[00:44] new touch OS coming 10/17/2013 | http://askubuntu.com/q/354617 [00:45] What kind of usable state it will be in... that's another matter entirely. === chihchun_afk is now known as chihchun === Namidairo`bnc is now known as Namidairo === anders3408 is now known as anders3408|afk === ryukafalz_ is now known as ryukafalz === Namidairo is now known as Namidairo`bnc [03:56] why would phablet-bootstrap delete android AOSP files I downloaded? === Namidairo`bnc is now known as Namidairo [04:21] OrokuSaki: hello === Namidairo is now known as Namidairo`bnc === Namidairo`bnc is now known as Namidairo [04:40] Is there a way to restore the TP to one partition without resorting to Doctor-ing? === chihchun is now known as chihchun_afk === cul- is now known as cul === chihchun_afk is now known as chihchun === joely is now known as Guest17487 === deathcrawler is now known as LockeAnarchist [06:34] good morning === alan_ is now known as Guest52790 [07:21] hello!, anyone knows when will be disponible the desktop part of ubuntu via HDMI? [07:31] Guest31817, probably 14.04, but more likely 14.10 [07:36] Laney: hi! Thanks for fixing the libexec bug in signon-ui [07:37] Laney: can you please also assign this to yourself and resolve it as approriate? https://bugs.launchpad.net/ubuntu/+source/signon-ui/+bug/1235337 [07:37] Ubuntu bug 1235337 in signon-ui (Ubuntu) "signon-ui ships files in a wrong directory" [High,New] [08:00] Morning === fmasi_afk is now known as fmasi [08:04] mardy: hey, seems Fix Committed/Released to me (not sure if it's uploaded) [08:12] jodh: hey! [08:13] jodh: there's a pretty serious bug logged by ogra_ on touch images: upstart eats memory like crazy; he didn't spot anything piling up in the list of apps, or crazy log files, so would be worth your investigation I think [08:13] ogra_: what's the bug id? [08:13] bug 1235649 [08:13] bug 1235649 in upstart (Ubuntu Saucy) "session upstart leaks massive amounts of memory on Ubuntu Touch" [Critical,Confirmed] https://launchpad.net/bugs/1235649 [08:44] Good morning all; happy Habitat Day! :-D === psivaa-afk is now known as psivaa [09:19] dbarth: would you think someone could help debug the issue Robert was seeing in https://bugs.launchpad.net/ubuntu/+source/cordova-ubuntu-tests/+bug/1235393 ? [09:19] Ubuntu bug 1235393 in cordova-ubuntu-tests (Ubuntu) "autopilot tests crash halfway through." [Undecided,New] [09:19] dbarth: rather than him doing the debug thing etc. [09:23] lool: looking [09:24] lool: yeah [09:24] lool: i need to merge in his packaging change, and then update the test suite to only test the supported apis anyway === nerochiaro_ is now known as nerochiaro [09:24] lool: but feels like an installation problem; anyway, mazim and i are on it === ritz is now known as ritz|away === oSoMoN_ is now known as oSoMoN [09:36] anyone can point me in the right direction on how to enable MIR in current phone builds ? [09:38] nerochiaro, touch /home/phablet/.display-mir ... reboot [09:39] ogra_: fyi - cpu spinner: bug 1236247. [09:39] bug 1236247 in touch-preview-images "brcm_patchram_plus runs at 100% CPU on nexus 7 (grouper) on build #82." [Undecided,New] https://launchpad.net/bugs/1236247 [09:39] jodh, thats one for cyphermox [09:39] ogra_: ack. === vrruiz_ is now known as rvr [10:01] Morning all [10:05] dbarth: ok === hikiko is now known as hikiko|lunch [10:23] seb128, let's talk about bug 1233682 [10:23] bug 1233682 in ubuntu-system-settings (Ubuntu) "Updates panel should not say Ubuntu Phone" [High,New] https://launchpad.net/bugs/1233682 [10:24] seb128, is the phone image "Ubuntu 13.10" in any sense? [10:26] mpt, we are going to have a v1 release for 13.10, and it's the same base OS/built from the Ubuntu archive [10:26] mpt, so yes, it's just a very different image than the desktop one [10:28] seb128, so if it's "v1", what is it "v1" of? :-) [10:28] mpt: we have to updaters now should be called application updater and the other system or something? Also do you know if there is a design to allow for auto updates with the application updater? [10:28] We could have a separate build number, that's no problem [10:28] mpt: official marketing will be "Ubuntu for Phones" [10:28] mpt, we call it v1, let's say it's a phone image built from the ubuntu 13.10 archive [10:28] mpt: so i guess "Ubuntu for Phones v1 (rev. 75)" or some such. [10:29] So no 13.10 anywhere? [10:29] mpt: i don't know, I would have thought it would be 13.10 none-the-less. [10:29] mpt: but do mind that very shortly after 13.10 (e.g. within 1-3 months) we will roll over all phones to the next development release (T-series) [10:30] mpt: thus it will be a very "short" support release. (Little Term Support) [10:30] (phones only that is) [10:31] So the one based on 14.04 won't be an LTS [10:32] mpt: 14.04 LTS status for products hasn't been decided yet. We will know that once we get there. [10:33] mpt: i assume phone will go LTS once we ship it pre-installed on devices. [10:33] which may or may not co-inside with 14.04 release date on desktop/server. [10:33] My only point in asking about LTS would be to tell whether the Y.MM version number was relevant at all to the phone image [10:35] mpt: the points that are relevant on the phone is: what channel and what image # one has installed. Cause in November we will have: stable, testing, development channels which can have same or different image # [10:36] mpt: stgraber/barry from system image updater can tell you actual names of channels, and image formats. [10:37] mpt: cause unlike desktop/server one will not be able to selectively upgrade, and only choose which channel of updates to subscribe to. [10:41] xnox, do you have a reference for the "Ubuntu for Phones" branding? [10:41] mpt: informal quotation [10:43] mpt: you have webteam in the office who are preparing website updates around you ;-) [10:48] ogra_: bug 1236292 # seen that on maguro? [10:48] bug 1236292 in indicator-sound (Ubuntu) "Using volume buttons on phone triggers spurious popup" [Undecided,New] https://launchpad.net/bugs/1236292 [10:50] popey: ogra_: confirmed over the weekend you get a popup saying indicator sound [10:50] will look at 83 when it has finished flashing [10:54] popey, ogra_ : http://ubuntuone.com/2vEr9fbeFbErYJrNrn5hy5 [10:54] davmor2: ta === chihchun is now known as chihchun_afk === hikiko|lunch is now known as hikiko === chihchun_afk is now known as chihchun [11:13] seb128, xnox: So where does the "v1" come from? What determines when it changes to 1.1 or 2.0? [11:14] popey, ogra_ am I mistaken or should the new image be on mir now, or did it get postponed? [11:14] probably tonight [11:14] bug 1235649 [11:14] bug 1235649 in upstart (Ubuntu Saucy) "session upstart leaks massive amounts of memory on Ubuntu Touch" [Critical,Confirmed] https://launchpad.net/bugs/1235649 [11:14] and there are a few test issues still i think [11:15] mpt: v1 was widely mentioned and used in the PRD spreadsheet w.r.t. which features go into first public release aka "v1" vs "later" [11:15] ogra_: ah thanks, that might explain why my maguro just stops working after a while then [11:15] xnox, I mean in the code. [11:15] mpt: but no idea if that's purely internal planning. [11:15] davmor2, yeah [11:15] mpt: none, as far as I know. === MacSlow is now known as MacSlow|lunch [11:15] davmor2, i think jodh and jibel are on it [11:17] seb128, so how about "Ubuntu 13.10 (r79)"? [11:17] mpt: pragramatically we ship os-release and lsb-release files that define the installed machine as "Ubuntu", code name "Saucy Salamander", version "13.10", human name "Ubuntu 13.10" [11:18] mpt: which is the same across all types of Ubuntu installation. [11:18] xnox, well, you would rather want the build stamp here too i guess [11:19] which lives in /var/log/installer/media-info [11:19] but that wont help on touch, wheer you also want the system-image number [11:20] Is the system-image number the one that's currently 75 or so? [11:20] Or is that something else entirely? [11:21] well, 82 but yes [11:21] root@ubuntu-phablet:/# system-image-cli -i [11:21] current build number: 83 [11:21] device name: maguro [11:21] channel: saucy-proposed [11:21] last update: 2013-10-06 12:08:27 [11:21] version version: 83 [11:21] version ubuntu: 20131005 [11:21] Error: Ubuntu bug 20131005 could not be found [11:21] version device: 20131003.2 [11:22] silly bot :P [11:22] i assume you would like some of this output in the UI [11:22] ogra_: no, we do not want media-info, as that's only original installation media and is optional. I know it's updated on RO/system-image updates, but it isn't on desktop/servers. [11:23] xnox, it isnt updated [11:23] (on touch) [11:23] ogra_: isn't that file part of the read-only image?! [11:23] the system-image info is, media-info always stays the same [11:23] ogra_: interesting. [11:23] that gives you the info from which initial rootfs you installed [11:24] (same as for d-i) [11:24] ogra_: so yeah, we don't care about media-info, as that's the original installation media, which may not be what is _currently_ running on the device. [11:24] right [11:24] ogra_: but very nice semantics, I like how the meaning was preserved =) [11:24] you want the system-image info [11:24] mpt, that works for me, it means the version is going to change with every bug fix update rolled out, but that's ok [11:25] root@ubuntu-phablet:/# grep version_detail /etc/system-image/channel.ini [11:25] version_detail: ubuntu=20131005,device=20131003.2,version=83 [11:25] that is where system-image-cli gets its info from [11:25] seb128, that's what it's supposed to do ... So someone can refer to this screen in saying "Problem X occurred in r84 but not in r83" [11:26] mpt, ok, that seems fine to me then [11:26] (ubuntu means the rootfs, device means the android build version, version means system-image ID) [11:28] tvoss_, can you take a look at bug 1235419 ? i wanted to wait for your input before unseeding [11:28] bug 1235419 in ubuntu-touch-meta (Ubuntu) "seed should not include ubuntu-location-service-examples" [Medium,Confirmed] https://launchpad.net/bugs/1235419 [11:28] ogra_, hey there. Looking [11:29] ogra_, I need to fix the packaging then ... [11:29] go ahead then :) [11:29] ogra_, I will find time to look into it tomorrow my morning [11:29] ogra_: examples should typically go into "sdk" package, as it's mostly interesting for developers, and not be seeded on the rest (sdk-libs, sdk-libs-dev, touch) [11:29] ogra_, just to give you an eta [11:29] k [11:29] xnox, well, its not clear its actually an example :) [11:30] ogra_: ah =) [11:30] examples shouldnt ship upstart jobs either :) [11:30] seb128, xnox, ogra_: Okay, spec updated. [11:30] mpt, thanks [11:31] looks good [11:35] mpt: I still think we need to list the mobile phone number for the device on that page [11:38] ogra_, popey: it's a side effect of the notification being added back for desktop when you wheelscroll over the indicator icon, Lars is looking at disabling that on the phone profile, that should fix it [11:39] seb128: great! [11:39] I knew you'd know, but decided not to disturb you directly ㋛ [11:39] ;-) [11:43] hi everybody. It's a know bug that with the build marked as current the wifi scan doesn't work ? [11:43] on nexus 4 [11:43] synny: what build id is this? system-image-cli -i [11:43] should work from the indicator [11:44] works here [11:44] the latest build on the ftp. Actually i'm on android [11:44] oh ok. I'll try to reinstall [11:44] how are you installing? [11:44] with multirom [11:45] what image are you downloading? [11:45] http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/current/ [11:45] ok [11:46] hi === fmasi is now known as fmasi_afk [11:46] @ogra what do you mean by from the indicator ? [11:46] synny, from the panel [11:46] can you port a device to Ubuntu touch even if it has no cm ROM? [11:47] in system settings ? [11:47] I have access to bottom.img, so on so forth [11:47] synny: no, drag down from top right corner [11:47] synny, from the top panel [11:47] over the wifi one [11:48] sorry I meant boot.img [11:48] hum stange, when I've tested it could'nt drag the wifi symbol, there was only the location ... [11:48] linux_unix-10, well, you definitely need the full source code for the android bit [11:48] I try to reinstall it so [11:49] MacSlow|lunch: Hey, how far did you get on the Mir input issues? [11:49] lool, Saviq beat me to it [11:50] lool, but we've discovered a bug in unity-mir [11:50] lool, but there's an issue with InputFilterArea itself, which is used to solve this, that needs to be sorted out first... [11:50] lool, that we need to tackle for this to get fixed [11:50] that ↑↑ [11:50] another issue too, I've delete the pin lock from security menu in android but I wasn't able to connect to cellular network and make call ... It's the right method that I've done ? [11:51] Saviq: Ok; so I guess I should prioritize landing the ubuntu-mir stuff then [11:51] lool, yeah... and afaik Gerry is working on that InputFilterArea issue [11:51] Saviq, did you check unity-mir trunk ? there is a bunch of stuff not landed yet [11:51] ogra_, yeah, unrelated I'm afraid [11:51] k [11:51] ogra_: there isn't much not landed [11:51] just the bring to front stuff I think [11:52] ok [11:52] MacSlow|lunch: Ok; thanks for the update [11:52] i just remembered there was something so i thought i should mention it [11:52] can you decompile the kernel and binaries? [11:52] Saviq: Do we know how many different issues we're actually looking at input-wise? [11:52] MacSlow|lunch, Gerry is on QtDevDays starting today, I'll get an update, but probably someone else will need to take over [11:52] it seemed they could possibly be related [11:52] lool, two, unrelated [11:52] Saviq, doh... right... forgot that [11:52] does anyone know how to deal with the following error: /usr/bin/system-settings: error while loading shared libraries: libSystemSettings.so.1: cannot open shared object file: No such file or directory [11:52] lool, https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1233245 [11:52] Ubuntu bug 1233245 in unity-mir "[mir] key events not working through input devices" [Undecided,In progress] [11:52] lool, https://bugs.launchpad.net/unity-mir/+bug/1235383 [11:52] Ubuntu bug 1235383 in unity-mir "InputFilterArea not always calculating geometry with respect to scene" [Critical,Confirmed] [11:53] thanks [11:53] linux_unix-10: you can't "port" android -> ubuntu, like you "port" android -> android. If you don't have source, then give up [11:53] lool, well, that one https://bugs.launchpad.net/unity-mir/+bug/1233245 is most probably related [11:53] it's the same one as the first one [11:53] lool, or well, the same, as the first one [11:54] lool, hmm I might've not filed a separate one then ;) [11:55] that's ok, if it's the same bug :-) [11:55] Saviq: I was looking at https://bugs.launchpad.net/ubuntu/+source/mir/+bug/1234903 and https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1233245 [11:55] Ubuntu bug 1234903 in telephony-service (Ubuntu Saucy) "[mir-only] On incoming calls the greeter remains and dialer-app is not started" [Critical,In progress] [11:55] Ubuntu bug 1233245 in unity-mir "[mir] key events not working through input devices" [Undecided,In progress] [11:56] Saviq, lool: btw... regarding Volume Up/Down... with the latest image I see indicator-sound triggering a notification (probably a left-over confirmation notification from desktop NotifyOSD) [11:56] MacSlow|lunch: Yes; it's also a completely broken one [11:56] MacSlow|lunch: how do we fix this? [11:56] MacSlow|lunch, yeah, not supposed to be there at all [11:56] MacSlow|lunch, see bug #1236292, it's being fixed [11:56] bug 1236292 in indicator-sound (Ubuntu) "Using volume buttons on phone triggers spurious popup" [Undecided,Confirmed] https://launchpad.net/bugs/1236292 [11:56] Saviq, lool: there are no confirmation notifications on the Phone [11:56] MacSlow|lunch, yup [11:57] Saviq, MacSlow|lunch: when do we get sync notifications in unity8? ;-) [11:57] Saviq, lool: just remove it triggering any notification [11:57] seb128, they're not supposed to be there on touch at all [11:57] MacSlow|lunch: what package am I supposed to remove? [11:57] MacSlow|lunch, it's already being adressed, please don't put more people to dup work [11:57] seb128, there's not use/Design for those [11:57] ah [11:57] lool, re: phone app not starting, will have to investigate - didn't see that one [11:57] seb128 has it alraedy [11:58] Saviq, what's the mechanism to have e.g sound level bars on screen feedback? [11:58] Saviq: I dont know whether it's related [11:58] lool, likely not, and I can see boiko has something there [11:58] Saviq, any other phone OS tell you what's the value so you can have a clue where you are [11:58] lool, no... indicator-sound just needs to be patched to not trigger a notification at all [11:58] seb128, it's there in the panel [11:58] MacSlow|lunch, IT'S ALREADY DONE [11:58] seb128, oh... ok :) sorry [11:58] seb128, not supposed to do anything else atm [11:58] Saviq, it's not really easy to read on suck a small icon [11:59] seb128, nothing else planned currently [11:59] Saviq, ok, shame [11:59] ogra_: so the upstart thing is broadcom patchram? [12:00] lool, ? [12:00] ogra_: I thought LP #1236247 was what James and you discussed [12:00] lool, i dont think they are related [12:00] Launchpad bug 1236247 in bluetooth-touch (Ubuntu) "brcm_patchram_plus runs at 100% CPU on nexus 7 (grouper) on build #82." [Undecided,New] https://launchpad.net/bugs/1236247 [12:00] ogra_: Ok; I didn't think so either [12:00] ogra_: which is why I was curious [12:00] james just pointed it out [12:01] jodh: did you reproduce the upstart issue ogra_ filed? [12:01] at least i didnt understand jodh to mean it would be related [12:05] ogra_: ok, was just the only traffic I saw on this [12:05] well, there was some traffic on the bug [12:05] jibel is collecting an strace [12:06] ogra_, I attached it a minute ago [12:06] jodh, ^ [12:06] :) [12:06] hey guys, I swipped down from the top panel and in network part I've only flight mode and cellular settings... do you think that myissue is related to multirom ? [12:07] synny, could well be, or to the rw image we dont ose them anymore [12:07] *use [12:07] (and we also dont test them anymore, so there might be issues that dont show up on ro images) [12:09] ok, I will flash the zip without multirom. If I flash these saucy-preinstalled-touch-armhf.zip and saucy-preinstalled-touch-armel+mako.zip does it will wipe my sdcard ? [12:10] why you don't test the image anymore ? [12:10] we only test the system-image [12:11] which has a completely different filesystem setup to the zips from cdimage [12:11] ah ok [12:12] so if I flash the system-image, I guess that this will erase my internal storage ? [12:13] yes === psivaa is now known as psivaa-afk [12:14] Okay, thanks for your help ! Have a nice day ;) === MacSlow|lunch is now known as MacSlow [12:15] jibel, ogra_, jodh: seems to be doing some dbus traffic with powerd [12:16] or rather, triggering some system events related to power [12:16] 1592 sendmsg(21, {msg_name(0)=NULL, msg_iov(2)=[{"l\4\1\1\256\0\0\0b\5\0\0a\0\0\0\1\1o\0\23\0\0\0/com/ubu"..., 120}, {"\26\0\0\0:sys:cpu-device-online\0\0\216\0\0\0"..., 174}], msg_controllen=0, msg_flags=0}, MSG_NOSIGNAL) = 294 [12:16] 1592 sendmsg(23, {msg_name(0)=NULL, msg_iov(2)=[{"l\4\1\1\31\3\0\0\310\0\0\0a\0\0\0\1\1o\0\23\0\0\0/com/ubu"..., 120}, {" \0\0\0:sys:power_supply-device-cha"..., 793}], msg_controllen=0, msg_flags=0}, MSG_NOSIGNAL) = -1 EAGAIN (Resource temporarily unavailable) [12:16] I wonder if it's basically the device waking up and sleeping super often [12:16] and that causing a lot of upstart system events [12:17] onTryRegister [12:17] thats what i get in my dbus.log [12:18] once or twice a second [12:18] no idea where it comes from :/ [12:18] popey, hey, is the rssreader supposed to be called Shorts only in it's friendly name? [12:18] sergiusens: shorts is technically one view in the rss reader [12:18] ogra@anubis:~/Devel/packages/powerd-0.13+13.10.20130926$ grep -r onTryRegister * [12:18] ogra@anubis:~/Devel/packages/powerd-0.13+13.10.20130926$ [12:19] not a powerd function at least [12:19] popey, well the app is called shorts in the lens; is the plan for it to be multiple apps? [12:19] no [12:19] it's one app [12:19] I'd keep it as "Shorts" [12:20] popey, something like com.ubuntu.rssreader_shorts_x.x ? [12:20] lool, the kernl notifies about power status once a minute === alan_g is now known as alan_g|lunch [12:20] sergiusens: if that matches the fomat of other apps, sure [12:20] (in dmesg that is) [12:20] besides source code, is it a requirement to have a cm 10.1 ROM running successfully or not? [12:21] popey, all other apps have pkg name ~= appname; this one is the only one a bit different [12:21] sergiusens: the package name is largely a "technical name" though isn't it? [12:22] sergiusens: or are you just after consistency? [12:22] ogra_: telepathy-ofono [12:22] popey, it affects where you can write files to too (confinement follows package name) [12:22] ah [12:22] qDebug() << "onTryRegister"; [12:22] in connection.cpp [12:23] aha [12:23] sergiusens: so lets be consistent and call it shorts [12:23] so why is it printing debug messages in the first place ? [12:23] ogra_: it's printing debug in every function it's entering there [12:23] can you make a flipped image with only binaries? [12:23] right [12:23] that one is probably the poll to check for data coming up or something [12:23] or sim card, but I'd think data [12:24] ogra_: I'm not too worried about this line every 5 seconds; not nice, but wont generate much logs [12:24] it pollutes a bit [12:24] but it's not in itself the source of the bug [12:24] yeah [12:25] can you port Ubuntu touch with only binaries? [12:25] popey, ack [12:25] linux_unix-10, no [12:25] linux_unix-10, you need the srouce of the android OS plus the binary drivers it uses [12:26] can you decompile the binaries? [12:26] it doesnt need to be cyanogenmod (note that porting will be a lot harder without CM though( [12:26] no [12:27] can I decompile and compare to aosp to make changes? [12:27] no [12:27] you cant decompile binaries [12:28] what can I do to get source code? [12:28] ask the vendor of your phone, they have to give it to you [12:29] good thing that the device's android is very close to aosp [12:29] it uses mostly vanilla configurations with only a few additions === _salem is now known as salem_ [12:31] lool, ogra_ I think I saw a lot of upower battery stat events (that and nm) [12:31] can I use source code from a device with identical hardware? (device is just rebranded) [12:31] sergiusens, where ? [12:31] ogra_, on the system bus [12:32] linux_unix-10, try it, if you manage to build a booting image for your device from that source that would be a good step forward [12:32] sergiusens, well, our bug makes the user session init eat all ram ... not sure if the system bus can cause this [12:33] i would expect fallout for this on the system side insterad [12:33] *instead [12:33] the partitioning of the two devices are different, is this a problem? [12:33] you will likely need to adjust it in the source code [12:33] ogra_, saw the bug; probably not related; I just chimed in on 'waking up often; [12:34] yeah, its definitely another one to look at [12:35] ogra_ there are more identical devices that were just rebranded [12:35] maybe I can use some source code from them when available [12:36] alecu: Re: #1235784, I'd like to gather as much as possible debug information right now as to be able to reflash my phone if I need to [12:36] but enough so that it can be resolved [12:37] Hey any way yet to enable 3G browsing on a Nexus 4? [12:37] one of the source codes for one device has code for the same partition as my device, can I merge it with the original device's code to make it compatible? === psivaa-afk is now known as psivaa [12:39] lool: what's the image # where this happens? [12:43] alecu: #83 [12:43] alecu: also see #ubuntu-ci-eng backlog [12:43] dobey: Let's perhaps move the debugging here [12:43] * alecu looks [12:43] oh ok [12:44] lool: what about signing into https://myapps.developer.ubuntu.com/dev/ ? [12:44] this is very much sounding like the server problem we found at the end of last week [12:44] dobey: I think that hole has already been fixed server side [12:44] alecu: has it been deployed to production? or only staging? [12:45] dobey: still the same [12:45] I wasn't signed in, clicked sign in from desktop, didn't get prompted for 2nd factor [12:45] still nothing happening on a different click on the phone [12:45] lool: unless you have 2fa set to "always" on that account, you shouldn't get prompted for it there (or on the software-center.ubuntu.com page either) [12:46] lool: can you paste unity-scope-click.log and ubuntu-download-manager.log from .cache? [12:46] lool: and you're definitely logging into those web pages with the samea ccount as you signed into on the device? [12:47] dobey: I only have one account [12:47] albeit I can't see the email address in the list of configured accounts [12:48] http://paste.ubuntu.com/6204781/ click-scope.log [12:48] I dont see the download-manager log [12:49] lool: perhaps .cache/ubuntu-download-manager/ubuntu-download-manager.log ? [12:49] lool: it's ~/.cache/ubuntu-download-manager/ubuntu-download-manager/ubuntu-download-manager.log i think [12:49] slangasek: nothing in fuser -vm / 2>&1 | grep F; might be android side? [12:50] it's actually http://paste.ubuntu.com/6204794/ [12:50] /home/phablet/.cache/ubuntu-download-manager/ubuntu-download-manager/ubuntu-download-manager.log [12:50] crazy [12:52] the unity-scope-click.log looks either incomplete, or like it's hanging on getting the credentials [12:53] lool: if you kill system-settings then go back to the accounts page, does it show your e-mail address for the u1 account? [12:54] dobey: no [12:54] it lists one account, no email [12:55] lool: remove the account, and re-add it. [12:55] hello [12:55] dobey: pretty sure that will fix the bug though :-) [12:55] I was wondering what will be the video/audio app on Ubuntu Touch ? [12:56] lool: how did you create the account? [12:56] dobey: how is that empty account being created? [12:56] Is it plan to developp something related to WebRTC ? [12:56] dobey: as usual, prompted to go to settings to add account, added account with 2nd factor [12:56] as "major" video/audio app won't work on Ubuntu Touch [12:57] lool: I think there was a bug where the account ended up being created empty, when the user used the "Back" button from the toolbar instead of "cancel" in the u1 dialog [12:58] copied some video onto my device.. should I expect it to show up in the video lens yet? [12:58] dobey: is there a way to list the token in lool's UOA entry? [12:59] alecu: yeah, that sounds more like it [12:59] alecu: on the device, i think it's stored in an sqlite db [12:59] alecu: was this actually fixed? [13:00] did you use the back button? [13:00] lool: I don't. let me ask ralsina or mmcc (when he starts) [13:01] alecu: the back fix? it's in trunk and in the latest release but it's not in the image yet [13:01] popey, I saw them show up in the home lens [13:01] popey, don't recall video though [13:03] lool, alecu: the back button issue is fixed, yes [13:04] dobey: do you have a reference? [13:04] bzr revision or such [13:04] lool, alecu: in ubuntuone-credentials 13.10-0ubuntu1 [13:04] ralsina: ^^ did you add that to the landing requests page? [13:04] from one hour ago [13:05] dobey: Hmm what landing slot was this? [13:05] dobey: not yet [13:05] lool: I was going to ask for it to be in the general click stack landing you mentioned earlier [13:05] ralsina: yes but note that this package is not under CI, so it just went straight through proposed and into archive [13:06] lool: yes, we do manual testing on it, will get it into CI after the release. Too much process breakage at this point :-/ [13:07] ralsina: I could have arranged to stage it in proposed though [13:07] lool: didn't know that, sorry. Let's do it for the next one [13:07] well, there shouldn't be a next one :) [13:08] (if there is a next one before release, which is not certain) [13:08] right :-) [13:08] just the man.. jhodapp|afk (when you're not afk) - should I expect to see videos from /home/phablet/Videos in the dash, and should I be able to play them? [13:08] lool: for reference, here's how to check that credentials have been stored right: [13:08] as phablet user: http://pastebin.ubuntu.com/6204852/ [13:09] stgraber, ping [13:09] stgraber, flashing devel-proposed-customized didn't seem to pull the sevilerow tarball === jhodapp|afk is now known as jhodapp [13:14] alecu: thanks, wanted to check this === fmasi_afk is now known as fmasi [13:15] alecu: first line returns 1||ubuntuone|0 [13:15] alecu: so that's probably the bug dobey fixed in ubuntuone-credentials [13:18] lool: did you use the back button on the toolbar to "cancel" adding the acccount? [13:18] lool: that last zero means "disabled", so there's probably a bug in how we are handling the account there [13:18] either in the scope or in libsignon [13:21] cwayne_: pong [13:21] cwayne_: let me check, may be related to the QA Jenkins changing IP [13:23] bfiller: Good morning! [13:23] Is there a port for samsung galaxy note 2 sgh-i317m(Canadian Model) [13:23] bfiller: good news: ubuntu-ui-toolkit is in archive with the enw openUrl stuff [13:23] bfiller: it would be lovely to land all the application changes to add the support; would you have a list handy? [13:23] I think the merge proposals were up already [13:24] cwayne_: yeah, so that looks like a mix between a bug in system-image server code (wrong behaviour when a file is missing) and the fact that it can't reach the Jenkins server [13:24] xnox, if you're interested, the upstart DBUS env thing is happening