== Meeting information == * #ubuntu-meeting Meeting, 27 Mar at 15:01 — 15:37 UTC * Full logs at [[http://ubottu.com/meetingology/logs/ubuntu-meeting/2014/ubuntu-meeting.2014-03-27-15.01.log.html]] == Meeting summary == === lightning round === The discussion about "lightning round" started at 15:02. * ''LINK:'' http://www.imdb.com/title/tt0051381/ === AOB === The discussion about "AOB" started at 15:24. == Vote results == == Done items == * (none) == People present (lines said) == * slangasek (39) * cjwatson (29) * bdmurray (24) * stgraber (24) * jodh (16) * ubottu (14) * doko_ (11) * barry (8) * caribou (6) * meetingology (3) == Full Log == 15:01 #startmeeting 15:01 Meeting started Thu Mar 27 15:01:35 2014 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. 15:01 15:01 Available commands: action commands idea info link nick 15:02 [TOPIC] lightning round 15:02 $ echo $(shuf -e barry doko stgraber jodh bdmurray slangasek cjwatson xnox caribou infinity) 15:02 doko slangasek infinity cjwatson bdmurray caribou barry jodh xnox stgraber 15:03 doko_: hi 15:03 - another round of GCC updates 15:03 - python 3.4 update, python 3.3 removal 15:03 - cleaning up a bit in -proposed, nbs 15:03 - scan for compiler ICEs in the 4.9 test rebuild, file upstream issues 15:03 - some random updates and ftbfs fixes 15:03 (done) there was more, can't remember all ... 15:04 \o/ for 3.3 removal 15:04 a few days remaining to remove ruby1.9.1 and update to ruby2.0 ;-P 15:05 oh? 15:05 * catching up on the world after being heads-down on customer work for two weeks 15:05 * moving forward on the Java role; expected to begin interviewing next week 15:05 * participating in good discussions about the current landing process and bottlenecks around image promotion (cf. Qt5.2) 15:05 * discussions around some late FFes coming in 15:05 * performance review discussions using the new and improved process 15:05 (done) 15:05 doko_: are we switching to ruby2.0 for trusty, or is it your turn to troll? :) 15:06 infinity's not here, so cjwatson: 15:06 Finished landing upstart-app-launch performance improvements via libclick. 15:06 Clawed away some of the backlog of uninstallables on arm64/ppc64el. 15:06 grub-installer fixes for ppc64el. 15:06 well, I assume security would like to only support one version, but nobody is pushing for it, and the puppet people won't look at it as usual :-/ 15:06 Continuing work on making ubuntu-sdk-libs-dev:armhf installable for "click chroot". Landed: dee, libaccounts-glib. Current blockers: libfriends, qml-friends. 15:06 ubiquity fixes for final beta (reviewed bug 1297312, worked around bug 1296697). 15:06 bug 1297312 in ubiquity (Ubuntu Trusty) "ubiquity-dm fails with: FileNotFoundError: [Errno 2] No such file or directory: '/usr/lib/x86_64-linux-gnu/indicator-application-service'" [Critical,Fix released] https://launchpad.net/bugs/1297312 15:06 Started trying to get my juju cross-build environment back up on trusty. 15:06 bug 1296697 in ubiquity (Ubuntu Trusty) "Trusty Desktop installer crashed with "Encrypted Home" set" [Critical,Fix released] https://launchpad.net/bugs/1296697 15:06 .. 15:07 ruby updates are a long way behind - I spent some time looking at various bits, it's a mess 15:07 like ruby-gnome2, we can fix it, realistically it'll take a new upstream version which'll need an FFe 15:07 and it's kind of all like that 15:08 but maybe I just don't know it well enough 15:08 doko_: who are "the puppet people"? 15:09 * barry thinks that sounds like a great band name or horror movie title 15:09 doko_: in the future, can we please make such questions a UDS session, so that we discuss with the server team at the beginning of the cycle and get a committment from them to include this work on their roadmap for the cycle? 15:09 http://www.imdb.com/title/tt0051381/ 15:10 cjwatson: speaking of puppets, I had bug #1292628 escalated to me in person; this seems to be a regression in the interaction between grub-install, grub-installer, and kickseed following the recent grub-installer rewrite. Do you think this should be on the radar for 14.04? 15:10 bug 1292628 in grub2 (Ubuntu) "'grub-install --force "(hd0)"' fails on trusty during installation" [High,Triaged] https://launchpad.net/bugs/1292628 15:11 slangasek, how? the plans for 2.0 did start in debian only in January ... 