=== micahg_ is now known as micahg === Logan_ is now known as Guest50062 === Logan__ is now known as Logan_ === chuck_ is now known as zul === lag` is now known as lag === ivoks_ is now known as ivoks === doko_ is now known as doko === inetpro_ is now known as inetpro === psivaa_ is now known as psivaa === nigelb_ is now known as nigelb === shadeslayer_ is now known as shadeslayer === cprofitt_ is now known as cprofitt === magespawn_ is now known as magespawn === Darkwing_ is now known as Darkwing === ogasawara__ is now known as ogasawara === davmor2_ is now known as davmor2 === soren_ is now known as soren === ogasawara is now known as Guest37522 === Zic_ is now known as Guest8512 === jono is now known as Guest19882 === leoquant is now known as Guest50782 === ashams is now known as Guest15032 === jrib is now known as Guest88569 === debfx_ is now known as debfx === lan3y is now known as Laney === Guest19882 is now known as jono === jono is now known as Guest85454 [15:03] * slangasek waves [15:03] #startmeeting [15:03] Meeting started Wed Jun 5 15:03:37 2013 UTC. The chair is slangasek. Information about MeetBot at http://wiki.ubuntu.com/meetingology. [15:03] 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 [15:05] [TOPIC] lightning round === meetingology changed the topic of #ubuntu-meeting to: lightning round [15:05] $ echo $(shuf -e barry doko stgraber jodh ev bdmurray slangasek cjwatson xnox stokachu) [15:05] cjwatson xnox barry stokachu jodh doko ev stgraber slangasek bdmurray [15:05] * slangasek waves the starting flag [15:05] phew [15:05] Investigation of delta updates for click packages. Playing with debdelta, which looks promising, including working on a Python 3 port. [15:05] Refactored click-package into the pattern of a single master command with subcommands (will land shortly). [15:05] Lots of merges/syncs, including giant debhelper merge. automake1.13 was some fun due to a buildd still running ext3. Helped to fix firefox on powerpc. [15:05] Finished several transitions (ghc, libgd, net-snmp, bits of ocaml). [15:05] Sorted out some confusion with ubuntu-touch-meta. [15:05] Several Debian NMUs / QA uploads to try to reduce deltas when I'm bored of merging things. [15:05] Proposed Debconf BOF on Ubuntu daily quality and how it overlaps with Debian's CUT project. [15:05] Next week: really, really get autopkgtest/proposed-migration integration going. [15:05] .. [15:06] (and by "promising" I think I mean "makes my eyes bleed but seems to fill the right slot") [15:07] please skip me. not ready yet. [15:07] LP: #1185170, python-gnupg 0.3.3-0ubuntu1 and 0.3.4-0ubuntu1 [15:07] Launchpad bug 1185170 in python-json-pointer (Ubuntu) "setup.py stumbles about an umlaut in jsonpointer.py when running under Python 3" [Undecided,Fix released] https://launchpad.net/bugs/1185170 [15:07] dmb meeting [15:08] emacs 24.3 upgrade (yay! but it broke a few minor things that had to be tracked down) [15:08] image based upgrades: gpg regime (blacklists, api refactoring, keyring download/verification/unpacking) lp:~barry/+junk/gpg for now [15:08] todo: continue working on the gpg spec [15:08] done [15:08] barry: sorry barry about emacs =) [15:08] xnox: no, no, no... *thank* you. the upgrade fixed more bugs than it caused :) === G4MBY is now known as PaulW2U [15:10] no stokachu today (at a sprint AIUI) [15:10] jodh: [15:10] * foundations-1305-upstart-app-launching [15:10] - Reworked lp:~jamesodhunt/upstart/libupstart to use an unpacked [15:10] ABI dump. [15:10] - Lots of rework on lp:~jamesodhunt/upstart/allow-multiple-cmdline-confdirs [15:10] to improve test code-sharing and make writing new functional tests easier. [15:10] - Raised MP for lp:~jamesodhunt/upstart/bug-1159895. [15:10] * misc: [15:10] - LPC13 planning. [15:10] * TODO: [15:10] - Ideally get lp:~jamesodhunt/upstart/serialise-remaining-objects merged [15:10] upstream today. [15:11] - Go on holiday (from tomorrow, back on 20th). [15:11] ʚ [15:11] - android cross