[00:01] cyphermox: would be good to trigger daily release for the phone stack as well to pick up all the latest stuff [00:03] <_polto_> sorry to ask for N-time, but I was not able to find any reply... Any chance to get WebGL working in the browser ? You can test here : http://www3.elphel.com/eyesis4pi_samples [00:04] <_polto_> I am trying on Nexus7 device. === chriadam|away is now known as chriadam [00:09] <_polto_> also, is it already possible to access to GPS data and to accelerometers ? [00:10] I think the GPS & accelerometer data is possible, yeah. [00:10] I don't know our browser code, but I suspect WebGL will take a bit of effort [00:15] <_polto_> RAOF, too bad for webgl. why is n't it based on firefox ? [00:16] <_polto_> will it be possible to run plain firefox ? [00:16] <_polto_> or chrome ? [00:16] No; there's no X server [00:17] <_polto_> RAOF, any docs for GPS pls ? [00:17] They'd be on the wiki; I don't know where they are, sorry. [00:30] <_polto_> are there any plan for VNC (or similar) support ? [00:42] bfiller: yes, i'll do it soonish [00:43] cyphermox: unity8 is merged [00:43] yep [00:44] I'll start pushing the changes to config but we'll wait for the next run to pick things up [00:44] it's in one hour [00:44] great [00:44] things are starting to take pretty much all four hours to get built and tested and everything with all the depends [00:57] Greetings. I'm trying to figure out how to read the changelog for ubuntu touch preview, and it seems there are little to no changes between from saucy-41 to saucy-43. Am I correct in my reading? http://cdimage.ubuntu.com/ubuntu-touch-preview/daily-preinstalled/current/saucy-preinstalled.changelog === bfiller is now known as bfiller_afk [01:42] rsalveti: it should, yes... the default route would only change, but I can double check === sergiuse1s is now known as sergiusens [01:55] A QML question (hope it's appropriate here): Is it possible to individually style the values in a ValueSelector? [01:55] If I copy and modify the ValueSelector code, I can make it work, but that seems silly. [01:55] Is there a way to do it with inheritance or accessing the child elements? [02:17] cyphermox: anything missing for the unity8 daily release? [02:17] rsalveti: not afaik [02:17] let's just hope all the tests pass for every stack it depends on ;) [02:18] I'm watching the jobs [02:18] hahah [02:18] cool [02:18] just waiting that to trigger a new image :-) [02:18] will keep watching [02:41] rsalveti: unity8 tests are running... though it's likely to be stuck due to other stacks [02:42] ah, woot [02:42] it depends only on platform and it's all green [02:42] awesome [02:42] so stay tuned, probably another 15 minutes I guess? [02:42] let's hope [02:42] this is prime minecraft time ;) [02:43] haha [02:44] wrong window === chihchun_afk is now known as chihchun === Namidairo`bnc is now known as Namidairo [03:52] rsalveti: publishing unity8 [04:14] cyphermox: cool [04:14] Does anyone know if unity8 is supposed to run on saucy? [04:15] I tried it yesterday and no luck. I'm about to try again on a fresh install. [04:16] rsalveti: unity8 has landed with bfiller's changes, AFAICT [04:16] rsalveti: and also address-book-app [04:16] yeah [04:16] just waiting to be published and will trigger a new image [04:16] but our job is done here :-) [04:17] they're not published yet, right [04:17] btw, I think I really fixed the provisioning bug [04:17] should be in a few minutes [04:17] awesome, what was the issue? [04:18] I'll start a build now, and upload to the same people.u.c url [04:18] well the issue is that apparently ofono is very unreliable with its dbus signals [04:18] and for some reason the file watches don't work so great [04:19] so so far, I'm having to try to read contexts when we get the signal, or if the ContextManager interface comes up, or if the modem goes Attached + Powered [04:19] that should cover all bases [04:19] (I hope) [04:19] right [04:19] great [04:20] patch: http://paste.ubuntu.com/6039033/ [04:20] the whole shebang is on the packaging branch already, and that's synced with the latest stuff in the archive now [04:21] cool [04:21] looks sane, just need testing [04:21] please email folks once you get the package [04:21] so we can get awe and sergio to test that during tomorrow's morning as well [04:24] yup [04:25] it will build during the night, I'll send an email in the morning [04:25] it's getting late now :) [04:27] everything got published, starting a new build === HelenaKitty is now known as INeedSomeSex === INeedSomeSex is now known as HelenaKitty [07:08] any prgress yet on ubuntu tablets? [07:16] lotuspsychje: we have an image for the nexus 10 - which is a tablet ☻ [07:16] lotuspsychje: and the nexus 7 which is a smaller tablet [07:16] popey: does the nexus7 work flawlesly on touch? [07:17] popey: or is still bith edgy? [07:18] lotuspsychje: all the images are in the same roughly similar state [07:18] i cant wait to use an ubuntu touch device.. [07:18] i use one daily ☻ [07:18] popey: but you can browse the web and such nicely? [07:22] so if i buy the nexus 7 or 10 i can use ubuntu touch good? [07:23] popey: what is the default password of the ubuntu touch image? [07:23] phablet [07:23] thnx [07:23] btw does camera work on a nexus 4? [07:23] lotuspsychje: it's not finished yet so there's some bugs, rough edges [07:23] yes nik90_ [07:23] installed months ago with no problem on my nexus 7 and the desktop images as well. [07:23] wilee-nilee: nice one, happy with it? [07:24] lotuspsychje, I only used it shortly, as suggested was a bit rough then, not sure now though. [07:25] touch seems more phone orientated I liked the desktop it was raring with unity. [07:25] wilee-nilee: but could your browse the web decently with it [07:25] and terminal jobs [07:26] lotuspsychje, The touch thing was a bit rough with raring not real functional at that time, but was momths ago. [07:26] i should probably wait bit longer then :p [07:26] popey: do you know if it is possible to install .click without root access? [07:27] I think both will be fine on release, they are working hard on it. [07:27] bzoltan: i do not. not played with click much yet [07:27] cant wait wilee-nilee :p [07:27] bzoltan: cjwatson is the man to ask [07:27] I like the idea of ubuntu on the nexus [07:27] i like the idea of ubuntu on any machine :p [07:27] I likes the idea of the edge and docking [07:27] yes for sure [07:28] terminal power on a tablet [07:28] that will look nice [07:29] to own your tablet without restrictions [07:48] wilee-nilee: https://www.youtube.com/watch?v=cfDghJV9seg [07:49] looks pretty decent to me [08:12] hello [08:12] nik90_: http://popey.com/~alan/device-2013-08-29-091214.png - set my location to Farnborough, where it thinks it's midnight [08:12] hello ax562 [08:13] watching a doc on h@cker electron [08:13] so was everyone sad about the ubuntu touch [08:13] nik90_: oddly it sees the right time for copenhagen http://popey.com/~alan/device-2013-08-29-091338.png [08:14] Good morning peeps :) [08:14] yo [08:15] sup bigcalm? [08:15] That's odd. Since reflashing my Galaxy Nexus yesterday (due to boot loader looping), I can't start the adb shell or ssh from qtcreator [08:16] Is there a step I missed? [08:16] iain@snafu2:~$ adb shell [08:16] error: insufficient permissions for device === schwuk_away is now known as schwuk === Nitrodex is now known as Guest92121 [08:20] Unplugging/replugging the usb cable fixed the adb shell problem [08:20] But I still can't ssh from qtcreator [08:21] Output of detecting devices in qtcreator http://pastebin.ubuntu.com/6039514/ [08:21] bigcalm: you might need to go to the Devices tab and redetect the device and enable the development mode again [08:23] bzoltan: this is the output of clicking enable developer mode: http://pastebin.ubuntu.com/6039518/ [08:24] bigcalm: adb shell;apt-get update; apt-get update --fix-missing [08:24] Ta [08:25] It does make me wonder why it got into that state [08:25] bigcalm: for the request of many I made the "Enable development mode" lighter and faster ... the cost is that the image is out of sync withe Saucy archive, so people need to do manual update [08:25] Ah, okay [08:25] Thanks :) [08:25] bigcalm: I guess I will add back the updating part ... causes problems for too many [08:26] bigcalm: and sorry for the trouble :) [08:27] bzoltan: not a problem really. I know everything is still somewhat fresh :) [08:29] I'm just pleased that I can do ctrl+f12 once more :D [08:29] Right, back to paid work :( [08:30] bzoltan: pkcon install-local foo.click [08:30] cjwatson: thanks [08:30] cjwatson, hey [08:30] cjwatson, do you still plan to add the icon/installed-size to the click manifest in august? [08:31] seb128: eh, vacation intervened. maybe I can squeeze it in on Friday ... [08:32] cjwatson, next week is fine as well, just making sure it's still on your list (and having it in the next weeks would be useful, I need it to enable the click codepath in system settings) [08:33] Right, understood [08:33] icon is IIRC just a matter of convention, installed-size requires code [08:35] cjwatson, icon would be a good start, I can't probably do without the installed-size (I'm going to eventually add code to get the actual size of directories, including app datas anyway) [08:35] I can* [08:48] cjwatson: is there something i do wrong? https://pastebin.canonical.com/96552/ [08:51] bzoltan: You're on a r/o system image, right? [08:51] cjwatson: no [08:51] Because it looks rather like https://bugs.launchpad.net/ubuntu/+source/lxc-android-config/+bug/1217462 [08:51] Launchpad bug 1217462 in lxc-android-config (Ubuntu) "packagekit crashes when installing clicks on read-only image" [Undecided,Fix released] [08:52] ah, maybe not quite [08:52] bzoltan: it works for me, anyway (r/o system image after applying the fix for that bug) [08:52] cjwatson: I wonder if I need to implement two different installation script, one for RO and one for devmode [08:53] ccjwatson: I flashed dimage-touch [09:00] cjwatson: so right now on cdimage-touch I need root access to install .click. Is that correct? [09:01] bzoltan cjwatson: seems like a CK issue [09:02] packagekit should be autospawned, but you don't have access because you're not considered a privileged user [09:02] bzoltan: You shouldn't need two different scripts; please don't do that :) [09:02] I thought this was borken from unity session too, so kind of surprizing it does work actually [09:02] bzoltan: Is this via adb or the terminal app? [09:03] Or something else? [09:03] cjwatson: I am not doing it :) that is why I am asking how does it expected to work [09:03] cjwatson: adb [09:03] bzoltan: You are absolutely not expected to need root access to install .click; I haven't tried cdimage-touch very recently but it worked for me not that long ago [09:04] bzoltan: However I think you need somebody who understands the consolekit (etc.) level of things better than I do [09:08] cjwatson: OK, thanks. I keep receiving bugs against the SDK that it does not support click installation...I wish to provide a solution for the application developers but i have no knowledge on how it should work. [09:11] bzoltan: It should work as I described. Occasionally there are bugs in the stack I'm depending on. [09:12] cjwatson: right, thanks [09:12] bzoltan: (pkcon is the command-line approach; there's also the PackageKit D-Bus API which is preferred for cases where you already have convenient D-Bus bindings.) [09:18] ogra_: what happened to the builds? floodgating worked it seems :) [09:19] yeah [09:19] lets see how 29.1 comes out, looks like 29 didnt even run through [09:21] is it the time already? [09:22] lool, 8:00 UTC is the build ... the dashboard picks up about 30-45min later [09:22] 8:02 indeed [09:22] yeah [09:22] there's a 29 build on cdimage [09:22] thats from last night [09:23] 29.1 is the oending one [09:23] *pending [09:23] cjwatson: is pkcon the way SDK should install a Click, or should it simply run click install? [09:24] it needs to run as phablet anyway [09:24] you want the dbus session address etc [09:24] So... is pending or current built without pulse or something? [09:24] I don't think you need the dbus session address [09:24] it's more of a permission issue I suspect [09:24] if you want to run the app remotely after installing it you do [09:24] Morning! [09:25] also the upstart session env [09:25] hey OrokuSaki [09:25] Sup ogra! [09:25] Trying pending for fun [09:25] OrokuSaki, pending and current have the new pulse setup now [09:25] sweet!!!! gracias [09:25] hmm click install would need either root or clickpkg user I guess [09:25] be back in 30 or an hour... tablet charging [09:25] lool: pkcon or the packagekit api [09:26] dbus api that is [09:26] but looking at http://reports.qa.ubuntu.com/smokeng/saucy/touch/ the last build wasnt so great [09:26] lool: the correct click install invocation to get user registration right is possible but awkward [09:26] lool: and yes, click install would have to be started as root [09:26] so the SDK must not rely on it [09:27] (it needs to be clickpkg to unpack the files, and to be the user to register them) [09:56] barry: So I adb pushed systemimage/state.py and bag.py and I could see the update in system-image-cli -v -v --dry-run and then in the system settings, but I pressed download the UI just sat there rather than saying that the download was in progress; afterwards it did offer to reboot though, and it's now deploying [09:56] this might be due to the translation changes? anyway, it's better than not working at all with daily-proposed [09:57] barry: would love if you could push this in the archive :-) [10:15] Love the new first run tour! Well done whoever did that! === chihchun is now known as chihchun_afk === chriadam is now known as chriadam|away [11:00] asac, i must say my local testing of 29.1 can't really confirm the bad looking dashboard (and i bet popey can't either for mako) [11:01] * ogra_ thinks the image never worked that well before [11:05] ogra_, did asac's message mean that there is a new image in current? [11:05] ogra_: dashboard? [11:06] ogra_: yes [11:06] rickspencer3: yes [11:06] yeah === gusch is now known as gusch|lunch [11:06] ! [11:06] ogra_: i assume they had landings [11:06] ogra_: i have 29.1 on mine.. [11:07] of new apps and tests were not updated? [11:08] ogra_: did you check what changed on the image? [11:10] ogra_: network seems fudged on 29.1. I get an IP but it doesn't pass packets http://paste.ubuntu.com/6039923/ [11:10] * popey restarts network-manager [11:16] ogra_: i think a script to show diffs on the packages file for two builds would be nice :) [11:18] oSoMoN_: any idea if you landed something big on the browser front like yesterday? [11:18] oSoMoN_: http://reports.qa.ubuntu.com/smokeng/saucy/image/3825/webbrowser-app-autopilot/ [11:18] seems to be a bit unhappy [11:19] oSoMoN_: also maybe check with others i your team about the other apps: http://reports.qa.ubuntu.com/smokeng/saucy/image/3825/ [11:20] the user agent changed didn't it? [11:22] "Mozilla/5.0 (Ubuntu; Mobile) WebKit/537.21" is the UA now [11:24] popey: you say that might cause test failures? [11:26] asac, you mean the manifests ? [11:27] the seeds change so rarely that i never found it worth it .... i thought about that before [11:29] asac: nothing big landed since last week, I’ll take a look at the failures [11:29] asac: looking at those tests, maybe not. [11:29] WARNING: ignoring unknown switch "--desktop_file_hint=/usr/share/applications/webbrowser-app.desktop" [11:29] thats interesting. when did that become deprecated? [11:30] when we switched to apparmor wrapping ? (did we for non click yet ?) [11:30] ah okay. [11:31] well, note the question marks in that sentence :) [11:31] oSoMoN_: nothing landed ... or nothing big :)? [11:32] asac: actually, nothing (the last meaningful revision is from the 19th, all the other are automatic translation updates from LP) [11:32] asac: looks like it might be a regression in the UITK, I’ll have a closer look [11:33] package diff seems to be: http://paste.ubuntu.com/6039989/ [11:33] asac, the android package was synced with phablet.u.c after being out of sync for a bit ... that might perhaps have some impact as well [11:34] oSoMoN_: yeah i suspect [11:34] ogra_: i think uitk feels like the most reasonable story given how many failures we see across the board [11:35] yeah, could be [11:35] @ogra yep.. music app cannot play, not even with pulse shut off [11:35] how would it play with pulse shut off ? [11:35] play\pause no workie just like the report said. =) [11:35] pulse is the only audio service left now [11:35] it did on my last image... from 8-27 [11:35] sil2100: Mirv: can we get a couple of app autopilots run during sdk stack daily-release? [11:35] at best all :)? [11:36] And it does not play with pulse running either [11:36] wonder if this would have been catched (in case it is an issue) [11:36] 0827 was only half there [11:36] before it would play without pulse running, now it does not... that is my status report =) [11:36] side stage works better though [11:36] 0829 is the first one with everything in proper order, all obsolete bits dropped and all new bits working (on the supported devices) [11:36] (wrt sound) [11:37] ran 8-29.1 [11:37] me too here [11:37] do you have sound? [11:37] * ogra_ likes the new phone app stack [11:37] sure [11:37] hmmm [11:38] I noticed on