| doflaherty | yeah, it's purely cosmetic | 00:01 |
|---|---|---|
| doflaherty | that makes sense | 00:01 |
| === japaric_ is now known as japaric | ||
| === chriadam|away is now known as chriadam | ||
| dobey | who knows how to debug why certain attributes are not appearing in /sys/class/power_supply/battery/ ? | 02:26 |
| === chihchun_afk is now known as chihchun | ||
| === vying is now known as Guest21178 | ||
| === duflu_ is now known as duflu | ||
| dholbach | good morning | 07:04 |
| === chriadam is now known as chriadam|away | ||
| ybon | I can't get to make Internet work since some images (devel, mako, Nexus 4), is it just me or is it everyone? | 07:39 |
| JamesTait | Good morning all; happy Leprechaun Day! :-D | 08:01 |
| * ogra_ hides his pot full of gold under the desk | 08:02 | |
| davmor2 | Morning all | 08:29 |
| ybon | One more question: it seems that recently, when I try to update the sound level from the top settings, the device freeze for a long. Anyone knows something about this? | 09:04 |
| popey | ybon: known bug about the sound indicator, it's laggy | 09:08 |
| popey | ybon: morning btw | 09:09 |
| ybon | morning popey :) | 09:09 |
| ybon | and thanks for the info :) | 09:09 |
| ybon | popey: I've bought another Nexus 4, for my girlfriend, so I'm a little more looking at details of usability ;) | 09:10 |
| ogra_ | Tassadar_, thanks a lot for the pointer !! | 09:11 |
| popey | ybon: excellent. | 09:11 |
| ogra_ | teach her to use the volume button for the time being :) | 09:12 |
| ybon | ogra_: :) | 09:16 |
| ybon | The fact is that it **seems** that the phone is sometime ringing sometime on a call or message | 09:16 |
| ybon | and I'm trying to investigate | 09:16 |
| ybon | (from one phone call to the other, not that I'm full time on it) | 09:17 |
| ybon | sometimes ringing sometimes not* | 09:17 |
| ogra_ | yeah, there are issues ... davmor2 or popey should know the bug number | 09:18 |
| ogra_ | btw, do we have a bug for the non-working alarms/meeting notifications ? | 09:19 |
| davmor2 | ybon: https://bugs.launchpad.net/ubuntu/+source/telephony-service/+bug/1318724 | 09:20 |
| ubot5 | Ubuntu bug 1318724 in telephony-service (Ubuntu) "indicator-message is not always showing new message notifications" [High,Confirmed] | 09:20 |
| davmor2 | ogra_: not that I'm aware of feel free to write one :) | 09:20 |
| popey | ybon: can you take a look at https://code.launchpad.net/~yohanboniface/ubuntu-calendar-app/AgendaView/+merge/212164 - victor left a couple of comments. | 09:22 |
| ybon | gasp, alarms again :s | 09:24 |
| ogra_ | well, the alarms themselves work *if* the clock works ... | 09:25 |
| ybon | popey: => todo list for tonight, to have a look on it | 09:25 |
| ybon | (still at Mozilla today) | 09:25 |
| popey | ybon: thanks. | 09:25 |
| ybon | popey: btw, do you know what will be the stack for the ubuntu touch app store? I see it very close to the Firefox Marketplace / addons site (on which I working these days), maybe some synergy is possible | 09:26 |
| popey | ybon: sounds like a question for JamesTait / bueno | 09:27 |
| ybon | ok :) | 09:27 |
| popey | I don't actually know what's underneath the store. | 09:27 |
| * ogra_ thought we dont plan an actual "app store" in the traditional way | 09:28 | |
| ybon | ogra_: the Mozilla way of an appstore is a community managed one | 09:28 |
| ybon | i.e. developpers reviews each others | 09:28 |
| ybon | -s | 09:28 |
| ogra_ | ah | 09:28 |
| ogra_ | i thought you meant something like google play | 09:29 |
| ybon | Noooo ;) | 09:29 |
| popey | ah okay | 09:29 |
| popey | so the ubuntu store will be fully automated (in theory) | 09:29 |
| ybon | without any validation or only automated validations? | 09:29 |
| popey | developer develops in qtcreator, presses a button which uploads to the store, processes on the store do checks, and then publish | 09:29 |
| popey | yes, automated checks | 09:30 |
| ybon | ok | 09:30 |
| popey | the developer can run the same checks locally too | 09:30 |
| popey | so they can pre-validate before uploading to save time | 09:30 |
| popey | but the server will still do the checks anyway | 09:30 |
| ybon | great workflow | 09:30 |
| popey | at the moment it's semi-automatic | 09:30 |
| popey | the developer uploads, some checks are done, and then it goes into a queue for manual approval. | 09:30 |
| ybon | yeah, I think I know who is the "semi" of this story ;) | 09:31 |
| popey | hah | 09:31 |
| popey | its a few of us. | 09:31 |
| popey | looking forward to being replaced by a very small shell script | 09:31 |
| ybon | huhu :) | 09:32 |
| ogra_ | semi-popey-driven :) | 09:32 |
| popey | its fun seeing the new stuff arrive in the store. someone uploaded a port of a GameBoy Advance emulator today | 09:32 |
| ybon | popey: is there a RSS somehow? | 09:33 |
| popey | there's JSON ☻ | 09:33 |
| ybon | what a modern stack! ;) | 09:34 |
| popey | https://search.apps.ubuntu.com/api/v1/search | 09:34 |
| popey | returns the first 100 hits | 09:34 |
| ybon | popey: ordered by uploaded time desc? | 09:34 |
| popey | uh, pass | 09:35 |
| Stskeeps | popey: careful, http://www.techradar.com/news/gaming/nintendo-s-new-piracy-tsar-takes-on-nokia-654330 (from maemo.org experience) | 09:35 |
| popey | Stskeeps: I'm sure we will respect all necessary legal take-downs ☻ | 09:35 |
| Stskeeps | nod, just noting some history | 09:35 |
| popey | Interesting that Ouya store has Nintendo emulators, as does the Play store | 09:36 |
| Stskeeps | i think the problem was that nokia used it for advertisement, too | 09:36 |
| Cimi | ogra_, hi! I have issues with my mako, wifi says connected but it's not, I heard you had routing issues, did you resolved? | 09:36 |
| davmor2 | popey: as does the Ubuntu Software Center through the deb package | 09:37 |
| popey | Stskeeps: oh dear, that seems unwise. | 09:37 |
| ogra_ | Cimi, a week ago ... around image 15 or earlier even | 09:37 |
| Cimi | ogra_, did you solve? | 09:38 |
| ogra_ | there landed a fix in the image, yes | 09:38 |
| ogra_ | (around image 15 as i said) | 09:38 |
| Cimi | I have 21 but same issues | 09:39 |
| Cimi | maybe I don't have this bug | 09:39 |
| ogra_ | well, to check: adb shell route -n | 09:39 |
| ogra_ | you should see only one line starting with 0.0.0.0 | 09:39 |
| ogra_ | if you see two it is the same bug | 09:39 |
| Cimi | ogra_, completely empty | 09:39 |
| ogra_ | looks like no connection is up then | 09:40 |
| ogra_ | different bug | 09:40 |
| Cimi | ogra_, I swear the wifi is shown on | 09:40 |
| ogra_ | try switching it off/on to see if it connects then | 09:40 |
| Cimi | yeah I basically click on wifi but performs nothing apart changing the icon to connected | 09:40 |
| Cimi | route table is still empty | 09:41 |
| ogra_ | weird | 09:41 |
| Cimi | ogra_, I can connect to new access points | 09:41 |
| Cimi | ogra_, but old stop to work | 09:41 |
| Cimi | (I have 4 AP at home with different SSID | 09:41 |
| Cimi | so I can see that I can connect to new ones | 09:42 |
| ogra_ | works fine here and i havent heard about such an issue yet | 09:42 |
| ybon | I possibly have the same issue with both my Nexus4 | 09:42 |
| ogra_ | sounds very new | 09:42 |
| Cimi | I have this issue since few weeks | 09:42 |
| ogra_ | well, file a biug against network-manager ... | 09:42 |
| Cimi | btw | 09:44 |
| Cimi | /etc/NetworkManager/system-connections | 09:45 |
| Cimi | no encryption for passwords?? | 09:45 |
| zyga | hi, is policykit expected to work on ubuntu touch? | 09:46 |
| zyga | https://bugs.launchpad.net/plainbox/+bug/1300828 | 09:46 |
| ubot5 | Ubuntu bug 1300828 in PlainBox (Toolkit) "policy kit dialogs do not show on tablet image" [Undecided,Confirmed] | 09:46 |
| seb128 | Cimi, normal users can't open those files | 09:46 |
| ogra_ | zyga, yes, else we wouldnt have ofono ir network-manager working | 09:46 |
| ogra_ | *or | 09:47 |
| zyga | ogra_: including interactive dialog prompts? | 09:47 |
| Cimi | seb128, adb can though | 09:47 |
| Cimi | :) | 09:47 |
| Rienzilla | hmm | 09:47 |
| ogra_ | ... or anything else that needs a privileged backend | 09:47 |
| ogra_ | zyga, no | 09:47 |
