[02:50] <extraymond> aia, but got stuck at finding b2g desktop for linux.
[02:50] <extraymond> Hi! I was trying to test tv version of gaia.
[02:50] <extraymond> But got stuck at finding b2g desktop for linux
[02:50] <extraymond> Anyone know how to do so?
[02:50] <extraymond> I'
[02:50] <extraymond> I'm following https://developer.mozilla.org/en-US/Firefox_OS/TVs_connected_devices
[02:51] <extraymond> ... I posted at the wrong forum... just ignore me...
[04:56] <ClashTheBunny> Hi, I'm curious if the current status of the Nexus 5 is actual on the Touch/Devices page.  It has said the same thing for quite a while.  Has there been no progress?  Where are hardware specific bugs kept?
[05:20] <duflu> ClashTheBunny: I don't have one, but can only see one bug open: https://bugs.launchpad.net/mir/+bugs?field.tag=nexus5
[05:39] <sturmflut-work> duflu: Searching for "Nexus 5" on bugs.launchpad.net yields a lot more results for me, https://bugs.launchpad.net/bugs/+bugs?field.searchtext=nexus+5
[05:40] <duflu> sturmflut-work: Yes, although I was going for hardware blockers, which would be mostly in Mir if at all
[05:42] <sturmflut-work> duflu: In my humble opinion things like https://bugs.launchpad.net/camera-app/+bug/1423045 are pretty big blockers too, even though they do not exactly seem to be hardware problems
[07:03] <cylonmath2> how can i use my google calendar on the phone?
[07:39] <pundir> hi, how tightly bound this mtp implementation https://launchpad.net/ubuntu/+source/mtp is to the Android USB gadget driver?
[07:40] <pundir> can it work with a generic MTP configfs driver as such?
[07:40] <pundir> rsalveti, ^^
[09:01] <sturmflut-work> Does the "Developer mode" setting change anything besides enabling ADB?
[09:02] <ogra_> well, enabling adb changes a lot :)
[09:02] <sturmflut-work> Haha
[09:02] <ogra_> but no, nothing beyond that
[09:05] <JamesTait> Good morning all; happy Tuesday, and happy Ferret Day! 😃
[09:17] <cylonmath2> u2
[10:07] <mcphail> I think it would be great if /usr/local was mounted rw and was not touched by system upgrades. It would be particulary useful if/when UT becomes multiuser. Would anyone agree?
[10:14] <mpt> tvoss, cyphermox: An architecture question. Most WPA Enterprise Wi-Fi authentication methods require one or more certificate files or a PAC file. What things do we need to provide for people to do that on Ubuntu Touch?
[10:15] <mpt>  I’m guessing (a) a file handler thingy so that when you download a cert/PAC file, System Settings lets you save it; (b) a method for the Wi-Fi auth dialog (which has many callsites, not just System Settings) to ask System Settings for the list of available certs/PAC files; (c) a method for Network Manager to extract the desired cert/PAC file from System Settings storage.
[10:15] <mpt> But not being an engineer I’m probably amusingly incorrect
[10:16] <tvoss> mpt, I'm not a 100% sure, but integration with content hub sounds like the easiest way forward. With that, the call site could just say: "Need certificate file of type...", and content hub takes care of resolving
[10:21] <mpt> tvoss, that’s a possible solution for (b), but the drawback is that the picker would be a secondary dialog on top of the auth dialog. I’d rather just have a menu inside the auth dialog.
[10:22] <tvoss> mpt, sure, we shouldn't assume a common shared location of data though. Thus I proposed the content hub
[10:22] <pitti> ogra_: hey Oliver, wie gehts?
[10:23] <mpt> tvoss, why can’t we assume a common shared location of these two particular types of files? It’s not as if any other app will do anything useful with them.
[10:23] <pitti> ogra_: do you happen to know how I can change the kernel boot params on mako?
