16:03 #startmeeting 16:03 Meeting started Thu Nov 7 16:03:34 2013 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 16:03 16:03 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 16:04 [TOPIC] Lightning round 16:04 $ echo $(shuf -e barry doko stgraber jodh bdmurray slangasek cjwatson xnox stokachu) 16:04 doko stgraber xnox cjwatson jodh barry stokachu slangasek bdmurray 16:05 doko_: please :) 16:07 - Linaro Connect 16:08 - visiting the client sprint on Friday 16:08 - some initial toolchain work 16:08 (done) 16:08 Merges: 16:08 - Done with most of my usual set, helping with a couple more 16:08 - Slowly processing the resolvconf one 16:08 LTSP hackfest (Wednesday afternoon till Sunday): 16:08 - Discussed the plans for LTSP 6 16:08 - Made some progress on libnss-sshauth and libpam-sshauth (modules allowing a machine to authenticate based on the result of an ssh connection to another machine) 16:08 - Worked with vagrant to try and reduce the Debian/Ubuntu delta 16:08 - Looked into possible new features for LTSP, mostly per-user session containers and checkpoint/restart support 16:08 LXC: 16:08 - Reworked the whole LXC CI environment, triggered once per hour if a commit is there, uploading to our PPA up to twice a day, test building on Ubuntu and Debian for amd64, i386 and armhf, also building with coverity and doing a clang based test build 16:08 - Various code reviews and bug fixes 16:08 - Helped debug a few 3.12 related issues 16:09 - Took a close look at lmctfy, trying to figure out a detailed plan for our own cgroup manager 16:09 System images: 16:09 - Introduced two new channels for a different set of customized images 16:09 - Changed the devel-* aliases to point to trusty (expect a full update next time) 16:09 Release: 16:09 - Tried to empty most of the queues as best I could, I think we're pretty much caught up on pending SRUs now at least 16:09 - Planned a bunch of sessions for vUDS (release, lxc, cgroup management) 16:09 Other: 16:09 - Been working on IPv6 patches for a few server related packages I use and that were failing quite badly in IPv6-only environment 16:09 16:09 TODO this week: 16:09 - resolveconf merge 16:09 - Get Android test builds in the LXC CI infrastructure 16:09 - Try to integrate Oracle RPM builds to the LXC CI infrastructure 16:09 - LXC work on container autostart 16:09 - Prepare for LXC 1.0~alpha2 16:09 (DONE) 16:10 * Transitions: bost1.49 removed, boost1.53 removal pending 16:10 mini-transition resolutions, db5.1 removal pending ubuntu-keyboard 16:10 release with fix to work with pinyin4. 16:10 * Many upstart code reviews done. 16:10 * Emulator: 16:10 - does have adbd/networking, and in the archive. 16:10 - rsalvetti/sergiusens is digging into GL issues (segfault when 16:10 accessing ThreadLocalStorage OpenGL pre-allocated slots :-/) at the 16:10 moment recompiling kernel & bionic and some such. 16:10 - From segfault we are up at -> error EGL_NOT_INITIALIZED 16:10 * Doing merges, from interesting ones crytpsetup done (brings 16:10 TrueCrypt support) and mdadm is still todo (also see mdadm fakeraid 16:10 vUDS session) 16:10 * Next week Debian Mini-conf UK Thursday-Sunday. 16:10 * TODO: - ubiquity code reviews (not done from last meeting, more came 16:10 xnox: >Many upstart code reviews done: thanks! 16:10 in) 16:10 (DONE) 16:11 Canonical client planning sprint. 16:11 Merged Brian's chroot management support into lp:click and got it working well enough for a sprint demo (which was well-received). Generally updated people outside Foundations on our plan for cross-building/emulation for app testing. 16:11 Merged ben from unstable and ported transition tracker to new version in trusty, gaining us Depends: foo:any support, better level calculation, and a better index. 