toolchain uploaded to PPA [15:11] - GCC 4.8.1 update, now updating cross compilers [15:11] - Mon and Tue two days off [15:11] (done) [15:12] ev: [15:12] drats [15:12] - https://wiki.canonical.com/Cassandra [15:12] - Rewrote and sent the announce email for the errors.ubuntu.com NDA: [15:12] https://lists.ubuntu.com/archives/ubuntu-devel-announce/2013-May/001039.html [15:12] - Processing quite a few NDA signatures :). [15:12] - Investigated and resolved a problem where the NDA signature team couldn't [15:12] access errors.ubuntu.com - we accidentally dropped a cowboy. [15:12] - Attended a Cassandra meetup. Trip report here: [15:12] https://lists.launchpad.net/canonical-tech/msg06936.html [15:12] doko: do you know if ogra (or anyone) has started using the cross-toolchain yet? I think he might be busy still with other pieces of the puzzle [15:12] - Worked with webops to get OOPS reporting to oops.canonical.com after many [15:12] configuration headaches: [15:12] https://oops.canonical.com/reports/WHOOPSIE-PROD/ [15:12] This means we no longer get 404s on some reports, and can scale to N wsgi [15:12] workers without having OOPS reports in N places. [15:12] - Debugging why we were getting authentication failures to Swift when [15:12] submitting core dumps with the webops team. It looks like RED kicked in [15:12] because we are so behind on processing retraces (it fires on high disk usage) [15:12] and that somehow caused the auth failures further down the chain. Still [15:12] looking into this one. [15:12] - Code review for Brian. [15:12] - Got webops to start graphing the time to process HTTP requests on [15:12] daisy.ubuntu.com after discussions with Tom. We want to be able to measure [15:12] the performance impact with the move to Ceph, which this should let us do [15:12] quite well. [15:12] slangasek, i'm to busy with the container flip still [15:13] - Debugging why the back population jobs were falling over. This turns out to [15:13] be HAProxy being a very bad idea when using pycassa: [15:13] https://rt.admin.canonical.com/Ticket/Display.html?id=62023 [15:13] - Working with webops to get a test Cassandra node provisioned to try [15:13] compression again, given learnings from the Cassandra meetup: [15:13] https://rt.admin.canonical.com/Ticket/Display.html?id=61877 [15:13] - Added code to provide OOPS reports on javascript exceptions in [15:13] errors.ubuntu.com. [15:13] ogra: ok :) [15:13] - Improved our OOPS reporting on errors in JSON API calls. We now include the [15:13] traceback for those. [15:13] - Discussion with Katherine on how we can provide people like Kees access to [15:13] the error data for investigating security vulnerabilities. [15:13] - Added RED support to the prodstack version of lp:daisy. Broken and in turn [15:13] preventing merges from landing; investigating. [15:13] - Working with Ted to get them using recoverable problems for glib critical [15:13] errors. [15:13] slangasek, i plan to try it out for the updater-binary (most important bit for zip creation) before end of the week [15:13] - More chatting with the server team on getting error reporting up and running. [15:13] Sounds like a d-i module for automatic error reporting and something in motd [15:13] to point at outstanding reports are good places to start, and I think we have [15:13] consensus. Next step will be talking with the juju team to see how we can [15:13] make it easy to enable from there. [15:13] - Added metrics to support monitoring usage of errors.ubuntu.com, as laid out [15:13] in the NDA. [15:13] ogra: great! [15:13] More, I'm sure, but I ran out of time typing :) [15:13] (done!) [15:13] slangasek, no, will ask [15:14] ahh ... [15:14] ev: getting any NDA requests from people who shouldn't be given access to the data? :P [15:14] slangasek: there's been at least two people who tried to provide an incomplete or nonexistent address. Just a reminder folks, "The Internet" is not a place we can send snail mail to. [15:14] I've