| mhall119 | no, not yet, let me put that on my todo list for tomorrow | 00:02 |
|---|---|---|
| Chocanto | mhall119: No sorry, It's not a hurry, take your time ! | 00:04 |
| mhall119 | no need to apologize Chocanto :) | 00:07 |
| Chocanto | mhall119: Ok ok ^^ | 00:15 |
| AskUbuntu | Ubuntu Touch for Thinkpad x200 tablet | http://askubuntu.com/q/321327 | 00:18 |
| === thomi is now known as thomi|lunch | ||
| DJJeff | is ummmmm usb tethering possible? http://i.imgur.com/FszXfdO.png | 00:24 |
| balloons_ | vthompson, yes you still need to assign it an objectname to image | 00:51 |
| balloons_ | upi dpm | 00:51 |
| balloons_ | you don't need an objectname for ubuntushape | 00:52 |
| === Gh0sT is now known as mahabal | ||
| IllegalOpcode | nsi | 01:25 |
| kanwisher | phatablet seems to wait for ever on "waiting for device" after it restarts the device | 01:25 |
| kanwisher | i can even see a device in 'adb devices' | 01:26 |
| DJJeff | fuck ya, finally got reverse USB TETHERING to work so I can get internet on my Ubuntu Touch over USB :) | 01:32 |
| DJJeff | gonna make a howto/video and post it up later :) | 01:32 |
| kanwisher | DJJeff: very cool | 01:33 |
| DJJeff | only took putting eth0 and usb0 on my ubuntu laptop into bridge mode then on my android doing netcfg rndis0 dhcp | 01:35 |
| DJJeff | inside of a root@android shell prompt | 01:36 |
| DJJeff | fired up wireshark on usb0 and was able to see the DHCP DISCOVER packets | 01:36 |
| === thomi|lunch is now known as thomi | ||
| annerajb_ | rsalveti: i been trying to deploy the raring zip and it's running out of space on the device. should i skip installing that and just use the bootstrap i build? | 01:53 |
| annerajb_ | also the install script dosnt fail the frecovery on a error untaring it goes along fine | 01:53 |
| annerajb_ | http://pastebin.com/4NsCPqd7 | 01:54 |
| vthompson | balloons_, Yea, I'm still stuck. | 01:57 |
| vthompson | balloons_, I have the following branch in the works if you want to give it a go: lp:~vthompson/music-app/autopilot-test-for-play-track | 01:58 |
| rsalveti | annerajb: well, you need to install the ubuntu image somewhere where you have enough space at least | 01:59 |
| annerajb_ | rsalveti: or start deleting stuff from the image till it installs :P ima look into it make sure i didnt made any mistakes and you where mentioning earlier of a test to run before installing the kernel that's in the bootstrap you remember about that? | 02:00 |
| rsalveti | annerajb: sure, but seems surfaceflinger is running just fine in your env, so it should be working | 02:07 |
| rsalveti | so just need to enable the kernel configs and install the ubuntu zip there | 02:07 |
| annerajb_ | i tried to run it by hand and it seg fault. rsalveti already did the kernel changes but now the ubuntu zip is running out of space. | 02:07 |
| annerajb_ | thought it may have been out of space without the changes i did since it's not failling when the tar command fails on the deploy script | 02:08 |
| rsalveti | right | 02:08 |
| annerajb_ | btw rsalveti not sure if this is normal but it's using around 900mb (probably more since it never finishes extracting) does that sound about right??? | 02:09 |
| rsalveti | yeah, it uses a bit more than 1gb | 02:10 |
| annerajb_ | Oo i found a 9mb log file in /var/log/ | 02:12 |
| annerajb_ | in the raring-preinstalled*.zip | 02:12 |
| rsalveti | yeah, we still need to optimize that | 02:14 |
| dejello | Hello all | 02:23 |
| crocket | It seems people are not psyched about ubuntu devices. | 02:31 |
| kanwisher | is there a way to enable hdmi out on the nexus 4 with ubuntu | 02:36 |
| === chihchun_afk is now known as chihchun | ||
| === Namidairo`bnc is now known as Namidairo | ||
| johny___ | hello | 05:40 |
| johny___ | i have a nexus 7, whoi is the correct dist, for this nexus? | 05:41 |
| wilee-nilee | johny___, YOU can install the touch or saucy | 05:47 |
| wilee-nilee | as far as what this channel is orientated with | 05:47 |
| johny___ | thankssss | 05:48 |
| === Guest81623 is now known as Zic | ||
| oSoMoN | mzanetti: ping | 07:12 |
