/srv/irclogs.ubuntu.com/2013/10/09/#ubuntu-touch.txt

loolracarr: there's one thing I'm missing00:03
loolracarr: could you confirm how to test if current device supports mir?00:04
loolracarr: or do I need to hardcode mako + maguro?00:04
loolI've pinged kgunn on this earlier but no luck00:04
kgunnlool: sorry, thot you got an answer....i would say you'd need to hardcode00:06
kgunnlool: afaik, there's not knowledge of product in mir....only knowledge of hwc versions00:06
kgunnwhich in essence relate to a product00:07
kgunnkdub: ^ any thots ?00:07
loolok, I think I'll test ro.product.model=maguro00:07
loolor mako00:07
kdublool, mir does autodetect the backend it should run on, but i'd imagine you're trying to work around mir not supporting a feature00:08
loolkdub: I'd like to only turn on Mir on devices that supportit00:09
loolnot on grouper00:09
kdubah00:09
kdubwell, yes, i think we'd need a device-specific approach for the forseeable future00:09
racarrlool: Ok.yeah I need to start looking at the unity8 failures00:10
RAOFlool: robert_ancell semi-recently wrote blacklist support for unity-system-compositor; a similar approach should work for unity800:11
* RAOF wonders if this is functionality common enough to be factored into Mir itself.00:11
racarrseems like probably00:13
racarr...definitive language from me there :p00:13
racarrum...hmm00:14
racarryeah seems like something we should have infrastructure for00:14
RAOFIt likely needs to be platform-specific, but that's doable.00:15
mhall119nik90: looks like boiko has been heavily involved in the phone app00:27
=== chriadam|away is now known as chriadam
mhall119nik90: I'm betting you'll need some additional apparmor stuff to get to the ringtone files though00:28
mhall119jdstrand: ^^ can you confirm or deny?00:28
jjohansenmhall119, nik90: grep DENIED /var/log/syslog00:29
jjohansenif you see denied messages then yes00:29
=== anders3408 is now known as anders3408|afk
slangasekxnox: note that I am querying the outcome of 'set-env -g' by using 'list-env -g'; this has nothing to do with individual jobs losing the state01:04
racarrkgunn: lool: Afk for a bit can reach me on phone at g+01:10
loolracarr: ok01:10
loolracarr: I wont have time today I'm afraid01:10
loolwill be tomorrow01:10
racarrno worries :D01:11
racarrso cloooose01:11
jdstrandmhall119: don't know-- would need to see the denials. that said, the dialer shouldn't be confined01:13
jdstrandmhall119: but yeah, what jjohansen said01:13
mhall119jdstrand: the dialer may not be confined, but the clock app presumably will be, which will stop it from accessing the dialer's ringtones01:48
=== Namidairo`bnc is now known as Namidairo
robert_ancellfginther, do you know why https://code.launchpad.net/~robert-ancell/lightdm/handle-no-xrdb/+merge/189971 hasn't merged?02:11
robert_ancellthomi, ^?02:13
* thomi looks02:14
thomiugh. what's the new jenkins IP address?02:15
thomirobert_ancell: http://10.97.0.26:8080/job/lightdm-autolanding/156/console02:19
robert_ancellthomi, uh, that's just blank for me02:21
thominot for me02:22
robert_ancellhmm, refresh gave it content02:23
robert_ancellso it's just running super slow?02:23
thomiI triggered it manually02:23
thomiIt wasn't running before. Not sure why02:23
robert_ancellcheers02:23
thominw02:24
ollido we know if build #88 is done?02:56
AskUbuntuHow can one go about integrating messaging into an Ubuntu Touch App? | http://askubuntu.com/q/35564703:07
nhainesmhall119: certainly the SDK should provide ringtone and notification support?03:11
=== zenvoid_ is now known as zenvoid
=== _salem is now known as salem_
=== salem_ is now known as _salem
=== plars_ is now known as plars
=== chihchun_afk is now known as chihchun
lucenutI've been reading about installing ubuntu-touch on a nexus 10. Kinda confusing.03:36
lucenutDo you think the install will be any easier with the release on the 17th?03:37
lucenutI'm dying to get it running. It's why I bought the nexus 10.03:38
nhaineslucenut: I don't suspect the install process will change.  Which is kind of too bad, since I did consider what a nice install wizard would look like.03:41
RAOFlucenut: Hm. It was pretty easy to install on a N7 and N4; what problems are you having?03:41
lucenutI don't know where to start!03:42
lucenutDo I have to "root" it first?03:42
lucenutIf something goes wrong can I just hold the volume and power buttons and restore jellybean?03:43
lucenutThe installation videos showed you have to install a backup app first?03:43
RAOFlucenut: https://wiki.ubuntu.com/Touch/Install describes the process; it's reasonably straightforward.03:43
lucenutWill this be "dual-boot"?03:45
RobbyFnot dual boot03:46
lucenutThe only ubuntu computer I have is an Xubuntu netbook. Will that work for this install?03:46
RobbyFIt will work for install03:46
lucenutIs dual-boot possible?03:46
RobbyFwhen you restore back to jelly beat then you'll be fresh install03:46
RobbyFdual boot might be possible but WAY more complicated.03:47
RobbyFnot worth it and not supported.03:47
lucenutOK03:47
RobbyFgive it a try03:47
lucenutScary!03:47
lucenut$500 tablet.03:47
nhaineslucenut: not scary.  It's simple to jump back to Android.  :)03:47
lucenutI can always restore android right?03:48
lucenutK03:48
lucenutThanks.03:48
RobbyFyes you can easily restore android with a click of a button03:48
lucenutIf I do it before the 17th, will the developer version update to the real one?03:49
RobbyFmost likely03:49
RobbyFif not it won't be hard to flash that.03:49
lucenutK03:49
nhaineslucenut: in the last week or so, Ubuntu's gotten really nice on phones, too.  Should be fun on the tablet.03:50
lucenutI imagine if they are really releasing on the 17th it should be pretty close to locaked down.03:50
lucenutIs the preview a nightly build?03:51
nhaineslucenut: no, they build 4 times a day and promote "blessed" builds.04:08
=== SuperLag_ is now known as SuperLag
=== SuperLag is now known as Guest95884
guest___________The install guide says " We currently do not support the newer '2013' Nexus 7."  does that mean it won't install at all?05:42
RAOFguest___________: Correct05:44
guest___________How do I verify that I have the new version and not the old version?05:44
RAOFguest___________: I'm not sure; when did you buy it?05:46
guest___________Few weeks ago, so probably the new one huh?05:46
guest___________Kinda bummed I got it just for linux.05:46
RAOFguest___________: You could also just try flashing the Touch image with phablet-flash; it'll detect what device you have connected and bail if it's unsupported.05:46
RAOFYou probably do have the shiny new one, yes.05:47
guest___________yeah I tried that if I give it the -d grouper command it will attempt to flash it and I have to bring it back to factory05:47
nhainesguest___________: on the other hand, it's a *very* nice tablet, and Ubuntu will probably be ported eventually.05:47
guest___________if I don't give it -d  I get ERROR:phablet-flash:Unsupported device, autodetect fails device05:48
guest___________Yeah, I suppose i'll just have to be patient05:48
RAOFYeah; it'll probably be ported fairly soon. It shouldn't be hard to port to.05:48
nhainesguest___________: Google used the same name, but the hardware's completely different.05:48
guest___________Well sounds like I tried to do everything correctly so its out of my hands at this point05:49
RAOFYup.05:49
RAOFWell, not exactly.05:49
RAOFYou could provide the port ☺05:49
guest___________well mines an asus nexus 7 is that the same as google nexus?  Get lost in the branding05:50
guest___________haha i'm not that good of a programmer05:50
guest___________found my box says google nexus think I answered my own question05:51
guest___________alright well thanks again05:51
=== chihchun is now known as chihchun_afk
=== chihchun_afk is now known as chihchun
=== dk is now known as Guest36601
=== jasoncwarner_ is now known as jasoncwarner
=== sporkeee is now known as wilee-nilee
dholbachgood morning07:34
JamesTaitGood morning all; happy World Post Day! :-D08:11
=== iahmad is now known as iahmad|afk
=== fmasi_afk is now known as fmasi
sergiusensmhall119, jdstrand dconf/dbus call was a missing 'calendar' profile08:30
AskUbuntuHow to run Ubuntu touch apps on Ubuntu desktop? | http://askubuntu.com/q/35572808:40
sergiusensnik90, are you around?08:41
nik90sergiusens: yup08:45
sergiusensnik90, https://code.launchpad.net/~sergiusens/ubuntu-clock-app/confinement_fixes/+merge/19004408:46
sergiusensnik90, if you don't mind taking a look08:46
sergiusensnik90, the dconf thing was evolution btw08:46
nik90sergiusens: ah for the alarms08:46
nik90sergiusens: will do08:47
sergiusensty08:49
nik90sergiusens: is com.ubuntu.clock the standard way of naming applications?08:57
sergiusensnik90, yeah, let me give you some docs08:57
=== fmasi is now known as fmasi_afk
nik90sergiusens: I notice weather and calendar following the same convention. But calculator is named ubuntu-calculator-app. I guess i can do a MP for that.08:58
sergiusensnik90, yes please08:59
sergiusensnik90, http://developer.ubuntu.com/publish/apps/security-policy-for-click-packages/08:59
nik90sergiusens: top approved. thnx09:00
Mirvhow can I use larger than 80x24 terminal sanely with adb shell which seems to somehow assume that?09:01
sergiusensMirv, hey, doanac had a shell script that resized the terminal09:01
Mirvsergiusens: sounds promising09:03
Abhishekhi all09:07
AbhishekI want to port the Android source code to create Ubuntu touch images09:08
AbhishekMy development board is having SoC "Qualcomm Snapdragon S4 Pro APQ8064"09:09
dholbachAbhishek, Dmitrjs answered your mail on the mailing list: https://lists.launchpad.net/ubuntu-phone/msg04516.html09:09
AbhishekI am not able to clearly understand09:10
dholbachAbhishek, and it looks like you sent your mail multiple times09:10
dholbachah no, it looked like it was multiple times, but it was different questions09:10
Abhishekhttps://lists.launchpad.net/ubuntu-phone/msg04531.html09:11
AbhishekPlease answer this query09:11
AbhishekI have complete Android source code09:11
dholbachAbhishek, is https://wiki.ubuntu.com/Touch/Porting of any help?09:11
AbhishekI synced the ubuntu touch source code using "phablet-dev-bootstrap"09:12
AbhishekThis is helping somewhat09:12
AbhishekMy device is not supported in CyanogenMod09:14
AbhishekI am having the development board with Android09:14
AbhishekThe source code of this android is available09:14
Mirvtsdgeos: no landings at moment if not fixing tests/bugs/crashes, #ubuntu-ci-eng for more.09:14
tsdgeosMirv: well it does fix a bug and a crash09:15
Mirvtsdgeos: but does it fix a bug and a crash also now that the unity8 change is not in trunk?09:15
tsdgeosMirv: it does fix the bug that the feature is missing in unity8 :D09:15
Abhishekdholbach: I want to use the source code of Android to create ubuntu images09:15
tsdgeosMirv: but we can commit the code again if you prefer :-)09:16
dholbachAbhishek, I'm afraid I'm not much of a help here - I'm not an expert when it comes to porting at all - I just noticed the discussion on the mailing list09:16
Mirvtsdgeos: haha :) anyway it's 'no' at the moment, but feel free oto come to the channel to discuss it. I guess the most interesting thing is that if it currently would happen elsewhere but has not been identified as coming from that.09:16
AbhishekDholbach, can you please let me know who can help me09:17
Mirvtsdgeos: it's most of all now about fixing mir-by-default related issues/crashes09:17
tsdgeosMirv: of course it would happen somewhere else09:17
tsdgeosit's a bug in Qt09:17
tsdgeosit's just that noone uses QML as extensively as us09:17
tsdgeosanyway, if it's no09:17
dholbachAbhishek, the mailing list is probably a good place - this channel is quite high traffic so thing easily get lost here09:18
tsdgeosit's no, i know how this management decisions work09:18
nik90sergiusens: https://code.launchpad.net/~nik90/ubuntu-calculator-app/applicationname-fixes/+merge/19006009:18
SaviqMirv, so we essentially got a freeze two days before the freeze?09:18
nik90sergiusens: fixed it for the calculator app. can you review and check if it is okay09:18
Saviqawesome09:18
MirvSaviq: something along those lines09:19
AbhishekCan someone help me in porting the Android source to create Ubuntu images ...which I can load on my development board ....and see ubuntu running09:19
sergiusensnik90, looks good, let me test this09:19
MirvSaviq: we've been in freeze with 'only via landing asks' for some time already09:19
AbhishekI raised the query in https://lists.launchpad.net/ubuntu-phone/msg04531.html09:19
SaviqMirv, yeah well...09:19
=== iahmad|afk is now known as iahmad
AbhishekPlease answer my query ...I want to port Android source code to create Ubuntu images09:34
AbhishekI raised the query in https://lists.launchpad.net/ubuntu-phone/msg04531.html ....Please answer it09:38
sergiusensnik90, feel free to top approve that one09:48
sergiusensnik90, I'm marked as community in lp, so can't09:48
xnoxAbhishek: please stop sending identical emails, multiple times.09:49
xnoxAbhishek: that's considered rude, and you don't give enough time. Many people only read mailing lists infrequently, e.g. every weekend.09:49
nik90sergiusens: I cannot do it either. I am only admin of clock app :)09:51
Abhishekxnox: I will not do it again ...sry for that09:52
Abhishekxnox: I did it beacuse I am eager to know that it is possible or not ?09:53
xnoxAbhishek: no problem. And from your replies, I can only recommend you to unpack your device boot.img (and the ramdisk from inside it) to actually study and understand how your device boots.09:53
xnoxAbhishek: it must already have a boot loader pre-installed - otherwise it will not boot at all.09:53
xnoxAbhishek: and then you just need to ticker and work out how can you point the ramdisk to e.g. boot alternative partition where you'd have an ubuntu rootfs.09:54
Abhishekxnox: I'll try to do it now09:54
xnoxAbhishek: maybe study ac100 boot images / tarball installer.09:54
xnoxAbhishek: it has it's own ramdisk inside bootimg that points to unpack ubuntu rootfs from a tarball from sd-card onto userdata partition and then boot that.09:55