16:11 Working on FusionIO disk detection in GRUB (bug 1237519). 16:11 bug 1237519 in grub2 (Ubuntu) "Grub2 fails to install to non-standard device path" [High,Triaged] https://launchpad.net/bugs/1237519 16:11 Some multiarch conversions to help with targeted cross-build situations. 16:11 LP adjustment to rely only on one other machine for running germinate rather than two (https://code.launchpad.net/~cjwatson/launchpad/configurable-germinate-base/+merge/194340). 16:11 To do: finish FusionIO work, catch up on other GRUB OEM bugs, parallelise germinate across architectures in publisher, release new click version to trusty 16:11 .. 16:11 * upstart 16:11 - Reworked lp:~jamesodhunt/upstart/bug-1238078 (now merged). 16:11 - Merged lp:~jamesodhunt/upstart/bug-1235649 16:11 - Discussions with CI re Upstart DEP-8 testing. 16:11 - Wrote script to detect new Upstart jobs hitting the archive 16:11 oh, I also showed off the emulator in the sprint demo, albeit with a disclaimer about lack of graphics :-) 16:11 (with a view to reviewing jobs): 16:12 http://people.canonical.com/~jhunt/upstart/scripts/check_for_new_upstart_jobs.sh 16:12 Now, just need to find a way to automatically detect changes to 16:12 upstart jobs without having to parse trusty-changes. Ideas? 16:12 - Overcoming ptrace limitations: Fixed 2 bugs in preliminary branch. 16:12 Currently working adding additional ptrace tests to ensure expected behaviour. 16:12 ʝ 16:12 system-image: LP: #1217098. system-image 2.0 testing (looking good - waiting on LP: #1215586 for u/i and landing slot). nexus 10 installation and testing. 16:12 Launchpad bug 1217098 in Ubuntu system image "Avoid re-downloading data files" [High,Fix committed] https://launchpad.net/bugs/1217098 16:12 Launchpad bug 1215586 in Ubuntu system image "The u/i cannot handle the array-of-dictionaries update descriptions" [High,In progress] https://launchpad.net/bugs/1215586 16:12 vUDS: core-1311-python3-roadmap. re-engaging with xapian upstream (lots of progress there recently). started updating python3 porting plan for 14.04 (spreadsheets, wiki). http://tinyurl.com/mtf9jrb 16:12 other: LP: #1248417. started looking at LP: #1165104. python-coverage sponsorship in debian. converted enum34 (0.9.19-1) to pybuild and updated LibraryStyleGuide for pybuild. PEP 8 updates. dmb meeting. 16:12 Launchpad bug 1248417 in python3-defaults (Ubuntu) "Sync python3-defaults 3.3.2-17 (main) from Debian unstable (main)" [Undecided,Fix released] https://launchpad.net/bugs/1248417 16:12 done 16:12 Launchpad bug 1165104 in oneconf (Ubuntu) "oneconf is only showing the pc you are on in raring and isn't sharing to other machines" [High,Triaged] https://launchpad.net/bugs/1165104 16:13 jodh: i thought we had deduplication running somewhere and possibly ability to fetch md5sums of files. 16:14 jodh: as it md5sums of all /etc/init/* 16:14 xnox: sounds great - just define somewhere! :) 16:14 jodh: conditional branch is out of range 16:15 xnox: aha; the backtrace sergiusens first shared did look like it was pointing to a TLS problem, you've confirmed now and are making headway? Is there anything I can do to help with that, or is the debugging env sufficiently arcane that I should just stay clear? 16:15 xnox: ? 16:17 slangasek: there is progress, needs recompiled bionic. I didn't try it yet myself. serge and rsalveti are on it.... they did tweak TLS slots before to make bionic work at all. 16:17 not bionic, for hybris to work. 16:17 xnox: ahh, ok 16:19 jodh: without parsing trusty-changes> grab Contents-i386.gz for the list of all packages containing jobs, and download Packages.gz to get the list of current versions; at interval of your choice, re-download the Contents and Packages; for each newly-introduced job file, or for each package version that's changed, download and compare 16:20 slangasek: ack - thanks! 