| dholbach | good morning | 07:29 |
| rickspencer3 | hi dholbach | 07:30 |
| rickspencer3 | btw, I'm back in Berlin today :) | 07:30 |
| sil2100 | gusch: hello! I see some updates on the bug #1202201 | 07:30 |
| ubot5 | bug 1202201 in Ubuntu UI Toolkit "Share dialog does not appear with latest SDK" [High,Confirmed] https://launchpad.net/bugs/1202201 | 07:30 |
| sil2100 | gusch: it seems that simply the way things are done changed, so I guess gallery-app will have to be modified instead? | 07:31 |
| dholbach | hi rickspencer3 | 07:31 |
| gusch | sil2100: in the works https://code.launchpad.net/~schwann/gallery-app/gallery-no-album-share/+merge/175275 | 07:32 |
| gusch | sil2100: but jenkins coudn't land it yesterday | 07:32 |
| sil2100 | Jenkins is a b*tch lately | 07:32 |
| sil2100 | Since my other branch can't land as well, because of: ERROR:pbuilderjenkins:Error during build execution | 07:33 |
| sil2100 | Whatever that means | 07:33 |
| sil2100 | Need to poke QA | 07:33 |
| rickspencer3 | gema, ^ | 07:34 |
| gusch | sil2100: if hope that now jenkins has a low load, I'm more lucky ... | 07:34 |
| * gema has been poked and is looking at it | 07:35 | |
| gema | jibel: sil2100 is having autolanding issues, could you or didrocks look into that? | 07:40 |
| didrocks | gema: what is autolanding? | 07:41 |
| gema | didrocks: that's what I was wondering :P | 07:41 |
| didrocks | gema: we told to not use those terms, upstream merger or daily release | 07:41 |
| gema | didrocks: ack, then I need to go back to my dash guys and undo a comment I made yesterday | 07:41 |
| gema | didrocks: he is having problems with upstream merger, I guess? | 07:41 |
| didrocks | gema: I think he's talking about upstream merger, which is managed by your team FYI ;) | 07:41 |
| sil2100 | ;D | 07:41 |
| sil2100 | https://jenkins.qa.ubuntu.com/job/qtubuntu-cameraplugin-fake-saucy-armhf-autolanding/1/console | 07:42 |
| sil2100 | Errors like these | 07:42 |
| didrocks | gema: yeah, it's your/fghinter's team | 07:42 |
| gema | didrocks: understood | 07:42 |
| didrocks | gema: to come back to terminology, do you mind working with fghinter to remove all the terms of "ci" or "autolanding" from jenkins jobs and docs? | 07:44 |
| gema | sil2100: I have heard of that error yesterday, we changed martin's key for a differnt one and some builds have been giving problems | 07:44 |
| didrocks | gema: people are confused and we never know if they are talking about usptream merging or daily release | 07:44 |
| gema | sil2100: (you can see it if you search on the console for "error") | 07:44 |
| didrocks | (or even image build sometimes) | 07:44 |
| gema | sil2100: I am going to have to wait for fginther to come online later to fix (or some of the other folks in the US) | 07:45 |
| gema | sil2100: but it is an easy one to fix, if I am reading the log right | 07:45 |
| gema | sil2100: sorry about that | 07:45 |
| gema | didrocks: no wonder, I am confused as well | 07:46 |
| gema | didrocks: I am starting to gain an understanding, though | 07:46 |
| didrocks | gema: from my side, I've already done this cleanup, so I think the upstream merger one is the one remaining to clean | 07:46 |
| didrocks | gema: then, it will take time for people to change their terminology, but with patience and repetition… :p | 07:47 |
| gema | didrocks: it is going to help that we put the right terminology on the dash we are landing this week | 07:47 |
| gema | I am going to make sure of that in today's meeting | 07:47 |
| didrocks | gema: I hope so :) | 07:47 |
| didrocks | thanks! | 07:47 |
| gema | didrocks: so your stuff (landing on the distro) is daily release, right? | 07:48 |
| didrocks | gema: exactly | 07:48 |
| gema | didrocks: ack | 07:48 |
| didrocks | gema: distro or "feature ppa" | 07:48 |
| didrocks | but yeah, called daily release :) | 07:48 |
| gema | didrocks: ok, it makes sense | 07:49 |
| gema | autolanding was an overloaded term, that's the problem | 07:49 |
| didrocks | right | 07:50 |
| didrocks | as CI | 07:50 |