sergiusensdpm, can you look at https://code.launchpad.net/~nik90/ubuntu-calculator-app/applicationname-fixes/+merge/190060 ?09:55
sergiusensor mihir_ https://code.launchpad.net/~nik90/ubuntu-calculator-app/applicationname-fixes/+merge/19006009:56
dpmsergiusens, still at the conference and about to change rooms, not sure I'll manage until tomorrow :/09:56
davmor2Morning all09:57
=== alan_g is now known as alan_g|tea
dpmsergiusens, if popey is around he can probably help you find someone for the review09:57
mihir_sergiusens: is it regarding change in app name ?09:58
dpmI've added a comment to the MP nevertheless09:58
pittidid anyone actually try and add an ubuntuone account? addition seems to work initially, but it immediately seems to forget it again once I go back in the system settings10:01
seb128pitti, I did some days ago and it worked, let me try today10:06
pittiah, I now tried to not go back, but just end system settings10:07
seb128pitti, wfm but I'm still on yesterday's image10:07
pittiand that seems to work10:07
pittiseb128: me too10:07
seb128"go back"?10:07
pittiseb128: once you added the account, I opened the menu and used "go back" to get back to the accounts list10:07
seb128pitti, when you press the "add" button it should bring you back to the account panel with the list of accounts10:07
seb128e.g no need to do anything10:07
pittiyes, and that worked10:07
pittibut afterwards it forgot the account10:08
pittiI'll reinstall with today's device and try to reproduce this more systematically10:08
seb128ok, but somewhere I guess10:08
=== alan_g|tea is now known as alan_g
seb128pitti, I'm going to keep an eye on it as well10:08
sergiusensmihir_, yes10:13
mihir_sergiusens: I just did testing :) and approving MR :)10:15
sergiusensmihir_, can you top approve as well?10:18
sergiusensmihir_,  thanks by the way10:18
mihir_sergiusens: done !! Sorry i forgot to do that10:19
nik90mihir_: thnx10:19
mihir_nik90: welcome :)10:19
morrowynMorning10:35
loolpitti: Did you test upower on touch?10:39
pittilool: the tests run during package build, so it tested itself on ARM10:42
pittilool: the only effective delta was in the glib client library, I guess we aren't even using that on touch10:42
pittiah, powerd does10:43
loolpitti: the diff is big; I understand one patch was merged upstream, but we'd like a heads up to test such uploads seeded on touch10:43
pittilool: it's not that big; we just replaced the bit "git backports" with a new upstream release, the two effective patches are http://cgit.freedesktop.org/upower/commit/?id=db89e5a3 and http://cgit.freedesktop.org/upower/commit/?id=ecc4e379ccc10:44
pitti(and this API got test coverage in 0.9.22 now)10:44
loolpitti: what are the potential properties hit by this in practice?10:44
pittilool: the actual property values  don't change10:45
pittilool: this just fixes the return value of that method10:45
loolpitti: yes, so I guess it changes the code path for a good reason10:45
pittithat was utterly broken before, so apparently our consumers didn't even bother to check it10:45
pittibut some GNOME consumers rely on it10:46
loolpitti: I'm expecting this got fixed because some functions didn't return what they were meant to return10:46
pittiright10:46
loolpitti: which properties did we witness this on that are going to be reported correctly when they used not to?10:46
pittibut if powerd etc. would actually check it, we would have noticed it ages ago10:46
loolmaybe we will notice a regression now  :-)10:46
loolor an improvement10:46
pittilool: none, we build with --enable-deprecated10:46
pittithe second commit is a no-op for us10:47
loolpitti: I'm speaking of the first one10:47
pittilool: that doesn't change the properties themselves, only the return value of up_client_get_properties_sync()10:47
pittilool: so, if it helps I can upgrade my phone to that upower and check if anything breaks10:48
loolpitti: I understand the properties dont change10:49
loolpitti: sorry, dont want to be a PITA10:49
pittiupgraded, rebooting10:51
loolpitti: the reason I'm bringing this up is that we're trying to check everything going in at this point because we are trying to fix issues in the image; it's super painful to do so, but this specific change either didn't change anything or changes something and then it should be checked on touch  :-)10:51
loolpitti: Thanks for testing10:51
pittiupower --dump unchanged, power indicator works with and without USB cable (i. e. "AC"), suspend works10:52
Laneybattery panel in system-settings uses the glib library10:53
loolpitti: great, thanks for testing10:53
loolpete-woods: Is https://code.launchpad.net/~pete-woods/unity-mir/window-stack-get-property/+merge/189984 just API/ABI addition?10:53
loolpete-woods: Can I land this safely when it's in trunk?10:53
pittilool: that update wasn't really aimed at fixing anything in touch, but we need it to fix desktopy bits (and also drop our huge patches and delta)10:53
loolpete-woods: If yes, I'd like to get that through if it's needed for the other merge10:54
loolpete-woods: (Was this tested somehow?)10:54
loolpitti: If you could just test things that are on touch image on touch before uploading or give us a heads up with the diff, what would be nice10:54
loolpitti: (so little remaining...)10:54
pete-woodslool: yes, it's just addition10:54
pittilool: well, I did consider the effective change and found it safe10:55
pittilool: and I explained the change in #u-release (who reviewed and accepted it)10:55
pete-woodslool: there is no ABI, though, as it's just a DBus interface10:55
loolpitti: we'd like a heads up on the touch side10:55
pittilool: but it's a pain to test it before upload10:55
* lool lunch &10:55
pittilool: the way we set up images make them rather hard to actually build stuff on10:56
pittiso for such rather easy changes I don't usually bother10:56
pitti(and for changes which already have test coverage anyway)10:56
jodhlool: fyi - bug 1235649 is partly caused by unity it seems.10:57
ubot5bug 1235649 in upstart (Ubuntu Saucy) "uevent spam causes session upstart to consume massive amounts of memory on Ubuntu Touch" [Critical,Confirmed] https://launchpad.net/bugs/123564910:57
ogra_jodh, well, i just made a most intresting observation while testing a very old image to verify something else10:59
ogra_jodh, with image 69 my session init process stays at 1.6M constantly, no matter what i do10:59
ogra_(sadly i think due to lool's recent image buold 69 is now gone from system-image.u.c)11:00
jodhogra_: ok. I wonder what version of unity-panel-service 69 was using?11:00
* jodh afk11:00
* popey coughs up http://popey.mooo.com/mirror/system-image.ubuntu.com/saucy/11:00
popeyogra_: ^^11:00
popeyooh, 69 is missing!11:01
popeyoh, i only mirror saucy, not the -proposed images ☹11:01
ogra_popey, right, and system-image only keeps a certain amount11:02
popeyogra_: i run wget every day though11:02
ogra_ah11:03
popeyso i guess 69 was never promoted11:03
ogra_yeah11:03
ogra_jodh, any idea what package that would be in ?11:06
ogra_ah, got it11:08
ogra_jodh, unity-services 7.1.1+13.10.20130927.1-0ubuntu111:08
ogra_jodh, but that job shouldnt be used by touch at all11:09
ogra_start on desktop-start DESKTOP_SESSION=ubuntu11:09
ogra_and it didnt change since11:10
gemaogra_: what is the first image with mir enabled?11:10
ogra_gema, the current one11:10
gema89?11:11
ogra_(in saucy-proposed)11:11
ogra_yes11:11
gemaogra_: ack, thanks11:11
=== fmasi_afk is now known as fmasi
popeyIt's a _lot_ slower.11:12
asacjibel: could you find AP regressions over SF that dont have crashes?11:13
asac -> we should tackle crashes (as those create flakiness) and AP tests that regressed, but dont have crashes.11:13
=== MacSlow is now known as MacSlow|lunch
ryanteckHey11:18
ryanteckMy Nexus 4 (8GB) has been sitting with a black screen at the stage "INFO:phablet-flash:Waiting for install to finish on device. Please do not unplug device until phablet-flash finishes." for about 5 mins11:20
ryanteckanyone know if this is an issue?11:20
popeyryanteck: it can take a little while11:20
ryanteckAhh ok11:21
looljodh: ok11:23
loolSaviq: I know little about unity-panel-service, but it seems to be related to the upstart memory leak we're seeing11:24
loolSaviq: the leak might be in libdbus, but doesn't happen when panel-service isn't used; what is it for?  could it be generating a huge amount of events?  can we disable it temporarily?11:24
ryanteckwhew all flashed11:24
ogra_popey, see if it is also a lot slower if you do --no-backup11:27
Saviqlool, panel-service is a proxy between indicators and unity711:27
ogra_popey, for me Mir is not much slower than SF ... while there is a slight slowdown it is nearly not noticeable11:28
Saviqlool, but that's the extend of my knowledge11:28
Saviqlool, we're not using it on touch11:28
ogra_lool, how could it be, it will never run11:28
popeyogra_: i notice it most on screens which have transparency, like the initial welcome thing11:28
ogra_lool, as i stated above11:28
=== forcev is now known as FunkyPenguin
looljodh: ^11:29
loolSaviq: Ok, so I guess is from jodh's debugging on the desktop11:29
gemaseb128: are you aware of a unity crash in image 89 when bluetooth is disabled and enabled again?11:30
ogra_right, that wont help us on touch though11:30
loolSaviq: he reproduced the upstart leak in an easier to work with environment (desktop image by sending events in a loop)11:30
gemaseb128: I lose the whole UI11:30
Saviqlool, k11:30
loolbut probably not the source of the events then11:30
ogra_unless the issue is caused by *not* using that job11:30
loolwhich seemed to be the vsync stuff from kenrel anyway11:30
dholbachpopey, same here11:30
seb128gema, there is a know but with Mir yes11:30
seb128gema, well, I think it's known, charles and lool were discussing it yesterday11:31
ogra_lool, well, mako doesnt have vsync uevent spam and also exposes it11:31
seb128gema, it's a Mir issue as far as I'm concerned11:31
loolgema: LP #123670511:31
ubot5Launchpad bug 1236705 in unity8 (Ubuntu) "unity8 crashed with SIGABRT in __gnu_cxx::__verbose_terminate_handler(), thrown from mir::DefaultServerConfiguration::the_connector()" [Critical,Confirmed] https://launchpad.net/bugs/123670511:31
gemalool, seb128 thanks!11:31
ogra_lool, *specifically* on 88 ... i had the session init have 30MB RSS occupied after playing for 10min11:31
ogra_on mako11:31
ogra_(starting and stopping apps and scrolling around a bit in the UI)11:32
ogra_so while the vsync stuff on maguro is ugly and adds to teh ram consumption, i dont think thats the main issue11:33
* ogra_ gets coffee11:34
OrokuSakiMorning11:56
OrokuSaki@ogra I have a customized graphics.c for my device.. Do you think that could be causing the EGLBad_Conext?11:56
OrokuSakiI bet that is the cause. =)11:56
=== fmasi is now known as fmasi_afk
OrokuSakiquiet in here this morning11:57
loolplars, doanac or jibel: I think we want to work on a phablet-test-run flag to "stop powerd" as root before startng the unity8 tests; it's not enough for them to pass, but it's required for anything to happen11:57
=== alan_g is now known as alan_g|lunch
=== MacSlow|lunch is now known as MacSlow
OrokuSakimy permissions for kgsl should be system graphics according to android12:13
OrokuSakibut with udev, the apply as system root12:14
OrokuSakichown system graphics kgsl-*12:14
OrokuSakiand when I ls... it is system root12:14
OrokuSakiIs that right?12:14
=== _salem is now known as salem_
jdstrandsergiusens: so, the calendar policy group is a) reserved and b) doesn't actually allow the dconf access-- it denies it. the qtcalendar with eds plugin tries to access dconf for some eds settings that aren't actually needed by the calendar12:30
=== fmasi_afk is now known as fmasi
xnoxogra_: i am confused by all the ways things get mounted.12:54
jdstrandmardy: hey, was curious about bug #1232097. is that going to be fixed for 13.10?12:54
ubot5bug 1232097 in libaccounts-glib (Ubuntu Saucy) "accounts-qml-module requires read/write access to accounts.db" [High,In progress] https://launchpad.net/bugs/123209712:54
xnoxogra_: is there separate DATA and USERDATA partitions in android?12:54
ogra_xnox, not that i know of12:55
=== chihchun is now known as chihchun_afk
xnoxogra_: cause my container is complaining about /data not available and/or not writtable.12:55
ogra_xnox, i think stgraber introduced /userdata with the system images12:55
xnoxogra_: ok.12:55
ogra_xnox, that should come from /etc/fstab for a normal flipped image12:55
ogra_system-imagwe works completely different here12:56
mardyjdstrand: sure, I've already made the fix upstream and I just need a MP to be approved12:57
xnoxogra_: stgraber: is there any way to debug lxc container's initramfs boot? for some reason it claims "to not find /sbin/adbd" when it is clearly present in the android-ramdisk.img12:58
loolpopey: what's the way to enter the sim pin again?12:59
popeylool: dunno, never enabled that12:59
loolpopey: you have no sim pin on your sim?12:59
ogra_xnox, thats wanted :)12:59
looldoes someone know how to unlock12:59
ogra_xnox, ignore this message ... we rm /sbin/adbd before firing up the container13:00
xnoxogra_: !!!!!! ARGH!13:00
xnoxogra_: where?13:00
ogra_xnox, in the lxc startup scripts13:00
xnoxogra_: thanks a bunch.13:01
ogra_xnox, since it breaks the world if you try to run two adbd's13:01
ogra_xnox, /var/lib/lxc/android/pre-start.d/13:01
xnoxogra_: well, I have none at the moment ;-)13:01
ogra_and you have no USB device anyway13:01
ogra_so why do you care :)13:02
xnoxbecause adb from the lxc-container should be able to us QEMUD instead USB and give me adbd into the container.13:02
ogra_if you want to debug the container use the logs :) .... /system/bin/logcat -d ...13:02
xnoxonce i get that to work, i can transfer qemud/adbd emulator jobs to ubuntu13:03
xnoxogra_: somehow my /system/ gets mounted empty =(13:03
ogra_xnox, do you have it in fstab ?13:03
ogra_for flipped, just add a line for it13:03
popeylool: no13:03
ogra_same for /data13:03
xnoxogra_: it does get mounted, it somehow has nothing in it though =/13:04
xnoxogra_: and kernel complains a bit.13:04
ogra_werid13:04