| Rienzilla | is there any docs on how to contribute to ubuntu? | 09:47 |
| zyga | ogra_: why not and what's the replacement? | 09:47 |
| Rienzilla | I wanted to make a sogo account provider for mail/calendar/contacts sync | 09:47 |
| seb128 | Cimi, well, encryption wouldn't help, since the code to decrypt is opensource, you could just apply the decription algo | 09:47 |
| seb128 | that would only give you a false sense of security | 09:48 |
| ogra_ | zyga, why would you have/want an interactive prompt ? | 09:49 |
| popey | Rienzilla: hiya, sounds interesting, have you taken a look at some of teh existing code for that? (also, we should move to #ubuntu-app-devel maybe) | 09:49 |
| zyga | ogra_: that's what we currently do on the desktop (to confirm stuff) | 09:49 |
| ogra_ | (i doubt there is anything planned for interaction, things have to work or not ... ) | 09:49 |
| zyga | ogra_: the user needs to ack that potentially dangerous tests are to be started | 09:49 |
| zyga | ogra_: if we have no prompts at all, anyone on the system can run, .e.g. reboot test or suspend test | 09:50 |
| ogra_ | well, we dont have reboot support yet but i assume it will (like everything else) give the local user full permission | 09:50 |
| Rienzilla | popey: not at all yet | 09:50 |
| Rienzilla | I was just asking here so someone could get me up to speed a little | 09:51 |
| ogra_ | if you have the HW in your hand nothing stops you from holding the power button to reboot | 09:51 |
| zyga | ogra_: so a flashlight app can reboot the tablet? | 09:51 |
| ogra_ | lol | 09:51 |
| ogra_ | no | 09:51 |
| ogra_ | an app cant do *anything* | 09:51 |
| zyga | ogra_: right, so apps need those secure helpers | 09:51 |
| ogra_ | they can no operate outside of their own context ... you should neot even be able to run a shell | 09:51 |
| Rienzilla | popey: I joined #ubuntu-app-devel | 09:51 |
| zyga | ogra_: and secure helpers cannot be apps themselves, right? | 09:51 |
| ogra_ | (there are still some bugs that allow you to) | 09:52 |
| zyga | ogra_: so how do we do what we need to do? install standalone unsandboxed click app with everything and just have no prompts anywhere? | 09:52 |
| ogra_ | your app needs to ship all libs it needs etc .... or be a static one ... you wont be able to do anything outside of your confinement | 09:52 |
| ogra_ | zyga, i doubt this will work at all | 09:52 |
| zyga | ogra_: so how can we run tests to certify a tablet device in the future? | 09:53 |
| ogra_ | talk to the security team ... by default your app has no direct access to anything from the system | 09:53 |
| zyga | ogra_: ok, thanks | 09:53 |
| ogra_ | only if you have direct access to the device i-e- via adb | 09:53 |
| zyga | ogra_: do you know any particular contact people? | 09:53 |
| ogra_ | try jdstrand | 09:53 |
| zyga | thanks | 09:53 |
| ogra_ | i guess what you want to do would need a writable image and a deb ... or a completely unconfined click package ... which we usually only allow for very few apps ... probably yours needs to be one then | 09:54 |
| ogra_ | wrt policykit you want to ship a file that allows the local user to do everything | 09:56 |
| ogra_ | (and if you want that to work in a readonly image your file needs to be shipped in the image, nothing of the polkit paths is writable by default) | 09:57 |
| ogra_ | janimo, hmm, just seeing your example script in https://code.launchpad.net/~jani/goget-ubuntu-touch/runscript/+merge/219313 ... how would #!/bin/bash work there ... given we dont ship bash in the recovery img | 10:01 |
| ogra_ | i dont see any chrooting going on in the code | 10:01 |
| janimo | ogasawara, that is run host-side | 10:01 |
| ogra_ | oh ! | 10:01 |
| ogra_ | ok | 10:01 |
| janimo | ogra_, so override the default steps by poking at the recovery shell | 10:01 |
| ogra_ | i thought it hooked into system-image-upgrader ... sorry then | 10:02 |
| janimo | ogra_, np :) | 10:02 |
| Mirv | Saviq: ok then, thanks for looking into it. did you try on device yet? | 10:05 |
| Mirv | damn me, always on wrong channel | 10:06 |
| === Tassadar_ is now known as Tassadar | ||
| janimo | stgraber, hi, I sent a few changes to gerrit agains system-image-upgrader in pahblet-4.4 . It seems it's behind phablet-trusty wrt some fixes | 10:08 |
| Tassadar | ogra_: you're welcome. I've got a few more, I like this one especialy: 0x7f63f4d2b000 | 10:09 |
| nik90 | charles, tvoss: Are we on track to merge https://code.launchpad.net/~thomas-voss/platform-api/hw-alarms-api/+merge/210592 today as planned? | 10:23 |
| ogra_ | will that finally make alarms and calendar notifications work ? | 10:23 |
| nik90 | ogra_: yes even when the phone is in deep sleep | 10:24 |
| * ogra_ is really tired of missing meetings because no alarams go off when the screen is not on | 10:24 | |
| ogra_ | cool | 10:24 |
| ogra_ | land it !!! | 10:24 |
| tvoss | nik90, we are waiting on an iteration of the platform api to switch to gcc 4.8 | 10:24 |
| tvoss | but the mp is good to go | 10:24 |
| nik90 | tvoss: ah ok..so this will go in the silo after that? | 10:25 |
| tvoss | nik90, yup, that's the plan | 10:25 |
| nik90 | tvoss: thnx..sorry for being annoying but the MP is my golden unicorn for this cycle :) | 10:25 |
| davmor2 | ogra_: you could always use one of the million calendar apps on erm well your desktop :P | 10:26 |
| davmor2 | ogra_: or is desktop dead to you ;) | 10:26 |
| ogra_ | davmor2, well, i am not always near my desktop when the reminder goes off ... but i always have my ubuntu phone in my pocket ;) | 10:26 |
| ogra_ | evolution does fine on the desktop ... but i kind of got used to use my phone for this | 10:27 |
| nik90 | thostr_: Are you still able to reproduce this bug? I believe that the EDS fix should have resolved it. But just making sure with the original reporter | 10:27 |
| davmor2 | ogra_: haha | 10:27 |
| thostr_ | nik90: I'll give it a shot | 10:31 |
| thostr_ | nik90: once it's in the silo | 10:31 |
| nik90 | thostr_: sry forgot to attach bug report link https://bugs.launchpad.net/ubuntu-clock-app/+bug/1309057 | 10:31 |
| ubot5 | Ubuntu bug 1309057 in Ubuntu Clock App "Clock app crashes when trying to access a saved alarm" [Critical,Confirmed] | 10:31 |
| nik90 | thostr_: I was referring to that ^ | 10:31 |
| thostr_ | nik90: will test that as well :) | 10:32 |
| thostr_ | supposed to be in which image? | 10:32 |
| nik90 | thostr_: its in the image since past few days | 10:32 |
| thostr_ | nik90: ok | 10:32 |
| nik90 | thostr_: thnx | 10:33 |
| davmor2 | nik90: alarms show up for me now I'll take a look at editing once I know if this one goes off or not :) | 10:34 |
| nik90 | davmor2: okay :) | 10:35 |
| davmor2 | nik90: out of interest clock seems to guess where I am. It's not right but the country is at least. However it seems that unless I change the timezone in settings the clock is out due to bst. Should the clock app not set the timezone? | 10:36 |
| davmor2 | nik90: I thought it did at one point but maybe not | 10:36 |
| nik90 | davmor2: it was decided that the clock should follow the system timezone. So if the system timezone is incorrect, then the clock app will show the wrong time. | 10:37 |
| nik90 | davmor2: Once we have assisted gps support land, then I can add a check where if the detected location and the current time don't match, the clock app can request the user to go to the settings app to set the correct timezone | 10:38 |
| davmor2 | nik90: oh man that is going to get really confusing then, It quesses your City correctly and then displays the wrong time | 10:38 |
| nik90 | davmor2: currently the location is determined using geoIP which is sometimes not reliable which is also why I did not implement the above ^ | 10:39 |
| nik90 | davmor2: but I can understand your situation | 10:39 |
| davmor2 | nik90: ho hum nevermind then | 10:39 |
| nik90 | davmor2: but there is a bug report for this in mind, I think I can get to it during the sprint | 10:40 |
| davmor2 | nik90: nice | 10:40 |
| mandel | Elleo, morning! any news about browser + udm? | 10:41 |
| Elleo | mandel: once alex-abreu has time to do some final testing of the browser branch we should be ready to start landing everything (assuming he doesn't discover any issues) :) | 10:43 |