16:20 note that Contents is only updated daily as it's pretty expensive to compute 16:20 xnox: right, TLS for bionic + hybris is something special... the work they did there is actually quite scary ;) 16:20 cjwatson: ack 16:21 xnox: hmm, I wonder if that work could actually be interfering with native thread support on the bionic side? Since most of it *did* involve "neutering threads" 16:21 * Client sprint last week, lots of discussions... won't try to give you a systematic update here, you'll certainly see the discussions reflected in the UDS sessions coming up in two weeks 16:21 * good discussions about how to support go on the client, which we will see more of this cycle 16:21 * dates set for the Ubuntu Core sprint, please get your travel approvals in :) 16:22 * helped with upstart MP reviews, we finally have the queue under control (mostly due to xnox!) and will see a new upstream release soon 16:22 * upstart discussions in Debian: the question of default init system has now gone to the Debian Technical Committee 16:22 * was out sick Tuesday - too much travel is bad for the body 16:22 * upcoming: 16:22 * Monday is Veterans' Day 16:22 * I'm up for jury duty on Tuesday (and possibly longer) 16:22 * looking at the emulator, and possibly even helping 16:22 (done) 16:22 ubuntu-release-upgrader bug triage 16:22 set up ubuntu security team virtual machine environment for bug testing 16:22 SRU verification of bug 459418 16:22 SRU verification of update-manager bug 1071057 16:22 tested slangasek's fix for bug 1069019 (didn't seem to work) 16:22 bdmurray: 16:22 uploaded and tested SRU fixes for unzip bug 580961 and bug 1199239 16:22 bug 459418 in lazr.restfulclient (Ubuntu Quantal) "Cache is not safe for concurrent use (by processes or threads)" [High,Fix released] https://launchpad.net/bugs/459418 16:22 oops :) 16:22 bug 1071057 in update-manager (Ubuntu Saucy) "people report bugs about update-manager thinking it still does dist upgrades" [Medium,Fix committed] https://launchpad.net/bugs/1071057 16:22 bug 1069019 in python-apt (Ubuntu Saucy) "[software-properties-gtk] can not delete, enable or modify any software source with non-ASCII characters in the comment" [Medium,Triaged] https://launchpad.net/bugs/1069019 16:22 bug 580961 in unzip (Ubuntu) "unzip fails to deal correctly with filename encodings" [Critical,Triaged] https://launchpad.net/bugs/580961 16:22 bug 1199239 in unzip (Ubuntu Raring) "[SRU] unzip list utf-8 (non-ascii) filenames as ??" [High,In progress] https://launchpad.net/bugs/1199239 16:22 investigation into apt bug 957231 / bug 854090 16:22 investigation into bug 984785 regarding glib2-0 (hot bug) 16:22 irc discussion with pitti regarding bug 1084979 16:22 investigation into foundations-bugs currently receiving the most duplicates 16:22 bug 957231 in apt (Ubuntu Trusty) "update-manager crashed with SIGSEGV in debListParser::LoadReleaseInfo()" [Medium,Triaged] https://launchpad.net/bugs/957231 16:22 submitted a merge proposal for ubiquity fixing bug 1051935 16:22 bug 854090 in apt (Ubuntu) "aptd crashed with SIGSEGV in debListParser::LoadReleaseInfo()" [High,Confirmed] https://launchpad.net/bugs/854090 16:22 bug 984785 in glib2.0 (Ubuntu Precise) ".goutputstream files polluting $HOME" [Low,Triaged] https://launchpad.net/bugs/984785 16:22 bug 1084979 in apport (Ubuntu) "Submitting error report asks confounding questions" [Medium,Triaged] https://launchpad.net/bugs/1084979 16:23 bug 1051935 in ubiquity (Ubuntu Saucy) "Fails with SystemError when too many files are