[15:04]  * slangasek waves
[15:05] <ev> hi
[15:06]  * stgraber waves
[15:06] <stokachu> slangasek: i dont have anything pressing today
[15:06] <ev> The list of things I've forgotten to do in the past several months now includes: enabling support for recoverable error reports server-side :-/
[15:06] <ev> (RT 65174)
[15:06] <slangasek> #startmeeting
[15:06] <meetingology> Meeting started Thu Oct 10 15:06:40 2013 UTC.  The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[15:06] <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
[15:07] <slangasek> [TOPIC] Lightning round
[15:08] <slangasek> $ echo $(shuf -e barry doko stgraber jodh ev bdmurray slangasek cjwatson xnox stokachu)
[15:08] <slangasek> bdmurray xnox stgraber slangasek cjwatson barry jodh ev doko stokachu
[15:08] <slangasek> bdmurray: hello!
[15:08] <slangasek> stokachu: ack :)
[15:08] <bdmurray> bug triage
[15:08] <bdmurray> tested update-manager bug 1119427
[15:08] <bdmurray> removed usage of gksu from ubuntu-release-upgrader - bug 1210649
[15:08] <bdmurray> added phased-update knowledge to update-notifier - bug 1223321
[15:08] <bdmurray> click chroot modifications
[15:08] <bdmurray> reviewed / approved P SRU of libreoffice
[15:09] <bdmurray> ⌁ done
[15:09] <cjwatson> bdmurray: Are you finished with the chroot work?  I think I was waiting for a response to my most recent comment before circling back
[15:09] <xnox> * emulator:
[15:09] <xnox>  - adbd into android available out of the box
[15:09] <xnox>  - serial into ubuntu available out of the box
[15:09] <xnox>  - everything else is built and boots correctly
[15:09] <xnox>  - emulator on the people.c.c is updated, but is kind of hand-build
[15:09] <xnox>    (some patches are not in saucy)
[15:09] <xnox>  - todo: /data/ is not mounted inside the android-lxc container
[15:09] <xnox> * bugs:
[15:09] <xnox>  - fixed up u1 page bug in ubiquity (couldn't login/sign-up)
[15:09] <xnox>  - putting spam plug into upstart-udev-bridge (and related debugging)
[15:09] <xnox>  - todo: spam plug on the udev source in systemd-udevd
[15:10] <xnox>  - fixing FTBFS in main
[15:10] <xnox> ..
[15:10] <stgraber> No fancy report today, fighting with dbus/kernel/lxc over in #ubuntu-touch (not sure why since it doesn't affect touch though ;)) ...
[15:10] <bdmurray> cjwatson: no, I was going to have a poke at it again today
[15:10] <stgraber>  
[15:10] <stgraber> Quite a lot of upload queue reviews
[15:10] <stgraber> Looked into a couple of ubiquity bugs (xnox, what's the status on the webkit related one?)
[15:10] <stgraber> LXC code reviews and cherry-picked a bunch of bugfixes (including trying to sort out the dbus/apparmor mess introduced by recent kernels...)
[15:10] <stgraber> Added logging support to system-image
[15:10] <stgraber> Started to work on some of the changes required for ports support for system-image
[15:10] <stgraber> Some more touch related work (debugging udm, system-image, ...)
[15:10] <stgraber> (DONE)
[15:11] <cjwatson> bdmurray: ok, cool
[15:13] <xnox> stgraber: right, it's on my todo list. but i don't think i'll get to it today.
[15:14] <stgraber> xnox: yeah, no worries, I kind of gave up on that one after I told cjwatson and slangasek I'd take care of it so I at least keep an eye on it to make sure we make progress (our initial guess was something related to logind I may have been able to help with, but now that it's clearly a webkit issue, I'm not too useful ;))
[15:15] <xnox> stgraber: i see.
[15:15] <xnox> stgraber: my guess was to move loading up T&C webpage, only if/when people click to see it. Instead of on module import. Thus at that point we at least know if network is available and don't fail the install horibly.