| mandel | Elleo, great news | 10:43 |
| mandel | alex-abreu, I dont thin we need a MIR for udm since it the qml findings are currently being seed to the phone, I think we should be able to use the html5 bindings without problems | 10:45 |
| Saviq | Mirv, no, not yet | 10:46 |
| === tsdgeos_ is now known as tsdgeos | ||
| === jhodapp|afk is now known as jhodapp | ||
| === MacSlow is now known as MacSlow|lunch | ||
| dbarth | i'm getting "QUbuntu: Could not create application instance" suddenly on #24 while trying to go into settings | 11:34 |
| cwayne | t1mp, so new-header support is in the image now? | 11:57 |
| ogra_ | no | 11:58 |
| ogra_ | in the next one | 11:58 |
| vthompson | ogra_, meaning toolbar buttons in the header? | 11:59 |
| ogra_ | dunno what this includes yet ... but the UITK change for this landed | 11:59 |
| === vying is now known as Guest26769 | ||
| === chihchun is now known as chihchun_afk | ||
| === alan_g is now known as alan_g|lunch | ||
| vthompson | ogra_, cool. cwayne: some of the new header stuff landed way back in r9: http://people.canonical.com/~ogra/touch-image-stats/9.changes | 12:01 |
| cwayne | ogra_, i mean the latest -proposed | 12:01 |
| vthompson | Music app is already using it | 12:01 |
| ogra_ | cwayne, me too :) | 12:01 |
| cwayne | that's what i was asking about though, the uitk support for new-headers | 12:01 |
| vthompson | cwayne, then it is mostly available, but it sounds like it might be finished off when the next proposed image, r25 (?) | 12:03 |
| ogra_ | right | 12:03 |
| t1mp | cwayne: yes, but more features and bug-fixes are still coming | 12:03 |
| cwayne | ok great, thanks guys! | 12:04 |
| === gnu is now known as Guest32552 | ||
| === MacSlow|lunch is now known as MacSlow | ||
| bzoltan | cjwatson: hello, do you know if there is a way to create x86 click chroot? I keep seeing this error: http://pastebin.ubuntu.com/7457207/ | 12:23 |
| ogra_ | seb128, poke ... i will need to remove teh creation of $HOME/.cache/upstart/dbus-session from the dbus session upstart job, do you know if anything uses that on desktop ? | 12:24 |
| seb128 | ogra_, why do you need to remove it? | 12:24 |
| ogra_ | seb128, because it keeps the session from starting if ~/ is 100% full | 12:24 |
| seb128 | ogra_, can't you just || true it? | 12:25 |
| ogra_ | i will move it to XDG_RUNTIME_DIR | 12:25 |
| seb128 | oh, wfm | 12:25 |
| ogra_ | seb128, no, there are other jobs reading it | 12:25 |
| seb128 | nothing use it on desktop that I know about | 12:25 |
| seb128 | yeah, moving it there makes sense | 12:25 |
| ogra_ | i just want to make sure i dont break anything non-touch by moving | 12:25 |
| ogra_ | great, thanks | 12:25 |
| jodh | ogra_, seb128: looks like hud + window-stack-bridge.conf use it | 12:26 |
| ogra_ | jodh, yep | 12:26 |
| seb128 | jodh, how did you find that? ;-) | 12:26 |
| ogra_ | http://paste.ubuntu.com/7452072/ | 12:26 |
| jodh | seb128: grep :) | 12:26 |
| ogra_ | thats my current patchset to fix bug 1316978 | 12:26 |
| ubot5 | bug 1316978 in ubuntu-touch-session (Ubuntu) "ubuntu-touch-session needs to be able to start even if the homedir is filled to 100%" [Critical,Confirmed] https://launchpad.net/bugs/1316978 | 12:26 |
| ogra_ | the issue is that there are most likely other bits like autopilot using it too | 12:27 |
| * ogra_ tries hard to find all occurences | 12:27 | |
| seb128 | http://ubuntu-codesearch.surgut.co.uk/search seems not happy | 12:27 |
| seb128 | that would be useful to look for that one | 12:28 |
| ogra_ | does that go over the whole archive ? | 12:28 |
| ogra_ | i thought only seeded stuff ... | 12:28 |
| seb128 | good question, I though it was the archive | 12:28 |
| ogra_ | (specifically the autopilot stuff and the specific tests arent seeded) | 12:29 |
| cjwatson | bzoltan: could you file a bug please? we'll need some work to create a click chroot for an architecture that doesn't actually require cross-compilation | 12:32 |
| bzoltan | cjwatson: Sure I will | 12:32 |
| cjwatson | bzoltan: please make it clear which architecture you were running this on (I *think* I'm correct in deducing that you were running it on an i386 system, not amd64?) | 12:33 |
| cjwatson | those are actually two slightly different cases really | 12:33 |
| bzoltan | cjwatson: yes | 12:33 |
| bzoltan | cjwatson: I will produce logs on both | 12:33 |
| cjwatson | no need, thanks | 12:33 |
| cjwatson | I expect they'll fail similarly right now, but they need slightly different fixes | 12:34 |
| bzoltan | cjwatson: OK | 12:34 |
| cjwatson | well, the second is a superset of the first, probably | 12:34 |
| vthompson | jdstrand, the music app has a MP to use the Thumbnailer (via Mediascanner2.0 metadata) and currently we get a dbus error because we don't have it in our apparmor profile. Running the app unconfined allows the Thumbnailer to work. Do you know what should be added to the profile for Thumbnailer to work? | 12:37 |
| jdstrand | vthompson: hi, can you paste the output of 'grep DEN /var/log/syslog'? | 12:39 |
| vthompson | jdstrand, let me rebreak the app, give me a minute | 12:39 |
| vthompson | jdstrand, http://paste.ubuntu.com/7457293/ | 12:41 |
| dobey | where does battery indicator get the charge level from? | 12:42 |
| jdstrand | vthompson: thanks | 12:43 |
| === alan_g|lunch is now known as alan_g | ||
| seb128 | dobey, upower | 12:50 |
| alex-abreu | mandel, yeah seeded on the phone but have to make sure that it works in desktop where people try to run their apps, work as in "fail gracefully if not found" | 12:50 |
| vthompson | jdstrand, are you investigating the Thumbnailer issue? Need anything else from me, bug filed, etc? | 13:01 |
| jdstrand | vthompson: I am-- it brought up a larger question in buf #1303962 with Satoris who is not in this channel | 13:04 |
| jdstrand | s/buf/bug/ | 13:05 |
| jdstrand | I'll get back to you | 13:05 |
| popey | mardy: https://blueprints.launchpad.net/ubuntu/+spec/mobile-s-password-storage is that still valid? DanChapman_ is asking about it for trojita email client | 13:06 |
| vthompson | jdstrand, ok thanks! I see you're central time as well. I won't be back on until after business hours. | 13:06 |
| ogra_ | t1mp, bzoltan, so with the new changes to input boxes (i..e. the red marker etc) it doesnt seem to be possible to do "select all" with a double-tap ... is that wanted ? | 13:08 |
| mardy | popey: in a meeting, will come back to you soon | 13:08 |
| ogra_ | (not that it matters much with totally non-functional copy paste ... but it feels a bit like a regression) | 13:09 |
| popey | mardy: k | 13:09 |
| bzoltan | zsombi: ^^^^ | 13:10 |
| ogra_ | bzoltan, heh, i generally ping the wrong people in your team :P | 13:11 |
| zsombi | ogra_: double tap always selected a work only | 13:11 |
| bzoltan | s/work/word/ | 13:11 |
| ogra_ | not in the browser ... | 13:11 |
| zsombi | yeah :) | 13:11 |
| ogra_ | i could always select the whole url | 13:12 |
| jdstrand | Satoris: hey, let's say I just put Ubuntu on my phone and copied a Katy Perry file over to ~/Music. when I launch music-app, it sees the Kety Perry song. how is that supposed to happen in the new world of media-hub, mediascanner2 and thumbnailer? (in the past, it had direct access to ~/Music, but that is going away) | 13:12 |
| jdstrand | opps, not here yet | 13:12 |
| ogra_ | heh | 13:12 |
| jdstrand | satoris: hey, let's say I just put Ubuntu on my phone and copied a Katy Perry file over to ~/Music. when I launch music-app, it sees the Kety Perry song. how is that supposed to happen in the new world of media-hub, mediascanner2 and thumbnailer? (in the past, it had direct access to ~/Music, but that is going away) | 13:12 |
| satoris | I can't speak for media-hub but for the other two it goes like this. | 13:13 |
| satoris | Mediascanner2 scans the file system. It stores its data into ~/.cache/mediascanner-2.0. | 13:13 |
| zsombi | ogra_: are you sure? | 13:13 |
| satoris | An app does a query for music. For this it needs read access to said directory or to the music scope. | 13:13 |
| satoris | Now it has a path to a music file. | 13:13 |
