[01:34] <nyl> make: *** No rule to make target `frameworks/base/nfc-extras/com.android.nfc_extras.xml'
[01:35] <nyl> do i need to edit manifest.xml?
[01:37] <nyl> erm now it works o.O
[01:41] <computerwizard0> make: *** No rule to make target `out/target/product/tostab03/android-boot.img', needed by `out/target/product/tostab03/obj/PACKAGING/target_files_intermediates/cm_tostab03-target_files-eng.username.zip'.  Stop.
[01:41] <computerwizard0> Everything else builds just fine.  Any thoughts on where to begin looking for a fix to this?
[01:52] <nyl> make: *** [/home/nyl/ubuntu-touch/out/target/product/jfltexx/obj/SHARED_LIBRARIES/libstagefright_intermediates/QCUtilityClass.o] Error 1
[02:03] <nyl> make: *** [/home/nyl/ubuntu-touch/out/target/product/jfltexx/obj/SHARED_LIBRARIES/libstagefright_intermediates/AwesomePlayer.o] Error 1
[02:06] <OrokuSaki> Anyone know how to set the natural orientation for my device?? or any device??? or how to check it?
[02:06] <OrokuSaki> portrait vs landscape?
[02:06] <OrokuSaki> I remember quantal had a way...
[02:15] <mhall119> OrokuSaki: I don't think there ever was a way to change the default orientation of the shell
[02:16] <mhall119> nor will there be until we switch to Mir
[02:16] <mhall119> or so I've been told
[02:17] <OrokuSaki> I guess my issue is the camera app.. it always wants to display my camera as landscape, when it should be portrait..
[02:18] <RobbyF> mhall119, I'll paypal $15 bucks for a g+ app that works like gmail.
[02:18] <OrokuSaki> Maybe I can recompile the camera app...
[02:18] <mhall119> RobbyF: a webapp?
[02:18] <RobbyF> yeah - lol
[02:18] <mhall119> it'd just need a .desktop file and icon
[02:19] <RobbyF> I started to look at the directions and the first thing said SDK and API then I said that sounds wrong.
[02:19] <RobbyF> packaged for the click webstore?
[02:19] <mhall119> RobbyF: by hand
[02:25] <OrokuSaki> http://pastebin.com/TjXa08NM So I notice some good information from the source code to the camera app..
[02:25] <OrokuSaki> I am like the Nexus with a sideways mounted camera... How can I correct the orientation? Recompile the camera app for my device?? Set the orientation of the camera to -90???
[02:37] <annerajb> nyl, you solved your errors?
[02:42] <OrokuSaki> The SDK is nice.
[02:50] <mhall119> RobbyF: very nearly got it
[02:50] <mhall119> RobbyF: I'll try again in the morning
[02:50] <RobbyF> k
[02:51] <RobbyF> easiest 15 ever, but i'd be more impressed if you made it a click package so i just tapped install :)
[02:51] <mhall119> RobbyF: http://ubuntuone.com/27mCO1Epcfixa5gXAaI21i if you want to see what I have so far
[02:51] <mhall119> as a click package
[02:52] <RobbyF> wifes yelling at me. toss it in the beta web store :)
[02:52] <RobbyF> bug ya tomorrow
[02:56] <annerajb> damm it i wonder why upstart is getting stuck
[03:54] <Mirv> xnox: hmm, that does not sound familiar, so no, no idea.
[03:55] <Mirv> xnox: google gives cross-compilation related results regarding that first. so are you doing such, or something else than debuild like own ./configure line?
[04:49] <Guest90470> hello
[04:49] <Guest90470> any guidance on installing touch on galaxy tab 2?
[07:03] <dholbach> good morning
[07:06] <lool> oy
[07:08] <lool> stgraber: When you come up, let's look at renumbering the daily{,-proposed} images and announcing this image update thing   :-)
[07:10] <xnox> Mirv: nope, just what's currently in the archive. I'll try the pre-released 5.1.1 packages now instead.
[07:11] <Mirv> xnox: ok. check https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-beta2/+packages if you want the latest packaging (mostly resyncs with Debian), or simply  lp:~kubuntu-packagers/kubuntu-packaging/qtbase-opensource-src
[07:12] <Mirv> I'll copy them over to the in-use PPA once I'm ready with the new round of builds
[07:32] <dholbach> does anyone know why the ureadit app can't be opened?
[07:33] <dholbach> (on 20130902.1)
[07:44] <dholbach> dpm, can you install ureadit from the software store?
[07:44] <dholbach> ^ maybe lool, too
[07:44] <dholbach> or rather, can you launch it, once it's installed
[07:45] <asac> ogra_: moin
[07:45] <asac> ogra_: any idea if our .1 build fell apart yesterday?
[07:46] <spixon> Hi all
[07:46] <dpm> dholbach, let me update to 20130902.1 - is that the image you get with --pending?
[07:46] <asac> psivaa: hi! :)
[07:46] <dpm> yep, it is
[07:46] <asac> psivaa: http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/20130902.1/ - that build never made it to the dashboard
[07:46] <asac> any idea?
[07:49] <dholbach> dpm, it is - or just try installing it on whatever you're on
[07:49] <lool> dholbach: got a manifest error
[07:49] <lool> dholbach: but not sure I'm on the right image
[07:49] <dholbach> lool, me too
[07:50] <dpm> flashing now
[07:50] <dholbach> beuno, mhall119: do you get the manifest error too, when launching ureadit after having installed it from the software store?
[07:50] <spixon> Hi asac..that link you gave generic image ...can stright install in tablet?
[07:51]  * lool reaches out straight for strace this time
[07:51] <dpm> lool, if I've got a C++ QML extension I want to cross-compile for arm on my desktop... how do I do it? Do we have instructions somewhere to set up an environment for cross-compiling for arm somewhere?
[07:52] <lool> dpm: so we want to make this super easy, but for now the only thing we have are older instructions for cross-building debian/ubuntu packages
[07:52] <lool> dpm: qmake in particular might introduce extra pain
[07:52] <dholbach> dpm, IIRC xnox wanted to put together some instructions (or at least some bullet points?)
[07:53] <lool> dpm: one doc I found recently but didn't have to review in full is https://wiki.linaro.org/Platform/DevPlatform/CrossCompile/CrossbuildingQuickStart
[07:53] <asac> spixon: well, depends which tablet basically
[07:53] <ogra_> asac, looks all fine
[07:53] <asac> spixon: you should use phablet-flash
[07:53] <lool> it covers sbuild cross-compilation and also xdeb
[07:53] <dpm> lool, yeah, I understand and I agree that we should make it super easy, I'm just wondering how to do it at all (simply to figure it out myself) beforehand
[07:53] <ogra_> no idea why utah didnt pick up
[07:53] <lool> dpm: I'm interestedin the outcome of your research of where we stand, in fact I wanted to do something similar this week
[07:53] <ogra_> http://people.canonical.com/~ogra/touch-image-stats/20130902.1.changes
[07:54] <ogra_> gnome-control-center is gone ...
[07:54] <asac> spixon: https://wiki.ubuntu.com/Touch/Install
[07:54] <spixon> asac.. I new in ubuntu....try to get some idea and basic guide
[07:55] <asac> spixon: great! welcome :) so if you have a device that we currently support give it a try. check out the wiki above. one prerequisite we currently assume is that you have ubuntu installed on your laptop/desktop already
[07:56] <spixon> asac Thanks
[07:56] <lool> dholbach: http://paste.ubuntu.com/6058027/
[07:56] <lool> dholbach: so just update your client to latest version
[07:57] <dholbach> karamba!
[07:57] <lool> dholbach: it's something worth bringing up with cjwatson once he's around
[07:57] <lool> dpm: do you have a branch with the app to cross-build?
[07:57] <lool> dpm: would actually help me to have a real life app to look at for today  :-)
[07:58] <dpm> dholbach, manifest error
[07:58] <dholbach> lool, you mean passing through a more descriptive error message?
[07:58] <lool> dholbach: well, how do we deal with this in the user experience
[07:58] <psivaa> asac: the jenkins publisher plugin had died last night. i've restarted it now.
[07:58] <dholbach> lool, hum, there's no newer click in the apt repository
[07:58] <lool> dholbach: certainly click will be updated post 13.10, we need a way to support this mismatch
[07:59] <psivaa> asac: there is a large queue at the moment. so it may take a while for the whole results to show up in the dashboard
[07:59] <lool> either we make sure SDK releases include a click version included in our supported images (in production), or we allow multiple versions in the appstore and have the client specify its supported versions etc.
[07:59] <dpm> lool, yeah, I'm using PPAs to do the cross-building for now. Here's the branch for my QML C++ extension (https://code.launchpad.net/qzxing/) which is used by this QML frontent (https://launchpad.net/qreator/touch)
[08:00] <dholbach> dpm, what did you mean by "update your client to latest version"?
[08:00] <lool> dholbach: my device has 0.3.4 and saucy has 0.4.1
[08:00] <lool> dholbach: perhaps I'm getting a different error than you then?
[08:00] <lool> dholbach: let me retry after flashing yesterday's image
[08:00] <dholbach> I have 0.4.1 installed on my device
[08:00] <dholbach> hang on let me double check
[08:00] <dpm> dholbach, where did I say that?
[08:00] <lool> dholbach: aha, so different issue than mine
[08:01] <asac> psivaa: thx!
[08:02] <lool> dpm: you're build-deping on cross-compilers to do the cross-compilation in a PPA or you're using a PPA which builds natively for armhf but with qemu?
[08:02] <lool> the latter isn't really cross-compilation
[08:02] <lool> (not trying to make a point but more understanding what you do)
[08:02] <dholbach> lool, http://paste.ubuntu.com/6058044/
[08:03] <lool> dholbach: is this with click install launched as root?
[08:03] <dholbach> yes
[08:05] <dholbach> interesting, /opt/click.ubuntu.com/com.ubuntu.developer.mhall119.ureadit/0.9.2/ only contains a ".click" directory
[08:05] <dpm> lool, I simply asked in #launchpad to enable armhf builds for that PPA, and I've got a recipe in LP that does daily builds. So I believe it's the latter (although I don't know the internals of LP as to whether it uses qemu)
[08:07] <lool> dholbach: is this a clean install?  I think it could either be a permission issue with running click in different ways, or it could be another bug in the LD_PRELOAD stuff
[08:07] <lool> dholbach: I'm flashing to latest version to reproduce
[08:07] <lool> dpm: yes indeed
[08:07] <dholbach> let me --wipe
[08:07] <lool> dpm: thanks for the source, will try cross-building it too, albeit I see it's qmake
[08:08] <WebbyIT> hi timp, I see you did the header element in SDK, shall you help me with a similar thing, please?
[08:08] <dpm> lool, yeah, I used SDK and Qt Creator as much as possible to exercise our tools (and to make it easier for me :)
[08:09] <lool> dpm: rightfully so
[08:09] <lool> xnox: did you have a specific qml extension in mind to serve as a template for cross-compilation?
[08:10] <OrokuSaki> Trying to create a package in qtcreator.. anyone know how to get past this?? FAIL!  : QmlTests::tst_ViewFinderGeometry::compile() module "QtTest" is not installed
[08:10] <xnox> lool: just similar to the current one: bog standard empty extension template, or bog standard empty extension with page stack template.
[08:10] <OrokuSaki> I have QtTest installed
[08:21] <dholbach> lool, freshly installed system - same problem
[08:21] <dholbach> lool, I'll file a bug
[08:22] <lool> dpm: so I tried cross-building it with sbuild
[08:22] <lool> dpm: found a couple of bugs
[08:22] <lool> dpm: it does find the cross-compiler just fine
[08:23] <dholbach> lool, maybe click can't deal with symlinks?
[08:23] <lool> dpm: but this was written for Debian and just sets APT::Architecture to list armhf which would require http://archive.u.c to have armhf; so I had to edit the chroot to list deb [arch=armhf] http://ports.u.c saucy main... and disable apt-get update (--no-apt-update); then this failed while resolving cross-bdeps, presumably because I had disabled updates
[08:23] <lool> dpm: I'll have to patch sbuild to go further
[08:24] <lool> dholbach: ok, finished flashing here and looking at getting a strace
[08:25] <dholbach> there's a lot of stuff in the click which could probably be stripped out ... if we had a proper plan for that
[08:25] <dholbach> (not just .bzr)
[08:27] <dpm> ok, thanks lool
[08:28] <lool> dholbach: hmm with latest image from daily-proposed (equivalent of pending), it sticks at 100% and doesn't install
[08:29] <lool> dholbach: xda worked and launched; trying ureadit again it stays at 0%
[08:30] <lool> grmpf
[08:31] <lool> this is so fragile
[08:31] <dholbach> lool, eventually the scope tells me that it's installed (on grouper if that matters)
[08:31] <lool> on grouper too
[08:31] <lool> dholbach: usually it takes a little while but installs
[08:31] <dholbach> yep
[08:32] <lool> but it didn't here
[08:32]  * lool ctrl-alt-deletes his grouper
[08:33] <dholbach> on the other hand we were bound to run into corner cases in the beginning
[08:34] <lool> ok, upon reboot and without strace, installation happened quickly
[08:35] <lool> dholbach: do you have a bug id?
[08:36] <dholbach> no, not yet - I was waiting for confirmation that somebody else ran into http://paste.ubuntu.com/6058044/ as well
[08:36] <lool> dholbach: I could reproduce the issue under strace now
[08:37] <dholbach> same traceback?
[08:39] <asac> Mirv: sil2100: hi
[08:39] <asac> Mirv: sil2100: any idea if we hav ea rpoblem with upstream merger?
[08:39] <asac> https://jenkins.qa.ubuntu.com/job/generic-mediumtests-builder-saucy-i386/2908/console
[08:39] <asac> that one seems a bit unhhappy
[08:41] <sil2100> asac: hi! Not that I am aware of, try a re run for this - since it looks to me like one of those transient errors
[08:41] <sil2100> asac: we had that once when Martin was leaving
[08:41] <asac> sil2100: the log seems to be full of retries: https://code.launchpad.net/~schwann/gallery-app/gallery-atest-toolbar-opened/+merge/183195
[08:43] <asac> gusch: have you retried?
[08:43] <asac> i see two tries aftrer your last merge
[08:43] <asac> maybe we can retry a third time?
[08:43] <lool> dholbach: No, I don't understand why it fails yet
[08:43] <lool> dholbach: I see dpkg-split failing in the strace
[08:43] <gusch> asac: well I retried several times before the last change
[08:44] <ogra_> xnox, happy birthday !
[08:44] <gusch> asac: and it's alway the same exception during the tab switching
[08:44] <asac> xnox: HB!
[08:45] <lool> dholbach: ah nevermind, this exit 1 error code of dpkg-split seems normal operation
[08:45] <gusch> asac: I started a rebuild, but I'm almost sure it will fail again
[08:45] <asac> gusch: so the merge requests only do build testing
[08:45] <asac> no?
[08:45] <asac> gusch: e.g. you are not stuck at the tab-switching-being-unreliable state
[08:46] <lool> dholbach: ok: 1893  lchown32("/opt/click.ubuntu.com/com.ubuntu.developer.mhall119.ureadit/0.9.2/debug/hot.json.dpkg-new", 0, 0) = -1 EPERM (Operation not permitted)
[08:46] <asac> but rather that the upstream merger bails out during building,no?
[08:46] <lool> dholbach: could it be that the .click includes files for another user?
[08:46] <gusch> asac: I'm totally stuck on this tab switching issue
[08:46] <asac> gusch: this merge request is stuck? https://code.launchpad.net/~schwann/gallery-app/gallery-atest-toolbar-opened/+merge/183195
[08:47] <lool> dholbach: hmm no, these files are indeed just symlinks
[08:47] <sil2100> asac, gusch: ah, this one... I still don't know what's wrong with it, I was waiting for US to be back from holidays
[08:47] <gusch> asac: yes - and therefore I can't merge https://code.launchpad.net/~schwann/gallery-app/gallery-conent-url/+merge/182861 either
[08:47] <lool> dholbach: well I think you're right, it's an issue specific to symlinks
[08:47] <asac> gusch: right. so how can i find the tab switching issue from there?
[08:47] <dholbach> lool, I'll file a bug
[08:47] <asac> gusch: i only see that UNSTABLE jobs fail because they have problems with GPG keys from archive
[08:48] <xnox> ogra_: asac: cheers!
[08:48] <gusch> sil2100: I managed to fix the friend component issue - now it's another one
[08:48] <asac> sil2100: i have not idea where to see the tab switch issue in the logs above
[08:48] <gusch> asac: https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-saucy/2471/testReport/junit/gallery_app.tests.test_album_view/TestAlbumView/test_add_photo_to_new_album_with_mouse_/
[08:48] <lool> dholbach: thanks, I'll attach the strace and quote the relevant part once you have the LP
[08:48] <sil2100> gusch: by changing the dependency, yes?
[08:49] <JamesTait> Good morning all, happy Welsh Rarebit Day! :-D
[08:49] <sil2100> gusch: so now it's blocked by the thing that asac pointed out?
[08:49] <gusch> sil2100: yes - I now set the version number qtdeclarative5-ubuntu-ui-extras0.1 (>= 0.1+13.10.20130829-0ubuntu1),
[08:49] <gusch> sil2100: correct
[08:49] <asac> agesil2100: there are more than one UNSTABLE job in the log. I just picked a random one and found that there is GPG key busted
[08:49] <asac> sil2100: there are more than one UNSTABLE job in the log. I just picked a random one and found that there is GPG key bustage
[08:50] <sil2100> It's really hard to track those issues, since every time I see something different
[08:50] <sil2100> gusch: since look, the  latest mako mediumtests:
[08:50] <gusch> sil2100: the moment the firend issue was solved, the tabs stopped working on jenkins - they work fine locally (even on my GN)
[08:50] <sil2100> https://jenkins.qa.ubuntu.com/job/generic-mediumtests-runner-mako/442/?
[08:51] <gusch> asac: well - that GPG happens from time to time - a rebuild usually fixes that
[08:51] <asac> gusch: what does UNSTABLE mean?
[08:51] <gusch> asac: FAILED
[08:51] <asac> doesnt it mean that the job succeeded and just previous ones failed?
[08:51] <asac> gusch: the log before explicitely says: FAILURE
[08:52] <sil2100> asac: UNSTABLE means that the tests were run but had failures
[08:52] <sil2100> asac: a FAILURE would mean that the tests even didn't really run, or something bad happened
[08:52] <asac> yeah
[08:53] <asac> gusch: sionce when do you see this ui-toolkit issue?
[08:53] <gusch> asac: I think if it says FAILURE, then it's a jenkins "internal error" - UNSTABLE means the tests failed (or at least one)
[08:53] <gusch> asac: I have seen it from time to time since I use the SDK autopilot funkction - but since yesterday they are there more or less constant
[08:54] <Laney> mpt: https://wiki.ubuntu.com/SecurityAndPrivacySettings#Phone "Unlock the phone using:" titles - is title case right there?
[08:55] <gusch> sil2100: yea - so there is a friend issue and the tabs issue combined
[08:55] <asac> gusch: do i see the test that fails for you during merge here as well: http://reports.qa.ubuntu.com/smokeng/saucy/image/3923/gallery-app-autopilot/ ?
[08:55] <gusch> sil2100 asac I filed a bug for the tabs switching autopilot issue https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1220064
[08:56] <sil2100> gusch: I wonder if Omer could help us out here
[08:56] <asac> omer seems to be not on
[08:56] <gusch> sil2100: I asked him yesterday - he recommended a rebuild - then I went to bed
[08:57] <seb128> gusch, hey
[08:58] <seb128> gusch, you are looking at those gallery tests issues?
[08:58] <gusch> asac: the one failing test there is the one that I'd like to get fixed with that MS ...
[08:58] <gusch> seb128: yes - thanks for your help yesterday - setting the version seems to have fixed that issue
[08:58] <asac> gusch: ok but in the MP it is something else failing?
[08:58] <gusch> seb128: now there is a new one
[08:58] <seb128> gusch, great!
[08:58] <sil2100> Ok, so we have a big depdendency chain
[08:58] <asac> or is it exactly that test?
[08:59] <asac> wow, those MP even have videos for every single test run :)
[08:59] <sil2100> Failure-dependency chains
[08:59] <mpt> Laney, yes
[08:59] <sil2100> seb128: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1220064
[08:59] <gusch> asac: yes - exactly that test that should get fixed - but I'm not able to land that fix, as all the time some tests fail :(
[09:00] <Laney> mpt: okely dokely, ta
[09:00] <gusch> asac: those videos can be very helpful :)
[09:00] <bzoltan> gusch:  we need a bit more context than this bug report :)
[09:01] <sil2100> oSoMoN: ping
[09:01]  * sil2100 just wants gallery-atest-toolbar-opened merge in trunk already
[09:02] <oSoMoN> sil2100: pnog
[09:02] <oSoMoN> pong
[09:02] <sil2100> oSoMoN: hi! ;)
[09:03] <gusch> bzoltan: what context would you like to get?
[09:03] <sil2100> oSoMoN: so, we have a situation here in the AP world, and we're not really sure if it's the app test doing something or the SDK itself
[09:03] <oSoMoN> sil2100: can you elaborate on the situation?
[09:03] <bzoltan> gusch: like how can I reproduce the issue locally
[09:04] <sil2100> oSoMoN: it seems jenkins/otto is hanging up after a failure in one of the webbrowser AP tests when running on the SDK stack, something like this:
[09:04] <sil2100> oSoMoN: http://paste.ubuntu.com/6057961/
[09:04] <gusch> bzoltan: I'm not able to reproduce it locally - not even on my galaxy nexus
[09:04] <sil2100> oSoMoN: Mirv filled in a bug probably related to that: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1220056
[09:04] <bzoltan> gusch: then how can I help you?
[09:05] <lool> cjwatson: Hey, I saw you poked sbuild's lib/Sbuild/ResolverBase.pm but I'm surprized things worked for you: for me when using --host=armhf on a saucy chroot, it adds armhf to APT::Architecture and fails to apt-get update; it tries to download armhf from deb http://mirror.d.o/ubuntu saucy (I also have an ubuntu-ports mirror); I guess it would happen the same with archive.u.c vs. ports.u.c so I'm curious on how you got it working?  or were you testing
[09:05] <sil2100> oSoMoN: bzoltan said to ping you first, but I guess this might not be completely related to the webbrowser-app
[09:05] <gusch> bzoltan: maybe find someone to debug it on jenkins
[09:05] <dholbach> lool, filed bug 1220125
[09:05] <Mirv> sil2100: so, it seems like the apps tests actually only now started to be run with the sdk stack, even though no changes were visibly done in cu2d-config? or did you redeploy sdk stack yesterday or something?
[09:06] <Mirv> sil2100: after all, there has been a critical bug against gallery-app about tests since Friday
[09:06] <sil2100> Mirv: no, I didn't redeploy anything yesterday - we had the change that SDK runs some apps tests since a month already
[09:06] <oSoMoN> sil2100: this error looks suspicious: file:///usr/share/webbrowser-app/Browser.qml:201 Invalid property assignment: "opened" is a read-only property
[09:06] <Mirv> sil2100: did you look at the previous "green" check runs of SDK stack already? it just says that they were skipped
[09:06] <sil2100> Mirv: we added those apps tests to SDK to make sure SDK doesn't cause regressions in existing applications
[09:07] <oSoMoN> sil2100: it seems to indicate that the Panel API changed
[09:07] <Mirv> sil2100: but now only yesterday sdk stack really started running those tests
[09:07] <gusch> bzoltan: maybe the SDK autopilot test should be extended to add a test with an app like gallery - 3 tabs, and the second one is active - maybe that'll fail as well (on jenkins)
[09:07] <oSoMoN> timp: hey, was there a recent change in the SDK’s Panel API?