xnoxogra_: i'll try to produce two system.img files one as ext4 -> for embedding for the lxc container13:04
xnoxogra_: and one in yaffs2 for direct mounting.13:04
=== vila is now known as vila-laaaate-lun
ogra_just make sure they dont hang around in / :)13:05
ogra_or /userdata rather13:05
=== chihchun_afk is now known as chihchun
danielholmhey, how do I install the docviewer on latest images? is there a click package to be found? feeling lost13:08
awafaaerm, is one supposed to flash cdimage-touch before flashing ubuntu-system?13:09
ogra_popey, or mhall119 might know13:09
ogra_awafaa, nope13:09
awafaaogra_: interesting, ok i'll try and work out why i cant flash system image then13:10
ogra_the installed android should be fine13:10
popeyawafaa: what's on it currently?13:10
jdstrandmardy: awesome, thanks! was asking because right now policy still blocks the 'w'13:10
popeydanielholm: which image?13:10
awafaapopey: android, i'll try with --no-backup13:10
danielholmpopey: ubuntu-sytem13:10
popeyogra_: i have seen others report that flashing from android -> ubuntu-system isn't reliable, and they have gone android -> cdimage-touch -> ubuntu-system to make it work13:11
danielholmI cant remember who it was that I talked with about installind the doc-vieweer13:11
popeydanielholm: i think it hasn't been converted to a click package yet.. sergiusens will know13:11
awafaapopey: looks like with --no-backup does the trick13:11
popeycoolio13:11
ogra_popey, it once was like that ...13:11
popeyah13:12
ogra_popey, but sergiusens didnt sleep :) he fixed all issues13:12
nik90charles: ping13:12
danielholmpopey: ok. branch the trunk and create my own from sdk?13:12
popeyin the dim and distant past...13:12
popeyof last month13:12
ogra_hehe13:12
popeydanielholm: that'll work13:12
ogra_hmm13:12
ogra_my maguro looks pretty weird on the apps lense13:12
awafaalooks like phablet-flash looks for a specific data dir to abckup before actually doing anything13:12
=== Namidairo is now known as Namidairo`bnc
ogra_i only have 12 preinstalled apps !13:13
popey--no-backup makes no sense the first time you run the flashing process tbh13:13
charlesnik90: pong13:13
ogra_and no click stuff at all13:13
danielholmpopey: lovley. but it should be in the image in the future, right? it's been away for a while13:13
ogra_popey, ^^^ is that ok on your 89 install ?13:13
popeyogra_: you sure? /usr/share/click/preinstalled ?13:13
nik90charles: regarding the argument that you send to the clock app, would you be sending it in the format qmlscene ubuntu-clock-app --alarm?13:13
popeyogra_: i have lots installed13:13
=== vila-laaaate-lun is now known as vila-late-lunch
ogra_root@ubuntu-phablet:/# ls /usr/share/click/preinstalled/|wc -l13:13
ogra_1513:14
charlesnik90: the plan was that indicator-datetime wouldn't have any intrinsic knowledge of what was being passed:13:14
ogra_popey, my apps lens is pretty empty13:14
popeyogra_: lock/unlock13:14
charlesnik90: it would take the URL that was attached to that ECalComponenet, and pass it to url_dispatcher()13:14
popeymine looked empty for a moment earlier13:14
nik90charles: oh yeah13:14
ogra_and searching for anything doesnt bring up anything from the store13:14
nik90charles: so url_dispatcher() has support for alarm:/// implemented by ted13:14
popeymine didnt show stuff from store, does now13:14
popeybut its slow13:14
ogra_lock unlock doiesnt change it13:14
popeywonder if it's a backend thing, icons not loading13:15
nik90charles: however will url_dispatcher() open the clock app?13:15
charlesnik90: so whatever arguments you pass into that url, you should get back13:15
popeytaking an age13:15
nik90charles: scratch my last question13:15
ogra_we dont cache them i think13:15
charlesnik90: yes, ted's got url_dispatcher's patch merged13:15
* ogra_ reboots to see if that changes anything 13:15
nik90charles: this is regarding the list of alarm in the indicator-datetime13:15
charlesnik90: I'm testing the datetime alarm dispathcer right now13:15
nik90charles: so when the user clicks on an alarm manually what happens?13:15
nik90charles: I thought that the clock app was supposed to be open with the alarm tab?13:15
charlesnik90: in the meeting, we discussed having clicking the menuitem also dispatching the same url13:16
charlesnik90: will that do what you need?13:16
nik90charles: yes I am just trying to figure what to do in my end to receive the url sent by the url_dispatcher13:17
charlesok13:17
charlestedg: ^13:17
=== alan_g|lunch is now known as alan_g
nik90tedg: so I saw the MP regarding the support for alarm:///13:17
nik90tedg: so does this call the clock app?13:17
tedgnik90, It should13:18
nik90tedg: okay, but how does the clock app know that it is been called by the url_dispatcher() to do something special like showing the alarms tab?13:18
tedgnik90, There is a bit of transition, so until the branch lands in the image it'll only call the clock app if it's a .deb not a .click13:18
awafaathere needs to be a splash screen or something on booting - i thought things were borked as i had a black screen for a while13:19
tedgnik90, Assuming it got a "%u" on its Exec line in its desktop file that %u will be replaces with the "alarm:///" URL13:19
popeyogra_: with mir, if I have a scope open then lock the screen, then unlock, i briefly see the scope before I see the welcome screen13:19
* nik90 is checkign the desktop file13:19
popeyogra_: looks like it's just dog slow.13:20
nik90tedg: it is right now only Exec=qmlscene /usr/share/ubuntu-clock-app/ubuntu-clock-app.qml13:20
ogra_popey, i always had that13:20
ogra_even with SF13:20
nik90tedg: I am assuming it should Exec=qmlscene %u /usr/share/ubuntu-clock-app/ubuntu-clock-app.qml13:20
ogra_if i switch on the screen i always see the lock screen slide in13:21
tedgnik90, I'm not honestly sure where that goes, I can only say that it'll get replaced :-)13:21
danielholmpopey: cant seemt to get the doc viewer to work. now lets see.. ping sergiusens13:21
* ogra_ has never seen it behave any different 13:21
tedgnik90, I'm guessing after the qml file, but I don't know.13:21
popeydanielholm: what's not working?13:22
tedgnik90, Also there's an internal API that loicm added to give you a signal if you're already running and get a URL.13:22
danielholmpopey: installed on device its just a white screen. trying to run from sdk to get output shows: org.docviewer.file not installed, and its read only, so stuck :P13:22
nik90tedg: I am looking at calendar app which had to add the same argument support13:23
asacogra_: can you reproduce those AP crashes on mako?13:26
asacogra_: can you see if that goes away if we kill the hud from the image?13:26
mhall119danielholm: docviewer isn't installed until we can decide on the proper way to call it13:29
danielholmmhall119: aha, ok. I thought it might be something like that13:29
danielholmmhall119: hope it gets there soon so I can read the articles for school on my tab :P13:30
danielholmso another question: is there a way of getting all the names of available icons from theme? need to change them for the music app13:30
danielholmand I suppose this is the right one to use: http://developer.ubuntu.com/api/ubuntu-13.04/qml/ui-toolkit/qml-ubuntu-components0-icon.html13:31
ogra_asac, which ones ?13:31
asacogra_: the hud13:31
asacall :)13:31
asacall that doesnt work13:31
ogra_asac, what doesnt work and what should i test ?13:32
danielholmnot work?13:32
asacogra_: we get all those AP failures with crashes13:32
asacogra_: check if you can remove the hud and see if the APs then start succeeding13:32
RedPandaFoxWhen attempting to sync contacts from gmail to Ubuntu Touch using Gooogle Contacts importer (or via syncevolution via term) I get an error "Sync result Sync error:" is this a bug or just not supported with two step yet? (I use two step authentication)13:33
ogra_asac, you cant "remove" the hud13:34
=== vila-late-lunch is now known as vila
ogra_asac, i can try to link the binary to 7bin7ture or some such, but i'm busy with a bunch of other things, this will have to wait until i have a phone free13:34
ogra_asac, also why shouldnt i try the fixed hud instead ?13:36
stgraberxnox: it's pretty hard to figure out what android does... some stuff get logged into the kernel ring buffer, the rest you have to figure out based on ps output and the state of the filesystem in the container13:39
asacogra_: you can try both13:40
asacwith fixed hud and without13:40
asac:)13:40
asacanyway, lets tryu the fixed first13:40
asachehe13:40
sergiusensdanielholm, mhall119 did you grab the click from jenkins?13:45
danielholmsergiusens: I did not. Where is it and how do I install it?13:45
sergiusensdanielholm, pkcon install-local package.click ?13:46
danielholmsergiusens: thank you. and link to the actual package?13:46
danielholmI have never installed a click package from cli13:46
sergiusensdanielholm,  https://jenkins.qa.ubuntu.com/job/docviewer-app-click/13:46
danielholmlovely, thank you13:46
danielholmsergiusens: but it is still not that usable on device?13:48
danielholmsine the calling is not done, right?13:48
sergiusensdanielholm, what calling?13:50
sergiusensdanielholm, the uri handler stuff you mean?13:50
dobeymhr3: can you re-review https://code.launchpad.net/~dobey/unity-scope-click/finite-search/+merge/189963 please?13:51
danielholmsergiusens: mhall119 said something about the app is note done since you dont know how to call it, from which I took it like you dont know how to open a file from the file manager?13:51
dobeydanielholm: perhaps he meant "how to call it" as in "what to name it"13:52
mhall119no, I meant how to get from the filemanager to the docviewer13:52
sergiusensdanielholm, yeah, well that needs to be defined with tedg13:52
danielholmsergiusens: ok13:53
danielholmI just wanted to read some pdfs later when I'm out :p13:53
=== fmasi is now known as fmasi_afk
mhall119talking to tedg and pmcgowan, they decided it shouldn't be in urldispatcher, it should be in contenthub, but after talking to kenvandine the contenthub isn't likely to have that capability by 13.10's release13:53
mhr3dobey, install and uninstall adds/removes local app and that should be shown in home, so in those two instances invalidating both was correct13:53
danielholmsergiusens: try again later again13:53
danielholmok13:53
sergiusensballoons, hey, who can help out with filemanager tests for the click app?13:56
dobeymhr3: sigh. and i can't test that at all, because i can't close the keyboard when searching on home, and swiping to a different scope view resets the search. but trying to scroll what little bit of the scope remains above the keyboard, doesn't show the installed apps in the home view. :(13:57
mhr3dobey, yea the osk is weird.. especially with mir13:57
mhr3dobey, btw the flickering is very visible with 8813:58
dobeyit's not exactly flickering, but yes, i see what the bug is. and i have image 89 installed13:59
mhr3dobey, then you have mir and that's why osk is so terrible13:59
mhr3dobey, and one more tiny comment14:00
mhr3in the mp14:00
=== dandrader is now known as dandrader|afk
dobeysearching in home doesn't even show any click apps. it only shows the ones which appear on the home page by default14:01
mhr3and that's a bug we'll have to fix14:01
alex-abreumzanetti, ping14:05
dobeymhr3: pushed14:06
=== fmasi_afk is now known as fmasi
ogra_Saviq, do you know an answer to the last question ?  https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/1235649/comments/4214:09
ubot5Ubuntu bug 1235649 in upstart (Ubuntu Saucy) "uevent spam causes session upstart to consume massive amounts of memory on Ubuntu Touch" [Critical,Confirmed]14:09
=== barry` is now known as barry
Saviqogra_, https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/1235649/comments/4414:15
ubot5Ubuntu bug 1235649 in upstart (Ubuntu Saucy) "uevent spam causes session upstart to consume massive amounts of memory on Ubuntu Touch" [Critical,Confirmed]14:15
Saviqogra_, can be applied straight on device, in /usr/share/unity8/14:16
ogra_Saviq, awesome !14:17
ogra_thanks14:17
nik90tedg: can you review https://code.launchpad.net/~rpadovani/ubuntu-clock-app/09Oct2013/+merge/190139 and see if that's okay from the url_dispatcher() point of view14:20
* tedg clicks14:22
tedgnik90, Do you need to add the %u to the desktop file as well?14:23
nik90tedg: I believe not..only required if it is a default argument14:24
nik90as in the argument is obligatory14:24
nik90which the alarm argument is not14:24
tedgnik90, No, it's telling the launcher where to put the URL if it has one.  Otherwise it doesn't know.14:25
ogra_pete-woods,14:25
ogra_root@ubuntu-phablet:/# ls /var/crash/14:25
ogra__usr_bin_maliit-server.32011.crash14:25
ogra__usr_lib_arm-linux-gnueabihf_hud_hud-service.32011.crash14:25
ogra_pete-woods, tedg, thats with the new hud package installed14:25
ogra_a fresh boot that had a wiped /var7crash14:25
tedgogra_, By new you mean which version?14:26
ogra_tedg, the one that just enters the archive14:26
nik90tedg: this was implemented by WebbyIT who also did it for the calendar app in a similar fashion14:26
nik90WebbyIT: ^^14:26
nik90WebbyIT: Can you confirm the above14:26
ogra_tedg, https://launchpad.net/ubuntu/saucy/+source/hud/13.10.1+13.10.20131009-0ubuntu114:26
ogra_doesnt seem like it helped :/14:26
ogra_i get the hud icon in the browser, but no actions14:27
tedgogra_, Hmm, so with the updated Upstart job.14:27
ogra_(indeed no hud-service)14:27
ogra_tedg, exactly14:27
jibelalecu, on mako with build #89 bug 123743314:27
ubot5bug 1237433 in unity-scope-click (Ubuntu) "[regression] Click apps in 'More suggestion' doesn't have icons" [High,New] https://launchpad.net/bugs/123743314:27
tedgogra_, Can you upload that crash file?14:27
jibelalecu, please don't tell me that I must file a bug for each individual app ;)14:28
ogra_tedg, http://people.canonical.com/~ogra/ubuntu-touch/_usr_lib_arm-linux-gnueabihf_hud_hud-service.32011.crash14:28
tedgogra_, Thanks!14:28
=== alan_g is now known as alan_g|tea
tedgogra_, I think that's a different crash, I think that one is because you're using Mir and we're still linking to platform API.14:31
ogra_tedg, we'Re on Mir by default14:31