| ogra_ | zsombi, yeah, because i often tried to copy them in the past (just to note that there is no copy/paste yet) | 13:13 |
| satoris | In addition it also gets other meta info, such as artist and album. | 13:14 |
| zsombi | ogra_ copy/paste was there, it's just segfaults now... | 13:14 |
| ogra_ | zsombi, might actually have been single tap ... not sure about that one | 13:14 |
| satoris | Now it can ask thumbnailer to get album art. For this it needs read access to the music file. | 13:14 |
| ogra_ | oh, waiit ... that actually works | 13:14 |
| zsombi | ogra_: afair the AddressBar in the browser has nothing that would suggest single/double tap to be select-all... | 13:15 |
| ogra_ | just if i tap again to make the copy/paste dialog come up it unselects again | 13:15 |
| satoris | It can also ask for album art with the metadata it got earlier. This is a single query to the thumbnailer dbus service with artist and album name. | 13:15 |
| ogra_ | zsombi, yeah, it selects all on focus it seems ... but if i tap again to copy it unselects | 13:15 |
| satoris | The thumbnailer service then downloads the image from last.fm, stores it in cache and returns a read only fd to the file. | 13:16 |
| zsombi | ogra_: long-press on the input, that opens the popup, then chose select all | 13:16 |
| === chihchun_afk is now known as chihchun | ||
| satoris | And that's about it in a nutshell. | 13:16 |
| zsombi | ogra_: yes, that wasn't working previously | 13:16 |
| ogra_ | zsombi, well, except that there is no popup ... :) but thats the segfault i guess | 13:16 |
| satoris | jdstrand: anything left unclear? | 13:17 |
| zsombi | ogra_ nope, the segfault is when you select Copy :) | 13:17 |
| jdstrand | jhodapp: hey, how does this work with media-hub? how does the music-app find out what music is in the media collection without having read access to the files? (I'm guessing via the dbus api) | 13:17 |
| ogra_ | zsombi, hmm, then something is still wrong, i cant get a popup here | 13:17 |
| zsombi | ogra_ with the longtap you must be careful, as tap usually produces mouse moves, and that si no longer long tap :( | 13:17 |
| ogra_ | yep | 13:17 |
| zsombi | ogra_ patience | 13:18 |
| jdstrand | satoris: that is clear however it requires that an app requires special permissions to the media collection | 13:18 |
| ogra_ | i suffer from that since we have no HUD close option anymore | 13:18 |
| jhodapp | jdstrand, the music-app implements its own playlist support right now, I've not looked at how it accesses the filesystem though | 13:18 |
| zsombi | ogra_ these issues are unfortunately not identified with any test suite, unit tests or autopilots | 13:18 |
| satoris | jdstrand: music-app uses the Qml data provider, which uses libmediascanner. Meaning it needs read access to ~/.cache/mediascanner-2.0. | 13:18 |
| ogra_ | closing an app is massively hard ... since you cant easily make the close X appear in recent-apps | 13:18 |
| zsombi | ogra_ touch usually produces mousePress and mouseMove events | 13:18 |
| ogra_ | without having a super steady hand at least | 13:19 |
| jdstrand | I was able to confine the music-app without it needing access to the files in ~/Music (but this was with mediascanner1 and I did have to give access to files in the mediascanner directory and media-art) | 13:19 |
| === nerochiaro is now known as nerochiaro_lunch | ||
| jhodapp | ogra_, I agree it's hard to close an app...I think we should implement the swipe up on the app switcher view from the right edge gesture to close an app, similar to the iPhone | 13:19 |
| satoris | jdstrand: we have toyed with the idea of having a trusted dbus service for querying the Mediascanner db, but have not done it yet because just giving the ro capability to the db directory has been sufficient. | 13:19 |
| ogra_ | jhodapp, yeah, something like that | 13:19 |
| zsombi | ogra_ : so, I have a fix for that as well, it's just we cannot figure out why and what on earth is failing on jenkins... once we get that, yopu will have a more precise pressAndHold handling there... and a proper unit test!!! | 13:20 |
| ogra_ | whee ! | 13:20 |
| zsombi | ogra_ so for now, pressAndHold gently (:)) | 13:20 |
| vthompson | jdstrand, we did add "music_files_read" to the music app for confinement with mediascanner1.0 | 13:20 |
| zsombi | ogra_: then get the popover, and use select all | 13:20 |
| ogra_ | well, i try to ... i cant bring up the dialog in the browser | 13:20 |
| jdstrand | vthompson: I know. but the initial MP didn't have that, and it worked | 13:21 |
| jdstrand | that was added for compatibility with trusty images | 13:21 |
| ogra_ | if i release the thumb the whole url is selected and the kbd comes up ... if i just hold it down nothing happens | 13:21 |
| vthompson | jdstrand, right because of media-hub | 13:22 |
| jdstrand | satoris: the thrust of bug #1303962 and bug #1315381 is about apps needing no special permissions (ie, no direct access to the media collection) without the user knowing. | 13:23 |
| ubot5 | bug 1303962 in mediascanner2 (Ubuntu) "please integrate mediascanner2 and media-hub with apparmor" [High,Confirmed] https://launchpad.net/bugs/1303962 | 13:23 |
| ubot5 | bug 1315381 in mediascanner2 (Ubuntu) " please integrate mediascanner2 and media-hub with trust-store" [Undecided,New] https://launchpad.net/bugs/1315381 | 13:23 |
| jdstrand | media-hub gets us part way there, since it is the one doing the playing, not the app itself | 13:24 |
| zsombi | ogra_: it took some time 2 me too... :( | 13:24 |
| ogra_ | you actually manage to get a popup in the url bar ? | 13:24 |
| zsombi | ogra_ yup | 13:24 |
| ogra_ | weird | 13:25 |
| zsombi | ogra_ you need to be verrrrry gentle... | 13:25 |
| ogra_ | is the kbd up when you do that ? | 13:25 |
| ogra_ | or before ? | 13:25 |
| zsombi | all the time | 13:25 |
| satoris | jdstrand: in order to access the list of media files in Mediascanner2, you need read access to ~/.cache/mediascanner-2.0. If you don't have that right, you can't access the list. | 13:25 |
| ogra_ | no go here :( | 13:26 |
| jdstrand | satoris: so, based on my understanding, mediascanner2 does all the scanning, and then an app needs access to ~/.cache/mediascanner2 to know what was scanned | 13:26 |
| jdstrand | satoris: then the app can create a playlist, request a thumbnail, etc | 13:26 |
| satoris | jdstrand: yes, exactly. Which reminds we that we don't have an apparmor profile for the scanner daemon yet... | 13:27 |
| zsombi | ogra_ yeah, it's really bad... | 13:27 |
| jdstrand | satoris: access to ~/.cache/mediascanner2 is then the issue. that contains all the stuff for data mining, etc that I brought up before | 13:27 |
| ogra_ | zsombi, are you on a newer version perhaps ? i'm on #22 | 13:27 |
| jdstrand | (which I think gets everyone on the same page for the bugs I mentioned) | 13:28 |
| * ogra_ doesnt know how he could be even more gentle to make the popup come up | 13:28 | |
| jdstrand | satoris: so, we can continue to special case the music-app so it isn' | 13:28 |
| jdstrand | t broken | 13:28 |
| ogra_ | should i say kind words all the time perhaps ? | 13:28 |
| ogra_ | :) | 13:28 |
| zsombi | ogra_: :D | 13:28 |
| ogra_ | oh ! | 13:28 |
| ogra_ | the 100s time it just worked now | 13:29 |
| jdstrand | satoris: but you mentioned that you toyed with the idea of mediascanner2 having a dbus api. that is required for bug 1303962 and bug 1315381 | 13:29 |
| ubot5 | bug 1303962 in mediascanner2 (Ubuntu) "please integrate mediascanner2 and media-hub with apparmor" [High,Confirmed] https://launchpad.net/bugs/1303962 | 13:29 |
| ubot5 | bug 1315381 in mediascanner2 (Ubuntu) " please integrate mediascanner2 and media-hub with trust-store" [Undecided,New] https://launchpad.net/bugs/1315381 | 13:29 |
| ogra_ | but now i cant make the paste disalog come up inside the G+ app :P | 13:30 |
| jdstrand | satoris: because the dbus service can have the apparmor integration and prompt the user "The facebook app wants to access your media collection. Proceed?" | 13:30 |
| satoris | jdstrand: I was just about to ask that. :) | 13:30 |
| jdstrand | satoris: and that opens the door for any app in the app store to be able to be unprivileged and access the media collection] | 13:31 |