[09:07] <bzoltan> gusch: MRs are welcome :)
[09:08] <sil2100> Mirv: that's basically impossible, since that change is old and Didier redeployed the stack at least twice during that time
[09:08] <cjwatson> lool: https://wiki.ubuntu.com/CrossBuilding has instructions for building a working chroot
[09:08] <gusch> bzoltan: yea - looking at it
[09:08] <sil2100> And yesterday no one did anything with the stacks, if Robert didn't of course
[09:08] <cjwatson> lool: I have used this quite extensively, it does work
[09:08] <bzoltan> gusch: I pinged timp, who is behind the tabs component. He is offline for some time, but will jump on it when he is back.
[09:08] <cjwatson> lool: mk-sbuild puts the right [arch=foo] bits in sources.list etc.
[09:08] <sil2100> Mirv: and I remember seeing App test failures on the SDK stack before too
[09:08] <Mirv> sil2100: I'm just looking at http://10.97.0.1:8080/view/cu2d/view/Head/view/SDK/job/cu2d-sdk-head-2.2check/
[09:09] <cjwatson> dholbach: sigh, I fixed that yesterday but didn't upload
[09:09] <Mirv> sil2100: so they've probably been run before but maybe they just were blocked on other stacks during the weekend or something like that, but still showed up as green
[09:09] <cjwatson> seems five people find the same bug in parallel an awful lot
[09:09] <oSoMoN> sil2100: got it, revision 734 in the UITK: CHANGED IN Panel property bool opened TO readonly property bool opened
[09:09] <sil2100> Mirv: https://jenkins.qa.ubuntu.com/job/autopilot-saucy-daily_release/1404/label=autopilot-intel/testReport/?
[09:10] <Mirv> sil2100: yes, I'm seeing app tests succesfully ran on Friday http://10.97.0.1:8080/job/autopilot-saucy-daily_release/1452/label=autopilot-intel/artifact/results/autopilot/autopilot.log
[09:10] <sil2100> Mirv: this shows that those were run
[09:10] <lool> cjwatson: Thanks, will check what mk-sbuild does
[09:10] <oSoMoN> sil2100: I wasn’t aware of this change, I’ll have to update the browser’s code, and I’m guessing that other applications will have to be adapted to
[09:10] <oSoMoN> too
[09:10] <lool> dholbach: added details to the bug
[09:10] <cjwatson> lool: please, bug 1220125 is already under control, you can stop :)
[09:10] <Mirv> sil2100: so it's just that the bug you opened against gallery-app on Friday hasn't been fixed, and even if there weren't additional failing tests that would still block SDK stack
[09:10] <sil2100> oSoMoN: ok! Thanks! Not sure if this will fix our hang-up bug, but at least we know now what happened
[09:10] <lool> cjwatson: I've stopped; this is information from this morning's debugging
[09:10] <seb128> bzoltan, ^ the change of propery in http://bazaar.launchpad.net/~ubuntu-sdk-team/ubuntu-ui-toolkit/trunk/revision/734 is creating those issues
[09:11] <oSoMoN> sil2100: well, that’s at least one problem identified, let’s hope it’s the only one
[09:11] <cjwatson> it was under control yesterday, I just ran out of time to upload
[09:11] <sil2100> Mirv: yes, the bug I opened has a merge for fixing that bug, but it can't get merged in because of mediumtests problems now
[09:11] <lool> cjwatson: I hadn't realized it was reported yesterday
[09:11] <sil2100> Mirv: here's the fix for the gallery issue: https://code.launchpad.net/~schwann/gallery-app/gallery-atest-toolbar-opened/+merge/183195
[09:11] <sil2100> Mirv: btw. ^ see my discussion with oSoMoN
[09:11] <seb128> bzoltan, just as a fyi, seems like oSoMoN is going to fix the browser
[09:11] <lool> cjwatson: it was filed this morning, wasn't aware you knew about the symlink issue, was it mentioned here and I missed it?
[09:11] <sil2100> Mirv: bzoltan SDK guys changed something again and it will be fixed by oSoMoN in the app code, not sure if that will help in unblocking SDK or not
[09:12] <sil2100> seb128: ^
[09:12] <oSoMoN> sil2100: just to make it clear, I’ll fix it in the browser, but other apps might need updates too
[09:12] <cjwatson> lool: yes, it was
[09:12] <lool> cjwatson: I see the commit in lp:click now
[09:13] <Mirv> sil2100: oSoMoN will fix browser, gallery-app merge request needs to go in and we still need a fix for notes-app, right?
[09:13] <lool> cjwatson: indeed, I see this in the backlog
[09:13] <lool> ok
[09:13] <Mirv> specifically notes_app.tests.test_create.TestCreate.test_note_expand_and_unexpand(with mouse)
[09:13] <lool> well that was an hour wasted this morning then
[09:13] <oSoMoN> nerochiaro: ping
[09:13]  * lool &
[09:13] <Mirv> and oSoMoN is apparently pinging nerochiaro for that precisely :)
[09:13] <oSoMoN> yup :)
[09:14] <cjwatson> click 0.4.2 uploaded now
[09:16] <nerochiaro> oSoMoN: hello ?
[09:17] <bzoltan> oSoMoN: sil2100:  timp has made the opened property to be read-only ... is your app writingt that property?
[09:17] <oSoMoN> bzoltan: yes, and I’m taking care of updating the app’s code
[09:17] <bzoltan> oSoMoN: sil2100:  timp said that he checked no app is writing it
[09:17] <nerochiaro> oSoMoN: something broken in notes tests ?
[09:17] <oSoMoN> bzoltan: then he didn’t check thoroughly :)
[09:17] <bzoltan> oSoMoN:  we can revert that change in no time, if you want
[09:18] <oSoMoN> nerochiaro: an API change in the SDK, the 'opened' property of the toolbar/panel was made read-only
[09:18] <bzoltan> oSoMoN:  It seems so... and he will be severely punished for that :)
[09:18] <nerochiaro> oSoMoN: i don't think i set that explicitly in tests, but i'll check
[09:18] <oSoMoN> nerochiaro: it seems the notes-app was writing to this property, can you take care of updating the code to the new API (see rev 734 in uitk)
[09:19] <oSoMoN> nerochiaro: not in tests, but in NotesApp.qml, line 72: page.tools.opened = false
[09:20] <asac> gusch: sil2100: one thing i noticed is that we still add the qt edgers ppa for those test runs
[09:20] <nerochiaro> oSoMoN: bzoltan: got it. so how does one force the toolbar to close ?
[09:20] <OrokuSaki> Morning all...
[09:20] <oSoMoN> nerochiaro: there is a close() method now
[09:20] <OrokuSaki> Anyone help me out with this one?  Trying to create a package in qtcreator.. anyone know how to get past this?? FAIL!  : QmlTests::tst_ViewFinderGeometry::compile() module "QtTest" is not installed
[09:20] <nerochiaro> oSoMoN: excellent
[09:21] <asac> gusch: check out what it pulls in here from ppas: http://10.97.2.10:8080/job/generic-mediumtests-runner-saucy/2472/consoleFull
[09:21] <bzoltan> oSoMoN: nerochiaro: soon there will not be... I am reverting that bloody change
[09:21] <OrokuSaki> Trying to compile the camera-app
[09:21] <oSoMoN> bzoltan: what?
[09:21] <nerochiaro> bzoltan: can't we just wait then ?
[09:22] <oSoMoN> bzoltan: you mean you are reverting revision 734, and we shouldn’t update the apps’ code?
[09:22] <bzoltan> oSoMoN: Well.. removing a writable property is certainly an API break what should not be tolerated without proper documentation and notice to the app developers.
[09:22] <oSoMoN> bzoltan: agreed
[09:23] <oSoMoN> bzoltan: ETA of the revert?
[09:23] <asac> ogra_: Get:78 http://ppa.launchpad.net/phablet-team/ppa/ubuntu/ saucy/main gir1.2-messagingmenu-1.0 i386 1:13.10.0phablet1ubuntu1 [3,112 B]
[09:23] <gusch> asac: I don't see what pullt from that ppa, but who can remove it?
[09:23] <asac> gusch: you can try with the ppa locally
[09:23] <asac> and see if you can reproduce
[09:23] <bzoltan> oSoMoN:too bad, that the close(), open() functions and the hideTimeout property will go away for now too
[09:24] <asac> gusch: Get:24 http://ppa.launchpad.net/ubuntu-sdk-team/ppa/ubuntu/ saucy/main qtdeclarative5-unity-action-plugin i386 1.0.0+13.10.20130716bzr29saucy0 [21.5 kB]
[09:24] <asac> just search for ppa.launchpad.net
[09:24] <asac> you will find the entries
[09:24] <ogra_> asac, i386 ?!?
[09:24] <asac> ogra_: well, we test on all archs. yes
[09:25] <asac> primarily in a fake env on x86 until we have the emulator
[09:25] <ogra_> heh, that will need a lot of work to work ... seed cleanup etc
[09:25] <asac> ogra_: but still this gir thing is scary
[09:25] <bzoltan> oSoMoN: few minutes
[09:25] <oSoMoN> bzoltan: thanks
[09:25] <asac> ogra_: cant we just kill stuff from the ppa?
[09:25] <asac> like delete everything that is in the archive?
[09:25] <ogra_> asac, agreed, still the ubuntu touch seed wont "juts work"
[09:25] <asac> ogra_: i dont know what they pull in and how ... :)
[09:25] <oSoMoN> sil2100: looks like the change in the SDK will be reverted, so app tests should work again
[09:25] <bzoltan> oSoMoN: well.. it is not something we deserve thanks
[09:26] <asac> gusch: do you know how the stuff is run in the automation at all?
[09:26] <ogra_> asac, lets wait for sergiusens and rsalveti ... that PPA isnt only needed for images and not only for armhf
[09:26] <asac> gusch: e.g. how to reproduce something close to that?
[09:26] <oSoMoN> bzoltan: that’s ok, we’re humans, we all make mistakes, what’s really important is to recognize them and fix them asap
[09:26] <sil2100> bzoltan, oSoMoN, gusch: so, the gallery-app fix won't be needed?
[09:26] <ogra_> asac, i'd rather just see the demo packages fixed and in the archive so we can just disable the PPA completely
[09:26] <bzoltan> oSoMoN:true :)
[09:27] <asac> ogra_: disabling? see how much stuff explicitely pulls it in
[09:27] <ogra_> but again that needs cleanup first
[09:27] <asac> ogra_: we want to remove it
[09:27] <gusch> asac: I can run the test locally and run them on the device - but no idea how it's done on jenkins
[09:27] <asac> kill it
[09:27] <asac> dead
[09:27] <ogra_> asac, nonsense
[09:27] <asac> gusch: they run them in a chroot
[09:27] <asac> gusch: with the sdk and phablet ppa enabled
[09:27]  * sil2100 doesn't have too much knowledge about mediumtest jobs
[09:27] <ogra_> asac, its not the buckets fault that you spilled its water on the street :P just dont use the PPA
[09:28] <asac> ogra_: well, we cleaned it
[09:28] <asac> now lets kill it
[09:28] <asac> we can move to another one
[09:28] <asac> at least i want to find all infrastructure that still pulls it
[09:28] <asac> so be a nomad
[09:28] <ogra_> asac, no that will most likely break jenkins test images
[09:28] <asac> yes, and thats what i want :)
[09:28] <asac> after giving warning and trying to move away of course
[09:28] <ogra_> asac, the seeds and metapckage need cleanup ... they both have it enabled
[09:29] <ogra_> but we only can clean that up once the debo stuff is sorted
[09:29] <ogra_> *demo
[09:29] <asac> gusch: so what you do is what is in the beginning of the log
[09:30] <asac> sil2100: do you know where the code is?
[09:30] <asac> that deals with setting up the jenkins?
[09:30] <asac> like apt-get install, pinning etc. all the stuff i see there
[09:30] <asac> ?
[09:31] <ogra_> asac, https://launchpad.net/~phablet-team/+archive/ppa/+packages?field.name_filter=&field.status_filter=published&field.series_filter=saucy
[09:31] <sil2100> asac: sadly I would have to start looking for that, since as I said I don't have too much to do with mediumtests, it's something that Francis and the touch guys have most experience with
[09:31] <ogra_> asac, i'm not sure what you guys are doing there, but there is definitely no gir
[09:31] <asac> ogra_: they are pinning
[09:31] <asac> so might be fallout from that
[09:32] <ogra_> asac, what are they pinning ? there is no such package
[09:32] <asac> gusch: so ... maybe really try the latest SDK and qt5 edgeers ppa content
[09:32] <asac> locally
[09:32] <asac> maybe you can reproduce the problem
[09:32] <gusch> asac: I'll check
[09:32] <asac> gusch: would be cool to know because if it really causes this we have to hold back further sdk releases etc.
[09:33] <gusch> asac: I did at least an update this morning to the latest SDK - but not edgers
[09:33] <asac> gusch: which sdk version is that?
[09:34] <asac> Get:7 http://ppa.launchpad.net/ubuntu-sdk-team/ppa/ubuntu/ saucy/main libunity-action-qt1 i386 1.0.0+13.10.20130716bzr29saucy0 [56.5 kB]
[09:34] <asac> Get:24 http://ppa.launchpad.net/ubuntu-sdk-team/ppa/ubuntu/ saucy/main qtdeclarative5-unity-action-plugin i386 1.0.0+13.10.20130716bzr29saucy0 [21.5 kB]
[09:34] <asac> Get:31 http://ppa.launchpad.net/ubuntu-unity/daily-build/ubuntu/ saucy/main ubuntu-ui-toolkit-theme i386 0.1.46+13.10.20130903.1-0ubuntu1 [188 kB]
[09:34] <ogra_> asac, so inspecting further this looks like a massive failure in launchpad !
[09:35] <asac> gusch: also seems we pull in some unity ppa as you can see above
[09:35] <asac> lots of stuff coming down from there
[09:35] <ogra_> asac, there actually is gir1.2-messagingmenu-1.0_13.10.0phablet1_i386.deb ... but it is in the raring pocket
[09:35] <ogra_> asac, this package should never be provided by LP in saucy
[09:35] <asac> ogra_: guess it automatically gets linked/copied if yoyu dont do anything?
[09:35] <ogra_> cjwatson, ^^^ do you have an idea ?
[09:35] <cjwatson> ogra_: In what archive?
[09:36] <gusch> asac: version is 0.1.46+13.10.20130830.1-0ubuntu1
[09:36] <ogra_> asac, no, it doesnt show up in any saicy search and according to the UI it ios only available in quantal and raring
[09:36] <asac> we see:
[09:36] <asac> Get:78 http://ppa.launchpad.net/phablet-team/ppa/ubuntu/ saucy/main gir1.2-messagingmenu-1.0 i386 1:13.10.0phablet1ubuntu1 [3,112 B]
[09:36] <asac> gusch: yeah thats the old version
[09:36] <asac> gusch: e.g. its not the same that we have in the run
[09:36] <ogra_> cjwatson, https://launchpad.net/~phablet-team/+archive/ppa
[09:37] <seb128> asac, who the heck added an epoch to a ppa package?
[09:37] <gusch> asac: I'm updateing to latest from ubuntu-sdk-team-ppa
[09:37] <ogra_> cjwatson, https://launchpad.net/~phablet-team/+archive/ppa?field.series_filter=saucy only these three packages should come from saucy
[09:38] <asac> cjwatson: search for: saucy.*gir.*messaging here: http://10.97.2.10:8080/job/generic-mediumtests-runner-saucy/2472/consoleFull
[09:38] <asac> hmm. you need qalab intranet for that
[09:38] <cjwatson> ogra_,asac: please give me a few minutes to investigate.
[09:38] <asac> sure, if you have enough info L:)
[09:38] <asac> seb128: is that an epoch taht we dont have in archive?
[09:38] <asac> interesting
[09:38] <seb128> asac, it is
[09:38] <asac> seb128: blame the ogra and rsalveti cowboy crew :)
[09:39]  * ogra_ raises his hands in innocence
[09:39] <asac> ogra_: there is no such thing as innocence :)
[09:39] <ogra_> haha
[09:39] <asac> you have been sitting there right when it happened
[09:39] <asac> you had the chance to prevent it :)
[09:40] <Mirv> sil2100: gusch: if the latest SDK change is reverted, will the gallery-app branch need additional changes so that it only fixes the earlier problem and not the new ones from yesterday evening?
[09:40] <ogra_> preventing something means to know about it
[09:40] <ogra_> :)
[09:40] <asac> ogra_: you know what the population said after WW2?
[09:40] <asac> :)
[09:40] <asac> they didnt know
[09:40] <ogra_> anyway, the epoch shouldnt matter as soon as thge PPA is dropped
[09:40] <asac> seb128: what impact can a bogus gir-messaging have?
[09:41] <seb128> asac, messaging menu integration for clients not working
[09:41] <asac> seb128: can it impact switching tabs in autopilot on qt difficult?
[09:41] <ogra_> asac, was that a subtile way to pull a goodwin ?
[09:41] <asac> ok ic
[09:41] <noviolence> Hiho, Sbdy here I can bug with annoying questions? Like: How ready is Ubuntu Touch for Real Life Usage?
[09:41] <seb128> asac, no
[09:41] <asac> ogra_: probably
[09:41] <asac> :L)
[09:41] <ogra_> heh
[09:41] <gusch> Mirv: don't think so
[09:42] <sil2100> Mirv, gusch: I re-ran CI for the merge proposal, maybe now we won't hae any builder problem
[09:42] <sil2100> s
[09:42] <asac> jibel: you know about upstream merger? where is the code?
[09:43] <jibel> asac, not really it's a fginther thing
[09:43] <asac> jibel: he is the one and only person that knows where the code for this is?
[09:43] <asac> :)
[09:44] <cjwatson> ogra_: Please can you try redeleting indicator-messages/saucy from https://launchpad.net/~phablet-team/+archive/ppa/+delete-packages ?  wgrant thinks that the source may have been deleted just before its binaries were published.
[09:45] <gusch> asac: argh - went into package issues - I need some time to resolve these
[09:45] <asac> ok sry
[09:45] <ogra_> cjwatson, hmm how would i do this ? i dont see it in saucy at all with the UI ?
[09:45] <asac> gusch: do you use a chroot for this stuff?
[09:45]  * ogra_ checks the URL again
[09:45] <cjwatson> ogra_: Hm.  Let me see if I can prepare a script for you.  (Or somebody could add me to ~phablet-team temporarily and I can work it out ...)
[09:46] <Mirv> gusch: ok, then
[09:47] <noviolence> Looking for a noob-proof way to get ubuntu touch on a nexus 4 using mac or windows.
[09:47] <ogra_> cjwatson, bfiller, sergiusens or pmcgowan are admins ... i fear none of them is around yet
[09:47] <ogra_> i guess we need an admin in a .eu TZ :)
[09:47]  * ogra_ will ask for that 
[09:48] <asac> ogra_: i am getting access
[09:48] <asac> asking dbarth right now
[09:48] <asac> ogra_: anyone in https://launchpad.net/~pspmteam
[09:48] <asac> can do it
[09:48] <asac> lool: plz add cjwatson to phablet-team
[09:49] <asac> thostr_: ^^
[09:49] <asac> Mirv: ^^
[09:49] <cjwatson> meh, don't worry
[09:49] <cjwatson> $ lp-shell production devel
[09:49] <asac> kk
[09:49] <cjwatson> >>> lp.load("https://api.launchpad.net/devel/~phablet-team/+archive/ppa/+sourcepub/3439950").requestDeletion("re-deleting due to binary publication race")
[09:49] <cjwatson> ogra_: ^- run that please
[09:50] <ogra_> TypeError: Method must be called with keyword args.
[09:50] <cjwatson> sorry
[09:50] <cjwatson> >>> lp.load("https://api.launchpad.net/devel/~phablet-team/+archive/ppa/+sourcepub/3439950").requestDeletion(removal_comment="re-deleting due to binary publication race")
[09:50] <cjwatson> copied and pasted the wrong line :)
[09:50] <ogra_> done
[09:51] <Mirv> asac: no rights
[09:51] <thostr_> asac: I cannot add anybody to pspmteam
[09:52] <asac> Mirv: thostr_: no ... i wanted you to add him to the phablet-team :)
[09:52] <asac> not the pspm team
[09:52] <Mirv> asac: pmcgowan, sergiusens and bfiller would have https://launchpad.net/~phablet-team/+members
[09:52] <asac> (sorry if i wasnt clear)
[09:52] <cjwatson> Mirv,thostr_: don't worry about it
[09:52] <thostr_> cjwatson: ok :)
[09:52] <cjwatson> ogra_: that should take effect once the PPA publisher gets to it, then
[09:52] <ogra_> asac, psp team has no admin over phablet-team ... at least according to LP
[09:52] <asac> Mirv: hmm
[09:52] <asac> oh
[09:52] <asac> erro
[09:52] <asac> ogra_: its the OWNER :)
[09:52] <asac> https://launchpad.net/~phablet-team
[09:52] <ogra_> ah
[09:52] <ogra_> k
[09:53] <asac> so it can make itself admin in worst case
[09:55] <oSoMoN> bzoltan, sil2100, timp: the API change is also breaking the calendar app’s autopilot tests, for information
[09:55] <gusch> asac: no - working on my main system (I know it's risky)
[09:56] <asac> gusch: make a chroot :) ... and do what you see in the log
[09:56] <asac> gusch: i am 99% sure you will be able to reproduce then
[09:56] <asac> and fix the problem for us :)
[09:58] <gusch> asac: I am already able to reproduce it (SDK version 0.1.46+13.10.20130830.1bzr736saucy0)
[09:58] <cjwatson> ogra_,asac: those publications have gone now
[09:58] <gusch> bzoltan: ^
[09:58] <asac> gusch: nice... so lets tell sdk folks to backout their stuff :)
[09:58] <asac> cjwatson: rock on
[09:58] <asac> gusch: first thing is to disable those ppas in testing
[09:58] <asac> gusch: instead the sdk team upstream merger should test all app tests
[09:59] <asac> so they dont land regressions like this
[09:59] <asac> i will sort that... just hold the line :)
[09:59] <gusch> asac: cool thx
[10:00] <asac> gusch: would be cool if you could help identify the commit that broke the sdk for you
[10:00] <asac> so bzoltan can back that out
[10:00] <gusch> asac bzoltan I can see the reason - now the toolbar is open on startup
[10:00] <asac> think is easier to unblock you that way before we remove the ppas later today
[10:01] <asac> bzoltan: do you know the change that did that? can we back it out so we unblock our upstream merger for other projects?
[10:01] <gusch> asac bzoltan so the first click on the tabbar is used to dismiss the toolbar - not to bring up the tabs
[10:01] <asac> gusch: we could fix it on our side if thats the expected behaviour?
[10:02] <gusch> asac: you mean in the tests? yes - for sure
[10:03] <ogra_> Setting up python3.3-minimal (3.3.2-3ubuntu1) ...
[10:03] <ogra_> Fatal Python error: Failed to read bytes from /dev/urandom
[10:03] <ogra_> qemu: uncaught target signal 6 (Aborted) - core dumped
[10:03] <ogra_> Aborted (core dumped)
[10:03] <ogra_> dpkg: error processing python3.3-minimal (--configure):
[10:03] <ogra_> ARGH !
[10:03] <ogra_> what a day :/
[10:03] <asac> ogra_: kernel?
[10:03] <ogra_> asac, qemu chroot
[10:03] <ogra_> trying a testbuild for a package change
[10:03] <asac> ogra_: syscall?