tedgogra_, I realize, and pete-woods is working quite literally day and night to fix that.14:32
pete-woodstedg: we're not linking to the platform API in HUD any more14:32
popeyogra_: just noticed the same issue of apps not showing in the apps lens, then collapse then open again and they show14:32
ogra_popey, they dont for me once it is in that state14:33
popeyhmm, odd14:33
tedgpete-woods, In 20131009 ?14:33
pete-woodsohh14:33
pete-woodstedg: sorry, stupid brain fail14:33
pete-woodstedg, ogra_: I refer you to https://bugs.launchpad.net/unity-mir/+bug/123399214:34
ubot5Ubuntu bug 1233992 in hud (Ubuntu) "HUD does not support unity8's ApplicationManger" [High,In progress]14:34
ogra_pete-woods, tedg, erm, you are aware that all the testing we did in the last days only happened on Mir ?14:35
ogra_since SF is dead14:35
sergiusenspopey, do you know who's writing tests for the filemanager?14:35
pete-woodsogra_: yes, I have been pinged about once every 30 seconds for the last 3 days about this14:36
popeysergiusens: we missing some?14:36
sergiusenspopey, all of them fail when testing them as click; and the setup logic is really big14:36
popeyugh14:36
popeydefault position is ping balloons14:36
sergiusenspopey, terminal is fine btw :-)14:37
ogra_pete-woods, ok, but why would you expect it to work ?14:37
sergiusenspopey, and I'm making some cleanups in music app14:37
ogra_pete-woods, if it sint even linked against the right stuff14:37
ogra_*isnt14:37
pete-woodsogra_: I only got assigned this task on Friday, I'm not even supposed to have anything to do with HUD any more14:37
popeysergiusens: cool, we have a logo added to weather app, and m-b-o is bumping to "1.0" for release, considering it "done"14:37
WebbyITtedg: arguments work also without %u in Exec, but if you say that is mandatory I add it14:37
dbarthlool: are you open to consider landing this: https://bugs.launchpad.net/unity8/+bug/122955114:38
ubot5Ubuntu bug 1229551 in unity8 (Ubuntu) "Facebook launcher doesn't work on Home scope" [Medium,In progress]14:38
sergiusenspopey, tell him to mark the 1.0 in manifest.json as well14:38
danielholmanyone who knows what the icon for expand/collapse is named?14:38
popeysergiusens: he has.. https://code.launchpad.net/~martin-borho/ubuntu-weather-app/v1.0/+merge/19014014:38
looldbarth: looking14:42
alecujibel: wow, that's a huge bug14:43
alecujibel: yup, no need to fill one for each app..... :-(14:43
=== alan_g|tea is now known as alan_g
looldbarth: Yes, but please stage it for after Mir work is done (tomorrow I guess)14:43
looldbarth: add an ask14:43
looldbarth: this is high priority, but top priority is fixing mir specific regressions right now14:43
dbarthok14:45
dobeyalecu, jibel: that's probabl a regression in unity, since the scope doesn't download those images, but simply passes the URL to the dash, no?14:45
gemalarsu: ping14:50
jibeljodh, the patch removes the panel right?14:54
mandelbarry, ping14:55
barrymandel: pong14:56
m-b-opopey: you can have a look now at the merge proposal14:57
popeyok14:57
mandelbarry, hello, I'm fixing bug #1233435 ATM I'm pausing the downloads when there is a connectivity issue (we are switching connections, lost connection etc..) and emitting the pause signal, once the connection is back I resume the downloads and emit the resumed signals, do you see an issue with that approach??14:57
ubot5bug 1233435 in ubuntu-download-manager "u-d-m downloads stall if the network configuration changes" [Critical,Triaged] https://launchpad.net/bugs/123343514:57
mandelbarry, if the user tries to manually resume it, I'll ignore that until we are back online14:58
popeydone14:58
m-b-opopey: thanks14:59
barrymandel: currently, i don't do anything with pause and resume signals.  the worst that can happen is that i'll timeout if the download doesn't finish in 10m15:02
mandelbarry, hm, how bad is that?? I could happen if the user has a network issue15:02
barrymandel: they'll just have to start again15:03
barrynot terrible i think15:03
mandelbarry, ok, then I think is a valid case, although it would be nice to fwd that state to the ui15:03
jodhjibel: looks like it to me. Saviq ?15:03
mandelbarry, so that if the user is wondering why is taking that look he can check if the download is paused15:03
Saviqjodh, jibel removes the indicators, yes15:04
barrymandel: yeah15:05
danielholmwhere can I find the icon project for ubuntu touch?15:05
danielholmoh wait15:06
mandelbarry, should I add a bug for that?15:07
barrymandel: what would the bug be?15:07
mandelballoons, paused and resumed signals from u-d-m should be fwd to the ui?15:07
barrymandel: yes, i think so.  it's related to LP: #1237360 but from the other direction i think15:09
ubot5Launchpad bug 1237360 in Ubuntu system image "Can't pause the download" [High,Triaged] https://launchpad.net/bugs/123736015:09
mandelbarry, correct15:09
jibelSaviq, jodh it doesn't make much difference, init uses 23MB after 40 start/stop cycles of dialer http://ubuntuone.com/66RSNScXhfbyMaFotBavth15:10
mandelbarry, I wonder, just wonder... but wouldn't it be nicer (guessing here) that the ui has way to ask image update for the object path of the download and just connect the ui to that object path signals?15:10
barrymandel: i talked with didrocks about that way back, and he said he'd rather just connect to s-i15:11
mandelbarry, weird.. I guess we can convince him the other way as soon as we have the download manager client lib15:11
mandelbarry, I see no benefit in fwd signals and methods other than abstracting the use of u-d-m15:12
barrymandel: i would generally agree.  it certainly doesn't make any sense to duplicate all of udm's api in si15:12
mandelbarry, increases lines of code, bugs and work to be done... will talk with him about it15:13
mandelbarry, by the way I fixed the issue with system settings and the dict array yet I have issues running it on the desktop, any idea if that is possible?15:14
jodhjibel, Saviq: are we *sure* that fix stops libIndicatorsQml.so from loading?15:14
barrymandel: you mean, it is fixed in system settings but it doesn't work with the current s-i?  that makes sense, since i disabled that.  it would be easy to re-enable it  (LP: #1215586)15:15
ubot5Launchpad bug 1215586 in ubuntu-system-settings (Ubuntu) "The u/i cannot handle the array-of-dictionaries update descriptions" [High,Triaged] https://launchpad.net/bugs/121558615:15
Saviqjodh, pretty sure, yes, you can remove it if you like15:15
barrymandel: i was mostly waiting for the ubuntu-system-settings bug task on that to get fixed released15:15
barry(or some coordination)15:15
jodhjibel: could you try that? (I don't have the appropriate h/w).15:15
barrymandel: it's just uncommenting some code :)15:15
mandelbarry, that I knew, but I can't seem to be able to run system-settings when I compile it locally, it complains about a shared lib yet I have used make install15:16
mandelbarry, I wonder if you have seen something similar15:16
mandelbarry, if you tell me what that dict those I will also do the changes to update the ui15:16
cyphermoxrsalveti: ogra_: I think I figured out why brcm-patchram is broke on nakasi... seems to be firing twice. Once for android, and once for touch15:16
barrymandel: i haven't15:16
barrymandel: but i haven't really tried to do that either15:16
mandelbarry, ok15:17
ogra_cyphermox, so kill the android one :)15:17
cyphermoxogra_: yeah, just trying to find out where it gets started15:17
barrymandel: i could pretty easily put together a branch that fixes 1215586, then you could install that to test the whole thing together15:17
mandelbarry, that would be awesome15:17
alecudobey, jibel: the icon issue seems to be a change in the servers: the icon_url is no longer being returned in the json.15:17
alecudobey, jibel: I'm discussing this right now with the server guys15:18
barrymandel: cool.  i need to reboot, but will get back on irc and ping you when it's ready15:18
ogra_slangasek, i think bug 1235649 isnt complete, while xnox' fix hides the spam from the dbus, udevd still picks it up which results in udevd consuming ~5% CPU constantly while the screen is on15:18
ubot5bug 1235649 in upstart (Ubuntu Saucy) "uevent spam causes session upstart to consume massive amounts of memory on Ubuntu Touch" [Critical,Confirmed] https://launchpad.net/bugs/123564915:18
cyphermoxogra_: any idea where the init scripts for the android side are located? I can only find init.goldfish ?15:18
mandelbarry, thx15:18
* barry -> brb15:18
dobeyalecu: oh. weird!15:18
cyphermoxfound it15:19
ogra_cyphermox, /var/lib/lxc/android/rootfs ... put a snippet to mangle init.rc into /var/lib/lxc/android/pre-start.d15:19
charlesjfunk: ping15:24
jibeljodh, Saviq I removed libIndicatorsQml.so, but not real improvement. 15MB after 35 cycles15:24
jibelthen the device becomes too slow to continue and I have to reboot15:24
charlesjfunk, could you take a look at https://bugs.launchpad.net/ubuntu/+source/indicator-bluetooth/+bug/1226903/comments/2 when you get time15:25
ubot5Ubuntu bug 1226903 in indicator-bluetooth (Ubuntu) "remove bluetooth indicator for 13.10" [Critical,Incomplete]15:25
jfunkcharles, sure thing15:26
charlesjfunk, cool15:27
Saviqjibel, right, so that rules out that library then...15:27
jfunkcharles, that was a misunderstanding and is quite old, sorry for the confusion, put it to invalid15:27
charlesjfunk: :) will do15:27
jfunkI've done that15:27
jfunkcharles, ^^ no need to do anything15:27
charlesok15:28
jdstrandkgunn: hey, do you know if /system/bin/screencap is supposed to work on mir?15:32
ogra_jdstrand, nope15:33
jdstrandogra_: is there an alternative?15:34
ogra_jdstrand, i dont think there is yet15:34
ogra_use a phone to take a pic ;)15:34
jdstrandheh15:36
jackcyI there anyone who can assist me to configure syncevolution for google contacts?15:39
spindleyjackcy: http://sergiusens.github.io/posts/google-contacts-on-ubuntu-touch.html15:40
jibelfb2png15:41
jibeloops, wrong window15:41
wesnothI am looking for help in configuring screen resolution and rotation. Can anybody help me?15:45
jibeljdstrand, if you can get something with cat /dev/fb0 > dump.raw then convert with ffmpeg15:46
jibeljdstrand, colors might be a bit off though15:46
=== fmasi is now known as fmasi_afk
mamenyakahello! can someone help me with backporting apparmor?15:49
PrimeMinister91Does the CM base for an Ubuntu Touch port have to be 10.1 or can it use 10.2?15:49
=== fmasi_afk is now known as fmasi
jibeljdstrand, cat /dev/fb0 > mir.fb then on another machine ffmpeg -vcodec rawvideo -f rawvideo -pix_fmt rgb32 -s 768x1280 -i mir.fb -f image2 -vcodec png screenshot.png15:51
jibelplaying with options of ffmpeg might help to correct the colors15:52
jackcyspindley that link worked fine with a few modifications - thank you.15:52
skinuxWhat exactly is ubuntu-touch? A complex app, ROM, or what?15:53
jodhjibel: ok, thanks for testing.15:53
=== dandrader is now known as dandrader|lunch
jjohansenmamenyaka: https://wiki.ubuntu.com/SecurityTeam/AppArmorForPhabletKernels15:54
jjohansenmamenyaka: though the zinc tree got corrupted and I am in the middle of rebuilding the backport kernel branches15:54
jjohansenso they won't be there for an hour or two15:54
mamenyakaI checked out the mako kernel15:55
mamenyakafound the patch15:55
=== tkamppeter_ is now known as tkamppeter
mamenyakabut I am having difficulties with the rebase15:55
mamenyakajjohansen, shouldn't the patch appear in the log after the rebase if I check out my kernel?15:57
jjohansenmamenyaka: the mako kernel does have the patches, however they are not as easy to get at15:59
=== fmasi is now known as fmasi_afk
=== gatox is now known as gatox_lunch
jjohansenmamenyaka: git log security/apparmor15:59
jjohansenwill show the the patch set15:59
jjohansenmamenyaka: what won't get shown is the individual backport patches that have been folded in to the main patch for mako16:00
mamenyakajjohansen, isn;t this the one? b89e510 UBUNTU: SAUCE: (no-up) apparmor: Sync to apparmor 3 - alpha 4 snapshot16:01
jjohansenmamenyaka: for the reference backport kernels we have a presquash tree which breaks out the main patch from the individual patches used to get a backport16:01
jjohansenmamenyaka: yes, its the main patch + backport patches to bring it back to the 3.4 kernel16:02
jjohansenmamenyaka: which kernel version are you porting to16:02
mamenyaka3.416:02
mamenyakajjohansen, 3.4.016:03
jjohansenmamenyaka: heh okay, well mako or manta is a good starting point then16:04
mamenyakajjohansen, how can I get the presquash branches?16:04
jdstrandjibel: ah, thanks :)16:05
skinuxI did read online about "Ubuntu for Android". However, I still didn't understand what exactly it is. I'm guessing It's Canonical's answer to Android??16:05
popeyskinux: ubuntu for android has not been released16:05
jjohansenmamenyaka: it will be on the tree on zinc, as soon as I finish cleaning it up, like I said we had corruption in the tree, I've only managed to push a few branches up yet16:06
mamenyakajjohansen, ah okay16:06
jjohansenmamenyaka: how I usually do a port, is to checkout a kernel, rebase to the base apparmor3 patch, and just copy the apparmor directory over. It avoids the merge conflicts, and then I will apply the patches ontop16:07
jjohansenmamenyaka: depending on what you are doing you may want to just copy the kernel over whole sale16:07
mamenyakajjohansen, so when I copied the security/apparmor dir from the (no-up) apparmor: Sync to apparmor 3 checkout, is that it?16:08
mhall119sergiusens: when will we be able to deliver core apps updates via the click store?16:09
jjohansenmamenyaka: you need to make sure you get any other patches that are on top of it.  But if you checkout mako and just copy over the apparmor directory that should get you everything, except the patch history16:10
jjohansenlike I said, I copy from the base sync patch and then apply the others to it16:10
=== kInOzAwA_ is now known as kInOzAwA
mamenyakajjohansen, I see, I wasn't sure that the patches like "apparmor: fix memleak of replacedby struct" were meant for my kernel too16:11
jjohansenmamenyaka: the apparmor backport has been done is such a way that it only touches the apparmor directory16:11