to me again, and I'm up to date [13:25] well, up to date enough to have the dbus conf fix anyway [13:26] cwayne_: unfortunately there isn't much I can do to fix it without messing with existing devices even more... the best way forward is to have RT:#65040 fixed ASAP, then the import run after that will get things back to normal [13:26] lool: yup, mediaplayer and browser have MR's ready. let me find them [13:27] lool: any chance we can get IS to hurry a bit on RT:65040 (customized images are broken until it's...) [13:28] stgraber, hmm, ok. who should we talk to to get that RT fixed? [13:28] jodh, hey ! ... stopping the upstart-event-bridge makes init stop eating ram over here, on both devices [13:29] ogra_: right, lending weight to it being a kernel issue. [13:29] cwayne_: maybe lool can help getting this fixed a bit faster [13:29] on both kernels ? [13:29] stgraber: Can try [13:29] let me look [13:29] jodh, maguro uses a 3.0, mako a 3.4 kernel [13:29] ogra_: have you seen the linked bug? [13:30] ogra_: bug 1234743 [13:30] bug 1234743 in linux (Ubuntu) "omapfb module floods system with udev events on samsung galaxy nexus when playing mp4" [Undecided,Incomplete] https://launchpad.net/bugs/1234743 === alan_g|lunch is now known as alan_g [13:30] ogra_: h/w specific [13:30] jodh, i see it on mako [13:30] oSoMoN: the url handling has landed in sdk, can you test and approve https://code.launchpad.net/~abreu-alexandre/webbrowser-app/expose-freedesktop-application-open-caps-on-dbus/+merge/188729 [13:30] ogra_: do you see it on grouper? [13:30] jodh, it doesnt differ from maguro though [13:31] stgraber: feedback is that Spads would be on it already [13:31] i dont have my grouper around atm ... since we dont really care for tablets atm [13:31] cwayne_: ^ [13:31] ogra_: can you recreate on grouper though? I can't [13:31] lool: ok [13:31] bfiller: will do [13:32] ogra_: well, I've asked a few extra questions on the bug that are still unanswered (I only have a grouper device, no phone). [13:32] jodh, thats bad, someone should get you a phone [13:32] ogra_: I guess they've got a few days to do so :) [13:34] jodh, oh, note that mako uses the interactive cpufreq governor, i am pretty sure each input is supposed to create an event for this [13:34] usually that governor is tied to touchscreen input [13:36] ogra_: I've seen no evidence that upstart is misbehaving yet: as I've explained on the bug, upstart is simply passing the udev message through the system. And those 2 devices seem to flood the system with udev messages :) [13:36] ogra_: ah - I see we have device-specific upstart jobs. What is in /etc/init/bluetooth-touch-maguro.conf ? [13:36] jodh, well, init consumes about 200MB here after using the system with Mir for 20min [13:36] after 25min the syystem is dead [13:36] onn both phones [13:37] sounds pretty much like a problem with init to me that it can do so [13:38] jodh: how could we track why init is using so much memory? [13:39] not matter if this is fallout of some other bug or HW specific, the session init constantly grows (on surfaceflinger as well, just at a slower rate so you will hit the issue after a day or two) [13:39] jodh: could it be it's piling events which aren't processed? [13:39] lool: the question is whether the memory drops down when the upstart-event-bridge is stopped - ogra_ ? [13:39] nope [13:39] it is kept [13:39] it just stops growing [13:40] lool: the mediaplayer MR is here and Ijust approved: https://code.launchpad.net/~renatofilho/mediaplayer-app/fix-1230404/+merge/189513 [13:40] ogra_: are any of the logs in ~/.cache/upstart/ updated at all? [13:40] jodh, nope [13:41] ogra_: can you create a job that specifies "exec hello world", start the job and see if you get a logfile? [13:41] ogra_: "exec echo hello world" even :) [13:41] oh, they are surely updated on session start [13:41] just not excessively beyond that ... i can see dbus.log updating etc [13:41] popey: Hmm dpm doens't seem to be around and I'm looking for someone to review https://code.launchpad.net/~lool/music-app/url-handler/+merge/188937 would you have any ida? [13:41] *idea [13:42] bfiller: awesome; any other one? [13:42] * popey looks [13:42] lool: oSoMoN is testing this one for the browser https://code.launchpad.net/~abreu-alexandre/webbrowser-app/expose-freedesktop-application-open-caps-on-dbus/+merge/188729 [13:42] jodh, I attached the info you requested to 1235649 [13:43] bfiller: Ok; listing these in landing #94 [13:43] bfiller: Note that upstart-app-launch will land a bit later [13:43] lool: what about the Unity8 changes to support promoting already running app? wondering if that has landed === pete-woods is now known as pete-woods-back- [13:44] bfiller: that's unity-mir, no it's not landed [13:45] bfiller: it's getting close, but I think we want to land with new upstart-app-launch when Mir input is fixed, should be tonight [13:45] jibel: thanks - looking. === kentb-out is now known as kentb [13:49] guys the music scope who deals with search and the wheel? If you have music on your device and search to lower it to an individual track and click on it it just moves around the wheel rather than opening [13:49] and what are the names of the elements so I can file a bug please [13:49] "these music things" [13:49] :) [13:50] davmor2: works here [13:50] for me too [13:50] although with no art, I have no clue what track I am playing until it plays [13:50] * ogra_ has art [13:50] * popey has _some_ art [13:50] popey: oh I wonder if it got fixed in 83 I'll retry it something I hit over the weekend testing media [13:50] sadly i have one album only ... which means all songs have the same art [13:50] heh [13:50] looks quite boring [13:51] ogra_: I'm with popey, I have some album art [13:51] http://www.jamendo.com/en/artist/343348/intercontinental-music-lab [13:51] get theirs [13:51] mad, and has art [13:51] (collaboratively made online too, just like Ubuntu) ☻ [13:51] popey, ogra_: I copied the 3.8GB of music I have from U1MS [13:52] does embedded album art work yet? [13:52] or does it still jsut pull it from last.fm [13:52] why dont you play it from your U1 store ? [13:52] :) [13:53] Grrr! W/Adreno200-EGL( 1293): : EGL_BAD_CONTEXT [13:54] Wish I knew what is causing that, I swear it is causing my 1080p VIDC not to work [13:57] ogra_: because you get the stupid popup asking for permission to continue after every track [13:57] the wonderful world of 2fa [13:57] zsombi: is the uitk feature to have slide-to-delete ask confimation already included in today's image ? [13:58] ogra_: no not 2fa, the one in the browser I'll grab a screenshot after [13:59] My udev rules are good.. checked over and over.. firmware is there.. but for some reason.. I still get EGL_BAD_CONTEXT. [14:00] @Ogra.. are you guys able to play videos? I can hear audio, but no video... [14:00] @ogra =) [14:00] ogra_: so in rhythmbox to transfer the music you have in mp3 from u1ms to the phone it want to convert everything to ogg vorbis oops [14:01] I was able to play 480p, but now I cannot play a thing.. after the new gstreamer update [14:01] I can play audio just fine though [14:01] OrokuSaki, i can play http://people.canonical.com/~jhodapp/sintel_trailer-1080p.mp4 when pushing it to the Movie folder and selecting it in the filemanager [14:02] Yeah.. I cannot play the sintel mp4 480p trailer [14:02] make sure yu have one of the latest images though ... media playback changed a lot recently [14:02] I am on the latest of the latest [14:02] (and also make sure to have your android build up to date) [14:02] yeah.. I did that.. saw hybris got updated, etc [14:02] OrokuSaki, what device are you on? [14:02] hp touchpad [14:02] right [14:02] popey, hey, can you do a final ack here? https://code.launchpad.net/~sergiusens/ubuntu-calendar-app/clicker/+merge/189580 [14:03] * popey clicks [14:03] I think the EGL_BAD_CONTEXT is probably preventing me from playing 720P and higher.. that is my guess [14:03] sergiusens: done [14:03] OrokuSaki, it's very possible there's some device specific bugs you're running into, unfortunately [14:03] I don't get that in android [14:03] does ubuntu-touch work for calls/sms/3g for anything outside US/europe? [14:03] popey, there's also this one here, not sure who to send to https://code.launchpad.net/~sergiusens/ubuntu-terminal-app/clicker/+merge/189578 [14:03] brazil, in my case [14:03] yeah... odd because I know hybris was being tested on a toucpad [14:04] krycek_, yes [14:04] OrokuSaki, yeah but typically there have been udev permission bugs to fix for devices [14:04] ogra_: without any extra config? [14:05] OrokuSaki, specific to the flipped container [14:05] I have tested all udev combo's.. well there is one issue I noticed [14:05] krycek_, as long as you dont need a PIn, yes ... else you need to unlock the PIN from cmdline [14:05] "/dev/pmem" has different permissions with user and group then it does in ubuntu touch [14:06] ogra_: nice, thanks [14:06] and my udev rules won't fix it.. the user or group doesn't seem to exist [14:06] lool, Are you willing to give a green light here? I'd like to get it landing in trunk so when release is ready, I'm there too :-) https://code.launchpad.net/~ted/url-dispatcher/app-id-url/+merge/187913 [14:06] OrokuSaki, yeah, that's an important one...just for kicks try changing the owner/group to root:root [14:06] tried that as well [14:06] OrokuSaki, did you try changing it to phablet:phablet? [14:06] .... no... [14:07] hmmm [14:07] or chmod a+rwx [14:07] should any udev rules be phablet phablet? [14:07] ogra_: so in mtp is there a setting for default music type or something? [14:07] davmor2, i'm not even sure music player mode is implemented in the server yet [14:07] davmor2, ask cyphermox if thats supposed to work at all yet [14:08] "/dev/pmem 0660 system graphics" "/dev/genlock 0666 system graphics" [14:08] that is from my uevent.d [14:08] ogra_: it wants to work it shows up and everything ;) [14:08] but I cannot seem to set those permissions in udev [14:08] not even with chown [14:08] davmor2: no, it doesn't work yet [14:08] davmor2, yeah, but its probably still on a leash :) [14:08] OrokuSaki, you can sudo chown phablet:phablet? [14:09] jhodapp: I just saw your ping via ogra_ re: brcm-patchram, I don't know why it's broken now... that was working quite well before [14:09] cyphermox, that wa jodh :) [14:09] yep [14:09] *was [14:09] evidently something changed somewhere and it broke brcm-patchram-plus -- I'll try to fix it [14:09] ahah :) [14:09] jodh: ^ [14:09] cyphermox: thanks! :) [14:09] call me dyslexic :) [14:09] OrokuSaki, and does that fix anything for you? [14:10] cyphermox: shame on you getting my hopes up ;) however I'm thinging that I might copy one of the u1ms files over to the ~/Music file and seeing if I could transfer that over :) You know just for the hell of it :) [14:10] cyphermox, am i supposed to find a headset already in the new BT UI ? [14:10] it tells me it scans for one but does never seem to find anything [14:10] service lightdm restart USER=phablet is this the correct way to restart the session these days? [14:10] ogra_: works for me [14:10] davmor2: well, you can definitely copy music files to the device to get them in the music player, but not via a music player on your desktop just yet [14:10] ogra_: for some, yes [14:10] ah, k [14:11] then mine is probably special :) [14:11] no go.. still bad context [14:11] ogra_: it should work for most, but perhaps there's something more fundamental about bluetooth first? [14:11] ogra_: hciconfig -a ? [14:11] ogra_: you got the qa headset right :D [14:11] OrokuSaki, there's some other bug then...if you pastebin the logcat output I could take a quick scan for you [14:11] davmor2, jabra halo 2 [14:11] Hello guys. I would like to try building Ubuntu touch for the New Nexus 7 flo. 2 questions. 1. can i build it from funtoo? 2. When it talks about pre-built images which one would i download as none are obviously labelled for the flo? [14:12] OpenGL doesn't start [14:12] after surfaceflinger [14:12] you can see a strace [14:12] I will show you [14:12] cyphermox, http://paste.ubuntu.com/6205076/ [14:13] OrokuSaki, can't really render video without OpenGL :) [14:13] cyphermox: should bluetooth be turn on by default? I'm assuming it sucks juice from the battery constantly checking for devices, but I think there is a bug if you disable BT from the indicator in that it never comes back up again [14:13] note that i dont have my headset around, i only noticed it on the weekend [14:13] ogra_: looks good. it's up to charles then :) [14:13] also why do we restrict ourselves to headsets ? [14:13] http://pastebin.com/5vsErDA4 Ubuntu Touch with TouchPad [14:13] davmor2: currently it is on by default yes, but I'd think the drain be minimal [14:13] * ogra_ would realy like to use his BT speaker [14:14] stgraber: where do downloaded updates end up on the filesystem of my device? [14:14] davmor2: honestly, I don't know. I figure the idea would be to fix the indicator first [14:14] jodh, I attached the output of get_state as user phablet, var/log/udev is already there in comment 19, or do you need another log? [14:14] ogra_: we shouldn't, but it seems like somehow it was coded that way so far? [14:14] popey: while downloading /tmp, then get moved to /cache/recovery right before rebooting, then they're removed by the upgrader once applied [14:14] well, thats what the new UI says now [14:14] yeah :/ [14:14] "connect a headset" ... [14:15] ogra_: I think it should just work with your speaker anyway [14:15] popey: (/tmp/system-image specifically) [14:15] http://pastebin.com/fcT9HFWq CM10.1 Booting on the TouchPad [14:15] cyphermox, it doesnt find it [14:15] You can see the difference after surfaceflinger [14:15] now that one i have next to me and could gather some data [14:15] ogra_: daft question you have set the headset to be discoverable right? I'm assuming yes but hey :) [14:15] charles: ^^ could you please check this out? I was wondering why only headsets as well... [14:15] stgraber: thanks [14:16] davmor2, yes, indeed :) [14:16] root@ubuntu-phablet:/# hcitool scan [14:16] Scanning ... [14:16] 00:0C:8A:6B:F7:F4 Bose Mini SoundLink [14:16] cyphermox, ^^^ [14:16] cmdline finds it [14:16] oops that last one is wrong [14:16] so i suppose its a UI thing [14:16] ogra_: yes [14:16] that's why I said charles :) [14:17] http://pastebin.com/SyrfuiNn CM10.1 booting the touchpad [14:17] http://pastebin.com/5vsErDA4 Ubuntu Touch with TouchPad [14:17] the spec only asks for headsets [14:17] https://wiki.ubuntu.com/Bluetooth#phone [14:17] bah [14:18] its not like the connection procedure differs much for other BT devices [14:19] E/Trace ( 1028): error opening trace file: Permission denied (13) Line 72 on the UbuntuTouch paste [14:19] I read that was because of google plus not being there or something??? [14:20] mzanetti: i forget how to have use that tool in autopilot that shows the obejct