[10:03] <asac> ic
[10:03] <asac> yeah thats bad
[10:03] <ogra_> or rather not trying but watching debootstrap die
[10:04] <asac> ogra_: change in python triggered that?
[10:04]  * ogra_ goes to build on real HW 
[10:04] <ogra_> asac, no idea what triggered it, can be qemu, can be python ...
[10:04]  * ogra_ will file a bug, i dont have the time to investigate now 
[10:05] <asac> ogra_: you could check what was uploaded last :)
[10:05] <ogra_> asac, sure, and fall even more behind with my TODO :P
[10:05]  * ogra_ will file a bug and either me or someone else will investigate later
[10:06] <cjwatson> qemu user mode is generally awful
[10:07] <ogra_> cjwatson, well, basic deboostrap used to work
[10:07] <cjwatson> yeah, but good luck getting somebody to figure it out ...
[10:07] <ogra_> though this is a stacked setup ... two chroots
[10:07] <ogra_> the outer debootstrap worked fine
[10:07] <noviolence> Looking for a easy to follow Tutorial to get Ubuntu Touch onto a Nexus 4 using Mac or PC. Any Help appreciated.
[10:07] <mpt> zsombi, I have replied to your questions about the time and date pickers.
[10:08] <asac> Mirv: sil2100: http://paste.ubuntu.com/6058389/
[10:08] <ogra_> i wonder if python3 requires /proc, /dev or /sys at install time ... the inner chroot definitely doesnt have that
[10:08] <asac> Mirv: sil2100: do you use that hooks: field for something in daily-release as well?
[10:08] <ogra_> (thought it didnt before i think)
[10:09] <asac> Mirv: sil2100: could we just drop all ppas there?
[10:09] <asac> for sure: qt5 edgers
[10:09] <asac> and phablet-team
[10:09] <asac> and sdk
[10:09] <cjwatson> ogra_: I would expect that the hash randomisation security fixes require Python to get random bytes from somewhere
[10:09] <ogra_> aha
[10:10] <cjwatson> So it will need to do that any time a Python process starts
[10:10] <cjwatson> Which indeed happens in the postinst
[10:10] <ogra_> which means you need /dev/(u)random
[10:10] <cjwatson> Yes
[10:10]  * ogra_ adds a bind mount
[10:11] <bernd> where can i finde a list of tablets which can bee flashed
[10:12] <sil2100> asac: we use the hooks as well from some projects, some I was already wondering about dropping - the edgers we should drop indeed, but having them on doesn't break anything
[10:12] <sil2100> asac: as it has no newer packages for saucy
[10:12] <asac> sil2100: it does for us now
[10:12] <asac> sil2100: sdk and qt had busted stuff in ppa
[10:12] <asac> blocking app testing
[10:13] <asac> sil2100: i guess we only would want the ppas enabled fro the stacks that own those ppas
[10:13] <sil2100> https://launchpad.net/~canonical-qt5-edgers/+archive/qt5-proper?field.series_filter=saucy <- which one of these broke stuff?
[10:13] <sil2100> qtconnectivity?
[10:13] <asac> sil2100: maybe it wasnt qt this time, but ... once we start adding proper testing
[10:13] <asac> sil2100: its sdk ppa this time
[10:13] <asac> next time it will be qt :)
[10:14] <sil2100> Right, I guess let's have those removed - I'll poke Francis to coordinate, so that we have both daily-release cleaned and mediumtests
[10:14] <asac> nice
[10:15] <asac> bzoltan: can you confirm that the change of tab behaviour is abug?\
[10:16] <bzoltan> asac: It is not a bug. Apps were using an internal property what was by mistake replaced with proper methods. This change caused an unintended API break. This change is now reverted.
[10:16] <dholbach> cjwatson, I ran this as root
[10:17] <asac> bzoltan: so its reverted on trunki?
[10:17] <asac> nice
[10:17] <asac> bzoltan: is an updated package in-flight into the sdk ppa?
[10:17] <asac> guess that will unblock gusch once its there
[10:17] <bzoltan> asac: Yes, the trunk now is clean
[10:18] <cjwatson> dholbach: click drops privileges before the code I was referring to in my commit message.
[10:19] <asac> bzoltan: can you kick off a build so the ppa gets the cool stuff?
[10:19] <asac> not sure how that works with launchpad daily
[10:19] <cjwatson> dholbach: Maybe I should have exposed fewer internal implementation details in my commit message.
[10:19] <davmor2> Morning all
[10:20] <ogra_> ugh
[10:21] <ogra_> so that /dev bind mount inside two qemu chroots now leaves me with:
[10:21] <ogra_> ogra@anubis:~/Devel/packages$ scp initramfs-tools-ubuntu-touch_0.41* 192.168.2.100:/media/ogra/08fc34a7-1f78-4a36-bcd7-d8a9adf26e7f/ubuntu-touch/saucy/packages/
[10:21] <ogra_> PRNG is not seeded
[10:21] <ogra_> lost connection
[10:21] <dholbach> cjwatson, ok, thanks a lot for looking into it!
[10:21] <gusch> bzoltan: this triggers the tab switching issue in the uitk autopilot as well http://bazaar.launchpad.net/~schwann/ubuntu-ui-toolkit/uitk-autopilot-tab-switch-issue/revision/736
[10:21] <ogra_> (after unmounting both bind mounts and removing the chroots
[10:21] <ogra_> )
[10:22] <oSoMoN> bzoltan: has the change been reverted yet?
[10:23] <bzoltan> oSoMoN: Yes, the trunk is good
[10:24] <bzoltan> oSoMoN:  the SDK PPA has the new release - 0.1.46+13.10.20130830.1bzr737
[10:25] <oSoMoN> bzoltan: I’m looking at the latest trunk, and the last revision (737, the one supposed to revert the change in 734) appears empty
[10:25] <ogra_> wow, i actually had to rm /dev/urandom and mknod it manually to get it working again
[10:25] <seb128> bzoltan, that revert seems buggy
[10:25] <seb128> what oSoMoN said
[10:25] <seb128> bzoltan, look at the diff on https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/removing-branch-734/+merge/183600 as well
[10:25] <bzoltan> Ohh fuck
[10:26]  * ogra_ glares at "[Alioth] Disk usage on alioth" in his inbox 
[10:26] <ogra_> i dont have an alioth account ...
[10:26] <seb128> ogra_, you must have one...
[10:27] <ogra_> well, probably someone abused my mail address
[10:27] <ogra_> i cant remember ever creating one
[10:27] <seb128> did you ever participate to some debian pkg team?
[10:27] <ogra_> hmm
[10:28]  * ogra_ wonders if someone added him to the ltsp team 
[10:28] <asac> ogra_: maybe it got forwarded to a debian-XXX list and tahts how you got it?
[10:29] <ogra_> nope, its directly to ogra@ubuntu.com
[10:29] <seb128> ogra_, maybe somebody from the ltsp team added you there so you had access to some packaging vcs
[10:29] <ogra_> i guess seb is right and someone added me to something like ltsp uploaders
[10:30] <ogra_> ltsp VCS is in LP ... but probably to allow me debian uploads
[10:30] <ogra_> (which i never used)
[10:32] <asac> Mirv: sil2100: you think its possible to run unity-phablet-qmluitests-saucy also as part of the sdk daily-release/merger?
[10:32] <mpt> dednick, I've followed up on your network menu MR
[10:33] <sil2100> asac: where are those located?
[10:34] <dednick> mpt: thanks for that
[10:34] <annerajb> morning
[10:34] <asac> sil2100: http://s-jenkins:8080/job/unity-phablet-qmluitests-saucy/ is a job
[10:35] <asac> that runs them
[10:35] <annerajb> ogra_, cjwatson ? does that line run upstart? http://pastebin.com/52AvbzHm
[10:35] <annerajb> (the exec)
[10:35] <cjwatson> annerajb: I cannot help further with your problem
[10:35] <cjwatson> SOrry
[10:36] <annerajb> cjwatson, allright
[10:36] <cjwatson> Hopefully somebody else can.  I was just jumping in yesterday to comment on one specific thing
[10:36] <ogra_> annerajb, and i'm still pretty sure upstart cant claim /dev/console and dies immediately on startup
[10:37] <annerajb> i got the printout of /dev let me grab it
[10:37] <ogra_> (i.e. you need to put console=tty1 last ... which then will make you fly blind again)
[10:37] <annerajb> http://pastebin.com/N4nNy1Ks
[10:38] <annerajb> let me grab the cmdline it hink it last right now
[10:39] <annerajb> ogra_, http://pastebin.com/5xGUrzf0
[10:39] <annerajb> the serial number get's added after tty1 does that matter? or does it only matter taht is after ttysac2?
[10:39] <ogra_> it doesnt matter
[10:39] <annerajb> also all those verbose didnt seem to do anything
[10:39] <annerajb> but...
[10:40] <annerajb> i swear i had some verbose flag somewhere where every command ran in the init printed it's return value.
[10:40] <annerajb> from reading around that's what supposedly INIT_VERBOSE=yes does.. but it never worked again.
[10:41] <ogra_> well, your problem is that init dies before it can print anything verbose :)
[10:41] <gusch> bzoltan asac you wanted an MR, you get an MR ;) https://code.launchpad.net/~schwann/ubuntu-ui-toolkit/uitk-autopilot-tab-switch-issue/+merge/183625
[10:41] <asac> bzoltan: ^^
[10:42] <asac> gusch: i hope thats still applicable
[10:42] <bzoltan>  gusch: nice, thanks
[10:42] <asac> gusch: i believe sdk is backing out something ... ah good.
[10:42] <asac> seems zoltan is happy ... so i am happy too :)
[10:43] <gusch> asac: I'm happy once gallery is able to land again ;)
[10:43] <asac> gusch: so you said you could reproduce the thing?
[10:43] <asac> exactly the same errors?
[10:43] <asac> with the ppa ... but without them it goes away?
[10:44] <ogra_> annerajb, hmm, wait., how was that log at http://pastebin.com/N4nNy1Ks created ? was that a normal boot or did you break= anywhere and did any manual stuff
[10:44]  * asac wonders if we should just merge manually saying that all issues seen are understood
[10:44] <gusch> asac: yes - when I installed the latest SDK
[10:44] <annerajb> no no breaks anything thats with the cmdline i printed up there
[10:44] <asac> :)
[10:44] <asac> gusch: and your patch above fixes that as well?
[10:44] <annerajb> ogra_, that's using the cmdline up there. also the upstart log show some upstart messages
[10:44] <annerajb> like trying to start ureadahead
[10:44] <gusch> asac: so far I only tested if for the uitk
[10:44] <annerajb> so i am confident upstart is starting to run.
[10:44] <gusch> asac: let me check
[10:45] <ogra_> annerajb, well, looks like upstart is starting ...
[10:45] <ogra_> yeah
[10:46] <asac> oSoMoN: do we have a single clue by now what is going on with webbrowser?
[10:46] <ogra_> annerajb, try disabling /etc/init/lxc-android-config.conf (with a manual override)
[10:46] <asac> psivaa: ?
[10:46] <ogra_> annerajb, and see if it gets any further
[10:46] <annerajb> ogra_, can you be more specific like i know where the file is but not sure what is a manual override commenting it?
[10:47] <ogra_> echo manual | sudo tee -a /etc/init/lxc-android-config.override
[10:48] <oSoMoN> asac: I can’t tell as the dashboard doesn’t even display test results (it says it didn’t run any for maguro), but it seems the preferred hypothesis yesterday was that unity8 is eating up a lot of CPU, thus making a number of tests fail in unexpected ways
[10:48] <Mirv> asac: all the PPAs could be dropped, except for packages that build for precise/quantal/raring too. so it's more like the PPAs should be cleaned (largely done already)
[10:48] <asac> oSoMoN: right
[10:48] <asac> mzanetti: hey... so the logs are there
[10:48] <asac> mzanetti: and unity is looping :)
[10:49] <Mirv> asac: there are not many saucy packages in any of the PPAs, and they were cleaned some time back but there are still a few lingering
[10:49] <asac> mzanetti: like here: http://reports.qa.ubuntu.com/smokeng/saucy/image/3930/mediaplayer-app-autopilot/
[10:49] <asac> at the end
[10:49] <asac> Mirv: i dont think we want them in our daily-release used though
[10:49] <asac> Mirv: right now we have non, but three demo packages in ppa that ar ein the image
[10:50] <asac> Mirv: so we can start testing exactly against the archive for integration
[10:50] <annerajb> ogra_, i dont have a lxc-android-config.override should i copy the non override one?
[10:50] <Mirv> asac: generally not, it's just that for those packages that are backported we're using same configs. for the rest they should be all removed at least.
[10:50] <Mirv> so that there is not a chance that something from PPA would be picked up on saucy
[10:51] <lool> cjwatson, asac: Added cjwatson to ~phablet-team
[10:51] <asac> kk
[10:51] <asac> guess he didnt want to anymore :)
[10:51] <asac> hehe
[10:51] <asac> thanks anyway
[10:51] <cjwatson> I didn't, but oh well
[10:51] <lool> cjwatson: enjoyable benefits: 10MB of merge proposals a week  :-)
[10:51] <lool> cjwatson: can remove you if you like
[10:51] <cjwatson> ... yeah, how about I leave the team :)
[10:51] <cjwatson> sorry for the extra work
[10:51] <asac> Mirv: i dont understand the last argument
[10:51] <cjwatson> left it now
[10:51] <asac> kMirv:
[10:51] <cjwatson> I did try to tell you not to :)
[10:52] <psivaa> asac: no sure if you've noticed but the test job that i created with maguro-02 also took 34 mins where as our normal timeout is 30 mins.
[10:52] <asac> Mirv: i think you say: since we have those ppas enabled during image build, we should have them enabled during testing?
[10:52] <lool> I didn't read all the backlog, just hte few lines after asac asked
[10:52]  * cjwatson nods
[10:52] <asac> Mirv: i believe the only ppa that is enabled in image build is the phablet-team ppa
[10:52] <sil2100> bzoltan: how's the SDK revert going? :)
[10:52] <asac> psivaa: can we bump our timeout?
[10:52] <asac> psivaa: for just that job?
[10:52] <asac> psivaa: otherwise we dont see any result
[10:52] <bzoltan> sil2100, seb128: https://code.launchpad.net/~bzoltan/ubuntu-ui-toolkit/reverse_cherrypicking_rev734/+merge/183628
[10:53] <cjwatson> lool: the multi-database work in click is going quite well btw - I should be able to get that uploaded today if I don't get too horribly distracted
[10:53] <asac> psivaa: and i want to see systemsettle after etc.
[10:53] <psivaa> asac: ok, sure will do now
[10:53] <asac> psivaa: guess only for that webebrowser job if thats possible
[10:53] <lool> cjwatson: that's nice
[10:53] <asac> psivaa: give it 45  i guess
[10:53] <lool> cjwatson: that way we can transition to it before 13.10 for the preinstalled packages
[10:53] <cjwatson> lool: I don't yet have the "register for all users" operation, and I don't yet have a notion of "whiteout" so that a user can remove something registered for all users
[10:54] <cjwatson> But it shouldn't take *too* long
[10:54] <lool> cjwatson: if you like, ping sergiusens, I'm sure he'd be interested in moving the preinstalled clicks over
[10:54] <bzoltan> sil2100, seb128: that should be the opposite of this > https://code.launchpad.net/~tpeeters/ubuntu-ui-toolkit/toolbar-reveal/+merge/183448
[10:54] <psivaa> asac: gave it 45 mins now
[10:54] <cjwatson> I will but not yet
[10:54] <asac> psivaa: can you rereun?
[10:54] <lool> cjwatson: sorry, I meant to write "ping sergiusesns when it's done"
[10:54] <asac> psivaa: or rather re-enqueue :)
[10:54] <psivaa> asac: it's already done :)
[10:54] <psivaa> i mean on the queue
[10:55] <asac> oSoMoN: http://reports.qa.ubuntu.com/smokeng/saucy/image/3930/mediaplayer-app-autopilot/ that one a known failure? otherwise we should retry given that after the test unity8 is eating CPU
[10:55] <lool> ogra_: signal 6 is SIGABRT I think, so probably not what you're looking for to fix; apparently the error is that python can't read /dev/urandom, perhaps this is what you should look after?  or find the source of the ABRT; it sounds like qemu is just passing the signal of some binary deciding it is in error condition
[10:55] <asac> psivaa: for the ones where we see systemsettle-after failing, we should check with the owner and retry if they cant explain the failyure
[10:56] <asac> psivaa: could well be that the unity8 process started to eat CPU early and then the tests failed
[10:56] <oSoMoN> asac: not a known failure, afaik
[10:56] <asac> psivaa: but always check with the owner and give up after trying 2 times :)
[10:56] <dholbach> with the new click, I can install ureadit now, but can't launch it - is that the case for anyone else?
[10:56] <asac> oSoMoN: ok. we will retry then
[10:56] <lool> dholbach: does it show up in click list?
[10:56] <asac> given that settle is bad afdter
[10:56] <lool> dholbach: what's the error launching it?
[10:56] <gusch> asac: my patch fixes that issue, but the toolbar stealing the "click" still causes other issues :(
[10:56] <ogra_> lool, well, i tired a bind mount of /dev ... but the armhf chroot doesnt seem to get along with that either ... and after i leave the armhf chroot /dev/urandom on the x86 host is busted (i need to rm and mknod to get it working again)
[10:56] <asac> gusch: i think the toolbar stealing is reverted now, no?
[10:57] <asac> bzoltan: ?
[10:57] <psivaa> asac: ack, i'll re-run that as well. unity8 is at the top of the top-after
[10:57] <lool> ogra_: that seems odd
[10:57] <ogra_> lool, yep
[10:57] <dholbach> lool, how would I launch it using the shell? (mimicking normal behaviour?
[10:57] <asac> psivaa: can you log the URLs of those that you retry?
[10:57] <bzoltan> asac: there is an MR what fixes that bloody tab issue... All I need an SDK dev to approve it
[10:57] <asac> psivaa: i would like to accumulate a few and then investigate
[10:57] <psivaa> asac: ack, will do
[10:58] <gusch> bzoltan: you mean my MR?
[10:58] <ogra_> lool, i end up with "PRNG is not seeded" on the x86 host
[10:58] <bzoltan> gusch: Your MR will land after this reverting hustle is done
[10:58] <cjwatson> dholbach: do check "click list" first
[10:58] <ogra_> anyway, no biggie, i built the package on real HW and the buildd will also be real HW
[10:58] <davmor2> ogra_: but that's an awesome ring tone whats wrong with you
[10:58] <asac> bzoltan: and SDK dev is blocked on lunchtime?
[10:59] <asac> bzoltan: you can also call their mobile if you feel they should be around :)
[10:59] <ogra_> davmor2, ask qemu
[10:59] <sil2100> bzoltan: it seems ok the revert, approved
[10:59] <dholbach> cjwatson, lool: it's listed there
[10:59] <gusch> bzoltan: ok - as the open toolbar causes some more issues ...
[10:59] <bzoltan> gusch: yes
[10:59] <asac> gusch: i believe the revert will revert that part
[10:59] <Mirv> asac: no. since we have daily release config handling also builds for precise, quantal and raring, we should keep them enabled at those parts of config for now. otherwise dropping the PPAs for packages that aren't build for P/Q/R.
[11:00] <Mirv> asac: but if it's possible that the autolanding does not require those ppa rules and they're actually only used on saucy, then all of them should be dropped
[11:00] <Mirv> I'm not that intimate with the configuration that I'd know, but fginther will know
[11:01] <Mirv> so it's best that he does the modification of the config
[11:04] <asac> Mirv: i think an autolanding component should have the ppa enabled that is used for daily-release...
[11:04] <asac> we also might have some flags that allow us to enable other ppas for just one job, in case there are weird interdependencies
[11:04] <asac> lets wait till francis comes on and recap the reasons why we have those
[11:05] <asac> Mirv: m
[11:05] <asac> Mirv: err... on p/q/r ... thats for host/sdk tools, right?
[11:05] <asac> our image integration should only run on saucy
[11:07] <asac> oSoMoN: latest mako build has a test failyure on webbrwoser: http://reports.qa.ubuntu.com/smokeng/saucy/image/3931/webbrowser-app-autopilot/
[11:07] <asac> without any business after
[11:08] <asac> err without being busy after :)
[11:08] <Mirv> asac: yeah, the problem has is that sdk tools starts using more and more of the new packages so we're gradually backporting more packages. but I'd guess that actually the ppa rules in the config files don't affect the those landings as those P/Q/R builds build within itself in the SDK Release PPA
[11:08] <asac> Mirv: right. i hope we can revisit our distribution strategy for the sdk somewhat
[11:09] <asac> the idea is to think of the sdk as a "single", managable release artifact somehow
[11:10] <asac> for now, noything to change on that front. just that we dont need the ppas for any imgae job
[11:10] <asac> image
[11:10] <annerajb> ogra_, http://pastebin.com/ZkErfwaz that's the last thing it prints.
[11:10] <asac> login.ubuntu.com is unhappy :()
[11:10] <asac> OOPS-...
[11:11] <ogra_> annerajb, and adb doesnt come up after this ?
[11:11] <oSoMoN> asac: yeah, about this one failure, no idea, I know it’s not a satisfactory explanation, but I’d say it’s an intermittent, unreproducible failure, i.e. if we run the tests again it will pass :/
[11:12] <asac> oSoMoN: whats your bet :)?
[11:12] <oSoMoN> asac: all in :)
[11:12] <ogra_> stop gambling !
[11:12] <ogra_> :)
[11:12] <asac> lol
[11:12] <oSoMoN> asac: if we had videos of the test runs it would help understanding what’s going on, it might be that the app failed to launch for whatever reason…
[11:12] <asac> wonder what all in means in this context
[11:12]  * asac thinks about a nice bug that oSoMoN fixes during all night :)
[11:13] <oSoMoN> asac: all the keys on my keyboard :)
[11:13] <asac> oSoMoN: right. emulator is needed for that i figure
[11:13] <asac> i doubt we can reasonabily screenrecord
[11:13] <asac> we could take screenshots
[11:13] <oSoMoN> asac: emulator, or lots of webcams…
[11:13] <asac> not sure if we could figure a good timing for those
[11:13] <ogra_> student interns !
[11:13] <asac> maybe just a live video feed
[11:14] <asac> with clearly numbered devices
[11:14] <asac> utah-live :)
[11:14] <mpt> OMG the phrase "Home Scope" actually appears in the UI
[11:14] <annerajb> ogra_, yeah it says device not found :(
[11:14] <annerajb> should var/log/upstart have a log for each upstart file it runs?
[11:15] <ogra_> for each one that has any output
[11:15] <asac> mpt: you say we dont want to define a new brand?
[11:15] <asac> :)
[11:15] <asac> "Scope"
[11:15] <annerajb> there are only http://paste.ubuntu.com/6058536/
[11:15] <annerajb> hmm let me try adding output to all of them
[11:16] <mpt> asac, only if it's something cool, like, oh, "Ubuntu"
[11:16] <asac> Scopes would be similar to Windows :)
[11:16] <ogra_> annerajb, is ypour fstab ok ?
[11:16] <oSoMoN> ogra_: you mentioned an e-mail from alioth, right? just got one too, and I have no clue what my relationship to alioth is
[11:16] <asac> might be a fun name if we want to start this project from scratch
[11:16] <annerajb> ogra_, the nosuid change??
[11:16] <ogra_> oSoMoN, heh, probably some kde upload team membership
[11:16] <ogra_> annerajb, your fstab in ubuntu
[11:17] <annerajb> let me check
[11:17] <asac> psivaa: so lets see if oSoMoN is right and retry the webbrowser job on mako too
[11:17] <asac> he has no story, but believe :)
[11:17] <oSoMoN> ogra_: last I checked I wasn’t a kde uploader, but well, I guess I’ll just safely ignore it
[11:17] <ogra_> same here
[11:18] <psivaa> asac: ack, will do.