=== kInOzAwA is now known as Guest61597
jjohansenmamenyaka: yep, give me a few more minutes and I will push the clean 3.4 branch16:11
mamenyakajjohansen, yeah, I read it in the guide, it's really great to have it this way16:12
jjohansenmamenyaka: its the same version of apparmor on all the kernels, and then backport patches are applied to make it work with that kernel version16:12
mamenyakajjohansen, now I understand, I also have a 3.0 kernel to "backport" apparmor to16:12
mamenyakajjohansen, any click app included in ubuntu touch?16:14
jjohansenmamenyaka: sure, we have that too. I have clean 3.4, 3.1, 3.0 in my tree that where the basis of the mako/manta, grouper, and maguro backports. The presquash trees have each patch required for the backport broken out16:14
jjohansenmamenyaka: click app?16:15
mamenyakajjohansen, click package16:15
mamenyakajjohansen, sorry, I think I am mixing up some terminology16:15
mamenyakajjohansen, in the testing apparmor part: launching the click application works16:16
jdstrandmamenyaka: twitter is a reasonable test16:17
cyphermoxogra_: there's still an underlying problem in bluetooth on grouper even if I fix the current issue16:17
=== kInOzAwA_ is now known as kInOzAwA
jjohansenmamenyaka: okay there should be a v3.4-backport-of-apparmor3 branch on that tree now16:18
ogra_cyphermox, well, not eating the CPU is still better than nothing16:18
cyphermoxyup16:18
ogra_cyphermox, if you could fix that one :)16:18
cyphermoxwhat package do I ship a fix in?16:18
ogra_whats the brcm_patchram stuff in ?16:19
cyphermoxI'm also wondering if the android side isn't an artifact of how we fixed bluetooth initially before flipped16:19
cyphermoxogra_: oh, you'd ship the pre-start.d file in there as well?16:19
cyphermoxsure16:19
jjohansenmamenyaka: iirc mako is the same with the backport patches squashed into the base patch, some of the other backports are exactly the same. eg. maguro differs from the 3.0 backport because maguro has some patches applied to the kernel that required some changes to the backport patch set16:19
ogra_cyphermox, well, else lxc-android-config, but make sure its only used on grouper16:19
cyphermoxalthough that might break maguro... going to need to test it some more16:20
cyphermoxogra_: either way it needs to be only grouper16:20
jjohansenthat is why we keep a clean backport and the specific branches16:20
ogra_cyphermox, right, just use getprop to get the device before doing anything in your script16:20
cyphermoxyucl16:20
cyphermoxogra_: IIRC I did request a change to the android rootfs before, so perhaps that just needs to be reverted16:21
mamenyakahow can I enable adb?16:21
MacSlowmterry, regarding your branch... is there a recommended/needed order to do that run-time dependencies?16:21
mterryMacSlow, for building the other branches?16:22
mterryMacSlow, you need upstart-app-launch, then unity-mir, then the rest I think16:22
mterryshould have noted that in MR16:23
mamenyakajjohansen, got the v3.4-backport-of-apparmor3 branch16:23
slangasekogra_: udevd seeing kernel uevent spam is hardly a udev bug...16:23
ogra_slangasek, well, xnox said that Keybuk seems to have planned some netlink filter functionallity for such cases16:24
ogra_to supppress such noise if needed16:24
slangasekfiltering it where?16:25
MacSlowmterry, you listed all the required branches... but not mentioned anything about the order so I'm assuming top-to-bottom16:25
mterryMacSlow, I will reorder  :)16:25
MacSlowmterry, thx16:25
ogra_slangasek, as i understood on the netlink socket16:25
mterryMacSlow, done16:25
ogra_slangasek, ask xnox for details, he dug up some docs about that (i dont have the link anymore)16:26
Laneyhow do I get the current image number?16:32
ogra_system-image-cli -i16:32
ogra_(on the device)16:32
wesnothIs there any way to rotate screen and touch together?16:32
Laneyogra_: any API way?16:33
mamenyakaogra_, how do I enable adb?16:33
ogra_Laney, there is a file that holds the value16:33
Laneyalso, that requires root?16:33
ogra_Laney,  grep build_number /etc/system-image/channel.ini16:34
LaneyI see16:36
xnoxslangasek: right, but udev can choose not to emit those events from kernel on the "udev" source?16:43
jjohansenmamenyaka: hold on I missed a patch in the v3.4 backport16:44
slangasekxnox: it can, but is that actually the cause of the CPU usage?16:44
mamenyakajjohansen, I already compiled the new image, but aa-status shows not loaded16:44
mamenyakajjohansen, is that patch he cause or am I doing something wrong?16:45
jjohansenmamenyaka: is securityfs mounted?16:45
mamenyakaoh16:45
mamenyakajjohansen, any chance to have adb?16:45
xnoxslangasek: when ogra_ tested upstart with this change in http://launchpadlibrarian.net/153093319/upstart_1.10-0ubuntu6_1.10-0ubuntu7.diff.gz systemd-udev was still spinning at 5% CPU. pitti was suspecting the same, as well udev would constantly parse and try to apply rules against those events.16:46
jjohansenmamenyaka: the missing patch is just a bug fix for a segfault that could happen in some circumstances, so its not the cause16:46
xnoxslangasek: so since udevd is using udev_monitor_new_from_netlink, i thought it should be possible to put a netlink filter in place.16:46
=== dandrader|lunch is now known as dandrader
jjohansenmamenyaka: init needs to mount securityfs, and run the apparmor initscripts16:47
xnoxslangasek: but i haven't looked into libudev yet.16:47
mamenyakajjohansen, how do I check it if it's mounted?16:47
jjohansenmamenyaka: I use adb all the time with this so that isn't the problem16:47
jjohansenmamenyaka: mount | grep securityfs16:48
mamenyakajjohansen, I used adb all the time too16:48
mamenyakajjohansen, okay, that's what I tought16:48
mamenyakajjohansen, somehow I managed to check it with the terminal app, but on my tablet it's so messed up16:52
mamenyakajjohansen, so, securityfs isn't mounted16:52
jjohansenmamenyaka: on the tablet, what is the value of /sys/module/apparmor/parameters/enabled16:53
mamenyakajjohansen, give me 10 mintues to type that16:53
mamenyakajjohansen, I'm guessing that it doesn't exists, no apparmor in module/16:54
jjohansenmamenyaka: in case I wasn't clear, by value I mean16:54
jjohansen   cat /sys/module/apparmor/parameters/enabled16:54
jjohansenmamenyaka: okay, so your compile does not have apparmor enabled, you are going to have to edit the configs16:55
mamenyakajjohansen, make menuconfig?16:55
jjohansenmamenyaka: how are you building your kernel?16:56
mamenyakajjohansen, I'm brunching the whole thing16:56
jjohansenmamenyaka: okay the 3.4 backport is repushed with the missing patch UBUNTU: SAUCE: apparmor: Use shash crypto API interface ...16:58
jjohansenI rebased it so the patch sequence follow the ubuntu master kernel and the backport patches are all applied after, that way its just easier to check that all the patches against the main kernel are also on the backport16:58
jjohansenmamenyaka: so you are using make directly instead of fakeroot debian/rules, or the other build scripts. In that case yeah make menuconfig16:59
=== gatox_lunch is now known as gatox
mamenyakajjohansen, got it: e940537 UBUNTU: SAUCE: apparmor: Use shash crypto API interface for profile hash17:01
mamenyakajjohansen, do you know where to look for apparmor?17:03
jjohansenmamenyaka: kernel security17:03
jjohansenmamenyaka: you want apparmor=y and default security=apparmor17:04
mamenyakajjohansen, Enable the securityfs filesystem17:04
mamenyakathis?17:04
jjohansenmamenyaka: yes17:04
mamenyakajjohansen, so in Enable different security models there is apparmor (NEW)17:05
mamenyakajjohansen, and I set Default security module (AppArmor)17:05
jjohansenmamenyaka: yes17:06
jjohansenmamenyaka: all the apparmor sub configs are at their default values17:06
mamenyakajjohansen, I got this in the config file: +CONFIG_SECURITY=y +CONFIG_SECURITY_APPARMOR=y17:07
=== alan_g is now known as alan_g|EOD
jjohansenmamenyaka: from the mako (v3.4) config I have these values17:09
jjohansenCONFIG_DEFAULT_SECURITY_APPARMOR=y17:09
jjohansenCONFIG_SECURITY_APPARMOR=y17:09
jjohansenCONFIG_SECURITY_APPARMOR_BOOTPARAM_VALUE=117:09
jjohansenCONFIG_SECURITY_APPARMOR_HASH=y17:09
jjohansen# CONFIG_SECURITY_APPARMOR_STATS is not set17:09
jjohansenCONFIG_SECURITY_APPARMOR_UNCONFINED_INIT=y17:09
nik90tedg: when u add a %u to the Exec in the desktop file, wouldn't it parse the arguments on every launch?17:09
tedgnik90, Not sure what you're saying.  We do resolve all the % entries on the Exec line every launch.  If we have no URI, we don't replace it with anything.17:09
mamenyakajjohansen, in mak menuconfig I saw that these are enabled, but in the config file only two new lines17:09
jjohansenodd17:10
mamenyakajjohansen, this was also the case with the basic changes from the proting guide17:10
nik90tedg: so if the Exec= qmlscene %u ubuntu-clock-app.qml, the %u will be used only when it is passed and all other times it will be ignored?17:10
jjohansenmamenyaka: I'd go back in and make sure that the default at least is set to apparmor17:11
mandelslangasek, I'm EOD, but could you please test the bug fix I added for bug #1233435,I have no access to a nexus 4 yet on the nexus 7 the fix seems to work17:12
ubot5bug 1233435 in ubuntu-download-manager "u-d-m downloads stall if the network configuration changes" [Critical,Triaged] https://launchpad.net/bugs/123343517:12
jjohansenotherwise you are likely to get apparmor, but it is disabled by default and you need a kernel boot param to enable it17:13
mamenyakajjohansen, did that with the swap config decades ago, no CONFIG_SWAP=y in the config file, but in menuconfig it's there17:13
tedgnik90, Correct17:13
jjohansenmamenyaka: okay, well give it a try then17:13
mamenyakajjohansen, but I will check it again after the brunch17:13
mamenyakajjohansen, still Default security module (AppArmor)  --->17:15
mhr3_am i the only one seeing17:15
mhr3_phablet   1316  0.5 13.4 103620 95184 ?        S    14:43   0:50 init --user17:15
mhr3_yes, that's 100mb of ram used by upstart17:15
mhr3_tedg, ^?17:15
jjohansenmamenyaka: okay it should be good then17:15
mhr3_jodh, ^^?17:16
tedgmhr3_, There's a bug for it.17:16
mhr3_oh that makes me feel so much better17:16
ogra_mhr3_, there is a fix for it.17:17
awafaahow does one disable screen timeout?17:18
awafaaor at least set it to a suitably high value?17:18
* awafaa cant see an option in settings17:18
jjohansenmamenyaka: okay v3.0-backport-of-apparmor3 branch has been pushed17:23
mamenyakajjohansen, great! just installing the new image17:24
slangasekmandel: so ubuntu-download-manager is using qmake instead of cmake for its build (which I believe is contrary to our policy for upstream projects, fwiw - I guess I need to nag tvoss to get that communicated more clearly :), which means I can't cross-build it.  So I don't have time to do a manual native build and test it; I'll just wait for it to be available in the archive17:25
tvossslangasek, you nagged me enough now ;)17:26
slangasekhaha17:26
seb128slangasek, do you plan to port all qt upstream components to cmake? ;-)17:27
seb128slangasek, qmake seems quite popular there17:27
slangasekseb128: I don't plan to do any porting, if you ask me to port it you'll get autotools :P17:28
seb128\o/17:29
* seb128 takes autotools any day ;-)17:29
ogra_stgraber, err, argh ... i have just noted  that your / fstab entry enables fsck ... please swithc that off (the /dev/root line need to end in 0 0 ... not 0 1)17:30
mterryboiko, heyo!  So what's the landing plan for the various telephony/upstart-app-launch fixes?17:30
ogra_*needs17:30
boikomterry: as soon as we fix the remaining bugs :)17:30
mamenyakajjohansen, looks like something is not right, sill no modules/apparmor17:31
mterryboiko, do you want a hand?17:31
mterryboiko, reviewing or dev?17:31
boikomterry: so, bfiller found a problem that calling "view contact profile" from the call log opens the addressbook app, but doesn't show the contact details17:31
jjohansenmamenyaka: even if apparmor is not on if it built in the kernel it should show up in /sys/module/apparmor17:31
jjohansenmamenyaka: does it show up in your build output?17:32
mterryboiko, oh odd, I thought I tried that and it worked for me17:32
mterryboiko, did he try the branch after my addressbook fix?17:32
mamenyakajjohansen, build output? should I log it?17:32
boikobfiller: ^17:33
mterryaddressbook wasn't properly handling any incoming requests17:33
bfillermterry: trying the latest stuff that was built by jenks17:33
jjohansenmamenyaka: you get output like17:33
jjohansensecurity/apparmor/lsm.o17:33
jjohansensecurity/apparmor/...17:33
jjohansensecurity/apparmor.o17:33
bfillerlet me check17:33
mamenyakajjohansen, oh right17:33
jjohansenmamenyaka: just trying to see if kbuild even built it17:33
mamenyakajjohansen, here, right? /pollux_windy/obj/KERNEL_OBJ/security17:34
jjohansenmamenyaka: apparmor it self will issue the following messages into the kernel log as it starts up17:34
bfillermterry: testing with debs from here, should have renato changes: https://code.launchpad.net/~renatofilho/address-book-app/fix-1230404/+merge/188473/comments/43566117:35
mamenyakajjohansen, okay, guess no apparmor yet17:35
mamenyakasearched for apparmor.o17:35
mamenyakanothing17:35
jjohansenmamenyaka: okay its not building, something in the config17:35
jjohansenmamenyaka: you could just copy the config entries I pasted earlier directly into your config17:36
jjohansenmamenyaka: hrmm perhaps a build deps is not selected17:37
mterrybfiller, good, that's the fixed one17:37
mamenyakajjohansen, well, I'll try17:37
jjohansenmamenyaka: apparmor depends on security and net, and then it selects the other bits it needs17:37
=== Tassadar_ is now known as Tassadar
jjohansenmamenyaka: and security depends on sysfs17:38
mamenyakajjohansen, but as you can see: http://kepfeltoltes.hu/131009/apparmor_www.kepfeltoltes.hu_.png17:39
jjohansenmamenyaka: indeed, but for some reason kbuild isn't building it17:39
mamenyakajjohansen, where can I check that net?17:40
jjohansenmamenyaka: grep your .config for CONFIG_NET17:40
jjohansenmamenyaka: but really menuconfig shouldn't even show the option unless that is enabled17:41