| satoris | If that is needed then we can create one. | 13:31 |
| jdstrand | it is needed | 13:31 |
| satoris | I'll need to do some talk with managerial types about time tables and such. | 13:32 |
| jdstrand | we can in the short term special case music-app like I said though | 13:32 |
| jdstrand | the problem is that other apps won't have that privilege and so the platform doesn't support alternative media players, etc via the app store | 13:32 |
| mterry | ogra_, has much changed in utopic touch around user logins, XDG_RUNTIME_DIR, or such? I'm seeing a new oddity when testing split mode | 13:39 |
| ogra_ | mterry, not yet ... there is a rather big landing i'm preparing to move the dbus-session file to /run | 13:40 |
| ogra_ | mterry, when did that start ? | 13:40 |
| ogra_ | rsalveti dropped one lightdm start condition yesterday that was breaking the emulator | 13:41 |
| mterry | ogra_, not sure. I only started noticing it this week, so maybe it started sometime last week or the weekend | 13:41 |
| ogra_ | ah, no | 13:41 |
| ogra_ | then it isnt related to that change | 13:41 |
| ogra_ | mterry, http://people.canonical.com/~ogra/touch-image-stats/ has all the changelgs for utopic | 13:41 |
| ogra_ | you can probably go backwards through them to find any related changes if you roughly know the date when it started | 13:42 |
| === Ariadeno is now known as Guest16234 | ||
| mpt | Laney, I think the root cause of our confusion about system updates yesterday was bug 1319037 | 13:43 |
| ubot5 | bug 1319037 in ubuntu-system-settings (Ubuntu) "Update described as "Installing" when it's automatically downloading" [Undecided,New] https://launchpad.net/bugs/1319037 | 13:43 |
| ogra_ | wasnt he referring to the popup (that overlays the whole screen after the download is done) ? | 13:44 |
| kenvandine | i think it says installing while it's downloading | 13:48 |
| kenvandine | or maybe it's just automatically downloaded | 13:49 |
| kenvandine | and says installing while it's preparing to install... before the popup | 13:49 |
| ogra_ | it surely also says installing while downloading, i was referring to the specific conversation from yesterday though | 13:51 |
| * kenvandine wasn't involved in that, but i'd say that status is wrong | 13:51 | |
| ogra_ | you simply cant prevent the popup where it always asks you to install/cancel | 13:51 |
| kenvandine | so if it's downloading, it should say that instead of installing | 13:52 |
| === chihchun is now known as chihchun_afk | ||
| ogra_ | yeah, but if you select download only on wlan it should also not popup that question | 13:52 |
| ogra_ | but only download :) | 13:53 |
| kenvandine | ogra_, true | 13:53 |
| kenvandine | so two bugs :) | 13:53 |
| ogra_ | yeah | 13:53 |
| alex-abreu | mandel, so no plan to mir udm ? ... some bit in the main archive (desktop) pull it as a dep, and since it is not in main it wont really fly, so I have to make it a recommends or soemthing | 13:59 |
| ogra_ | alex-abreu, there was a block on MIRing touch stuff ... talk to asac if thats still the case | 14:00 |
| alex-abreu | ogra_, ok, | 14:01 |
| ogra_ | (i think it was lifted though ... ) | 14:01 |
| jdstrand | satoris: fyi, updated bug #1303962 (comment #7) for our conversation | 14:06 |
| ubot5 | bug 1303962 in mediascanner2 (Ubuntu) "please integrate mediascanner2 and media-hub with apparmor" [High,Confirmed] https://launchpad.net/bugs/1303962 | 14:06 |
| jdstrand | I think the thumbnailer task could probably be closed | 14:07 |
| jdstrand | satoris: fyi, I created bug #1319065, assigned it to me and gave myself a work item | 14:15 |
| ubot5 | bug 1319065 in mediascanner2 (Ubuntu) "please provide apparmor profile for media scanner service" [Undecided,New] https://launchpad.net/bugs/1319065 | 14:15 |
| jdstrand | ok, *now* I assigned it to me :) | 14:16 |
| === dholbach_ is now known as dholbach | ||
| === alan_g is now known as alan_g|tea | ||
| satoris | jdstrand: added requirements as a comment. | 14:28 |
| jdstrand | thanks! | 14:28 |
| mandel | alex-abreu, asac any idea of a possible mir on udf? | 14:29 |
| mandel | alex-abreu, I was just considering the html sdk use case in the phone at this point in time when I made the "no MIT clame" | 14:29 |
| mandel | s/MIT/MIR | 14:29 |
| alex-abreu | mandel, sure, I got that, I'll adjust the html api to be more flexible, posisbly hoping for a mir, but it shouldn't be much of an issue | 14:30 |
| === alan_g|tea is now known as alan_g | ||
| mandel | alex-abreu, ok, in any case when we move everything to the desktop we will have to do this | 14:31 |
| mandel | alex-abreu, for example, system image updates might be moved there and therefore udm is needed | 14:31 |
| alex-abreu | mandel, yeah, I think we should be fine for now, ... | 14:33 |
| dpm | hi dobey, quick question: where do the "See more | See less" strings I can see on the click scope when I'm looking at an app preview come from? I could not find them in the .pot file | 14:37 |
| dobey | dpm: not from the click scope. i think the dash adds them automatically when the description text is more than some number of characters or lines | 14:39 |
| dpm | ah, so that might come from unity8? | 14:39 |
| dobey | yes, afaik | 14:39 |
| === dandrader is now known as dandrader|afk | ||
| dpm | ah, yes, they're in the unity8 .pot, thanks dobey! | 14:43 |
| dobey | dpm: there are a couple of string changes in the click scope though, that you'll want to poke the zh_CN translator to get filled in asap, so we can land them in the image with translations for MAE | 14:44 |
| dpm | dobey, yes, I saw them, thanks for the heads up. Will ask translators to complete them | 14:46 |
| dobey | is anyone else seeing weird behavior with wifi on nexus4? | 14:50 |
| ogra_ | dobey, someone mentioned issues above | 14:50 |
| ogra_ | (all fine here ) | 14:51 |
| seb128 | dobey, Cimi was mentioning earlier than he has his icon set as being online but that there were no active connections | 14:51 |
| dobey | mine is literally 1 meter from the access point, and the icon shows very poor signal, and network is very poor. iwconfig says 9 Mb/s, but the access point is G/N | 14:52 |
| ogra_ | yeah, looks different to what Cimi saw | 14:53 |
| cimi_ | a router reboot solved my issue | 14:53 |
| cimi_ | for now | 14:53 |
| cimi_ | although all other devices were working fine | 14:53 |
| cimi_ | so it could have been the router or ubuntu waiting something | 14:54 |
| dobey | i thought maybe it was trying to switch to my other access point that's on the other side of my house, but the hex id isn't changing (i'm running iwconfig under watch) | 14:54 |
| dobey | yeah i have a nexus 5 sitting right next to the nexus 4, and it's working fine, no problems at all. has been since i got it | 14:54 |
| dobey | nexus 4 has been horrible with wifi since i got it though | 14:54 |
| ogra_ | dobey, in any case let cyphermox know ... | 14:55 |
| ogra_ | eh will surely want to know about wlan issues | 14:55 |
| ogra_ | *he | 14:55 |
| dobey | cyphermox: ^^ any idea how to debug this? | 14:55 |
| cyphermox | that's pretty much it; check what "iw dev wlan0 link", and any messages from wpa-suppliant/kernel in syslog | 14:58 |
| sergiusens | mandel: https://code.launchpad.net/~sergiusens/udm/packaging/+merge/219387 | 14:59 |
| mandel | sergiusens, did you review the other mrs in that? | 15:00 |
| sergiusens | robru: can you do a package review for https://code.launchpad.net/~sergiusens/udm/packaging/+merge/219387 please? | 15:00 |
| sergiusens | mandel: no, no email notif :- | 15:00 |
| sergiusens | :-P | 15:00 |
| mandel | sergiusens, no worries, I have the same problem | 15:06 |
| === dandrader|afk is now known as dandrader | ||
| === DanChapman_ is now known as DanChapman | ||
| mterry | ogra_, I think my boot oddities are due to the runlevel change in lightdm's override file in ubuntu-touch-session. Doing some more reboots to confirm | 15:18 |
| dobey | cyphermox: http://pastebin.ubuntu.com/7457986/ | 15:20 |
| cyphermox | dobey: I can't do anything about that | 15:21 |
| cyphermox | could be that there is lots of interference in your area | 15:21 |