graph of a qmlscene [14:20] mzanetti: can you remind me how to run it ? === dandrader_ is now known as dandrader [14:23] ogra_, charles, cyphermox: Meh so my BT Headset that was detected with Ease on Friday isn't today [14:24] davmor2: as in it's hard to detect or not detected at all? [14:25] cyphermox: on the setting app I see connect to headset: and the spinny icon then none detected appears [14:25] davmor2: ok [14:25] as before; what does hciconfig -a say? [14:26] and then check if hcitool scan finds it [14:26] cyphermox: http://paste.ubuntu.com/6205117/ [14:27] I really hadn't planned on bluetooth getting broken this late, especially when I don't change stuffs === Namidairo is now known as Namidairo`bnc [14:28] cyphermox: root@ubuntu-phablet:/# hcitool scan [14:28] Scanning ... [14:28] 00:1A:7D:54:32:8D BH99b [14:30] cyphermox: http://ubuntuone.com/55XfMNyfP3QtOy1q0H15X7 that's what I see and then the spinner goes away and the None Detected becomes bold again [14:30] davmor2: ok, so this is a UI issue [14:31] charles: ^ [14:31] davmor2: please file a bug about it [14:31] cyphermox: will do [14:31] davmor2: you can attach this screenshot and the output of hciconfig, it's very useful [14:31] oh, and hcitool scan as well :) [14:32] cyphermox: against what package settings, I'm assuming? [14:32] yep [14:41] cyphermox: https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1236397 [14:41] Ubuntu bug 1236397 in ubuntu-system-settings (Ubuntu) "Maguro: BLuetooth says None Detected in the ui but cli hcitool shows device" [Undecided,New] [14:43] bfiller, tmoenicke: re- the touch events going through the OSK in landscape mode, see bug #1236360 that has a standalone QML example to reproduce the issue [14:43] bug 1236360 in Ubuntu UI Toolkit "MainView.anchorToKeyboard doesn’t resize the contents in landscape orientation" [Undecided,New] https://launchpad.net/bugs/1236360 === dandrader is now known as dandrader|afk [14:51] oSoMoN: good thx [14:57] ogra_, popey: plug in headset that comes with your phone. play some music (works on the headset), now make a call (this happens on the phone not the headset for me, and I have to remove the headset to talk/hear on the phone) [14:57] oSoMoN: great, thanks [15:01] bfiller: do you know how I can test the URL handling in the browser? I tested with a sample QML file that does Qt.openUrlExternally("http://example.com"), but it opens a new browser instance === dandrader|afk is now known as dandrader [15:03] oSoMoN: opens a new instance even if browser is running already? === chihchun is now known as chihchun_afk [15:05] oSoMoN: we have the single instance flag set properly in the browser desktop file? [15:05] bfiller: yes, and yes [15:05] tedg: is the single instance stuff working yet in the image? or still waiting on stuff to land? [15:06] bfiller: there's a fake version of it right now [15:06] bfiller: the real implementation is held back [15:06] because it's mir specific and will regress the SF experience [15:06] so waiting a tiny bit more to be closer to Mir, but should go in soon [15:07] lool: what does the fake version do? [15:07] bfiller: it upstart-stops the app is already running, then starts it with the new URL [15:07] s/is/if [15:08] oSoMoN: ^^^ sounds like you are not seeing this? [15:08] Can't it test whether Mir is in use and do one or the other [15:08] ? [15:08] Sounds like unnecessary coupling to wait for it [15:09] yes, that's possible [15:09] tedg: ^ [15:09] bfiller: nope, not seeing this, I’m seeing two instances of the browser [15:10] tedg: would you want to reinstance the upstart-stop for the SF case since we only have "resume and bring to the front" for the Mir case? [15:10] bfiller: shall we discuss this race condition in the notes-app AP tests? [15:10] tedg: that would allow landing upstart-app-launch + unity-mir right now [15:10] oSoMoN: finishing a hang out, will be 5 more minutes or so [15:11] bfiller: ok [15:11] bfiller, cjwatson: (Note that this was also blocked on SDK implementation to allow a running app to open an URL, which is just making it to the image, but apps aren't all updated for that) [15:12] davmor2: I don't have a headset with my phone [15:12] cjwatson, lool that's what we came up with with ricmm and tedg with upstart app launch [15:12] * tedg is confused [15:13] tedg: rolling back a bit [15:13] popey: man that sucks any wired headset that is mic and earphones that fits the jack if there is one then :D [15:13] tedg: UI toolkit open url thing is landing [15:13] I thought we were dropping SF today. [15:13] tedg: apps using it are landing [15:13] tedg: well, we're trying, but it's not completely done [15:14] we could land most upstart-app-launch + unity-mir updates if we kept the saucy implementation on SF and the current implementation for Mir [15:14] I'm not sure what "saucy implementation" is. [15:15] Or why we care about SF for a day. [15:15] \o/, got my fitbit app working with usermetrics :D [15:15] cwayne_, Wow, cool! [15:16] tedg, erm, all ports run SF [15:17] (and surely will for a while) [15:18] ogra_, Well, then, let's give them a reason to port to Mir! [15:18] * ogra_ doubts they wont be able without a lot of support from the Mir team [15:18] *they *will* be able [15:24] bfiller: https://code.launchpad.net/~renatofilho/mediaplayer-app/fix-1230404/+merge/189513 getting a sceneselector failing test on mako! :-) [15:28] When Mir comes out.. we will be using Hybris to talk to Mir instead of SurfaceFlinger? [15:28] Where does the egl driver come from? the lxc container not running flinger? [15:29] Mir is already in the official images [15:30] many of us use it (you can switch it on optionally) [15:30] Extend existing QML Component from C++ plugin | http://askubuntu.com/q/354895 [15:30] (ask in #ubuntu-mir about more details :) ) [15:32] dobey: can you try this: 1) add U1 account, 2) add another U1 account, 3) remove one account, 4) try removing the second account [15:32] the latter doens't work for me, but that's using the old ubuntuonline-accounts [15:35] lool: just added a second account and removed both here, and it worked ok. [15:35] dobey: ok thanks! [15:36] lool: there was a weird crash i was getting with the online accounts code when doing something similar in testing, on my workstation, though. DONE: reviews, u1-creds release, debugging/ranting [15:36] TODO: releases, maybe some bug fixes [15:36] BLCK: None. [15:36] https://bugs.launchpad.net/bugs/1234348 [15:36] Error: ubuntu bug 1234348 not found [15:36] crap [15:36] ignore the done/todo stuff [15:36] lool: there was a weird crash i was getting with the online accounts code when doing something similar in testing, on my workstation, though. https://bugs.launchpad.net/bugs/1234348 [15:37] is what that should have been :) [15:37] right [15:37] I'm getting what appear to be crashes too [15:38] possibly the same crash then [15:38] seb128, remember the issue with the Back button in System Settings screens accessed from indicators? [15:38] need to get mardy to look at that bug i guess [15:40] mpt, yes [15:40] mpt, did you have that meeting? what's the output? [15:41] seb128, the conclusion from the design meeting today is that the Back button should always behave as an Up button [15:41] mpt, ok, so we should have it when directly opening a panel as well then? [15:41] seb128, so for example, it doesn't matter whether you accessed the "Location access" settings from the location indicator or from the "Security & Privacy" screen, the "Back" button should take you to "Security & Privacy" regardless. [15:42] lool: we are getting failures in random tests of media player because the codecs are crashing the media player app, rsalveti, jhodapp we need some help figuring this out [15:42] bfiller: I'll investigate that later today still [15:42] bfiller: ok [15:43] bfiller: it's certainly unrelated, so you might want to force this in [15:43] OH.. wow.. https://gitorious.org/mertp/kernel-clean/source/e32dd173f13bb9f2961567a0fe719c9c6150482b: [15:43] https://wiki.merproject.org/wiki/Adaptation/Touchpad [15:43] I should try that kernel! === fmasi is now known as fmasi_afk [15:43] mpt, the issue was more "what if do -> settings, that opens directly the battery control" [15:43] lool: yup, fginther can you have a look at this MR and get it mreged? https://code.launchpad.net/~renatofilho/mediaplayer-app/fix-1230404/+merge/189513 [15:44] fginther: seems tests failures are due to known issue with media playback crashes [15:44] seb128, it goes to the top-level System Settings screen [15:44] mpt, should it have a "back" and should the back take you to... (back would be the indicator) [15:44] mpt, ok, thanks [15:44] Laney, ^fyi [15:44] jodh, notify-disk-writeable, is that as user or root ? [15:45] ogra_: non-user. jibel already tried it - didn't help [15:45] seb128: okay then [15:45] I can do that patch [15:46] ah, k [15:47] lunch, bbiab [15:47] Laney, thanks ;-) [15:48] jodh, hmm, what i see on maguro is that while the screen is on each VSYNC from the graphics driver seems to create an event [15:49] ogra_: that's the same as the referenced bug (1234743). [15:49] yeah, likely [15:49] iirc Mir needs these events though [15:49] wont be easy to work around that [15:49] ogra_: the memory failing to fall issue almost feels like a compiler issue as upstart does automatic garbage collection and the state dump lgtm. [15:50] ogra_: it would be reeeeeeeally useful to have a device over here :) [15:51] jodh, yes :( [15:51] dont we have someone near you who has one ? [15:51] ogra_: where are the build logs for this build for maguro/mako/grouper? [15:51] you mean image build ? [15:52] http://people.canonical.com/~ubuntu-archive/livefs-build-logs/saucy/ubuntu-touch/20131005/livecd-20131005-armhf.out === gatox is now known as gatox_lunch [15:52] bfiller, lool, done [15:53] fginther: thanks [15:53] fginther: thanks [15:59] ogra_: out of interest why do we still have elements of gstreamer 0.10 on touch with 1.0 installed? [16:00] davmor2, because some Qt bits require it still [16:00] patches accepted :) [16:01] ogra_: no I was just looking into why rad.io didn't work any more and got a gstreamer listing so thought I would ask :) [16:04] jhodapp: hi, do you have any idea what changed from gst 0.10 to 1.0 ? [16:11] cjwatson, if I unregister an app, click list doesn't show it listed anymore but click list --manifest does; is that something you know about? === Dazzozo` is now known as Dazzozo [16:17] seb128: ping [16:18] mfisch, hey [16:18] seb128: have you guys made progress/decisions on how you will configure maliit? [16:18] seb128: specifically how can users switch the keyboard language [16:19] mfisch, ubuntu-keyboard config is going to be gsettings [16:19] daker, yes, quite a bit [16:19] seb128: so maliit.conf will be unused? [16:19] lool: android should still show up in the output of that call, AFAIK [16:19] mfisch, e.g https://code.launchpad.net/~schwann/ubuntu-keyboard/keyboard-gsettings/+merge/187722 [16:19] jhodapp: my app doesn't work anymore, it was working with 0.10 /cc davmor2 [16:19] daker, take a look at this: http://cgit.freedesktop.org/gstreamer/gstreamer/tree/docs/random/porting-to-1.0 [16:20] did anyone hear of issues doing calls (like only the second call actually gets through, or I can only hear the other person on the second call), possibly only on maguro? [16:20] daker, what's your app? any reason you started with 0.10 and not 1.0 right away? === MacSlow is now known as MacSlow|afk [16:20] slangasek: didn't find what was holding the fs read-write then [16:20] mfisch, , not sure, better to check with tmoenicke or attente, I didn't follow the details [16:20] cjwatson, nevermind, just a goof of mine [16:20] lool: hmm :/ [16:21] seb128: moving this into gsettings is perfect, did I miss this in your "settings update" email? [16:21] jhodapp: i am getting empty page http://i.imgur.com/DPNmrq5.png [16:21] mfisch, I didn't do those recently, I was away for holidays and then thing went a bit crazy, I should resume them [16:21] jhodapp: no particular reason... [16:22] jhodapp: here is the c++ code https://bazaar.launchpad.net/~rad.io-devs/rad.io/trunk/view/head:/plugin/player.cpp [16:22] seb128: we were shipping a customized maliit.conf which will no longer work, but we hated that solution anyway [16:22] mfisch, ok, sorry about the lack of notice, those are still moving parts [16:23] seb128: sure, np [16:23] sergiusens: ... good, that sounded bizarre :) === dandrader is now known as dandrader|afk [16:24] seb128: okay, so people asked us if you can switch keyboard language w/o a reboot, do you know if that will work? === dandrader|afk is now known as dandrader|lunch [16:24] ogra_: stgraber: it seems like something funny is going on in my android image, it somehow doesn't get mount namespaces or some-such as I have to mount "system" in ${rootmnt}/system ( ubuntu system.img loop.mnt ) for android's init inside lxc, to find /system/bin/* binaries. [16:25] xnox, why dont you concentrate on normal flipped before trying system image [16:25] mfisch, I don't know, I've not really be involved in the osk work, attente is doing the settings side and tmoenicke/bfiller know about the osk itself and what work is going there (gusch might be involved as well seeing the mp I pointed you) [16:26] mfisch, so better to ask those guys [16:26] seb128: will do thanks [16:27] mfisch: language change is on the list of tmoenicke - should work without reboot as far as I know [16:27] gusch: thanks, I'll email to confirm [16:29] ogra_: because normal flipped didn't even boot to rootfs for me. With system.img I can have yaffs2 -> file -> loop mount normal ext4. I cannot do direct ext4 on the mtd. =( [16:29] xnox, that wont really get you further though [16:29] you need at least boot into the ubuntu rootfs to be able to move on [16:30] mfisch, yw [16:30] i'll see if i can find some time this week to look into the emu stuff [16:30] ogra_: yeah, i have rootfs and android booting now. [16:30] xnox, then you should have adbd too [16:30] ogra_: nope, somehow it's not built for goldfish device =) [16:30] ?? [16:31] its an armhf ubuntu binary [16:31] ogra_: i have serial console to ubuntu rootfs, so I am ok. [16:31] ogra_: well yeah, but to talk to the host via qemu, it cannot do adb-over-usb, it must do it adb-over-qemud. and qemud is not running at the moment. [16:31] ah k [16:31] ogra_: http://paste.ubuntu.com/6205627/ [16:32] why cant you attach a USB host device to a qemu one ? [16:32] ogra_: no. [16:32] hey there; just curious, do you guys have a tip to activate your sim card pin automatically? [16:32] ogra_: no support for such things in the android's emulator. [16:32] thats a generic qemu feature [16:33] ogra_: passing -usb qemu options do nothing. =/ [16:33] at leats it used to be when i still poked around in beagle images years ago [16:33] bah [16:33] hm..... [16:34] popey: any suggestion for the review of https://code.launchpad.net/~lool/music-app/url-handler/+merge/188937 ? [16:34] at the moment kernel panics are not helping me. === pete-woods-back- is now known as pete-woods [16:35] lool: perhaps request a review from the music app devs? [16:36] lool: ahayzen is around (music dev) - how can he test it? [16:37] ahayzen: Hey there [16:37] lool, o/ [16:37] ahayzen: could you review https://code.launchpad.net/~lool/music-app/url-handler/+merge/188937 ? [16:37] ahayzen: to test, you'd want latest proposed image #84 [16:37] lool, is there a way of testing on the device easily? [16:38] popey, ogra_: #84 is up by the way :-) [16:38] lool, just reflash it? [16:38] super [16:38] ahayzen: or update to it [16:38] flashing... === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk [16:41] You guys think I should include this patch? 