| didrocks | that's why I kept repeating to not using them as soon as I see them being used :) | 07:50 |
| sil2100 | gema: thanks ;) Too bad we have to wait, but good that it's fixable | 07:50 |
| didrocks | sil2100: gema: won't that be a case of pushing to trunk to unblock the stack? | 07:51 |
| rickspencer3 | gema, there is no other way to get sil2100 unblocked other than waiting for 3+ hours for someone in the US? | 07:51 |
| sil2100 | There's also gusch's branch that is a bit more important that probably fails because of the same reasons | 07:53 |
| sil2100 | https://code.launchpad.net/~schwann/gallery-app/gallery-no-album-share/+merge/175275 <- same thing during autolanding, the armhf lander fails | 07:53 |
| didrocks | asac: please ensure that fghinter is around on Monday's meeting, I see a lot of merging failures lately (and we had to push unity to trunk as well directly) | 07:54 |
| gema | rickspencer3: right now there isn't , but this is only going to happen today, tomorrow we won't have this problem again | 07:54 |
| sil2100 | didrocks: would you mind if we merge directly to gallery-app's trunk gusch's fix? | 07:55 |
| rickspencer3 | gema, how come? there's only one person who can fix the issue? | 07:55 |
| sil2100 | didrocks: CI jenkins said it's OK, just one of the autolanders dies | 07:55 |
| didrocks | sil2100: same, as long as you build locally and we ensure to have a daily just after that, I'm personnaly fine with it | 07:55 |
| gema | rickspencer3: no, there's plenty of them, they are all in the US | 07:55 |
| didrocks | sil2100: ok, sounds good then :) | 07:55 |
| gema | rickspencer3: or in NZ and already gone | 07:55 |
| sil2100 | gusch: you want to push it to trunk or should I do it ;) We have a green light! | 07:55 |
| didrocks | rickspencer3: gema: FYI, we are workarounding with merging directly to trunk manually, as we did have build runs successfully first and daily release. | 07:56 |
| rickspencer3 | didrocks, good | 07:56 |
| gema | didrocks: ack, in any case I need to fix this later today | 07:56 |
| rickspencer3 | every day is precious | 07:56 |
| didrocks | gema: oh completely, please get that fixed asap :) | 07:56 |
| rickspencer3 | we can't lose a whole day due to infrastructure issues | 07:56 |
| gusch | sil2100: I approved (again) | 07:56 |
| gema | didrocks: will do | 07:57 |
| sil2100 | gusch: oh, not sure if that will help ;/ It'll fail on armhf again | 07:58 |
| === alan_g|EOD is now known as alan_g | ||
| sil2100 | gusch: ok, then once it fails again I'll merge directly | 07:58 |
| sil2100 | gusch: so I'll take care of it | 07:58 |
| gusch | sil2100: ok - thx | 07:58 |
| timp | is image 20130717.2 working? | 08:03 |
| timp | gusch: I saw the uitk bugs that showed up in gallery-app | 08:06 |
| timp | gusch: the toolbutton.enabled I'll check later; I guess it will be easy to fix | 08:06 |
| gusch | timp: the it's weired to once have the caller, and once not | 08:07 |
| gusch | timp: but in the CHANGES file, I didn't see that API change | 08:07 |
| Saviq | timp, current is 20130716, so that's the latest "known to work" | 08:07 |
| timp | Saviq: do you know what is not working in 20130717.2? Or when there will be a new image that works? | 08:08 |
| === schwuk_away is now known as schwuk | ||
| timp | gusch: triggered can still take a parameter (not called caller), so I didn't think the API changed. But I did not take into account that in onTriggered: you use the name of the variable | 08:11 |
| timp | my mistake.. should have been in CHANGES | 08:12 |
| Saviq | timp, today's image is in the works around now | 08:12 |
| Saviq | timp, should be up in ~2hrs I think | 08:12 |
| Saviq | timp, not sure what's (not) working in 17.2 - just flashed now to see | 08:13 |
| timp | Saviq: ok, thanks | 08:13 |
| timp | Saviq: I also flashed it just now, I need some stuff that's not yet in 0716 | 08:13 |
| Saviq | timp, well, shell came up, so that's 90% success | 08:13 |
| timp | Saviq: yes, for me too :) | 08:13 |
| timp | gusch: ToolbarButton still has the caller parameter, but I think that can be removed as well. In PopupUtils.open(), the id of the button can now be passed directly | 08:15 |