mamenyakajjohansen, whole buch of config_net is there17:41
mamenyakajjohansen, CONFIG_NET=y17:41
jjohansenyep that is it17:41
jjohansenmamenyaka: touch security/apparmor/apparmor/include/apparmor.h17:42
jjohansenand then run your build again17:43
mamenyakajjohansen, I am using an included script: https://github.com/mamenyaka/android_kernel_sony_apq8064/blob/phablet-saucy/make_defconfig.sh for make menuconfig17:43
jjohansenmamenyaka: that will update the time stamp on those files, but not everything else, so you should be able to easily see apparmor building17:43
jjohansenmamenyaka: hrmmm, does you .config differ from your arch/arm/configs/${DEFCONFIG_FILE}17:46
mamenyakajjohansen, I was directly editing the defconfig file, I have no .config17:47
jjohansenmamenyaka: are you sure? It sounds to me like kbuild is using a different config file17:48
mamenyakajjohansen, check the script I linked17:48
mamenyaka./make_defconfig.sh fusion3_pollux_windy_defconfig17:48
jjohansenmamenyaka: generally I would do a17:49
jjohansen  make defconfig               #copy default config to .config17:49
jjohansen  make17:49
jjohansenmamenyaka: and what does make_defconfig.sh look like?17:49
mamenyakajjohansen, go up ten lines17:50
=== anders3408|afk is now known as anders3408
mamenyakajjohansen, https://github.com/mamenyaka/android_kernel_sony_apq8064/blob/phablet-saucy/make_defconfig.sh17:50
jjohansenmamenyaka: right, okay that is editing the config, I mean actually building the kernel17:50
jjohansenmamenyaka: make defconfig, copies the default config to the .config file that is going to be used for building17:51
mamenyakajjohansen, default config file?17:52
jjohansenmamenyaka: are you building in the source tree or are you setting a build directory?17:52
jjohansenmamenyaka: you edit script, edits the default config file17:52
jjohansenmamenyaka: what is the command you are using to do the build17:52
mamenyakajjohansen, brunch pollux_windy17:53
jjohansenmamenyaka: okay, so I have never dug into what these wrappers are doing exactly, where are the object files of the build?17:57
mamenyakajjohansen, in out/target/product/17:58
mamenyakajjohansen, but funny thing is, I tried the 3.0 kernel for my other port17:58
jjohansenmamenyaka: is there a out/target/product/.config file?17:58
mamenyakajjohansen, no17:59
mamenyakajjohansen, I don't want to make your life any harder, the 3.0 kernel is behaving a little bit better17:59
mamenyaka-# CONFIG_SECURITYFS is not set17:59
mamenyaka+CONFIG_SECURITYFS=y17:59
mamenyaka CONFIG_SECURITY_NETWORK=y17:59
mamenyaka # CONFIG_SECURITY_NETWORK_XFRM is not set17:59
mamenyaka-# CONFIG_SECURITY_PATH is not set17:59
mamenyaka+CONFIG_SECURITY_PATH=y17:59
mamenyaka CONFIG_LSM_MMAP_MIN_ADDR=3276817:59
mamenyaka CONFIG_SECURITY_SELINUX=y17:59
mamenyaka # CONFIG_SECURITY_SELINUX_BOOTPARAM is not set17:59
mamenyaka@@ -3213,11 +3213,16 @@ CONFIG_SECURITY_SELINUX_AVC_STATS=y17:59
mamenyaka CONFIG_SECURITY_SELINUX_CHECKREQPROT_VALUE=117:59
mamenyaka # CONFIG_SECURITY_SELINUX_POLICYDB_VERSION_MAX is not set17:59
mamenyaka # CONFIG_SECURITY_TOMOYO is not set17:59
mamenyaka-# CONFIG_SECURITY_APPARMOR is not set17:59
mamenyaka+CONFIG_SECURITY_APPARMOR=y17:59
mamenyaka+CONFIG_SECURITY_APPARMOR_BOOTPARAM_VALUE=117:59
mamenyaka+# CONFIG_SECURITY_APPARMOR_STATS is not set17:59
mamenyaka+CONFIG_SECURITY_APPARMOR_UNCONFINED_INIT=y18:00
mamenyaka+CONFIG_SECURITY_APPARMOR_HASH=y18:00
mamenyaka # CONFIG_IMA is not set18:00
mamenyaka-CONFIG_DEFAULT_SECURITY_SELINUX=y18:00
mamenyaka+# CONFIG_DEFAULT_SECURITY_SELINUX is not set18:00
mamenyaka+CONFIG_DEFAULT_SECURITY_APPARMOR=y18:00
mamenyaka # CONFIG_DEFAULT_SECURITY_DAC is not set18:00
mamenyaka-CONFIG_DEFAULT_SECURITY="selinux"18:00
mamenyaka+CONFIG_DEFAULT_SECURITY="apparmor"18:00
mamenyakaoops, sorry18:00
jjohansenmamenyaka: heh np, that does look better18:00
jjohansenmamenyaka: I'm not sure what is going on with your 3.4 build but it is something to do with the config18:00
jjohansen or setup18:01
mamenyakajjohansen, let's see what I get from 3.018:01
mamenyakajjohansen, also, are there any ports with apparmor backported?18:01
jjohansenmamenyaka: yes but I am not sure which ones18:05
jdstrandsergiusens: fyi: applicationName is not correct: Oct  9 18:04:45 ubuntu-phablet kernel: [  365.868757] type=1400 audit(1381341885.333:125): apparmor="DENIED" operation="mkdir" parent=1369 profile="com.ubuntu.sudoku_sudoku_0.4.3" name="/home/phablet/.local/share/sudoku-app/" pid=2366 comm="qmlscene" requested_mask="c" denied_mask="c" fsuid=32011 ouid=3201118:06
mamenyakajjohansen, is this because of apparmor? scripts/kconfig/zconf.tab.c:206:24: fatal error: zconf.hash.c: No such file or directory18:07
mamenyaka #include "zconf.hash.c"18:07
mamenyaka                        ^18:07
mamenyakacompilation terminated.18:07
jdstrandsergiusens: that is on 8918:08
barrymandel: https://code.launchpad.net/~barry/ubuntu-system-image/lp121558618:12
mterrybfiller, any luck with the updated addressbook?18:14
jjohansenmamenyaka: I don't think so but, you can try disabling "SHA1 hash of loaded profiles (NEW)" in the menu config, it depends on hashing, disabling it disables the creation of policy hash file, its not critical, it can just be used to compare policy in userspace vs what is loaded in the kernel18:15
mamenyakajjohansen, I got around it, but now I am getting apparmor related errors18:15
jjohansenmamenyaka: that is a subconfig of apparmor18:15
jjohansenmamenyaka: what are the errors?18:15
bfillermterry: there are issues, salem_ was looking into it. seems to work ok if address-book is already running but when it's not looks like some sort of race18:16
mamenyakajjohansen, just a sec18:16
larsugema: hi, what's up?18:17
salem_bfiller, mterry I am trying a local fix. I am confident it will fix the issue.18:17
mamenyakajjohansen, http://paste.ubuntu.com/6214836/18:18
jjohansenmamenyaka: okay the kernel you are porting to has a backport of 02125a82, which changed the d_path api, you need to drop the patch18:20
jjohansen  UBUNTU: SAUCE: apparmor: 3.1 backport __d_path api intro'd in 02125a8218:20
=== dandrader is now known as dandrader|afk
mterrysalem_, awesome.  If you need a testing guinea pig, let me know18:20
mamenyakajjohansen, so revert it?18:22
jjohansenmamenyaka: yeah you can revert or just drop that patch18:22
mamenyakajjohansen, how can I just drop a patch?18:22
jjohansenmamenyaka: git rebase -i XXXX, but if you don't know how to use it revert is easier18:23
mamenyakajjohansen, I not yet used revert, so I'll stick to revert18:24
jjohansenmamenyaka: XXXX would be something like HEAD~10, in this case. Then the command pops into an editor with a list of patches, and you delete patch lines18:24
jjohansenmamenyaka: git log, find the sha1 of the patch, then git revert <sha1>18:24
mamenyakajjohansen, already done18:25
mamenyakajjohansen, lsm.c:801:4: error: expected specifier-qualifier-list before '.' token18:26
mamenyakalsm.c:810:35: error: 'path' undeclared (first use in this function)18:26
jjohansenmamenyaka: what is at line 810 of lsm.c18:27
* jjohansen thinks he knows18:27
mamenyakareturn aa_path_perm(op, label, &path, 0, mask, &cond);18:28
jjohansenmamenyaka: yep, okay messed up the merge conflict on patch18:29
jjohansen  UBUNTU: SAUCE: apparmor: 3.3 backport revert path in unix_sock 40ffe6718:29
mamenyakajjohansen, this will be a long night I guess...18:32
mamenyakajjohansen, lsm.c:806:36: error: 'struct unix_sock' has no member named 'path'18:32
jjohansenmamenyaka: in lsm.c, a few lines up from 810 change18:32
jjohansen  struct path {18:32
jjohansento18:32
jjohansenstruct path path ={18:32
Saviqev, ping18:32
mamenyakajjohansen, oh, so I don't need to revert that one?18:33
mamenyakajjohansen, okay18:33
jjohansenmamenyaka: you could revert it and apply the new version, or just directly edit your choice18:33
jjohansenmamenyaka: I am making the edits and will push the revised versions18:34
=== dandrader|afk is now known as dandrader
ogra_stgraber, did you see the ping above ?18:37
pc_magasHello do you know a way to make an Ubuntu Touch app have conectivity over 3g?18:37
ogra_why would the app care ?18:38
ogra_the system will provide networking to the app, why would it care which kind of networki is used18:38
pc_magasorga_ actually I want to make an app that will be able to connect to wikipedia and get some data18:39
stgraberogra_: /dev/root is fine as 0 1 since /dev/root doesn't actually exist as a device and we need the entry to match that from /lib/init/fstab with only the ro flag added18:40
ogra_pc_magas, right, it should just rely on the system to provide the access18:40
stgraberogra_: however all the others should be 0 0 and not 0 1, so we need to change that part18:40
ogra_stgraber, you tell mountall to run fsck on mount with the 118:40
ogra_stgraber, we had a vUDS session about this18:41
stgraberogra_: I know, that's why I say we should have 0 0 for all but /dev/root18:41
ogra_stgraber, mountall will do that even for /dev/root once it remounts18:41
stgraberogra_: /dev/root is not mounted by mountall and isn't a valid path anyway18:41
pc_magasorga_ yes but how can I initialize an http connection via js?18:41
stgraberogra_: and it won't ever remount it since it keeps it read-only18:41
pc_magascan it done like AJAX?18:41
ogra_stgraber, see /lib/init/fstab :) ...18:41
ogra_stgraber, if you boot a PC without fstab it uses this file ... and forces an fsck18:41
ogra_on first boot18:42
stgraberogra_: sure because a normal PC has a root device18:42
ogra_our kernel provides onr too18:42
salem_mterry, bfiller try append a %u to the Exec=address-book-app in /usr/share/applications/address-book-app.desktop18:42
ogra_*one18:42
mamenyakajjohansen, looks good, image built18:42
cyphermoxogra_: rsalveti: http://paste.ubuntu.com/6214937/18:42
cyphermox^ permission to upload plz :)18:43
ogra_stgraber, i would prefer to be on the safe side here ... in case there is really an fsck we could end up in a deadlock due to not having any plumouth input abilities18:43
ogra_cyphermox, put it on the spreadsheet, uploads are blocked until Mir is sorted18:43
stgraberogra_: sure, just s/0 1/0 0/g in scripts/touch and upload then18:43
cyphermoxwhat, is mir still broke?18:43
rsalveticyphermox: looks fine, that's grouper specific right?18:43
rsalveticyphermox: just sync with lool and push18:44
ogra_lool, ^^^ cyphermox has a fix for the BT eats 100% CPU on grouper18:44
cyphermoxrsalveti: no, it looks at all the files, but it only changes things on grouper and avoids us getting into the same issue on maguro18:44
rsalvetiright18:44
cyphermoxrsalveti: but it's safe on all devics18:44
ogra_stgraber, thanks18:44
rsalvetiyeah, cool18:44
bfillersalem_: let me try18:46
mterrysalem_, bfiller: works for me18:46
mterry(and I reproduced the problem before changing)18:47
salem_mfisch, bfiller yes, that fixes part of the problem, adding a new contact from messaging-app is still broken. the main view is pushed over the edit view right after contacts app is launched.18:48
bfillermterry, salem_ : do dialer and messaging need the %u as well?18:48
salem_mterry, ^18:48
salem_bfiller, I think boiko added the %u to those apps already18:48
mterrysalem_, bfiller: they have them already18:49
bushcHello, I installed Ubuntu Touch 12.10 to nexus 7 (2012) wifi only and now my Touchscreen isn't working. I installed it over cwm recovery. pls help me18:50
mamenyakajjohansen, did you update v3.0-backport repo?18:53
jjohansenmamenyaka: I just pushed to it again, hopefully its right this time18:54
bfillersalem_: when editing a contact or creating new, seems like pressing cancel/save not working under MIR. you seeing this?18:55
kenshiroHi, there will be some instant messaging software installed in Ubuntu Touch in October 17th release? I would like to know too if Google Maps and Google Contacts will work.18:55
salem_bfiller, yes, I am seeing the same thing.18:55
mterrybfiller, it is finicky18:56
mterrybfiller, it will work18:56
mterrybfiller, but not always18:56
bfillermterry, salem_ : works if you swipe away the keyboard18:56
mterrybfiller, but I think that's a separate issue from the MR18:56
bfilleryup18:56
mamenyakajjohansen, is it only me but I got: <<<<<<< HEAD18:56
mamenyakastruct path {18:56
mamenyaka=======18:56
mamenyakastruct path path = {18:56
mamenyaka>>>>>>> be3e89295c63a7ac1939109c1ef6d1470dbdd5b818:56
bfillermterry: is there a bug open on that do you know?18:56
mterrybfiller, I don'tk now18:56
jjohansenmamenyaka:18:56
jjohansenmamenyaka: let me check18:57
jjohansenmamenyaka: its good here, are you getting a merge conflict?18:58
mamenyakajjohansen, I got forced update18:58
mamenyakajjohansen, oh, okay, then there is the merge conflict18:58
jjohansenmamenyaka: yeah, I did a rebase and forced the update18:58
mamenyakajjohansen, sorry, it's okay now18:59
bfillermterry, salem_ : looks likes its already being fixed https://bugs.launchpad.net/ubuntu-keyboard/+bug/123460018:59
ubot5Ubuntu bug 1234600 in unity-mir "Contacts Cancel/Save buttons don't work with OSK showing when running Mir" [High,In progress]18:59
mterrybfiller, awesome19:01
mamenyakajjohansen, guess what!19:01
timppais there a bug open on SDK + MIR, "run on device does not work" ?19:01
mamenyakajjohansen, root@ubuntu-phablet:/# aa-status apparmor module is loaded.19:01
mterryseb128, are there known problems with changing the background?  I'm running mir and when I pick a new background, my screen goes black and doesn't come back on19:05
mterryseb128, also, we can't have separate welcome/dash backgrounds anymore?19:05
mterrybfiller, so do the branches work OK now?  What is next step for the suite of updates?19:06
seb128mterry, the black background is a known Mir issue19:06
seb128mterry, the "no separate welcome/dash background" is because we got told that changing the lock screen image was not going to be supported for v119:07
bfillermterry: sounds like one remaining issue salem_ is working with address book19:07
bfillermterry: then we should merge these and get them released19:07
seb128mterry, I think the rational was "random images don't work nicely with the artwork on the greeter"19:07
