and send a bunch of lxc patches for review (split libcap, bionic support, ...). 16:02:50 - Go through the isc-dhcp and resolvconf bugs. 16:02:53 - Try to apply redhat's fiberchannel patchset to our isc-dhcp. 16:02:55 - Holidays 16:02:58 - I'm working tomorrow (European time) and then will be off until the 2nd of January. 16:03:01 (DONE) 16:03:28 debian bug 695707 (virtualenv broken, with fixes for python2.7 #695958 and 3.3 #695959 proposed). emacs bug 12765. worked on whoosh for py3 (needs upstream fixes). debugging friends service crashes for internal customer. done. 16:03:31 Debian bug 695707 in libpython2.7-stdlib "[libpython2.7-stdlib] Breaks python-virtualenv" [Normal,Open] http://bugs.debian.org/695707 16:03:32 bug 12765 in nautilus (Ubuntu) "Use full resolution thumbnails on desktop" [Wishlist,Fix released] https://launchpad.net/bugs/12765 16:03:57 Mostly on leave and/or in a virus-induced haze this week. On leave for the rest of the year starting Thursday afternoon. 16:04:00 Fixed incorrect colours in d-i images (bug 1089331). 16:04:02 bug 1089331 in cdebconf (Ubuntu) "ubuntu server d-i installer is blue, not purple" [Medium,Fix released] https://launchpad.net/bugs/1089331 16:04:03 Test-built all libssl-dev reverse-build-dependencies with a couple of different candidate patches for making libssl-dev Multi-Arch: same: 16:04:06 - Moving all headers to /usr/include/$(DEB_HOST_MULTIARCH) caused far too many failures (67). 16:04:10 - Moving just causes two failures in unstable, both of which are easily tractable. I've filed bugs with patches for both. 16:04:13 Working again on britney/autopkgtest integration, with Jean-Baptiste's help. Not sure if I'll have something by EOY but we'll see. 16:04:17 Working on powerpc image changes requested by Ben Collins. 16:04:19 .. 16:04:53 * upstart sprint: 16:04:53 - finish xdg config dirs branch, now merged 16:04:53 - progressing on overrides from any config dir 16:04:53 - fixing unit tests & daily builds / adt 16:04:53 - refactor fixes from lp:ubuntu/libnih & propose them into lp:libnih 16:04:54 Sprint progress at http://pad.ubuntu.com/j5DPzIp5KP 16:04:56 * started cross-building fixes spree (3.5 so far, but it's fun) 16:05:00 .. 16:05:05 * Upstart Sprint: http://pad.ubuntu.com/j5DPzIp5KP 16:05:05 ✄ 16:05:21 surely that's cheating 16:05:45 different area codes so its ok 16:05:52 * jodh rummages for generate-random-buzzwords.pl... 16:05:56 at least thats what ive been told 16:06:01 - updated armhf cross toolchain 16:06:01 - built arm64 and powerpc biarch cross toolchains 16:06:01 - rewritten gcc control.m4 to make it easier to use runtime libs for the target arch, both for 4.7 NS 4.8 16:06:01 - fixing gcc-4.8 (upstream) build and install issues. 16:06:01 - GCC 4.8 ppa now ready for the test rebuild 16:06:04 (done) 16:06:14 - dput uploaded (bug 633109) needs approval 16:06:15 - packaged firefox-esr 10/17 (this took up all of my time) 16:06:15 - applied for Contributing Developer http://wiki.ubuntu.com/AdamStokes (testimonies welcomed :D) 16:06:15 - Started backport process for sssd/sudo to Precise because of the new sudo ldap+caching features 16:06:18 introduced. Still going through the rdeps list for sudo to see what needs to be tested. 16:06:19 bug 633109 in dput (Ubuntu Precise) "No progress bar for sftp uploads" [Undecided,New] https://launchpad.net/bugs/633109 16:06:21 - I expect high visibility bugs from me to pick back up at beginning of new year. 16:06:24 16:06:32 .. 16:09:25 stokachu: hmm, you're only applying for contributing developer? 16:09:35 I thought this was for core dev 16:09:39 slangasek: micah recommended i did that 16:09:50 but contributing developer gives you no upload rights :) 16:10:43 slangasek: what membership board should i apply to 16:10:57 * xnox thought contributing is a bit on the low side. Motu or core-dev is better. 16:10:59 well, we can harrass micahg (and stgraber) about this afterwards 16:11:07 stokachu: the membership board is the DMB in all cases 16:11:11 and ogra and barry =) 16:11:34 slangasek: ok cool just point me in the right direction b/c i wasn't totally sure 16:11:47 slangasek: it was initially posted to the main membership board 16:11:51 stokachu: or ppu 16:11:55 so what do you get by being a contributing developer if you don't get upload rights? A nice fruit basket? 16:11:59 (per-package upload rights) 16:12:10 but it doesn't seem optimal to me to go through the process for contributing developer if you're going to be applying for core-dev shortly regardless 16:12:15 ev: Ubuntu Membership 16:12:19 fruit basket optional 16:12:23 ... 16:12:41 slangasek: looks like i can just add my name to the core dev in the agenda 16:12:43 oh dear 16:12:46 applying is the same it looks like 16:13:03 stokachu: different secret handshake 16:13:04 stokachu: yeah process is the same for all developer levels. 