| timp | gusch: I see there is already an MR that fixes it :) | 08:15 |
| Saviq | timp, everything seems to be working, not sure why it's not current | 08:16 |
| Saviq | ogra_, popey, any reason why wasn't 17.2 promoted to current? | 08:16 |
| gusch | timp: yep, and using ids of the buttons directly (I like that more than using caller) | 08:16 |
| timp | gusch: ok, good. The code is more clear like that. | 08:17 |
| ogra_ | Saviq, because the automatic tests didnt pass http://reports.qa.ubuntu.com/smokeng/ | 08:19 |
| popey | \o/ | 08:19 |
| Saviq | ogra_, ah, so that's automagic now :) | 08:19 |
| ogra_ | nope | 08:19 |
| timp | ogra_: why didn't they pass? | 08:19 |
| ogra_ | but i'm not allowed to pull the trigger until they pass | 08:19 |
| ogra_ | timp, i just got to the kbd, no idea :) | 08:20 |
| * popey flashes it anyway to test | 08:20 | |
| ogra_ | i will too | 08:20 |
| timp | ok | 08:20 |
| timp | I don't understand http://reports.qa.ubuntu.com/smokeng/ . since when is 28/106 45.7%? | 08:21 |
| ogra_ | that QA math, ask them :) | 08:21 |
| timp | (pass rate for maguro) | 08:21 |
| rickspencer3 | ogra_, did I understand correctly that today's image built, but there were smoke test failures? | 08:21 |
| ogra_ | probably different tests have different weight | 08:21 |
| rickspencer3 | for 20130718 | 08:21 |
| rickspencer3 | ? | 08:21 |
| ogra_ | rickspencer3, 18 is still in the press | 08:21 |
| ogra_ | rickspencer3, thats 17.2 | 08:22 |
| * ogra_ is happy it built, that means there should be the first click packages on it | 08:22 | |
| rickspencer3 | ogra_, ok, so we're still waiting to see if 18 passes? | 08:22 |
| ogra_ | rickspencer3, i watch every image atm, so yes, 18 too | 08:23 |
| ogra_ | looks like the utah dbus issues are still there :/ | 08:23 |
| ogra_ | Jul 18 01:24:30 ubuntu-phablet utah: s.Error.NoServer: Failed to connect to socket /tmp/dbus-JIoEvd8Big: Connection refused#012}}} | 08:23 |
| ogra_ | sigh | 08:23 |
| timp | 17.2 seems to work for me | 08:24 |
| ogra_ | timp, well, its an error in the test suite | 08:24 |
| timp | ah | 08:24 |
| timp | gusch: can you run gallery on the phone using the "run app on device" option in QtC? | 08:24 |
| timp | gusch: it tells me: sed: can't read /home/phablet/dev_tmp/gallery-app/gallery-app.desktop: No such file or directory | 08:25 |
| ogra_ | when switching to upstart user sessions some hardcoded stuff broke | 08:25 |
| timp | gusch: is that caused by some cmake config issue that should create the .desktop file? | 08:25 |
| gusch | because the desktop file in in /home/phablet/dev_tmp/gallery-app/desktop/gallery-app.desktop (desktop sub dir) | 08:25 |
| gusch | timp: and that ;) | 08:26 |
| timp | I only see gallery-app.desktop.in there | 08:26 |
| gusch | timp: so it's the cmake issue, and the sub-dir | 08:26 |
| gusch | timp: I never used the QtCreator to run my app on the device | 08:27 |
| timp | gusch: why do you use .desktop.in? is it for the tr()? | 08:28 |
| popey | 17.2 is painfully slow on first boot | 08:28 |
| Saviq | ogra_, re ↑, how do I stop unity8 on the device remotely? if I try `ssh stop unity8` or `ssh "bash -ic 'stop unity8'"` it doesn't cut it | 08:28 |
| gusch | timp: correct - for translation | 08:28 |
| timp | hm | 08:28 |
| popey | 3fps when unlocking | 08:28 |
| popey | if that | 08:28 |
| Saviq | ogra_, I generally end up with "Unknown job: unity8" | 08:28 |
| timp | gusch: so if you change the language after installing gallery-app, the translation is not updated? | 08:28 |
| Saviq | popey, maybe it's still loading the scope data? | 08:29 |
| gusch | timp: that should work - but the translation there includes quite some "magic" | 08:29 |
| Saviq | popey, seems to work fine here | 08:29 |
| popey | its okay after a minute | 08:29 |
| * Saviq flashes again | 08:29 | |
| gusch | timp: ask oSoMoN if you need to know more about the .desktop translation | 08:30 |
| Saviq | gusch, timp no magic, really :) | 08:31 |