[15:16] <slangasek>  * working to try to reproduce bug #1234731 in upstart
[15:16] <slangasek>   * being hamstrung by bug #1228406
[15:16] <slangasek>  * freeze reviews
[15:16] <slangasek> (done)
[15:17] <cjwatson> Lots of archive cleanup.  NBS now zero, amd64 and armhf zero uninstallables, saucy_outdate_all.txt almost cleared, proposed-migration backlog improved, component-mismatches-proposed tidied up, fixed various build failures.
[15:17] <cjwatson> ARM64 porting.  Lots of build supervision and scheduling.  Loads of config.guess/config.sub updates.  Fixed dlopen segfault in statically-linked programs which is harmless but confused the fault watcher we have to have in our builds at the moment.
[15:17] <cjwatson> Investigated bug 1237519.
[15:17] <cjwatson> Working on displaying Launchpad builder versions in the UI so that upgrades are slightly less painful to monitor.
[15:17] <cjwatson> To do:
[15:17] <cjwatson>  * Finish reviewing and merging Brian's chroot management branch (carried over).
[15:17] <cjwatson>  * Hunt through GRUB code to try to chase down bug 1229458 (carried over).
[15:17] <cjwatson>  * Merge Steve's GRUB patch (in progress).
[15:17] <cjwatson>  * Look for last-minute installer bugs to hoover up.
[15:17] <cjwatson> ..
[15:17] <cjwatson> (I can't take credit for all the archive cleanup obviously :) )
[15:17] <barry> short week due to illness.
[15:17] <barry> system-image: LP: #1215586, LP: #1206588, LP: #1235975
[15:18] <barry> done
[15:18] <stgraber> xnox: not sure if that'll be enough though, here I clearly had network access and ubiquity would work fine if started under sudo, so it seems like we've got a permission issue somewhere (write to cache, config, similar stuff?)
[15:18] <jodh> * upstart
[15:18] <jodh>   - Fixed bug 1234898.
[15:18] <jodh>   - Investigated bug 1235231.
[15:18] <jodh>   - Investigated bug 1234743.
[15:18] <jodh>   - Investigating bug 1235649 - multiple issues: relates to 1234743 and
[15:18] <jodh>     what looks to be an incorrect use of the D-Bus API (ongoing).
[15:18] <jodh> ಠ
[15:18] <xnox> stgraber: set a11y?!
[15:19] <xnox> (cause that's one thing that webview does)
[15:19] <xnox> jodh: i think mhr3 can be a dbus expert, he seems to be working a lot with e.g. for https://launchpad.net/dee
[15:20] <ev> - Chat with Rick on the 1SS move.
[15:20] <jodh> xnox: ack, thanks.
[15:20] <ev> - Chat with James on the 1SS move.
[15:20] <ev> - Call with Chris and Andy on getting a role account in canonistack for team
[15:20] <ev>   projects. Chatted with Tom Haddon over the approach. Submitted an RT.
[15:20] <ev> - Call with Chris on the vanguard implementation.
[15:20] <ev> - Landing task force standups.
[15:20] <stgraber> xnox: could be
[15:20] <ev> - Work with Chris on getting OOPS support into lp:qa-dashboard.
[15:20] <ev> - Some light code review.
[15:20] <ev> - Chased Canonistack role account (RT 65080).
[15:20] <ev> - Debugging GNetworkMonitor waking up frequently with apw (LP: #991481).
[15:20] <ev>   Suggestions welcome.
[15:20] <ev> - A few meetings with Alice on the move.
[15:20] <stgraber> xnox: if sudo strips some of that from the environment, that may be the different
[15:20] <stgraber> *difference
[15:20] <ev> - Set up a project to manage the new canonistack role account:
[15:20] <ev>   https://launchpad.net/ci-engineering-canonistack
[15:20] <ev> - Call with Rick, Larry, pinky, Herb, and James Troup on the 1SS move.