[11:18] <asac> psivaa: and record it in our new list with a comment: test flakiness aka no busy system
[11:18] <asac> or something
[11:18] <psivaa> asac: i haven't seen the backscroll on this http://reports.qa.ubuntu.com/smokeng/saucy/image/3931/camera-app-autopilot/
[11:18] <psivaa> asac: camera app test on mako. no systemsettle issues there though
[11:18] <asac> that one fails miserably ... as if camera is broken
[11:18] <asac> popey: is camera working for you on mako at all?
[11:19] <asac> psivaa: the failure pattern looks similar to what we got when we broke the whole camera middleware during pulse landing
[11:19] <asac> so scary
[11:19] <mpt> asac, "Windows" is a nonsense brand on a phone too, because there aren't any. :-)
[11:20] <asac> mpt: right. hence we call our OS Scopes :)
[11:20] <mpt>  /rename #ubuntu-touch #scopes-touch
[11:20] <asac> hehe
[11:21] <asac> oSoMoN: can you check the camera test failures?
[11:21] <asac> oSoMoN: http://reports.qa.ubuntu.com/smokeng/saucy/image/3931/camera-app-autopilot/
[11:22] <asac> Mirv: sil2100: did we start running the autopilots on the sdk yet?
[11:23] <oSoMoN> nerochiaro, gusch|lunch: can one of you guys have a look at the camera app failures? http://reports.qa.ubuntu.com/smokeng/saucy/image/3931/camera-app-autopilot/
[11:28] <annerajb> ogra_, http://paste.ubuntu.com/6058559/ does taht look ok?
[11:28] <bzoltan>  oSoMoN, sil2100: now it is landed
[11:29] <ogra_> annerajb, /data is missing
[11:29] <ogra_> enable it again (with the proper device indeed)
[11:30] <annerajb> and change ext4 to yaffs?
[11:30] <ogra_> beyond that it looks fine (if /system is actually on that device)
[11:30] <ogra_> oh, yeah, probably
[11:30] <oSoMoN> bzoltan: excellent, thanks!
[11:35] <nic-doffay> Anyone had issues booting into recovery before? Whenever I try I just get sent back to the google logo...
[11:35] <nerochiaro> oSoMoN: the page you liked is all green
[11:36] <psivaa> nerochiaro: that's because the re-run came all passed
[11:36] <oSoMoN> nerochiaro: huh, sorry for the noise then, I was only forwarding a request from asac, I should have checked before
[11:36] <oSoMoN> aha
[11:36] <psivaa> nerochiaro: let me paste you the failed job
[11:37] <Mirv> asac: selected apps autopilot tests are still being run, but no new ones added since the last week's problems
[11:37] <psivaa> nerochiaro: https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-camera-app-autopilot/97/
[11:38] <psivaa> asac: oSoMoN: we've now got results for maguro webbrowser tests with 45 mins timeout: http://reports.qa.ubuntu.com/smokeng/saucy/image/3930/webbrowser-app-autopilot/
[11:39] <asac> sysmte doesnt look busy
[11:39] <nerochiaro> psivaa: the console log is really hard to read ( https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-camera-app-autopilot/97/console ) , is there anywhere else i can see what tests failed in a more human readable way ?
[11:39] <asac> oSoMoN: gfiven that the sdk directly impacts everything, i am tempted to put all back
[11:39] <asac> sorry
[11:39] <asac> Mirv: ^^
[11:40] <oSoMoN> psivaa: they shouldn’t take that long to run anyway, even 30min is way too much, on my galaxy nexus they run in ~500secs, i.e. 8.5min
[11:40] <asac> oSoMoN: so systemsettle is running 6 minute before
[11:40] <asac> and 5 minute after
[11:40] <asac> that means we already are at 20'ish
[11:41] <asac> oSoMoN:  i think it is roughly ok... we do mroe stuff before we run the test
[11:41] <asac> psivaa: whats the shortest run we get?
[11:42] <asac> ok address book is this: https://jenkins.qa.ubuntu.com/job/saucy-touch-maguro-smoke-address-book-app-autopilot/12/?
[11:42] <asac> thats 3min :)
[11:42] <lool> dholbach: with the aa-launch rune from the .local/applications .desktop file
[11:42] <psivaa> asac: let me check
[11:42] <asac> 4 min
[11:42] <asac> so lets say we have 4 min obverhead then we have 25 :)
[11:42] <psivaa> nerochiaro: would https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-camera-app-autopilot/97/artifact/clientlogs/utah.yaml/*view* help?
[11:44] <oSoMoN> asac: right, so the initial timeout of 30min should be fine. I guess 45min doesn’t hurt, but it’s not solving the problem anyway
[11:44] <asac> oSoMoN: well, i am saying that the time is ok aish
[11:45] <asac> i can see how it takes 32minutes
[11:45] <nerochiaro> psivaa: it's equally hard to read to be honest
[11:45] <popey> asac: yes
[11:45] <oSoMoN> asac: fair enough
[11:45] <asac> oSoMoN: seems the job took 29 minutes this time https://jenkins.qa.ubuntu.com/job/saucy-touch-maguro-smoke-webbrowser-app-autopilot/113/?
[11:45] <asac> so not sure
[11:45] <asac> that sounds pretty lengthy :/
[11:46] <oSoMoN> asac: I could randomly remove half of the autopilot tests, it might make the situation better :)
[11:46] <asac> not the right answer
[11:46] <sil2100> seb128: re-running SDK now, fingers crossed
[11:46] <popey> asac: http://popey.com/~alan/device-2013-09-03-124616.png
[11:46] <asac> :P
[11:46] <popey> Ubuntu Saucy Salamander (development branch) - armhf (20130831)
[11:46] <asac> popey: not --pending :)
[11:47] <asac> popey: talking about 03
[11:47] <nerochiaro> oSoMoN: psivaa: but trying to read into it anyway, it seems the failed tests are pretty random. i can't see any common pattern in the failures
[11:47] <popey> no, read only image, will re-flash
[11:47] <psivaa> asac: oSoMoN: the tests on 20130828.1 on the same device took only 16 min (all inclusive)
[11:47] <psivaa> http://reports.qa.ubuntu.com/smokeng/saucy/image/3807/webbrowser-app-autopilot/
[11:47] <oSoMoN> asac: looks like we’re back to the initial issue, for some reason the toolbar fails to open (in all the failures), and I’m using the standard emulator provided by the SDK to open it, so there’s something odd going on
[11:48] <nerochiaro> psivaa: oSoMoN: however, please note that there are a number of bugs still open on tablet devices for camera. they were not addressed yet as tablet haven't been a priority for a while.
[11:49] <psivaa> nerochiaro: re-running  the tests made it all pass. but if you want i could re-run that again to make it fail, since it's random :)
[11:49] <oSoMoN> nerochiaro: the smoke tests we are interested in are running on mako (n4) and maguro (galaxy nexus) only
[11:50] <nerochiaro> psivaa: oSoMoN: ok, then i'm not really sure what's going on. are they failing on maguro too ? that's the only place where i have been testing camera, and gusch|lunch too i think. we don't have makos
[11:50] <asac> oSoMoN: so we have the same issue with the SDK fix?
[11:50] <asac>  backout?
[11:50] <asac> oSoMoN: did the busted sdk land in the image?
[11:51] <oSoMoN> asac: no, I don’t think it even landed
[11:51] <oSoMoN> asac: so it’s not the same issue
[11:51] <awafaa> am i correct in thinking the only way of adjusting time/date settings is via the command line and not the settings menu?
[11:52] <gusch|lunch> oSoMoN nerochiaro psivaa no idea what went wrong there, and only have a maguro
[11:53] <asac> oSoMoN: you talk about the webbrowser?
[11:53] <ogra_> awafaa, currently, yes ... i usually use the terminal and run "sudo dpkg-reconfigure tzdata"
[11:53] <asac> besides that we have gallery-app fix in the pipeline (waiting for sdk backout)
[11:53] <asac> bzoltan: did the sdk backout land as a package now?
[11:53] <asac> in the ppa i guess
[11:53] <awafaa> ogra_: thanks, just wanted to confirm
[11:54] <asac> gusch|lunch: i think the new ui-toolkit is in the ppa
[11:54] <asac> gusch|lunch: so your MP might work better now
[11:54] <gusch|lunch> asac: I triggered a rebuild a few minutes ago
[11:54] <asac> cool
[11:55] <asac> gusch|lunch: you have nothing to do with friends/notes app?
[11:55] <oSoMoN> asac: yeah, I’m saying that the failures seen in the browser tests are not related to the busted SDK, since it was not in the image
[11:55] <gusch|lunch> asac: that's nerochiaro
[11:55] <psivaa> asac: oSoMoN: so the slowness in webbrowser in maguro started with 20130829. not sure if that's when the 'swipe intro' was integrated
[11:55] <asac> nerochiaro: you have friends/notes app failures on your radar?
[11:55] <asac> check http://reports.qa.ubuntu.com/smokeng/saucy/image/3931/
[11:56] <asac> also other apps that you might own
[11:56] <asac> psivaa: interesting
[11:56] <asac> mzanetti: any idea when swipe intro landed?
[11:56]  * asac wonders what effect that has on testing anyway
[11:56] <psivaa> asac: and i dont see the unlock screen getting rid of the intro
[11:56] <ogra_> yay, finally my package merge worked
[11:56] <mzanetti> asac: thursday or friday
[11:57] <asac> mzanetti: not earlier? :)
[11:57] <mzanetti> asac: let me check the logs
[11:57] <psivaa> thrursay was 29th
[11:57] <asac> right
[11:57] <asac> so that might match
[11:57] <asac> mzanetti: any idea what the swipe intro might do to the device if not closed?
[11:57] <asac> will it consume CPU? or even break tests?
[11:58] <nerochiaro> oSoMoN: can it be that what asac just pointed out is a failure due to that SDK API change that we were discussing before, and that will be reverted ? i haven't changed anything in notes app in a while
[11:58] <mzanetti> asac: Thu 2013-08-29 04:04:04 +0000
[11:58] <pmcgowan> nerochiaro, oSoMoN what sort of failure is happening?
[11:58] <mzanetti> asac: it will consume CPU while the demo animation runs, yes
[11:59] <mzanetti> asac: the question is if the unlock_screen.sh (iirc) is still able to unlock, yes
[11:59] <mzanetti> asac: are tests failing?
[12:00] <popey> asac: camera is find on mako on 20130903
[12:00] <popey> *fine
[12:02] <asac> nerochiaro: unlikely. the sdk issue only happened in the MP and dail-release gates
[12:02] <asac> nerochiaro: the tests i pointed at fail on the image and the "bogus" sdk hasn't made it there afaik
[12:02] <nerochiaro> pmcgowan: the part of the test that fails is clicking on a note and waiting for it to expand. but this same thing is done in many other tests and never fails
[12:02] <asac> but could be you talk about other API issues then what we backed out today
[12:02] <oSoMoN> nerochiaro: no, the API change in the SDK is not in today’s image afaict
[12:02] <nerochiaro> oSoMoN: ^
[12:03] <asac> right
[12:03] <nerochiaro> oSoMoN: asac: ok then it's a very weird failure. it's doing something that's done in *all* the tests in that suite and none of the others fail
[12:03] <asac> nerochiaro: you have a device and can try reproducing?
[12:03] <asac> nerochiaro: we can give the jobs back once
[12:04] <asac> to see if its a random thing
[12:04] <nerochiaro> asac: sure. do i need the latest image ?
[12:04] <asac> nerochiaro: otherwise you can see what changes landed in which image: http://people.canonical.com/~ogra/touch-image-stats/
[12:04] <asac> nerochiaro: sure you need the image that was used to run the tests :)
[12:04] <greyback> tmoenicke: ping
[12:05] <oSoMoN> asac: can we check the device used to run the tests in the lab? I’ve been consistently running the browser tests on my maguro without any failure, a bunch of times since yesterday, everytime with the latest image, so it might be worth looking into the test device itself
[12:05] <asac> oSoMoN: we moved the webbrowser to a new device yesterday
[12:05] <asac> had the same results
[12:05] <asac> psivaa: can you confirm?
[12:05] <annerajb> ogra_, that didnt work changed to yaffs2 and fixed the path of the device
[12:05] <asac> oSoMoN:  we can ask rfolwer to do something though
[12:05] <asac> he will come on soon i guess
[12:05] <psivaa> asac: yea i can. also i ran a couple of times locally today and they fail as well
[12:06] <asac> oSoMoN: ^^
[12:06] <asac> check with psivaa on how to reproduce
[12:06] <nerochiaro> asac: is it the one i would get now by running phablet-flash ?
[12:06] <asac> nerochiaro: phablet-flash cdimage-touch --pending
[12:06] <asac> to get the latest not-blessed image
[12:08] <ogra_> annerajb, you mean it regressed or it just didnt improve ?
[12:09] <pmcgowan> nerochiaro, oSoMoN do the failed tests have anything to do with exposing the toolbar?
[12:09] <annerajb> the serial output is the same i dont get any adb
[12:09] <nerochiaro> pmcgowan: no
[12:09] <pmcgowan> ok
[12:14] <noviolence> One quick question everybody. I want to have the newest Version on my Nexus 4. I followed a tut and didn't make sure it had the newest files linked. So now I've got the mwc-demo on it which is from february
[12:14] <noviolence> there are so many options here http://cdimage.ubuntu.com/ubuntu-touch-preview/
[12:14] <noviolence> I want a new version with system updates
[12:17] <oSoMoN> pmcgowan: they do, at least in the case of the browser
[12:17] <oSoMoN> pmcgowan: all the failed tests are actually failing on opening the toolbar
[12:18] <pmcgowan> oSoMoN, it may not be related but the HUD reveal changed recently,it is very easy to have it appear
[12:18] <lool> cjwatson: ok, so with a fixed schroot I get as far as cross build-essential being correctly installed before build, but then things fail when it tries to apt-get build-dep because it's not a source package in the archive, but a local one (SDK generated one)
[12:18] <pmcgowan> oSoMoN, but I think automation would not hit that
[12:18] <cjwatson> lool: sbuild >= 0.64.0-1 doesn't require the source package to be in the archive.  I fixed that.
[12:18] <lool> cjwatson: I guess we could generate an apt source repo (just like there's a binary one for the bdeps) in the chroot and use that
[12:18] <pmcgowan> noviolence, make sure you have the latest phablet-tools, then check the help, you want ubuntu-system
[12:19] <cjwatson> lool: We do as of 0.64.0-1.
[12:19] <cjwatson> lool: See http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=695203 .
[12:20] <lool> cjwatson: haha awesome, so just need to merge this in Ubuntu then
[12:20] <noviolence> pmcgowan, are these files here http://cdimage.ubuntu.com/ubuntu-touch-preview/daily-preinstalled/current/ the latest?
[12:20] <pmcgowan> noviolence, no, those are quite old
[12:20] <pmcgowan> drop the -preview from the url
[12:20] <cjwatson> lool: Oh, yes, I missed that that had never happened
[12:20] <noviolence> ah, i see
[12:20] <noviolence> so here i go http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/current/
[12:21] <cjwatson> lool: Bug jodh or mdeslaur as they're the last uploaders
[12:21] <bzoltan> asac:  yes
[12:21] <lool> cjwatson: yup exactly
[12:21] <cjwatson> I seem to remember bugging mdeslaur about this before, but maybe not ...
[12:22] <mardy> Kaleo: hi! I cannot find a component for making a combo-box in the SDK; is it planned to have one?
[12:23] <noviolence> pmcgowan, i can't find any phablet files in the ubuntu-touch, they are only on the touch-preview sites
[12:24] <lool> cjwatson: anything else you did in the last 6 months that I should avoid redoing before I notice it's already done?   ;-)
[12:24]  * lool apt-get changelog cjwatson
[12:25] <pmcgowan> noviolence, the naming is slightly different now
[12:25] <pmcgowan> its best to rely on phablet-flash
[12:25] <cjwatson> lool: :-) Uh ...
[12:25] <mardy> kalikiana: maybe you also know: is it planned to have a combobox component in the SDK?
[12:26] <lool> ok, /me goes back to this ureadit thing
[12:26] <cjwatson> lool: Sorry, I ought to have noticed that this one wasn't in Ubuntu, but since I'm an sbuild committer I've been using macros that cause me to use sbuild from a git checkout, so I'd forgotten to check what was in Ubutnu
[12:26] <cjwatson> *Ubuntu
[12:26] <psivaa> mzanetti: asac: oSoMoN: i ran the webbrowser test once after skipping the swipe intro locally and it completed in roughly 18 mins with all tests passing. running it again to confirm
[12:26] <lool> cjwatson: it's fine, it's actually good news that everything is done; I expected more real rough edges such as cross-compilers not being installable
[12:26] <lool> I guess the only issue will be with qmake, will see once I'm there
[12:27] <lool> perhaps SDK packages' multi-arch flags too
[12:27] <oSoMoN> psivaa: what is the "swipe intro"?
[12:28] <cjwatson> lool: As we discussed at vUDS, we will need to convert to cmake
[12:28] <asac> oSoMoN: havent seen it :) ... get a pending imgae
[12:28] <lool> cjwatson: yup
[12:28] <cjwatson> lool: qmake is highly problematic for cross-building, because you need to generate a cross-qmake for each architecture combination
[12:28] <asac> oSoMoN: its an animationm running on device to explain users how to use it :)
[12:29] <asac> (in particular how to swipe the launcher away)
[12:29] <cjwatson> I wish click were cross-compilable, but it's a bit too hard to get the Python build-dependencies working :-(
[12:29] <lool> cjwatson: I think we will need to work on this to cross-build qt, but I agree we should move SDK templates to cmake
[12:30] <lool> it's just that I wanted to start from a real-life QML extension (currently using qmake) to see how it would break and how much work it is to convert to cmake
[12:30] <cjwatson> lool: Qt builds its own qmake internally, so it's a quite different matter
[12:30] <noviolence> pmcgowan, thanks for your help so far, it's great not being ignored. So, Phablet files don't seem to exist anymore. Would the equivalent now be this: http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/20130903/saucy-preinstalled-touch-armhf.zip  ?
[12:30] <lool> cjwatson: I see
[12:30] <oSoMoN> asac: ah, right, I’ve seen it indeed, I just didn’t know it was called the swipe intro
[12:31] <pmcgowan> noviolence, right, then there is a device specific file as before
[12:31] <asac> oSoMoN: hehe. well i just picked the name up a few lines above :)
[12:31] <asac> just guessed it would be that
[12:31] <lool> cjwatson: good to know, I wont spend a single second trying to fix a qmake package to cross-build or on a cross-qmake; will just look at converting it once that's the next blocker
[12:31] <psivaa> oSoMoN: asac: http://www.youtube.com/watch?v=HreKruXnn_A from popey
[12:31] <lool> I guess SDK bdeps will come up first
[12:32] <psivaa> that kind of confirms that this was introduced on the 29th
[12:32] <noviolence> thanks pmcgowan,  and for the specific file i should take the saucy-preinstalled-system-armel+mako.img instead of  saucy-preinstalled-touch-armel+mako.zip  ? You said sth with system before. Or do I need both?
[12:33] <pmcgowan> noviolence, those files you are seeing are the images that are not doing system upgrades, those still have a rw file system
[12:33] <pmcgowan> the images that are not yet passing tests are in /pending
[12:34] <pmcgowan> and the ubuntu-system images are in another location
[12:34] <pmcgowan> but phablet-flash will show you where
[12:34] <oSoMoN> psivaa: so you’re saying that the swipe intro is interfering with the autopilot tests, thus the failures?
[12:34] <asac> balloons: hey ... whats the update on tests from the core apps front?
[12:34] <pmcgowan> noviolence,  you do a "-D" to get the files but not flash them
[12:34] <asac> balloons: any news/progress/need4help?
[12:35] <psivaa> oSoMoN: i think so. just running another test just to confirm
[12:35] <asac> balloons: ... given that we are now in Feature Freeze we might want to use this milestone to raise the bar here?
[12:44] <psivaa> asac: oSoMoN: the second webbrowser tests on maguro without swipe intro also passed in 17 mins.
[12:46] <asac> psivaa: nice!
[12:46] <asac> mzanetti: any idea how we can stop that intro thing during autopilot tests?
[12:46] <mpt> lisettte, https://wiki.ubuntu.com/SystemSettings
[12:47] <pmcgowan> ogra_, when does ubuntu-system become default?
[12:48] <ogra_> pmcgowan, dunno, thats in lool's hands i think
[12:48] <lisettte> mpt: merci
[12:48] <mzanetti> asac: dbus-send --system --print-reply --dest=org.freedesktop.Accounts /org/freedesktop/Accounts/User32011 org.freedesktop.DBus.Properties.Set string:com.canonical.unity.AccountsService string:demo-edges variant:boolean:false
[12:48] <mzanetti> asac: before starting unity. is that possible?
[12:48] <mpt> lisettte, clicking here should do the needful: #ubuntu-design
[12:48] <ogra_> pmcgowan, i heard something like "this week" though there is no default anymore, it will just be the default recommendation :)
[12:48] <asac> mzanetti: i think we cant do stuff before
[12:49] <asac> mzanetti: any way to close it properly? thorugh a similar dbus?
[12:49] <pmcgowan> ogra_, what do you mean no default? cuase there is no alternative I suppose
[12:49] <asac> pmcgowan: ogra refers to the tools
[12:49] <ogra_> mzanetti, boo ... touch /home/phablet/.demo-edges-done .... would be so much better :)
[12:49] <asac> pmcgowan: hwoever, we have a focus and that hasn't moved over
[12:49] <pmcgowan> ah right
[12:49] <asac> we wait for green light
[12:50] <mzanetti> asac: hmm... can you point me to the script that does the unlocking currently?
[12:50] <ogra_> pmcgowan, phablet-flash has no default anymore
[12:50] <pmcgowan> ogra_, of course
[12:50] <asac> pmcgowan: i was expecting to hear: we feel we are ready to move over... and then we would move our focus to system-only
[12:50] <asac> on dashboard etc.
[12:50] <pmcgowan> right, thought target was last friday
[12:51] <ogra_> asac, pmcgowan, as i understood it they want a test phase first ... and just recommend people to use system-image for a while but not switch the default hard
[12:51] <pmcgowan> makes sense
[12:51] <asac> pmcgowan: i believe there are still some details lool and friends are working on
[12:51] <ogra_> not sure what that means for QA
[12:51] <mamenyaka> diwic, hi, the ucm priority patch when will land?
[12:51] <ogra_> but i guess to watch both images in parallel for a while
[12:52] <diwic> mamenyaka, when ogra_ sponsors the upload *poke*
[12:52] <ogra_> mamenyaka, oops, thats my fault, sorry, i forgot to do the upload
[12:52] <asac> lool: whats the summary on things TODO before we can move focus over to RO/system-images?
[12:52] <diwic> mamenyaka, but I was just thinking of you for another thing
[12:52] <mamenyaka> diwic, did I do something bad?
[12:52] <diwic> mamenyaka, I've discovered that the nexus 10 has a different jack sensing mechanism. Perhaps one of your devices had that mechanism too?
[12:53] <mamenyaka> you mean the h2w/status thing?
[12:53] <diwic> mamenyaka, yeah
[12:53] <mamenyaka> diwic, yeah, the Sony Tablet Z
[12:53] <mamenyaka> diwic, so how does it detect?
[12:53] <diwic> mamenyaka, so the jack detection on Nexus 10 can be detected on /dev/input - wanna test if that applies to your device too?