the report the music app couldn't play\pause for.. mako or something === hikiko is now known as hikiko|lunch [11:38] sil2100: did you say something about eg. unity8 tests not possible to run at the moment for sdk? is it the dependency chains? [11:38] oh. qcom has some proprietary lib that enables the sound device... the android HAL used to use that lib so sound worked. Now we have only ALSA left, that part is missing (and calling into proprietary libs from the kernel is not allowed).. maybe I can add a simple call-proprietary-lib-tool on startup :/ [11:38] * ogra_ ddoesnt have any music on the phone ... i re-flash/wipe to often [11:40] http://reports.qa.ubuntu.com/smokeng/saucy/image/3825/music-app-autopilot/ [11:40] no pause\play [11:41] OrokuSaki, i think thats just because the buttons are gone from the UI [11:42] the new music app doesnt show any UI elements if you dont have music imported [11:42] i would guess the test was just not updated [11:43] mpt: howdy. Check https://wiki.ubuntu.com/SystemSettings#A.2BIBw-Reset_Phone.2BIB0- — the alert text for the second entry is the same as the first. I guess that's not intended. [11:44] OrokuSaki, oh, in that specific report: Error: "Your GStreamer installation is missing a plug-in. [11:46] asac: when running some webbrowser-app tests locally on maguro, I’m seeing this error message, which seems to confirm my suspicion of a regression in the UITK: file:///usr/lib/arm-linux-gnueabihf/qt5/qml/Ubuntu/Components/Panel.qml:333: Error: Cannot assign to non-existent property "savedLocked" === oSoMoN_ is now known as oSoMoN [11:51] can you provide factory image for samsung galaxy s duos device | http://askubuntu.com/q/338850 === alan_g is now known as alan_g|lunch [12:05] Laney, thanks, fixed. [12:06] mpt, hey, did you see my question/note about silent mode and how to enable it yesterday? do you want me to open a bug somewhere to track that? [12:06] seb128, I didn't. I was assuming it had a hardware switch. [12:08] mpt, hum, I'm not sure that's a right assumption [12:08] well, even if it has we probably need a piece of UI to see the status/confirm it? [12:08] on my android phone there is a dialog that is display that let you see flight mode/silent mode/... when you keep the power button pressed some seconds [12:10] oSoMoN: thx [12:16] I've posted new screenshots of blackjack app: https://plus.google.com/110587021591787005452/posts/CS5ppRPo3TW [12:17] @okra Hmmm I have the UI an I have one song... gstreamer... is it me or is it the build? [12:17] gstreamer is.. a codec? [12:17] or a library to play music like ffdshow? === _salem is now known as salem_ [12:29] ogra_: 29.1 is a good one [12:29] rsalveti, not according to the dashboard though [12:29] new phone-app, improved pulse support and a few other fixes [12:29] yeah, details :P [12:29] guess we need to adjust the test cases for that as well === hikiko|lunch is now known as hikiko === gusch|lunch is now known as gusch [12:32] barry: doing another run from the UI; cli reports build number: 20130834, but UI offers Version 20130834 for download; dowload failed for blacklist in the first try, trying again [12:33] rsalveti, that, and seems popey saw wlan issues [12:33] barry: and again on second try; so seems like a bug [12:33] barry: I'm against daily-proposed with the two files I mentioned deployed on the device [12:33] Can't play a wav file either... mixer has no volume control, stays muted [12:34] msm-audio does show up in alsamixer [12:34] barry: I get Download failed: /tmp/system-image/blacklist.tar.xz [12:34] stgraber: ^ [12:34] lets try aplay [12:35] alecu, ralsina: does searching/installing/launching click apps work for you in today's image? [12:35] it didn't work for me on today's image - I couldn't find the xda app for example [12:36] barry: BTW /var/log/system-image/client.log is empty [12:38] http://paste.ubuntu.com/6040179/ [12:38] pmcgowan: do you know about contacts search in the contacts-app? [12:39] sergiusens, maybe not, whats the issue? [12:39] pmcgowan: I can't find a search box [12:39] sergiusens, flashing right now [12:39] pmcgowan: I have a 400 contacts list to stress it [12:40] pmcgowan: faster and dialer doesn't hang (although I had tons of messages I filled it up with too and with the backend change lost) [12:40] @ogra "Failed to drain stream: Timeout" when playing a wav [12:40] well, is pulse running ? [12:41] you could try https://wiki.ubuntu.com/PulseAudio/Log (as phablet user) and file a bug with the collected data [12:41] yeah.... can you try paplay with a wav? What happens with you? [12:42] I started pulse as phablet with pulseaudio --start [12:42] waited for it to start.. then used paplay [12:42] barry: ah hold on, this might relate to me getting switched back to daily since I got an updated client.ini from the upgrade, tricky [12:42] paplay should work [12:42] hmmmm [12:43] I noticed in alsamixer a soundcard was missing... I guess that is what changed [12:43] I had 2.. one for lxc or something and the other was msm-audio, now I have 1 [12:47] ogra_: popey: do we have bugs for the wlan issues? [12:48] let me also flash my mako [12:48] i dont think so [12:48] not yet [12:48] happy to file === jhodapp|afk is now known as jhodapp [12:49] barry: same problem after changing to daily-proposed [12:49] * popey reflashes mako to test [12:50] sergiusens, seems search not enable here either [12:50] Powered off Galaxy Nexus. Pressing power button shows battery state image. Long pressing does the same upon release. Is this a weirdness of the phone or an issue for ere? [12:50] pmcgowan: I'll log a bug [12:51] boiko, are you working on contacts ui? [12:51] dholbach: Hm, yeah, now I'm seeing the same thing bzoltan was earlier, after upgrading [12:51] bigcalm: remove USB cable and try holding the power button for a couple of seconds after battery shows up [12:51] I like the new phone app [12:51] s/and/or [12:51] boiko: hi! I've the qtpim update ready and smoke-tested, I'd need someone to sponsor it https://code.launchpad.net/~timo-jyrinki/ubuntu/saucy/qtpim-opensource-src/new_snapshot/+merge/182874 [12:51] rsalveti: a lot smoother and less invasive to the eyes IMO [12:51] yeah [12:51] rsalveti, did you see the edges tour? [12:52] yeah, awesome [12:52] lool: this is with the usb cable not plugged in [12:52] dholbach,bzoltan: I think this is a seed bug - the touch images need to have packagekit proper rather than python3-aptdaemon.pkcompat [12:52] bigcalm: odd; I don't have Galaxy Nexus but have something similar with USB cable plugged on N7 where I have to time the power button right to boot it [12:52] bigcalm: don't know for USB unplugged then, sorry [12:53] This is getting weirder. Just pressed the power button and it showed the battery state again. But this time is showing it as though it's being charged. Also, the back of the phone, around the camera area, is hot! [12:53] cjwatson, wow - well spotted - I'll try that out in a bit [12:53] Time to pull the batter [12:53] bzoltan: did you log a bug for the pkcon issue with packagekit dbus activation? would be good to escalate this one [12:53] y [12:53] lool, the 2A power supply might help [12:53] ogra_: we can now move away from the jenkins build, the android tarball is already reflecting the latest stuff [12:53] lool: I'm already fixing it [12:53] cjwatson: ah cool; wasn't sure [12:53] rsalveti: \o/ [12:53] At least I'm pretty sure this is the cause [12:53] sergiusens: is there a sorta one stop shop for all the click packages you are working on? I'd like to test apparmor policy on all of them [12:54] OrokuSaki, please ping me if you get msm-audio working [12:54] cjwatson: oh just saw your line about packagekit above [12:54] bigcalm: probably run away process which ate the cpu and sucked the battery [12:54] bigcalm: commonly ueventd is the runaway process in question, but can also be unity8 [12:54] lool: /var/lib/sudo is mode 041777 on today's system image. Did the initrd fix not land in time? [12:54] cjwatson: it didn't; next image will have it [12:54] popey: I should this happen if I have turned the phone off with a long press of the power button? [12:54] jdstrand: http://10.97.2.10:8080/view/click/? [12:55] sergiusens: ie, you gave me an app specific URL for dropping letters yesterday. how can I enumerate the apps? [12:55] lool: OK, good [12:55] bigcalm: well, if the battery is well and truly dead it may not have enough juice to power up [12:55] cjwatson: I had a typo and the android package had a FTBFS, these things delayed it too much, but .1 image has it [12:55] popey: fully battery [12:55] (also common) [12:55] or even 29 image [12:55] bigcalm: hmm, not seen that but i dont have a gnex [12:55] sergiusens: perfect, thanks! :) [12:55] cjwatson: we don't have a transition snippet, but I don't think we care [12:55] lool: I don't think it's needed [12:55] popey: I'm starting to wonder if anybody does :P [12:55] sergiusens: I'll try to remember this one-- I'm pretty sure you gave it to me once before :) [12:56] bigcalm: ogra_ has one, and i think davmor2 does too [12:56] lool: Incidentally, if I remount rw and install a package then go back to ro, do you know if the system image upgrader will notice and make sure to deploy a full image? [12:56] Or if there's some way to force it? [12:56] cjwatson: it wont [12:56] rsalveti: where do you want this network manager bug filed? [12:56] rsalveti, awesome, i'll try to figure out the cdimage stuff then [12:56] davmor2 is on holiday. But he has his irssi proxy set up the same as mine. So just by saying his name, he's getting emails :D [12:57] cjwatson: there is a bug open to force a full, but don't think it's done; what it should notice in the future is when you're in writable_image mode, it should offer to switch back to read-only by deploying a full [12:57] cjwatson: but these are future features [12:57] popey: can be against the nm package itself, or the touch-preview-images [12:57] bigcalm: poor davmor2 [12:57] ok rsalveti [12:57] jdstrand: no sweat... I've been looking at some other apps that are arch dependant and it seems they only use it to create the launcher... so I might add those soon too (they'll be in that list) [12:58] cjwatson: it's covered in https://bugs.launchpad.net/ubuntu-system-image/+bug/1208909 and https://bugs.launchpad.net/ubuntu-system-image/+bug/1192575 I think [12:59] Launchpad bug 1208909 in Ubuntu system image "Add support for forcing full image update" [High,Fix committed] [12:59] Launchpad bug 1192575 in Ubuntu system image "Support for changing channels and reinstalling" [Wishlist,Triaged] [12:59] Are there plans to add an image (animated or otherwise) after the Google logo and before the phone is useable? I would love to know that my phone is actually booting and not just dead [12:59] popey: is the issue that's not asking for password? [13:00] the interesting thing is that it worked just fine with maguro [13:00] but with mako it's not asking for the AP password === alan_g|lunch is now known as alan_g [13:00] rsalveti: no, probem is that it has an IP but I can't ping out of the phone [13:00] hm, rifght [13:00] let me reboot this [13:01] rsalveti: https://bugs.launchpad.net/ubuntu/+source/network-manager/+bug/1218365 [13:01] Launchpad bug 1218365 in network-manager (Ubuntu) "network half-working after flashing mako 20130829.1" [Undecided,New] [13:02] rsalveti: added a comment, the default route looks screwy to me [13:03] cjwatson: I've apt-get installed packagekit after reproducing the pkcon install-local issue, pulled quite some packages and got a bunch some conffile prompts, rebooted read-only and could confirm that it proceeds now! [13:03] cjwatson: good catch [13:03] I hadn't passed --no-install-recommends; it pulled:gdebi-core libelfg0 libglib2.0-bin libglib2.0-data packagekit packagekit-backend-aptcc python-apt python-chardet python-debian python-six python3-packagekit [13:04] hope the conffile stuff didn't matter [13:04] bzoltan: so pkcon should work soon; when relevant packagekit packages are pulled in the image and image is rebuilt [13:06] cyphermox: awe: bug 1218365 [13:06] bug 1218365 in network-manager (Ubuntu) "network half-working after flashing mako 20130829.1" [Undecided,New] https://launchpad.net/bugs/1218365 [13:06] that's with 3g and wifi [13:06] ogra_, ^^ [13:06] ogra_, oops, forgot to ping you... we're on mumble [13:06] oops [13:07] * ogra_ rushes to the office [13:07] lool: recommends makes no difference [13:07] lool: I guess I can always flash cdimage-touch and then ubuntu-system to force it ... [13:07] cjwatson: yeah, just mentioning it case it did since we're not pulling them in the touch image I think [13:08] cjwatson: you could try changing /etc/ubuntu-build to version 0, this should be enough to force system-image to do a full [13:08] cjwatson: BTW you can flash ubuntu-system directly too :-) [13:08] Doesn't that try for a delta if it can? [13:08] there wont be a delta from version 0 [13:08] but it will try :-) [13:09] Oh, sure, I meant that flashing ubuntu-system without changing /etc/ubuntu-build wouldn't work here [13:10] cjwatson: flashing ubuntu-system with phablet-flash shouldn't look at etc/ubuntu-build all? not sure I got that last one [13:10] lool: I didn't get any conffile prompts, fwiw [13:10] lool: What I mean is that I have seen "phablet-flash ubuntu-system" use deltas [13:10] I meant either you may use system-image on the image after setting ubuntu-build to 0, or you may phablet-flash ubuntu-system [13:10] I'm fairly sure [13:10] uh [13:10] it shouldn't [13:10] Maybe I was hallucinating [13:11] I've not really stopped working very much this week so it's possible :-P [13:11] nah, just checked the code in bzr and it doesn't [13:12] it only looks for "full" images in the json, and it doesn't poke at the current version to compute deltas either [13:12] ogra_: seb128: so is https://launchpad.net/ubuntu/+source/webbrowser-app in main or universe :) [13:12] the launchpad page is a bit sneaky in that it says "release (universe)", but Component: main :) [13:12] cjwatson: ? [13:12] asac, universe [13:12] asac: look at +publishinghistory for the latest thing [13:12] seb128: and what does "Component:* main" mean? [13:13] cjwatson, " *actual publishing details may vary in this distribution, these are just the package defaults. " [13:13] I think that's probably the component in the source or something. I can check if you like [13:13] ups [13:13] But universe is correct [13:13] asac, ^ [13:13] oh right :) [13:13] hehe [13:13] cjwatson, sorry [13:13] probably bogus component [13:13] wonder who got that wrong :) [13:13] cjwatson: thanks! [13:13] asac: nobody got it wrong - packages do not typically explicitly say universe [13:13] asac: this is usually applied in overrides [13:14] cjwatson: hmm. i see [13:14] for a momoent i thought Component: was what "Section: is" [13:14] it's more obvious if you look at something like https://launchpad.net/ubuntu/+source/trn4 where the Component listed is not an Ubuntu one [13:14] right [13:14] its the debian componbent [13:14] i now remember [13:14] i think :) [13:15] It's one of those things that makes perfect sense when you know the model and is a bit surprising when you don't. Anyway [13:16] lool: uploaded ubuntu-touch-meta with the extra packagekit dep [13:16] dholbach: I just flashed with "phablet-flash cdimage-touch --wipe", and after enabling network and trying a search I can see both apps [13:16] dholbach: xda gets stuck when installing, I suppose due to the bug in packagekit that was fixed yesterday [13:17] lool: deltas> I guess I misread the output [13:17] dholbach: which image did you flash? [13:17] cjwatson: packagekit >> thanks [13:18] alecu: try "sudo apt-get install packagekit" first [13:18] (will be fixed in the next image) [13:18] sergiusens: hey, so I'm finding a few bugs with the click packaging. should I file them somewhere? [13:18] tedg: would you think you could handle https://bugs.launchpad.net/ubuntu/+source/upstart-app-launch/+bug/1215478 ? [13:18] Launchpad bug 1215478 in upstart-app-launch (Ubuntu) "upstart-app-launch-desktop hook fails when ~/.local/share/applications doesn't exist" [High,New] [13:18] sergiusens: if so, where? [13:18] cjwatson: you are right, this is the error in the downloader log: "Failed to contact PackageKit: Error calling StartServiceByName for org.freedesktop.PackageKit: GDBus.Error:org.freedesktop.DBus.Error.Spawn.ExecFailed: Cannot launch daemon, file not found or permissions invalid" [13:19] tedg: I first experienced it when we were deploying clicks; it doesn't seem to be present in the current way we setup images, but it seems a trivial robustness fix; would be safer/more efficient if you'd prepare the mkdir mp if you don't mind :-) [13:19] cjwatson: perhaps we can have unity-scope-click depend on packagekit? [13:19] lool, Yeah, I thought we fixed that when cjwatson reviewed the code... but perhaps that was something else. [13:19] tedg: I reproduced with current versin at least; recipe in the latest comment [13:20] lool, We reshuffled some code, I'm guessing the fix got lost in the reshuffle. [13:20] ack [13:22] alecu: No, I fixed it in ubuntu-touch-meta instead. [13:23] alecu: unity-scope-click would be the wrong place for this; it should only depend on the API and the details of which PK implementation is present should be dealt with by the system (i.e. the metapackages). [13:23] alecu, --pending [13:24] cjwatson: ack. Now I installed packagekit, and pkcon exits ok, but the package is not installed right. [13:24] cjwatson: the package folder is created (ie, 0.1.5), but it's empty [13:25] sergiusens: is --pending installing the RO images? [13:26] dholbach: I'll try reflashing with --pending [13:26] I'm just doing a dist-upgrade and installing packagekit [13:26] to try how that changes things [13:27] alecu: seems to work for me ... [13:28] alecu: which package exactly? [13:28] (url) [13:29] oh, there's a 29.1 image - I'll try that [13:29] in any case: no dice up until now [13:29] pmcgowan: renato is the one who owns the address-book-app, but salem_ is doing some UI improvements on it [13:29] ping nik90_ [13:29] Mirv: I will test that one [13:30] cjwatson: https://public.apps.ubuntu.com/download/com.ubuntu.developer.mhall119/xda-developers-app/com.ubuntu.developer.mhall119.xda-developers-app-0.1.5.click?noauth=1 [13:31] cjwatson: I've flashed this morning with "phablet-flash cdimage-touch --wipe", I'm trying now with "--pending" === tetris4_ is now known as tetris4 [13:33] salem_, do you know about having search enabled in the new contacts app? [13:34] alecu: Seems to work for me with pkcon [13:34] (today's ubuntu-system with packagekit installed) [13:34] pmcgowan, it's not yet implemented. [13:35] salem_, is there an eta for it? a bit of a regression [13:35] alecu: Let me know if you need me to try with cdimage-touch [13:35] cjwatson: are you using ubuntu-system? [13:35] Yes [13:35] Usually ubuntu-system is a strictly tougher test though :-) [13:36] great [13:36] awe: cyphermox: tedg: bug 1218385 [13:36] bug 1218385 in indicator-network (Ubuntu) "Network indicator doesn't ask for the AP password" [Undecided,New] https://launchpad.net/bugs/1218385 [13:37] tedg, ^^ [13:37] boiko: it got sponsored now! [13:37] rsalveti: ogra_ https://code.launchpad.net/~sergiusens/ubuntu-seeds/libqtfolks1--/+merge/182900 [13:37] rsalveti, on mako, my Wi-Fi toggle is set to 'off' [13:38] Mirv: ok :) [13:38] jdstrand: evilapp isn't in appstore? [13:38] even though the NM props are enabled [13:38] ;( [13:38] seems it's always showing itself as off by default [13:38] at least I get that at every first boot [13:38] lool: not yet, no [13:38] oh yea, and networka appear in the list [13:38] s/networka/networks/ [13:38] lool: it has compiled code. waiting on tha to be worked out [13:38] Hmm, I think that Unity 8 is still the secret agent there, right dednick? [13:38] I wonder if having two messed it up. [13:39] tedg: it does [13:39] tedg, did the indicator change wrt to the wifi toggle? [13:39] awe, No [13:39] hmmm [13:39] But the thing that did change is we installed our stub secret agent. [13:39] mine shows off, yet networks are listed, and NM's props show enabled [13:39] pmcgowan, christina said she needs to talk to oren about it. We dont have the visual design for it yet. [13:39] sergiusens: can we safely remove that? [13:40] rsalveti: well no one directly depends on it [13:40] rsalveti: salem_ ^^ [13:40] salem_: can we remove libqtfolks1 from the images? [13:40] it only depends on the seed [13:41] rsalveti, Can you see if "stop indicator-secret-agent" does anything? I'd also be curious if it had data in its log. [13:41] sergiusens, I think so, phone-app was the only package depending on it. I believe it's safe to remove it. [13:41] boiko, ^ [13:41] sergiusens: some of the apps are missing policy groups. should I file bugs? [13:41] tedg: will check [13:41] We may need to disable it until we transition. [13:41] jdstrand: yes please [13:41] I thought it'd be benign... [13:41] sergiusens: where shall I? [13:42] sergiusens: yeah, I think it can be removed [13:42] alright [13:42] jdstrand: that's a good question, either an umbrella one to touch-preview-images or to each lp project individually [13:43] sergiusens: shall I assign you? [13:43] jdstrand: yes [13:43] thanks === dandrader is now known as dandrader|afk [13:44] rsalveti: ah, last time I went through this we weren't on the pulse image, no libqtgstreamer-0.10-0 isn't pulled in [13:45] asac, http://paste.ubuntu.com/6040362/ .... comapre-manifests [13:47] ogra_: rsalveti 3 packages need to be in universe to get the PPA out https://launchpad.net/~phablet-team/+archive/ppa?field.series_filter=saucy [13:48] tedg: worked fine after stopping indicator-secret-agent [13:48] asac, what do you think ? should we just pull them into universe and finally get rid of the PPA ? ^^^ [13:49] ogra_: well, the deadline for that was like 1 month ago, so if its doable, do it :) [13:49] but talk to folks about their apps :) [13:49] asac: their apps? [13:49] wait a bit [13:50] asac: oh... the 3 things in there are demo content [13:50] sergiusens: check with bfiller. i think we either agreed to remove them [13:50] or to remove stuff from it so it can go into archive [13:50] rsalveti, Okay, I'll disable that upstart job for now. We'll migrate to it when the Unity 8 stuff is fully ready. [13:50] asac, like sergiusens says, all demo content ... imho that can as well dangle around in universe [13:51] asac: right, forgot... ogra_ demo-content can't go into the archive as it is [13:51] ogra_: is it ok licensing wise? [13:51] i know that was one concern [13:51] that we ship like copyrighted movie images :) [13:51] why not ? it is in a public PPA so it cant really have anything offending or illegal [13:52] i think archive is a bit tighter about what is OK and what not than ppa ;) [13:52] not really [13:52] in practice at leaast (in theory you are right i guess) [13:52] ogra_: yeah, lets wait for bfiller :) [13:52] bfiller: ah ... as we speak :) [13:52] speaking of the devil [13:52] bfiller: we wondered about demo-assetys and webapps-demo [13:53] was that cleaned up so we can move it to universe/archive? [13:53] or did we say we kick it out? [13:53] (or neither) [13:53] asac: I think we said we don't need demo-assets [13:53] asac: and that webapps-demo should be renamed and added to the archive [13:53] the image will look quite empty though [13:54] without the demo assets [13:54] bfiller: what name? [13:54] bfiller: can you just do it, or do you need help by ogra/didrocks to get this changed? [13:54] isnt sintel completely open ? [13:54] asac: I need to find the email thread we had going on this topic a while back, can't remember [13:54] we could just drop the other movies [13:54] bfiller: yeah that would be cool. i also have pretty vague memories only :) [13:55] bfiller, the demos are the last bits that make us keep the PPA now (which is why it came up) [13:55] yup [13:55] wil look into it after off my calls [13:55] bfiller: check latest image [13:55] pmcgowan: might have some recollection of that ^^^^ === kentb-out is now known as kentb [13:56] bfiller, btw, the new phone stuff is *beautyful* !! great work === Ursinha is now known as Ursinha-afk [13:56] ogra_: thanks :) [13:57] asac, webapps will be redone, not sure if we want those "demo" packages once they are [13:57] awe: rsalveti cyphermox so far, that fix seems to be working a lot better === Ursinha-afk is now known as Ursinha [13:58] awe: rsalveti cyphermox twice so far and got my gprs setup in less than 30 secs [13:59] do we have a plan what happens if there comes one moe ? [14:00] *more [14:00] err [14:00] * ogra_ notes that the top panel is now packed with icons [14:00] ogra_, asac webapps demo should become individual click packages for real webapps [14:00] sergiusens: great [14:00] that was rather out of order [14:00] the bluetooth one seems to be quite unstable as weel [14:00] well === dandrader|afk is now known as dandrader [14:05] sergiusens: aye [14:09] cjwatson, alecu, lool, reflashed with --pending, dist-upgraded, installed packagekit, but I still don't see click apps in the scope [14:10] dholbach: so you tap the SEARCH button at the top of Apps screen on the right of the Home scope, and then search for e.g. "xda" or "hello" and you dont see anything? [14:10] dholbach: or you install stuff, and then it stops at some point? [14:11] dholbach: may I see your .cache/unity-scope-click.log ? [14:12] lool, normally it would just show up, but searching doesn't show it either [14:12] dholbach: FYI, things are backed up and restored nowadays, so for a true wipe you need --no_backup; not that it would help here, just mentioning it while I think of it [14:12] lool: --no-backup [14:12] dholbach: I'm flashing a ./phablet-flash ubuntu-system --revision -1 --no_backup [14:13] sergiusens: --no_backup is what I get here [14:13] sergiusens: --no-backup fails :-) [14:13] lool: ohhh... blast :-/ [14:13] sergiusens: happier if you change it to --no-backup though [14:13] alecu, http://paste.ubuntu.com/6040437/ [14:13] dholbach: dumb question, but are you on a network? [14:13] ditto [14:14] dholbach: I just flashed like you did (with--pending), and had to connect twice, because somehow I got disconnected [14:14] dholbach: with cable plugged in and if you have wifi on lappy, run phablet-network [14:14] sergiusens, yes [14:15] since I never got my network password asked, I resorted to: nmcli -pretty dev wifi connect password [14:18] sergiusens, using the browser just worked [14:19] well, there are wlan issues on todays image [14:19] (if i understand the backlog right., removing /usr/share/upstart/sessions/indicator-secret-agent.conf and rebooting should work) === alan_g is now known as alan_g|tea [14:32] sergiusens: have you noticed that during the bootup with NM 0ubuntu20, ofono crashes? [14:33] cyphermox: hmm, let me check the logs [14:34] check /var/crash too [14:34] cyphermox: I have the 1 network manager .crash [14:34] cyphermox: want it? [14:35] yeah, why not [14:35] slangasek, ogra_, rsalveti: fyi - bug 1218415 === Namidairo is now known as Namidairo`bnc [14:35] I still don't know how I'm going to retrace them though, seems like the porter box doesn't want to load a chroot for me [14:35] bug 1218415 in linux-maguro (Ubuntu) "upstart needs subreaper prctl support for touch kernels < v3.4" [Undecided,New] https://launchpad.net/bugs/1218415 [14:36] jodh, omg ... [14:36] jodh, that means even more kernel hacks for our community ports :( [14:36] yeah [14:36] ogra_: well technically it doesn't *need* it, but it would be extremely helpful [14:37] jodh: hmm. I thought we had "sensible" fallback behavior when we don't have subreaper? [14:37] at least it seems to be just a small patch [14:37] cyphermox: well in case it's useful http://people.canonical.com/~sergiusens/_usr_sbin_NetworkManager.0.crash [14:37] yeah [14:37] yep, I'll manage [14:37] but they also need apparmor now [14:37] slangasek: yes, we don't hard-enforce, but the point is that currently some touch kernels are >3.4 and some are not (nexus 7 = 3.1) [14:37] the patchset grows [14:37] cyphermox: let me run it a couple more times, but it's only one crash from multiple runs [14:37] alecu, does it work for you now? [14:38] ogra_: is apparmor really required? [14:38] jodh: but if the fallback is sensible, is it worth worrying about backporting subreaper? [14:38] sergiusens: ack [14:38] rsalveti: absolutely, the app isolation model relies on it heavily [14:38] sergiusens: hmm no preinstalled clicks in 28.1? [14:38] slangasek: maybe not, but it could well confuse folk comparing two different touch systems as behaviour would be different [14:38] I mean, would it work without apparmor? [14:38] rsalveti, not really [14:39] sigh [14:39] lool: image based upgrad image? [14:39] you can slightly evil hack around it by modifying all .desktop files [14:39] I doubt we have complete failure handling in the absence of apparmor... if you don't have apparmor, you don't have the Touch platform so there's really no reason to fail gracefully in that case [14:39] sergiusens: ./phablet-flash ubuntu-system --no_backup [14:39] and remove the aa-exec [14:39] that won't work with application lifecycle === alan_g|tea is now known as alan_g [14:39] slangasek, tell that to our 60 or whatnot ports [14:39] (cause upstart doesn't use the desktop files). the evil hack there is modifying the upstart job [14:39] dholbach: after playing with nmcli to connect, I got the scope to work. And, I was able to start the installation of the xda app. [14:39] sergiusens: hmm seems it didn't have latest apparmor-easyprof-ubuntu [14:40] slangasek, of which some still work with a 2.x kernel [14:40] slangasek: right, but forcing porters to backport apparmor to a bunch of different kernels is really painful [14:40] just to get touch working [14:40] yeah [14:40] I'd prefer that for a product, apparmor would be a requirement [14:40] fyi, jjohansen added some preliminary notes on how to do the backport. also, apparmor upstream keeps track of a bunch of patches for different kernel versions [14:40] dholbach: but it seems to be failing again on the installation, I'll check exactly how after this UDS session. [14:40] but not for porters [14:40] ok [14:40] rsalveti: the porting all the way back to 3.0 is done [14:40] ogra_: just because people will try running Touch on kernels that don't provide functionality that's part of our fundamental design doesn't mean we should cripple the platform to support them [14:41] jjohansen: I know, but we have different kernels using 3.1, 3.2, 3.3, even 2.x [14:41] jjohansen, well, many of our porters just find their way around ... but most are not kernel hackers [14:41] and the porter might not necessarily know how to deal with kernel issues and backporting [14:41] yeah [14:41] lool: why don't you use the unstable channel? [14:41] slangasek: not saying we should cripple the platform, just saying it'd be nice for it to work without apparmor as well [14:42] with some restrictions [14:42] ++ [14:42] I think that counts as crippling the platform ;) [14:42] * ogra_ would call it "adding options" [14:42] rsalveti: it can work without apparmor. there are notes in the porting guide on how to do it. but the port is not finished until apparmor is integrated [14:42] :) [14:42] jdstrand: right, that's fine [14:42] yeah [14:43] I just don't want it to not work at all without apparmor [14:43] rsalveti: the requirement is for click packages [14:43] rsalveti: because click packages have unreviewed code [14:43] everything else should work [14:43] right [14:43] heh [14:43] everything else [14:43] soon everything will be click apps [14:43] sergiusens: we don't have an unstable channel yet? [14:43] yeah [14:44] rsalveti: not sure if you saw this-- I added it this week: https://wiki.ubuntu.com/Touch/PortingFlippedInProgress#AppArmor [14:44] no, not yet [14:44] rsalveti: and jjohansen added the notes on how to patch older kernels [14:44] sergiusens: I wanted to test a read-only image specifically [14:44] lool: isn't daily-proposed built out of /pending? [14:44] thats cool [14:44] sergiusens: Yes [14:44] sergiusens: I could try this again [14:45] * ogra_ knew the aa entry but not the patching guide [14:45] sergiusens: just trying to understand whether daily / /current worked or not [14:45] jdstrand: cool, quite a nice set of instructions [14:45] lool: that' what I meant with unstable... I'll install that [14:45] ogra_: jj added it later [14:45] thanks [14:45] yep [14:45] feel a bit better now :-) [14:45] really nice [14:46] ok good :) [14:46] hi :) [14:46] sergiusens: will also switch to daily-proposed now, was just wondering why preinstalled clicks were gone [14:47] sergiusens: cause I was also expecting click installation to work with that image, and that seemed like a possible explanations [14:47] any nexus s users around? :) [14:47] lool: I tried yesterday's and it seemed to be related to pkit [14:48] dholbach: so I can't get things to work with 28.1 either (/current) [14:49] * ogra_ notes that cjwatson just added policykit to the touch seed ... you might want to install it [14:49] packagekit [14:49] mardy: howdy, just wondering if you have a recipe for creating a symlink in qmake? [14:49] oh, right [14:49] some kit with p [14:49] :P [14:49] but I got things working before that, which is why I'm puzzled [14:49] lool, 29.1 the same - it seems like packagekit needs to be installed [14:49] maybe because I installed packagekit-dbg while debugging [14:50] there was also some speculation about some network issues [14:50] I've installed packagekit too now [14:50] but that doesn't seem to be a problem for me [14:50] dholbach: I can find packages in the appstore and they download for me at least [14:50] dholbach: in every single case [14:50] wow [14:50] no, thats only on the 29 images [14:50] I can't get them to show up [14:50] dholbach: adb shell and ping stuff from there perhaps? [14:50] cyphermox: I'm checking for crashes in my test script, none so far [14:51] lool, I could do an apt-get upgrade and stuff :) [14:51] sergiusens: alright [14:51] dholbach: so after installing packagekit and up-to-date apparmor, I can install it, but it fails to open; grr [14:51] dholbach: no moving to 29.1 [14:51] via daily-proposed [14:52] I still can't see it in the lens :/ [14:52] scope [14:52] lool, anything in /var/crash ? [14:52] too many old bugs still in this 28.1 image (I thought more fixes had made it there) [14:52] * ogra_ just notices some upstart applauncher stuff in there [14:53] dholbach: so after unlocking, you pull to see Apps on the right, then you tap SEARCH, then you type "hello" and you don't see the ubuntu logo hello world app proposed? [14:53] lool: what is 'it'? [14:54] jdstrand: in