61c482e: tty: add 'active' sysfs attribute to tty0 and console device [16:42] https://wiki.merproject.org/wiki/Adaptation/Touchpad [16:47] ogra_: plus "imitating" system.img layout is not that bad, given that I touch ".writable_image" ahead of time. === tvoss_ is now known as tvoss|quick_dinn [16:51] xnox, well, as long as you are sure you dont mix and mash the two systems [16:51] 753371 [16:51] bah [16:51] ogra_: quite. [16:52] popey, generate a new token [16:52] :þ [16:55] lool: 84 feels good [16:56] back in 15 [16:56] not seeing 84 on http://reports.qa.ubuntu.com/smokeng/saucy/touch_ro/ yet? [16:57] yeah, it takes a while until they show up on the dashboard [16:58] * ogra_ is still waiting for the download to finish [16:59] mfisch: keyboard layout switching not working yet [16:59] for langs that is [17:00] ogra_: bah, my systemimage doesn't have a label, thus update-grub doesn't find it! hence the manual hackery. [17:00] bfiller: ok, but eventually it will work w/o a reboot? [17:00] ogra_: will fix. [17:00] mfisch: yes [17:00] thanks === gatox_lunch is now known as gatox [17:00] great === alan_g is now known as alan_g|EOD [17:02] mhall119: Did you get the app of yours that we talked about upgraded to Friends 0.2? [17:02] cjwatson: not yet, no, but it's in a click package not deb so don't wait for me to remove Friends 0.1 [17:02] ogra_: ^- [17:03] yeah, i thought so [17:03] clock apps are luckily not bound to the release schedule [17:03] *click [17:03] I'll even unpublish it if that makes things easier/better, since Karma Machine is a better Reddit app than mine anyway [17:03] your call ... [17:03] if it doesn't cause a problem for you, I'll leave it there to encourage me to upgrade it :) === Ursinha is now known as Ursinha-afk [17:04] it doesn't given that nothing has any way to check === Ursinha-afk is now known as Ursinha [17:05] well, it gives us the opptortunity to add checks now that we have a known broken app :) [17:05] Kaleo: ping [17:11] ahayzen: did you get the update? [17:11] lool, i've reflashed just trying to build a deb and get it installed on the device [17:13] ahayzen: cool [17:14] lool, should i be using start application APP_ID=music-app APP_URIS="file:///home/phablet/Music/woot"? to test? [17:17] ahayzen, You probably should use either upstart-app-launch utility or the url-dispatcher one. [17:17] tedg, how would i do tht? [17:17] Calling start directly will probably work... but not advisable. [17:18] ahayzen, $ url-dispatcher file:///home/phablet/Music/woot [17:18] ahayzen, You might need url-dispatcher-tools package, not sure if that one is there or not. [17:18] tedg, bash: url-dispatcher: command not found :/ [17:18] ahayzen: right; this simulates how url-dispatcher opens music:// URLs, you could either do that or click on music files from scopes; note that you need to put a real pathname there, not woot obviously :-) [17:19] ahayzen, $ upstart-app-launch music-app file:///home/phablet/Music/woot [17:19] ah my debian install seems to have failed :/ damn [17:19] ahayzen: the cmdline is the only way to test sending multiple URLs, but we don't really use this right now [17:19] tedg: upstart-app-launch isn't seeded though [17:20] ahayzen: what's interesting is that the first time it should start the app with the file on the cmdline, the second-time it should tell it to switch playing something else [17:20] lool, Oh, I thought click was using it... [17:21] lool, i'm getting qmlscene: could not find a Qt installation of '' when i run music-app must have done something wrong :/ === dandrader|lunch is now known as dandrader [17:21] Fine: $ gdbus call --session --dest com.canonical.URLDispatcher --object-path /com/canonical/URLDispatcher --method com.canonical.URLDispatcher.DispatchURL file:///home/phablet/Music/woot [17:21] :-) [17:24] tedg: if it is it's broken, could you check? [17:24] ahayzen: is this on a phone? [17:24] ahayzen: are you running this as ~phablet? [17:24] lool, i'm running from adb shell [17:25] ahayzen: as root? [17:25] lool, yeah [17:25] ahayzen: should be run as ~phablet [17:25] lool, ah ok [17:25] ahayzen, sudo -u phablet -i [17:25] that will make sure you get all the environment you need [17:26] module PowerD is not install :/ [17:26] i'll rebuild without the import [17:26] lool, Not sure how to do that. cjwatson, do you know if click needs upstart-app-launch-tools? If so, does it need to be seeded? [17:28] tedg: it can take advantage of it if it's installed, to fine-tune removal [17:28] tedg: it's not a hard dependency, but I think it should be seeded if you're happy for it to be installed by default [17:28] cjwatson, Yeah, I think so. It's also useful for debugging. [17:29] Sounds like JFDI :) [17:29] I should write some man pages.... [17:29] cjwatson, Heh, yup. [17:30] tedg, lool is just doing seed changes anyway, probably he could add that [17:32] tedg: let me pick this up [17:32] Oh, I just proposed a merge, either way. [17:33] Wait, it errord. You do it :-) [17:33] seeded [17:33] well needs an upload still [17:34] balloons: ping [17:35] m-b-o, hello [17:35] balloons: can the test for switching tabs be unskipped? [17:35] you remember, the loop? [17:35] m-b-o, yes [17:35] that bug has long been fixed. [17:36] however there is/was a new bug with switch by index, is it using that/ [17:37] balloons, hmm, there's a "switch_to_tab_by_index()", I guess that's an emulator call [17:38] m-b-o, yep that's it [17:38] m-b-o, the changes should have landed by now, but I don't have the bug handy [17:38] meaning it should work [17:39] balloons: https://code.launchpad.net/~zsombi/ubuntu-ui-toolkit/tab-index/+merge/188769 ? [17:39] m-b-o, yep [17:40] okay, will unskip the test [17:47] ahayzen: no luck? [17:47] lool, no it froze when i ran music-app :/ [17:48] ahayzen: like when launching it? [17:48] lool, and it is like the app is too big for the screen very strange [17:48] ahayzen: ah that sounds like you don't have the right SDK [17:48] lool, yeah when i run music-app in the terminal on the phone [17:48] ahayzen: is this with #84? [17:48] lool, how can i check [17:48] ahayzen: system-image-cli -i [17:49] lool, FileNotFoundError: [Errno 2] No such file or directory: '/var/log/system-image/client.log' [17:49] pmcgowan: where is the API for accessing the infographics? [17:50] ahayzen: which device is this on? [17:50] lool, Nexus 4 [17:50] ahayzen: how did you flash it? [17:50] lool, sudo phablet-flash ubuntu-system -b [17:50] ahayzen: ok, so I think you're on the "stable" image [17:50] mhall119, not sure ask bfiller, but we are restricting its usage right now [17:51] ahayzen: the updated ui-toolkit is only in the "proposed" image [17:51] lool, and then i did adb shell touch /userdata/.writable_imag [17:51] lool, how do i get tht? [17:51] bfiller: ping [17:51] ahayzen: you may either flash to latest image, or upgrade some packages manually albeit in my experience this often ends in fs corruption [17:51] ahayzen: phablet-flash ubuntu-system --channel devel-proposed -d mako [17:51] that should backup /some/ files [17:51] but not everything [17:52] lool, not worried about backup the files tbh its a dev phone so just gets wiped each time :) [17:52] lool, thts running i'll see wht happens [17:53] lool, wht i did last time was branch ur code... build a deb package on the laptop then push the deb to the device and install it...tht is correct? [17:53] mhall119: Would you happen to have an idea how would we show of Qt integration when develping apps for Ubuntu phone? Something that is not obviously QML, but shows off why we're talking about *Qt* specifically. [17:53] ↑ question to anyone with an idea how to demo Qt integration on Ubuntu touch :) [17:53] karni: I don't quite understand the question [17:54] ahayzen: sounds good to me! [17:54] mhall119: How can we show of leveraging power of Qt on Ubuntu Touch? [17:55] mhall119: We have examples of QML and HTML5, how easy it is to use them. We were wondering together with achiang how can we show leveraging Qt integration on Ubuntu touch. [17:55] mhall119: ong [17:55] Qml is Qt ? [17:55] bef0rd: not really [17:56] karni: ah, you mean using C++ and Qt? [17:56] QML is a Qt module [17:56] mhall119: perhaps. for instance, a Unity plugin is far to complex to show off on one slide [17:56] karni: we plan to add a section for C++, like we have for QML, as soon as we can put together the content for it [17:56] bfiller: hey, do we have API docs for the infographics? [17:56] bfiller: also, what are the requirements/restrictons on their use [17:57] mhall119: Right. Perhaps you might have a simple idea for me what we could show off? The conference is next Monday, so we won't be able to wait for the c++ section. [17:57] mhall119: don't know about api docs but it's in code libusermetrics I believe has the qml bindings, pete-woods would know [17:57] karni: well you can use some of the default apps, like gallery-app [17:58] bfiller, mhall119: libusermetrics does have QML bindings, yes [17:58] mhall119: I see. Could you tell me under which project can I find it on lp? Is it under ubuntu-touch or some sort of core apps project? [17:58] bef0rd: FYI :) http://en.wikipedia.org/wiki/QML [17:59] bfiller, mhall119: http://bazaar.launchpad.net/~unity-team/libusermetrics/trunk/view/head:/examples/metrics_declarative.qml [17:59] thanks pete-woods [18:00] mhall119: found it :) thanks! [18:01] pete-woods: how do I build the docs for that? [18:02] mhall119: they should already be being published: http://developer.ubuntu.com/api/devel/ubuntu-13.10/cplusplus/usermetrics/ [18:02] but it looks like the qml example isn't there === MacSlow|afk is now known as MacSlow [18:07] thanks pete-woods [18:10] karni: you want to show code on a slide, or just something that works and is written in c++/qt that is integrated into a qml thing? [18:11] dobey: Ideally at least a snippet of code + screenshot. If a simple example is rather hard to find, probably the latter. [18:12] dobey: That's in the more technical part of the presentation. === tvoss|quick_dinn is now known as tvoss_ [18:13] mhall119: Thanks for finding the infographics information. I was looking for it after seeing cwayne_'s post. What are the restrictions on usage? [18:13] karni: well, C++ is going to take a lot more LOC than QML will, so showing a tiny bit of code that does much isn't going to be very easy. you could maybe write a simple Hello World app in C++ or something, but it's not going to really show much in the way of "integration" [18:14] lool, ok so the app is now the right size, but when i run it from the terminal on the phone it seems to freeze after startup [18:14] lool, and if i run if from the adb shell it doesn't even appear on the phone [18:15] dobey: I thought so. Thanks for your opinion, dobey! [18:15] ahayzen: how do you run it? [18:15] lool, $ music-app [18:15] ahayzen: yeah, that's expected not to work from adb shell [18:16] ahayzen: concerning the freezes, did the one installed before that work? [18:16] dobey: karni: i think i have a perfect example... i'll grab some exif parsing code from the gallery app [18:16] karni: the u1 account stuff in the online accounts settings though, is a combination of c++ and qml. [18:16] achiang: \o/ [18:16] bfiller: pete-woods: can either of you comments towards what apps can and can't do with the infgraphics API as a matter of policy? [18:16] dobey: karni: now to figure out in the qml where we might display that info [18:16] lool, idk i'm not sure if it because i'm running from inside the terminal app [18:16] ahayzen: for it to work from adb shell, you need the "start " syntax I mentioned earlier [18:16] ahayzen: and that needs to be launched as ~phablet [18:17] ahayzen: does "start application APP_ID=music-app" work? [18:18] ahayzen: freeze >> I would suspect something else is going on, but can't tell why [18:18] mhall119, wellsb would it be helpful for me to write a quick blog post about using the qml bindings? it's super simple to understand from those examples, but they're not very well publicized [18:18] mhall119: I'm not sure, think they need an apparmour profile that allows it, jamie? [18:18] lool, it says its running in a process but have yet to see anything appear on the device [18:19] mhall119: apps aren't supposed to be using it AFAIK, I thought it was an internal API [18:19] cwayne_: that'll depend on the restrictions on their use [18:19] cwayne_: not much reason to tell people how to use it, if we're not going to let them [18:19] ahayzen: hmm that's odd [18:19] ahayzen: maybe kill all the running apps first [18:20] lool, none displayed as running in the app [18:21] mhall119, fair enough, although im not sure what the point of writing qml bindings would be if we're not gonna let people use them [18:21] dobey: karni: huh, i'm not actually seeing any QML code in the gallery app at all ;) [18:21] dobey: nvm, we found it [18:22] cwayne_: it could be a temporary prohibition on their use until the security aspect is fully undesrstood [18:22] dobey: sorry, i am dumb! [18:23] mhall119, fair enough [18:23] ahayzen: hmm not sure what's going on [18:24] lool, neither do i :/ [18:26] achiang: i know nothing about the gallery app :) [18:27] dobey: where do you use c++ in the u1 app? [18:27] Hey guys. Who's familiar with the hud? Is it hard to add a context button to hud? [18:28] achiang: all of the backend work is in C++. qml only provides the UI we embed into system-settings for adding the account [18:28] dobey: i see [18:28] dobey: i think i'll keep plugging away trying to grok the gallery app source, because that example will make a lot more sense to a room of app devs [18:28] dobey: thanks for your help though! [18:29] sure [18:30] lool, have u managed to get it to work on ur device? [18:31] ahayzen: so with #84, not with my own branch, music-app starts when I "start application APP_ID=music-app" [18:31] but then it spins loading files forever [18:31] which sounds like a bug [18:31] lool, do u actually have any music on the device? [18:32] no [18:32] lool, lp:~andrew-hayzen/music-app/library-empty-fix will hopefully fix tht :) [18:32] ahayzen: good :-) [18:32] ahayzen: building a package now [18:33] lool, needs some work to stop it flickering up when u actually do have tracks but its getting there...cool [18:41] ahayzen: so I've put some music [18:41] ahayzen: current app kind of starts on #84 now [18:42] lool, kind of? [18:42] ahayzen: there are issues in the music-scope [18:42] not relted to music-app [18:42] ahayzen: the start thing works for me [18:42] with no file [18:43] ahayzen: ah the URI thing doesn't work [18:43] lool, :/ [18:43] lool, wht happens? [18:43] ah sorry it worked on second try [18:43] not sure why [18:43] might be timing, I killed it and ran it immediately [18:44] phablet 5064 31.6 4.1 376716 78596 ? Ssl 18:44 0:04 \_ /usr/lib/arm-linux-gnueabihf/qt5/bin/qmlscene /usr/share/music-app/music-app.qml --file= --desktop_file_hint=/usr/share/applications/music-app.desktop [18:44] it's not getting the --file= thing [18:44] uhoh [18:45] ah my bad, forgot an S in APP_URI [18:46] ahayzen: ok, this works in current #84 image: start application APP_ID=music-app APP_URIS=file:///hme/phablet/Music/bilan-et-perspectives-cd-cd-03-monde-de-droite.flac [18:46] with a valid URI [18:47] lool, start: Env must be KEY=VALUE pairs....oh spaces.. [18:48] ahayzen: now trying my .deb [18:48] lool, i get application start/running, process 2396 but nothing happens on the device [18:49] ahayzen: yeah it's broken somehow [18:50] now idea how though [18:50] lool, how did u do it the first time? [18:53] mhr3: where does the existing broken apps scope live exactly? is it unity-lens-applications? or something else? [18:53] dobey, yes [18:53] unity-lens-applications [18:54] ahayzen: I hand deployed the updated qml files [18:54] lool, ok... are the core apps deployed via click package on the device rather than debian? [18:54] no [18:55] not yet [18:55] soon deployed as click though [18:55] :) [18:55] wonder wht is happening then [18:55] ahayzen: ah I know [18:55] ahayzen: you need to install qtpowerd bindings [18:55] lool, i commented out the import [18:56] ahayzen: there is code in the qml file to comment out too [18:56] ahayzen: you want qtdeclarative5-qtpowerd0.1 [18:56] lool, cause thts wht i have to do on the laptop [18:56] how is it that the dep isn't there [18:56] file:///usr/share/music-app/music-app.qml:25 module "Powerd" is not installed^M [18:56] that should be qtpowerd [18:56] ahayzen: let me update my branch [18:56] lool, music-app.qml:25 module "Powerd" is not installed is wht u get [18:57] ahayzen: yes, that's because it got renamed [18:57] lool, ah [18:58] lool, did https://code.launchpad.net/~lool/music-app/qtpowerd-rename/+merge/188728 get reverted or something? [18:58] ahayzen: try pulling my branch, I merged with trunk now [18:58] building myself a .deb [18:59] davmor2: poke? [19:00] ahayzen: so now it starts and plays the first time [19:00] lool, did u run from a phablet user inside a adb shell? [19:01] ahayzen: but when trying to change URL, it doesn't work -- because I don't have the latest upstart-app-launch [19:01] ahayzen: yes [19:01] lool, wht cmd did u use? [19:01] lool, so i can try [19:01] ahayzen: start application APP_ID=music-app APP_URIS=file:///hme/phablet/Music/bilan-et-perspectives-cd-cd-08-septembre.flac [19:01] but just start application APP_ID=music-app should work [19:04] ahayzen: so to sum up because it's a bit complex, using the latest branch revision + #84, it should start with or without URL, and it will be killed if you pass it another URL while running; if you update to upstart-app-launch packages from ppa:ubuntu-unity/daily-build it will change the playing song to the newone [19:04] lool, nothing :/ i've probably done something to the install... when i run music-app i get the debugging in the terminal but yet again nothing on screen :( [19:04] ahayzen: do you run it with start as above? [19:04] i ran phablet@ubuntu-phablet:~$ start application APP_ID=music-app [19:04] ahayzen: check ~phablet/.cache/upstart/application-legacy-music-app-.log [19:04] lool, ** (process:2887): ERROR **: Unable to submit Zeitgeist Event: GDBus.Error:org.freedesktop.DBus.Error.ServiceUnknown: The name org.gnome.zeitgeist.Engine was not provided by any .service files [19:04] lool, and ** (process:2886): WARNING **: Unable to exec: No such file or directory [19:05] lool, they are both repeated loads of times [19:05] ahayzen: the Zeitgeist one yu can ignore [19:05] ahayzen: what are the last lines? [19:06] lool, wht i posted [19:06] lool, it goes file not found then Zeitgeist [19:06] ahayzen: want to try my .deb? [19:06] lool, yh [19:07] ahayzen: http://people.dooz.org/~lool/music-app_0.7_all.deb [19:07] lool, thx [19:08] what happened to the image size ? It was like 204MB now its 248Mb what changed ? I hope we are not pulling in something useless ? [19:08] lool, same :/ probably something else i have broken [19:11] om26er: langpakcs [19:11] ahayzen: did you install stuff? [19:11] ahayzen: does the app start if you press the icon? [19:12] lool, thanks, that would make sense. [19:12] lool, the icon doesn't even exist, i've had this before after running apps from developer mode tht the main one breaks [19:13] lool, it was there a second ago but has gone now :( [19:13] hey guys, any working version for the nexus 4 outthere ? [19:15] ahayzen: that's bad [19:16] ahayzen: if you rebooted a bunch of times in write mode, you might have had fs corruption like I do often [19:16] ahayzen: maybe reflash + install this .deb [19:16] lool, it happens quite often just assumed it was to do with me using the developer mode so much to test other branches [19:16] lool, ok i'll try tht [19:18] stgraber, lool: I'm about to propose a change for adding custom fonts by dropping in a /etc/fonts/local.conf that points to /custom. It works fine, but the whole font list would be faster if /var/cache/fontconfig was writable, any objections to me adding that to writable-paths? [19:18] "faster" as determined by running fc-list [19:18] mfisch: we're running out of time [19:19] lool: we can probably do this post 13.10 if it's an issue, I know we're pretty much locked [19:19] mfisch: yeah; sorry, it's too much risk to lose cycles unbreaking stuff [19:20] mfisch: so one thing that bothers me a bit is that our read-only image contains files in there, so if we are to make this writable, we'd have to confirm first that those get properly re-generated on first boot [19:20] lool: understand, when do the floodgates re-open? [19:21] stgraber: they're being rebuilt on boot [19:21] stgraber: on every boot from what I see [19:21] mfisch: 19th or so? [19:21] lool: sure we can delay until then [19:21] mfisch: but we might need a 13.10 + changes base [19:21] mfisch: rather than 14.04 [19:22] lool: we will certainly have that. From our POV once we've ID'd a technical solution it's low risk so I can move on to something else [19:22] stgraber: I should clarify, when it's writable, it gets regenerated on reboot [19:22] but whether I do that is orthogonal to the real change I want to make, a local.conf for fonts === _PaoloRotolo_ is now known as PaoloRotolo [19:24] mfisch: ok, sounds good to me, so feel free to add to writable-paths whenever lool is happy with this landing [19:25] stgraber: can /etc/fonts/local.conf go in lxc-android-config or do we have another config-type package I can drop it in? [19:26] mfisch: lool: i do not think there is a strict need to target 13.10 [19:26] achiang: agreed [19:26] mfisch: lool: so we'd be fine to go in as part of 14.04 development [19:28] Cool [19:29] achiang, mfisch: Thanks, that's less pressure on the landing pipeline [19:29] Saviq: Hmm so I seem to have an unity8 regression in #84 with music playback: I can't see the title of songs anymore on music-scope, and these aren't listed in home scope either [19:29] looks like mediascanner isn't doing its job [19:36] lool, i just reflashed it but i think it has restored everything back cause the music app icon still doesn't exist :/ [19:38] lool: are you keeping the landing pipeline after 13.10 ships? [19:39] ahayzen: that's ... odd [19:40] ahayzen: I dont see what would be causing this [19:40] ahayzen: anyway, if you want to not keep a backup, pass --no-backup === anders3408|afk is now known as anders3408 [19:40] lool, i'll try tht cause i seem to have all my music as well [19:40] mfisch: Not for 14.04, not in this way; don't know about 13.10 updates [19:40] ahayzen: Hmm it sounds like you haven't really flashed [19:40] lool, i can see it restoring everything back at the end of the process [19:41] lool, trying now with the extra arg [19:41] sergiusens, any idea why I wouldn't be able to flash my device from the 'jenkins' user on my local machine? [19:41] fginther, use sudo; you need to have the seat [19:41] fginther: android-tools sets up some udev rules for the active user [19:42] fginther, so I mean, either sudo or active user [19:42] sergiusens, lool, ah. thank you [19:42] popey: we should consider infographics data for core apps 2.0 features [19:47] mhall119: good call [19:47] lool: tests finished? [19:50] popey: looking! [19:50] popey: do you put music up there? [19:50] popey: I seem to lack song names on the scopes, do you? [19:50] popey: there's something wrong, the passrate is way too high [19:52] lool, music app is back! [19:52] ahayzen: cool! [19:52] ahayzen: Quick! break it! ;-) [19:52] ahayzen: /me crosses finger on the .deb [19:52] lool, ok i left the normal music-app there and ran start application APP_ID=music-app and tht worked [19:53] lool, now to copy ur deb [19:55] lool: just copyng some music & videos now [19:56] popey: pmcgowan says webapps don't work [19:56] indeed twitter doesn't come up [19:56] hmm, missed that [19:57] yup, confirmed [19:57] lool, right tht worked! now to put some music on and actually test wht we were supposed to be testing :) [19:58] lool, it started the app but didn't play it :/ [19:59] hello, how do you make the ubuntu-system flash read/write? (as it seems to be read only by default) [20:00] touch /userdata/.writable_image [20:00] then reboot [20:00] thanks! [20:00] you lose the ability to do OTA updates when you do that though [20:00] that shouldn't be much of a big deal, im trying to test out a package :) [20:00] popey, lool what changed? [20:01] pmcgowan: can't see what would affect it here http://people.canonical.com/~j-lallement/touch/changes/20131007.html [20:03] popey, new browser went in friday [20:04] alex-abreu, webbrowser-app tells me it doesnt know the enable-back-forward or webapp arguments [20:06] pmcgowan: http://people.canonical.com/~ogra/touch-image-stats/20131007.changes === diegoyam_ is now known as diegoyam [20:06] pmcgowan: twitter works on the image we put out today though [20:07] built on saturday iirc [20:07] so #83 works and #84 doesn't work [20:07] yes [20:07] are these click? [20:07] click-apparmor changed [20:08] yes [20:08] [10426.055942] type=1400 audit(1381175623.667:90): apparmor="DENIED" operation="open" parent=1550 profile="com.ubuntu.developer.webapps.webapp-twitter_webapp-twitter_1.0.3" name="/android/system/etc/media_codecs.xml" pid=7917 comm="webbrowser-app" requested_mask="r" denied_mask="r" fsuid=32011 ouid=0 [20:08] i assume so given it's in /usr/share/click/preinstalled/com.ubuntu.developer.webapps.webapp-twitter [20:08] jdstrand: click-apparmor regression it seems [20:08] that is not because of click-apparmor [20:08] pmcgowan: ^ [20:08] lool: no [20:08] lool, not so sure [20:08] that is underlying changes to the system [20:08] why doesnt the browser take the webapp switch? [20:08] alex-abreu, ^^ [20:08] the pending apparmor-easyprof-ubuntu package will have a fix for that denial [20:09] (click-apparmor doesn't ship any policy) [20:09] also, I'm pretty sure the app works without that denial being fixed (I saw it friday) [20:10] what is the failure? [20:10] how can i launch pre-installed apps via upstart? [20:11] popey: should be the same: start application APP_ID=... [20:11] popey: turns out my LP email settings didn't have be subscrived to the avengers list, I should start getting those emails now :) [20:11] popey: the trick is knowing the APP_ID [20:11] yeah, its the .desktop file AIUI [20:12] iirc, it is everything before .desktop [20:12] but maybe it includes .desktop, not sure otoh [20:12] lool: what is the failure? [20:12] jdstrand: blank screen when launching twitter app [20:12] nothing when launching facebook app [20:12] so start APP_ID=com.ubuntu.developer.webapps.webapp-twitter [20:13] what goes in the gap? [20:13] jdstrand: I'm getting other failures [20:13] popey: 'application' [20:13] jdstrand: sorry other denials [20:13] [10425.405675] type=1400 audit(1381175623.016:88): apparmor="DENIED" operation="open" parent=1550 profile="com.ubuntu.developer.webapps.webapp-twitter_webapp-twitter_1.0.3" name="/home/phablet/.cache/gstreamer-1.0/registry.arm.bin" pid=7917 comm="webbrowser-app" requested_mask="r" denied_mask="r" fsuid=32011 ouid=32011 [20:13] jdstrand: I wonder if it's due to new thumbnailer [20:13] ah [20:13] duh [20:13] pmcgowan: ^ [20:13] lool: that denial is harmless [20:13] lool: that came because of the new gstreamer [20:13] pmcgowan, it should ... all webapps should roughly be "--webapp --enabled-back-forward" etc. [20:13] jdstrand: we did add a thumbnailer plugin to SDK [20:13] and also: [10425.411657] type=1400 audit(1381175623.026:89): apparmor="DENIED" operation="exec" parent=7917 profile="com.ubuntu.developer.webapps.webapp-twitter_webapp-twitter_1.0.3" name="/usr/lib/arm-linux-gnueabihf/gstreamer1.0/gstreamer-1.0/gst-plugin-scanner" pid=7934 comm="webbrowser-app" requested_mask="x" denied_mask="x" fsuid=32011 ouid=0 [20:14] lool: but I will have it silenced in the next apparmor-easyprof-ubuntu upload [20:14] alex-abreu, if I use those options browser says it doesnt know them, if I remove them then it runs [20:14] lool: that last one I think needs to be fixed. also because of gstreamer [20:14] man, this gstreamer update is a serious pain [20:15] jdstrand, we should be backing it out tomorrow, at least from the sdk [20:16] pmcgowan, doesn't make sense at all, this code hasn't changed for some time already ... having a look [20:17] pmcgowan, upi launch it directly thru ssh ? [20:17] upi=you ... [20:17] alex-abreu, I ssh in and run webbrowser-app [20:17] webbrowser-app --help, or --webapp [20:17] yeah [20:18] with desktop file jint [20:18] hint [20:18] lool: you are testing that manually? [20:18] jdstrand: No, this is what I found in logs when running twitter app [20:19] jdstrand: didn't know what regressed, there was a click-apparmor update, I thought I'd blame it on you as usual ;-) [20:19] lool: you didn't launch the twitter app on your system? [20:19] jdstrand: I launched the twitter app from image #84, it didn't start, I grep-ed dmesg on mako and found these [20:20] I see [20:20] yes, some are harmless. one I think may not be [20:20] but it seems it's a webbrowser-app cmdline flag [20:20] jdstrand: the gst-plugin-scanner thing I think is called when the