followed up and convinced them to do the right thing. [15:15] pff, snail mail [15:15] but there haven't been people who claimed to be part of some nonexistent project yet [15:15] it's mostly been names I recognise [15:15] and some Canonical people (I'm adding them in, in case they ever jump ship and still want access) === Riddelll is now known as Riddell [15:15] Blueprint-related work: [15:15] - Image based updates (BLUEPRINT: foundations-1305-image-based-updates) [15:15] - Waiting on review of the GPG implementation plan, need to get the keys generated and signed after that. [15:15] - Waiting for IS to setup the image based update server. [15:15] - Update the differ to be pure python. We no longer need to unpack the tarballs for the checksuming part. [15:15] - Rebuild the test server (http://phablet.stgraber.org) to match the current version of the spec. Includes all the needed keyrings and real test tarballs. [15:16] - Added a new function, generate_version_tarball, which generates our version tarball (containing /etc/ubuntu-build) [15:16] - The testsuite now covers all the server side code. [15:16] Other work: [15:16] - release/cdimage [15:16] - Quick production update of the QATracker (css update by knome) [15:16] - Wrote a script to extract flavours participating in the current milestone (to be integrated in seeded-in-ubuntu) [15:16] - Fixed broken (outdated) invoked-rc.d in ubuntu-release-upgrader [15:16] - Ubuntu touch [15:16] - Tweaked adbd to use nss to find what shell to use [15:16] slangasek: I'm going to do a few drives for people to fill it out though. This isn't going to be a throw the mail over the wall and be done with it thing. [15:16] - Discussed the partitioning plan and fs layout for touch devices. [15:16] - Upstart [15:16] - Looked at and commented on a few merge proposals. [15:16] - Discussed SRU plan for the serialization work. [15:16] - Wrote upstart user session jobs for startkde, xfce4-session and startlxde [15:16] - Tested upstart user session on Kubuntu, Xubuntu, Lubuntu, Ubuntu Studio, Ubuntu GNOME and Edubuntu [15:16] - The plan is to switch all those sessions to run under upstart in a couple of weeks [15:16] ev: ah, cool [15:16] - LXC [15:16] - Answered some questions from the QA/desktop team on using LXC for their desktop tests [15:16] - Some more autostart discussions upstream, I think we have a plan now, just need to implement it [15:16] - Some code reviews [15:16] [15:16] TODO: [15:17] - Finish self-rebuilds feature implementation on nusakan [15:17] - Implement some of the remaining bits for the image based update server tools [15:17] ev: if you're doing drives for legal documents, you could get them to sign the CLA too while they're at it ;) [15:17] - Process some pending merges (ifupdown and resolvconf) [15:17] (DONE) [15:17] stgraber: image-based update server - AIUI from the ticket (and my sync call with IS), this ticket has been dispatched to the DC engineer and we should see some movement soon [15:17] slangasek: how far up the hill do you want me to push that boulder then? [15:18] slangasek: yep, I got pinged by deej yesterday, looks like we've got hardware, just need to have it setup [15:18] ev: :) [15:18] :D [15:19] * in a long-running battle with bug #1152736 [15:19] * worked with stgraber to validate, upload nfs-utils merge; seem to have gotten myself comaintainer status in Debian as an unfortunate side-effect [15:19] bug 1152736 in linux (Ubuntu Saucy) "system swapping itself to death in raring for no good reason" [High,Confirmed] https://launchpad.net/bugs/1152736 [15:19] * working with the phonedations team on the ubuntu-touch container flip: [15:19] * current model is using chrooting under userdata partition, not the read-only system partiton; would like to fix this, aiming to have the system as pure Ubuntu as possible [15:19] * questions about whether udev or ueventd should own the firmware loading [15:20] * Code review of lp:~jamesodhunt/upstart/serialise-remaining-objects [15:20] * discussions with PES about arm64 bring-up; we should be close now to being able to bring the emulated buildds online [15:20] * LPC travel sorted out [15:20] (done) [15:20] slangasek: [fx: anxious noises wanting to play with the emulator] [15:20] cjwatson: heh [15:20] phonedation? [15:20] submitted apport bug LP: #1185515 regaring duplicate signature for apport-package bugs [15:20] Launchpad bug 1185515 in apport (Ubuntu Raring) "SRU of updated apport-package duplicate signature calculation" [Medium,Fix released] https://launchpad.net/bugs/1185515 [15:20] uploaded P, Q and R SRUs for bug 1185515 [15:20] P, Q, R SRU verification of bug 1185515 [15:20] created a tool for checking for error buckets for duplicates of a launchpad bug [15:20] investigation into errors bug 1186215 [15:21] bug 1186215 in Errors "Timed out while loading subscriptions for most-common-problems" [High,Triaged] https://launchpad.net/bugs/1186215 [15:21] doko: the Phone Foundations team, AKA fōndations [15:21] review of phased-updates report [15:21] submitted merge proposal for errors version table sorting issue [15:21] reported bug 1186376 regarding software-properties UI for phased updates options [15:21] bug 1186376 in software-properties (Ubuntu) "should support setting of whether or not to include phased updates" [Medium,Triaged] https://launchpad.net/bugs/1186376 [15:21] searched for pyc corruption bugs with error message in DpkgTerminalLog (e.g. apport-package crashes) [15:21] doko: ogra's team. [15:21] research into bug LP: #1175637 [15:21] Launchpad bug 1175637 in unattended-upgrades (Ubuntu) "Kernel updates are being marked as manually installed" [Undecided,Confirmed] https://launchpad.net/bugs/1175637 [15:21] tested ubuntu-release-upgrader bug LP: #1181470 [15:21] modified sru-review to reject packages too [15:21] emailed xubuntu-devel regarding update-notifier and auto-launch [15:21] removed update-notifier's support for choosing not to auto-launch applications [15:21] f [15:21] Launchpad bug 1181470 in ubuntu-release-upgrader (Ubuntu) "deb-src repositories are not disabled for unknown distributions" [Low,Triaged] https://launchpad.net/bugs/1181470 [15:21] oops [15:21] merge, uploaded my changes to update-notifier that create and use upstart user sessions [15:21] slangasek: ditto on emulator anxiety. :P [15:21] patch piloting [15:21] was attempting to type fóndations to see if I prefer that spelling [15:21] ✔ done [15:21] * progressed with most merges (btrfs-tools left from important ones in [15:21] main, atm updating to merged btrfs-tools fails to boot) [15:21] * demoed U1 plugin ubiquity to design/U1 teams over google hangout a [15:21] few issues got raised that I'm filing bugs/fixing. Also [15:21] incompatability between sso/u1 was discovered, thus still no go [15:21] ahead to enable the plugin by default. [15:22] * helping out with automatic-raid preseeding case for UA/PES [15:22] * implemented ubiquity autopilot proof of concept, for QA team source [15:22] writing tests. [15:22] * reviewing upstart serialisation branch, raised a couple of questions [15:22] * boost1.49 demoted to universe & synced (thanks infinity) [15:22] * started looking into trimming gcc-android cross-toolchain (got rid [15:22] of java dependancy already, still to trim more & proposed patches to [15:22] doko) [15:22] * Generated PDFs of some raring sources & binaries, apparently a few [15:22] more needed. [15:22] done [15:22] infinity, cjwatson: so last I saw, vanhoof was having problems with the git checkout because he couldn't make git ignore ssl certificate verification hard enough. If either of you have access to calabash, maybe you could help [15:23] cjwatson: "fóndations": http://www.youtube.com/watch?v=KhR4RZMWbxw [15:23] I tried, can't get to calabash [15:23] hahahahaha. Susan. [15:23] slangasek: ha [15:23] I, apparently, have access to calabash. [15:25] slangasek: You have more info on that you want to pass me out