[12:53] <mamenyaka> diwic, have it right here, a moment
[12:54] <mamenyaka> diwic, I have event0-event9 files in there, adn a by-path dir
[12:54] <diwic> mamenyaka, if so, install the evtest package, and run "sudo evtest /dev/input/event0". Now plug/unplug and see if anything happens. If not, continue with "sudo evtest /dev/input/event1", and event2 etc
[12:56] <daan> Hello
[12:56] <daan> Is this the place to ask questions about the installation of Ubuntu Touch ?
[12:57] <mamenyaka> diwic, hah, I have like volume up/down in there
[12:57] <mamenyaka> diwic, event6 is the jack detector
[12:57] <netcurli> daan: yes
[12:58] <mamenyaka> diwic, http://paste.ubuntu.com/6058791/
[12:58] <daan> Hi, I followed the guide on the Ubuntu Wiki step by step but the command "phablet-flash (cdimage-touch|cdimage-legacy|ubuntu-system|community) -b" won't work
[12:58] <OrokuSaki> Anyone know how to get QTTest to work with cmake for the Ubuntu-SDK? =)
[12:58] <popey> daan: phablet-flash cdimage-touch -b
[12:59] <daan> It says: "phablet-flash (cdimage-touch|cdimage-legacy|ubuntu-system|community) -b"
[12:59] <popey> does that work?
[12:59] <OrokuSaki> Or does QTTest have to be compiled?
[12:59] <ogra_> daan, you need to pick one of the options in brackets
[12:59] <diwic> mamenyaka, I might look into reviving the support for /dev/input jack detection then...but I don't promise too much at this point
[12:59] <popey> daan: those are options
[12:59] <wellsb> jdstrand: mdeslaur: My app can't play sound under apparmor or access onlineaccounts.  I just get a bunch of denied messages.  I've added accounts, networking, and audio
[12:59] <mamenyaka> diwic, okay, thank you
[12:59] <daan> that seems to work
[12:59] <daan> now installing
[12:59] <daan> thanks :)
[12:59] <mamenyaka> diwic, so until then we use the priority hack?
[13:00] <diwic> mamenyaka, I had pulseaudio patches for that like a year ago or two, but they were never upstreamed as a different kernel/userspace API was implemented instead
[13:00] <diwic> mamenyaka, and also, the /dev/input/event things need to be readable for the phablet user
[13:00] <mamenyaka> diwic, and what about the volume key event? can we have support for that? I guess the nexus 10 also has it this way
[13:00] <lool> asac: we have the weekly meeting right now
[13:01] <lool> asac: on this topic
[13:01] <mdeslaur> wellsb: could you please file a bug against apparmor-easyprof-ubuntu with the relevant logs and then ping jdstrand with the bug number? thanks
[13:01] <wellsb> Certainly
[13:02] <diwic> mamenyaka, the volume keys are tracked in bug 1219057, but long story short, I'm not the right person to fix it
[13:02] <mamenyaka> diwic, evtest works without sudo as phablet user
[13:02] <diwic> mamenyaka, oh, interesting
[13:02] <tmoenicke> greyback: pong
[13:03] <ogra_> asac, the meeting is in #ubuntu-meeting if you want to read along
[13:03] <mamenyaka> diwic, phablet@ubuntu-phablet:~$ evtest /dev/input/event6
[13:04] <greyback> tmoenicke: hey, as mentioned last week, we've having problem with OSK orientation changes while running on Mir. After lots of trials, we're unable to get the OSK positioning itself correctly all the time
[13:05] <daan> My installation gave an error: "autodeploy.zip not found"
[13:05] <tmoenicke> greyback: i'll check it today
[13:05] <tmoenicke> greyback: there is a hangout later today about this
[13:05] <greyback> tmoenicke: instead we're proposing this idea: have the OSK just use the fullscreen surface, but only draw to a portion of it  - as opposed to changing the QQuickView geometry with setGeometry
[13:05] <greyback> tmoenicke: I know. I'm just giving you a heads-up on what we're thinking
[13:06] <tmoenicke> greyback: sounds good to me
[13:06] <tmoenicke> greyback: it would match what maliit-server does
[13:07] <greyback> tmoenicke: we could use the MainView supplied by the SDK, so the orientation is handled by that code-path instead.
[13:08] <tmoenicke> greyback: the orientation needs to be synced with the applications mainview
[13:08] <daan> My Ubuntu Touch installation gave an error: "autodeploy.zip not found"
[13:08] <greyback> tmoenicke: how do you mean synced up? As in, if application doesn't support rotation, then OSK should not either?
[13:08] <tmoenicke> greyback: yep
[13:08] <nerochiaro> psivaa: oSoMoN: asac: i flashed the newest image, then autopilot run notes_app on maguro. it seems no tests fail
[13:09] <ogra_> daan, on the phone screen ? thats normal if you boot into recovery for the first time
[13:09] <greyback> tmoenicke: makes sense. Ok, you know best about that part
[13:09] <ogra_> daan, as long as it moves on, ust ignore :)
[13:09] <ogra_> *just
[13:10] <daan> Also in terminal: "ERROR:phablet-flash:Command 'adb push /home/daan/Downloads/phablet-flash/ubuntu-touch/20130828.1/saucy-preinstalled-touch-armhf.zip /sdcard/autodeploy.zip' returned non-zero exit status 1""
[13:10] <daan> Phone hangs on Google screen
[13:11] <mhall119> dholbach: ah, now my problem us ureadit is the .desktop file it's generating
[13:11] <mhall119> cjwatson: on installing uReadIt I get the following desktop filename: com.ubuntu.developer.mhall119.ureadit_uReadIt_0.9.2.desktop
[13:11] <diwic> ogra_, as mamenyaka discovered, it seems like the permissions for /dev/input are less restrictive on phone than on desktop. If this is intentional, I could just read the volume up/down hotkeys directly in PulseAudio.
[13:12] <mhall119> but currently Unity8 requires all lower-case .desktop file names
[13:12] <psivaa> nerochiaro: yea notes test are passing in the latest run
[13:12] <ogra_> diwic, hmm, not intentionally, but we should make sure the phablet user can read them i guess
[13:12] <tmoenicke> greyback: ok good. are you gonna be in the hangout today?
[13:12] <nerochiaro> psivaa: so what was going on with the failed ones ?
[13:12] <cjwatson> mhall119: Unity8 bug.  click explicitly permits upper-case
[13:13] <diwic> ogra_, right, that's what it can do currently. On desktop, only root can read these files.
[13:13] <cjwatson> "the application name (must consist only of characters for a Debian source package name, Debian version and [A-Z])
[13:13] <cjwatson> "
[13:13] <psivaa> nerochiaro: i dont know, probably random failures
[13:13] <ogra_> diwic, hmm, they should be managed by udev-acl for users
[13:14] <diwic> ogra_, /dev/input is not managed by udev-acl on the desktop.
[13:14] <nerochiaro> psivaa: lovely heisenbugs ;)
[13:14] <ogra_> diwic, yeah, i see
[13:14] <greyback> tmoenicke: yes I'll be there
[13:14] <psivaa> nerochiaro: :), please ignore the latest run on mako. i've kicked another run
[13:15] <discopig> hi
[13:16] <daan> Okay, my installation failed due to "autodeplay.zip not found". This error also appeared in terminal and the installation stopped after that error. Now my phone is stuck on Google startup screen. What should I do? Thanks
[13:17] <mhall119> jdstrand: is there an apparmor profile I need to launch webbrowser-app from a .desktop file?
[13:18] <diwic> ogra_, so, I don't mind writing some support for volume up/down in pulseaudio by just reading them from there, but if so I should probably know that this privilege is not going to disappear in the near future
[13:19] <ogra_> diwic, yeah, let me find out why it is readable in the first place, i dont think it shoudl differ from desktop
[13:19] <ogra_> diwic, oh, i see
[13:20] <ogra_> diwic, the device is owned by the android_input group
[13:20] <ogra_> which phablet is a member of
[13:20] <asac> psivaa-lunch: plars: can you give the touch_ro images for maguro and mako alsio some love
[13:20] <ogra_> so i think it is possible we will keep that
[13:20] <asac> psivaa-lunch: in the sense of retrying if something completely failed?
[13:21] <ogra_> rsalveti, /dev/input permissions on touch differ from desktop (using the android_input group to allow direct user access). do you think it is likely that we'll keep it that way ?
[13:21] <plars> asac: yes, working my way around to that
[13:22] <mhall119> dholbach: lool: beuno: https://bugs.launchpad.net/unity8/+bug/1214428 is what's stopping uReadIt from launching now, but it installs fine
[13:23] <mhall119> cjwatson: where is click getting the capitalized uReadIt from?  the manifest title?
[13:24] <cjwatson> mhall119: The manifest entry immediately inside "hooks"
[13:24] <cjwatson> mhall119: Which is presumably generated by QtC
[13:25] <asac> plars: yeah. dont start digging deep if you dont know whats going on
[13:25] <asac> plars: i believe lool and the system image team might have ideas as soon as they know what the problem is
[13:25] <asac> or see the log
[13:25] <mhall119> cjwatson: ok, I'll try changing that
[13:25] <OrokuSaki> To get past QTTest with Cmake and the Ubuntu-SDK sudo apt-get install qtdeclarative5*
[13:27] <gusch> ogra_: what's the max file size on the N10?
[13:28] <plars> asac: well, retrying on this one, it seems at least in one of them it was a pretty nasty launchpad/bzr problem: https://jenkins.qa.ubuntu.com/job/saucy-touch_ro-mako-smoke-install-and-boot/58/artifact/clientlogs/utah.log
[13:28] <ogra_> gusch, whatever ext4 defines i'd say
[13:29] <annerajb> ogra_, what do you write on a upstart job so it writes a log file or something?
[13:29] <ogra_> annerajb, http://upstart.ubuntu.com/cookbook/#console
[13:30] <gusch> ogra_: ok
[13:31] <daan> okay, I'm able to get in recovery mode, but I can't boot my phone. How can I restart the Ubuntu Touch installation ?
[13:31] <asac> plars: guess show that to lool if its reproducible
[13:32] <plars> asac: yeah, retrying now, but I suspect it's not
[13:32] <cjwatson> plars: What timestamp was that?
[13:32] <asac> plars: yeah retry then
[13:33] <cjwatson> plars,asac: Saturday, I see - that was around the time when there was heavy datacentre maintenance related to Launchpad
[13:33] <plars> cjohnston: the timestamp I have on the jenkins build itself (not that exact line) is Aug 31, 2013 7:30:40 PM
[13:33] <cjwatson> SAN rearrangement, firewall restarts, etc.
[13:33] <ogra_> daan, keep the phone in recovery and run "phablet-flash cdimage-touch -d mako"
[13:33] <cjwatson> So failures around that time are pretty much expected.  Just retry
[13:34] <asac> cjwatson: cool. thanks for that precious info
[13:34] <daan> Thanks, I'll try that
[13:39] <oSoMoN> psivaa-lunch: have you managed to confirm that the edge swipe intro animation is what makes the browser tests fail?
[13:42] <daan> Oh yeah, it's working. Thank you very much Ogra :)
[13:44] <dholbach> jdstrand, lool: do you know if scopes and webapps as click packages are going to be any additional work?
[13:45] <stgraber> lool: import is running
[13:45] <lool> dholbach: in terms of implementation in click or?
[13:45] <ogra_> daan, enjoy
[13:46] <psivaa> oSoMoN: well i ran it twice with the animation and the tests had a lot of failures and it took more than 30 mins to complete. i also ran it without the animation a couple of times and the all the tests passed within 18 mins
[13:46] <ogra_> pmcgowan, so who owns splitting of the webapps-demo package ? it would be nice if we could drop the PPA this week
[13:46] <lool> dholbach: they need a little bit of integration, like generating the right .desktop (or none), having a way to remove them, reloading scopes and such; there's the whole disucssion about security arch too
[13:46] <pmcgowan> ogra_, alex-abreu, we are working on it
[13:46] <asac> doanac: do you have any insight on what for system-updates your team has to do?
[13:46] <ogra_> pmcgowan, ah, great, sorry, i wasnt aware, didnt mean to appear pushy
[13:47] <lool> pmcgowan: is he also looking after dropping demo-assets?
[13:47] <lool> or just webapps-demo?
[13:47] <asac> doanac: i think stgraber and lool wait for something on our side there
[13:47] <pmcgowan> just webapps
[13:47] <pmcgowan> lool, ^^
[13:47] <lool> ogra_: ^^
[13:47] <oSoMoN> psivaa: seems to confirm your hypothesis then
[13:47] <ogra_> lool, webapps-demo needs splitting into individual apps
[13:47] <psivaa> oSoMoN: if we want more confidence in that we could ask rfowler to get rid of the animation in the lab and rerun the tests :)
[13:47] <ogra_> lool, we will most likely have to drop the demo-assets
[13:47] <dholbach> lool, do you have any blueprints or links for the discussion?
[13:47] <oSoMoN> psivaa: that’d be great
[13:48] <lool> doanac, plars: we're ready to announce new images (read-only), but we need a decent dashboard before we tell people they can use these
[13:48] <lool> doanac, plars: We can discuss in the meeting later today
[13:48] <lool> in 2 hours
[13:48] <plars> lool: sounds good... I'm still getting the same failure though, and I can pull those branches by hand just fine. Trying to sort out why it's failing in jenkins
[13:49] <stgraber> plars: btw, expect the switch in version numbers to happen in a couple of hours, at that point, I'd expect the daily channel's max version to be "2" and the one for daily-proposed to be "3"
[13:50] <stgraber> plars: also not sure if you're the right person to discuss that with but the way daily-proposed works, the same version number is re-used until it gets promoted to daily, so the image in there will always be "3" but with a different content until "3" is published to daily
[13:51] <stgraber> plars: I have a feeling the dashboard and your scripts probably won't like that :)
[13:51] <xnox> Mirv: so yeah the 5.1.1 packaging works much better. I have a few patches and qmake is cross-compiling stuff here now. I need to move all the bits in the package and then I'll make a merge proposal to further flesh it out.
[13:51] <plars> stgraber: well, yes that could cause some confusion
[13:52] <mhall119> beuno: can you review 0.9.3 of uReadIt?  I'm hoping that'll fix the .desktop filename
[13:52] <plars> stgraber: especially if you want to differentiate between the "3" that ran yesterday and the "3" that ran today given that they could be two totally different images
[13:52] <stgraber> plars: I think we should instead use "<system image id> (<rootfs version>/<android version>)" as the version number on the dashboard for touch_ro
[13:52] <ogra_> ++
[13:53] <ogra_> stgraber, except that we dont have the android version atm
[13:53] <plars> stgraber: what's the best place for us to grab (<rootfs version>/<android version>) from?
[13:53] <stgraber> plars: so for today's proposed build we'd have "3 (20130903/20130903)" (assuming android got rebuilt today, not sure)
[13:53] <sil2100> gusch: hi! What's the final verdict about the gallery-app merge request?
[13:53] <sil2100> gusch: since fginther is around now, so I'm sure he could help out with the mediumtests bits
[13:53] <stgraber> ogra_: not on the target system, but on system-image we sure do
[13:53]  * lool &
[13:53] <ogra_> stgraber, we need to dump the android package version somewhere during build i think
[13:53] <sil2100> om26er: ^
[13:53] <lool> brb
[13:53] <ogra_> stgraber, oh, you pull that in directly ?
[13:54] <stgraber> ogra_: and I believe the QA stuff already parses index.json so they can get the version of all the bits from there
[13:54] <plars> stgraber: I would assume we can just use media-info for the rootfs version (just not *after* the upgrade) right?
[13:54] <ogra_> ok
[13:54] <gusch> sil2100: the merginf failed :( (some tests on mako failed) - I'm trying it once again
[13:54] <asac> fginther: hi :)
[13:54] <asac> sorry was wrong channel before
[13:54] <asac> fginther: so today we wondered how to disable PPAs
[13:54] <sil2100> gusch, om26er, fginther: I'm referring to https://code.launchpad.net/~schwann/gallery-app/gallery-atest-toolbar-opened/+merge/183195 , not sure what's the issue, I also re-approved it again
[13:54] <stgraber> plars: I think whatever script you have that monitors system-image's index.json should be generating the long version number, then you use for everything else
[13:54] <asac> fginther: seems we have a bunch of ppas still on during MP processing
[13:55] <asac> fginther: i believe for componetns that go on the image we should only have PPAs on that are included on the image
[13:55] <fginther> asac, right, sil2100 is working on that now
[13:55] <asac> ah ok cool
[13:55] <asac> thanks
[13:55] <asac> thats good enough for me to know
[13:55] <fginther> asac, excellent. thanks again sidnei
[13:55] <om26er> sil2100, gusch fginther the problem is with the uitoolkit emulators, they are somehow failing to switch tabs reliably
[13:55] <fginther> asac, excellent. thanks again sil2100
[13:55] <asac> :)
[13:55] <plars> stgraber: we're just looking for change, I don't think we parse it at the moment
[13:55] <om26er> I have see a similar issue yesterday for clock app as well
[13:55] <fginther> sidnei, sorry for the misfire
[13:55] <plars> stgraber: it's causing us some problems still though too
[13:55] <plars> stgraber: because it appears to change twice
[13:56] <plars> stgraber: so all the touch_ro jobs get kicked off twice
[13:56] <gusch> om26er: SDK reverted already because of that
[13:56] <stgraber> plars: yeah, cjohnston mentioned that, I'm really not sure why that's happening since I confirmed I only trigger rsync once per build on my side...
[13:56] <om26er> gusch, then I guess the above issue will vanish with that as well
[13:57] <fginther> om26er, gusch, do we have to wait for a new image with the reverted sdk?
[13:57] <gusch> om26er: it was started after the revert - I guess mako (where the test fail) has some config issues
[13:58] <om26er> gusch, ughh, I'll look into it now
[13:58] <stgraber> plars: anyway, I think you should parse it, it' trivial to extract the last full image "sorted([image for image in images if image['type'] == full], key=lambda image: image['version'])[-1]", then grab the version of the rootfs and android from the files and if you haven't tested that combination yet, then trigger a test
[13:58] <fginther> om26er, thanks
[13:58] <stgraber> plars: I've got to run for a few minutes but I can give you a python script which does that parsing and returns the long version number we discussed earlier for the latest build in daily-proposed
[13:58] <plars> stgraber: we can take a look, I think cjohnston was going to try to figure out *what* exactly was changing in that json each time
[13:58] <plars> depending on the outcome of that, things might get easier
[13:59] <plars> but first things first, need to figure out why it's failing to run at the moment
[14:00] <stgraber> plars: because currently we can't generate that long version number based on what's on the device, so you need it to come from index.json (as android is built on Jenkins so doesn't know its final version number and repacking a .img on cdimage isn't that trivial)
[14:00] <stgraber> anyway, be back in a bit
[14:00] <kalikiana> mardy: a combo button is planned - it's not exactly the same as a combo box, though. see at the bottom http://design.ubuntu.com/apps/building-blocks/buttons
[14:00] <plars> I need to step away for a moment also... if I don't force myself to find something to eat for breakfast it will probably be 6pm before I remember again
[14:01] <kalikiana> tvoss_ who would review sensors patches? https://code.launchpad.net/~kalikiana/qtubuntu-sensors/feedback/+merge/183456
[14:02] <tvoss_> kalikiana, don't know for sure, best to ask bzoltan1
[14:03] <kalikiana> k thanks
[14:07] <sil2100> asac: yes, working on that as well, I already removed the qt5-proper PPA, now for the others
[14:07] <dholbach> mhall119, do you know if "0.4" as "Click-Version" was put in there automatically? (is it "x.y" of the click package version?)
[14:08] <cjwatson> dholbach: it's hardcoded in click - currently I make sure it always matches the x.y prefix of the click package version
[14:08] <dholbach> cjwatson, gotcha, thanks
[14:12] <doanac> lool, asac: I believe plars and cjohnston: turned off touch_ro testing because we couldn't test much. There wasn't enough free space to install touch $app-autopilot packages required to run the tests
[14:12] <sergiusens> doanac: which brings in the question of, did you test the click based testing mechanism yet?
[14:12] <cjohnston> I didn't turn them off, but I think someone did.. But yes, we can't test the system images as they currently are due to space
[14:13] <doanac> sergiusens: thomi is blocked by a bug #1217700
[14:13] <sergiusens> doanac: also, wrt to autopilot-qt, are we seeding or piloting a dep?
[14:13] <doanac> sergiusens: i think it might be in today's  image. I have an MP out for it
[14:14] <sergiusens> doanac: let's hope so, not sure if the beta block is still in effect
[14:15] <doanac> sergiusens: looks like thomi didn't review: https://code.launchpad.net/~doanac/autopilot/autopilot-touch-depend/+merge/182769
[14:15] <doanac> i'll have to ping him
[14:16] <sergiusens> doanac: I'll review
[14:16] <plars> doanac: it doesn't appear to be disabled, so I was wondering if he turned it back on with running from host
[14:16] <sergiusens> doanac: reading the bug report now
[14:17] <plars> doanac: this is just in the install test though, but it could be due to space issues
[14:18] <gusch> om26er: really only failing on the mako job ... https://code.launchpad.net/~schwann/gallery-app/gallery-atest-toolbar-opened/+merge/183195
[14:18] <sergiusens> doanac: seems the bug is incomplete and it might be that thomi is not importing the dbus address
[14:19] <om26er> gusch, I am logged into a mako and trying to run the test there manually
[14:19] <plars> doanac: the error doesn't look to be space related though
[14:19] <sergiusens> stgraber: what's the ETA for version reset? no rush, just want to know
[14:19] <cjohnston> maguro looks like it was failing to fetch. normally the first 2 to 3 tests would run before it ran out of space
[14:22] <josepht> sergiusens: phablet-flash cdimage-touch --pending -b on mako gives me:
[14:22] <josepht> Ubuntu Saucy Salamander (development branch) - armhf (20130828.1)
[14:23] <sergiusens> josepht: interesting, can you log a bug? ubuntu-bug phablet-tools
[14:23] <om26er> gusch, ah, the device seems to have an older version of the toolkit
[14:23] <josepht> sergiusens: same happens when I manually install today's image
[14:24] <sergiusens> josepht: oh, wait, what?
[14:24] <stgraber> sergiusens: the initial import is running at the moment
[14:24] <josepht> sergiusens: it
[14:24] <sergiusens> stgraber: thanks
[14:24] <josepht> 's strange I know
[14:24] <gusch> om26er: yesterday the problem was that it had an older version of the ui-extras
[14:25] <sergiusens> josepht: can you run that same command again and add a --debug and pastebin it for me? Are you behind some sort of caching proxy?
[14:25] <stgraber> sergiusens: it should actually have finished a few minutes ago but apparently something went racy and decided to do a cleanup run before it finished, removing a bunch of files under its feet, so I'm re-running now with cron disabled
[14:25] <josepht> sergiusens: no proxy that I'm aware of
[14:25] <cjwatson> jdstrand: I think https://bugs.launchpad.net/ubuntu/+source/apparmor/+bug/1219885 is urgent for the system image work
[14:25] <om26er> gusch, we flash new image whenever there is a new image available at cdimage.ubuntu.com so any updates that are released between the time when new images are generated are not installed
[14:26] <om26er> gusch, we could go with apt-get update;apt-get upgrade
[14:26] <cjwatson> jdstrand: ... maybe I misunderstood, ignore that :)
[14:26] <gusch> om26er: so if you don't find the reason why ther is the old version, I could set the version needed by gallery (helped for friends yesterday)
[14:27] <seb128> cjwatson, do you think we could have "pkgdir" in the click list manifest?
[14:28] <om26er> gusch, found the issue :)
[14:28] <om26er> gusch, someone disabled that specific mako to be flashed daily so we had an older version of stuff hence the failure
[14:29] <gusch> om26er: so can I restart the job, did you fix it already?