registry is missing to update it, or when the registry is out of date [20:20] ok, well, that's good, but I am trying to get the apparmor and apparmor-easyprof-ubuntu tested and ready so all these will go away [20:21] jdstrand: it's in the home dir, so can't pregenerate on package install; the first gstreamer app starting generates it [20:21] lool: I have grouper with 83-- is just apt-get dist-upgrading going to get me what is in 84? [20:21] jdstrand: it should, might pull in slightly more stuff though [20:22] lool: re pregenerating-- fyi, apps are allowed to execute that but the aren't allowed to write out the registry, but can read the registry [20:23] ok [20:23] it's worked fine until now-- I think the denial is harmless [20:23] but I will verify [20:24] nik90: ping [20:28] lool, alex-abreu not sure how this could be, but it seems like an old binary of the browser app [20:28] popey: does shorts refresh work for you? [20:28] * popey checks [20:28] installing directly from the archive to my desktop has the new options [20:28] mine seems to get stuck spinnings it's wheels but going nowhere [20:28] works here [20:28] on #84 [20:29] hmm, spins on #83 [20:30] mhall119: i also get a hang on refresh [20:30] on #84 [20:30] * mhall119 is on #83 [20:31] now it works [20:31] it worked on first launch, then hung on refresh (didn't spin), then worked on subsequent runs + refreshes [20:35] so I guess I need to wait for build >= #84 to be promoted [20:38] pmcgowan: I can get twitter if I run: webbrowser-app --enable-back-forward --webapp --webappUrlPatterns=https?://mobile.twitter.com/* https://mobile.twitter.com/ --desktop_file_hint=/usr/share/applications/webbrowser-app.desktop [20:38] mhall119: or take it from -proposed [20:38] mhall119: probably wont be promoted since we have a webapp regresion [20:38] *regression [20:38] lool, I cannot and neither can alex-abreu [20:38] lool: I'll wait then, I'm supposed to only use stable releases for the avengers stuff [20:39] lool, what version of the image? and browser package? [20:39] pmcgowan: I just ran this on #84 from adb shell sudo -i -u phablet [20:39] webbrowser-app 0.22+13.10.20131004.1-0ubuntu1 [20:39] mhall119: ok [20:39] mhall119: we're about to enter Mir mode [20:39] might not be a stable image for a while [20:39] \o/ [20:39] probably switching tomorrow morning [20:39] how the lightdm stuff, has that landed? [20:39] lool, humor me, if you run webbrowser --help do you see all those options for webapp [20:40] pmcgowan: no [20:40] I see: Usage: executable [--form_factor_hint={desktop, phone, tablet}] [--stage_hint={main_stage, side_stage, share_stage}] [--desktop_file_hint=absolute_path_to_desktop_file] [20:40] thats not new stuff [20:40] pmcgowan: this is with webbrowser-app --help [20:40] wrong code [20:40] pmcgowan: http://people.canonical.com/~platform/cu2d/results has an updated staged [20:41] but seems only like a couple of commits [20:41] lool, did it report unknown options on the command line [20:41] pmcgowan: oh yes [20:41] indeed [20:42] this is the output from same package on the desktop [20:42] http://pastebin.ubuntu.com/6206639/ [20:42] pmcgowan: so I guess I'm not seeing what I should be seeing; just the twitter site in browser [20:42] the binaries are 10MB different [20:42] that paste matches the source code [20:42] how could that be? [20:42] indeed [20:43] that's interesting [20:44] this binary is a couple months old maybe [20:44] sergiusens, ^^ [20:44] or at least several weeks [20:46] pmcgowan: wasn't there some unconfined wrapper? [20:46] pmcgowan: you remember that unconfined click thing [20:48] lool, we didnt do that [20:48] lool, seems to be the package build failed and copied old bits or something inexplicable [20:48] seems to me [20:48] lool, we are just using the deb and having an icon link [20:50] pmcgowan: https://launchpadlibrarian.net/152489739/buildlog_ubuntu-saucy-armhf.webbrowser-app_0.22%2B13.10.20131004.1-0ubuntu1_UPLOADING.txt.gz has same code size has the one on the image [20:51] Kaleo, dont suppose youre around? [20:52] pmcgowan: also: grep webappUrlPatterns /usr/bin/webbrowser-app matches [20:52] lool, ok so that part was right, so? [20:53] lool, oh you mean in the binary [20:53] hmm [20:53] for me it's more a startup sequence problem [20:53] like some different init path depending on the qpa [20:54] or perhaps misbuilt on armhf [20:54] lool, heres the thing, I run /usr/bin/webbrowser-app and its command line parsing is not the same as the desktop [20:54] I cannot explain that from environment [20:55] and it provides an actual response which matches old source code [20:56] lool, what am I missing? [20:56] lool, pmcgowan yeah that's what I am thinnking although I don't know what can be put in between [20:57] no aa-click since we are launching manually [20:57] so there must be something [20:57] pmcgowan, oh interesting, try "webbrowser-app -h" [20:58] alex-abreu, doesnt work for me on phone [20:58] pmcgowan, ? no output [20:59] no core dump [20:59] works on desktop as expected [21:00] pmcgowan, I get this when using -h http://pastebin.ubuntu.com/6206726/ and your (desktop_hint etc.) output when using --help [21:00] alex-abreu: ah right, -h works [21:00] interesting [21:01] lool, any idea what can be "hijaking" the launch ? [21:01] no, I tried stracing, but got limited success [21:01] or rather I failed utterly [21:03] alex-abreu, but whats running instead? and from where? [21:03] is there a bad click app installed or something? [21:03] no idea [21:04] I am thinking more about e.g a conflicting plugin path, or some sort of qpa issue [21:04] I see a bunch of additioanl libs on the touch version [21:05] ranging from /system/lib/libwpa_client.so to /usr/lib/TRIPLET/libgflags.so.2 [21:06] lool, so something again with the toolkit? [21:06] lool, as runtime deps for webbrowser-app ? I don't see anything in ldd's output [21:07] pmcgowan: curious-- is it all webapps or just twitter? [21:07] jdstrand, all [21:10] jdstrand, pmcgowan not even webapp but webbrowser-app apparently [21:10] that is weird [21:10] indeed [21:11] are the md5sums correct? [21:12] guys I need to drive home before some storms hit, looking forward to your analysis! [21:15] but browser app itself works [21:15] ah /usr/lib/TRIPLET/qt5/plugins/platforms/libqubuntu.so is in there [21:15] lool: note, this denial is bug #1235444 : [21:15] Oct 7 21:13:45 ubuntu-phablet kernel: [ 536.137702] type=1400 audit(1381180425.183:115): apparmor="DENIED" operation="mkdir" parent=1364 profile="com.ubuntu.developer.webapps.webapp-twitter_webapp-twitter_1.0.3" name="/home/phablet/.cache/com.ubuntu.developer.webapps.webapp-twitte/" pid=2528 comm="webbrowser-app" requested_mask="c" denied_mask="c" fsuid=32011 ouid=32011 [21:15] bug 1235444 in thumbnailer (Ubuntu Saucy) "pkg_name calculated incorrectly" [Critical,New] https://launchpad.net/bugs/1235444 [21:15] lool: Satoris is aware of it, there is a patch in a branch [21:16] alex-abreu: I wonder if it's platform-api, but it didn't change [21:17] lool, that's what I was thinking it must be something like that [21:17] alex-abreu: cause it's doing all this complex hybris stuff, parsing cmdline too [21:17] alex-abreu: also, the output is suspiciously like it [21:17] alex-abreu: {"form_factor_hint", required_argument, NULL, uninteresting_flag_value}, [21:17] {"stage_hint", required_argument, NULL, uninteresting_flag_value}, [21:17] yes [21:18] {"desktop_file_hint", required_argument, NULL, uninteresting_flag_value}, [21:18] lool, alex-abreu: fyi, I dist-upgrade grouper and the twitter app launches [21:18] is this only mako? [21:18] (or some other device?) [21:18] alex-abreu: see src/ubuntu/mircommon/application_options_mir.cpp in platform-api or android/default/default_ubuntu_application.cpp [21:19] my test was surfaceflinger btw [21:19] mfisch, ssweeny, cwayne_, lool: I have now landed a fix that makes any missing file a critical failure for an image. That'll prevent breakage of customized images when Jenkins goes down (that also means that whenever something like this happens again, no updates will be pushed to that channel until all files are reachable again) [21:19] android/hybris/ubuntu_application_api_for_hybris.cpp has that actually [21:19] stgraber: cool [21:19] stgraber: is firewall fixed? [21:19] no [21:19] stgraber: do you think you could take a backup of our current set of images? [21:19] stgraber: we'd like to keep the last good SF images "forever" [21:19] stgraber, awesome,thanks [21:19] stgraber: perhaps we can do a channel copy to a saucy-sf? [21:19] and never purge it [21:19] stgraber, any update on that RT by any chance? [21:20] lool: I'd rather not waste space, we don't have so much of that on the public servers [21:20] alex-abreu: "executable" strikes me as coming from android/hybris/ubuntu_application_api_for_hybris.cpp [21:20] cwayne_: no updates, ... [21:20] now who calls u_application_options_new_from_cmd_line [21:20] stgraber: where could we keep it? [21:20] lool: so however wants those around will have to provide me with a place to put them in, system-image is not an archiving server [21:21] lool, indeed ... the output matches quite nicely [21:21] stgraber: I'm worried they will be hard to write to devices if they aren't on system-image.u.c [21:21] lool: I have a mirror of devel-customized and devel-propoosed-customized [21:21] symbol=u_application_options_new_from_cmd_line; lookup in file=webbrowser-app [0] [21:21] stgraber: I also have a small shell script to do the mirror, but it's all wget based [21:22] mfisch: recursive wget works fine for mirroring, everything is staatic [21:22] alex-abreu: so it's /usr/lib/TRIPLET/qt5/plugins/platforms/libqubuntu.so [21:22] stgraber: well I only wanted mirror certain builds and so I parse the json and pull what I need [21:23] stgraber: a full mirror is much simpler obviously [21:23] lool, might "just" be, ... mmh does platform-api do something to the args before "passing them down" (not sure what the mechanism is here to the app ? [21:23] lool: what annoys me the most with keeping those on system-image (besides the fact that the service was never meant for that) is that it means some of the bugs I fixed last week won't ever be fixed [21:24] lool: as the bugfix depends on the fact that all existing files will one day be expired and so the bug will clear itself within 25 images. Keeping references to those in a channel will prevent that from happening. [21:24] alex-abreu: that's what I'm trying to track down [21:25] lool: so if it's short term with a fixed removal date (say 2 months from now), fine, if not, then they'll have to be saved somewhere else [21:25] alex-abreu: so qtubuntu/src/platforms/ubuntu/ubuntucommon/integration.cc does this in QUbuntuIntegration::QUbuntuIntegration(QUbuntuInputAdaptorFactory*) [21:25] lool, I am pretty sure that it does ... and wrongly ... from the output [21:25] alex-abreu: I think it's meant to pick up its own options, but ends up hijacking things it shouldn't [21:26] lool, there is an extra space apparently ... that is being introduced somehow ... e.g. '--enable-back-forward' -> '-- enable-back-foward' [21:27] mmh or not [21:27] alex-abreu: Might just be adb playing tricks [21:27] yeah [21:27] alex-abreu: I think the logic leap is that it's meant to pick up just its options, but interprets far too much [21:27] alex-abreu: e.g. rejects unknown options and takes over --help [21:28] yes [21:28] in fact it exit(EXIT_SUCCESS); [21:30] alex-abreu: but this must have been going on for months? [21:30] lool, I don't think that it rejects other options though ... the "unknown option -- " does not seem to come from the webbrowser-app ... [21:30] so webbrwoser-app seems to take it [21:31] alex-abreu: right, --help is a bad test [21:31] lool, when I try w/ a bad option e.g. --enable-whatever [21:31] alex-abreu: We're just diving into why --help is different when really it's just a specific case [21:31] you can see the warning message from webbrowser-app [21:31] lool, yes [21:32] alex-abreu: I get the same: WARNING: ignoring unknown switch "--foo" [21:32] in both cases [21:32] unknown option -- webapp [21:32] that's interesting [21:32] lool, there is a slight issue there w/ --help being eaten, but the initial thing was that Pat could not launch any webapp on his image [21:33] lool, I don't think that its is of interest either [21:33] not sure where it comes from though [21:33] couldn't find anything in platform-api [21:33] alex-abreu: isn't --webapp meant to be supported? [21:33] lool, yeah, but this message does not come from webbrowser-app I think [21:34] comes from someone hooking into args somewhere [21:34] since webbrowser-app emits a warning when there is an arg parsing filure [21:34] failure [21:34] OUCH [21:34] oops [21:35] massive failure [21:35] alex-abreu: /usr//lib/arm-linux-gnueabihf/libQt5Core.so.5.0.2 was updated on 7t [21:35] 7th [21:35] and has that string [21:35] pretty supsivious [21:36] mmh [21:37] alex-abreu: so we changed QSortFilterProxyModel in latest qtbase upload [21:37] lool, I cannot test webapps yet though ... wifi on the phone does not seem to work, or at least does not persist my wifi settings ... is it a known issue? [21:37] lool, yup [21:37] alex-abreu: wifi should work [21:37] and should persist your settings [21:38] I would be surprised that it does not, but something is odd [21:38] lool, i'm back... running start application APP_ID=music-app APP_URIS=file:///hme/phablet/Music/01.\ Mylo\ Xyloto.flac doesn't start the track in the app :/ but it does open the app which is an improvement [21:38] ahayzen: try %20 instead of "\ " [21:38] lool, ok :) [21:38] ahayzen: also it's home not hme [21:38] lool, whoops [21:38] ahayzen: the adb shell is tricksy; ssh is safer; "sudo start ssh" then ssh phablet@ip address [21:38] password: phablet [21:39] alex-abreu: let's revert to an older qt5core [21:39] lool, start application APP_ID=music-app APP_URIS=file:///home/phablet/Music/01.