| Saviq | gusch, timp the .desktop file gets CMake-mangled to remove the tr() from around the fields | 08:31 |
| Saviq | gusch, timp, and gettext extracts it from there before that | 08:31 |
| Saviq | gusch, timp, the "receiving side" then uses the X-Ubuntu-Gettext-Domain domain to translate those strings runtime, is all the magic :) | 08:32 |
| timp | ah, ok. sounds good :) | 08:32 |
| gusch | translate those strings runtime == magic ;) | 08:33 |
| seb128 | hum | 08:35 |
| seb128 | trying to use qtmultimedia to play a sound I get | 08:35 |
| seb128 | defaultServiceProvider::requestService(): no service found for - "org.qt-project.qt.mediaplayer" | 08:35 |
| === schwuk is now known as schwuk_away | ||
| seb128 | does anyone know what that's about? | 08:36 |
| JamesTait | Good morning all, happy Nelson Mandela Day! :-D | 08:48 |
| ogra_ | \o/ | 08:49 |
| Saviq | seb128, I'd say you need gstreamer | 08:49 |
| seb128 | Saviq, it's my desktop, I've gstreamer and rhythmbox and totem play ogg just fine (which is what I'm trying to play) | 08:49 |
| Saviq | seb128, ah | 08:49 |
| ogra_ | cjwatson, hmm, is the click installation really so quiet ? i'm looking at http://people.canonical.com/~ubuntu-archive/livefs-build-logs/saucy/ubuntu-touch/20130717.2/livecd-armhf.out ("Setting up click packages", nothing more) | 08:51 |
| cjwatson | ogra_: There's normally some output | 08:51 |
| ogra_ | hmm, then i guess something is wrong | 08:51 |
| ogra_ | we need a "click.manifest" :) | 08:52 |
| cjwatson | ogra_: There must be, since if you look at the hook it should say "Setting up $package" for each one | 08:52 |
| cjwatson | ogra_: So clearly nothing to do with click - it's just failing to actually call it | 08:52 |
| ogra_ | yeah | 08:52 |
| Saviq | seb128, http://pastebin.ubuntu.com/5886881/ works fine here, does it, for you? | 08:52 |
| Saviq | seb128, http://pastebin.ubuntu.com/5886883/ here's my set of installeds | 08:53 |
| cjwatson | ogra_: I don't know exactly why, but the lb_chroot_hooks step is taking a suspiciously long time compared to the previous build, so I'd suspect a timeout | 08:54 |
| cjwatson | i.e. network problem | 08:54 |
| asac | anyone checked why mako is completely busted now? | 08:54 |
| ogra_ | yeah, i wonder if the machine is actually reachable under the external name | 08:54 |
| ogra_ | probably needs to say lillipilly | 08:55 |
| cjwatson | The name shouldn't matter | 08:55 |
| cjwatson | It's the same IP address either way | 08:55 |
| cjwatson | You might try archive-team.internal though; but before messing about I suggest checking with IS | 08:56 |
| ogra_ | well, the seed mirror is people.u.c | 08:57 |
| cjwatson | Oh, but archive-team.internal doesn't serve user home directories | 08:57 |
| ogra_ | so it should work | 08:57 |
| cjwatson | I hope it's people.c.c, not people.u.c | 08:57 |
| cjwatson | (Besides, archive-team.internal will soon become a different machine) | 08:57 |
| ogra_ | but the seeds are processed in config ... which is outside of the build chroot i think | 08:57 |
| cjwatson | What have the seeds got to do with anything? | 08:57 |
| ogra_ | they are pulled from people.u.c | 08:57 |
| ogra_ | so it should work without issues | 08:57 |
| cjwatson | Oh, preinstalled does a germinate run | 08:58 |
| ogra_ | yes | 08:58 |
| cjwatson | Seeds are pulled from people.c.c, not people.u.c | 08:58 |
| ogra_ | but in the config step | 08:58 |
| cjwatson | Please make that distinction clear in your mind as they are separate machines | 08:58 |
| ogra_ | oh, yeah,, sorry | 08:58 |
| ogra_ | it ususally is .... finger memory | 08:58 |
| cjwatson | *.ubuntu.com) MIRROR=http://ftpmaster.internal/ubuntu/ | 08:58 |
| cjwatson | SEEDMIRROR=http://archive-team.internal/seeds/ | 08:59 |
| seb128 | Saviq, | 08:59 |
| seb128 | $ qmlscene sound.qml | 08:59 |
| seb128 | defaultServiceProvider::requestService(): no service found for - "org.qt-project.qt.mediaplayer" | 08:59 |
| ogra_ | oh, where is that | 08:59 |
| cjwatson | So in fact they are pulled from archive-team.internal, not people.c.c | 08:59 |
| cjwatson | top of live-build/auto/config/ | 08:59 |