| dobey | cyphermox: are those messages what's causing my problem? | 15:22 |
| cyphermox | not necessarily what's causing, more like maybe a symptom | 15:23 |
| ogra_ | mterry, that change is only in the last two images though | 15:23 |
| dobey | well, the messages not being the cause obviously | 15:23 |
| cyphermox | it's just the kernel background scanning mentions it didn't find the AP it was looking for | 15:23 |
| mterry | ogra_, I only noticed the change yesterday, that fits | 15:23 |
| ogra_ | mterry, if you saw the issues before it cant be that ... also the "runlevel PRELEVEL=S" line still makes sure it only starts at runlevel change | 15:23 |
| ogra_ | so efferctively nothing changed for you | 15:24 |
| ogra_ | unless you kind of rely on a certain runlevel number (which i doubt anyone does nowadays) | 15:24 |
| mterry | ogra_, I only noticed this problem after I silo 002 rebuilt ubuntu-touch-session to update with those changes. So I'm sure it *could* be those changes | 15:24 |
| mterry | ogra_, the boot problem seems to be related to when /run/user stuff gets set up? | 15:25 |
| ogra_ | it only removed the check for the runlevel number | 15:25 |
| ogra_ | it doesnt change the timing or anything | 15:25 |
| mterry | ogra_, I'm just telling you what I see. Adding that line back in makes the problem go away. I'm still investigating why | 15:26 |
| ogra_ | strange | 15:26 |
| mterry | ogra_, the PREVLEVEL=S is an or condition, not an and one | 15:26 |
| mterry | ogra_, so it could change timing I think | 15:27 |
| ogra_ | i could understand that if the "runlevel PRELEVEL=S" was dropped too | 15:27 |
| ogra_ | mterry, but it didnt, i have bootcharts :) | 15:27 |
| mterry | ogra_, with split greeter, maybe timing changes | 15:27 |
| ogra_ | http://people.canonical.com/~ogra/touch-bootcharts/ | 15:27 |
| mterry | ogra_, I'm not noticing a problem with normal images | 15:27 |
| ogra_ | (22 vs 24 ... lightdm starts in both at ~15sec) | 15:28 |
| mterry | ogra_, does the normal user session rely on /run/user stuff at all? Would we notice if that wasn't being set up for the unity8 session? | 15:28 |
| Rienzilla | ik ben bijna klaar... | 15:28 |
| ogra_ | mterry, some bits do ... i'm currently working on making it fully rely on that btw | 15:29 |
| ogra_ | ripping out all bits that write to ~/ on session startup | 15:29 |
| Rienzilla | neus | 15:29 |
| Rienzilla | wwe | 15:29 |
| mterry | ogra_, b/c there are two possibilities -- /run/user is racy on normal images now and we don't notice, or only racy with split greeter. But it's definitely racy with split greeter at least | 15:30 |
| mterry | Split greeter relies on it | 15:30 |
| ogra_ | hmm, not sure what actually sets it up, i thought it only gets created on login/autologin | 15:30 |
| Rienzilla | sorry about that :-) | 15:31 |
| mterry | ogra_, I *think* logind manages it... when does that start? | 15:31 |
| popey | Rienzilla: having problems? ☻ | 15:31 |
| ogra_ | way way before lightdm | 15:31 |
| ogra_ | look at the bootchart ... somewhere along with the system dbus | 15:32 |
| Rienzilla | i typed in the wrong terminal | 15:32 |
| Rienzilla | regarding the account plugin im getting some assistance from mardy | 15:32 |
| === gatox is now known as gatox_lunch | ||
| ogra_ | mterry, in any case Mir uses the dir for the private socket so at that time it definitely exists | 15:35 |
| ogra_ | MIR_SERVER_FILE=$XDG_RUNTIME_DIR/mir_socket | 15:35 |
| mterry | ogra_, my problems show up pre-mir-launch but sure | 15:35 |
| ogra_ | well | 15:36 |
| ogra_ | we didnt use that dir before unity8 started in the past | 15:36 |
| mterry | ogra_, specifically, my problem is that upstart's init for the lightdm user never creates its /run/user/XXX/upstart directory (I'm guessing because it wasn't available at the time) | 15:36 |
| mterry | ogra_, well phablet's upstart must've | 15:36 |
| seb128 | mterry, for normal session I think it's pam creating the directory | 15:36 |
| mterry | seb128, right.... But via systemd's pam plugin I think? | 15:37 |
| seb128 | mterry, correct | 15:37 |
| ogra_ | root@ubuntu-phablet:~# grep XDG_RUNTIME_DIR /usr/share/upstart/sessions/* | 15:37 |
| ogra_ | /usr/share/upstart/sessions/unity8.conf: MIR_SERVER_FILE=$XDG_RUNTIME_DIR/mir_socket | 15:37 |
| ogra_ | root@ubuntu-phablet:~# | 15:37 |
| ogra_ | nothing in the session handling uses or creates it | 15:38 |
| mterry | ogra_, but I think upstart itself uses it (puts session info in XDG_RUNTIME_DIR/upstart) | 15:39 |
| andrewbiolo_ | Hi, I am Andrea and I am Italian. I am studying bioinformatics and I love Ubuntu. I would like to learn QML and to develop a new Application. I think is important to contribute at the core apps, but I don't know how to start. Someone can give me any advice? :) | 15:39 |
| ogra_ | mterry, oh, right /etc/profile.d/upstart-phablet.sh | 15:40 |
| ogra_ | it actually exports it there | 15:40 |
| ogra_ | but thats only for shell access | 15:40 |
| ogra_ | i dont think profile.d is parsed for non bash/shell | 15:41 |
| ogra_ | (at least it shouldnt be) | 15:42 |
| === nerochiaro_lunch is now known as nerochiaro | ||
| ogra_ | andrewbiolo_, try #ubuntu-app-devel | 15:48 |
| andrewbiolo_ | thanks :) | 15:48 |
| === andrewbiolo_ is now known as andrewbiolo | ||
| === alecu_ is now known as alecu | ||
| robru | sergiusens, hey, just woke up. yeah I can take a look ;-) | 16:15 |
| === gatox_lunch is now known as gatox | ||
| === alan_g is now known as alan_g|EOD | ||
| === dpm is now known as dpm-afk | ||
| mterry | ogra_, hmm. "May 13 16:34:24 ubuntu-phablet lightdm: pam_systemd(lightdm-greeter:session): Failed to create session: Access denied" -- and I have also seen "Message did not receive a reply (timeout by message bus)" as the reason. Digging into it | 16:38 |
| ogra_ | mterry, wow, gross | 16:39 |
| ogra_ | which bus does it try to talk to ? system or session ? | 16:39 |
| rsalveti | weird | 16:39 |
| mterry | ogra_, probably system. I think it's talking to the logind daemon | 16:40 |
| ogra_ | wow | 16:40 |
| tvoss | nik90, ping | 16:42 |
| mterry | Now where does logind log to? | 16:43 |
| ogra_ | probably some binary systemd journal :P | 16:43 |
| ogra_ | i guess pitti would knwo but seems he is gone for the day | 16:43 |
| mterry | agreed, I've done this before and i remember not liking it | 16:43 |
| nik90 | tvoss: pong | 16:47 |
| tvoss | nik90, hey there, is the clock app using the location service to determine the location of the user? | 16:48 |
| nik90 | tvoss: no, it is using geoIP...GPS was too slow at that point when I tried. | 16:48 |
| nik90 | tvoss: I was waiting for Assisted GPS to land before using the location service | 16:48 |
| tvoss | nik90, ack and thx | 16:48 |
| nik90 | np | 16:49 |
| mterry | testing "and started cgmanager" | 17:04 |
| ogra_ | uh | 17:04 |
| ogra_ | where is that from ? | 17:04 |
| ogra_ | mterry, that should always be "cgmanager-ready" | 17:05 |
| ogra_ | unless it is cgproxy itself | 17:05 |
| mterry | ogra_, what does cgproxy do? | 17:05 |
| ogra_ | stuff for lxc | 17:05 |
| === dandrader is now known as dandrader|lunch | ||
| ogra_ | it sets up the cgroup management iirc | 17:06 |
| mterry | ogra_, and what's the story with cgmanager-ready? | 17:06 |
| ogra_ | thats emitted if it is actually ready for consumption | 17:06 |
| mterry | hrm | 17:06 |
| ogra_ | if you want to wait for cgmanager, use that event | 17:06 |
| ogra_ | started cgmanager is to early | 17:07 |
| mterry | k | 17:07 |
| mterry | it looks like logind needs cgmanager is all | 17:07 |
| ogra_ | well, then the system dbus should make sure to not start before cgmanager-ready | 17:11 |
| ogra_ | given that logind is started by a dbus service file | 17:11 |
| ogra_ | /usr/share/dbus-1/system-services/org.freedesktop.login1.service | 17:12 |
| mterry | ogra_, yeah getting "cgmanager: Error pinging manager: Did not receive a reply" in logind. Changing startup conditions for system dbus seems like a heavy hammer | 17:14 |
| ogra_ | yeah, logind seems to also start a lot after dbus ... looking at http://people.canonical.com/~ogra/touch-bootcharts/ubuntu-phablet-utopic-24.png nearly 4 seconds after dbus | 17:17 |