[10:24] <tvoss> mpt, because we generally try to avoid common hard-coded paths around the system. Also: the content picker import thingy on top of the auth dialog would be a one-time operation iiuc
[10:27] <mpt> tvoss, you say hard-coded path, I say “method for asking for the list of available certs/PACs” :-)
[10:28] <tvoss> mpt, to me a method is a call to an interface :) not listing the contents of a probably well-known directory
[10:28] <mpt> exactly
[10:29] <ogra_> pitti, append-cmdline-arg ... and remove-cmdline-arg ... from the abootimg package
[10:29] <tvoss> mpt, so (b) and (c) could obviously be solved with services. (a) is a content-hub operation
[10:29] <mpt> tvoss, also, this probably won’t be the last time that the picker is annoyingly slow as a way of accessing files of a particular type. <http://photos2.appleinsidercdn.com/gallery/10509-2769-140917-Messages-2-l.png>
[10:29] <ogra_> pitti, just on the running phone ... (and reboot indeed)
[10:30] <tvoss> mpt, that's again a larger platform discussion that we need to have
[10:30] <pitti> ogra_: hm, "abootimg --help|grep cmdline" shows nothing; but thanks for the hint, I'll look in that direction
[10:30] <tvoss> or better: if we want to change the current behavior and alter the primitives we have introduced
[10:30] <pitti> ogra_: (might be different on dual-boot)
[10:31] <ogra_> pitti, ah, no idea about dualboot, we dont really support it
[10:32] <ogra_> pitti, with plain abootimg you would use: abootimg -u /path/to/boot/partition -c "cmdline=foo"
[10:33] <ogra_> but i guess in the case of dualboot thats hacked up differently so the boot partition might not be what you think
[10:33] <pitti> $ cat /proc/cmdline
[10:33] <pitti> console=ttyHSL0,115200,n8 androidboot.hardware=mako lpj=67677 user_debug=31 uart_console=enable lcd_maker_id=primary lge.hreset=off lge.reset=mode_reset gpt=enable lge.kcal=0|0|0|x lge.rev=rev_11 androidboot.emmc=true androidboot.serialno=01dfbefacd5202ac androidboot.bootloader=MAKOZ30f androidboot.baseband=mdm bootreason=recovery
[10:33] <pitti> ogra_: yeah, it looks fairly magic like that
[10:34] <ogra_> well, the cmdline you see is assembled from different places ... kernel, boot.img and bootloader each set parts of it
[10:34] <pitti> ogra_: anyway, I just took the plunge and installed systemd-sysv :)
[10:35] <pitti> it boots enough to get me a phablet-shell, so I have everything I need
[10:35] <ogra_> good
[10:39] <pitti> ok, the session crashes on startup, but the rest looks good; so running systemd on kernel 3.4 is nowhere near as complicated as I was afraid of originally
[10:39] <pitti> ogra_: ^ context: it seems $HIGH_ABOVE wants us to move touch to snappy, so we need to accelerate the init move, I guess :)
[10:39] <ogra_> pitti, not only HIGH_ABOVE :)
[10:41] <ogra_> pitti, well, the only tricky bits i see is the massive amount of upstart job hacks we have and the interaction with the container ... beyond this it shouldnt differ from server or desktop
[10:41] <ogra_> (and indeed the lack of kernel features)
[10:41] <pitti> ogra_: right; I was a bit afraid it required the cgroup reorg from 3.7, but turns out it doesn't
[10:41] <pitti> ogra_: it only needs kernel-side firmware loading as it was dropped from udev; but we don't do/need that on touch anyway
[10:42] <pitti> and even if we need to, that's really easy to put back
[10:42] <ogra_> well, not sure how cgroups and systemd would get into pla with ubuntu-app-launch here
[10:42] <ogra_> *play
[10:42] <pitti> ogra_: well, that's session, not system, isn't it?
[10:42] <ogra_> yeah
[10:42] <pitti> ogra_: AFAIK the biggest thing that's not just a JFDI is session upstart jobs listening for property changes
[10:42] <ogra_> but i dont know if UAL for example relies on any system upstart features
[10:43] <ogra_> well, there are session jobs that heavily depends on system job events too
[10:44] <ogra_> we need ot find some ways to bridge that gap again
[10:45] <pitti> ogra_: right, that's what I meant with the above property changes
[10:45] <ogra_> (i.e. mtp-server is rather broken due to that and just constantly running nowadays)
[10:46] <pitti> ogra_: we have some MPs like https://code.launchpad.net/~xnox/location-service/drop-sys-events/+merge/246050 to eliminate the need for the upstart system bridge
[10:46] <ogra_> for mtp and adb i have worked on a new udev based system that doesnt rely on system jobs ... but there are still some issues
[10:46] <ogra_> pitti, yeah, that broke the world
[10:47] <pitti> ogra_: ok, so I think we are by and large on the same page about the outstanding issues
[10:47] <ogra_> yeah
[10:47] <pitti> $ cat .xsession-errors
[10:47] <pitti> init: unrecognized option '--user'
[10:47] <pitti> haha
[10:47] <pitti> that might explain why the user session crashes :)
[10:47] <ogra_> fun
[10:48] <ogra_> i didnt even know we create that file :P
[10:48] <ogra_> is the container running properly btw ?