| Saviq | seb128, do you have the QtMultimedia plugins installed? | 08:59 |
| seb128 | Saviq, I didn't have phonon from those packages, but still the same issue after installing it | 08:59 |
| Saviq | seb128, hmm | 08:59 |
| cjwatson | I think you need to get a firewall hole explicitly opened to people.c.c | 08:59 |
| seb128 | Saviq, http://paste.ubuntu.com/5886905/ | 08:59 |
| cjwatson | archive-team.internal is soon not going to be on lillypilly, so you cannot rely on having Sergio's home directory available from that | 08:59 |
| ogra_ | hmm, or make /~ available | 08:59 |
| cjwatson | No, that will break soon | 09:00 |
| asac | sergiusens: have you checked my phablet-tools contributions? | 09:00 |
| ogra_ | yeah, ttyped to fast :) | 09:00 |
| Saviq | seb128, libqt5multimedia5-plugins | 09:00 |
| Saviq | seb128, those are the plugins that actually talk to different backends | 09:00 |
| gema | ogra_: I saw you kicked the build yesterday night in the end, what status did you and andy leave the tests ? | 09:00 |
| gema | ogra_: everything moved to pending? | 09:00 |
| ogra_ | gema, nope, still everything failing | 09:01 |
| seb128 | Saviq, that fixes it, thanks! | 09:01 |
| seb128 | Saviq, shouldn't qtdeclarative5-qtmultimedia-plugin depends on it? | 09:01 |
| ogra_ | 20130718 should come out soon (daily job) | 09:01 |
| gema | ogra_: yes, I see that, but are they running on the pending images? | 09:01 |
| cjwatson | I mean, /~user/ is available from archive-team.internal right now - but we *will* be breaking that soon and I'll have no sympathy at all if you relied on it :-) | 09:01 |
| Saviq | seb128, not really | 09:01 |
| ogra_ | lets see, the former image night have hit a race with updating utah | 09:01 |
| Saviq | seb128, as you could provide a different set of backends | 09:01 |
| Saviq | seb128, as we do on the device | 09:01 |
| cjwatson | So I think opening a firewall hole is the right answer at the moment | 09:01 |
| ogra_ | gema, if i can belive the dashboard, they are | 09:02 |
| Saviq | seb128, but it should be recommended (and I do have it installed, not manually) | 09:02 |
| gema | ogra_: the dashboard doesn't lie (yet) | 09:02 |
| gema | ogra_: thanks | 09:02 |
| seb128 | Saviq, ubuntu-ui-toolkit-examples qtmultimedia5-examples mediaplayer-app recommends it | 09:02 |
| Saviq | seb128, yeah | 09:03 |
| Saviq | seb128, we might need to fix something there indeed | 09:03 |
| seb128 | Saviq, should ubuntu-system-settings do the same? | 09:03 |
| ogra_ | cjwatson, where does /seeds live on that machine ? we could probably just have another /click there | 09:03 |
| seb128 | Saviq, seems wrong to recommends it, especially if that's wrong for the touch image | 09:03 |
| Saviq | seb128, yeah indeed | 09:03 |
| cjwatson | Well, it's in ~ubuntu-archive/public_html/ | 09:03 |
| seb128 | Saviq, seems like we should have a virtual "qtmedia-backend" provided by libqt5multimedia5-plugins and whatever we use on touch and recommends that | 09:04 |
| seb128 | Saviq, anyway, if that works out of the box on the touch image no big deal | 09:04 |
| cjwatson | But I'm not sure that's going to be at all convenient for you to update | 09:04 |
| seb128 | Saviq, thanks for the help ;-) | 09:04 |
| Saviq | seb128, cheers | 09:04 |
| cjwatson | I dunno, I guess we could create a mirror job | 09:04 |
| ogra_ | so cant we just link the dir (or copy it) for now .... if the machines change there will be changes needed anyway and we also dont plabn to keep it in a homedir i think | 09:04 |
| cjwatson | How does Sergio update his repository at the moment? | 09:05 |
| cjwatson | Just manually? | 09:05 |
| ogra_ | hmm, good question | 09:05 |
| cjwatson | Hmm, click_copy.py | 09:06 |
| ogra_ | might be cronned, i'm not sure | 09:06 |
| tester56 | As ubuntu touch relies for drivers partly on android stuff: google has reportedly a backdoor to access phones ... will this also be part of ubuntu touch? | 09:06 |
| cjwatson | Not sure of the proper arguments though | 09:06 |
| asac | ogra_: so what happened to mako? is that completely broken now? | 09:07 |