15:11 slangasek: I'll have a look 15:11 doko_: hmmm. put the UDS session on the agenda, and make the server team do their homework on ruby versions before the meeting? :) 15:12 ok 15:12 surprising that the grub-mkdevicemap code would have broken; so hopefully it's not too hard 15:12 the planning for ruby2.0 may have started in January, but I guess if someone would have asked the Debian ruby team in November they would've had a good guess 15:12 * bdmurray is ready but waiting 15:12 I vaguely recall the Debian ruby maintainer mentioning 2.0 at debconf in July/August or so 15:13 cjwatson: sorry, it's not grub-mkdevicemap that's the problem per se, the problem is that kickseed is feeding us '(hd0)' as an override and grub-install never has a device map on disk so seems to not do anything sensible 15:13 though it was in passing and I don't think he mentioned a timescale 15:13 slangasek: right, but the grub-mkdevicemap calling code in grub-installer is meant to deal with that just as it did in prior releases 15:13 hmm 15:13 ok 15:13 stgraber: ? 15:13 I'm not disbelieving the bug or anything, just saying it's hopefully shallow 15:14 yes, met him at Connect a few weeks ago. I think we can't remove it, but likely demote it if puppet works with 2.0 15:14 slangasek: ? 15:15 I think I was next so I'll start now 15:15 confirm prodstack data is appearing on graphite 15:15 reported daisy bug 1295400 regarding retracer sandboxes 15:15 submitted merge proposal to daisy to remove sandbox dir data after each retrace (bug 1295400) 15:15 bug 1295400 in Daisy "retracer requires a lot of storage for instance sandboxes" [High,New] https://launchpad.net/bugs/1295400 15:15 merged thedac's changes to the daisy-retracer charm 15:15 updated daisy's submit.py to reject crashes from armel systems as we have no retracer for them 15:15 verified that bug 1267112 is in fact fixed since we have a bucket for the ubiquity / autopilot-gtk bug 15:15 bug 1267112 in Daisy "errors indicates a retraced crash but isn't showing a stack trace" [High,Fix released] https://launchpad.net/bugs/1267112 15:15 investigation into hplip-data crashes missing package and sourcepackage information 15:15 tested daisy (haproxy config) changes to reduce occurrence of 502 errors in whoopsie 15:15 worked with thedac regarding missing OOPS reports from prodstack version of errors 15:15 updated my apport retrace-package-versions merge proposal 15:15 sent error tracker armhf announcement 15:16 investigation into armhf crashes on errors (webbrowser-app failed to retrace, likely due to outdated debug symbol package for liblttng-ust0) 15:16 SRU queue work with arges 15:16 uploaded a fix for bug 1294392 15:16 bug 1294392 in ubuntu-release-upgrader (Ubuntu) "PAE check prevents Precise to Trusty upgrades on non-Intel ports" [High,Fix released] https://launchpad.net/bugs/1294392 15:16 ␗ done 15:16 research into update-manager test suite failures (bug 1295392) 15:16 bug 1295392 in update-manager (Ubuntu) "test suite failures" [Medium,Fix committed] https://launchpad.net/bugs/1295392 15:16 stgraber: hah, yes, somehow I thought you were next - sorry :) 15:16 caribou: anything you'd like to share today? 15:17 I'm next, it'll be quick 15:17 * Completed uvtool documentation in the server guide 15:17 *btmp logging re-enabling (Debian bug #742775) 15:17 Debian bug 742775 in openssh-server "openssh-server: Please re-enable btmp logging" [Normal,Open] http://bugs.debian.org/742775 15:17 * Misc bug fixing 15:17 Done 15:17 short week due to holiday 15:18 bdmurray: we should have noisemakers and party hats for the armhf retracers 15:18 phone: py3 autopilot: gallery-app, address-book-app. trying to further debug phablet-test-run issues. 15:18 ubuntu/debian: LP: #1295392. LP: #1296325. pystache FTBFS. pip 1.5.3-1 (blocked on python-distutil in stuck in NEW). ensurepip/pyvenv for debian. other general debian work 15:18 Launchpad bug 1295392 in update-manager (Ubuntu) "test suite failures" [Medium,Fix committed] https://launchpad.net/bugs/1295392 15:18 Launchpad bug 1296325 in oneconf (Ubuntu) "oneconf-query needs generic /usr/bin/python3 shebang" [High,Fix released] https://launchpad.net/bugs/1296325 15:18 other: debugging python import problems for security team (http://tinyurl.com/l7ps5br). code reviews for other internal projects. 