[10:48] <pitti> ogra_: yes, I already fixed that during the porting sprint
[10:48] <ogra_> (there should be /init in your processlist)
[10:49] <pitti> ogra_: I had the emulator working under systemd all the way up to unity8
[10:49] <pitti> now I wanted to try on a real device
[10:49] <ogra_> oh, nice !
[10:50] <pitti> ogra_: http://paste.ubuntu.com/10989309/ FYI
[10:50] <ogra_> sweet !
[10:50] <pitti> lxc-ls also says "android  RUNNING  -     -     -       NO"
[10:51] <pitti> ogra_: so, next I'll figure out why the session crashes right away
[10:51] <pitti> a phone is a tad more useful without just a black screen :-P
[10:51] <ogra_> wow, i wasnt aware how much crap we run in the container ... that could use some cleanup
[10:52] <pitti> is it? I thought we'd still use most of that
[10:52] <ogra_> sdcrad service is definitely unused
[10:52] <pitti> (ok, no idea what qcks is..)
[10:52] <ogra_> debuggerd and installd too
[10:53] <ogra_> hmm, i thought healthd was drooped ages ago
[10:53] <ogra_> but probably ricardo just disabled its logging
[10:53] <pitti> well, there's a wholly new wily release to upload fixes to :)
[10:54] <ogra_> yeah, not sure we care enough for N4 to optimize that :)
[10:54] <ogra_> its a workhorse for automation ... as long as it fulfills that purpose ...
[11:00] <pitti> yay, that was an exceptionally simple fix
[11:00] <ogra_> pitti, oh, and there is a UOS session about snappy-phone today ..
[11:00] <pitti> unity8 starting fine now
[11:08]  * pitti files https://code.launchpad.net/~pitti/ubuntu-touch-session/systemd/+merge/258240
[11:10]  * ogra_ approves
[11:13] <pitti> ok, now port /etc/init/ofono.override and I should have a somewhat usable phone :)
[11:18] <pitti> rockin', works
[11:19] <pitti> 3G data and telephony
[11:20] <pitti> ogra_: lxc-android-config is good old dput? no train?
[11:20] <ogra_> yeah, but you need to check the train if some silo blocks it
[11:21] <ogra_> if so, yu need to hold back til that silo is landed
[11:23] <pitti> ogra_: yeah, I figure we need to wait anyway until wily is truly open and the CI train moved to it, etc.
[11:23] <pitti> ogra_: I just wondered about doing an MP vs. just attaching a patch, etc.
[11:23] <ogra_> yeah, just dput
[11:23] <pitti> (FTR, there is a silo)
[11:23] <ogra_> i think rsalveti has a landing pending for that package though
[11:23] <ogra_> yeah
[11:23] <pitti> not urgent
[11:23] <ogra_> for tethering and WIFI AP support
[11:24] <pitti> I'll just keep collecting bugs and patches/MPs at https://bugs.launchpad.net/ubuntu/+bugs?field.tag=systemd-boot-touch
[11:24] <ogra_> though i'm not even sure if there will be wily silos at all for the phone
[11:24] <ogra_> sounded to me like the phone is supposed to stay on vivid
[11:24] <pitti> yeah, I understand we want to keep it vivid based, and then move to something snappyish for 16.04
[11:25] <ogra_> so not sure how wily will get into play with all this
[11:25] <pitti> well, we'll need some kind of vivid overlay, presumably we'll do a new RTM release?