| ogra_ | asac, i have no idea, lets please wait for 18 to come out, it might have been that utah was upgraded in the middle of the 17.2 runs or some such | 09:07 |
| ogra_ | i'm not sure the fixes made it in time for that image | 09:08 |
| asac | when is 18 | 09:08 |
| asac | ? | 09:08 |
| asac | can we produce one now? | 09:08 |
| ogra_ | sat in the queue when i looked 1h ago, lets see | 09:08 |
| ogra_ | still, server is building atm .... should be done soon | 09:09 |
| pitti | hey all | 09:09 |
| cjwatson | Yeah, it's waiting to build, can't speed it up any more | 09:09 |
| asac | lets overclock :) | 09:09 |
| asac | the panda | 09:09 |
| asac | ;-P | 09:09 |
| asac | cant be more flaky anyway | 09:10 |
| asac | panda on 4GhZ with big gamer FAN | 09:10 |
| pitti | right after booting on the n4, rild permanently uses > 80% CPU and ofonod takes > 10% (no wonder it runs out of juice so quickly); that's with an unlocked SIM card; can somebody confirm? | 09:10 |
| ogra_ | yeah, but we'll need a fridge for it then | 09:10 |
| didrocks | asac: I propose water cooling :) | 09:10 |
| asac | i guess you need both :) | 09:10 |
| ogra_ | pitti, what takes 80% CPU ? | 09:10 |
| cjwatson | http://people.canonical.com/~ubuntu-archive/livefs-build-logs/saucy/ubuntu-server-omap4/latest/livecd-20130718-armhf.out <- waiting for that to finish | 09:10 |
| pitti | ogra_: rild, as I said | 09:10 |
| ogra_ | oh | 09:10 |
| cjwatson | (the logs only mirror every hour, so don't read too much into that) | 09:10 |
| * ogra_ was looking directly on cadejo | 09:11 | |
| cjwatson | Yep | 09:11 |
| pitti | ogra_: I was about to strace it to find out what these two actually do when I do a call, but ofono constantly spews thousands of polls, reads, and writes even when not doing (visually) anything | 09:12 |
| ogra_ | pitti, better wait for tony, its his area of expertise ... we have a bug open for rild issues though, just trying to find it | 09:12 |
| pitti | killing ofonod respawns it with the same result | 09:12 |
| pitti | ogra_: do you get the same? (mostly just looking for a confirmation here, not debugging yet) | 09:13 |
| ogra_ | pitti, i dont have a SIOM | 09:13 |
| ogra_ | *SIM | 09:13 |
| ogra_ | and i only have a galaxy nexus ... they use different rilds | 09:14 |
| ogra_ | pitti, bug 1197656 | 09:14 |
| ubot5 | bug 1188404 in touch-preview-images "duplicate for #1197656 ofono between 30% and 40% CPU usage when roaming" [Critical,In progress] https://launchpad.net/bugs/1188404 | 09:14 |
| asac | ogra_: which fixes? | 09:15 |
| ogra_ | asac, dbus | 09:15 |
| asac | ogra_: ok... that might prevent installing with phablet-flash? | 09:15 |
| ogra_ | no | 09:15 |
| asac | just checking because thats what we see | 09:15 |
| ogra_ | what do you see ? | 09:16 |
| ogra_ | and wghgat do you run exactly | 09:16 |
| asac | ogra_: did you run phablet-flash --pending on mako on .2? | 09:16 |
| ogra_ | *what | 09:16 |
| ogra_ | asac, no mako | 09:16 |
| pitti | ogra_: ah, thanks; I think I have something else, I'll file a bug | 09:16 |
| ogra_ | only maguro here | 09:16 |
| asac | ogra_: see see phablet-flash falling over on mako | 09:16 |
| asac | we see | 09:16 |
| ogra_ | popey, ^^^ | 09:16 |
| ogra_ | asac, i think popey tested yesterday evening | 09:17 |
| cjwatson | Does anyone know what IP address lillypilly might be able to use to contact the PS Jenkins? http://10.97.2.10:8080/ doesn't appear to work from there | 09:17 |
| * ogra_ doesnt | 09:17 | |
| pitti | ogra_: acually no, I just see more debug messages; I think it's the same, I'll try the fix | 09:18 |
| pitti | ogra_: thanks for pointing out! | 09:18 |
| asac | ogra_: 11:07 < ogra_> asac, i have no idea, lets please wait for 18 to come out, it might have been that utah was upgraded in the middle of the 17.2 runs or some such | 09:18 |
| ogra_ | :) | 09:18 |
| asac | oops | 09:18 |
| asac | ogra_: --pending seems to download 16 image | 09:18 |
| asac | not 17.2 | 09:18 |
| asac | can this be true? | 09:18 |
| asac | maybe our /pending doesnt work? | 09:18 |