16:13:08 stokachu: yeah, same board, different agenda item. Just move your application to coredev. 16:13:41 stgraber: do i need to resend out an email to the list? 16:13:44 stokachu: yes; though we should figure out why micahg is still steering you away from core-dev, since he gets to vote on the question of whether you get in ;) 16:13:46 my original one said core dev 16:13:49 stokachu: coredev will give you ubuntu membership automatically, so no need to first apply for contributing developer. 16:14:26 stokachu: just reply to your e-mail to devel-permissions saying that you instead want to be considered for coredev. 16:14:28 he prepares harder questions for coredev applicants :) 16:14:29 stokachu: if informal polling of the DMB suggests you're not ready for core-dev, you probably want to find out what more they want to see from you, and focus on that so you don't have to go through the process twice 16:14:49 ok 16:15:11 stokachu: I haven't looked at your recent history, but last I checked the only thing that was missing was the "sustained" part which has definitely be solved since ;) 16:15:20 I guess it's my turn to spam the channel 16:15:22 (as I remember you last poking me 6 months or so ago) 16:15:24 * xnox did go contributing -> core-dev jump. But I did apply for contributing dev way back when i was "green" 16:15:30 * Following up on Samsung laptop EFI bricking; looking like we may want to disable samsung-laptop module as a workaround 16:15:33 * UEFI Secure Boot updates. Latest upstream version of gnu-efi+shim give a black screen and a reboot. Still tracking down. 16:15:36 * Doing my "fix 5 package cross-builds" work item; for one of them, fixed valac for use as a cross-build-dep 16:15:39 * Working on gobject-introspection porting to python3, possibly tackling the cross-build-dep problem after that 16:15:42 (done) 16:16:12 ev: 16:16:17 oh hi 16:16:23 slangasek: did you have a look to see whether any of the packages in question improved in the latest cross-builder run? 16:16:28 - Going on holiday tomorrow. I'll be back in January. Have a wonderful holiday 16:16:28 and New Year! 16:16:29 - The retracers have been moved to autorestart hell, as they were bothering 16:16:29 webops too much on the weekends. This was mostly because a Cassandra node was 16:16:29 taken down, which was causing timeouts. Still, we'll need to make the 16:16:29 retracers more resilient to database failure if we want to get them back into 16:16:29 a normal service operation. 16:16:36 - The queue is still massive (~6-8k per arch). We've started to remove some 16:16:36 cores from it and Tom is working on getting Rabbit queue size hooked up to 16:16:36 graphite (http://graphite.engineering.ubuntu.com). Trying to get everything 16:16:36 ready for prodstack as quickly as possible to better remedy this. 16:16:36 - Spent much of the week getting our charms and daisy in shape for deployment 16:16:36 to prodstack, following the advice that Tom gave in RT 58019. 16:16:41 - Fought gunicorn to the death. Discovered the hard way that gunicorn from 16:16:41 precise is busted. Backported from Quantal. 16:16:41 - Refactored our separate wsgi scripts for receiving crash reports and core 16:16:41 dumps, unifying them under a single wsgi service and merging in the S3 16:16:44 cjwatson: I didn't save the 'before' view :) 16:16:46 support along the way. This makes running them under the gunicorn 16:16:47 subordinate charm a lot easier. 16:16:47 - Taught the daisy charm to handle multiple Cassandra nodes coming and going. 16:16:51 - Daisy is *almost* ready. I just need to wire the Apache frontends to the 16:16:51 HAProxy instances. Everything else seems to be working. Getting the 16:16:54 retracer and Errors charms updated should be largely a matter of copying 16:16:54 code around. 16:16:54 - Worked a bit with Brian to dig at why getBugSubscriberPackages was not 16:16:54 returning anything. Brian ultimately tracked down the root cause. 16:16:56 (done) 16:17:15 slangasek: or "succeeded", even 16:17:16 err graphite.engineering.canonical.com 16:18:12 any questions on status? 16:18:45 [TOPIC] AOB 16:18:56 anyone have any plans for the end of the world on Friday? 16:19:27 I was planning to go to the pub tomorrow evening, if that counts :P 16:19:43 https://plus.google.com/u/1/103242523324036094862/posts/AKBdNSnyrv5 16:19:59 +1 16:20:11 lol 16:20:16 :-) 16:20:39 * xnox is flying the day after, so I will be packing and doing last minute shopping. 16:20:40 :) 16:20:54 as long as we dont get an emp from a solar flare and i can still get on the internet im good 16:21:06 ok, sounds like that's it for the year then :) 16:21:18 happy holidays, everyone! don't drink too much eggnog 16:21:26 (drink just the right amount of eggnog) 16:21:28 #endmeeting