the 28.1 image [14:54] lool: no, you said 'but it fails to open' [14:54] lool: what is 'it' in that context? [14:54] lool, nope [14:54] jdstrand: the app that I just install [14:54] lool: right, which app :) [14:55] jdstrand: this was with /current (28.1) from daily (read-only); then added easyprof update + ran hooks + installed packagekit, it did install but got a manifest error [14:55] jdstrand: hello I thnk [14:55] xda wouldn't show up anymore for some reason [14:55] only showed once [14:55] now flashing daily-proposed 29.1 [14:55] lool: what version of apparmor-easyprof-ubuntu do you have? [14:56] jdstrand: I upgraded to 1.0.21 [14:56] ok. well I tried hello-world just now on grouper on non-ro, and it worked [14:57] jdstrand: yeah, probably the changes I did weren't enough; 28.1 seems too old even with these two changes [14:57] jdstrand, is "non-ro" what I get when I use --pending? [14:57] let's see how 29.1 fares [14:57] dholbach: yes [14:57] I am trying the other apps now-- there are a few apparmor dbus denials but the apps still open [14:57] jdstrand: lool I'm installing daily-proposed without backup right now, first thing after network setup is installing that app [14:57] sergiusens: same here [14:57] I even used -b --wipe to be sure [14:57] sergiusens: but you'll beat me as usual [14:57] dholbach: I think so-- I don't think we are on ro yet, but lool would know better [14:58] lool, sergiusens: be nice guys - no beating please! [14:58] jdstrand: we're switching like tomorrow; well people can still use the cdimage images afterwards [14:58] but we want to recommend the new ro ones [14:58] dholbach: :-) [15:01] right, I'm just using the default atm [15:02] jdstrand: would be good to try the read-only ones since we're about to switch the default :-) [15:02] lool: I've been meaning to. what is the phablet-flash command? [15:04] sergiusens: cyphermox: 3g data connection should in theory be enabled by default right after boot, right? [15:04] yeah [15:04] rsalveti: yes [15:04] there's no way to disable/enable it anymore via indicator [15:04] with very little delay [15:04] rsalveti: just not the default gw if you are on wifi too [15:04] rsalveti: sure there is [15:05] not with 29.1 [15:05] I can only enable/disable wifi [15:05] jdstrand: ubuntu-system as first arg [15:05] rsalveti: confirmed, not there [15:05] the mobile switch was dropped? I thought it was just at the end of the menu? [15:05] not there at all [15:05] that's bad... we should definitely keep it [15:06] installing ubuntu in virtualbox [15:06] how much ram will the build need [15:07] does anyone know if a LG-P500h [15:07] can run ubuntu phone? [15:07] sergiusens: can confirm, no preinstalled apps in 29.1 either [15:07] sergiusens: could download xda app, but couldn't install it [15:07] now will try adding packagekit [15:08] lool: seems that dash for apps only works after a reboot after enabling networking [15:08] cyphermox: who remove it? [15:08] tedg: ? [15:08] no idea :P [15:08] dholbach: ^^ [15:08] sergiusens: it worked without reboot for me [15:08] sergiusens: at lesat I could search for xda [15:08] but it took a couple of seconds to appear [15:08] if it was removed it's because someone asked for it to be removed, or it was part of the design for it not to be there [15:09] maybe dash is now network bound xD [15:09] rsalveti, It's never been in the design. It was only there until the cellular settings landed. [15:09] sergiusens, lool: I still can't see the app - it's not auto-shown and not through searching - I'm happy to reflash or do whatever else [15:09] seems far to go to turn mobile data off [15:09] (also on N7) [15:09] but whatever ;) [15:09] cyphermox, Seems like an odd thing to do :-) [15:09] tedg: not really [15:09] just a little bit tied up in hosting sessions, taking notes, writing summaries of sessions of yesterday and other bits [15:09] lool: dholbach jdstrand this is why it fails: phablet@ubuntu-phablet:/$ pkcon install-local /usr/share/preinstalled/click/com.ubuntu.dropping-letters_0.1.2.2_all.click [15:10] Failed to contact PackageKit: Error calling StartServiceByName for org.freedesktop.PackageKit: GDBus.Error:org.freedesktop.DBus.Error.Spawn.ExecFailed: Cannot launch daemon, file not found or permissions invalid [15:10] I'm past my data limit for this month... I don't want extra charges :) [15:10] http://pastebin.ubuntu.com/6040614/ [15:10] oops, meant to paste the bin :-/ [15:10] aha! [15:10] cyphermox, We've talked to your carrier and they think you should pay for more data ;-) [15:10] sergiusens, do you think you can file a bug about it? [15:10] tedg: yeah I know ;) the problem is I don't have a spare 300$ on my [15:10] *me [15:10] sergiusens: dist-upgrade should fix this one, tyring that [15:10] 750mb data plan here [15:11] xD [15:11] dholbach: I think that's the one Colin fixed earlier today [15:11] lool: no apt-get :-P [15:11] tedg: how's your french btw? :) [15:11] sergiusens: is packagekit-plugin-click installed? [15:11] lool, aha [15:11] sergiusens: of course there is ! [15:11] cyphermox: maguro connected right after booting, after a clean flash + your packages [15:11] * ogra_ agrees that not being able to turn off 3G easily is pretty awful and will get very expensive for many people [15:11] let me now try with wifi [15:11] cyphermox, "fucking carriers" <- good I think ;-) [15:11] jdstrand: yes [15:11] yeah, I know it's not part of design, but we should keep that until we can enable/disable in the system settings [15:11] jdstrand: but seems packagekit itself is missing [15:11] on the nexus s on wich i try to run ubuntu goodness i have unlimited data plan [15:12] tedg, who made that decision ? [15:12] sergiusens: ah, that might do it. there might be a bug on that somewhere (perhaps ask cjwatson (context: http://pastebin.ubuntu.com/6040614/)) [15:12] ogra_, Well mpt writes the specs, but then they get reviewed. If you want to object, you'd probably start with him. [15:14] tedg, the super cheapo prepaid card i use for testing offers an 100MB limited flatrate for 10 euro .... then each further MB costs serious money .... if i would just use it as i use my main phone with real flatrate i would end up with a 500 euro bill [15:14] cyphermox: didn't connect to 3g automatically after rebooting with a valid wifi ap connection [15:14] let me get the logs [15:15] yeah, and that was the main reason we had a way to disable/enable that [15:15] even if not part of the design [15:15] ogra_, seems like you don't need a "quick mobile data off switch" you need a "permanent mobile data off switch" ;-) [15:16] The question isn't whether you should be able to turn it off, it's how often and how quickly. [15:16] tedg, i need a switch that allows me to enable it fast on demand, i i can stay within the 100MB limit [15:16] dholbach, sergiusens: with dist-upgrading, I can get to the "No manifest found" when opening the app now [15:16] ok [15:16] alecu, ralsina ^ [15:16] I'll try [15:16] lool: ok, that's alecu [15:17] lool: argh? [15:17] lool: that is a broken app (again) [15:17] ralsina: that's hello world [15:17] lool: at least hello world did that because it was broken 2 days ago [15:18] ralsina: but that's today [15:18] should be 0.6 [15:18] jdstrand: install packagekit [15:18] jdstrand: I already fixed it in ubuntu-touch-meta, next image will have it [15:18] sergiusens: ^- [15:18] cjwatson: yup, ty [15:19] cool [15:19] shouldn't the app scope depend on packagekit? [15:19] cyphermox: sergiusens: awe: http://paste.ubuntu.com/6040628/ [15:20] with latest cyphermox packages, and with a valid wifi connection already in place [15:20] rebooted, and now 3g data connection is not enabled [15:20] jdstrand, lool, sergiusens: --pending + dist-upgrade + installing packagekit + rebooting: it still doesn't show click apps and I can't search for them (on a N7) [15:20] works fine if I don't have a wifi connection already in place [15:20] Aug 29 15:13:34 ubuntu-phablet NetworkManager[770]: [1377789214.953753] [nm-settings-connection.c:1523] nm_settings_connection_read_and_fill_timestamp(): failed to read connection timestamp for 'd64bd694-4617-7e59-8927-fe9a15b99963': (3) Key file does not have key 'd64bd694-4617-7e59-8927-fe9a15b99963' [15:20] rsalveti: hmmm, let me destory my wifi settings [15:20] Aug 29 15:13:34 ubuntu-phablet NetworkManager[770]: (/ril_0) failed to disconnect modem: Operation already in progress [15:21] dholbach: I saw something like that the other day. I had to play with which scopes were enabled [15:21] aha [15:21] under Dash Plugins I think [15:21] Aug 29 15:13:35 ubuntu-phablet NetworkManager[770]: OFONO connection failed: (32) Operation already in progress [15:21] Aug 29 15:13:35 ubuntu-phablet NetworkManager[770]: [1377789215.65808] [nm-modem-ofono.c:515] translate_mm_error(): in translate_mm_error [15:21] Aug 29 15:13:35 ubuntu-phablet NetworkManager[770]: [1377789215.65839] [nm-modem-ofono.c:541] translate_mm_error(): unmapped dbus error detected: 'org.ofono.Error.InProgress' [15:21] Aug 29 15:13:35 ubuntu-phablet NetworkManager[770]: (/ril_0): device state change: prepare -> failed (reason 'unknown') [40 120 1] [15:21] cyphermox: ^ [15:22] dholbach: in my case, I disabled one of them and rebooted. at the time reenabling it means clicking 'Enable' and rebooting [15:22] rsalveti: that makes no sense, there is no link between wifi and 3G [15:22] ralsina, alecu: and I get it with xda app too [15:22] lool: xda has the same manifest issue [15:22] aha [15:23] beat me [15:24] cyphermox: I think it's just a race somehow [15:24] lool, cjwatson: after installing packagekit, the installation succeeds, but the click directory is empty (/opt/click.ubuntu.com/com.ubuntu.developer.mhall*/0.1.5) [15:24] sergiusens, dholbach: BTW I think the network issue is if you open the search box while wifi is not connected, then it doesn't work [15:24] I have to close it, then go back to it, and then it works [15:24] cyphermox: because nm is already busy setting things up for wifi [15:24] rsalveti: the only way this can be a race is if ofono is going something [15:24] rsalveti: this is not an error from NM but from ofono [15:24] it's saying it's already inprogress [15:24] right [15:25] cyphermox, did you see popey's bug this morning? [15:25] but why is it inprogress? [15:25] yes [15:25] looks like things got really fubar'd [15:25] rsalveti: how would I know? [15:25] rsalveti: this is where we need ofono debug logs [15:25] in that default routes were setup for both 3g & wifi at the same time [15:25] awe: see http://paste.ubuntu.com/6040628/ [15:25] lool, sergiusens: network worked for me [15:25] cyphermox: it's already in there [15:25] both nm and ofono are with debug [15:25] is there a command to reload the shell on the phone? [15:25] rsalveti, yea I saw... but it's kinda hard to debug a syslog w/out context! ;)- [15:26] awe: latest cyphermox packages, clean 29.1, set up a valid wifi connect, reboot [15:26] with sim card that can do data connection [15:26] rsalveti, what does 'nmcli c' and ofono 'list-contexts' show? [15:26] it fails to setup the data connection because ofono gives the inprogress error [15:27] nevermind I got it [15:27] awe: http://paste.ubuntu.com/6040660/ [15:27] awe: but I can connect just fine [15:27] jdstrand, that was it! [15:27] jdstrand, how can we auto-enable the click dash plugin? [15:27] gah, this is ridiculous [15:28] dholbach: oh, nice! :) [15:28] rsalveti, that's weird... it's showing the same gsm connection twice [15:28] ok, now I get "No manifest found for app_id: com.ubuntu.developer.mhall119.xda-developers-app"? [15:28] once with connected == "never" [15:28] dholbach: that is a good question. in my case, I disabled it when I was playing around with it [15:28] and once with a time-stamp [15:28] dholbach: now you are in the same place everyone else is [15:28] dholbach: I just phablet-flashed this morning (without --pending) and then dist-upgraded and it was enabled [15:28] ok... where does the bug come from? [15:29] do we have a report for it? [15:29] dholbach: lool is discussing with ralsina [15:29] perfect [15:29] thanks [15:29] dholbach, lool: alecu is looking into it [15:30] dholbach: can you please try to do "ls /opt/click.ubuntu.com/com.ubuntu.developer.mhall*/0.1.5" [15:31] dholbach: do you have any files there? [15:31] dholbach: afaict the click installer is failing (though it reports to the scope that it succeeded) [15:31] alecu, it's empty [15:32] alecu: I see both packages there personally [15:32] but I get this [15:32] root@ubuntu-phablet:/# click list --all | grep mhall [15:32] com.ubuntu.developer.mhall119.xda-developers-app 0.1.5 [15:32] but the dirs are empty [15:33] awe: yeah, cyphermox, any idea why that? [15:33] http://paste.ubuntu.com/6040660/ [15:33] rsalveti: yes [15:33] fixing now [15:33] lool, there's a .click in three [15:33] there [15:34] cyphermox: also, see the logs, there's a ril_request_setup_data_call after removing the connection [15:34] which makes no sense [15:34] that's why it's in-progress [15:34] no, it makes sense [15:34] why? [15:34] enable has ofono re-read the connections, and when one is found that was already there, it gets removed and re-added [15:34] that's logic I kept from ifupdown [15:35] right, but would that request a data call? [15:35] but in this case, it's wrong because now this parsing happens more than just on startup [15:35] rsalveti: it's not [15:35] the data call is a previous request [15:35] Aug 29 15:13:34 ubuntu-phablet NetworkManager[770]: Activation (/ril_0) starting connection '/724031151801422/context1' [15:35] oh, right [15:35] then it removes and readd the file [15:36] and tries to connect again [15:36] so it should work once you fix this remove/add logic [15:37] yes [15:37] basically, you got a connection already, so ofono tries to activate it when the modem goes online [15:37] but when the modem goes online, it also tries to re-read contexts [15:38] the removal makes the connection disappear, which causes NM to kill off the connection prepare step [15:39] yeah [15:40] mterry, Hi, sorry I just realised I missed our meeting === dandrader is now known as dandrader|lunch [15:41] katie, no worries. We talked yesterday... [15:41] mterry, :) [15:41] katie, I didn't have more questions. Except to note that the edge demo bits landed this morning [15:41] mterry, yes, I saw [15:41] looks good! [15:42] katie, jounih reminded me that I didn't do the "label moves with swipe", so I put that on my TODO [15:42] mterry, great [15:43] katie, ooh, I see videos of the demo on my G+ now :) [15:45] cyphermox, it's at the top level of the Cellular screen, so it couldn't be much more prominent. And keeping track of your monthly usage is why I designed the "Data usage" screen. :-) [15:46] cyphermox, actually it will get slightly more prominent when I move the carrier selection from the top of that screen to the bottom. [15:48] mpt: hey, I'm not the designer, you know best ;) [15:48] boiko: has this landed in upstream? [15:48] boiko: https://codereview.qt-project.org/#change,62404 [15:50] cyphermox, not so much ... engineers like seb128 and Laney and charles are finding mistakes in my System Settings designs at the rate of about two per day. So please ask if something seems weird. [15:55] mpt: not weird, I'd just personally make it more prominent by being directly in the network indicator [15:55] but it's all a matter of opinion [15:55] rsalveti: trying to build with some small changes [15:57] cyphermox: ok, let me know once you got the packages in hands [15:57] daker, hi, do you have any success with gstreamer and metadata? [15:58] codinho: yes [15:59] good [15:59] cyphermox, space is precious in the indicator menus since I'd rather they seldom scrolled. So something you turn off once a month, when you near your quota, probably doesn't qualify. :-) But it could be neat to have settings inside "Data usage" to turn off 4G/3G/data-at-all automatically when you're just about to go over quota. [15:59] codinho: do you need help? [16:00] mpt: +1 [16:00] daker, no [16:01] mpt, btw, what do we do if we run out of panel space ? seems the number of indicators grows every day :) [16:01] codinho: ah sorry, i remember you!! [16:01] cjwatson, alecu: It looks like clicks are installed, just not for the current user [16:01] ogra_, hey, if it was up to me, we'd have only one. [16:01] daker, :) [16:01] haha [16:01] * lool tries something [16:01] mpt: which one ? :) [16:01] codinho: the thing is that i don't want to use gstreamer via c++ [16:02] ogra_, but seriously, designing overflow is on my todo list. [16:03] the maguro screen has actually reached the limit [16:03] daker, why? [16:03] mako could still fit one or two in [16:03] wonder if we also want the indicator for something that's not enabled [16:03] like bluetooth [16:03] codinho: the more it's QML the more it's easy... [16:04] what happens if we disable bluetooth, do we still want it as an indicator? [16:04] codinho: and the app is a compiled app [16:04] rsalveti, well, i guess the indicator would be the place where you enable it [16:04] rsalveti: should be avoided, I'd say [16:04] cyphermox, pollen count. We'd include a pollen counter in every handset. [16:04] ogra_: there is the bluetooth settings [16:04] mpt: !!! [16:04] lool: how can you tell the click is installed? The main directory for each click is empty [16:04] else you need to endlessly click through settings [16:04] but then we need all the indicators [16:04] daker, actually qt is the very complicated thing [16:04] sergiusens: fyi, bugs 1218392 1218394 1218397 1218398 1218399 [16:04] bug 1218392 in touch-preview-images "missing policy group for music app" [Undecided,New] https://launchpad.net/bugs/1218392 [16:04] * cyphermox registers indicator-pollen-count [16:04] cyphermox: +1 [16:05] daker, I wandering how people do love this :) [16:05] rsalveti, "When Bluetooth is off, to save space the Bluetooth indicator should not be present at all." [16:05] great [16:05] * ogra_ actually likes the android shortcuts ... i wouldnt want to have to go through system settings all the time [16:05] codinho: and i need to submit the app to the store which doesn't support compiled apps for 13.10 [16:06] ogra_: we can have one indicator for the system settings :P [16:06] alecu: it's under a .click subdir [16:06] would be nice to have some similar concept in ubuntu touch [16:06] rsalveti, yeah, i wouldnt mind that [16:06] cjwatson: what would explain that clicks get deployed under /opt/click.u.c but no .desktop file is generated? [16:06] indicator-toggle [16:06] :) [16:06] :-) [16:06] hi all [16:07] cjwatson: in fact, there's no phablet user subdir under /opt/click.u.c/.click [16:07] jdstrand: saw them in email [16:07] jdstrand: thanks [16:07] lool: that would be the register step being broken. can't look now, later [16:07] lool: I can't see the actual click content, only metadata: http://pastebin.ubuntu.com/6040814/ [16:08] sergiusens: that was a useful exercise. ran all of those on touch and desktop and was able to snag some more policy refinements too [16:08] lool: if the clicks were installed, I would expect the .qml files at least [16:08] alecu: right [16:09] daker, I still wandering about how could I apply my port to the qt repo, I need some help from the qt guys but they looks like not interested .. [16:10] codinho: you mean the ubuntu qt guys ? or the real qt guys ? [16:10] the real ones [16:11] sergiusens: also did you see preinstalled clicks are missing too? [16:11] codinho: :( [16:11] lool: checking [16:11] awe: here I'm getting some delays bringing up data, but it would be in ofono's hands [16:11] waiting in the prepare step to get an IP address back from ofono [16:12] excessive delays? Is your code timing out? [16:12] codinho: maybe some badbuzz will make them interested... [16:12] sergiusens's script is timing out [16:12] awe: actually, scratch that [16:12] something's not right, the context is active but NM doesn't notice it for some reason [16:13] lool: adb shell rm /home/phablet/.ubuntu-build && adb reboot [16:14] cyphermox, ok. I'm pretty sure it's just a property update ( 'Settings' ) when the context becomes active [16:14] lool: I have all 3 after that [16:14] let me know if I can help [16:14] lool: at least with 'click list' [16:15] codinho: anyway, just keep up the good work ;) [16:15] daker, yeah, you too :) [16:16] does ubuntu touch run on the new nexus 7 :) [16:17] awe: right, it's just a property update [16:17] awe: I'm trying to remember how we notice that the context is ready [16:21] lool: desktop files are already there [16:26] sergiusens: I don't get these! [16:26] sergiusens: click list empty for me [16:26] sergiusens: oh sorry, after rm /home/phablet/.ubuntu-build && adb reboot [16:27] lool: yes, after, since the version didn't change and the stamp was created it won't run again [16:34] tedg, has a fix for indicator-network/indicator-secret-agent been uploaded already ? [16:35] tedg: hey [16:35] ogra_, No, needs a review. [16:35] when will wifi/3g be fixed? [16:35] charles, https://code.launchpad.net/~ted/indicator-network/disable-secret-agent/+merge/182914 [16:35] what is the phablet-flash target for ro? [16:35] we want to roll an image :) [16:35] that is good [16:35] tedg: where is the fix? do you have a link? [16:36] asac, It's the merge above ^ [16:36] asac, its just dropping the upstart session job [16:36] Yup [16:36] awe: I can't explain it, it's as if I never get the dbus response from ofono [16:36] jdstrand: phablet-flash ubuntu-image [16:36] jdstrand: phablet-flash ubuntu-system [16:36] sorry [16:36] tedg: is indicator using daily-release? [16:36] tedg, ack [16:36] tedg, why the sleep ? [16:36] sergiusens: thanks [16:36] awe: if it was timing out, there would be an error [16:36] ogra_, asac, keeping it simple :-) [16:36] awe: this is really weird [16:37] breakfast command not found o_O [16:37] ogra_, I wasn't sure what would happen if I put true there. [16:37] ogra_, I figured that was safer. [16:37] tedg: through which stack is that going? [16:37] i would just have rm'ed the file :) [16:37] asac, indicator stack [16:37] tedg: http://reports.qa.ubuntu.com/daily/ [16:37] that one currently fails somewhat [16:37] ogra_, Heh, we'll hopefully need it working RSN :-) [16:37] oh, its in phablet-flash -h :) [16:37] heh, ok [16:37] tedg: do you know what broke it? [16:37] :-( [16:37] No [16:37] tedg: can we backout so we can land this? [16:38] * tedg looks [16:38] tedg: i found it is talking about https://launchpad.net/~ubuntu-unity/+archive/daily-build/+build/4917407 [16:38] causing issues [16:38] so qmenumodel blocks the indicators stack [16:38] Yeah, looking at that... no build log? [16:39] sil2100: Mirv: kenvandine: https://jenkins.qa.ubuntu.com/job/cu2d-indicators-head-2.1build/441/console [16:39] can you check that one out? The build job failed in an odd manner afaict [16:39] maybe it just needs a retry? [16:39] can someone help me build my kernel [16:39] breakfast doesn´t seem to show [16:40] cyphermox: what ETA for 3G fix? do you have an idea yet what caused it? [16:40] e.g. what upload/change? [16:40] Worked on ARM and i386. https://launchpad.net/~ubuntu-unity/+archive/daily-build/+sourcepub/3453107/+listing-archive-extra [16:40] asac: depends what 3G fix you're talking about [16:40] bfiller: don't think it landed [16:40] cyphermox: thought something is broken right now [16:40] cyphermox: more broken than yesterday [16:40] whatever is more broken :) [16:40] cyphermox: ddoes that mean anything to you? [16:40] breakfast crespo and it doesn´t find breakfast command [16:41] o.O [16:41] sergiusens: right, click list reports them after this reboot, thanks [16:41] asac: no [16:41] sergiusens: what stopped them from installing in the first place? [16:41] it could be the iproute change from a few days ago in distro [16:41] asac: I'm working on another bug right now, should fix it at the same time [16:41] sergiusens: I tried reinstalling xda package afterwards, but didn't help [16:42] cyphermox, that might explain popey's bug? ( iproute2 )? [16:42] cyphermox: ok. so not a regression from /current? [16:42] awe: possibly. it's the only thing I can thing of [16:42] lool: you needed to install packagekit [16:42] I mean, NM hasn't changed in quite a few days, so if something is more broken, it's not because of NM [16:42] asac, we need to talk bug numbers... too many problems being discussed [16:42] lool: aka broken system [16:43] lool: xda plugin is broken due to scope/store issues [16:43] asac, yeah, it should get picked up in the next tick, unless we need it sooner [16:43] sergiusens: ah right [16:44] NYL, ". build/envsetup.sh" [16:45] Waiting for authentication [=========================] [16:45] Fatal error: Failed to obtain authentication. [16:45] that's with pkcon [16:45] I think there was a conffile change allowing me to bypass onconsole= [16:45] (which I didn't do, but was there) [16:46] charles, thanks! [16:47] Waiting for package manager lock[=========================] [16:47] kenvandine: when is that? [16:47] Fatal error: W: Not using locking for read only lock file /var/lib/dpkg/lock [16:47] we want it in next image for sure :) [16:47] that's probably the problem now [16:47] pete-woods, I rejected your merge, if we leave it in the package then we can test on the image and such as we get closer to landing. I don't want to stop building it. [16:47] kenvandine: i feel we want it sooner :-P, but give me the time of the tick and also if this job is run early or late during the tick :) [16:48] tedg: we still build and test the agent in that branch [16:49] we just don't install the upstart config file [16:49] Ah, I see. Sorry. [16:49] asac, i think in less than an hour, and indicators are probably early [16:49] cyphermox, do you know? [16:49] pete-woods, Read too quickly :-) [16:49] :) [16:49] * asac makes the math [16:50] kenvandine: too tight [16:50] kenvandine: in 2h it needs to be in to be picked up [16:50] om26er: ping [16:50] bfiller, pong [16:50] kenvandine: unless a manual ttick is super hard, i believe it would be better to give indicators a nudge [16:50] the next start is in an hour and fifteen minutes [16:50] once teds fix is in [16:50] ah, ten I mean [16:51] tedg: did you test if that fixes it? not that we forget somethign :) [16:51] asac, so a manual one now wouldn't be a good idea, since the next tick is about to start [16:51] asac, I didn't but rsalveti said it worked for him, and I trust him. [16:51] :-) [16:52] indeed, not a good idea, indicators take a while to test too [16:52] sergiusens: is there no 1.0+13.10.20130828-0ubuntu1 package yet? [16:53] tedg: i didnt hear it from rsalveti [16:53] tedg: soo... i take your word and will call you later :) [16:53] asac, if you scroll up a little ... [16:53] :) [16:53] asac, heh, sure. [16:53] vie phone [16:53] via [16:53] :) [16:53] if it works [16:53] :-P [16:54] hope i get ubuntu touch to run on my nexus s [16:54] :D [16:54] asac: sorry, what? [16:55] rsalveti, the 007^Wsecret-agent issue [16:55] need helping testing tedg's mr? [16:55] *needs help [16:55] rsalveti, asacwants confirmation that disabling it helps [16:55] well, I updated the bug [16:55] (and you tested that above) [16:55] asac: read the bug [16:55] :P [16:55] haha [16:55] there is a bug ? [16:56] bug 1218385 [16:56] bug 1218385 in indicator-network (Ubuntu) "Network indicator doesn't ask for the AP password" [Undecided,New] https://launchpad.net/bugs/1218385 [16:57] rsalveti: can you bless the patch submission that this will fix it :) [16:57] rsalveti: we land a patch hoping it will give us a good image [16:57] asac: the mr was already approved [16:57] but we land it without testing :) === schwuk is now known as schwuk_away [16:57] rsalveti: ah soo you acked it? === crypto is now known as Guest72495 [16:57] nops [16:57] just noticed it was approved [16:58] e.g. i can come to you if this MP does not help :) [16:58] I'd hope someone tested it [16:58] asac, charles did above [16:58] rsalveti: noone did [16:58] thats the point :) [16:58] tedg: can you please test it :-P [16:58] http://pastebin.com/XL4NqJBa [16:58] tedg: for real? like building the source, installing, confirming that stuff works? [16:58] ogra_: he tested the patch to the on-disk file or the source patch? [16:58] charles: ? [16:59] no idea, he approved the MP :) [16:59] right. [16:59] plars: done [16:59] i just hoped for confidence that this issue will be gone with the 2200 build [16:59] sergiusens: thanks! [16:59] but guess thats too much to ask [16:59] :) [16:59] well, who needs wlan if you have 3G working [16:59] lets hope [16:59] haha [17:00] well, I only approve stuff after testing [17:00] barry: Around? [17:00] especially since we keep 3G even when wifi is on [17:00] but it seems this is not a rule [17:00] we dont even get the power savings [17:00] so yeah :) [17:00] barry: did you get my messages about system-image upload and possible bugs? [17:00] rsalveti: thanks! please keep the bar high! [17:00] :) [17:00] lool: yes. lurking on arch-click but about to get some lunch [17:00] tedg: is this indicator stuff because of "hard-coded-password" ? [17:00] barry: ok [17:00] lool: i saw the bug [17:00] rsalveti: i think other belive the patch submitter has tested [17:00] rsalveti: once everybody belives the other is testing, noone tests [17:00] :) [17:01] everybody expecting someone tested [17:01] lol [17:01] barry: if it's not related to your latest changes, would you upload them so that we can use daily-propsoed? [17:01] right [17:01] so yeah. everybody has to test :) [17:01] lool: LP: #1218357 iirc [17:01] Launchpad bug 1218357 in Ubuntu system image "Can't upgrade in daily-proposed" [High,Fix committed] https://launchpad.net/bugs/1218357 [17:01] barry: good lunch! [17:01] there is no way around [17:01] the test fairy [17:01] cyphermox, I'm not sure what that is, so guessing not :-) [17:01] testing around the clock :) [17:02] lool: i'll go ahead and release 1.3 now, which should fix bug ^^ and let you upgrade to daily-proposed. then i'll get some lunch :) [17:02] tedg: I'm having an issue with the indicator for wifi on my device, though admittedly yesterday's image, where the passphrase for wifi is hardcoded, the auth dialog never shows [17:03] barry: right, it's just that I had issues when upgrading with the changes applied, but can't tell whether these are regressions or not [17:03] cjwatson: hey === alan_g is now known as alan_g|beer [17:03] cjwatson: so getting two packagekit issues now (see above) [17:03] cjwatson: First is when running pkcon from adb shell, I get: [17:03] 18:45 < lool> Waiting for authentication [=========================] [17:03] 18:45 < lool> Fatal error: Failed to obtain authentication. [17:03] * ogra_ points cyphermox at bug 1218385 which was discussed the last 50 lines :) [17:04] bug 1218385 in indicator-network (Ubuntu) "Network indicator doesn't ask for the AP password" [Undecided,New] https://launchpad.net/bugs/1218385 [17:04] lool: well, when I said "later" ... I need to use my lunchtime in figuring out foundations notes for the end-of-UDS plenary, apparently [17:04] cjwatson: which is expected; I dont know whether it works or not from click-scope, I guess it should [17:04] cjwatson: eh ok [17:04] ogra_: ah, good then [17:04] lool: there's supposed to be a .pkla/.policy in packagekit-plugin-click for that [17:04] cjwatson: and second thing is 18:47 < lool> Waiting for package manager lock[=========================] [17:04] 18:47 < lool> Fatal error: W: Not using locking for read only lock file /var/lib/dpkg/lock [17:04] lool: it may need tweaking. the real fix is to fix consolekit/logind/whatever already though [17:04] cjwatson: which looks like packagekit apt plugin trying to lock apt database and failing [17:04] err dpkg database [17:04] wifi doesn't work for me in 28.1 (aka current), is that a known issue? I get a segfault after ril_0 info in "nmcli dev list".. it seems different than the bug you're discussing right now [17:05] the click plugin is supposed to filter out click packages so that the aptcc backend doesn't see them, and skip aptcc entirely if there's nothing left other than click packages [17:05] cjwatson: haha [17:05] cjwatson: if I rename it to .click it works [17:06] but because it's named com.ubuntu.developer.mhall119.xda-developers-app-0.1.5.click?noauth=1 it doesn't [17:06] rsalveti: sergiusens: new update for the provisioning fix, should correct the issues you had with double activation [17:06] well sorry, it's not in click list either [17:06] need to debug this :-( [17:06] lool: there's *supposed* to be a [17:06] MIME type declaration in place [17:06] rsalveti: sergiusens: and on that, before the next tick starts, I'm going to go grab lunch [17:06] lool, cjwatson: https://bugs.launchpad.net/ubuntu/+source/click/+bug/1218483 [17:06] Launchpad bug 1218483 in click (Ubuntu) "Installation errors are not reported" [High,New] [17:07] cyphermox: pushed the packages to the same place? [17:07] mhall119: you appear to have built that package with an old version of click [17:07] yeah, need to eat here as well [17:07] rsalveti: yes [17:07] same place [17:07] awesome [17:07] I'm not quite sure how since it *claims* Click-Version: 0.3 [17:08] But it doesn't have the extra _click-binary ar member that makes MIME type detection work reliably [17:08] cyphermox: thanks, will check... but I need to eat too [17:08] Which is why lool is finding that he needs to rename it to end .click [17:09] cjwatson: indeed, I dont need that with the ar.com.beuno.hello-world-0.6.click?noauth=1 [17:09] doesn't show up with user=phablet though [17:09] (in click list) [17:09] I don't know how it's possible to have generated a Click-Version: 0.3 package without that; this was the very first change that went into the 0.3 seris [17:09] *series [17:09] Has somebody foolishly written their own build tool or something that doesn't use click build? [17:09] lool: where are you passing the "user=phablet" [17:09] alecu: I'm not [17:10] lool: ls -lR /opt/click.ubuntu.com please [17:10] cjwatson: this is good in that it has the clicks [17:10] lool: cjwatson what package is it? [17:10] cjwatson: http://paste.ubuntu.com/ [17:10] sergiusens: com.ubuntu.developer.mhall119.xda-developers-app-0.1.5.click [17:10] lool: I think I might need the paste ID too :) [17:10] This works, but it doesn't install the click for the phablet user, just under /opt: sudo pkcon install-local ar.com.beuno.hello-world-0.6.click?noauth=1 [17:11] Please don't run pkcon under sudo [17:11] Ever [17:11] cjwatson: oh wow, compiz bug [17:11] cjwatson: http://paste.ubuntu.com/6041054/ [17:11] or no, can't be [17:11] ubuntu-paste bug [17:11] You run pkcon as your user [17:11] how weird [17:11] hmmm, yeah, pkcon as user [17:11] cjwatson: I sudo-d it to workaround the dbus onconsole