[11:25] <ogra_> we have a vivid overaly PPPA already
[11:25] <ogra_> err
[11:25] <ogra_> PPA
[11:25] <pitti> but regardless where we'll upload to, I can still go ahead and collect patches/MPs there
[11:25] <ogra_> (not private :) )
[11:25] <ogra_> yeah
[11:26] <ogra_> our landing processes are such a mess :/
[11:26]  * ogra_ wishes we had something unique that suits all sides ... instead of multiple different ways to land something
[11:26] <ogra_> ... always stepping on each others toes ...
[11:27] <robin-hero> hi all! quick question. Does Facebook support notifications on the bq device with push client? Because I don't get notifications
[11:28] <davmor2> robin-hero: no currently
[11:28] <davmor2> not even
[11:29] <robin-hero> then just Telegram supported, right?
[11:29] <davmor2> robin-hero: and twitter
[11:30] <robin-hero> oh, thanks.I mixed with it :)
[11:31] <davmor2> robin-hero: you need a @<you> in twitter or a direct message for twitter to ping
[11:32] <robin-hero> And make it in Facebook is more difficulty than Telegram?
[11:33] <davmor2> robin-hero: I think there were issue with permissions on facebook so it was dropped.
[11:38] <ogra_> davmor2, i wonder why we expose facebook in the notification settings then ... we should hide it if it doesnt work
[11:40] <davmor2> ogra_: remnants of a happier time ;)  No idea to be honest maybe pete-woods might know more I think this is you right pete?
[11:40] <ogra_> well, sounds liek a bug to me :)
[11:41] <robin-hero> Remainders are listed there too.
[11:41] <robin-hero> *Remainder
[11:41] <pete-woods> davmor2: unfortunately I am responsible for none of facebook, notifications or online accounts
[11:42] <davmor2> pete-woods: why the hell did I think it was you then I'll go back to sleep :)
[11:42] <pete-woods> davmor2: I thought MacSlow|lunch was the notifications guy
[11:42] <pete-woods> davmor2: and mardy is the online accounts person
[11:42] <pete-woods> although maybe we need the system settings guy
[11:43] <pete-woods> really don't know for sure
[11:43] <davmor2> pete-woods: yeah but I thought it was the indicator that triggered them via messaging indicator
[11:43] <pete-woods> davmor2: oh right, well if it's indicators then maybe charles might be the right person
[11:44] <pete-woods> I only really do the network indicator
[11:44] <davmor2> pete-woods: yeah not sure about the pulling in bit though E:TOO_MANY_MOVING_PARTS
[11:47] <davmor2> ogra_: I will return to my default, it's your fault :P  Fix it :D
[11:47] <ogra_> well ...
[11:57] <seb128> ogra_, davmor2, the notifications entry for facebook is coming from the facebook webapp click
[11:58] <seb128> it's a push-helper hook from the click
[11:58] <ogra_> seb128, right, well, the apps shouldnt enable that helper if they cant use it
[11:58] <seb128> ogra_, right, just pointing to the component that need to be changed ;-)
[11:58] <ogra_> yeah
[12:00] <ogra_> pitti, oooh, i didnt know the main blocker was firmware loading with 3.4
[12:01] <ogra_> if it is just that ... we even forecfully diverted the udev rule for firmware handlin to leave all that to the container
[12:25] <robin-hero> is it possible that my phone discharging faster with turned off wifi/mobil data than with everything are turned on?
[12:27] <cwayne> robin-hero, i recall seeing such a bug somewhere, that discharge was faster in flight mode for some reason
[12:28] <robin-hero> cwayne: both tested without flight mode
[12:29] <robin-hero> with turned on wifi/mobile data discharing ~1% at night, with turned off it is about 4%.
[12:39] <cyphermox> mpt: tvoss: sounds indeed like something that could be solved by content-hub, even when it comes to displaying a list of known certificate data objects for the user. presumably there wouldn't be many of them
[12:40] <cyphermox> this of course depends on content-hub being able to return a list of certificate objects from wherever they may be stored when copied to the system via MTP or otherwise
[12:41] <tvoss> cyphermox, the content hub is file based, but that isn't an actual limitation
[12:42] <cyphermox> tvoss: what do you mean?
[12:42] <cyphermox> file based is fine, the certificates are files
[12:42] <tvoss> cyphermox, yup, the list is displayed by a different app, though
[12:42] <cyphermox> I'm saying "object" here because it can be varying types of files
[12:42] <tvoss> cyphermox, yup, I think we are in agreement here
[12:42] <cyphermox> ok
[13:21] <sturmflut-work> Is there any documentation on "ubuntu-device-flash core"?
