== Meeting information == * #ubuntu-meeting Meeting, 19 Mar at 15:02 — 15:22 UTC * Full logs at [[http://ubottu.com/meetingology/logs/ubuntu-meeting/2015/ubuntu-meeting.2015-03-19-15.02.log.html]] == Meeting summary == === Lightning round === The discussion about "Lightning round" started at 15:02. === AOB === The discussion about "AOB" started at 15:17. == Vote results == == Done items == * (none) == People present (lines said) == * slangasek (30) * mvo (23) * bdmurray (19) * robru (15) * jodh (15) * infinity (13) * stgraber (13) * ubottu (12) * caribou (12) * cyphermox (11) * doko (9) * xnox (6) * barry (5) * meetingology (3) == Full Log == 15:02 #startmeeting 15:02 Meeting started Thu Mar 19 15:02:08 2015 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 15:02 15:02 Available commands: action commands idea info link nick 15:02 [TOPIC] Lightning round 15:02 $ echo $(shuf -e barry doko stgraber jodh bdmurray slangasek caribou infinity mvo sil2100 robru cyphermox) 15:02 bdmurray cyphermox doko sil2100 jodh mvo stgraber barry slangasek infinity caribou robru 15:02 bdmurray: tag 15:02 investigation into daisy bug LP: #1431796 (only one thread in ThreadStacktace) 15:02 Launchpad bug 1431796 in Errors "Only one thread shown in Thread Stacktrace" [High,Fix released] https://launchpad.net/bugs/1431796 15:02 fixed errors bug LP: #1431796 (not displaying threadstacktrace) 15:02 resolved issues with whoopsie-preferences upstream branch (out of date and changes not uploaded to vivid) 15:02 tested gunicorn change of worker_class from gevent to sync to resolve too many interim responses issue 15:03 updated daisy-charm with worker_class change for gunicorn 15:03 submitted RT to have daisy charm updated so we can test worker_class change in staging 15:03 tested / worked with webops regarding gunicorn worker_class changes (seems good) 15:03 fixed an issue with the way daisy returned EoL error message and worker_class = sync 15:03 searched for strange stacktrace endings 15:03 tested DayBucketProposedCounts for recording crashes from -proposed users 15:03 updated oops-repository package in daisy-pluckers PPA (DayBucketsProposedCount) 15:03 submitted RT regarding updating oops-repository and daisy on staging 15:03 submitted merge proposal regarding component-mismatches svg (show team name) 15:03 worked on component-mismatch improvements (showing uploader) 15:03 uploaded apport SRU to enable automatic crash reporting from Trusty (LP: #1431058) 15:03 Launchpad bug 1431058 in apport (Ubuntu Trusty) "Stable Release Update to provide possiblity of automatically reporting crashes" [Medium,Fix committed] https://launchpad.net/bugs/1431058 15:03 investigation into apport bug LP: #1300235 (chromium-browser crashes reported about init) 15:03 Launchpad bug 1300235 in chromium-browser (Ubuntu) "init (chromium-browser) crashed with SIGSEGV" [High,Triaged] https://launchpad.net/bugs/1300235 15:03 submitted apport merge proposal fixing apport part of bug LP: #1300235 15:03 verified that bug LP: #1419061 is fixed 15:03 Launchpad bug 1419061 in ubuntukylin-default-settings (Ubuntu) "On Ubuntu Kylin detect all packages as not genuine" [High,Fix released] https://launchpad.net/bugs/1419061 15:03 ✔ done 15:03 * reviewing/testing installer multipath patches in bug 1430074 15:03 * upload partman-base, partman-multipath, grub-installer with multipath fixes 15:03 * testing and review for post-boot multipath delays and issues: 15:03 bug 1430074 in debian-installer (Ubuntu) "fix handling of multipathed disks" [Critical,In progress] https://launchpad.net/bugs/1430074 15:03 - Multipath devices take long to initialize during initramfs (bug 1431650) 15:03 - system drops to initramfs after install on multipath disk (bug 1429327) 15:03 - Ship the default /etc/multipath.conf on multipath-tools-boot (bug 1432062) 15:03 bug 1431650 in multipath-tools (Ubuntu) "Multipath devices take long to initialize during initramfs" [High,In progress] https://launchpad.net/bugs/1431650 15:03 bug 1429327 in debian-installer (Ubuntu) "ISST-LTE: system drops to initramfs after install on multipath disk" [Undecided,Confirmed] https://launchpad.net/bugs/1429327 15:03 bug 1432062 in multipath-tools (Ubuntu) "Ship the default /etc/multipath.conf on multipath-tools-boot (for user_friendly_names)" [Medium,Triaged] https://launchpad.net/bugs/1432062 15:04 ^ I got some good results last night with changes to the multipath-tools udev rules 15:04 * reviewed urfkill pull request to fix new issues with flight mode and WWAN. 15:04 * silo for urfkill 15:04 * patch for NM connection selection bug, wrong context for IMSI (bug 1431471) 15:04 bug 1431471 in network-manager (Ubuntu) "NetworkManager uses ofono GPRS contexts for wrong SIM" [High,In progress] https://launchpad.net/bugs/1431471 15:04 (done) 15:04 meh, still difference in summer time ... (I'll go last, and will be forgotten ...) 