open" [Undecided,Confirmed] https://launchpad.net/bugs/1051935 16:23 phased-updater work: 16:23 research into contact person issue with packages sync'ed from a staging ppa (e.g. hud) 16:23 updated phased-updater to add in Reply-To header and also CC me 16:23 xnox: have your fixes for upstart ppa recipe builds landed? Because I still get an awful lot of build failure spam :) 16:23 errors work: 16:23 submitted RT regarding updating errors and daisy revisions 16:23 submitted a fix for bug 1191182 (regarding version appending when selecting a package) 16:23 bug 1191182 in Errors "choosing a period on the main page appends a version to the query" [High,In progress] https://launchpad.net/bugs/1191182 16:23 reported and submitted a fix for errors bug 1248259 (package version drop down not displayed) 16:23 bug 1248259 in Errors "package_versions drop down not displayed if you use the &package url parameter" [Medium,In progress] https://launchpad.net/bugs/1248259 16:23 triaged errors bug 1219706 16:23 bug 1219706 in Errors "Follow state of bugs marked as duplicates" [Medium,Confirmed] https://launchpad.net/bugs/1219706 16:23 investigation into errors bug 1218670 regarding date range calendar 16:23 bug 1218670 in Errors "date range calendar does not work in firefox" [Medium,Confirmed] https://launchpad.net/bugs/1218670 16:23 pushed a daisy branch to enable saucy-proposed in the retracers 16:23 planned rls:pkg:pkg_version counter changes 16:23 ✔ done 16:25 any questions over the above? 16:26 jodh: and nicely done, taking my ptrace branch off the floor and making it un-broken :) Will be so nice to have that bug finally fixed 16:26 slangasek, did I miss any message about the sprint? 16:26 slangasek: thanks - most of the credit goes to you though :) 16:26 doko_: what kind of message? were you cc:ed on msm's mail? 16:27 17025 N T Nov 05 Michelle Surtee PLEASE READ: Ubuntu Core sprint - 27 - 31 Jan 2014 - London, Bluefin 16:27 (no comments on my lack of inbox zero please :-P) 16:28 slangasek: almost, there was broken commit on trunk. next set should be fine. 16:28 cjwatson: wow, that makes me feel better 16:28 slangasek, can't find it. forward would be appreciated 16:28 xnox: whoops. whose commit was broken? 16:29 doko_: does 'doko@canonical.com' reach you? 16:29 should 16:29 doko_: forwarded 16:31 jodh, xnox: does it actually make sense to do these upstart recipe builds across the set of releases we currently build for? (raring,saucy,trusty?) 16:31 slangasek: integration issue. trunk & MPs passed everything, the combination of the two with new automake in trusty did not, however. 16:31 ah 16:31 ok, cool :) 16:31 [TOPIC] AOB 16:31 slangasek: yes it does. I want to go back all the way to precise actually. 16:31 anything else? 16:32 now's a great time to discuss 16:32 xnox: hmm. I think we should build for precise, but not for raring 16:32 slangasek: ok. 16:33 slangasek: i'll check that it still works with quantal, then step down to precise, and then disable quantal & raring dailies. 16:33 ok 16:34 are these recipe builds on devirt? 16:35 (either way, it's good to be economical with builder time, although the virt pool is worse off at the moment and probably will be until we get scalingstack) 16:36 cjwatson: well upstart is ~8minute build. 16:36 we at least have one devirt PPA with builds, not sure if we still also have a virt one 16:36 xnox: sure, it does add up though 16:36 cjwatson: and armhf devirt has plenty of builders. 16:36 and it's easier to argue against builds that are genuinely abusive when we're more consistent :) 16:37 I agree it's not the worst problem by any means 16:37 ok. 16:39 alrighty, guess we've exhausted our capacity for conversation today ;) 16:39 #endmeeting