15:18 done 15:18 * foundations-1305-upstart-work-items 15:18 - cgroup support: Spent a lot of time debugging the async spawn issue 15:18 with help from cjwatson. As a result of that, we're now planning not 15:18 to pursue the async spawning feature due to the fact that to 15:18 implement this fully would require an extensive refactoring of the 15:18 Upstart core. As such, I'm now refocusing on the cgroup branch with 15:18 the intention of either having Upstart manage its own cgroups, or 15:18 working with hallyn to refactor cgmanager to allow Upstart to 15:18 leverage that work in some form. 15:18 * upstart 15:18 - Wrote draft release notes for Upstart 15:18 https://wiki.ubuntu.com/TrustyTahr/ReleaseNotes/TechnicalOverviewUpstart 15:19 - Reworked MP for logrotate improvements: 15:19 caribou: fwiw I don't think the claim in that btmp bug is totally true; Debian stable at least still has -rw-rw---- 1 root utmp 0 Mar 1 08:02 /var/log/btmp 15:19 https://code.launchpad.net/~jamesodhunt/ubuntu/trusty/upstart/periodic-logrotate-using-events/+merge/212569 15:19 - 15:19 Ⱆ 15:19 caribou: this isn't simple, need to track down exactly what changed and when 15:19 otherwise I'll cause regressions 15:19 cjwatson: ok, will check that out 15:19 no xnox so I guess that's my turn then 15:20 LXC: 15:20 - Finally tracked down our last 1.0.2 blocker, will release 1.0.2 today with cgmanager support enabled in Ubuntu (that may be our last LXC upload for the cycle)! 15:20 - Some more CI infrastructure work (almost everything runs in Jenkins now) 15:20 - Code reviews 15:20 - A couple of bugfixes for lxc-ls performance when > 2k containers and userns config fixes 15:20 - Fire fighting the latest apparmor breakage (they broke the parser causing quite a few bug reports for LXC upstream and in Ubuntu...) 15:20 logind: 15:20 - Unblocked as the cgmanager MIR was finally approved! 15:20 - Filed the FFe to get cgmanager support landed: bug 1297363 15:20 bug 1297363 in systemd (Ubuntu) "[FFe] Add cgmanager support" [Undecided,New] https://launchpad.net/bugs/1297363 15:20 network: 15:20 - Uploaded a new ifupdown fixing quite a few bugs reported by CTS. 15:20 - Uploaded a new vlan fixing another bug reported by CTS. 15:20 - Did some more triaging of our network packages. 15:20 - Answered some of lamont's OpenVPN questions and looked into a few NM OpenVPN crashes. 15:20 caribou: ah, actually that's not a problem per the code ... would still like to know when/where btmp changed perms though 15:20 system-image: 15:20 - some discussions and some code to get new image notifications delivered by the new push service. 15:20 other: 15:20 - Helping lttng upstream update their packages to a stable version for 14.04. 15:20 - Couple of CI-Train landings. 15:20 - Some FFe and queue review. 15:20 - Helped setup Beta 2, tested Edubuntu, fixed a couple of issues, added ppc64el to tracker + cdimage scripts, ... 15:20 (DONE) 15:21 caribou: ah, and openssh itself has started allowing group rw since the bug was originally filed, too, which was the other part of i 15:21 t 15:22 any questions over status? 15:22 btw, I should have mentioned, I'll be off Monday 15:24 [topic] AOB 15:24 anything else today? 15:25 Could someone have a look at bug 1296386 and related bug 1296373? 15:25 has someone started putting together beach recommendations for the May sprint? 15:25 bug 1296386 in casper (Ubuntu) "[PATCH] Remove 23etc_modules script" [Undecided,New] https://launchpad.net/bugs/1296386 15:25 bug 1296373 in hw-detect (Ubuntu) "[PATCH] Fix sound on PowerPC" [Undecided,New] https://launchpad.net/bugs/1296373 15:26 bdmurray: any reason those shouldn't be pushed to ubuntu-sponsors? 15:26 alternatively, you could check if infinity wants to have a look at this 15:26 caribou: ah, I found it, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=370050 15:26 Debian bug 370050 in logrotate "general: /var/log/btmp permission should be 660" [Wishlist,Fixed] 15:26 ubuntu-sponsors deals with branches and debdiffs not ordinary patches last I knew 15:27 ubuntu-sponsors largely shouldn't care about the format 15:27 those are native packages, the difference between a patch and a debdiff is negligible 15:27 I don't think it was the format so much as the volume 15:28 However, the system (of choosing who to subscribe) was setup quite some time ago 15:31 I think it's fine to have separate frontline review of patch-tagged bugs to make sure they're actually ready for sponsorship, but in this case I think it would be ready for sponsorship 15:33 slangasek: ah, that makes sense 15:35 okay, I'll ping infinity and depending on that subscribe sponsors 15:35 bdmurray: sounds good 15:35 aught else? 15:36 naught 15:36 except for wondering why the sudden Elizabethan English :) 15:36 ;) 15:36 sorry, "save for" 15:37 cjwatson: variety 15:37 #endmeeting Generated by MeetBot 0.1.5 (http://wiki.ubuntu.com/meetingology)