15:05 hmm :) 15:05 sil2100 is off sick 15:05 jodh: 15:05 * snappy 15:05 - lp:~jamesodhunt/snappy/logger 15:05 - Reworked after further review, including new tests. 15:05 - Created packaging branch for the juju "loggo" golang logger, 15:05 required for the snappy logger branch: 15:05 https://github.com/jamesodhunt/loggo.git 15:05 - Reworked lp:~jamesodhunt/snappy/privileged-operations (also with new tests). 15:05 * upstart 15:05 - Bug 1429756 15:05 bug 1429756 in linux (Ubuntu Vivid) "FTBFS: upstart test_job_process fails in majority of cases / Kernel returning unexpected EIO at end of file" [High,In progress] https://launchpad.net/bugs/1429756 15:05 - Collaborated with apw. 15:05 - Lots of testing. 15:05 - Wrote a C program + test script that surfaces the timing issue without 15:05 upstart+libnih: https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/1429756/comments/13 15:06 - Tested on a 3.18, 3.19 and modified 3.19 kernels. 15:06 ວ໌ 15:06 apt: 15:06 - Debug http errors on azure with smoser 15:06 - Look into size tracking bug (#777565) 15:06 - Merge debian/sid fixes for vivid and FFe (#1431877) 15:06 - New 1.0.9.7ubuntu1 uplaod (matches the debian 7 version) 15:06 - Trusty SRU for #1429041 15:06 software-center: 15:06 - debug/fix install issue with aftershot pro2 from the store 15:06 - identify a bunch of similra incorrect meta-data for apps in the 15:06 store that prevents installing them and fix them 15:06 snappy: 15:06 - native build and unpack landed in snappy-go, major step in 15:06 reducing the external dependencies 15:06 - lots of features, fixes, cleanup 15:06 - package/upload golang-ar 15:06 - discuss/implement new hashes support 15:06 - look into oem snap dtb handling 15:06 software-properties: 15:06 Help Brian with software-properties-gtk questions 15:06 misc: 15:06 - trainguard duty 15:06 - performance review 15:06 (done) 15:06 - LXC/LXD 15:06 - Released LXC 1.1.1 on Monday 15:06 - Released LXD 0.4 on Tuesday 15:06 - Various bugfixes and feature work on LXC and LXD 15:06 - Code reviews, bug triaging, other admin paperwork 15:06 - Now working on 0.5, due Tuesday (bugfix release) 15:06 - Looked into a LXC security issue (related to Seccomp) 15:06 - Still waiting on LXD-related FFes (archive is currently on barely usable LXD 0.1) 15:07 - system-image 15:07 - code reviews (some more to do this week too) 15:07 - Other 15:07 - Booked travel to Austin sprint 15:07 (done) 15:07 qa sprint 15:07 (done) 15:09 slangasek: ? 15:09 slangasek: back to you 15:09 was still reading scrollback ;) 15:09 and mine was so short! :) 15:10 * snappy: working on the developer story, how to cross-build a snap 15:10 * review season continues, look for 1:1 invites coming soon 15:10 * spent a bit of time on proposed-migration stalls, fixing maas (bug #1433697) to unblock syslinux and removing some never-gonna-build packages that are broken in Debian unstable 15:10 bug 1433697 in MAAS "maas depends on syslinux-dev, removed upstream" [Critical,Fix committed] https://launchpad.net/bugs/1433697 15:10 * worked with robru on debugging uci-engine bootstrap issues 15:10 * sprint planning: we have a date and a location 15:10 * discussions with IBM about multipath-related bugs for 15.04 15:10 (done) 15:10 * Lots more glibc mangling, should be uploading today 15:10 * Wrangle some transitions out of proposed 15:10 * Work with didrocks (after a conversation with xnox) on restructuring the upstart package split 15:10 * General AA and SRU things 15:10 * Argue with Canonical VPN configs 15:10 * Discussions with Colin/William about bugcontrol versus series nomination 15:10 * Work on planning the vivid release sprint 15:10 (done) 15:11 Bugfix : 15:11 - Investigating rsyslog memory leak on Trusty : running stress tests 15:11 * Noticed small leak (~27Mb) on trusty (got help from infinity) 15:11 * Running stress tests to reproduce (need to open bug for this one) 15:11 - Instance left in dying state : setting Havana test environment 15:11 Continued work on backport of runlevel S kdump fix over trusty for testing 15:11 infinity: note that smoser opened critical - that ADT testing fails as it tries to boot things with upstart, whilst systemd-sysv is installed and that fails. 