%20Mylo%20Xyloto.flac works! [21:40] lool, now wht do i need to test? should i be able to send 2 URIs and it queue the second? [21:41] lool, only change was indeed qsortfilterproxymodel [21:41] ahayzen: the 2 URI case is supported, but is not used in real life; also only works if you have updated upstart-app-launch [21:41] ahayzen: so you should try adding upstart-app-launch binaries from ppa:ubuntu-unity/daily-build [21:41] alex-abreu: I sponsored it :-) [21:43] lool, ah ok network is a bit poor on the device though :P [21:43] lool, as alex-abreu said the WiFi just drops out all the time [21:43] ahayzen, oh I am not the only one ? [21:43] alex-abreu: downgraded just qt5core and still getting unknown option -- webapp [21:44] lool, :/ ... [21:44] ahayzen: you have to keep it from sleeping to get good network [21:44] am I correct in assuming that to rollback and flash android, it has to be done through an ubuntu distro? [21:44] "powerd-cli active" in a shell [21:44] alex-abreu, mine drops out all the time but them i am to a hotspot from a laptop [21:44] alex-abreu: but might be another packge [21:44] alex-abreu, android doesn't even see the hotspot because it is adhoc not infrastructure :P [21:44] lool, yah ... [21:45] well....you can flash it outside Linux but you can't rollback outside it seems [21:46] btw...what in the hell is the default password? [21:48] p8triot, same as user [21:48] p8triot, phablet [21:49] thanks!!! where is that in the documentation? I didn't see it [21:50] jdstrand, hey, what do I need for this one? http://paste.ubuntu.com/6206914/ [21:50] I am running "saucy" is their a more stable rev or is this it? [21:50] sergiusens: friends policy group. will be in apparmor-easyprof-ubuntu 1.0.35. note, it is a reserved policy group (ok for trusted apps) [21:50] jdstrand, gallery app works fine as click (even hud as a surpise) , you can check the app out from http://10.97.0.26:8080/view/click/job/gallery-app-click/lastSuccessfulBuild/artifact/wrapped_click/com.ubuntu.gallery_2.9.1_armhf.click === jhodapp is now known as jhodapp|afk [21:51] sergiusens: nice! :) [21:52] jdstrand, great, I'll add the policy as soon as that's in [21:52] I am running "saucy" is their a more stable rev or is this it? [21:52] alex-abreu: downgraded everything and still getting: unknown option -- webapp [21:53] sergiusens: I'm trying to coordinate the landing now [21:54] lool, I don't think that *that* "unknown option" message comes from libqt5core ... I grepped qtbase src and found a few 'unknown option' but format differs from the one we see [21:54] alex-abreu: maybe the unknown option thing was alwyas there though [21:54] maybe [21:55] alex-abreu: ok, wrong track IMO [21:55] alex-abreu: just stared it from a shell now [21:55] alex-abreu: aa-exec-click -p com.ubuntu.developer.webapps.webapp-twitter_webapp-twitter_1.0.3 -- webbrowser-app --enable-back-forward --webapp --webappUrlPatterns=https?://mobile.twitter.com/* https://mobile.twitter.com/ [21:56] alex-abreu: getting: [21:56] (webbrowser-app:11908): GStreamer-WARNING **: External plugin loader failed. This most likely means that the plugin loader helper binary was not found or could not be run. You might need to set the GST_PLUGIN_SCANNER environment variable if your setup is unusual. This should normally not be required though. [21:56] ERROR: Caught a segmentation fault while loading plugin file: [21:56] /usr/lib/arm-linux-gnueabihf/gstreamer-1.0/libgstandroidmedia.so [21:56] so looks like the plugin db is the first thing I'd fix [21:56] anyone have any luck with the clock app and getting the alarm to stick? I set it and then it goes away [21:56] alex-abreu: also, that's in recent hybris code [21:57] -rw-r--r-- 1 root root 80308 Oct 3 17:53 /usr/lib/arm-linux-gnueabihf/gstreamer-1.0/libgstandroidmedia.so [21:57] rsalveti: hey [21:57] lool, yeah [21:57] lool: hey [21:57] lool, have to disconnect for small bit to handle the kids [21:57] rsalveti: launching: aa-exec-click -p com.ubuntu.developer.webapps.webapp-twitter_webapp-twitter_1.0.3 -- webbrowser-app --enable-back-forward --webapp --webappUrlPatterns=https?://mobile.twitter.com/* https://mobile.twitter.com/ [21:57] rsalveti: get us above warning and segfaults [21:58] rsalveti: actually hold on, can just try unconfined [21:58] I wanted to generate the registry, but even there it doens't help [21:58] lool: yeah, that would be my question :-) [21:58] as I tested just the browser, unconfined, and it works fine [21:59] it'd only load gst in case it's actually playing audio/video [21:59] alex-abreu: it starts outside of aa [21:59] jdstrand: so I do blame you! [21:59] anyone have any luck with the clock app and getting the alarm to stick? I set it and then it goes away [21:59] jdstrand: ah I knew I'd get you one day! [21:59] but it might init at least, which would cause these calls, to generate the registry.arm [21:59] rsalveti: aa prevents it [22:00] rsalveti: jdstrand said he would fix this [22:00] well [22:00] so the webapp template doesn't allow exec of gst-pugin-scanner [22:00] that I can fix [22:01] I'm a bit surprized this started occuring in #84 though [22:01] jdstrand: currently, all webapps are broken on image; if you had a fix like now, that would be cool [22:01] jdstrand: or something you'd upload later today and I'd include in image tomorro [22:01] ww [22:02] rsalveti: currently I have the following apparmor rules: http://paste.ubuntu.com/6206940/ [22:02] anyone have any luck with the clock app and getting the alarm to stick? [22:03] rsalveti: now, granted, the app can't currently exec gst-plugin-scanner [22:03] rsalveti: but are you saying that if the registry files don't exist, it won't fail gracefully? [22:03] well, it'd try to create it, but not sure why it crashed [22:03] the crash might be a bug as well [22:03] lool: I have an apparmor-easyprof-ubuntu upload with a bunch of fixes in it [22:04] this has been working for months [22:04] lool: I'd guess that this would work in case you run that test after testing mediaplayer-app [22:04] rsalveti: not sure, i ran it unconfined and it still crashes on startup [22:05] heh [22:05] hm, nothing changed in the gst side [22:05] jdstrand, rsalveti: I think the regression is addition of hybris [22:05] not sure [22:05] lool: which image are you using? [22:05] /usr/lib/arm-linux-gnueabihf/gstreamer-1.0/libgstandroidmedia.so is 3rd of October [22:05] rsalveti: #84 [22:06] lool: how to launch it unconfined? [22:06] ok, I deleted the registry files, but kept the rules to deny the writes [22:07] the app started fine [22:07] and can play audio [22:07] right [22:07] yeah, it just need to write the registry file [22:07] that's part of the gst init process [22:07] well, that is what I'm saying [22:07] I allow the exec but deny the write [22:07] no crash [22:08] (and no write [22:08] ) [22:08] rsalveti: remove the part until webbrowser-app, add --desktop_file_hint [22:08] on, no write at all, ok [22:08] yeah, ~/.cache/gstreamer-1.0 is empty [22:08] rsalveti: but it seems other apps cope better with this, and I see the code is regitering a bunch of codecs [22:08] I have a few more hardware denials as a result. let me check that [22:09] rsalveti: anyway, happy with an apparmor update [22:09] jdstrand: without the registry it might not necessarily use the right codecs, not yet sure though [22:09] lool: it could also be that in other tests the registry files were present already so the writes didn't have to happen [22:09] yeah [22:09] I'll create achild profile [22:09] jdstrand: but the registry file *is* present here [22:09] I ran this unconfined [22:09] then confined [22:09] and it crashes [22:10] lool: can you remove it and run unconfined again? === _salem is now known as salem_ [22:11] rsalveti: confirmed: removed, launched unconfined, launched confined => crash [22:11] ERROR: Caught a segmentation fault while loading plugin file: [22:11] /usr/lib/arm-linux-gnueabihf/gstreamer-1.0/libgstandroidmedia.so [22:13] what I'm a bit sad about all of this is that I've been trying to include the webapps testsuites for weeks [22:13] again not later than today [22:13] couldn't add them because they pull compiz [22:16] no crash when unconfined, let me test with aa-exec now [22:16] does someone have #83? [22:17] crashes all the time when launching confined, without a previous registry.arm file [22:17] rsalveti: probably not able to open it though [22:17] or is it? [22:18] it's trying to generate it [22:18] open is DENIED on /home/phablet/.cache/gstreamer-1.0/registry.arm.bin in my logs [22:18] might be because that plugins try to open /system/etc/media_codecs.xml [22:18] rsalveti: that's also DENIED [22:19] rsalveti: /lastlog DENIED to see the 4 [22:19] that plugin tries to open /system/etc/media_codecs.xml to decide what is supported in the android side [22:19] right, and that's probably what is causing the segfault [22:19] jdstrand: will you allow reading this file? otherwise plugins will be misdetected [22:19] jdstrand: /system/etc/media_codecs.xml [22:20] lool, if 83 is the latest stable I think I have it [22:20] sergiusens: could you check whether twitter works? [22:21] lool, works [22:21] lool, it's my dogfooding device btw [22:21] lool: I already am in the next upload [22:22] perhaps that is why it is working here [22:22] right, will test after the next upload then [22:22] yeah :-) [22:22] lool, so it's ro, never broke the rules device :-) [22:22] * jdstrand is testing his upload [22:22] brb [22:22] rsalveti, jdstrand: I wonder if the regression was due to thumbnailer being added [22:22] lool: what do you mean by “Works”? [22:22] new package: libthumbnailer0:armhf [22:23] nhaines: ? [22:23] nhaines: whether the webapp opens [22:23] it does [22:23] apparently [22:23] lool: oh, that works then. [22:24] Jusy tested it. :) [22:24] rsalveti, jdstrand, alex-abreu: So what I think happened: ui-toolkit grows thumbnailer dependency; that scans for gstreamer plugins on startup; results in a segfault under confinment since addition of gst-hybris plugin, but other apps like mediaplayer weren't as confined and hence weren't affected [22:24] we should have catched that by running the webapps testsuite [22:25] which I filed a bug on === RAOF_ is now known as RAOF [22:27] lool, mmh how come this wasn't caught outside of webapps testsuite (e.g. other app testsuite) ? [22:27] lool, also which webapps testsuite are your refering to ? :) [22:29] alex-abreu: other apps don't use multimedia I guess? [22:29] not sure why not actually [22:29] alex-abreu: unity-webapps-qml-autopilot [22:31] alex-abreu, lool you guys said it above, confinement [22:31] lool, aren't those for 'desktop' webapps? [22:32] lool: ok, I'm testing apparmor-easyprof-ubuntu with a fix for this [22:32] sergiusens: also seeded in touch === salem_ is now known as _salem [22:32] so washoping the testsuite would work [22:33] lool: I'll let you know how it goes. fyi, this upload is required for two other landing plans, so getting this into the image will be good [22:33] jdstrand: ok [22:37] jdstrand: I've added a landing slot for it [22:38] mzanetti: Around? [22:39] ahayzen: so no luck testing this? [22:39] lool, my device refuses to connect to WiFi :/ so it won't add the PPA [22:40] lool, it seems to get it for a few seconds then drop out === JanC_ is now known as JanC [22:43] lool, this is not a test suite for webapps, well yes, but not in the environment of e.g. confined webbrowser-app (which pulls uitk) [22:43] lool, I think that dbarth was working on a webapps-demo test suite [22:43] ahayzen, I get this too [22:43] lool, u say in the merge 'the track queue is the whole library; will file a bug on this one' it is expected that the whole library is initially the play queue [22:43] ahayzen: even with the powerd-cli active tip? [22:43] lool, how do i do tht? [22:44] ahayzen: you run "powerd-cli active" in some adb shell or ssh window [22:45] lool, nope [22:45] lool, its more like it isn't getting an IP or something [22:45] ahayzen: you could download the packages on your desktop and adb push them [22:45] ahayzen: https://launchpad.net/~ubuntu-unity/+archive/daily-build/+packages?field.name_filter=upstart-app-launch&field.status_filter=published&field.series_filter=saucy [22:45] lool, thanks :) [22:46] ahayzen: you want the armhf ones, but not all [22:46] ahayzen: upstart-app-launch, upstart-app-launch-tools, libupstart-app-launch1 [22:46] lool, ok [22:46] -tools is optional actually [22:48] lool, right they are installed now wht am I testing? [22:49] ahayzen: so you start music-app from my branch in anyway [22:49] *in any way [22:49] ahayzen: or actually, start it with "start" as earlier [22:49] ahayzen: and then you run "start" again, with some APP_URIS [22:49] ahayzen: this should change the playing track to the URI(s) you pass there [22:50] lool, ok how do u do multiple? [22:50] ahayzen: separate by space [22:50] ahayzen: e.g. APP_URIS=file:///xyz file:///abc" [22:50] sorry APP_URIS="file:///xyz file:///abc" [22:54] lool, ok i started playing track A then told it to play track B,C and it cleared the queue and started playing B with C in the queue [22:54] Hi, I am new to ubuntu-touch. I am interested ubuntu-touch (my work will be on PulseAudio). Can somebody tell me the ARM based baords used for this [22:55] lool, but the bug u were talking about is tht the fact then when I played A ... B and C were in the queue? [22:56] ahayzen: nah [22:56] ahayzen: oh yes [22:56] ahayzen: correct [22:56] ahayzen: that _when starting with an URL_ the whole DB is in the queue [22:56] ahayzen: but that's a different code path [22:56] ahayzen: I would personally expect it to play just the one I played [22:56] I mean the URL I open [22:57] on startup, and on subsequent opens [22:57] ahayzen: that other code path is when parsing the cmdline arg [22:57] lool, u'll probably just need to modify... if (trackQueue.model.count === 0 && !argFile) [22:58] lool, to make sure that it is false when there is a file sent cause tht is where the library is loaded in the queue [23:00] ahayzen: I dont really want to touch this bug now though [23:00] ahayzen: I need to land the URL opening support [23:01] ahayzen: would you mind top approving the merge proposal if it works for you? [23:01] lool, ok think Daniel was working on tht area before i'll have a poke around see whts going on [23:01] lool, yep :) [23:02] lool, could u add a commit msg? [23:07] ahayzen: done [23:07] lool, thanks [23:07] ahayzen: thanks for reviewing [23:07] lool, i'll top approve so it lands [23:07] cool [23:08] lool, we got there eventually :) so the other issue is when using --file ? [23:09] lool, i'll have a look at tht and see whts going on [23:13] ahayzen: cool [23:13] ahayzen: yeah, it's pretty similar; the queues aren't setup right IMO [23:13] ahayzen: I think the argFile code should do something similar, but ran out of time to do it, and it was unrelated [23:14] lool, yeah [23:15] Ubuntu mobile questions | http://askubuntu.com/q/355084 [23:23] lool, think i've fixed it i'll push the branch [23:25] ahayzen: Great [23:29] lool, https://code.launchpad.net/~andrew-hayzen/music-app/file-args-queue-fix/+merge/189739 [23:29] ahayzen: ah that's exactly how I think it would look like, did it work in your testing? [23:29] /69 [23:30] bah [23:30] lool, i did $ qmlscene music-app.qml --file="file:///home/andy/Music/Coldplay/Mylo Xyloto/01. Mylo Xyloto.flac" and it worked [23:30] ahayzen: Ok; my device is running tests right now, I'll approve it now and top approve it when I can test [23:31] lool, ok thx :) [23:31] Is it just me, or am I missing a way to type straight quotes (") on Ubuntu Touch? [23:33] nhaines: yup [23:34] I am slightly annoyed by this. [23:34] * wilee-nilee looks for their worlds smallest violin [23:35] nhaines: ah, thanks for asking I was wondering the same :) [23:55] cyphermox: you're welcome. Just realized this monrning. === anders3408 is now known as anders3408|afk