| mterry | ogra_, I used cgmanager-ready but still got this issue | 17:18 |
| ogra_ | (lxc-start actually depends on cgmanager-ready ... as you can see that starts about 1sec before dbus) | 17:18 |
| ogra_ | right, i think cgmanager is a red herring ... | 17:19 |
| ogra_ | since that starts way before everything | 17:19 |
| mterry | ah ok. I should have looked at the bootchart | 17:19 |
| === gnu is now known as Guest1051 | ||
| mterry | But the cgmanager error certainly seems connected to the failure to create a session | 17:22 |
| ogra_ | that must be a relatively new dep then | 17:23 |
| ogra_ | cgmanager is only around since a month or so ... on the images | 17:23 |
| ogra_ | we used cgroups-lite before | 17:23 |
| ogra_ | (or perhaps 6 weeks) | 17:24 |
| elopio | ogra_: yes, you are right, apparmor denied http://paste.ubuntu.com/7458617/ | 17:32 |
| ogra_ | that must be new then | 17:33 |
| ogra_ | jdstrand, ^^^ | 17:33 |
| ogra_ | issues with the camera-app | 17:33 |
| jdstrand | oh, the camera-app is confined now? | 17:33 |
| ogra_ | it shoulldnt i think | 17:34 |
| ogra_ | but can apps become confined by accident ? | 17:34 |
| jdstrand | no | 17:35 |
| jdstrand | not unless it said, "hey, I want to be confined" | 17:35 |
| jdstrand | fyi, the camera app could be special and simply use the 'picture_files' policy group | 17:37 |
| jdstrand | note, that the denial is for creating the ~/Pictures directory | 17:37 |
| jdstrand | the picture_files policy group doesn't allow creating the directory, only reading it (as well as reading/writing files under ~/Pictures) | 17:38 |
| ogra_ | still weird though | 17:39 |
| jdstrand | so maybe it has been confined for a while but the ~/Pictures directory always existed but now it doesn't | 17:39 |
| ogra_ | rsalveti, you updated the camera-app twice the last week ... do you know if the click package was actually updated too ? | 17:39 |
| ogra_ | ah, not twice actually | 17:40 |
| elopio | ogra_, jdstrand: I created the Pictures directory and now it works. | 17:49 |
| ogra_ | well, it should exist ... thats weird | 17:49 |
| ogra_ | does ~/.config/user-dirs.dirs exist on your system ? | 17:50 |
| ogra_ | (and contain Pictures) | 17:50 |
| elopio | ogra_: yes. XDG_PICTURES_DIR="/home/phablet/Pictures" | 17:51 |
| ogra_ | hmm, weird | 17:52 |
| elopio | my /home/phablet is empty. | 17:52 |
| ogra_ | sergiusens, does --wipe actually remove the dirs ? | 17:52 |
| sergiusens | ogra_: --wipe just does a 'format data' in ubuntu_commands | 17:54 |
| ogra_ | i dont get how elopio could end up with an empty home then | 17:55 |
| sergiusens | ogra_: elopio I had that after running autopilot sometimes | 17:55 |
| ogra_ | oh | 17:55 |
| elopio | I will flash one more time. | 17:56 |
| sergiusens | ogra_: camera and gallery haven't been click uploaded | 17:56 |
| ogra_ | sergiusens, right, well, the missign dir is the issue anyway ... but i guess we need to get the apps back in sync with the debs at some point | 17:56 |
| rsalveti | ogra_: no, that was just packaging changes for the deb version | 17:57 |
| ogra_ | right | 17:57 |
| rsalveti | sergiusens: talking about camera and gallery, what do we need to do to upload the x86 version to the store? | 17:57 |
| sergiusens | rsalveti: do a refactor with fginther, basically what we talked about last Friday | 17:58 |
| sergiusens | rsalveti: I planned for that next week | 17:58 |
| rsalveti | sergiusens: ok, will let that to you then :-) | 17:59 |
| === dandrader|lunch is now known as dandrader | ||
| seb128 | Cimi, did you see Saviq's needs fixing on https://code.launchpad.net/~unity-team/ubuntu-system-settings/wizard.wifi/+merge/212675 ? | 18:00 |
| elopio | ogra_, sergiusens: freshly installed, with ubuntu-device-flash --channel devel-proposed --wipe | 18:00 |
| elopio | /home/phablet is empty. | 18:01 |
| elopio | well, it has the hidden dirs. | 18:01 |
| mterry | ogra_, ah... I think this is what happens -- for whatever reason, cgmanager is available on dbus, but not responding to pings. This due to a bug in logind is crashing logind. Which means that the pam module errors out and can't set up the lightdm sesion well. And something with that runlevel change is triggering the problem with cgmanager | 18:03 |
| mterry | Well when I say "on dbus" I guess I mean at the private dbus location of "/sys/fs/cgroup/cgmanager/sock" | 18:04 |
| ogra_ | mterry, hmm, that might explain why rsalveti has issues in the emulator but we dont get them in all other installs | 18:04 |
| mterry | ogra_, it doesn't seem to affect the autologin session, not sure why yet | 18:05 |
| ogra_ | perhaps some kernel support that cgmanager uses is missing in that kernel | 18:05 |
| rsalveti | I still don't get why upstart would not trigger lightdm | 18:05 |
| rsalveti | bug 1318681 btw | 18:06 |
| ubot5 | bug 1318681 in upstart (Ubuntu) "lightdm not started when running the ubuntu-touch x86 emulator" [Undecided,New] https://launchpad.net/bugs/1318681 | 18:06 |
| mterry | ogra_, is there anything other signal besides cgmanager-ready I could listen to? That didn't seem to fix the problem | 18:06 |
| ogra_ | rsalveti, well, read above ... seems logind needs a fully working cgmanager now | 18:06 |
| seb128 | mterry, hey, I just commented on the wizard mp that starts using mir apis ... can you explain/comment there to explain why we need that? It would be nice to be able to keep developing/running u-s-s on desktop environments | 18:07 |
| ogra_ | and if logind cant start the session cant | 18:07 |
| rsalveti | but how is the event related with upstart and lightdm? | 18:07 |
| mterry | ogra_, well, it can take no cgmanager fine. But a half-working one it can't (due to a bug that should be fixed regardless) | 18:07 |
| rsalveti | I mean, wouldn't that just be an issue after lightm starts? | 18:07 |
| rsalveti | *lightdm | 18:07 |
| mterry | seb128, maliit | 18:07 |
| seb128 | mterry, I see you are in the middle of another discussion, I'm going to ping you again later/another day, no hurry | 18:08 |
| ogra_ | rsalveti, our autologin is kind of "in" lightdm atm | 18:08 |
| ogra_ | we dont use a greeter so the login isnt separated | 18:08 |
| rsalveti | ogra_: right, but then we would at least be able to see lightdm starting in upstart, right? | 18:09 |
| ogra_ | it would attempt to start ... fall over and fail | 18:09 |
| rsalveti | right, we don't even see it starting | 18:09 |
| ogra_ | hmm, ok | 18:10 |
| mterry | rsalveti, reading that bug, the runlevel events are emitted just fine, but something weird about that combination of start-on triggers doesn't work? | 18:10 |
| rsalveti | mterry: exactly | 18:11 |
| ogra_ | well, android is emitted by the android container ... dbus should run since a long time already and filesystem was emitted by mountall | 18:11 |
| rsalveti | everything was emitted | 18:11 |
| ogra_ | right | 18:12 |
| mterry | ogra_, is cgmanager necessary for sessions normally? You say it's a new thing? | 18:14 |
| * mterry is wondering if fixing logind crash is enough | 18:14 | |
| ogra_ | it is used for lxc | 18:14 |
| ogra_ | which runs our android container | 18:15 |
| mterry | huh, so things will still not work, just in a new and different way | 18:15 |
| robru | mterry, hey. we just had a unity8 landing, is it going to screw you up if I rebuild unity8 in silo 2? | 18:15 |
| mterry | robru, well chances are there are conflicts. Let me confirm | 18:15 |
| robru | mterry, ok, no rush on my end, but it will need to be rebuilt before it can be published | 18:16 |
| mterry | robru, right | 18:16 |
| mterry | robru, yeah conflicts | 18:16 |
| ogra_ | mterry, lxc switched from cgroups-lite to cgmanager ... ask hallyn about details for that bit | 18:16 |
| ogra_ | i wasnt aware that logind even remotely uses cgmanager | 18:16 |
| robru | mzanetti, ping? I got you silo 5 for your unity8 landing, and building: https://ci-train.ubuntu.com/job/landing-005-1-build/55/console | 18:19 |
| ogra_ | lol ! | 18:27 |
| * ogra_ just saw popey's comment on G+ | 18:28 | |
| ahayzen | mterry, ping | 18:28 |
| mterry | ogra_, cgmanager support is an ubuntu add on apparently | 18:28 |