[13:32] <elopio> bfiller: meeting?
[13:36] <sturmflut-work> Hmmm. "ubuntu-device-flash touch --bootstrap --developer-mode --password=1234 --recovery-image /tmp/recovery.img" does not work on the production bq E4.5 device. It flashes the device, using the temporary ADB-enabled recovery image, but when the device comes up with the welcome wizard ADB is not enabled.
[13:45] <jgdxx> pitti, hey, how's it going?
[14:12] <dobey> sturmflut-work: is it not enabled, or is it blocked because it's not enabled until the device has been set up (seems like a reasonable expectation, but i am not sure if it works that way or not)?
[14:16] <davmor2> sturmflut-work: you haven't set a channel so it won't know where to get the image from
[14:30] <seb128> kenvandine, hey, do you have any recommended documentation/tutorial/example about content-hub hacking? also can you easily test c-h code on an u7 env?
[14:31] <kenvandine> it works on u7, but not as well as on the phone
[14:31] <kenvandine> mostly just app focus isn't as clean
[14:32] <seb128> kenvandine, ok, good ... what about the first half of the question? ;-)
[14:32] <kenvandine> seb128, d.u.c has docs and i have some hub examples in my +junk that are more detailed
[14:32] <seb128> kenvandine, thanks
[14:32] <kenvandine> https://code.launchpad.net/~ken-vandine/+junk/hub-exporter
[14:32] <kenvandine> https://code.launchpad.net/~ken-vandine/+junk/hub-importer
[14:32] <kenvandine> https://code.launchpad.net/~ken-vandine/+junk/hub-sharer
[14:32] <seb128> kenvandine, great
[14:50] <pitti> hey jgdx; pretty well, thanks! how about yourself?
[14:51] <sturmflut-work> dobey: Well, --developer-mode is supposed to be used to automatically install the device for automated testing, so I would expect that ADB gets enabled while the welcome wizard is displayed. The Touch Wiki even has an entry on how to do this, https://wiki.ubuntu.com/Touch/Testing#Preparation
[14:51] <jgdx> pitti, I'm well, thanks. Thing quited down for you yet? :)
[14:51] <pitti> jgdx: well, -ish; release sprint, post-release high-urgency regression bugs, now UOS :)
[14:51] <pitti> jgdx: I still have the two NM mock pull requests on my TODO list
[14:51] <sturmflut-work> davmor2: You don't need to specify a channel, it will automatically pick the default one for the attached device.
[14:51] <pitti> in case you want to ping about those
[14:52] <jgdx> pitti, :) Okay. Any estimate? We're really eager to release it
[14:52] <jgdx> *it being the thing depending on the new mocking
[14:52] <pitti> jgdx: is it urgent?
[14:52] <pitti> jgdx: oh? you can always modify an existing mock or ship a local one?
[14:52] <pitti> (and you have to for vivid anyway
[14:56] <jgdx> pitti, that's fine as long as it lands in your repo and gets a review from you
[14:56] <pitti> jgdx: yes, absolutely; I just wasn't aware that it's blocking something (and it shouldn't)
[14:57] <pitti> sorry for the delay! (too much stuff is happening.. :-) )
[14:57] <jgdx> pitti, dbusmock was one of four blockers :) Now it's one of three.
[16:59] <cylonmath2> how i can switch to my other sim?
[17:00] <ogra_> cylonmath2, in the system settings or in the dialer app
[17:01] <ogra_> (first permanent, second temporaray i think)
[17:02] <cylonmath2> thats the problem
[17:03] <cylonmath2> i could not find the setting
[17:03] <cylonmath2> ahhh okay i guess i gotta restart
[17:03] <ogra_> in the cellular settings you pick the default
[17:03] <ogra_> oh, yeah, i dont think it supports hotswapping of SIMs yet
[17:04] <cylonmath2> thats not an urgent feature i would say
[17:04] <ogra_> yeah, i guess you dont swpa them out very often
[17:04] <ogra_> *swap
[17:05] <cylonmath2> thx! it works now
[17:05] <ogra_> :)
[17:07] <elopio> rickspencer3: the argument is object_name, not objectName.