[14:29] <om26er> gusch, I disabled that device so yep, restart should be fine
[14:30] <cjwatson> seb128: I'm a bit uncomfortable with that since it isn't really a property of the app
[14:30] <cjwatson> seb128: you can't just call it?
[14:30] <cjwatson> seb128: or what else do you need it for?
[14:31] <cwayne_> Wellark, ping
[14:32] <sergiusens> josepht: check /cache/recovery/log please
[14:32] <Wellark> cwayne_: pong
[14:32] <seb128> cjwatson, to resolve the icons path. I can call "click pkgir <name>" though, no worry, I was just trying to lower the number of system calls, but it's not like it was going to make a big difference in practice
[14:32] <cwayne_> Wellark, any chance the launcher icons are actually using gsettings now? :)
[14:32] <davmor2> ogra_: is there an issue with click in todays image?  I'm not seeing the stockticker or the sudoku apps in the dash
[14:33] <Wellark> cwayne_: sorry, I'm kinda emotional right now.. Stephen Elop finally closed the task he was sent out to do :/
[14:33] <cjwatson> seb128: ultimately once click stabilises I'll probably be rewriting it as a C library with a much faster interface
[14:33] <ogra_> davmor2, no idea, i didnt havee time to even take a look at todays image :)
[14:33] <Wellark> cwayne_: it will use accounts service. mzanetti has more info
[14:33] <cjwatson> seb128: so if it's not a blocker for 13.10 then I don't think you need to worry about optimising it now
[14:33] <cwayne_> Wellark, daw, I saw that :(
[14:33] <psivaa> oSoMoN: http://reports.qa.ubuntu.com/smokeng/saucy/image/3930/webbrowser-app-autopilot/ is the webbrowser results in maguro after (manually)disabling the animation
[14:33] <seb128> cjwatson, ok, as said it's mostly non-user-visible optimization ... works for me
[14:34] <seb128> cjwatson, thanks ;-)
[14:34] <Wellark> maybe the old Nokia share holders finally get over to sue him
[14:34] <oSoMoN> psivaa: all green, I like that :)
[14:34] <psivaa> oSoMoN: it took only 15 mins to complete
[14:34] <mzanetti> cwayne_: what exactly do you need?
[14:34] <davmor2> popey: do you see stock ticker and sudoku on todays image?
[14:34] <oSoMoN> psivaa: so can we ensure that the animation is always disabled before running tests?
[14:34] <Wellark> I don't know about the US but at least in Finland it's illegal for a CEO to act against the best interest of his company
[14:35] <cjwatson> seb128: I just don't want to shift it out of a dynamic language when I'm still rewriting half of it once every week or two :)
[14:35] <cwayne_> mzanetti, i need to customize what apps are on the launcher
[14:35] <cwayne_> preferably via gsettings key
[14:35] <mzanetti> cwayne_: yeah. you can do that
[14:35] <cwayne_> Wellark, :(
[14:35] <mzanetti> cwayne_: the merge that reads dconf was approved like 10 minutes ago
[14:35] <seb128> cjwatson, yeah, that makes sense, it's going to be able to iterate easily while the code is still having frequent changes
[14:35] <mzanetti> cwayne_: so its not in the image yet, but should be in 4 hours at latest
[14:36] <psivaa> oSoMoN: asac and mzanetti were discussing that. i dont know the exact out come of that
[14:36] <cwayne_> mzanetti, :D thanks!
[14:36] <seb128> cjwatson, it's going -> it's good
[14:36] <seb128> cjwatson, anyway, I've my icons and installed size so I'm happy, thanks again for those!
[14:36] <cjwatson> great
[14:36] <popey> davmor2: no, missing
[14:37] <mzanetti> cwayne_: you can track status here: https://code.launchpad.net/~mzanetti/unity8/launcher-defaults-from-dconf/+merge/183420
[14:37] <davmor2> popey: I blame that cjwatson it's bound to be his fault ;)
[14:38] <cwayne_> mzanetti, thanks so much
[14:39] <josepht> sergiusens: phablet-flash --debug: http://paste.ubuntu.com/6059094/
[14:39] <josepht> sergiusens: recovery log: http://paste.ubuntu.com/6059095/
[14:41] <sergiusens> josepht: can you open /home/josepht/Downloads/phablet-flash/ubuntu-touch/20130903/saucy-preinstalled-touch-armhf.zip and check /etc/media-info ?
[14:42] <OrokuSaki> Does the SDK currently build click apps on the device??? Having a hard time with this
[14:43] <josepht> sergiusens: Ubuntu Saucy Salamander (development branch) - armhf (20130903)
[14:47] <sergiusens> mzanetti: is this a scope crashing? http://people.canonical.com/~sergiusens/empty_installed.png
[14:49] <sergiusens> lool: have you tested click packages in the latest ro images?
[14:49] <sergiusens> lool: I'm seeing stuff like this: https://bugs.launchpad.net/ubuntu/+source/apparmor/+bug/1219885
[14:49] <jdstrand> I'm fixing that now
[14:49] <sergiusens> jdstrand: ah, great
[14:49] <sergiusens> jdstrand: path changes?
[14:49]  * jdstrand just commented in the bug
[14:50] <jdstrand> stuff moving to /android/ without coordination with apparmor caused the problem
[14:50] <sergiusens> jdstrand: we need stgraber to sync those changes with you :-)
[14:50] <jdstrand> I fixed one of the accesses last week, but the other one is your bug
[14:52] <sergiusens> jdstrand: thanks
[14:53] <ogra_> jdstrand, with that fix /system and /vendor still work ?
[14:53] <sergiusens> josepht: I'm not sure what or where the problem is
[14:53] <ogra_> (we need to support both ways)
[14:53] <sergiusens> ogra_: it's symlinked
[14:53] <ogra_> ah, cool
[14:54] <sergiusens> ogra_: http://paste.ubuntu.com/6059136/
[14:58] <lool> sergiusens: only on grouper
[14:59] <jdstrand> ogra_: yes
[14:59] <lool> sergiusens: are we seeing these failures when running autopilot tests against clicks?
[14:59] <jdstrand> ogra_: http://paste.ubuntu.com/6059148/
[15:00] <jdstrand> ogra_: (I refactored it for simplicity, so I didn't paste the removed rules)
[15:00] <sergiusens> lool: nope, just by running them confined
[15:01] <jdstrand> lool: I would think this would affect autopilot tests only when those tests are running under confinement. as of late last week, work was being done on that front, but it hadn't landed yet
[15:02] <jdstrand> lool: ie, when autopliot tests are using upstart-app-launch
[15:03] <sergiusens> jdstrand: regarding thomi 's bug on autopilot, I'm looking into that
[15:04] <jdstrand> sergiusens: did you see my latest comment in the bug?
[15:05] <sergiusens> jdstrand: the one where you can't reproduce?
[15:05] <jdstrand> yes
[15:05] <mhall119> dholbach: yes I belive 0.4 for click was added automagically
[15:05] <dholbach> mhall119, fixed it in the review tools, thanks
[15:05] <sergiusens> jdstrand: I'm almost sure he is switching to the phablet user in a way that the DBUS_SESSION_BUS_ADDRESS is not imported
[15:05] <kgunn> ricmm: so...is video working?
[15:06] <lool> jdstrand: do you have some pointer on the usage of confinment for tests?  I mean a bug, or a person doing it?
[15:06] <jdstrand> sergiusens: that would definitely cause the bug
[15:06] <jdstrand> lool: see discussion between me and sergiusens right now :)
[15:06] <jdstrand> lool: thomi filed a bug on it. let me find it
[15:06] <lool> jdstrand: just want to make sure this is tracked with adequate priority: bitten by this 2-3 times already, and it breaks a bunch of click packages if not all on at least a couple of devices each time, sounds like a typical image test
[15:06] <lool> jdstrand: ok, so thomi; great
[15:07] <jdstrand> lool: bug #1217700 is an obstacle they hit. see sergiusens' comment for what might be happening. I don't know when this is supposed to land
[15:08] <stgraber> stgraber@castiana:~/Desktop$ python3 parse-systemimage grouper
[15:08] <stgraber> Current full image: 20130900 (ubuntu=20130903, grouper=20130903)
[15:08] <stgraber> Available deltas:
[15:08] <stgraber>  - From '20130835 (ubuntu=20130831, grouper=20130831)' to '20130900 (ubuntu=20130903, grouper=20130903)'
[15:08] <jdstrand> lool: his comment in this channel
[15:08] <stgraber> plars: ^
[15:09] <jdstrand> stgraber: so, all weekend I was trying to use system-image-cli to update, but kept getting gpg errors and it would bail (I am on daily-proposed). is this a known issue?
[15:10] <ricmm> kgunn: yea
[15:10] <plars> stgraber: does that work off the json file on the download server? or off of some local information in the image?
[15:10] <stgraber> jdstrand: hmm, not a known issue, no though I can't really help debug it at the moment because I'm re-generating all the channels for the new version numbers
[15:10] <ricmm> kgunn: should be
[15:10] <stgraber> plars: off the download server
[15:10] <kgunn> ricmm: \o/ greyback ^
[15:10] <plars> mterry: ping
[15:11] <lool> jdstrand: ok thanks
[15:11] <plars> stgraber: so... hopefully we don't hit it during an update
[15:11] <jdstrand> stgraber: that's fine. will that require reflashing the device? will you send an email that we need to do this?
[15:11] <stgraber> jdstrand: yeah, I'll blog + send an e-mail about it. I'll also clarify what daily-proposed is and how people are meant to use it (or rather, not use it).
[15:12] <cwayne_> sergiusens, so about my keyboard issues again :)  we have a config based on this https://wiki.ubuntu.com/Touch/ReleaseNotes#Keyboard and ever since 829.1 the keyboard wont launch unless this config is removed
[15:12] <cwayne_> sergiusens, just wondering who to ask about fixing
[15:12] <jdstrand> stgraber: ok. I'll wait until after all these changes and try again. if I still have the problem, I'll report it
[15:12] <sergiusens> cwayne_: ok, libmaliit-keyboard needs to change to libubuntu-keyboard ... since bfiller is on today he can give you more info
[15:13] <cwayne_> ah
[15:13] <plars> mterry: I'm trying to get rid of this intro screen using what you gave me, and it doesn't seem to be working
[15:13] <cwayne_> sergiusens, i was hoping it was something like that :D
[15:13] <cwayne_> sergiusens, ill try it out and if it works ill update the wiki
[15:13] <mterry> plars, ok
[15:14] <plars> mterry: I'm using: dbus-send --system --print-reply --dest=org.freedesktop.Accounts /org/freedesktop/Accounts/User32011 org.freedesktop.DBus.Properties.Set string:com.canonical.unity.AccountsService string:demo-edges variant:boolean:false
[15:14] <mterry> plars, ooh
[15:14] <bfiller> cwayne_: for now change libmaliit to libubuntu and things should be fine
[15:14] <sergiusens> cwayne_: something like this: https://code.launchpad.net/~thomas-moenicke/maliit/maliit-framework-ubuntu-serverconf/+merge/180612
[15:14] <mterry> plars, I have an idea.  Often the first request to accountsservice will only wake it up and it won't actually act on it (seems to be some bug in AS).  Maybe try the command twice?
[15:15] <stgraber> plars: so basically the idea is that this script tells you what needs testing. In the output I pasted above, it shows the latest image published to daily-proposed and tells you that you can flash it directly (full) or flash 20130835 and update to it. Internally you should track the long version number it gives you and if that changes, then re-trigger the tests.
[15:15] <jdstrand> stgraber: fyi, speaking to sergiusens comment about you syncing my fix for apparmor-easyprof-ubuntu. I just uploaded 1.0.26
[15:16] <plars> mterry: I ran it 5 or 6 times, then rebooted, still comes up with the intro
[15:17] <sergiusens> jdstrand: going to manually install that now to confirm the autopilot issue is a non issue
[15:18] <mterry> plars, OK.  I'm assuming you don't get any errors or interesting response from AS
[15:18] <jdstrand> sergiusens: sounds good
[15:18] <mterry> plars, if you do the following, I assume you get back 'false':   dbus-send --system --print-reply --dest=org.freedesktop.Accounts /org/freedesktop/Accounts/User32011 org.freedesktop.DBus.Properties.Get string:com.canonical.unity.AccountsService string:demo-edges
[15:18] <plars> mterry: all it gives me back is "method return sender=:1.24 -> dest=:1.49 reply_serial=2"
[15:18] <plars> mterry: no, I still get true back when I run that
[15:19] <mterry> plars, ah...  OK, hrm
[15:19] <plars> mterry: wait
[15:19] <plars> mterry: if I run it, then check, I get false
[15:19] <cwayne_> bfiller, thanks brochacho
[15:19] <plars> mterry: but if I reboot, it comes back as true
[15:19] <mterry> plars, curious
[15:20] <plars> stgraber: unfortunately, triggering off of a script output isn't something we can do too easily I don't think
[15:20] <plars> stgraber: once we know that somethings changed, we could inspect the json to see what we're testing (better to look on the image because then we know what we actually got vs. what we think we were supposed to get)
[15:21] <plars> stgraber: but detecting a change is much better done by watching the json file itself
[15:21] <mterry> plars, so AS keeps its data in /var/lib/AccountsService/users/*
[15:21] <mterry> plars, anything interesting in there?  Any reason it would be reset on reboot?
[15:22] <plars> mterry: I am trying this on one of the touch_ro (system images) at the moment... could it be that it just can't write to an appropriate location?
[15:23] <plars> yeah, can't write there
[15:23] <plars> ok, so if I mark this image as writable, it *should* work
[15:23] <plars> I just can't test it easily right now
[15:23] <plars> mterry: ok, thanks... I think I have what I need, I'll just have to wait until I can test it in a bit
[15:24] <plars> mterry: I have some people telling me that the intro is causing problems on the app tests, but I'm not seeing how at the moment. Everytime I've watched it run locally, the app comes up over the top of unity and the intro, so it doesn't seem to interfere
[15:24] <dholbach> and uShopper is in the software store (and it installs and launches!) \o/
[15:28] <sergiusens> jdstrand: after installing the apparmor-easy package I need to rerun the hooks, right?
[15:28] <jdstrand> sergiusens: yes. easiest is 'sudo aa-clickhook -f'
[15:28] <sergiusens> ah, thanks
[15:29] <jdstrand> sergiusens: we are going to fix that, but I want to see what is happening with stgraber's adjustments to ro updates before I do
[15:29] <nerochiaro> renato_: i pushed a couple more changes to that merge request. it should be finished now
[15:29] <ogra_> jdstrand, there are a few devices that i think the security team shoudl take a look at ... namely /dev/input but there are also others ... on these devices we need to use the android system groups for granting access to the android blobs ... this causes different access rights than we have on desktop (i.e. the phablet user is in the android_input group and thus has full rw access to /dev/input)
[15:31] <jdstrand> ogra_: that is exceedingly yucky sounding
[15:31] <ogra_> jdstrand, yeah, but the blobs demand that
[15:31] <jdstrand> ogra_: that said, I haven't seen /dev/input come up in any denials
[15:32] <jdstrand> ogra_: is this just something that mir will use?
[15:32] <ogra_> jdstrand, in any case i think the security team should be aware ... we are a) using android system groups the kernel demands and b) have android based udev rules that modify thepermissions compared to i.e. desktop
[15:32] <sergiusens> jdstrand: fwiw your apparmor-easyprof package fixes click apps for me
[15:33] <ogra_> jdstrand, it is something the kernel uses, Mir will use and the blobs that work across the container boundaries need
[15:33] <jdstrand> ogra_: can you send something to a public list mentioning that, cc'ing security@ubuntu.com?
[15:33] <ogra_> will do
[15:33] <jdstrand> thanks!
[15:33] <jdstrand> sergiusens: great! :)
[15:35] <xnox> Mirv: here are three issues that needs solving before we can cross-compile on the host against qt libraries. https://bugs.launchpad.net/ubuntu/+bugs?field.tag=qmake-cross i have something in the works for the first two, but not the last one just yet.
[15:35] <stgraber> ogra_: hey, so if I want a rebuild of the recovery image, what should I do nowadays?
[15:35] <stgraber> ogra_: is that still on Jenkins or is that the android package nowadays?
[15:35] <ogra_> stgraber, upload android
[15:35] <xnox> Mirv: would you be able to rebuild qt with those changes in a test ppa, for i386/amd64/armhf? cause due to datadir location change I think all qml plugins need to be rebuild.
[15:35] <ogra_> oh, indeed, for the cdimage images thats still on jenkins
[15:36] <ogra_> stgraber, so preferably both ... upload android and trigger a jenkins rebuild
[15:36] <ogra_> stgraber, i'll try to get cdimage fixed this week so we only have to use the package
[15:37] <stgraber> ok
[15:37] <stgraber> xnox: what should I do to the android source package to pick up a change I just made to android_bootable_recovery?
[15:37] <sergiusens> doanac: jdstrand: lool I marked https://bugs.launchpad.net/apparmor/+bug/1217700 invalid
[15:37] <AskUbuntu> Ubuntu Touch App developers | http://askubuntu.com/q/340909
[15:38] <doanac> sergiusens: i saw the problem orginally also, but it has gone away for me. autopilot still can't test the app properly, but I'll have to get some input from thomi on what's broke now
[15:38] <xnox> stgraber: wait for it to propagate into a http://phablet.ubuntu.com/export/android-*.tar.xz  & then android package needs uscan && uupdate
[15:38] <stgraber> sergiusens: hey, so what branch are we building android off from nowadays? phablet-10.1 or phablet-saucy? (looking at android_bootable_recovery)
[15:39] <xnox> stgraber: not sure if there are ways to kick off tarball export earlier.
[15:39] <sergiusens> stgraber: phablet-saucy
[15:40] <sergiusens> doanac: let's discuss in 20'
[15:41] <mpt> katie, oreneeshy, tedg: What's the project/package for reporting bugs on the HUD?
[15:41] <tedg> mpt, http://launchpad.net/hud
[15:41] <stgraber> sergiusens: ok, can you kick off an android rebuild?
[15:41] <oreneeshy> mpt: depends
[15:41] <stgraber> sergiusens: also, how do I manually trigger a new export on phablet.u.c?
[15:41] <mpt> thanks tedg, should have guessed :-)
[15:42] <tedg> mpt, I know, you were expecting a github link ;-)
[15:42] <mpt> tedg, the project page says "The backend for..." ... I wanted to report a UI bug
[15:43] <tedg> mpt, Hmm, then http://launchpad.net/unity8
[15:44] <stgraber> lool: I'd like to hold on the renumbering e-mail+blog until that android change lands (switches from 1.2GB system.img to 2GB)
[15:45] <stgraber> lool: as that change requires people to do a full update
[15:45] <stgraber> lool: (reason for the change is that we only have 20MB of free space at the moment)
[15:45] <josepht> sergiusens: I had to reinstall android then phablet-flash to get it working
[15:46] <lool> stgraber: yup
[15:46] <lool> stgraber: this will help QA apt-get install stuff, yes?
[15:47] <stgraber> lool: yep, they'll have 800MB of free space.
[15:47] <lool> cool
[15:48] <sergiusens> stgraber: lool: they shouldn't need to apt-get install stuff
[15:48] <lool> sergiusens: ultimately no, I agree
[15:48] <sergiusens> stgraber: I can kick on off, the export stuff it not under my radar, xnox can do that
[15:49] <stgraber> sergiusens: ok, please kick one now then.
[15:49] <lool> sergiusens: but we want to have our current tests on new images to announce them, and the quickest path is to let them continue to apt-get install there and then work out how to run more from host
[15:51] <stgraber> sergiusens: as for the export, I'm talking about http://phablet.ubuntu.com/export/ which xnox said earlier that he had no idea of what triggers those
[15:52] <xnox> sergiusens: stgraber: there is cron-job there..... not sure about naming scheme, i'm yet to see a DATE.N notation.
[15:52]  * xnox has ssh there, but i've never triggered it.
[15:52] <sergiusens> stgraber: going to take a bit, seems a hybris change broke the build
[15:52] <xnox> rsalveti: are you the p.u.c/export/ mastermind?
[15:54] <sergiusens> xnox: he is, and he's also sick
[15:54] <stgraber> sergiusens: fun... How long do you think it'll take to get that fixed? I was kind of hoping to send the renumbering announcement out in a couple of hours but I need a new build for that...
[15:54] <sergiusens> stgraber: not sure I didn't push this change
[15:55] <xnox> stgraber: I see. I can go in and re-run cron.
[15:55] <sergiusens> ricmm: rsalveti http://pastebin.ubuntu.com/6059339/
[15:55] <xnox> stgraber: our last upstream tarball used was from 0829, so I don't mind clobbering todays with further updates.
[15:55] <xnox> stgraber: are your changes all pushed on phablet now?
[15:56] <sergiusens> stgraber: it's in hybris so it means a package upload
[15:56] <stgraber> xnox: yep, my bits are in
[15:56] <stgraber> xnox: I'll re-check the resulting tarball to make sure they're in there as well
[15:57] <stgraber> sergiusens: do you have a link to that build failure?
[15:57] <sergiusens> stgraber: the failure is in that pastebin above ^
[15:58] <sergiusens> stgraber: full log http://10.97.2.10:8080/job/ubuntu-touch-image/90/console
[15:58] <ricmm> sergiusens: that sounds wrong, is that againstp phablet-saucy ?
[15:58] <ricmm> salveti had pushed the patches and it has been building right for the past few days
[15:59] <sergiusens> ricmm: it's failing since Aug 28
[15:59] <stgraber> xnox: btw, where did you find the cron job on hesperidium? I had a quick look but didn't find anything relevant.
[15:59] <ricmm> sergiusens: fail
[16:00] <sergiusens> ricmm: yup
[16:00] <sergiusens> ricmm: seems rsalveti applied that patch on Aug 28 as well
[16:00] <ricmm> sergiusens: can you link me to the commit ?
[16:01] <OrokuSaki> Anyone know the correct way to compile a QT app? I am trying to compile the camera-app on my device.. cmake -DCMAKE_INSTALL_PREFIX:PATH=/usr ???
[16:01] <shirow> Hi guys. I'm interested in trialing Ubuntu Touch on an x86 device with a touchscreen, can I build an image and just flash it to a CF card or something similar and have it boot? Or is the process more involved? Has anyone done this before?
[16:01] <sergiusens> ricmm: http://bazaar.launchpad.net/~ubuntu-branches/ubuntu/saucy/libhybris/saucy/revision/24
[16:01] <OrokuSaki> cmake can compile and install, but the camera-app is white and doesn't seem to load
[16:01] <ricmm> sergiusens: there should be a parallel patch against phablet-saucy in phablet-git
[16:02] <ricmm> sergiusens: http://phablet.ubuntu.com/gitweb?p=CyanogenMod/android_frameworks_native.git;a=commit;h=5d7499bc9ad4a7cf3bc2a03c2e4ae620daa3d719
[16:02] <ricmm> that needs to build, it looks like it isnt
[16:03] <nyl> any ideas
[16:03] <nyl> http://paste.ubuntu.com/6059357/
[16:04] <cjwatson> mhall119: Why does your ureadit package have the full icon path hardcoded?
[16:04] <ricmm> sergiusens: maybe the image builds against master?
[16:04] <cjwatson> mhall119: That is very very broken
[16:04] <cjwatson> mhall119: (in the .desktop file)
[16:05] <cjwatson> dholbach: ^- Is there a way we can cause such packages to be rejected?  Full paths will break the workflow for preinstalled apps
[16:05] <sergiusens> ricmm: I can check if someone did that, but the latest patches for android wouldn't be there
[16:05] <ricmm> right, well they need to be there for libhybris to build
[16:05] <dholbach> cjwatson, I'll file a bug for it - thanks
[16:06] <ricmm> maybe salveti expected them to be picked up when applying?