| mterry | ahayzen, hello! | 18:28 |
| ahayzen | mterry, you said you were having issues with mp3's in the music-app before correct? | 18:28 |
| ogra_ | mterry, well, it is handled by the lxc guys ... | 18:29 |
| mterry | ahayzen, uh yeah | 18:29 |
| ahayzen | mterry, does that still occur? | 18:29 |
| mterry | ahayzen, I'm not in a position to test this second | 18:29 |
| mterry | ahayzen, can I get back to you on that? | 18:30 |
| ahayzen | mterry, ok no worries, i tried a few mp3s on mine and couldn't get the same issue. When u have a moment and if u can confirm it would you be able to send the file to me? | 18:30 |
| mterry | ahayzen, k | 18:30 |
| ahayzen | mterry, thanks :) | 18:31 |
| doanac | sergiusens: just hit a problem with the x86 emulator and phablet-click-test-setup. seems like get_package_version may not be working for ubuntuuitoolkit | 18:33 |
| doanac | just started digging, but was curious if you'd seen this? | 18:34 |
| mterry | robru, if you want to rekick unity8, that's fine | 18:41 |
| mterry | I merged from trunk | 18:41 |
| doanac | sergiusens: i think the issue is that ubuntu-ui-toolkit isn't installed in the x86 image. | 18:42 |
| ogra_ | doanac, should be -gles ... | 18:47 |
| ogra_ | the x86 emu image has a bunch of packages that uuse -gles as suffix in the package name | 18:47 |
| ogra_ | UITK is one of them | 18:47 |
| doanac | ogra_: oh - thanks. i can patch that locally to unblock myself | 18:47 |
| Saviq | seb128, yeah he saw it, and it shouldn't be a problem to develop on X11 still | 18:48 |
| seb128 | Saviq, you are replying for others? ;-) | 18:48 |
| Saviq | seb128, the wizard needs to run as a Mir server when applicable, to be able to composite the OSK on itself | 18:48 |
| Saviq | seb128, I did get highlighted, too ;P | 18:49 |
| seb128 | oh, right | 18:49 |
| seb128 | thanks for the reply | 18:49 |
| Saviq | seb128, but if not under Mir, it will just use the xcb QPA and not try to be Mir | 18:49 |
| seb128 | great | 18:50 |
| sergiusens | doanac: yeah, that's probably it | 18:56 |
| robru | mterry, thanks | 19:10 |
| bzoltan | cjwatson: that is the bug https://bugs.launchpad.net/ubuntu/+source/click/+bug/1319153 | 19:21 |
| ubot5 | Ubuntu bug 1319153 in click (Ubuntu) "Failes to create x86 chroot" [Undecided,New] | 19:21 |
| rsalveti | doanac: yeah, we got a few packages duplicated for the x86 to work | 19:22 |
| rsalveti | ubuntu-ui-toolkit is one of them | 19:22 |
| rsalveti | just check with dpkg -l | grep gles | 19:22 |
| doanac | rsalveti: thanks. | 19:23 |
| rsalveti | so I'd guess we need to change that check to make it work with the x86 emulator, as it uses a different src package for it | 19:23 |
| sergiusens | rsalveti: doanac there's a couple more assumptions I need to deal with wrt x86 | 19:33 |
| doanac | sergiusens: ack. i think there might be some other stuff around our new unlock-script from mterry. i'm taking a step back from my daily image testing scripts and just playing with things by hand to get a better feel | 19:34 |
| Saviq | rsalveti, I just ran the x86 emulator on my machine and got scopes just fine, are you still seeing them miss the deadline on yours? | 19:58 |
| doanac | rsalveti: just found something subtle but important. i was running "ubuntu-emulator run" as my normal user. i noticed a warning message about it not having access to the kvm device. I re-ran things with sudo and its MUCH faster now | 20:04 |
| doanac | mterry: i think we need a change to your unlock_device script. "adb reboot" doesn't work against the emulator. we'll need to kill the emulator pid and re-start it. | 20:07 |
| doanac | so i think we need an option "--reboot-cmd" flag for this script | 20:07 |
| mterry | doanac, curious | 20:07 |
| rsalveti | Saviq: that only happens when you get bug 1318070 | 20:08 |
| ubot5 | bug 1318070 in linux-goldfish (Ubuntu) "TSC fails to calibrate against PIT" [Undecided,Confirmed] https://launchpad.net/bugs/1318070 | 20:08 |
| doanac | mterry: to be honest, just providing a "reboot-and-wait" hook might be easier. then we can do this with one thing | 20:08 |
| mterry | doanac, it would be nice if we could make the emulator fake being a device on adb | 20:08 |
| mterry | doanac, explain the reboot-and-wait bit? | 20:08 |
| Saviq | rsalveti, on both x86 and armhf or? | 20:08 |
| mterry | oh right, because we have the wait command? | 20:08 |
| rsalveti | Saviq: because then libpgm needs to run a small benchmark to calculate the cpu mhz | 20:08 |
| rsalveti | https://code.google.com/p/openpgm/issues/detail?id=35 | 20:08 |
| doanac | mterry: correct. | 20:08 |
| rsalveti | Saviq: that is only with x86, armhf happens all the time because it's way slower | 20:09 |
| Saviq | rsalveti, right, ok | 20:09 |
| rsalveti | so I believe if you fix the armhf issue, it'd probably fix it for x86 as well | 20:09 |
| rsalveti | doanac: hm, interesting, I'm able to run with KVM as normal user here | 20:09 |
| doanac | mterry: for example. before your unlock_device we had a more sophisticated approach to reboot-wait: http://bazaar.launchpad.net/~ubuntu-test-case-dev/ubuntu-test-cases/touch/view/head:/scripts/reboot-and-wait | 20:09 |
| rsalveti | maybe your user is not part of a special group or something | 20:10 |
| rsalveti | and yeah, the real speed is when you have kvm enabled :-) | 20:10 |
| doanac | rsalveti: i'm sure that's the problem | 20:10 |
| rsalveti | Saviq: when are you going to land the code that is currently in devel? | 20:10 |
| rsalveti | Saviq: that might already behave differently | 20:11 |
| Saviq | rsalveti, I have no devel branches ;D | 20:11 |
| rsalveti | thought the scopes also had a devel branch | 20:11 |
| Saviq | rsalveti, if you mean -scopes-shell that's mhr3 | 20:11 |
| doanac | rsalveti: is it feasible to get the emulator to support "adb reboot" ? | 20:11 |
| rsalveti | Saviq: right | 20:11 |
| rsalveti | doanac: might be, need to check | 20:12 |
| rsalveti | doanac: mind opening a bug for now? against the android package | 20:12 |
| rsalveti | will take a look later | 20:12 |
| doanac | rsalveti: sure. | 20:12 |
| doanac | mterry: i'll add you to the bug as well | 20:12 |
| === dandrader is now known as dandrader|afk | ||
| === beidl_ is now known as beidl | ||
| doanac | mterry, rsalveti: https://bugs.launchpad.net/ubuntu/+source/android/+bug/1319172 | 20:28 |
| ubot5 | Ubuntu bug 1319172 in android (Ubuntu) ""adb reboot" doesn't work with emulator" [Undecided,New] | 20:28 |
| rsalveti | doanac: thanks | 20:28 |
| === dandrader|afk is now known as dandrader | ||
| === Ursinha is now known as Ursinha-afk | ||
| === Ursinha-afk is now known as Ursinha | ||
| doanac | rsalveti: i have a small set of hacks in place, but I'm starting to see some really postive stuff with testing the x86 emulator. gonna keep digging, but it looks like you guys have done a pretty great job! | 21:47 |
| rsalveti | doanac: awesome! | 21:47 |
| rsalveti | doanac: yeah, would be nice to have a completely green image on the emulator now :-) | 21:47 |
| rsalveti | so we can start using it for real, when testing our landings | 21:47 |
| rsalveti | need to create a wiki page to describe what is still not working in there | 21:48 |
| rsalveti | so we can follow the progress for each component | 21:48 |
| doanac | yeah. that would be helpful | 21:48 |
| mhall119 | alecu: you couldn't find a more recent picture of yourself for the pioneers page? :) | 22:35 |
| alecu | lol | 22:35 |
| alecu | mhall119: I'm told Nicolás looks just like me without the beard, so I went with it. | 22:36 |
| alecu | mhall119: plus, inside the apps t-shirt, it's really me. | 22:36 |
| mhall119 | lol | 22:37 |
| MoPac | Hi all. I know this is going to be a dumb question, but I have contradictory ideas in my head and can't find an obvious reliable source. At this point, do the plans for Ubuntu Touch include the ability to run standard Android applications? | 23:10 |
| popey | MoPac: no | 23:17 |
| MoPac | popey: Thanks. I assume that also means that there won't likely be a wine-like layer for them either? | 23:18 |
| popey | no plan to | 23:22 |
| MoPac | Cool; good to know | 23:24 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