[17:07]  * elopio updates the docs.
[17:08] <rickspencer3> whoops
[17:08] <rickspencer3> :)
[17:08]  * rickspencer3 tries
[17:08] <elopio> balloons: any idea why the signature of the method shows instance, *args, **kwargs ?
[17:08] <elopio> this is the definition: def click_element(self, object_name, direction=None):
[17:08] <elopio> seems like sphinx going nuts.
[17:09] <balloons> elopio, right, that's what I was trying to get him to do.. but anyone, not sure
[17:09] <balloons> I can rebuild the docs
[17:10] <balloons> let me just do a quick rebuild and see what they say
[17:10] <rickspencer3> seems like it shouldn't be "objectName" in some places, and "object_name" other places
[17:11] <rickspencer3> but, in any case, it works now ;)
[17:11] <balloons> elopio, I tried to simplify and fix rick's __init__.py and launching: lp:~nskaggs/+junk/flash2
[17:12] <elopio> that should probably be an alias. Everywhere we take object_name we should accept objectName.
[17:12] <elopio> it's weird to mix the qml code style with the python code style.
[17:12] <balloons> indeed.. camelcase style vs underscores
[17:13] <balloons> can we do a highlevel alias for that?
[17:13] <elopio> balloons: not really. highest level we can do is a decorator.
[17:14] <elopio> I would just add a statement everywhere in the api where we accept object_name.
[17:15] <balloons> elopio, the new build looks the same
[17:15] <balloons> err, wait.. no args, kargs is gone
[17:15] <elopio> balloons: yes, we need to dig into sphinx to see wtf.
[17:15] <balloons> let me push this
[17:16] <balloons> the parameters of objectName is still off
[17:16] <balloons> but again, honestly you shouldn't need to use the object_name="blah" style
[17:17] <elopio> what would be cool is to receive any *args and **kwargs, and pass them to autopilot.
[17:17] <elopio> we can do it backwards compatible by the object_name alias objectName.
[17:17] <elopio> good ideas to play during the following weeks.
[17:17] <balloons> ok, page is updated
[17:18] <balloons> let's get bugs open and tagged then
[17:21] <elopio> balloons: https://code.launchpad.net/~canonical-platform-qa/ubuntu-ui-toolkit/fix1451945-click_element_doc/+merge/258287
[17:21] <elopio> for now.
[17:24] <balloons> rickspencer3, I changed your __init__.py around to making launching saner and added in a helper class for the app: lp:~nskaggs/+junk/flash2
[17:24] <balloons> that's more or less how I would setup things
[17:25] <balloons> merge it and see if you like / makes sense
[18:04] <dobey> mzanetti: hey, is there a bug for that weird netowrk routing issue you were talking about last week at the sprint? i tried vivid and hit the same issue
[18:05] <mzanetti> dobey, there are a couple of bugs, yes. let me try to find them
[18:06] <mzanetti> dobey, this one might be yours: https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1307981
[18:08] <mzanetti> dobey, read the last comment though
[18:08] <dobey> i'm not sure that is the issue i was seeing
[18:11] <dobey> for me, after a while, the route would just be completely empty. and turning on wifi would have a route, but only for LAN IPs and no default route for 0.0.0.0
[18:12] <dobey> and turning on flight mode, the [H] icon would stay in the indicator
[18:12] <ogra_> thats an ancient bug ...
[18:12] <ogra_> you won the lottery :)
[18:12] <ogra_> (no route at all is a rare occurence ... there are multiple bugs btw)
[18:35] <dobey> it wasn't rare for me :)
[20:12] <brunch875> oooo will we be starting to get convergence with wily?
[20:12] <ogra_> kind of
[20:12] <brunch875> nniiiijsse
[20:13] <ogra_> might not be available before wily + 0.5
[20:13] <ogra_> but the work starts right now :)
[21:19] <speck84> Hiya all.
[21:19] <speck84> I currently working on a HTML5 app does it somebody know how to setup the app to not make it resizable?
[21:28] <dobey> speck84: #ubuntu-app-devel is probably a better place for app dev questions :)
[21:28] <speck84> ok thx
[23:28] <extraymond> Hi! everyone! It seems that I can't use gps on v186 on nexus4 (mako), is that normal?