[16:06] <mhall119> cjwatson: because it wasn't working previously
[16:06] <sergiusens> ricmm: right, I see the issue, somethings wrong...
[16:06] <cjwatson> mhall119: Please undo it - if it wasn't working it needs to be fixed in click, working around it in your package is causing me problems
[16:06] <cjwatson> mhall119: It should be just "Icon=uReadIt.png"
[16:07] <mhall119> cjwatson: it is in 0.9.3 package
[16:07] <xnox> stgraber: hmm... seems like rsalvetti user is doing it using /home/rsalveti/bin/export_android.sh script
[16:07] <cjwatson> mhall119: aha, I was on 0.9.1
[16:07] <cjwatson> good
[16:07] <cjwatson> mhall119: do you have a URL?
[16:07] <stgraber> jdstrand: pushed an unblock for easyprof
[16:08] <davmor2> cjwatson: any idea why sudoku and stock ticker wouldn't be showing up in todays image?  They are the 2 click demo apps and seemed to of vanished
[16:08] <mhall119> cjwatson: https://myapps.developer.ubuntu.com/dev/click-apps/12/download/
[16:08] <cjwatson> davmor2: No, I see vestiges of them in ~/.local/share/applications/ but I don't know where they went
[16:09] <cjwatson> mhall119: access forbidden
[16:09] <xnox> rsalveti: to eliminate busfactor the cron-job should probably be run as a generic user
[16:09] <mhall119> there's something you don't have access to?
[16:09] <mhall119> I'm shocked
[16:09] <cjwatson> mhall119: apparently :-)
[16:09] <cjwatson> (public service announcement: if you ever find yourself tempted to type "/opt/click.ubuntu.com" anywhere, please check with me first)
[16:10] <cjwatson> (or at least in code you're distributing ...)
[16:10] <mhall119> cjwatson: well now I don't know the right URL
[16:10] <mhall119> it's published though
[16:11] <nyl> are the media frameworks needed by ubuntu touch
[16:11] <nyl> ?
[16:11] <cjwatson> beuno: ^- can you help?
[16:11] <cjwatson> (the myapps thing above)
[16:13] <shirow> Anyone here know if it is currently possible to flash Ubuntu Touch onto a flash card/DOM or something and load it on a regular x86 device?
[16:13] <shirow> All the guides are about flashing it onto an existing Android unit
[16:15] <dholbach> cjwatson, do you have 0.9.1 still lying around somewhere as a test case for me?
[16:15] <cjwatson> davmor2: I expect the preinstall-click stuff is buggy - I'm replacing all that at the moment anyway
[16:15] <cjwatson> dholbach: https://public.apps.ubuntu.com/download/com.ubuntu.developer.mhall119/uReadIt/com.ubuntu.developer.mhall119.uReadIt-0.9.1.click?noauth=1
[16:15] <dholbach> ahh ok, I was trying "ureadit", nevermind - thanks :)
[16:17] <mterry> bfiller, when an incoming call occurs, what piece of code launches the dialer app?  (also, is there a way to fake an incoming call for testing?)
[16:18] <sergiusens> doanac: I think what you saw on friday was the other bug I logged
[16:18] <doanac> sergiusens: you have the bug #?
[16:18] <sergiusens> doanac: https://bugs.launchpad.net/ubuntu/+source/apparmor-easyprof-ubuntu/+bug/1219885
[16:19] <sergiusens> doanac: saw it on Friday, but only got to log it on Monday
[16:19] <sergiusens> doanac: that bug basically blocked any click app from opening
[16:19] <doanac> sergiusens: not sure that was what i saw. looks like dropping letters crashed for you?
[16:20] <doanac> my app launched, but didn't run its autopilot tests
[16:20] <sergiusens> doanac: all apps as access to the video drivers was blocked
[16:20] <sergiusens> doanac: where is the autopilot branch for this?
[16:20] <labsin> shirow,  You can try and install unity8 package from saucy (universe)
[16:20] <nyl> is the S4 a lot similar to Nexus 4?
[16:20] <nyl> components wise
[16:20] <shirow> labsin: thank you, I will check it out
[16:21] <doanac> sergiusens: http://bazaar.launchpad.net/~thomir/autopilot/add-click-package-support/changes
[16:21] <doanac> plus i then i had to update the dropping-letters test case
[16:21] <sergiusens> doanac: ty, branching now
[16:21] <OrokuSaki> when I run the camera-app I compiled.. this is what it says... "Module 'CameraApp' does not contain a module identifier directive - it cannot be protected from external registrations." Any ideas? Does it need to be signed or something?
[16:22] <shirow> labsin: very interesting, thanks, I did not realize this was available, it will help me a lot
[16:22] <shirow> thanks!
[16:22] <labsin> shirow, It doesn't work well with me
[16:22] <shirow> labsin: in what way?
[16:22] <labsin> shirow, the sliding from the side of the screen doesn't work
[16:23] <shirow> ah.. so it 'works'.. it just has features that don't work yet?
[16:23] <shirow> can you build QML touch apps and load them if you're running unity8? or is it purely just an interface?
[16:25] <OrokuSaki> http://pastebin.com/m3wuk2MU camera-app output
[16:27] <dholbach> cjwatson, added a test in the click review tools
[16:27] <labsin> shirow, Actually, I don't know what it should do, I might miss a few things. My click apps are listing, but I can't run them. I had a better experiance running a self compiled Unity8 package, but the public branch had dependies to packages that are updated in sausy and I don't wat to mess with these.
[16:27] <OrokuSaki> How I am compiling the camera-app http://pastebin.com/j0WumWG8
[16:27] <cjwatson> dholbach: great, thanks
[16:29] <shirow> labsin: gotcha. well thanks a lot for the info, that helps
[16:55] <stgraber> xnox: so did you trigger an export?
[16:56] <plars> stgraber: did you just release something to daily?
[16:56] <plars> stgraber: it triggered a test, and its saying the checksum doesn't match
[16:56] <plars> also in daily-proposed
[16:56] <plars> ERROR:phablet-flash:Checksum does not match after download for /home/plars/Downloads/phablet-flash/imageupdates/daily-proposed/ubuntu/ubuntu-20130903.full.tar.xz and hash a3af2773e247a6246e6f171dbb1cc9fa5777167e0eee1b3f319c137fd55eb7b0
[16:57] <stgraber> plars: that was almost an hour ago, but yeah, I did the initial renumbering which meant a full import of all the files, and unfortunately different checksums for all of them
[17:03] <lool> stgraber: would you want to add your script at http://paste.ubuntu.com/6059420/ into ~ubuntu-system-image/ubuntu-system-image/server?  perhaps under an utils/ subdir or something
[17:03] <stgraber> lool: sure, I'll do that
[17:03] <lool> thanks
[17:12] <bfiller> mterry: pretty sure it's /usr/bin/telephony-service-approver process, code lives n the lp:telephony-service branch
[17:12] <bfiller> mterry: not sure if you got last message
[17:12] <bfiller> mterry: pretty sure it's /usr/bin/telephony-service-approver process, code lives n the lp:telephony-service branch
[17:13] <bfiller> mterry: I think you can test with the ofono simulator. let me try and dig up instructions (may be outdated)
[17:15] <xnox> stgraber: running.
[17:17] <sergiusens> stgraber: http://10.97.2.10:8080/job/ubuntu-touch-image/91
[17:17] <cyphermox> ChickenCutlass: pmcgowan: so, I can now successfully push images to the device, and have them end up in the gallery
[17:17] <cyphermox> ChickenCutlass: pmcgowan: now trying to decipher how to download the images
[17:18] <pmcgowan> cyphermox, awesome
[17:19] <stgraber> sergiusens: thanks, triggering a rebuild now
[17:21] <renato_> nerochiaro, I will take a look soon
[17:21] <cyphermox> pmcgowan: and that will probably work now too, I was just being dumb
[17:22] <nerochiaro> renato_: tx
[17:23] <ChickenCutlass> cyphermox, fantastic
[17:24] <cyphermox> yup, can download too
[17:25] <cyphermox> hmm.. with a small utf8 bug
[17:29] <lool> dpm: http://paste.ubuntu.com/6059598/ <- cross-build attempt of qzxing-plugin; so most of the way there, now need to move to cmake and wrap this up nicely
[17:30] <sergiusens> sforshee: https://code.launchpad.net/~sforshee/powerd/thermal-shutdown/+merge/183179 is done with ci
[17:32] <dpm> wow, nice one lool!
[17:33] <lool> dpm: actually just confirming that things were working as expected down to the package  :-)  the only gotcha outside of the way I created my local setup was a missing merge of sbuild to get us to latest version that Marc now uploaded today
[17:34] <lool> dpm: next step: moving to new templates and testing that things cross-build and work as well as a native build, then wrapping this up nicely
[17:34] <nyl> is it normal recovery to say no os installed
[17:34]  * lool walks away for today
[17:34] <dpm> lool, that is really awesome, have a nice evening!
[17:35] <pmcgowan> mterry, can you modify the daily landing stack definitions or is that only didier?
[17:44] <bfiller> mzanetti, greyback : can you point me at the code in the Apps lens that allows the section headers to stay visible beneath the header as you scroll? trying to do something similar in contacts
[17:44] <mzanetti> bfiller: we reimplemented our own ListView
[17:44] <greyback> bfiller: we use a custom component to do that. Lemme find it
[17:45] <mzanetti> bfiller: tsdgeos, that is
[17:45] <bfiller> ah
[17:45] <bfiller> thanks
[17:45] <mzanetti> bfiller: ListViewWithPageHeader (the most hated word in unity8 team btw) :P
[17:46] <greyback> bfiller: check out lp:unity8, in the "plugins" you'll see a directory named that
[17:46] <bfiller> mzanetti, greyback: thanks guys
[17:52] <cwayne_> mzanetti, hey, so this will just be merged and built automatically now that it's approved, right? https://code.launchpad.net/~mzanetti/unity8/launcher-defaults-from-dconf/+merge/183420
[17:52] <nyl> UT just won't load
[17:52] <nyl> http://paste.ubuntu.com/6059668/
[17:52] <mzanetti> cwayne_: yes, once our CI is fixed. I'm working on that right now
[17:52] <cwayne_> mzanetti, ah wonderful, thank you :)
[17:52] <OrokuSaki> Sweet.. successfully recompiled a working camera app... I must have messed up something or something
[17:53] <OrokuSaki> now I can try to modify it
[17:53] <plars> stgraber: so what's the best way to fix this checksum problem?
[17:53] <plars> stgraber: I assume clear my download cache? But that shouldn't be needed every time right?
[17:54] <stgraber> plars: yeah, wipe your download cache. This should be a one time thing.
[17:54] <stgraber> plars: the reason for it was that when renumbering all the files were re-imported from cdimage leading to different check sums, unless we decide to change numbering scheme again, it should happen anymore
[17:58] <stgraber> ogra_: so I just finished a rebuild of ubuntu-touch on cdimage but the resuting image didn't pick up the new recovery partition image from Jenkins, any idea what happened?
[18:02] <stgraber> sergiusens: actually, looks like it's a question for you, why is http://10.97.2.10:8080/job/ubuntu-touch-image-phablet-saucy/lastSuccessfulBuild/artifact/archive/saucy-preinstalled-recovery-armel+mako.img and http://10.97.2.10:8080/view/Phablet/job/ubuntu-touch-image/91/artifact/archive/saucy-preinstalled-recovery-armel+mako.img not the same file?
[18:04] <stgraber> stgraber@castiana:~/Desktop/test$ md5sum *
[18:04] <stgraber> eb3c0e61341bec3e8ec163a234748375  91.img
[18:04] <stgraber> b62580ce7e908c3963ac65e2d1ae8116  last.img
[18:04] <stgraber> sergiusens: ^
[18:04] <stgraber> that caused my rebuilt image to still contain the old recovery image, so we need to have this fixed then I can rebuild the image and actually get the right thing published
[18:06] <sergiusens> stgraber: let me check
[18:07] <stgraber> sergiusens: I think I found why, I wasn't checking the right project on Jenkins
[18:07] <stgraber> sergiusens: ubuntu-touch-image vs ubuntu-touch-image-phablet-saucy
[18:07] <sergiusens> stgraber: yeah
[18:07] <sergiusens> I'm seeing that now
[18:08] <sergiusens> stgraber: and I triggered the wrong job...
[18:08] <sergiusens> stgraber: ok, goes number 2, should be done in ten
[18:09] <stgraber> sergiusens: ok, thanks. I'll re-trigger an image build once that one's done.
[18:09] <sergiusens> stgraber: the switch was handled differently, I always kept ubuntu-touch-image as the official build
[18:09] <stgraber> plars: so you should have an image with 800MB of free space in the next hour or so (sorry for the delay...)
[18:10] <sergiusens> it was done in a hacked way since the building in the archives was supposed to happen days from the switch
[18:24] <anze> what about porting UTouch to motorola milestone 1?
[18:25] <anze> i got milestone 1 rooted, with Cyanogenmod 7
[18:28] <jdstrand> is there some trick to getting data to work on mako? I am up to date within a few days, but data doesn't work. I thought it did at one point... I get a nice 3g indicator, but data doesn't actually work
[18:28] <pmcgowan> jdstrand, I thought if you disabled wifi it would kick in
[18:29] <pmcgowan> cyphermox, would know
[18:29] <sergiusens> jdstrand: can you run http://bazaar.launchpad.net/~sergiusens/+junk/network/view/head:/network_gprs_provision_test.sh
[18:29] <jdstrand> pmcgowan: I tried that at one point
[18:29] <cyphermox> jdstrand: pick NM from proposed
[18:29] <jdstrand> sergiusens: sure
[18:30] <sergiusens> pmcgowan: data is always enabled (unless manually not), if wifi is on, it's the default gw
[18:31] <pmcgowan> right
[18:32] <pmcgowan> sergiusens, yeah not working here
[18:32] <plars> stgraber: do we ever have the possibility that we might build one target, but not others?
[18:32] <jdstrand> cyphermox: I still have 0.9.8.0-0ubuntu19. I guess I should have 0.9.8.0-0ubuntu20?
[18:33] <stgraber> plars: what do you mean by target?
[18:33] <cyphermox> jdstrand: it will very likely hep, yes
[18:33] <plars> stgraber: for instance, if we built a new system image for maguro but not mako
[18:34] <stgraber> plars: it's possible we'd rebuild android just for one of them, yes
[18:34] <sergiusens> pmcgowan: do what cyphermox suggests
[18:34] <stgraber> plars: which would lead to only that device getting an update (with android being the only thing changing)
[18:34] <pmcgowan> ack
[18:35] <sergiusens> stgraber: http://10.97.2.10:8080/view/Phablet/job/ubuntu-touch-image-phablet-saucy/49/console
[18:35] <stgraber> sergiusens: yep, I noticed, thanks!
[18:35] <sergiusens> stgraber: with one update from the git sources
[18:35] <sergiusens> Running repo sync
[18:35] <sergiusens> From git://phablet.ubuntu.com:9419/CyanogenMod/android_bootable_recovery
[18:35] <sergiusens>    b4760bf..7e03a15  phablet-saucy -> phablet/phabl
[18:35] <stgraber> sergiusens: confirming the recovery image is fine before triggering a build this time
[18:36] <stgraber> sergiusens: and it's good! triggering
[18:38] <AlanBell> anze: ask in the channel, I know a lot less than you do, but other people here might be able to help
[18:56] <pmcgowan> sergiusens, cyphermox upgraded to nm 20 and still no 3g data
[18:56] <cyphermox> hrm
[18:56] <cyphermox> after rebooting the phone and all?
[18:56] <pmcgowan> yes
[18:56] <cyphermox> what does /usr/share/ofono/scripts/list-contexts return?
[18:56] <pmcgowan> rebooted, connected to wifi, turned wifi off
[18:57] <sergiusens> pmcgowan: run that script I mentioned... http://bazaar.launchpad.net/~sergiusens/+junk/network/view/head:/network_gprs_provision_test.sh
[18:57] <pmcgowan> cyphermox, http://pastebin.ubuntu.com/6059905/
[18:58] <cyphermox> pmcgowan: according to this you'd be connected
[18:58] <cyphermox> pmcgowan: so the next step is "ip route"
[18:58] <sergiusens> cyphermox: maybe his default route is not setup? pmcgowan ?
[18:58] <cyphermox> sergiusens: ^
[18:58] <cyphermox> :D
[18:58] <pmcgowan> when would I set up the default route?
[18:59] <pmcgowan> ip route says
[18:59] <pmcgowan> 10.184.173.200/30 dev rmnet_usb0  proto kernel  scope link  src 10.184.173.201
[19:00] <pmcgowan> sergiusens, cyphermox ?
[19:00] <sergiusens> pmcgowan: it should be done for you
[19:00] <sergiusens> pmcgowan: I don't see a ril_0 dev there
[19:02] <sergiusens> stgraber: is logrotate supposed to work on these images?
[19:03] <stgraber> sergiusens: I think so, /var/log is writable
[19:03] <sergiusens> stgraber: error: error creating state file /var/lib/logrotate/status: Read-only file system
[19:03] <sergiusens> stgraber: should I log a bug?
[19:03] <sergiusens> stgraber: not urgent
[19:07] <stgraber> sergiusens: sounds like /var/lib/logrotate should be persistent then
[19:08] <sergiusens> stgraber: that's lxc-android-config, right? I'll make the change
[19:09] <stgraber> sergiusens: fixed
[19:10] <stgraber> (added and uploaded)
[19:10] <sergiusens> stgraber: great, thanks
[19:10] <sergiusens> you can tell I'm using the image more now ;-)
[19:13] <pmcgowan> cyphermox, sergiusens anything else I can check for that data connection?
[19:14] <cyphermox> pmcgowan: if you only get that one line as output of ip route, then your default route isn't set
[19:14] <pmcgowan> cyphermox, what sets it?
[19:14] <cyphermox> are you sure you restarted NM / restarted the phone after getting the network-manager packages?
[19:14] <cyphermox> pmcgowan: NM should
[19:14] <pmcgowan> cyphermox, I simply installed and rebooted
[19:14] <cyphermox> alright, just a second
[19:14] <pmcgowan> let me double confirm
[19:15] <cyphermox> pmcgowan: dpkg -l | grep 0ubuntu20
[19:16] <pmcgowan> network-manager                                       0.9.8.0-0ubuntu20
[19:16] <pmcgowan> cyphermox, hang on
[19:17] <pmcgowan> cyphermox, its working
[19:17] <sergiusens> pmcgowan: so you just needed time for it to really connect
[19:17] <sergiusens> ?
[19:17] <cyphermox> pmcgowan: ok... otherwise I'll need to look at /var/log/syslog, NM rarely is unable to set the default route, but in all cases it outputs something if it fails
[19:17] <pmcgowan> sergiusens, either that or I never rebooted, but
[19:17] <cyphermox> sergiusens: no.. the device route was established
[19:18] <cyphermox> possibly wifi was connecting by unsuccessfully?
[19:18] <cyphermox> I'm not sure what would cause the default route to be completely unset, I tested this specifically and it does switch between both with the right preference
[19:19] <pmcgowan> cyphermox, what would I look for in the logs
[19:21] <pmcgowan> ip route says this now
[19:21] <pmcgowan> default via 10.188.245.138 dev rmnet_usb0  proto static
[19:21] <pmcgowan> 10.188.245.136/30 dev rmnet_usb0  proto kernel  scope link  src 10.188.245.137  metric 13
[19:21] <pmcgowan> r
[19:56] <stgraber> lool: hmm, so we've got another small problem. We need a flipped image to be promoted to current before I can send out the e-mail...
[19:56] <stgraber> lool: that's because phablet-flash picks up its recovery image from /current/ and so won't have the 2GB upgrade until the current pending image is marked as good and promoted to current
[19:57] <rickspencer3> mdeslaur, how would I use a sqlite database that I want to ship with an app? (given app isolation)
[19:58] <mdeslaur> rickspencer3: you should be able to read it if it's in the app's install directory
[19:58] <rickspencer3> mdeslaur, can I write to it there?
[19:58] <mdeslaur> rickspencer3: no
[19:58] <rickspencer3> mdeslaur, so if I wanted to write to it, what should I do?
[19:58] <mdeslaur> rickspencer3: you'll have to read from the one you ship, and then create a per-user one which you can then write to
[19:58] <rickspencer3> (not actually certain that I want to, but I am interested in the answer just in case)
[19:59] <jdstrand> you could ship it, then copy it into your data directory
[19:59] <mdeslaur> right, what jdstrand said
[19:59] <rickspencer3> ok
[19:59] <jdstrand> note that there are two things preventing writes to the install dir: apparmor and unix permissions (the install directory is writable by the click user, not regular users)
[19:59] <rickspencer3> but if I just want to read from it, I can just leave it with the code
[20:00] <jdstrand> rickspencer3: yes
[20:00] <jdstrand> @{CLICK_DIR}/@{APP_PKGNAME}/@{APP_VERSION}/**  mrklix,
[20:01] <jdstrand> those permissions should work fine for reading ^
[20:01] <rickspencer3> jdstrand, if I just have it next to my qml, I should be able to just use a relative path, right?
[20:01] <rickspencer3> or no, because it get's executed from some whacky directory somewhere?
[20:02] <jdstrand> rickspencer3: yes
[20:02] <jdstrand> rickspencer3: you should be chdir'd to the install directory
[20:02] <rickspencer3> ok
[20:02] <rickspencer3> thanks jdstrand and mdeslaur
[20:02] <mdeslaur> yw
[20:03] <jdstrand> np
[20:05] <jdstrand> cyphermox, sergiusens, pmcgowan: ok, I finally got the new nm and rebooted. data seems to be working now. if I uncheck my wifi network (but leave wifi enabled), '3g' icon is displayed and I can ping www.ubuntu.com
[20:08] <pmcgowan> sergiusens, do we still plan to tag a version prior to Mir and RO becoming default? and will that then be legacy to get the tag?
[20:08] <sergiusens> pmcgowan: I thought we weren't going to tag any more images
[20:09] <cyphermox> jdstrand: cool
[20:09] <sergiusens> pmcgowan: there is no tag concept in the image based upgrade system
[20:09] <pmcgowan> sergiusens, I ask because I occasionally get requests for something known good for demo (like today)
[20:09] <jdstrand> cyphermox: curious how the '3G' icon is being picked. I should have T-Mobile's version of 4G
[20:09] <pmcgowan> sergiusens, prior to move to image upgrades
[20:10] <cyphermox> jdstrand: I think it's the same for 4G and 3G, which just one generic icon for that
[20:10] <cyphermox> jdstrand otherweise we can try something
[20:10] <sergiusens> pmcgowan: well, it wasn't planned; who's going to stamp the image as good?
[20:10] <pmcgowan> sergiusens, just waiting for one that is
[20:11] <jdstrand> cyphermox: ok thanks. this may be out of your responsibility, but do you know otoh if we will have signal strength for cellular?