check [17:11] That's not a valid workaround [17:12] lool: you need to login to phablet with a login shell [17:12] Also - do you not have an /opt/click.ubuntu.com/.click/ directory? That's not in your paste [17:12] I do sudo -u bash -l because I forget of a clever way to do it [17:12] "su - phablet" from "adb shell" works for this [17:12] cjwatson: I do, ls -lR didn't get it [17:12] lool: can I see that too? [17:12] cjwatson: ah, it didn't last time I tried (DBUS_SESSION_BUS_ADDRESS wasn't envset) [17:13] cjwatson: that's a find: http://paste.ubuntu.com/6041057/ [17:13] /opt/click.ubuntu.com/.click/users/root/com.ubuntu.developer.mhall119.xda-developers-app === alecu is now known as alecu_lunch [17:13] Right, there you go, it's registered it for root because you used sudo [17:13] don't do that :) [17:13] ok, with sudo -u phablet -s, pkcon works without sudo [17:14] will use that instead of su - phablet [17:14] and click list works [17:14] cjwatson: ^ [17:14] Great [17:14] trying com.ubuntu.developer.mhall119.xda-developers-app-0.1.5.click now [17:14] works too [17:15] yes [17:15] alecu_lunch: so hello world app works [17:15] and xda doesn't [17:15] cjwatson: so the remaining issues seem to be solved in apt-get dist-upgrading stuff + renaming xda .click [17:15] hello world works [17:15] dholbach: ^ [17:16] we should work out how that xda package was built [17:16] but that probably needs mhall119 to appear [17:17] If I'd known about the _-prefixed ar members trick to make the MIME type reliably detectable then it would have been in there from the start, but I only found out about it in passing via a thread on debian-devel :) [17:17] So it landed in 0.3 [17:17] cjwatson: thanks for helping debug this [17:18] now need to verify all of this again with a fresh image I guess :-) [17:18] ogra_: when is the next image due? [17:18] ogra_: Would you launch a build now? === dandrader|lunch is now known as dandrader [17:20] * rsalveti needs food === bfiller is now known as bfiller_afk [17:24] http://pastebin.com/RnvjqMBx [17:29] is it good or bad [17:33] cjwatson: what need me? [17:34] mhall119: How did you build com.ubuntu.developer.mhall119.xda-developers-app-0.1.5.click? It is very weird - it claims Click-Version: 0.3, but it doesn't have a particular file format tweak I made as the very first thing in the click 0.3 series [17:35] mhall119: Not to mention that the file name seems to be non-standard-ly formed, but perhaps that's the app store's fault, I don't know [17:36] sil2100: hi [17:37] sil2100: I'm trying to look into the dialer-app autopilot tests [17:37] sil2100: but I'm getting ImportError: No module named ubuntuuitoolkit [17:37] sil2100: that's just trying to list the autopilot tests [17:37] cjwatson: I manually unpacked an older version and made changes recommended by jdstrand, then re-packagd it manually with the new version [17:37] mhall119: manually how? [17:38] dpkg-deb IIRC [17:38] ugh [17:38] please don't do that again, I spent quite a while tracking down what was wrong with this! [17:38] this was all to test jdstrand's recommended changes to the way the SDK did stuff [17:38] sorry about that, it wasn't meant to be used as "the right way" to do things [17:38] it would be appreciated if you could upload a new version built with click build [17:38] because this is causing problems [17:38] plars: ah, I guess we might be missing some dependencies then [17:39] ubuntu-ui-toolkit-autopilot i'm guessing? [17:39] plars: since it probably requires ubuntu-ui-toolkit-autopilot installed [17:39] plars: which was probably missed from the packaging [17:39] plars: we're not running those tests yet on daily-release, so we didn't notice it yet [17:40] sil2100: don't we require the tests before daily release can happen? [17:42] lool, we're waiting for the indicator-network fix so people can actually use wlan [17:42] lool, but a manual build is planned then [17:43] plars: normally, yes, but in case of the phone apps we agreed to release before the tests got prepared [17:44] plars: since there has been a great concern to get those in before FF [17:44] (there's an FFe for that, but still, people wanted those in the archive already) [17:45] already ? [17:45] heh [17:45] (we were supposed to not use PPAs anymore a monyt ago) [17:45] Actually, we should already run them now, since there are some now at least, guess I'll add them ASAP [17:45] they are in use now [17:45] phone-app is gone [17:46] Hello guys ;) [17:46] ogra_: heeey! We were only told about enabling them for daily release mid-last week! [17:46] Oviii: hi [17:46] =) [17:46] sil2100, yeah, we're fast :) [17:46] sil2100, they couldnt go in until after latest indicators, then tests held them up [17:47] Wha, today I wanna test the Ubuntu Touch on my SGSS2 ;) [17:47] But kind of a difficult to install at the moment. [17:47] I guess in the future it will be more user-friendly [17:48] ... or not ;) [17:48] Haha :D [17:48] Someone hadalready tried on a SGS2 ? [17:48] the sgs2 is installable with phablet-flash [17:48] :o [17:48] I had test the install via CWN :/ [17:48] Not a great idea x) [17:48] though it is an old port i think, rather outdated [17:49] phablet-flash on Nux OS ? [17:49] phablet-flash is the ubuntu tool to flash ubuntu touh to phones [17:49] lool, let me test it [17:49] it comes with the phablet-tools package [17:49] Hmmm ok ok [17:50] Any idea torestore android ? In orde to re-install again UbuntuPhone ? [17:50] At the moment, black screen after samsung logo x) [17:50] even in recovery mode ? [17:51] with CM10.1, the recovery not work anymore with up button and power button... [17:51] Maybe another method ? [17:51] heimdall then [17:51] heimdall ? °° [17:51] well, what did you use to root the phone in the first place ? [17:52] oh, it was a very long time [17:52] https://github.com/Benjamin-Dobell/Heimdall [17:52] Maybe a windows tool at the time [17:52] oh great [17:53] with that you should be able to flash a recovery.img ... from there you should be able to then flash zips [17:53] Hmmmm OK OK [17:54] I must retore android before Ub.Ph ? [17:54] no, you just need a working recovery that can flash zips ... at least if you manually flash [17:55] for phablet-flash it helps to have an ubuntu recovery.img ... the wikipage for your port should point you to it [17:55] ogra_: actually, just a cwm one would work [17:55] ah, k [17:55] Understood [17:55] Well, I should restore android before, so è_é [17:57] Waw, I don't like Github ! [17:57] bfiller_afk: pmcgowan I have a contact with two numbers, if I choose the contact from the contacts tab in the dialer I'm taken straight into the call [17:57] SVN is better (design speaking) [18:00] sergiusens, you mean it just uses the first one or something without selecting? [18:00] Ha! My SGS2 is now in download mode haha :D [18:01] pmcgowan: first contact (one marked as other) no chance of selection [18:01] ogra_: https://wiki.ubuntu.com/Touch/Install this link ? ;) [18:01] Oviii, https://wiki.ubuntu.com/Touch/Devices/i9100 [18:02] sergiusens, fix committed 2 hrs ago https://bugs.launchpad.net/ubuntu/+source/address-book-app/+bug/1211714 [18:02] Launchpad bug 1211714 in dialer-app (Ubuntu) "[Address book] [Dialer] Contact details can't expand in Dialer" [Undecided,Fix committed] [18:02] and yeah, the manual install methid from the Install wikipage [18:02] sergiusens, do your contacts with no last name show as undefined [18:02] pmcgowan: great, thanks [18:03] pmcgowan: let me check... for osd or on the list? [18:03] sergiusens, on the list [18:03] pmcgowan: one sec, I just smashed the shell [18:03] k [18:04] sil2100: I guess s/contacts/address-book right? [18:05] plars: right, I probably just re-pasted what asac said in the e-mail [18:05] lool, hello world works for you? [18:06] pmcgowan: confirmed [18:06] xda + hello world both have the same issue for me [18:06] even after dist-upgrade + reboot [18:06] sil2100: the ulimate know how would be with bfiller_afk [18:06] :) [18:06] sil2100: so where can i see what autopilots are run for which stack? [18:07] is there way to find that easily from the dashboard? [18:07] sil2100: e.g. from here: http://reports.qa.ubuntu.com/daily/ [18:07] tedg: did your stuff land now (tested)? [18:07] ogra_: i think if its not in, we have to turn off autoscheduling and wait wit the image run until its in. [18:07] asac: not sure, since hm, we're not really using the dashboard, as it currently doesn't offer enough information on it [18:07] :) [18:07] ogra_: autoscheduling of the image build that is :) [18:08] * ogra_ sighs, thats the third all nighter for manual image crap i do this week [18:08] * ogra_ would really like to get to bed before 3am one day [18:13] ogra_: yeah. dont do that [18:13] ogra_: tell me when you want to be off and i will ensure you dont get harrassed anymore :) [18:13] asac, well, if i catherd the image i have to [18:13] ogra_: we need to fix that otheres can do that [18:13] i eally just want a button :) [18:14] to kick off a new run [18:14] asac, rsalveti and sergiusens can [18:14] well, they also are busy and overworked :) [18:15] but yeah [18:15] we could say they take over at 2000 :) [18:15] sergiusens: rsalveti: hi! :) [18:15] i'll disable the scheduler and watch -changes until the package is there [18:15] they just have to watch what comes out at the rear end [18:15] asac: hey, what do you need? [18:16] rsalveti: we want to pick up the indicator fix [18:16] rsalveti, an off schedule image build ... [18:16] that seems to be delayed slightly ... or we dont have enough intelligence right now :) [18:16] asac: just spin another image once that's in? [18:16] right, that's fine [18:16] I want to test that myself anyway [18:16] rsalveti: super easy [18:16] rsalveti, i'll be around for another few hours and take care of cron and all but it is likely that it takes longer apparently [18:16] rsalveti: well, i actually wanted to not build a new image in 1.5 hours, but rather kick it then off once the imag lands [18:16] rsalveti: of course only if we know that anything is happening at all :) [18:17] right [18:17] rsalveti, i'll ping you to take over the watch once i go afk [18:17] sounds good [18:17] tedg: :) whats the status? [18:17] where is that one liner ? :) [18:17] asac, i think you should rather hunt the CI people, ted is done :) [18:17] ogra_: no. his merge or whatever failed build tests :) [18:17] so he wanted to test it and fix it and resubmit afaict [18:18] at least thats what i would do :-P [18:18] hah, ok [18:18] cyphermox: kenvandine: do you have any insight about where our indicator fix is? :) [18:18] rsalveti: sergiusens: is my updated package better? [18:18] asac: do you know what commit? [18:18] no thats the point :) [18:18] i know the initial commit failed [18:19] i wondered if you know there is another in flight already [18:19] cyphermox: I had a bad hang on the first run, the second went ok, in sessions again now to give it a better go [18:19] ok [18:19] cyphermox: sorry, didn't test yet, in a meeting now will test once it's over [18:19] asac: indicators is running right now... [18:19] is there an equivalent of --pending for ubuntu-system? --help isn't saying there is [18:20] rsalveti: do you know what this indicator fix thing is? ^^ [18:20] cyphermox: that first bad state left my status in connecting(pending) and restarting network-manager just proke [18:20] is it still for indicator-netowrk? [18:20] jdstrand: yes --channel daily-proposed [18:20] rsalveti: lool very simple https://code.launchpad.net/~sergiusens/phablet-tools/options_fix/+merge/182979 [18:20] Hi, is there a possibility my device will work with Ubuntu Touch even if it is not on the list or am I screwed. [18:20] http://system-image.ubuntu.com/channels.json [18:21] sergiusens: thanks! :) [18:21] np [18:22] cyphermox: nice. at which stage? [18:22] where is it? :) [18:22] asac: building indicator-network and qmenumodel [18:22] cyphermox: https://code.launchpad.net/~ted/indicator-network/disable-secret-agent/+merge/182914 [18:23] cyphermox: whats the average time from this point until its proper in the archive? [18:23] * asac still needs to calibrate his sense of time a bit [18:23] I don't know... an hour maybe? [18:24] so if we are super lucky ... it might make it :) [18:24] ogra_: ^ [18:24] ogra_: you should go off duty anyway now (remember) [18:24] :) [18:24] asac, UDS goes for another hour :P [18:25] ogra_: right. thats uds-duty, not off- agreed :) [18:25] but just do that [18:25] yeah, i dont mind working long ... just not on duty :) [18:25] from a certain hour on [18:34] argh, system-image in ubuntu-system doesn't allow overriding the channel === alecu_lunch is now known as alecu [18:44] jdstrand, did the click apps launch for you in the end? === bfiller_afk is now known as bfiller [18:46] sergiusens: in dialer app when you click on a contact it should expand and show you the phone numbers and allow you to click either one [18:46] dholbach: I don't seem to have the click apps. eg, no dropping letters or sudoku preinstalled [18:46] oh, I meant "hello world" and the xda app [18:46] bfiller: yeah, pmcgowan forwarded me to a bug [18:46] bfiller: all good [18:46] sergiusens: it's working? [18:46] bfiller: I need to update the packages (was fixed 2 hours ago I think) [18:47] cool [18:47] I'm going to say 'no'. if I try to use 'Search', the key press goes through the keyboard and launches the icon underneath [18:47] which launches the app I 'tapped' and makes the search go away [18:48] ok, I managed to click 'Install' in hello world [18:48] it is hung though-- but this image is days old, so it doesn't have packagekit [18:51] jdstrand: talk to barry about switching channels, not sure it's possible still [18:51] and would be the same effect as flashing full === dandrader is now known as dandrader|afk [18:56] so how do i add files into the ubuntu-root folder for packaging in the initial ramdisk? [18:57] asac: https://launchpad.net/ubuntu/+source/indicator-network/0.5.0+13.10.20130829-0ubuntu1 [18:57] yeah, so just another hour or so [18:58] looks like it can make it in time for the image build [19:03] jdstrand, sergiusens once system-image 1.3 lands (or you install it) the following will force a channel change: system-image-cli -b 0 -c --filter=full [19:03] barry: well, if I install, won't that break it? [19:03] ie, I just change the image from what it is expecting [19:04] jdstrand: naw, because "what it's expecting" is just a build number, really [19:04] and installing the system-image 1.3 debs won't change that [19:04] dholbach: after dist-upgrade, hello-world worked for me, but then I had fiddled other stuff, so I ought to retest with latest image spin [19:05] I'll do that in a bit too [19:05] barry: oh, I thought that if I installed stuff it would blow things up (which is why it is readonly [19:05] ) [19:05] jdstrand: well, it won't blow up system updates. i make no claims about anything else ;) [19:08] fginther: here :) [19:08] ack [19:08] fginther: the location does not exist at all? [19:08] m-b-o, sorry, the directory is there, but there is no ini file for the weather app === dandrader|afk is now known as dandrader [19:09] m-b-o, when that happens, the test is supposed to launch the app to create a fresh db? [19:10] fginther: it will always try to delete the db-file beforehand. It can exist, but it doesn't have to [19:11] fginther: I've copied that portion of the test from the notes-app. [19:12] m-b-o, ack. I'm focusing on that are now [19:12] s/are/area/ [19:16] fginther: the db-file doesn't get created [19:16] m-b-o, it just worked on my box [19:25] nice... i think indicator made it :) [19:28] yeah should have [19:28] yup [19:28] there's still a few other things not yet done [19:28] cyphermox: what? [19:28] :) [19:28] unity8, friends, apps, media [19:29] will they miss image? not sure if i prefer them to miss :) [19:29] what do they fix? [19:29] cyphermox: unity etc. are still in daily-release? or are they in propposed etc.? [19:30] no, still running tests [19:30] good [19:30] so they will miss :) [19:30] or you know, wiating to build [19:30] cyphermox: wonder if the build of image starts in 30 minutes [19:31] might we pick up half of those? or will they be even later? [19:32] stgraber: did something change with switching to developer mode? [19:33] cjohnston: the name of the flag changed to .writable_image though the latest image should have compatibility code to move .developer_mode at boot time [19:33] plars: ^ [19:34] good to know... [19:35] stgraber: is there somewhere (other than following MPs) that we should have gotten notification on this change? [19:35] it will break testing [19:36] cjohnston: I was supposed to e-mail ubuntu-touch about it, but didn't get to it before we landed the backward compatibility change which made the announcement useless (since both .developer_mode and .writable_image work) [19:37] rsalveti: can we kick off a build befoore something else big lands? [19:37] * asac stills for feedback if the sdk fix we landed 10 hours ago actually has any effects [19:37] asac: I'm fine doing a build now [19:37] ogra_: ^^ [19:38] rsalveti: i know its just 20 minutes before regular tick, but... [19:38] there is big stuff in flight [19:38] and would like to get a checkpoint without a new unity8 etc. :) [19:38] if cheap and easy of coruse [19:38] otherwise lets wait and hope :) [19:38] that will automatically queue up