mterryseb128, ?19:07
mterryseb128, I thought that the semi-transparent infographic fixed that problem19:08
jjohansenmamenyaka: nice19:08
mterryseb128, that was the rationale for that change anyway19:08
seb128mterry, it could be, in fact changing the user background changes the greeter one atm I think?19:08
mamenyakajjohansen, but apps don't seem to work (calculator)19:08
salem_bfiller, mterry I actually asked renato to do the changes, as that MR belongs to him.19:09
mterryseb128, yeah I believe so (if I could change my background)19:09
mamenyakajjohansen, white then black screen19:09
seb128mterry, the code is still there, "USS_SHOW_ALL_UI=1 system-settings"19:09
=== beuno_ is now known as bueno
seb128mterry, we are hiding the controls with non functional widgets19:09
bfillersalem_: do you and renato know how to fix that last issue?19:09
seb128mterry, but that env enable them back19:09
salem_bfiller, yes, I think so.19:09
jjohansenmamenyaka: hrmm, grep DENIED /var/log/syslog19:10
mterrysalem_, bfiller: well that fix landed in the branch19:10
mterry(the %u one)19:10
salem_mterry, yes, but there's one more.19:10
mterrysalem_, ah  :)19:10
mterryalways one more19:11
mamenyakajjohansen, lot of DENIEDs19:11
salem_mterry, :)19:11
mterryseb128, will test in SF19:11
mamenyakajjohansen, ubuntu-phablet kernel: [  505.282110] c0 type=1400 audit(1381345734.744:130): apparmor="DENIED" operation="mkdir" parent=1 profile="com.ubuntu.calculator_calculator_0.1.3" name="/home/phablet/.local/share/ubuntu-calculator-app/" pid=3740 comm="qmlscene" requested_mask="c" denied_mask="c" fsuid=32011 ouid=3201119:11
seb128mterry, thanks, let me know if that works, it's one "visible" line to drop to enable the control back19:11
sergiusensjdstrand, ack19:12
pc_magasWhile I was Developing I cannot see the app.js File on the projects tree19:13
jjohansenjdstrand: shouldn't the base profile allow for mamenyaka ^ reject?19:13
pc_magasBut I can see into fiulesystem tree19:13
jdstrandjjohansen: yes and no. the calculator app is not setting its applicationName correctly so it is getting that denial19:14
jdstrandsergiusens: fyi, there's another one for you ^19:14
sergiusensjdstrand, the two games and stock?19:14
jjohansenjdstrand: ah thanks, mamenyaka did you see jdstrand's answer^19:14
sergiusensjdstrand, looking into those now19:14
mamenyakais there any reason I am getting kicked out from adb shell every minute?19:15
jdstrandsergiusens: honestly, I imagine it is everything your are packaging since it was developed before the sdk did the right thing19:15
mamenyakajjohansen, yes, but other apps are acting the same19:15
jdstrandsergiusens: but you'll only see the denial right now for things that are accessing local files19:15
sergiusensjdstrand, yup19:15
kenshiroHi, there will be some instant messaging software installed in Ubuntu Touch in October 17th release? I would like to know too if Google Maps and Google Contacts will work.19:15
mamenyakajjohansen, ubuntu-phablet kernel: [  846.234679] c0 type=1400 audit(1381346075.694:133): apparmor="DENIED" operation="mkdir" parent=1 profile="com.ubuntu.dropping-letters_dropping-letters_0.1.2.2" name="/home/phablet/.local/share/dropping-letters/" pid=4117 comm="qmlscene" requested_mask="c" denied_mask="c" fsuid=32011 ouid=3201119:16
jdstrandmamenyaka: kicked out-- I wonder if that is related to the mtp stuff. I find I get kicked out and I have a nautilus window showing me the device19:16
mamenyakayep19:16
mamenyakajdstrand, exactly19:16
jdstrandmamenyaka: after that initial kick out, I can connect for long periods19:16
jdstrandmamenyaka: that is the same issue as the calculator19:17
mamenyakajdstrand, for me it's no longer than a minute19:17
jdstrandmamenyaka: if you try twitter or facebook, they won't have the denials19:17
mamenyakaalso, any reason why the keyboard is popping up as if the device was in portrait mde?19:17
* jdstrand doesn't know anything about ubuntu-keyboard19:18
mamenyakaubuntu-phablet kernel: [ 1050.292322] c3 type=1400 audit(1381346279.754:134): apparmor="DENIED" operation="open" parent=1 profile="com.ubuntu.developer.webapps.webapp-facebook_webapp-facebook_1.0.5" name="/dev/ump" pid=4421 comm="webbrowser-app" requested_mask="rw" denied_mask="rw" fsuid=32011 ouid=100019:18
mamenyakaOct  9 19:18:00 ubuntu-phablet dbus[2430]: apparmor="DENIED" operation="dbus_method_call"  bus="session" path="/org/freedesktop/DBus" interface="org.freedesktop.DBus" member="RequestName" mask="send" name="org.freedesktop.DBus" pid=4421 profile="com.ubuntu.developer.webapps.webapp-facebook_webapp-facebook_1.0.5" peer_profile="unconfined"19:18
mterryseb128, when I pick a new bg, it always seems to change the Home one, regardless of which side (left/right) I clicked on19:18
jdstrandmamenyaka: that is a hardware device access. please see https://wiki.ubuntu.com/Touch/Porting#AppArmor for how to fix that19:19
seb128mterry, weird19:21
=== tdaitx_ is now known as tdaitx
seb128mterry, are you sure the greeter has its own bg? it's running as the user right... is it picking the user bg?19:21
mterryseb128, it pulls the bg from accountsservice19:22
mterryseb128, so whichever value is there, it grabs19:22
seb128hum19:22
seb128mterry, let me try19:22
mterryseb128, this doesn't have to be fixed for v1, just thought we had this feature already19:22
jdstrandrsalveti: btw, not sure if you saw, but https://wiki.ubuntu.com/Touch/Porting#AppArmor is updated with instructions on dealing with hardware-specific apparmor rules like we discussed before19:23
mterryseb128, we certainly are updating the welcome bg when user changes home bg, so we don't have an excuse not to have it, design wise.  But if it's buggy that's a different issue19:23
seb128mterry, http://bazaar.launchpad.net/~system-settings-touch/ubuntu-system-settings/trunk/view/head:/plugins/background/background.cpp#L6919:23
seb128mterry, it might be that we update both19:23
* seb128 checks19:23
mamenyakajdstrand, thanks, I'm on it, but adb isn't on my side19:28
mamenyakajdstrand, if the app is not starting at all, what do I do?19:28
jdstrandmamenyaka: if there are apparmor denials, fix those. if the app still doesn't launch, it is something else in the stack19:30
mamenyakajdstrand, no denials19:30
jdstrandmamenyaka: which app?19:31
mamenyakajdstrand, facebook19:31
jdstrandmamenyaka: does 'Browser' launch?19:31
mamenyakajdstrand, yes19:31
jdstrandmamenyaka: does twitter?19:32
mamenyakajdstrand, yes, but it's white19:32
jdstrandmamenyaka: can you do: sudo sysctl -w kernel.printk_ratelimit=019:33
jdstrandmamenyaka: then try again on both?19:33
mterryseb128, how do I launch the app on the phone with the SHOW_ALL_UI env var?  I just sed'd to get it done for the moment19:33
mterryseb128, i.e. using the upstart mechanisms, can I pass an env var?19:34
rutenldoes anyone know why my galaxy nexus just shows a black screen after flashing the ubuntu touch files, I did a factory reset before flashing it19:34
sergiusenspopey, jdstrand https://code.launchpad.net/~sergiusens/dropping-letters/click/+merge/19022819:34
mamenyakajdstrand, facebook starts, black19:34
mamenyakajdstrand, twitter also19:34
jdstrandmamenyaka: right, but do you have apparmor denials?19:35
mamenyakajdstrand, now yes19:35
jdstrandmamenyaka: what are they?19:35
mamenyakajdstrand, I'm on it19:35
jdstrandmamenyaka: note, I updated the Porting page just now, you might refresh19:36
seb128mterry, adb shell; su - phablet; system-settings --desktop_file_hint=/usr/share/applications/ubuntu-system-settings.desktop19:36
seb128mterry, with SHOW_ALL_UI=1 before the command19:36
jdstrandmamenyaka: basically it now says to look at https://wiki.ubuntu.com/DebuggingApparmor#Fixing_profile_bugs for more info on debugging profile issues19:36
jdstrandmamenyaka: (which has the sysctl tip)19:37
seb128mterry, it seems that it always update the gsettings key and never the accountsservice one...19:38
mamenyakajdstrand, okay now, working19:38
jdstrand\o/19:38
mterryseb128, ah forgot about the desktop_file_hint19:38
mamenyakajdstrand, where do I commit my file?19:38
mterryseb128, yeah OK.  That's what I saw when I sed'd the UI in19:38
mamenyakajdstrand, apparmor-easyprof-ubuntu_GT-N800019:38
seb128mterry, I'm going to debug it, let that to me, no need to duplicate work19:39
jdstrandmamenyaka: that is a good question-- you would be the first. I envisioned in lxc-android-config (and thus you could drop the 'apparmor-easyprof-ubuntu_' bit19:39
jdstrandrsalveti, ogra_: thoughts ^19:39
jdstrandmamenyaka, rsalveti, ogra_: I figure it makes sense in lxc-android-config because that is where the device specific udev rules are19:40
mterryseb128, OK, thanks19:40
jdstrandmamenyaka, rsalveti, ogra_: so it should ship the device-specific apparmor policy snippets19:40
* jdstrand planned to move all the apparmor-easyprof-ubuntu_* out to lxc-android-config after 13.10 release19:41
ogra_jdstrand, right19:41
ogra_mamenyaka, file a bug and attach the file19:41
rutenlAnyone, i did a factory reset and i flashed ubuntu touch on my galaxy nexus. and when i start my phone now it only shows the google logo for a while and then just a black screen. adb does see the device. what do i do?19:41
ogra_btw, aweome work !19:41
ogra_rutenl, wait a little longer ... the maguro is slow on boot19:42
* ogra_ vanishes back into the night 19:42
mamenyakaogra_, thank you! wouldn't been possible if it wasn't for the great backport of jjohansen19:42
ogra_++19:43
mamenyakawhere do I file the bug? lxc-android-config?19:44
WebbyITtedg: ping :)19:45
jdstrandmamenyaka: yes19:45
tedgWebbyIT, Howdy19:46
WebbyITtedg, hi, I have implemented Argument in clock app and in calendar app, and added %u in desktop entry, can you please review it?19:47
WebbyIThttps://code.launchpad.net/~rpadovani/ubuntu-clock-app/09Oct2013/+merge/19013919:47
mamenyakajdstrand, https://bugs.launchpad.net/ubuntu/+source/lxc-android-config/+bug/123761819:49
ubot5Ubuntu bug 1237618 in lxc-android-config (Ubuntu) "apparmor policy snippets for N8000, N8013" [Undecided,New]19:49
mamenyakaoh, fancy bot there19:49
WebbyITtedg, Added %u in .desktop Exec entry, before app command due bug #123155819:49
ubot5bug 1231558 in Ubuntu UI Toolkit "qmlscene prevents using default command line arguments after the .qml filename" [Medium,Confirmed] https://launchpad.net/bugs/123155819:49
bfillerkenvandine: do I need any content-hub fixes to test gush's MR: https://code.launchpad.net/~schwann/gallery-app/gallery-pick-return/+merge/19018119:50
pc_magasHello I amde an app havinh thias Index.html http://pastebin.com/2rNVNk4h and RThis App.js http://pastebin.com/v0fER4Aq for Ubuntu phone using html519:51
bfillerkenvandine: I tried it and not being returned back to calling app after gallery picks19:51
tedgWebbyIT, Dektop file looks good!19:51
pc_magasThis app get something from an input field and prints it19:51
kenvandinebfiller, to switch back you need another branch of content-hub19:51
kenvandinewaiting to be merged19:51
pc_magasBut when sucessfull prints it then whatever I gave disapears19:51
bfillerkenvandine: ok, can you point me at the MR?19:51
kenvandinebut you should be able to switch back yourself and see the background19:51
pc_magasand I do not know why19:51
kenvandinehttps://code.launchpad.net/~ken-vandine/content-hub/invoke_destination/+merge/19021719:52
kenvandinebfiller, ^^19:52
kenvandinebfiller, and you're on Mir right?19:52
WebbyITtedg, good, thanks19:52
kenvandinewithout Mir, you'll get multiple instances of the calling app19:52
bfillerkenvandine: yup, will try it now19:52
WebbyITtedg, I have to resolve a conflict in calendar app after today merge, I'll it to you soon19:52
=== chihchun is now known as chihchun_afk
sergiusensmhall119, the com.ubuntu ones? I am doing those today19:54
davmor2kenvandine: image 89 defaults to use mir now :)19:54
kenvandinedavmor2, yeah, that's why i proposed that branch :)19:54
sergiusenspopey, jdstrand https://code.launchpad.net/~sergiusens/sudoku-app/click/+merge/19023119:55
mfischstgraber: I've been thinking about where the update server is specified in /etc. We'll need to customize that setting eventually. How much work would it be to have your code look in /custom vs say moving it to gsettings?19:55
stgrabermfisch: you won't have to do anything special for that19:57
mfischstgraber: orly? how?19:57
stgrabermfisch: /etc/system-image/channel.ini is part of the version-XYZ.tar.xz tarball which is generated by the system-image server when publishing the image19:57
stgrabermfisch: so when using a private server, the private url will be in there and things will just work19:57
mfischstgraber: perfect, do the scripts that generate that need any work?19:58
=== diegoyam_ is now known as diegoyam
stgrabermfisch: nope, they'll just DTRT19:58
mhall119sergiusens: awesome, it'll be nice to get update of those while waiting for stable builds19:58
mfischokay nice19:58
anders3408stgraber:  are we closer to other than nexus devices being able to compile system-images yet ? :D ?19:58
stgrabermfisch: the only things that need to happen is the list of work I sent to schwuk after our meeting19:58
mfischstgraber: yep, thanks19:59
stgrabermfisch: I've started doing some of the bits myself (tired of waiting ;)) but there's still quite some work to do19:59
sergiusensmhall119, true19:59
sergiusensmhall119, that's actually a good reason for me to hurry19:59
mfischstgraber: does your update code read the h/w info to determine which bundle to pull down (for updates)?20:00
pc_magasHello how can I report probs to ubuntu SDK?20:00
kenvandinebfiller, did you see my comment on that gallery-app MP?20:02
bfillerkenvandine: yes, and I agree. will note that too20:02
kenvandinethanks20:03
kenvandinefwiw, i think the Mir bug that was triggered by the quit is fixed20:03
kenvandinewith the latest updates 30m ago20:03
stgrabermfisch: not sure I understand the question20:04
tvosskenvandine, yup20:05
kenvandinetvoss, i spoke too soon.. it isn't20:05
tvosskenvandine, damn20:05
kenvandinei thought i had the archive version of gallery-app installed20:05
kenvandinei didn't20:05
kenvandine:/20:05
tvoss:/20:05