[15:20] <ev> - Catch up call with Chris.
[15:20] <ev> - Reviewed some small changes from deej to the daisy and daisy-retracer charms.
[15:20] <ev> - Investigated using apport-retrace for the touch images for lool. Chat with
[15:20] <ev>   lool and pitti about how we can override UnreportableReason.
[15:20] <ev> - Chased jjo on https://rt.admin.canonical.com/Ticket/Display.html?id=64590.
[15:20] <ev> - Mail to the CI team, explaining the new Canonistack role account.
[15:20] <ev> - Arranged a meeting with the whole CI team to discuss the 1SS move.
[15:20] <xnox> stgraber: cause ubiquity-dm does process and sets/unsets a11y & modifies gsettings which wouldn't be there, if boot straight to desktop.
[15:21] <ev> - Mail to ubuntu-engineering, letting them know how they can engage the CI
[15:21] <ev>   engineering team.
[15:21] <ev> - Worked with Chris and Alex to further clarify our responsibilities in the
[15:21] <ev>   wiki.
[15:21] <ev> - Call with Alex, Steve, Leann, Michael, and Colin on the landing spreadsheet
[15:21] <ev>   and what we're doing post-release.
[15:21] <ev> - Created RT 65144 for setting up oops.canonical.com for reports.qa.ubuntu.com.
[15:21] <ev> - Mail to ubuntu-engineering explaining how to retrace crashes.
[15:21] <ev> - Chased retracers in prodstack. Nearly there!
[15:21] <ev> - Working with the AUS folks to find an acceptable meeting time.
[15:21] <ev> - Small review and merge from Steve for daisy armhf retracer config.
[15:21] <ev> - Helping the CI team chase a LXC/apparmor issue around dbus.
[15:21] <ev> - Chasing up statsd RT with alexlist, psivaa, and vila. Waiting for init script
[15:21] <ev>   fixes from jjo.
[15:21] <ev> - Chased up oops.canonical.com reporting from dashboard.qa.ubuntu.com (RT 65144)
[15:21] <ev> - Putting Larry and Rick in charge of lab access and discussing using IS' LDAP,
[15:21] <ev>   per Alex's request.
[15:21] <ev> - Chasing up why we're not seeing the autopilot documentation on
[15:21] <ev>   http://utah.readthedocs.org/en/latest/.
[15:21] <ev> - Helped jjo debug statsd configuration.
[15:21] <ev> - Fixes to lp:daisy to support recoverable problems server-side.
[15:21] <ev> - Investigated bugs in the retracers from deej that turned about to be wrong
[15:21] <ev>   package versions in the deployed charms.
[15:21] <ev> (done)
[15:22] <ev> I think there's more, but I forgot to write down whatever it was I did last week
[15:22] <cjwatson> oh, one other thing on my to-do, the germinate run at the end of the publisher is crashing with something arm64-related and I need to investigate why ASAP
[15:22] <slangasek> jodh: do you have all the remote access to a mako device that you need for reproducing the problem in place on the phone?
[15:25] <jodh> slangasek: yes, I do now thanks.
[15:25] <slangasek> jodh: ok, good
[15:25] <slangasek> doko: your turn :)
[15:25] <doko> - one more last (later?) gcc-4.8 update for saucy, fixing wrong code-gen for gmp and nss
[15:25] <doko> - update all our cross binutils and compilers
[15:25] <doko> - fixing build failures, ftbfs looks good (main), the test rebuild too besides some texinfo related ftbfs
[15:25] <doko> - Address remaing MIRs
[15:25] <doko> - Trying to prepare a Go meetup during Linaro Connect together with mwhudson and some google Go devs
[15:25] <doko> - Prepare an AArch64 test chroot
[15:25] <doko> - AArch64 upload orgies
[15:25] <doko> - last Thu was national holiday, was sick on Fri
[15:25] <doko> (done)
[15:28] <lool> ev: pitti suggested something simple for forcing report on .crash
[15:28] <slangasek> doko: so I didn't get back to you on your question about wrong code generation on armhf
[15:28] <ev> hm?