of band, or should I just poke vanhoof with "Steve said something cryptic involving calabash and WTFery"? [15:25] infinity: mail bounced [15:26] Ta. [15:26] any questions over the above statuseses? [15:26] infinity: do you happen to know if we've flipped that magic switch on ddebs? [15:27] I'm kind of on the wrong timezone to have regular chats with the launchpad folk [15:27] slangasek: my favourite so far http://www.youtube.com/watch?v=zzt2NLU3uSo&list=PLB4CCD4AB5BCBF4E3 [15:27] ddeb magic switch == powerpc cross-retracers === Guest8512 is now known as Zic [15:27] you know you want them [15:27] jodh: :) [15:28] [TOPIC] Bugs === meetingology changed the topic of #ubuntu-meeting to: Bugs [15:28] bdmurray: any course corrections for us? [15:28] slangasek: oops should have been http://www.youtube.com/watch?v=6usx5vS238Y [15:28] jodh: oh, heh [15:29] I could use some help investigating bug 1175637 [15:29] bug 1175637 in unattended-upgrades (Ubuntu) "Kernel updates are being marked as manually installed" [Undecided,Confirmed] https://launchpad.net/bugs/1175637 [15:31] bdmurray: you did that on a precise system with all package-manager-related SRUs up to date? [15:31] (does unattended-upgrade use aptdaemon or not?) [15:32] hmm, it doesn't [15:32] ev: Not flipped yet, no. [15:32] but it is written to python-apt directly [15:32] rubbish [15:32] slangasek: yes regarding package manager SRUs [15:32] thanks though [15:32] ev: Don't make me get elmo in here to dude you. [15:33] bdmurray: so I think it's probably a straightforward matter of unattended-upgrades not calling the right api to mark packages as auto when pulling them in as new deps [15:33] not that I know what that api actually is [15:34] infinity: I'm so not complaining. [15:41] slangasek: okay [15:41] mark_install(from_user=False) I think? But check ... [15:42] ah, there's a single call to mark_install() in u-a, so that would be nice and convenient if so [15:42] bdmurray: is that enough to be going on with? [15:42] slangasek: yep, just need some more kernel updates ;-) [15:42] heh [15:42] or test in a chroot! [15:42] you can mark_auto() on single packages too but you probably don't want to [15:42] right [15:42] bdmurray: anything else? === ubott2 is now known as ubottu === G4MBY is now known as PaulW2U [15:42] slangasek: nope [15:42] okie [15:42] [TOPIC] Ubuntu Touch saucy dogfooding === meetingology changed the topic of #ubuntu-meeting to: Ubuntu Touch saucy dogfooding [15:42] so the Phonedations Team and Desktop Team are at the point where they're ready to finally flip all the Touch development to target saucy exclusively [15:42] (up to this point, some things have been stuck on raring ppas for $reasons) [15:43] please take some time to test those images out when they become available and help iterate them, so that we can assist in nailing down any regressions as quickly as possible [15:43] there are several more major shifts in the pipeline, of which this is only the first [15:43] I gather this morning's images were busted? [15:43] I haven't seen [15:43] cjwatson: well, there is a .1 as well already for today. [15:43] http://cdimages.ubuntu.com/ubuntu-touch/daily-preinstalled/20130605.1/ [15:43] not ubuntu-touch, though [15:43] I'm talking about http://cdimages.ubuntu.com/ubuntu-touch-preview/daily-preinstalled [15:43] ah, yes, the jenkins images [15:43] the /next/ major shift is to get our actual images really and truly built via cdimage infrastructure [15:43] which I believe will also come with the container flip (to boot to Ubuntu first and run Android in a container, instead of the other way around) [15:43] ok. [15:43] if you want to help with this, there's ongoing discussion in #ubuntu-touch about how to make this happen [15:44] [15:44] [TOPIC] AOB === meetingology changed the topic of #ubuntu-meeting to: AOB [15:44] anything else? [15:45] are the containers flipped? [15:45] the spec looks like that's sort of happened, but I haven't updated yet [15:45] they are in the saucy images, when they boot [15:45] no [15:45] conflict! [15:46] stgraber: no, the container flip is separate from the move to saucy [15:46] there will be saucy images, still built in jenkins, ASAP [15:46] then on top of that, we still need to get the container flip done [15:46] I think stgraber means the ubuntu-touch/saucy images [15:47] yes - the images in /ubuntu-touch/ are still a work in progress; help is welcome, but those are separate from the daily images that everyone will be dogfooding right now [15:47] ah yeah, sorry for the confusion, I'm only looking at those from /ubuntu-touch [15:47] aren't the ubuntu-touch ones the ones I want? [15:47] this is confusing - I'm hoping we will be able to clearcut this confusion soon :) [15:47] ev: you probably want the ones that have been on the floor since their creation [15:48] er, /don't/ :) [15:48] ah, probably not :) [15:53] ev: ubuntu-touch-preview - these are the ones that are built in jenkins, are moving to saucy imminently, still have Android as the root, and phablet-tools will install them [15:53] ahhhh [15:53] ev: ubuntu-touch - these are still a work in progress, built on the livefs builders, will have Ubuntu as the root, and have bugs to be hammered out [15:53] right [15:53] thanks, that makes much more sense to me now [15:53] ok :) [15:53] * xnox vaguely wants to upgrade util-linux & include more namespace patches against shadow [15:53] #endmeeting === meetingology changed the topic of #ubuntu-meeting to: Ubuntu Meeting Grounds | Calendar/Scheduled meetings: http://fridge.ubuntu.com/calendar | Logs: https://wiki.ubuntu.com/MeetingLogs | Meetingology documentation: https://wiki.ubuntu.com/meetingology [15:53] Meeting ended Wed Jun 5 15:52:28 2013 UTC. [15:53] Minutes (wiki): http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-06-05-15.03.moin.txt [15:53] Minutes (html): http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-06-05-15.03.html [15:53] that's a wrap [15:53] thanks :) [15:53] thanks! [15:53] any reasonable now-how at testing those? [15:53] thanks. [15:53] thanks! === barry` is now known as barry_ [15:54] slangasek: looks like i got disconnected. i'm guessing the meeting ended just after your aob topic? ;) [15:54] barry_: http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-06-05-15.03.html [15:54] barry_: full log: http://ubottu.com/meetingology/logs/ubuntu-meeting/2013/ubuntu-meeting.2013-06-05-15.03.log.html === LjL^ is now known as LjL === lan3y is now known as Laney === shirgall is now known as Guest96061 [15:55] xnox: cheers! === Laney is now known as Guest71473 === barry_ is now known as barry === Guest71473 is now known as Laney === lool- is now known as lool === jibel_ is now known as jibel === plars is now known as plars-afk === tyhicks` is now known as tyhicks === hobana.freenode.net changed the topic of #ubuntu-meeting to: AOB === Guest88569 is now known as jrib === mmrazik is now known as mmrazik|afk === hallino1_ is now known as hallino1 === arosales_ is now known as arosales === jamespag` is now known as jamespage === tumbleweed_ is now known as tumbleweed === tumbleweed is now known as tumblewed === G4MBY is now known as PaulW2U === tiagoscd is now known as Guest15290 === Guest15290 is now known as tiagoscd === czajkows1i is now known as czajkowski === psivaa is now known as psivaa_AFK === Quintasan_ is now known as Quintasan === mmrazik|afk is now known as mmrazik === yofel_ is now known as yofel === Ursinha is now known as Ursinha-afk === Ursinha-afk is now known as Ursinha === inetpro_ is now known as inetpro === ubott2 is now known as ubottu === yofel_ is now known as yofel === tyhicks` is now known as tyhicks === s1aden is now known as sladen === elky` is now known as elky === Guest37522 is now known as ogasawara === hggdh is now known as ubotu-br === ubotu-br is now known as ubotu-br` === ubotu-br` is now known as hggdh === maxb_ is now known as maxb === Sp4rKy_ is now known as Sp4rKy