mfischstgraber: I'll just peruse the update code and read it, but I was curious what info is transmitted back to the update server (or do you never update the device specific piece)?20:05
kenvandinei just purged it and tried again, same crash :/20:05
stgrabermfisch: so for a separate server the idea is simply to configure it to monitor and import any new rootfs or device tarball from the public server, then generate a keyring tarball, a version tarball and import a customization tarball from an internal server, and publish the result as a new image20:07
stgrabermfisch: so a device will only talk to your server and get everything from it. Any update showing up on the public server will get imported into the private server a few minutes later and then be published.20:07
bfillerkenvandine: are you able to pick and save an avator for a contact? I can pick in address book and get the picture but when I press save the pic not saved20:09
mamenyakajdstrand, do the icons on the HOME screen work?20:10
jdstrandmamenyaka: afaik20:11
mfischstgraber: right and there are 3 pieces: main, customized, and device. someone had asked how the update code knows which device image to pull down, I assume it's the same way phablet-flash does it?20:11
rutenlogra, my phone has been booting up for 15 minutes now, and i still only have a black screen.. should i wait longer or is it something different.20:11
mamenyakajdstrand, because facebook doesn't work from there20:11
jdstrandmamenyaka: but it does fro installed applications?20:11
mamenyakajdstrand, yep20:12
jdstrandmamenyaka: that sounds like a bug. I guess against unity-scope-home20:12
mamenyakacan someone else confirm it?20:12
stgrabermfisch: the device simply looks at its channel.ini, the information is in there20:13
sergiusensmhall119, btw, can you check these three? https://code.launchpad.net/~sergiusens/dropping-letters/click/+merge/190228 https://code.launchpad.net/~sergiusens/sudoku-app/click/+merge/190231 https://code.launchpad.net/~sergiusens/stock-ticker-mobile-app/click/+merge/19023820:14
mhall119sergiusens: I assume you've tested and they work20:15
joachimvdaHi, I am trying to install Ubuntu Touch on a Nexus 7 (with 3G, marked tilapia)20:16
sergiusensmhall119, let me add a comment20:16
rutenlAnyone, my phone has been galaxy nexus up for 15 minutes now, and I still only have a black screen.. should i wait longer or is it something different.20:16
rutenloops20:16
mhall119sergiusens: line 25 of https://code.launchpad.net/~sergiusens/sudoku-app/click/+merge/19023120:17
joachimvdaSeems to work well (no error mesasges), but the boot stays at google logo20:17
kenvandinebfiller, i'll try20:17
mhall119why is that hard-coded to sudoku-app now?20:17
rutenlmy galaxy nexus has been booting up for 15 minutes now* sorry20:17
kenvandineis that in the image?20:17
sergiusensmhall119, because the debian packaging uses sudoku-app20:17
joachimvdaI read a not about changing cmdline (https://lists.launchpad.net/ubuntu-phone/msg04419.html), how do I do that?20:18
sergiusensmhall119, that's not relevant to click, just wanted to be backwards compat20:18
bfillerkenvandine: yes, it's in there20:18
mhall119ok20:18
bfillerwas working but think it regressed in address book20:18
sergiusensmhall119, forgot to change in one other location (desktop file also for desktop)20:19
mhall119sergiusens: stock-ticker-mobile doesn't have the apparmor changes the other 2 had, does it need them?20:20
sergiusensmhall119, stock ticker already had them20:21
mhall119ok20:21
mhall119sergiusens: okay, all top-approved20:22
=== Namidairo`bnc is now known as Namidairo
kenvandinebfiller, it's getting the image20:24
kenvandinebfiller, where does it save the image too?20:24
johnjohn101when this is released, will this become the support board for issues?20:25
bfillerkenvandine: into eds20:26
kenvandinedoes it copy the image somewhere first?20:26
bfillerkenvandine: I think so, renato do you know?20:26
kenvandineif not, the content transfer request needs to specify a store20:26
bfillerkenvandine: right, I think renato was storing it local himself20:27
bfillerkenvandine: you seeing image go away after saving?20:27
kenvandineyeah20:27
kenvandinei see it get stored in the temp cache20:27
kenvandineand purged too20:27
renatokenvandine, bfiller are you talking about the content hub20:27
renato?20:28
kenvandinerenato, yeah20:28
renatokenvandine, yes I am saving with QTemporaryFile20:28
=== salem_ is now known as _salem
bfillerrenato: the bug is after pressing save the image is disappearing now20:29
bfillerwas working but in latest image it's not20:29
sergiusensmhall119, great, also, do you know where to get bigger icon assets for these three apps?20:34
mhall119I'm not sure who made them in the first place....20:35
mhall119lucas (community guy) perhaps20:35
mhall119or sam hewitt or georgi20:35
kenvandinebfiller, so he's creating a QTemporaryFile to do the scaling, but that isn't a persistent file20:35
mhall119popey: do you remember where we got the icons for sudoku, stock-ticker and dropping-letters?20:36
kenvandinei assume where ever that gets saved gets cleaned up20:36
bfillerkenvandine: right, supposed to get persisted to eds after it's scaled, and then cleaned up after that20:37
bfillerrenato will track it down, thanks for confirming it20:37
kenvandineoh right, eds doesn't store a path20:37
kenvandinebut the actual image20:37
kenvandineyeah20:37
kenvandinenothing has changed in the hub recently20:37
kenvandineand it is getting a file back20:37
bfillerexactly20:37
kenvandineso something with storing it in eds20:37
WebbyITtedg, https://code.launchpad.net/~rpadovani/ubuntu-calendar-app/1231136/+merge/18800220:38
WebbyITthanks :)20:38
kedivessgood evening20:39
kedivessi just installed ubuntu touch on my nexus 420:39
mhall119bfiller: kenvandine: you guys talking about contacts images?20:40
kenvandinemhall119, yes20:40
mhall119is it going to crop then scale?20:40
mhall119because right now it isn't keeping perspective when it scales, so images are stretched or squashed20:40
kenvandinemhall119, no idea, bfiller ^^20:41
mhall119I haven't filed a bug yet, but if it's not on somebody's radar I will20:41
kenvandinei hadn't even realized the address-book-app with content-hub picker had landed :)20:41
bfillermhall119: thats fixed or will be20:41
mhall119bfiller: ah, then I won't file another bug :)20:41
kenvandineit's pretty tricky to click on it20:41
mhall119thanks20:41
kenvandinebfiller, is it just the little image on the bottom right of the default avatar?20:42
kenvandineto prompt the picker?20:42
mhall119kenvandine: yeah20:42
kedivessplease, i need some help....how to add my google contacts , thanks20:42
bfillerkenvandine: currently yes but MR has selecting anywhere in the default avatar launch it20:42
kenvandinegood20:42
mhall119kedivess: have you tried the google importer app?20:42
kenvandinebfiller, maybe that little image is too close to the edge20:42
bfillerkenvandine: yes and the touch area is a bit small as well20:43
kenvandineso that would help allot20:43
kedivesssorry but i don't know how20:43
kenvandinehttps://wiki.ubuntu.com/Touch/DailyDriver20:44
kenvandinekedivess, ^^20:44
kedivessthanks for the answer20:44
kenvandinelook at the section for adding google contacts20:44
kenvandinekedivess, no problem20:44
karniWhat is the latest qtcreator version? I remember there was a staging ppa.20:44
mhall119kedivess: in the "More Suggestions" section of the Applications lens there should be an app available to install named "contactsimporter" that claims to be able to sync contacts20:44
karniI badly need it.20:44
mhall119I haven't tried it though20:45
mhall119or at least there was such an app yesterday20:47
kedivessi have another problem....my wifi network has wpa2 password but for wifi connection on phone is set for wpa....what should i have to do?20:48
karnimhall119: Sorry to bug you, I'm in a bit rush. Do you know where's the staging ppa of qtcreator/ubuntu sdk?20:48
mhall119karni: https://launchpad.net/~ubuntu-sdk-team/+archive/staging probably20:48
rsalvetijdstrand: I saw the wiki, thanks for updating that20:48
karnimhall119: yes! thank you :)))20:49
mhall119np20:49
alecuis there a bug for skewed top-left corner of images in the dash?20:50
joachimvdafor Nexus 7 tilapia, found the "fix" script, but is not working20:51
joachimvdaline "fastboot -c "console=tty1 datapart=/dev/mmcblk0p10" flash:raw boot zImage initrd.img" says "cannot load 'zImage': No such file or directory"20:51
boikosergiusens: ping20:53
kedivesswhen i'm trying to install contactsimporter, the terminal says "unable to write to /var/cache/apt/ .....20:56
alecuwhat's the way to take screenshots with mir?20:57
mhall119kedivess: how are you trying to install it?20:58
kedivessvia terminal app20:59
mhall119and apt?20:59
sergiusensboiko, pong20:59
kedivessyes, writing " apt-get install contactsimporter "21:00
mhall119kedivess: yeah that's probably not going to work21:00
mhall119first of all, if you flashed with ubuntu-system you can't use apt at all21:00
mhall119secondly, the app I was referring to is in the click store, not the apt repositories (as far as I know)21:00
mhall119bfiller: at one point we had some automatic contacts syncing, from Facebook I think21:01
mhall119back when we had a contacts lens21:01
mhall119any idea when/if we'll get that back and working with Google?21:01
dobeythere's a gmail contacts importer app thing in the click app store21:02
dobeyi don't know anything about a people scope coming back though21:02
mhall119dobey: I think it may have been unpublished or something, I don't see it now21:02
kedivessthat's what i did...i flashed the ubuntu-system. In Applications lens i can't find the click store....21:03
mhall119kedivess: there's a section in the Applications lens labeled "More suggestions"21:03
mhall119that's the store21:03
dobeymhall119: oh, indeed it's missing now21:03
kedivesssorry, but i can't find "more suggestions" section in the applications lens....21:05
dobeyare you not connected to the internet?21:06
dobeyon the device, that is21:06
kedivessi see only installed applications and dash plugins....21:06
dobeyin the applciations scope view, there is Installed at the top, then "More suggestions" and "Dash plugins" at the bottom21:06
dobeykedivess: if you're connected to the internet, tap on the "Search" at the top, and press space key, then backspace, on the keyboard21:07
dobeyor go to System Settings, Updates, and install the latest update and reboot21:07
kedivessthanks  <dobey> it worked...21:08
joachimvdaanybody a clue about getting Ubuntu Touch running on a Nexus 7 3G21:13
kedivessnow i see the "more suggestions" but there isn't the contactsimporter app...21:14
Tassadarjoachimvda: I don't think that's gonna happen until Ubuntu switches to version 4.3 or higher with their Android parts21:15
Tassadarbecause flo's proprietary libraries are only for 4.3 and, presumably, higher21:15
bfillermhall119: we'll have google snync for next release21:16
Tassadarjoachimvda: that's assuming you're asking about 2013 version of Nexus 721:16
joachimvdait is still the "old" model, just with 3G. There are reports of it working...21:16
Tassadarwell, then there should be a guide on wiki21:17
joachimvdait hangs when booting, staying at the google prompt21:17
joachimvdaI found a mailing list message about it working https://lists.launchpad.net/ubuntu-phone/msg04421.html21:18
Tassadarthen current daily image does not boot on N7, probably, it happens sometimes - it is not yet stable21:18
joachimvdabut the suggested fix script gives a problem, line "fastboot -c "console=tty1 datapart=/dev/mmcblk0p10" flash:raw boot zImage initrd.img" says "cannot load 'zImage': No such file or directory"21:19
AskUbuntuBlack screen after installing Ubuntu Touch. Galaxy Nexus (maguro) | http://askubuntu.com/q/35600021:20
kedivessi tryed to install some apps in "more suggestions", but they won't install...21:20
joachimvdaThanks for the help <Tassadar>, I will try again tomorrow21:22
sergiusensmhall119, seems there was a merge error21:23
sergiusensmhall119, in the music app21:23
sergiusensmhall119, oh, nevermind; hmmm21:24
sergiusensmhall119, any idea how this happened? 1805=== removed file 'manifest.json' in https://code.launchpad.net/~andrew-hayzen/music-app/toolbar-rewrite-1/+merge/18891721:27
kedivessinstead of ubuntu-system image, should i install the cdimage-touch ? Maybe in that way i can import google contacts and i will able to install other apps...21:27
mhall119sergiusens: no idea, maybe he merged from trunk at one point and deleted the file?21:31
sergiusensmhall119, I just asked balloons as he top approved21:32
=== fmasi_afk is now known as fmasi
=== fmasi is now known as fmasi_afk
kedivessany help?21:41
cjohnstonkedivess: system-image is the preferred image.21:49
kedivesstanks....but i can't add my google contacts and i can't install the "more suggestions" apps...21:50
mhall119kedivess: you can't install any of them?21:52
kedivessno21:52
mhall119does it give you a reason?21:53
kedivessit says "unable to write to /var/cache/apt/"21:54
mhall119don't use apt in the terminal21:55
mhall119click on the icon in the dash to install it21:55
kedivessi did click on the icon in the dash to install it, but it did nothing...22:00
mamenyakajjohansen, ping22:06
jjohansenmamenyaka: whats up22:06
mamenyakajjohansen, I found out why wasn't the 3.4 working22:07
mamenyakajjohansen, there were two defconfig files: fusion3_pollux_defconfig and cm_fusion3_pollux_defconfig22:07
jjohansenah22:08
mamenyakajjohansen, of course I edited the wrong one, but idk why is there two of these files here22:08
mamenyakajjohansen, so sorry about that22:10
jjohansenmamenyaka: np, I'm glad you figured it out22:11
mamenyakajjohansen, building seems to go fine, no errors22:11
=== jhodapp is now known as jhodapp|afk
OrokuSakiytou have to recreate var/cache/apt22:43
OrokuSakithen do apt-get update22:43
OrokuSakithen do apt-get install reinstall dbconf or something like that.. you will see the errors for the package when you try to update22:44
OrokuSakior upgrade, actually22:44
OrokuSaki@kedivess22:44

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!