the cronned build [19:38] imo we could put the whole thing to alway manual [19:38] we want to build images more frequently i feel :) [19:39] if you do a manual build now and dont want two images you should suppress the cron build [19:39] right [19:39] thtas what i want [19:39] just never cron build [19:39] always do it manual for a while :) [19:39] hehe [19:39] no, let it run automatically too ... [19:39] rsalveti: so yeah, lets do it if you dont feel its wrong :) [19:39] or have it in cron doing one build per hour [19:39] lol [19:39] ogra_: sure. i dont have problems with multiple iumages [19:39] rsalveti: i think we were told we have to be sensible to not disturb other stuff running on that machine [19:39] if we can do something that check that something changed in the archive [19:40] * ogra_ can happily do it, is indicator-network promotoed already [19:40] so manual gives us a chance to skip 1 or two horus [19:40] rsalveti, yeah, that was my plan [19:40] ogra_: i think its in [19:40] something that watches the seeded packages [19:40] ogra_: and big landings are somewhere in the daily-release... i would prefer to not pick them up in this run :) [19:40] cool, then lets do one now [19:40] so rather hurry than wait [19:40] ogra_: yeah [19:40] and if a certain percentage changed, run a build [19:40] yeah, on demand would be best [19:40] i actually think ... we hsould trigger a tick based on stacks [19:41] thats whayt i said above :) [19:41] like heads.apps pushes to proposed ... tracks until everything is in and then kicks off an image [19:41] well, based on seeds [19:41] but you surely could break that up into smaller stacks [19:41] ogra_: stacks are different. we alreayd have a good, granular beet in daily-release [19:41] we can just leverage that [19:41] ogra_: like http://reports.qa.ubuntu.com/daily/ [19:41] whenever one of those buckets spit its stuff inthe archive, trigger an image [19:42] rsalveti, so do you or should i ? [19:42] ogra_: whatever you prefer, don't know how tired you are atm :-) [19:42] I still have a few hours to go [19:42] i'm fine [19:42] we said ogra is off/uds-duty [19:42] ogra_: no :) [19:42] i pinged rsalveti intentionally :) [19:42] right, let me do it then [19:43] ogra_: ^ [19:43] running :P [19:43] lol [19:43] lets make auto kick for everyone mentioning the nick ogra outside of working hours :) [19:43] haha, cool [19:44] haha [19:44] * asac tries to find a bot [19:44] wait.. there is time outside of ogra_'s working hours? [19:45] i'm totally happy to work until night ... i just dont like to be responsible for process things after 11pm since such stuff tends to strech often enough ) [19:45] cjohnston: yeah we had a session today and decided that we wanted to try to experiment with that idea [19:45] ogra_: cron suspended? [19:45] lool, yeah [19:45] lool: read backlog [19:45] :) [19:45] i'll re-enable it after 10 [19:46] * lool checks whether system-image migrated from -proposed [19:46] sloooow [19:47] stgraber: seems to require .writeable_image now [19:47] cjohnston: ^ [19:47] looks like it [19:47] lool: if you want us to include something of your work when manually moving our images forward let me know what [19:47] didnt know you wanted something in [19:47] plars: Yes, and we've readded support for .developer_mode, but this didn't make it into an image [19:47] plars: but the new name is .writable_image [19:47] lool: ah, ok [19:47] cjohnston: we may as well just change it then [19:47] ahh [19:47] plars: already working on a MP [19:49] asac, lool wanted the build earlier actually, we talked already [19:50] asac: it's good [19:50] --- also we're likely resetting image numbers on system-image tonight --- [19:51] which means updates will break once [19:51] or will have to be forced rather [19:51] lool: why? not enough numbers for future releases :)? [19:51] did we mean to install gnome-control-center on the touch devices? [19:51] did we pick 1 byte :)? [19:51] asac: yes, would overflow with 100 bulds in a month [19:51] lol [19:51] ok [19:51] also number were confusing folks [19:51] yeah. make a proper scheme [19:51] ++ [19:51] sized so the universe fits into it :) [19:51] will just start at 1 [19:51] and increment [19:52] but discussing that after plenary if you want to jin [19:52] join [19:52] lets see how i feel then :) [19:52] indicator-power seems to pull in gnome-control-center now [19:52] dholbach, oh crap [19:53] all the indicators depend on it [19:53] they probably should not [19:53] dholbach: since when? [19:53] tedg, ^^? [19:53] is that a regression from today or something? [19:53] asac, I just did an upgrade and it was pulled in [19:53] asac, i suspect since last week [19:53] sil2100: Mirv: ^^ [19:53] maybe a Depends on gnome-control-center | ubuntu-system-settings ? [19:54] sil2100: Mirv: do you remember if something like this went in? [19:54] i assume would (in theory) have required core dev ack [19:54] dholbach: could be [19:54] wow [19:54] dholbach: lets cehck if its on the image too [19:54] ogra_: what? [19:54] :) [19:54] and it has a list f depends as long as my arm [19:55] do a; "dpkg -l |grep gnome" on your phone [19:55] asac: I don't remember that anything like that got ACKed, I guess seb128 would do that ACK if needed, and it's impossible he would ACK something like that ;) [19:55] asac, so i thijnk i'll set up my new manifest comparison script with a cron job ... that way we should spot it eaily next time [19:55] m-b-o, If the db already exists, the tests work a lot better: http://91.189.93.70:8080/job/generic-mediumtests/94/testReport/ [19:56] i would have the image to grow massively in size due to that, but it intrestingly didnt [19:56] seb128 is really rigorous and firm in the ACK's and reviews ;) [19:56] what [19:56] seb128, indicator-power pulls in gnome-control-center on touch? [19:56] pmcgowan, asac, sil2100: side effect of landing the desktop indicators converged [19:57] which results in http://paste.ubuntu.com/6041552/ [19:57] seb128, maybe a Depends on gnome-control-center | ubuntu-system-settings? [19:57] http://paste.ubuntu.com/6041554/ [19:57] pmcgowan, asac, sil2100: dholbach: we need something similar to https://code.launchpad.net/~ted/indicator-datetime/split-out-control-center/+merge/182414 [19:57] aha [19:57] I can work on that tonight/tomorrow [19:58] okay got my new device at least booting, as im sitting in phablet shell, but no display. anyway to force via shell a display to try and load? [19:58] seb128, brilliant [19:58] I've also "looking at e-d-s pulling in webkitgtk/gtk" on my ff list [19:58] seb128: ok thanks. [19:58] lool: when is the discussion on the versions? [19:58] seb128: i assume its under control and not hurting short term to carry that? [19:58] cjohnston: in 5 mn [19:58] seb128: or should we look into backing the cause out? :) [19:58] stgraber: mind adding cjohnston? === Zic is now known as Guest41094 [19:58] seb128: i sense its not hurting, just asking :) [19:58] lool: ack.. stgraber plars and doanac` also possibly [19:59] asac, backing out ... like in, dropping all indicators ? [19:59] ogra_: not sure, we would systematically look :) [19:59] in worst case pop off all stuff that came in one by one :) [19:59] lol [19:59] but lets do that another time [19:59] that transition took several days when doing it forward, how much do you think it would take rooolling it back ? [19:59] :) [20:00] ogra_, this is my favorite subject from months ago, we need something watching the manifest [20:00] ogra_: not sure, i would like that data though :) [20:00] mhall119: Hey [20:00] asac, that's nothing new, not sure what you want to back out [20:00] asac, we know the cause, a direct depends [20:00] mhall119: could you rebuild the xda click? it seems it didn't get the right special hints in the .click to say it's actually a click; rebuilding with latest version of click build should be all there is to is + bump version and uploadin appstore [20:01] asac, gnome-control-center is there since we landed the converged indicators (not new) [20:01] g-c-c won some new depends today [20:01] pmcgowan, i have a script to do that http://paste.ubuntu.com/6040934/ ... i'll set up a cron job and publish this regulary from tomorrow on [20:01] which adds to the issue [20:01] mhall119: sorry for the hassle, but this breaks installation in latest image [20:01] ogra_, awesome [20:01] asac, but it's only disk footprint waste, not runtime issue [20:02] minro disk foorprint [20:02] *minor [20:02] ogra_, asac dashboard should have an image size and package count check [20:03] pmcgowan: +1 [20:03] pmcgowan, our image build system has an oversized check [20:03] ogra_, disk is cheap, 640G should me enough ;-) [20:03] fginther: hmm, strange. Did you started #94 by hand? the 2 tests failing should be fixed by a commit 30 minutes ago [20:03] not sure how much it takes to enable that for touch [20:03] pmcgowan, our days should also have 30 hours ;-) [20:03] seb128, haha ++ [20:04] seb128, apparently yours does today [20:04] lol [20:04] if only it was "today"... ;-) [20:04] but vUDS is over, so good [20:04] pmcgowan: yeah, let me add that to some list [20:04] we can get back to get work done [20:05] And right about ...... now ...... everyone misses beer time at physical UDS [20:05] +++++++++++ [20:05] and + [20:05] * josepht cries [20:06] not only the beer time [20:06] cjohnston: do we maintain a wishlist for dashboard somewhere? [20:06] asac: bugs [20:06] yeah, i kind of knew that was coming :) [20:06] asac: I don't think that anything currently provides us with that info, so someone would need to create something to provide that [20:07] lool, cjohnston: coming? [20:07] cjohnston: i really just want to collect loads of ideas atm and then distill them into real wishlist bugs :) [20:07] stgraber: I haven't gotten an invite? [20:07] cjohnston: let me invent something for that purpose :) [20:07] cjohnston: I sure invited a Chris Johnston, not sure which account that was though ;) [20:07] lol [20:08] cyphermox: testing your package now [20:10] any people here running UT on a Galaxy S? [20:14] asac, pmcgowan: this sounds like something that could be a smoke test and then just show up as any other test result like "share-app-autopilot"? [20:14] Wellark, ping [20:14] fginther: seems like the MP is merged now [20:14] doanac`, works for me [20:15] asac: so probably is less of a dashboard feature and more of asking someone to write such a test [20:15] m-b-o, #94 was done by hand. [20:16] m-b-o, and it works right now because there is a copy of the DB in place with the test starts now. I suspect this is a bug, but I wasn't able to reproduce it outside of the jenkins machine [20:17] fginther: tried on my device, both with db pre existing and deleted. worked both times [20:17] m-b-o, I tried to find a missing dependency, but nothing came up [20:18] so in adb devices my device is listed as offline, any tips? [20:18] already tried sudo adfb kill-server etc, and debug mode is on [20:18] doanac`: i think for now we want to see the information [20:18] and not a red/green test [20:19] not saying that it couldnt be a test :) [20:19] balachmar: is system-image-cli -b 0 -c daily-proposed --filter=full supposed to give me any output (eg, progress)? [20:19] fginther: it's even stranger, when a db-file already exists, it will be removed and a new db will be created... [20:19] asac, 29.2 is ready [20:19] place your bets :) [20:19] (well, nearly, still syncing to cdimage) [20:19] ooh, and I was never asked to accept a a host key [20:19] great [20:19] now its done [20:20] asac: okay. then lets add it to the qa-dashboard BP: https://blueprints.launchpad.net/ubuntu/+spec/qa-s-dashboard [20:20] doanac`: is utah in shape for 29.2 [20:20] jdstrand: I think you didn't mean to ask that from me [20:20] ouch question mark doesnt work anymore on keyboard :) [20:20] m-b-o, I don't get it either [20:20] balachmar: hehe, you are correct :) [20:20] barry: is system-image-cli -b 0 -c daily-proposed --filter=full supposed to give me any output (eg, progress)? [20:21] asac: i think so. [20:21] jdstrand: :) [20:21] doanac`: so you don't know :) [20:21] ? [20:21] that build was.. quick :o [20:21] why wouldn't it be? [20:21] it works again :) [20:21] odd [20:22] fginther: thanks anyways! [20:25] lool: just uploaded 0.1.6 of my app, click package created via QtCreator [20:25] m-b-o, I filed https://bugs.launchpad.net/ubuntu-weather-app/+bug/1218598 [20:26] Launchpad bug 1218598 in Ubuntu Weather App "Unable to run autopilot tests due to missing local storage db" [Undecided,New] [20:26] mhall119: thanks! [20:26] rsalveti: don't bother, it's borken [20:26] beuno: do click uploads require review for each version? if yes, would you mind approving mhall119's updated xda? [20:26] cyphermox: :-( [20:26] of course, it has the .bzr directory in it again, so 5x as big [20:26] ping me when it's fixed then [20:27] beuno: they do require review, yes [20:27] jdstrand: -v [20:27] jdstrand: to get some output [20:27] jdstrand: add -v or -vv [20:27] -vv is a bit too verbose now in my experience [20:27] rsalveti: now should be much better [20:27] like, downloaded bits [20:27] hmm. it's been running for nearly an hour... [20:28] cyphermox: cool, let me test [20:28] barry: if I ctrl+c will it continue where it left off? [20:28] jdstrand: you can ^C it, it will redownload from start [20:28] not resume, no [20:28] no resume yet [20:28] that is planned? [20:28] jdstrand: but at worst it's 300M, so if you have decent connectivity, it should not be 1h [20:28] yes, with the new download service [20:29] lool: you are assuming a lot about my connectivity [20:29] I'm not assuming, *if* you have decent connectivity ;-) [20:29] there is a reason why I mirror all devel and stable releases of Ubuntu, and it isn't because I'm a packrat [20:30] * lool mirrors latest stable and devel Ubuntu and ports and Debian too, but that's because I *used* to have bad connectivity [20:30] now I'm just wearing out these poor disks [20:30] that die one after the other [20:30] running with 2 disks out of the original 5 in raid 1+0 [20:30] mirror is soon gone! [20:31] jdstrand: I can send you a pidgin with an USB key of the latest delta [20:32] I'll take it! :) [20:32] it's not horrible by 2005 standards [20:33] one pidgin per day with a 64 GB USB key is pretty good Mbits [20:33] yes :) [20:33] 759 kbits [20:33] /s [20:33] oh sorry 7 Mbits/s [20:33] that's a really funny way to think about it :) [20:34] 759 kB/s [20:34] that's more than my DSL! [20:35] <_polto_> small wifi traffic with wpa2 uses 100% of CPU on nexus7 [20:35] hello [20:36] <_polto_> on one of the cores. [20:38] <_polto_> and /usr/bin/brcm_patchram_plus ---- 100% also. [20:39] <_polto_> system/bin/brcm_patchram_plus --enable_hci --scopcm=0,2,0,0,0,0,0,0,0,0 --baudrate 3000000 --use_baudrate_for_download --patchram /etc/firmware/bcm4330.hcd --no2bytes --enable_lpm --tosleep=50000 /dev/ttyHS2 ------ consume 100% CPU (one core) [20:39] yes, seen that in the test farm too [20:39] that's some broadcom driver getting crazy [20:39] we've seen it in the qa lab as well [20:39] <_polto_> lool, anything we can do about ? [20:40] cyphermox: working with with maguro, testing mako now [20:41] _polto_: I don't know I'm afraid; you could file a bug against the kernel perhaps, or rather against the package shipping that file [20:43] <_polto_> lool, I'll try some more tests without WPA2, ... [20:43] lool: why can't i set the milestone on LP: #1215959 ? [20:43] Launchpad bug 1215959 in Ubuntu system image "Report image versions" [Low,Triaged] https://launchpad.net/bugs/1215959 [20:43] rsalveti: ack [20:44] barry: are there milestones in this project? [20:44] hmm yes [20:44] lool: oops, no unreleased ones ;) [20:45] ah [20:51] cyphermox: ship it [20:53] awesome, this is getting better everyday [20:53] woot, new image [20:53] cwayne: pong [20:53] only issue now is that the data connection is on all the time by default, as long you have a valid sim card [20:55] barry: daily-barry is currently importing images, should be ready in ~45min (yeah, importing stuff is slow.... still no pxz on the machine) [20:55] Wellark, hey, should the unity launcher actually use the gsettings key now? it doesn't seem to [20:55] com.canonical.Unity.Launcher favorites seems to be ignored [20:55] stgraber: i was about to say that [20:55] lool: the server side change required for the new numbering scheme was a one line change, simple enough :) [20:55] watching top on nusakan [20:56] stgraber: :-) [20:56] lool, barry: actually, let me cheat a bit since we already have those images in daily [20:56] * lool goes taking some tea and then will do a final click test [20:56] stgraber: cool. [20:56] cwayne: not yet. the current plan is to have it working by Tuesday [20:56] lool, stgraber, cjohnston: LP: #1218612 [20:56] Launchpad bug 1218612 in Ubuntu system image "Support new version number scheme" [Critical,Triaged] https://launchpad.net/bugs/1218612 [20:56] thanks [20:57] stgraber: cool, thanks [20:57] cool thanks [20:59] stgraber: so it seems delta generation completed for 29.2, but I dont see it in the json; is it due to some internal mirrors? [20:59] (in http://system-image.ubuntu.com/daily-proposed/grouper/index.json that is) [21:00] barry: done [21:00] is there a way to re-enable the tour after taking it? [21:01] mhall119, still need that review? [21:01] beuno: I'm a slow learner :) [21:01] oh, review of my app [21:01] yes, it appears I do [21:01] lool: it's being repacked at the moment [21:02] barry: http://system-image.ubuntu.com/daily-barry/grouper/index.json [21:02] stgraber: yay! [21:03] mhall119, it grew in size quite a bit, is that expected? [21:04] mhall119, did not pass review: http://paste.ubuntu.com/6041759/ [21:05] includes the .bzr and is missing apparmor [21:06] beuno: I used QtCreator again, so all of .bzr is in there [21:06] mhall119, old version as well, as it doesn't have the apparmor bits? [21:07] I thought I was up to date [21:07] let me check again [21:09] popey: do you know of a way to get back to the phablet tour? [21:10] that phablet tour was pretty neat when I first saw it btw :) [21:10] yeah, it is [21:10] I want to test it out on family members [21:11] without having to phablet-flash again [21:20] sergiusens: do you know if the nexus 7 images work? mmcc on my team had issues with his 7 (it used to work fine last week) [21:21] sergiusens: details in bug #1218065 [21:21] bug 1218065 in Unity 8 "unity8 crashing repeatedly on nexus 7" [Undecided,New] https://launchpad.net/bugs/1218065 [21:24] pmcgowan, Fixed === Zic is now known as Guest54718 [21:26] tedg, which the control panel dep? [21:26] mhall119: latest xda app seems to work again, thanks [21:26] pmcgowan, Yeah, it'll take a while to get through the system though. [21:27] tedg, great, let seb know its all done so he doesnt need to look at it [21:27] cjwatson: click support seems functional in latest cdimage --pending build; trying r/o image now [21:27] mmcc: alecu long story short, dist-upgrading is not supported [21:27] lool: \o/ [21:27] pmcgowan, Will do, seems to be the last one. [21:27] mmcc: alecu you are missing hybris changes on the android side [21:27] pmcgowan, All the other indicators look good, I was able to remove gnome-control-center from my phone. [21:27] tedg, cool [21:28] sergiusens: ah, I get it now. makes sense, thanks! [21:28] sergiusens: aha. so, a "phablet-flash cdimage-touch --wipe" is called for then? [21:29] mmcc: yes, or no --wipe if you don't want to === salem_ is now known as _salem [21:30] sergiusens: that's confusing, since "phablet-flash cdimage-touch" without --wipe failed to help yesterday... [21:32] ping nik90 [21:33] Wellark, thanks, is that likely to happen? [21:34] mhall119: edit /var/lib/AccountsService/users/phablet, remove the demo-edge=false line, reboot [21:34] lool: beuno: so when does the app store go public? [21:34] mmcc: oh, I was just basing this out of your bug report [21:34] mhall119, we tried today but didn't make it, hoping for tomorrow [21:35] mmcc: there was an issue with the builds yesterday but with a reboot loop [21:35] beuno: cool, I'll ready uReadIt [21:36] alecu: thanks [21:36] sergiusens: hmm. my bug report was that 'dist-upgrade' had problems, and when I tried to fix with a 'phablet-flash cdimage-touch', it had the same problems :( [21:36] mhall119: I wanted to see it again, too :-) [21:36] mmcc: never mind, I'm just tried and headaching now [21:36] sergiusens: no problem. I'm currently re-flashing with today's image and will update the report [21:37] mmcc: ah, you may have caught the build when the switch from android from jenkins to android from ubuntu archives was being done [21:38] sergiusens: ok, interesting. [21:38] cyphermox: that script I made now takes a -l to enable extra nm logging fwiw [21:38] alecu: where do I file bugs against the tour? [21:39] mhall119: I've no idea! I just grepped the code to see how to enable it again. [21:39] ah, lol, ok [21:39] mhall119: I suspect it's lp:unity8, since that's where the qml code seems to be [21:39] ok [21:42] boiko, ping :) [21:42] cwayne: pong [21:43] boiko, hey, i noticed you just marked this fixed, should the fix be in today's image? or not til tomorrow? https://bugs.launchpad.net/telephony-service/+bug/1216139 [21:43] Launchpad bug 1216139 in telephony-service "phone app not respecting gsettings key for tones" [Undecided,Fix released] [21:43] cwayne: should be already fixed in the images that have the new apps (dialer-app and messaging-app) [21:44] boiko, ah, i still saw the bug today with the dialer-app [21:46] cwayne: really? /me checks the gsettings file [21:48] !$#@$@ [21:48] click still broken with r/o images [21:48] cwayne: well, the code is definatelly in the telephony-service, so maybe the config options are not properly saved? I'm double checking here [21:48] so preinstalled packages are there [21:49] thanks alecu, sending them an MP [21:51] alecu: still some breakage in latest image [21:51] less [21:51] [unity-scope-click] - DEBUG: click-scope.vala:183: Error building preview: No manifest found for app_id: ar.com.beuno.hello-world [21:52] lool: is that on the RO image? [21:52] yes [21:52] alecu: pkcon works as phablet over adb shell [21:53] alecu: checking whether pkcon installed apps works now [21:53] cwayne: found the problem: if you reboot the phone it is going to play the new ringtone for incoming call, but for messages it should pick the change for the next message already [21:53] alecu: lool that app_id is different from a click app_id btw [21:53] alecu: but at least it appeared in click list, when it didn't when installed from scope [21:53] as in, it looks different [21:53] sergiusens: which app_id? [21:53] lool: I'm midway flashing --pending, will try the RO image next [21:54] lool: app_id: ar.com.beuno.hello-world [21:54] lool, sergiusens: nevermind about that debug message, the name "app_id" is wrong there [21:54] cwayne: the problem is that the ringtone code in telephony-service is reading the incoming call sound only at startup, I will fix that, an MR should go in your direction really soon :) [21:54] alecu: and I confirm the app is there and can be launched [21:54] (some icon issues but it's there) [21:54] alecu: ok, seems just like the package name [21:54] sergiusens: it should say "package_name" in that debug message instead, and the id looks right. [21:54] alecu: great [21:55] boiko, awesome! man i missed working with you :P [21:55] alecu, sergiusens: The good thing is that it feels we're down to a much smaller number of bugs [21:55] I should confirm the xda one real quick [21:56] cwayne: might be that I am just happy that by switching to the new apps we got rid of lots of bugs :) [21:56] uh Fatal error: Failed to obtain authentication. [21:56] boiko, :) [21:57] mhall119: Hmm I might have jumped too happily on the xda package; I still get a different behavior than hello world; only renaming the download to .click allows installing it in the r/o image [21:58] right [21:58] mhall119: that's with 0.1.5 [21:58] mhall119: cjwatson said it had something to do with the click-build version you're using that should set the right things in the generated .click; if there's no old version lying around, then that might be a bug in click [21:59] cjwatson: ^ [21:59] mhall119: lool let me track down the qtcreator stuff for this [21:59] cjwatson: file --mime reports application/x-debian-package on all of them, not sure how to test exactly [22:03] alecu: I ran packagekit by hand with -v -v [22:03] alecu: saw this error: [22:03] 22:02:27 PackageKit internal error: failed, but no error code: unknown [22:03] lool: seems to be using click build [22:04] I wonder whether we're seeing inconsistent results because transacations are accumulated [22:04] ah, the joy of pkit [22:04] so the sdk fix didnt really help :( [22:04] ogra_: ^^ [22:04] http://reports.qa.ubuntu.com/smokeng/saucy/image/3832/webbrowser-app-autopilot/ [22:04] MAN [22:04] lool: ah, so it's not the scope after all... anyway, :-( [22:05] alecu: I dont know [22:05] alecu: the scope dies pretty much every time [22:05] plars: can you give back: http://reports.qa.ubuntu.com/smokeng/saucy/image/3833/webbrowser-app-autopilot/ [22:05] seems super busy [22:05] alecu: I just mean that tests are inconsistent because of state saved in transactions.db [22:05] somehow unity is looping [22:05] not sure [22:05] hope the next image has that fixed [22:05] asac: I'm giving back the whole thing, I just merged the new {dialer,messaging,address-book}-app tests [22:06] asac: but we also got a phablet-tools update [22:06] asac: not sure if it's causing problems yet, so standby [22:06] plars: the whole thing? [22:06] :) [22:06] plars: if you wish i am fine [22:06] ah it worked this time [22:06] but i guess there is not much hope [22:06] plars: so we thought this webbrowser stuff that is failing on maguro [22:07] without system-settle being wild [22:07] lool: afaict the scope is dieing of inactivity [22:07] was fixed by a ui-toolkit regression [22:07] asac: I needed to, because I needed to regenerate the whole set of jobs, and it's just easier that way... not many things have run yet on .2 [22:07] plars: can you maybe check if going back two or three uploads of the uitk helps? [22:07] alecu: I can never go back to apps scope when anything goes wrong [22:07] i really would prefer if we could isolate this issue first :) [22:07] plars: yeah sounds good [22:07] asac: if it fails again, I'd have to try that locally [22:07] plars: look maguro [22:07] no systemsettle problems [22:08] but still same shitty results as this morning [22:08] as if the ui-toolkit fix, fixed like nothing :) [22:08] asac: yeah, sensorservice was going nuts it seems [22:09] on mako at least [22:09] lool: 0.1.6 is the new XDA package, not 0.1.5 [22:09] mhall119: ah appstore still serves me 0.1.5 [22:09] 0.1.5 was my manually mangled package that cjwatson didn't like [22:09] lool: perhaps because beuno hasn't approved the 0.1.6 changes yet [22:09] mhall119, but I have not approved it,so the store still has 0.1.5, right? [22:09] right :) [22:10] lool: I've just flashed with --pending and --wipe. pkcon is happily reporting both installations as completed, but the app directories are empty. [22:10] I've dist-upgraded, so let me build 0.1.7 [22:10] lool: it's the same error I pointed a few hours ago: https://bugs.launchpad.net/ubuntu/+source/click/+bug/1218483 [22:11] Launchpad bug 1218483 in click (Ubuntu) "Installation errors are not reported" [High,New] [22:11] alecu: I didn't get this error [22:11] alecu: apps appeared for me [22:11] rsalveti: any idea what might be going on with sensor? [22:11] rsalveti: https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-webbrowser-app-autopilot/89/artifact/clientlogs/top_before.log/*view*/ [22:11] from: http://reports.qa.ubuntu.com/smokeng/saucy/image/3833/webbrowser-app-autopilot/317869/ [22:11] cwayne: https://code.launchpad.net/~boiko/telephony-service/fix_ringtone_updating/+merge/183044 [22:11] and: http://reports.qa.ubuntu.com/smokeng/saucy/image/3833/webbrowser-app-autopilot/ [22:12] rsalveti: feels like a race that happens only on some boots [22:12] cwayne: can you give it a try using the debs the CI job generates? (CI is going to take some time to run for the MR though) [22:12] rsalveti: the other boots here dont have that problem http://reports.qa.ubuntu.com/smokeng/saucy/image/3833/ [22:12] lool: ah, that's on the RO image, right? I'm still with the non-RO one [22:13] alecu: yes, but I tested the regular image earlier and it just worked [22:13] albeit I hadn't properly wiped my home [22:13] so I guess that was unconclusive [22:14] lool: afaict --wipe also erases the apps installed in /opt/click..., so yes, they might be the ones already installed. [22:15] boiko, sure thing [22:17] lool: I'm getting the same "internal error: failed, but no error code: unknown" you posted. [22:18] the scariest thing is pkcon returning OK! [22:18] ok, I need to be a dad for the rest the day. [22:18] alecu: ok [22:18] there's not much I can do at this point :-( [22:18] see you guys tomorrow! [22:43] lool: beuno: 0.1.7 uploaded [22:43] thanks === Ursinha is now known as Ursinha-afk [22:43] device doesn't boot anymore [22:43] np:/ [22:43] no my fault [22:43] so will leave it there for tonight [22:43] mhall119: must be the xda app! [22:43] I knew it was evil [22:44] I'm sure the extensive security review would have caught all the evil things that code is doing [22:44] mhall119, looking [22:44] in any case I can tell you it's one of the 2 most popular apps from the appstore [22:44] \o/ [22:45] still 3 stars in the scope though :( [22:46] mhall119, still has .bzr though [22:46] I'm going to add "Had the most downloaded app in the Ubuntu phone app store for over a week" to my CV [22:46] beuno: not my fault, blame bzoltan [22:46] I think that one isn't fixed yet [22:46] mhall119, :) [22:47] lool: mhall119 constructed his package by hand in a bogus way, hence the mime failure - he didn't use "click build" at all [22:47] ^^ was 0.1.5 [22:47] everything after that was built via QtCreator [22:47] looks like this has been addressed as I catch up on more scrollback [22:47] 0.1.6 even by the right version of qtcreator [22:48] 0.1.7 I meant [22:48] checking it now [22:49] "lint_package_filename_pkgname_match": "'com.ubuntu.developer.mhall119.xda-developers-app-0.1.7' != 'com.ubuntu.developer.mhall119.xda-developers-app' from DEBIAN/control" [22:49] mhall119, ^ [22:49] not sure if that breaks anything or not [22:49] the lint tool sure doesn't like it [22:50] am happy to approve if you think this is better than 0.1.5 [22:52] cjwatson: trying to figure out what makes the pk-click plugin / pk unhappy [22:52] where did DEBIAN/control come from? [22:53] cjwatson: seeing this "internal error" http://paste.ubuntu.com/6042085/ [22:53] not sure, probably click generates it? [22:53] now reflashing since it wont boot [22:53] beuno: it does [22:53] lool: no idea, not enough info [22:53] cjwatson: yeah, I'll strace it to see if I spot anything while click install runs [22:53] lool: I think I've seen that in apparently otherwise successful builds and shrugged [22:54] ah [22:54] beuno: I have no idea where "lint_package_filename_pkgname_match" comes from though - that's not me [22:56] beuno: does this actually mean "the package is called com.ubuntu.developer.mhall119.xda-developers-app-0.1.7.click when it should be called com.ubuntu.developer.mhall119.xda-developers-app_0.1.7_all.click"? I think something in QtC might be renaming the package under me, or maybe the app store - click build always calls it packagename_version_arch.click [22:57] Qtc created the file name com.ubuntu.developer.mhall119.xda-developers-app_0.1.7_all.click [22:58] is the lint tool generating the wrong expected filename? [23:01] boiko, just installed from the CI and rebooted, didn't seem to work [23:03] plars: seems not many tests were run at all? [23:04] i guess you restarted before we got more? [23:04] * asac assumes this is still running and says good night :) [23:04] 'night [23:18] boiko: https://code.launchpad.net/~lool/telephony-service/add-missing-dconf-cli-dep/+merge/183055 [23:19] alecu: so fresh image, pkconf install-local from terminal works but installing from scope doesn't [23:19] alecu: with hello-world package [23:21] Hello! [23:22] I'm trying to port Ubuntu Touch to a new device, Samsung GT-S7500. It's not supported officially by CM, but there is a [23:22] asac: yes, it's all running, I'll be checking on it tonight [23:22] an unofficial build. [23:24] I want to ask a question: after putting the local manifests in .repo/local_manifests, should i run breakfast [codename]? === Ursinha-afk is now known as Ursinha [23:43] Is there an Ubuntu tablet install wiki? === frals_ is now known as frals [23:44] Is there a list of Ubuntu approved tablets? [23:47] cjwatson: So I can see packagekit run click info on the .click, looking at the code it would extract name and maintainer and architecture from the json, but it doesn't contain architecture [23:49] cjwatson: it seems to call pk_package_id_build() on it [23:49] architecture is a bodge at the moment [23:49] hmm that's non-fatal in pk anyway, it is just replaced with "" [23:49] cjwatson: I straced packagekit calling when the scope requests installation [23:50] cjwatson: I confirmed the .click data is downloaded directly by download manager (it's named .click BTW), then it runs click info, but never goes past that [23:50] I've had three pints including some rather strong cider, any sense you get out of me is coincidence [23:51] cjwatson: oh best time for debugging :-) [23:51] I'm planning on buying a tablet and installing Ubuntu on it With a bluetooth keyboard and protfolio case it would be a very cost efficient touchscreen laptop. [23:51] CountryfiedLinux: Check the First link in the MOTD. [23:51] Gnome Shell would most likely be what I'd use [23:52] http://forum.xda-developers.com/showpost.php?p=45109578&postcount=3 [23:52] thanks Fishscene [23:52] ah! http://paste.ubuntu.com/6042220/ [23:53] any tips? [23:54] Fishscene, I don't see a how to wiki. I'm looking for a step by step how to. [23:55] thomi: so if I'm dipping my toes in on autopilot, trying to fix that chronically failing app test from the other day; and I cannot figure out how to run an autopilot test reliably on the device over adb. Is there a howto I should be following that tells me how to set up the environment? (Current failure, after manually passing in the dbus env vars: DBusException: org.freedesktop.DBus.Error.NoReply) [23:57] On that link, click on "Install" (Get started here!). Which leads you to here: https://wiki.ubuntu.com/Touch/Install (Which gives a list of currently "supported" devices, as well as a step-by-step how-to.