[20:11] <pmcgowan> sergiusens, I can ask asac about it and work it out
[20:11] <cyphermox> jdstrand: try /usr/share/ofono/scripts/list-operators I think
[20:11] <sergiusens> pmcgowan: I would think demos would want to show off upgrades too though
[20:11] <cyphermox> jdstrand: should be soon, tedg would know ^
[20:11] <sergiusens> you won't be getting the full expierience
[20:11] <stgraber> sergiusens: btw, in case you didn't see what I told lool above, I'll need one >= 20130903.2 flipped image to be marked as good before we can switch to system images by default
[20:12] <stgraber> sergiusens: that's so phablet-flash can grab a more recent version of the recovery partition (with the 2GB partition size)
[20:12] <stgraber> sergiusens: well, that or we need to teach phablet-flash how to grab the recovery partition from within a system-image, which would actually be better :)
[20:12] <sergiusens> stgraber: that's easier
[20:13] <sergiusens> stgraber: and more appropriate
[20:13] <sergiusens> stgraber: I'll work on it tonight
[20:13] <tedg> jdstrand, Yeah, we will.  Mostly waiting on reviews.  The Unity8 panel doesn't support multiple icons for an indicator yet.
[20:13] <stgraber> sergiusens: do you think you've got time to replace the recovery.img download by, 1) unpacking the device .tar.xz, grab recovery.img and push that one with fastboot?
[20:13] <stgraber> sergiusens: cool, thanks!
[20:13] <stgraber> that way we won't depend on cdimage at all from a user perspective, I like that!
[20:14] <sergiusens> yup
[20:14] <stgraber> lool: ^
[20:14] <jdstrand> tedg: ok, thanks. unrelated fyi, I submitted a patch for bug #1218107
[20:14] <jdstrand> tedg: (in the bug). not sure if I should let you know or someone else
[20:15] <tedg> jdstrand, Oh, hey, I was working on that, but I just screwed up my bazaar repo :-)
[20:15] <tedg> jdstrand, Can you submit it as an MR?
[20:15] <jdstrand> tedg: sure
[20:17] <tedg> jdstrand, I think you need to include some of the nih-dev or dbus-dev deps independently.  If I remember right they were deps of nih-dbus-dev
[20:17] <tedg> jdstrand, Jenkins will tell us for sure though.
[20:17] <jdstrand> tedg: I was flying by the seat of my pants a bit there. It compiled locally. That's about all I can say about it :)
[20:17]  * jdstrand is not a cmake expert
[20:18] <jdstrand> well, locally in sbuild
[20:18] <jdstrand> tbh, I was quite happy for that-- it took me a while to figure out what to do
[20:19] <tedg> Heh
[20:19] <tedg> The NIH memory stuff is interesting, but different.
[20:20] <jdstrand> tedg: so, you don't want a change to debian/changelog, correct? that will all happen when I do the merge request?
[20:20] <tedg> jdstrand, Correct
[20:20] <tedg> jdstrand, It will actually happen on release, but automatically none the less.
[20:20]  * jdstrand nods
[20:21] <jdstrand> I'll try to remember what to do this time
[20:21] <jdstrand> don't remind me :P
[20:24] <davjan> mmm getting            W: Failed to fetch http://ppa.launchpad.net/u2t/bleedingedge/ubuntu/dists/saucy/main/binary-armhf/Packages             on apt-get update
[20:24] <jdstrand> tedg: https://code.launchpad.net/~jdstrand/hud/hud-lp1218107/+merge/183756
[20:24] <jdstrand> tedg: meh, I forgot the LP refernce
[20:24] <jdstrand> reference
[20:25] <jdstrand> tedg: ok, added
[20:25] <tedg> jdstrand, Cool, I'll give Jenkins a chance to get to it.
[20:26] <jdstrand> tedg: thanks :)
[20:30] <sergiusens> doanac: I think I found the problem with autopilot
[20:32] <doanac> sergiusens: what's that? thomi ^
[20:32]  * thomi is listening
[20:32] <sergiusens> doanac: so I recall I had this issue one, but you are using dbus.SessionBus() which starts a new bus, the problem there is that you need to connect to the existing bus
[20:33] <stgraber> plars: so I published build "3" to system-image which contains the updated recovery partition, giving you a 2GB filesystem. Unfortunately we need a change in phablet-flash to have it use that recovery image instead of the one from cdimage, so it won't do you much good at the moment...
[20:33] <thomi> sergiusens: shouldn't DBUS_SESSION_BUS_ADDRESS env var take care of that?
[20:33] <plars> stgraber: ok, thanks for the update
[20:34] <stgraber> plars: if you want to manually test it, you can do a first flash, then on the device do "system-image-cli -b 0" which will trigger another full update and will give you the 2GB system.img
[20:34] <stgraber> plars: otherwise wait for tomorrow and we should have an updated phablet-flash which will use the right recovery image by default
[20:36] <sergiusens> thomi: I think you have to use  dbus.bus.BusConnection
[20:36] <sergiusens> thomi: I had this problem with ufa, remember?
[20:37] <thomi> sergiusens: I don't remember, sorry. I'm like 90% sure that it uses that env var if it's present
[20:37] <thomi> however, last time I checked, it wasn't exported for upstart jobs
[20:37] <thomi> (I assume that's the problem we're talking about)
[20:37] <sergiusens> thomi: I did a initctl set-env DBUS_SESSION_BUS_ADDRESS=$DBUS_SESSION_BUS_ADDRESS and nil
[20:38] <thomi> 'and nil'?
[20:39] <sergiusens> thomi: and nothing :-)
[20:39] <thomi> you mean it still didn't work?
[20:40] <sergiusens> thomi: yeah, I meand that
[20:40] <sergiusens> meant
[20:40] <thomi> OK :)
[20:40] <thomi> my device flash is finishing now, should be able to test it soon
[20:40] <thomi> hmmmm
[20:41] <thomi> sergiusens: is it possible that my device has hung mid-flash?
[20:41] <thomi> The spinning thing in the android has stopped... spinning
[20:41] <thomi> for liek 5 minutes now :-/
[20:41] <sergiusens> thomi: image based upgrades?
[20:42] <thomi> sergiusens: no, but just as I typed that, it came back
[20:43] <thomi> sergiusens: should click packages work on the grouper?
[20:43] <sergiusens> thomi: you should be able to check adb shell cat /tmp/recovery.log (for the temprorary install/flash log)
[20:44] <sergiusens> thomi: as far as I know, yes
[20:44] <sergiusens> thomi: mhall119 is an active n7/grouper user
[20:44] <sergiusens> and does click
[20:44] <thomi> ok, just trying your thing now :)
[20:46] <thomi> doanac: sergiusens: I notice libautopilot-qt isn't in the images, did you decide to leave it out?
[20:46] <sergiusens> thomi: doanac had an MR, I approved today
[20:47] <thomi> ahh cool
[20:47] <pmcgowan> cyphermox, one more question, I seem to have disabled the 3g with the settings panel but it does not re-enable it, any command line magic?
[20:48] <cyphermox> pmcgowan: try nmcli nm wwan off ; nmcli nm wwan on
[20:49] <pmcgowan> nope still off
[20:50] <slangasek> thomi: so I'm coming back around to trying known-good autopilot tests locally via 'adb shell'... and the first unity8 test I try complains about not having GRID_UNIT_PX set in the environment
[20:50] <slangasek> no idea what that's supposed to be set to
[20:51] <thomi> veebers: can you answer that? ^^
[20:51] <sergiusens> slangasek: that's imported from ubuntu-session-manager-touch
[20:51] <sergiusens> should be
[20:51] <thomi> slangasek: veebers is the QE for the unity8 team :)
[20:51] <thomi> huh, or sergiusens, whatever :)
[20:52] <slangasek> sergiusens: sorry, what does that mean?
[20:52] <slangasek> sergiusens: the problem is, I don't even care about running the unity8 tests, I'm trying to debug a *different* autopilot test and it was suggested I could use the unity8 tests as a baseline because these were known to work when 'autopilot run'ning from adb shell
[20:53] <sergiusens> slangasek: grep GRID /etc/ubuntu-touch-session.d/mako.conf
[20:53] <sergiusens> GRID_UNIT_PX=18
[20:53] <sergiusens> something like that
[20:53] <cyphermox> pmcgowan: tbh I'd normally just restart the phone at that point :/
[20:53] <cyphermox> sergiusens: did you have a trick for this?  ^
[20:53] <pmcgowan> cyphermox, hah ok
[20:53] <pmcgowan> weird that wifi control works but not wwan
[20:54] <sergiusens> cyphermox: I can't think of anything now but to restart nm _and_ ofono
[20:54] <sergiusens> used to work for me
[20:55] <sergiusens> thomi: I see in dbus_handlers.py there's a get_custom_bus
[20:55] <slangasek> sergiusens: ok; that lets me clear the immediate error, then it wants some other indicators-client package which is not installed <sigh>  iterating...
[20:56] <slangasek> sergiusens: thanks for the pointers
[20:56] <sergiusens> slangasek: the best test to test for me is the share-app
[20:57] <tedg> mhall119, cordova apps will be packaged as clicks, right?
[20:57] <mhall119> tedg: should be, yes
[20:58] <pmcgowan> slangasek, that variable will be set if you logged in as phablet user
[20:58] <tedg> mhall119, Cool, that's what I thought.
[20:58] <slangasek> pmcgowan: I'm specifically not logging in, I'm adb'ing in
[20:58] <pmcgowan> slangasek, yeah so not set
[20:58] <tedg> mhall119, Are we doing some Mozilla OpenWebApp to click converter?
[20:58] <sergiusens> slangasek: ah, that's a problem
[20:59] <slangasek> pmcgowan: well, that's the standard way to remote into a device, and TTBOMK is how we're driving the autopilot tests in the lab, isn't it?
[20:59] <sergiusens> slangasek: this is how you want to do it http://bazaar.launchpad.net/~phablet-team/phablet-tools/trunk/view/head:/phablet-test-run#L92
[20:59] <slangasek> sergiusens: ah
[20:59] <thomi> sergiusens: got a call, bbs
[20:59] <slangasek> sergiusens: is that what's used in the lab?
[21:00] <sergiusens> well, the options aren't necessary
[21:00] <mhall119> tedg: ask alex-abreu
[21:00] <sergiusens> slangasek: lab is a bit more complicated, they are migrating to this, but currently an upstart job is created that runs the tests, let me search for that code
[21:00] <slangasek> sergiusens: if the lab is migrating to this, that solves my main concern - thanks :)
[21:00] <pmcgowan> tedg, as I understand FF has made a bunch of extensions to the API
[21:01] <pmcgowan> tedg, and they would not look like nice ubuntu apps
[21:01] <tedg> pmcgowan, Ah, bummer.
[21:06] <sergiusens> jdstrand: thinking about how autopilot works, the tests would need access to the bus
[21:06] <sergiusens> thomi: ^^
[21:08] <slangasek> I believe there's an open bug about that
[21:08] <sergiusens> jdstrand: the app under test I mean, so it can export it's object tree over the bus
[21:08] <slangasek> bug #1217700 ?
[21:08] <slangasek> ah, not so open
[21:08] <slangasek> oh, and *you* closed it ;D
[21:09] <sergiusens> slangasek: there was a bug about testability driver and bus, not about this
[21:09]  * slangasek shuts up
[21:09] <sergiusens> slangasek: yeah
[21:09] <sergiusens> slangasek: but was mentioning the testability driver explicitly
[21:10] <jdstrand> sergiusens: yes, that was my conclusion. based on your comment, looks like nothing for the security team to do, correct?
[21:11] <sergiusens> jdstrand: so does an app under confinment have permissions to create export something new (my dbus terminology memory is blank now)
[21:12] <sergiusens> jdstrand: if not, regardless of being able to load the testability driver there would be nothing for autopilot to inspect
[21:12] <stgraber> plars: rootfs                          2.0G  946M  924M  51% /
[21:12] <sergiusens> jdstrand: since it is the application itself that needs to export the tree
[21:13] <stgraber> plars: so the "system-image-cli -b 0" trick works fine and gives you 2GB
[21:14] <jdstrand> sergiusens: I'm not sure I understand the question. an app can adjust its environment sure. the problem in this case was that DBUS_SESSION_BUS_ADDRESS wasn't in the environment that qmlscene was launched in, so it cranked up dbus-launch. so long as the env has DBUS_SESSION_BUS_ADDRESS when qmlscene is launched, then the app will have DBus access according to its policy
[21:16] <sergiusens> jdstrand: so all these apps would need to have 'content_exchange' ?
[21:16] <sergiusens> jdstrand: no, sorry, not that
[21:17] <jdstrand> sergiusens: apps have a certain level of access to the session bus
[21:17] <jdstrand> sergiusens: see /usr/share/apparmor/easyprof/templates/ubuntu/1.0/ubuntu-sdk
[21:18] <jdstrand> sergiusens: the DBus rules are towards the top. some policy groups add additional rules
[21:19] <sergiusens> jdstrand: hmm, so that's going to be the problem
[21:19] <jdstrand> what problem is that?
[21:21] <sergiusens> jdstrand: /com/canonical/Autopilot/Introspection
[21:22] <jdstrand> sergiusens: yes, that would be a problem. based on your comments in the bug, it looks like autopilot initially runs as root?
[21:23] <sergiusens> jdstrand: no, it should run as the phablet user
[21:23] <sergiusens> jdstrand: I'm not an autopilot expert though
[21:23] <jdstrand> hmm
[21:24] <sergiusens> jdstrand: I need to bike home before it gets dark, hopefully thomi can get back to you with this
[21:24] <sergiusens> jdstrand: I'm not sure how to solve this so that it is secure and testable though
[21:24] <jdstrand> if when setting up the environment, a root process could update /etc/apparmor.d/abstractions/base to add the necessary rules, then reload the policy
[21:24] <jdstrand> otherwise we'd need to think about that
[21:25] <sergiusens> jdstrand: it doesn't run as root, but could and then switch
[21:26] <jdstrand> that would make it testable. I'd like to review the policy adjustments in the test environment though
[21:28] <sergiusens> jdstrand: ack... I'll be back in 30' - 1h
[21:28] <jdstrand> sergiusens: ok, I may be eod then, but will check back
[21:29] <sergiusens> jdstrand: lets hope thomi reaches you before that then
[21:29] <sergiusens> but at least we have the problem in view now
[21:30] <jdstrand> yes
[21:30] <jdstrand> fix one problem, find another :)
[21:31] <pmcgowan> awe, I just entered this but maybe should be against ofono  https://bugs.launchpad.net/ubuntu-system-settings/+bug/1220450
[21:32] <thomi> erp, be with you in 30 minutes?
[21:37] <thomi> jdstrand: autopilot runs as the phablet user currently
[21:38] <thomi> and I'd be wary of changing that
[21:38] <thomi> jdstrand: did that answer your questions?
[21:39] <jdstrand> thomi: it does, but that is problematic because it sounds like dbus rules are needed for autopilot
[21:39] <jdstrand> thomi: and adding them to default policy is not ideal
[21:40] <thomi> just to be clear, this is rules for autopilot itself, or for the app under test?
[21:40] <RobbyF> mhall119, busy with that gplus app? :P
[21:40] <jdstrand> thomi: well, I was assuming for the app under the test based on what sergiusens said
[21:41] <jdstrand> thomi: we aren't confining autopilot itself, so there should be no issue
[21:41] <jdstrand> (with it)
[21:42] <thomi> OK, so by default apps under confinement cannot export objects over dbus?
[21:43] <jdstrand> thomi: see the dbus rules in /usr/share/apparmor/easyprof/templates/ubuntu/1.0/ubuntu-sdk for what they can do by default
[21:47] <thomi> jdstrand: and we can't change that unless we're root?
[21:48] <jdstrand> thomi: so is the testing methodology essentially: install some autopilot packages from the archive, set up some environment variables, run the test
[21:48] <jdstrand> thomi: that is correct
[21:48] <thomi> hmm
[21:48] <thomi> jdstrand: how about changing those rules as part of the installation of the 'autopilot-touch' package?
[21:49] <thomi> so, we open app confinement a bit if, and only if that package is installed?
[21:49] <thomi> seems like a reasonable stop-gap measure, until we can think of a better idea
[21:49] <jdstrand> thomi: I was thinking about something like that. to do that, two things need to happen: the rules need to be updated and the rules need to be reloaded
[21:51] <jdstrand> the problem is where to change the rules-- click package policy is in /var/lib/apparmor/profiles, other system policy is in /etc/apparmor.d. files in /etc/apparmor.d are conffiles, so we can't just edit the policy
[21:51] <mhall119> RobbyF: I think I need to do something extra to work within the apparmor profile
[21:51] <doanac> thomi: autopilot-touch is in the default image so things will just always be "moe open"?
[21:51] <doanac> more open
[21:51] <mhall119> running the Exec line by hand makes it work fine, but launching the .desktop from the Dash doesn't
[21:51] <RobbyF> aw
[21:52] <jdstrand> thomi: I could update the the template to include a directory, then you drop a file with the necessary rules into the directory, then we can trigger a policy reload
[21:53] <mhall119> jdstrand: when you have a minute, I have a question for you
[21:53] <jdstrand> autopilot-touch could do both of those things, but I'd like to discuss with my team first
[21:53] <jdstrand> thomi: what kind of timeframe are you looking at?
[21:54] <mhall119> jdstrand: I'm trying to build a click package that has a .desktop file which calls webbrowser-app (instead of containing it's own app), how can I do that within an apparmor sandbox?
[21:54] <jdstrand> thomi: and how does this work with ro images?
[21:55] <jdstrand> mhall119: see lp:~jdstrand/+junk/webapp-test
[21:55] <mhall119> jdstrand: "template": "ubuntu-webapp-experimental", is that the key piece?
[21:56] <jdstrand> mhall119: this is not finalized and the discussion hasn't started yet, but: https://wiki.ubuntu.com/SecurityTeam/Specifications/WebAppsConfinement
[21:56] <jdstrand> mhall119: essentially
[21:56] <thomi> jdstrand: got time for a hangout in 5 minutes or so?
[21:56] <jdstrand> mhall119: see the desktop file too, but that is likely what you already have
[21:57] <mhall119> jdstrand: ok, given that it's not finalized and still in discussion....will it work if I add that to my .json file on the phone?
[21:57] <jdstrand> mhall119: ubuntu-webapp-experimental is currently in apparmor-easyprof-ubuntu, yes
[21:57] <mhall119> cool, I'll try that
[21:57] <jdstrand> mhall119: but that will be renamed once finalized
[21:57] <mhall119> RobbyF: might have a solution :)
[21:57] <mhall119> jdstrand: that's fine
[21:58] <jdstrand> thomi: well, I could, but my guys can't. I'm happy to respond to the thread
[21:58] <jdstrand> thomi: is the idea of autopilot-touch shipping a file and calling something from its postinst to trigger a policy reload palatable to you?
[21:59] <thomi> jdstrand: yup, let's talk with doanac in 2 minutes or so
[22:01] <thomi> OK, inviting your guys now
[22:01] <thomi> jdstrand: https://plus.google.com/hangouts/_/184c8cbef7dfa5810eaaba1113d4989d08de8428?authuser=1&hl=en
[22:03] <awe> pmcgowan, ack.  I'll take a look, but most likely a NM thing
[22:04] <awe> pmcgowan, you mind added which device you used, and what 'command-line' you tried?
[22:05] <pmcgowan> awe, ok will fix
[22:06] <cwayne_> can phablet-flash still work on a local directory?
[22:06] <pmcgowan> awe, i did say mako and wwan on, but fixed it
[22:08] <awe> pmcgowan, sorry...
[22:08] <awe> fyi, the only command line you should be using is a dbus command to toggle the WwanEnabled property
[22:09] <awe> pmcgowan, never used that nmcli command, but that sounds reasonable too
[22:09] <awe> I'll check mine
[22:14] <lool> stgraber: Ok noted
[22:46] <RobbyF> mhall119, awesome! let me know your proper email address to send that too
[23:14] <sergiusens> thomi: hey
[23:15] <thomi> sergiusens: ho
[23:15] <sergiusens> thomi: did you get a chance to chat with jamie?
[23:15] <thomi> sergiusens: yes, he's going to talk to his team and make a recommendation to doanac, who has to fix things in phablet-tools
[23:16] <thomi> we can't fix it in autopilot, since by the time we get to autopilot we've already lost the privileges we need to make the fix
[23:17] <mhall119> sergiusens: pmcgowan: where should I file a bug to get u1db-qt included in the Ubuntu Touch platform?
[23:18] <sergiusens> thomi: yeah, one thing is to modiy the apparmor stuff in /etc from phablet-tools and you'd be tied to the tool from then on, not ideal in a production device as it would require root, but fine our purposes
[23:19] <thomi> yeah
[23:19] <sergiusens> mhall119: is u1db-qt in the archives?
[23:19] <sergiusens> mhall119: just create an MR against the seed it is
[23:19] <sergiusens> mhall119: or file the bug against ubuntu itsel
[23:19] <OrokuSaki> is it possible to run updates with a uImage kernel?
[23:19] <mhall119> sergiusens: I don't think it's in the archive, it's in the sdk-team PPA
[23:20] <mhall119> kalikiana: did u1db-qt make it into the archives for saucy?
[23:20] <sergiusens> mhall119: that's not going to fly... then a bug for the sdk to include in the archives
[23:20] <sergiusens> and MIR it
[23:20] <OrokuSaki> my boot partition holds many uImage files.. because it's awesome
[23:20] <OrokuSaki> each one boots a different OS
[23:20] <mhall119> ah, seems it's in universe
[23:21] <OrokuSaki> but.. I can never update because.. of the kernel.. so I redownload a daily image
[23:21] <OrokuSaki> is that about right?
[23:21] <sergiusens> mhall119: well an MR or the seed would do, but this would need to be in main eventually (soon)
[23:21] <OrokuSaki> I was successful at recompiling the camera and telling it to rotate the camera sensor 90 degrees in landscape mode. =)
[23:22] <OrokuSaki> It was easy... =)
[23:23] <mhall119> MR or the seed, or MR to the seed?
[23:23] <OrokuSaki> Still can't figure out why it takes my android lxc to take a long time to init.. I tell phablet to sleep 60 seconds before starting, and that usually fixes it
[23:23] <OrokuSaki> boot takes about 5 minues or so
[23:25] <mhall119> sergiusens: is lp;~ubuntu-core-dev/ubuntu-seeds/ubuntu-touch.saucy/ the correct target for that MR?
[23:27] <sergiusens> mhall119: yes, was searching or that link
[23:28] <mhall119> sergiusens: the 'touch' file?
[23:29] <sergiusens> mhall119: yes
[23:29] <sergiusens> mhall119: add it under apps
[23:29] <sergiusens> mhall119: then someone like stgraber can get it in (only person that I know that could be active now)
[23:33] <mhall119> sergiusens: stgraber: https://code.launchpad.net/~mhall119/ubuntu-seeds/add-u1db-to-touch/+merge/183784
[23:33] <mhall119> he's in South Africa now though isn't he?  So it's late for him too
[23:34] <sergiuse1s> mhall119: ah, wasn't aware of that
[23:35] <sergiuse1s> cyphermox: I finally got my mobile data to get that patch attached https://bugzilla.gnome.org/show_bug.cgi?id=707289
[23:42] <stgraber> mhall119: looking
[23:42] <stgraber> mhall119: and I'm in Canada, I think you're confusing me with highvoltage :)
[23:43] <stgraber> mhall119: we used to be roomate before he had to move back to South Africa and we both work on Edubuntu, so I can see how you got confused ;)
[23:44] <stgraber> mhall119: anyway, merged your branch, now refreshing the package
[23:45] <mhall119> stgraber: oh, must be, I thought it was you though
[23:46] <mhall119> thanks stgraber
[23:46] <mhall119> I honestly can tell you guys apart
[23:47] <stgraber> mhall119: good! I'm a swiss living in Canada, highvoltage is a South Africa who used to live in Canada ;)
[23:47] <stgraber> *south african
[23:48] <mhall119> all you edubuntu guys look the same to me
[23:48] <mhall119> :)
[23:53] <sergiusens> lol
[23:54] <stgraber> mhall119: meta package uploaded
[23:55] <mhall119> thanks again stgraber