[15:28] <slangasek> I think it's too late (and was already too late) to do much about this before release, besides release note it
[15:28] <lool> ev: just remove the unreportable reason with vi and submit it (ugly, but works); he didn't want an official flag as this approach is frowned upon in the first place
[15:28] <doko> yeah
[15:29] <ev> lool: seems reasonable
[15:29] <slangasek> doko: I was puzzled by something in your mail, you said that it affected 64-bit math, but the test case for it is only using 32-bit ints?
[15:30] <doko> was it? hmm ... just chatted with linora tcwg guys
[15:30] <slangasek> yeah
[15:30] <slangasek> so if it affects 32-bit math, that worries me a lot more :-)  Yet the problem was only reported in things like gmp
[15:31] <slangasek> and if we had a widespread problem with 32-bit math I would expect everything to be on the floor, and it's not
[15:31] <doko> yes, I add the list of packages that apparently were affected
[15:31] <slangasek> those affected packages were all ones that caught the bug in the testsuite though, right?
[15:32] <doko> yes, I'm sure about nss and the cyrpto package, didn't check the others
[15:33] <slangasek> so the part that worries me is the packages that might be affected but *don't* have test suites
[15:33] <doko> exactly
[15:33] <slangasek> but I don't think we currently can quantify this well enough to release note it
[15:33] <slangasek> "some math might be broken" is scaremongering, not a useful release note
[15:33] <doko> well, it's not broken anymore
[15:34] <slangasek> doko: can you work with the Linaro guys to try to identify the scope of this?  Any chance that there's a way to search the archive for packages that include the wrong code?
[15:34] <slangasek> or maybe they know that this is a very rare corner case that only affects heavy-duty math libs
[15:34] <slangasek> right, it's not broken for anything built with the current compiler
[15:34] <bdmurray> the ubuntu security team has code for searching archive packages
[15:35] <doko> slangasek, the corner case was the issue which did show up in mpfr. right, will check again, but afaiu just scanning the assembler for some instructions isn't possible
[15:35] <doko> will do that at connect
[15:35] <slangasek> bdmurray: yeah, what doko said, the hard part is scanning the binary for assembly sequences
[15:35] <doko> xnox, did you feel disturbed? ;-P never typing that much during meetings
[15:36] <slangasek> I know there are tools out there for scanning for instructions, but perhaps in this case it's too difficult to craft a proper match for the breakage
[15:36] <slangasek> anyway
[15:36] <slangasek> [TOPIC] AOB
[15:36] <slangasek> ?
[15:36] <xnox> doko: I use push-to-talk such that I don't make noise =)
[15:37]  * xnox ponders if Saucy Salamander, will be followed by Tasty Tintin
[15:38] <bdmurray> I could use some help with bug 1024590
[15:39] <slangasek> xnox: a tintin is an animal?
[15:40] <mdeslaur> tasty turkey
[15:42] <barry> tangy tadpole
[15:42] <slangasek> bdmurray: how can we help?
[15:43] <slangasek> reading the bug log, it seems you're already pretty far down the rabbit hole
[15:43] <bdmurray> and then I got stuck
[15:44] <slangasek> anyone have some spare cycles to help bdmurray with this?

[15:46] <slangasek> bdmurray: I can try to take a look this afternoon, but no promises :/
[15:47] <bdmurray> slangasek: okay, no problem
[15:47] <slangasek> anything else today?
[15:48] <slangasek> #endmeeting
[15:48] <meetingology> Meeting ended Thu Oct 10 15:48:44 2013 UTC.
[15:48] <meetingology> Minutes (wiki):        http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-10-10-15.06.moin.txt
[15:48] <meetingology> Minutes (html):        http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-10-10-15.06.html
[15:48] <slangasek> thanks :)
[15:48] <barry> thanks!
[15:48] <stgraber> thanks
[15:56] <jodh> thanks