15:11 Preparation for CTS sprint 15:11 sosreport juju plugin enhancement 15:11 Need to book travel before CTS sprint 15:11 infinity: marked as duplicate of the master bug. 15:11 (done) 15:11 * CI Train 15:11 - Implemented 'dirty silo' notification -- when one silo is published/merged, other silos containing conflicting packages are marked dirty so that people know not to QA them until they've been rebuilt 15:11 - Rewrote migration job from scratch with 100% test coverage, bringing the overall total test coverage up from 85% to 90% 15:11 - various tweaks and improvements in error reporting, logging, pylint cleanups 15:11 - significantly improved the clarity of the error message for the case when you try to publish a silo that contains a package that hasn't been built yet. 15:11 - cleaned up the code that generates package diffs (code used to call debdiff 3 different times 3 different ways, now it just calls it once and processes the output 3 different ways as necessary) 15:11 * lp:cupstream2distro-config 15:11 - dropped some transitional flags that haven't been used in nearly a year, shrinking the size of the entire project by 1/6th 15:11 * CI Engine 15:11 - continuing to bumble through deployment issues, not much luck. 15:11 (done) 15:12 xnox: You have bug numbers? I was basing this work on our conversation, not on actual reported bugs. 15:12 infinity: yes, we have bugs =) https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/1422681 15:12 Launchpad bug 1422681 in upstart (Ubuntu) "split out upstart-sysv" [Undecided,Triaged] 15:12 xnox: Ta. 15:13 infinity: note that bug report has irc log pastes in it ;-) 15:13 xnox: Even better. 15:13 infinity: well, paraphrased. 15:13 smoser: links to _actual_ logs in the duplicate bug. 15:13 robru: dirty silo notifications> great! btw, I guess there's still nothing that prevents two conflicting silos being submitted to QA in parallel, even though only one of these will be published? 15:13 robru: which flags were transitional? 15:13 caribou: your turn 15:14 slangasek: done already 15:14 before somebody forgets me ... 15:14 - finally updated openjdk-8 15:14 - look at firefox ftbfs on arm64, reduce test case, file issue 15:14 - look at linux ftbfs on ppc64el, gcc dropping -m32 (but why do we need 32bit support) 15:14 - test ceph bugfix on armhf 15:14 - fix auto pkg test for kde packages 15:14 - migrations, vivid ftbfs fixes, ... 15:14 - more GCC 5 work 15:14 caribou: oh yes, failing to read my list (or scrollback) sorry 15:14 slangasek: indeed, nothing is in place to stop two built silos from going to QA in parallel; QA will have to just know not to QA the same package in two different silos. but once one gets published the other is immediately marked dirty 15:15 slangasek: np 15:15 slangasek: and the transitional flags were 'daily_release: False' and 'use_stack_ppa: False', ancient stuff from way way back when ci train was first brought online. 15:15 robru: interesting, thanks 15:16 slangasek: note that lp:cupstream2distro-config is different (and currently unrelated to) lp:cupstream2distro 15:16 oh, ok 15:17 [TOPIC] AOB 15:17 anything else? 15:17 slangasek: they were related pre-train but they've diverged since the train went online. 15:18 anyone with spare cycles could put them to use on helping to get the archive in shape for 15.04: http://people.ubuntuwire.org/~wgrant/rebuild-ftbfs-test/test-rebuild-20150202-vivid.html http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html 15:18 a lot of packages stuck in -proposed right now 15:19 Looking forward to the relaxation of build failure hunting once I'm done with other tasks. 15:20 :) 15:20 indeed :) 15:22 ok, sounds like that's it then 15:22 get your travel requests in for the sprint! :) 15:22 #endmeeting Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/meetingology)