| ogra_ | ogra@chromebook:~/branches/livecd-rootfs-2.161$ get-touch-img-ver -p | 09:18 |
| ogra_ | 20130717.2 | 09:18 |
| ogra_ | that script uses the same code phablet-flash uses | 09:19 |
| ogra_ | so it should pick up 17.2 | 09:19 |
| asac | ogra_: ok here i see it goes for 17.2 | 09:20 |
| cjwatson | cdimage@nusakan:~$ ls -l cdimage/www/full/ubuntu-touch/daily-preinstalled/pending | 09:20 |
| cjwatson | lrwxrwxrwx 1 cdimage cdimage 10 Jul 17 23:44 cdimage/www/full/ubuntu-touch/daily-preinstalled/pending -> 20130717.2 | 09:20 |
| ogra_ | and the phablet-flash change from tonight seems to only touch test code | 09:20 |
| popey | ogra_: i flashed mako an hour ago | 09:20 |
| ogra_ | popey, with 17,2 i suppose | 09:20 |
| popey | yes | 09:20 |
| ogra_ | great | 09:20 |
| ogra_ | asac, fix the proxy :P | 09:21 |
| asac | good point | 09:21 |
| popey | sound indicator is empty | 09:21 |
| ogra_ | 18 started ... | 09:22 |
| ogra_ | popey, give it some time | 09:22 |
| ogra_ | there is a bug open for that | 09:22 |
| ogra_ | bug 1181299 | 09:22 |
| ubot5 | bug 1181299 in touch-preview-images "Sound indicator takes a while to load, not displaying the content sometimes" [Critical,Confirmed] https://launchpad.net/bugs/1181299 | 09:22 |
| popey | 09:23:42 up 57 min, 0 users, load average: 1.56, 1.63, 1.57 | 09:23 |
| popey | is that long enough for it to appear? ☻ | 09:23 |
| asac | ogra_: http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/ | 09:31 |
| asac | go there | 09:31 |
| asac | ogra_: the .img fils are timestamped 16 Jul | 09:31 |
| ogra_ | asac, i'm always there :P | 09:31 |
| asac | is that normal? | 09:31 |
| ogra_ | asac, thats android stuff | 09:32 |
| ogra_ | comes from a jankins build that runs before | 09:32 |
| asac | do we run that daily? | 09:32 |
| ogra_ | yep | 09:33 |
| ogra_ | we will soon start using packages for that bit | 09:33 |
| ogra_ | xnox is nearly there | 09:33 |
| ogra_ | then and android buits wont be daily anymore and you might see even older dates | 09:34 |
| ogra_ | (we could build the package daily indeed, but since it only needs rebuilds when it actually changes that would be a waste) | 09:35 |
| xnox | ogra_: correct. last stretch is to test if i can disable ICU from libmedia, and then it will be ready for security-team review for archive upload. | 09:35 |
| xnox | ogra_: well, it does at the moment still embed kernel, platform-api, hybris. so when those change, my package needs a rebuild. But the loger term plan is to combine those three at livebuild stage, to avoid this dependency triggered rebuilt. | 09:36 |
| ogra_ | exactly | 09:37 |
| pitti | ogra_: yay, a tame and working ofonod now \o/ | 09:55 |
| ogra_ | :) | 09:56 |
| === vrruiz_ is now known as rvr | ||
| === schwuk_away is now known as schwuk | ||
| davmor2 | Morning all | 10:02 |
| popey | Just | 10:03 |
| davmor2 | popey: tis always morning on the tinterwebz | 10:04 |
| davmor2 | popey: is https://docs.google.com/a/canonical.com/spreadsheet/ccc?key=0ArLs7UPtu-hJdDZDNWliMmV1YUJ3Zk1pQlpDdGp4VFE#gid=0 actually up-to-date cause according to it the last time the phone were tested was the 12, there was no image for maguro since the 16 yet mako had one for 17.2, manta just doesn't have a build and the n7 is way back | 10:16 |
| davmor2 | popey: or is this something that is manually meant to be updated and you're the only one doing it for the n4 ;) | 10:18 |
| popey | correct | 10:19 |
| popey | you can request edit access | 10:19 |
| popey | pat owns it | 10:19 |
| popey | davmor2: what do you mean "the n7 is way back" ? you mean it's more broken? | 10:21 |
| davmor2 | meh ignore me I saw it as 11 not 17.1 | 10:21 |
| davmor2 | popey: I only gave it a cursory glance flicking between screens :) | 10:25 |
| davmor2 | popey: odd question is there an option on the page to request write permission? if so where please | 10:25 |
| popey | top right i think | 10:26 |
| popey | click share | 10:26 |
| ogra_ | 20130718 is up | 10:29 |
| * ogra_ goes to watch the dashboard | 10:30 | |
| davmor2 | ogra_: out of interest why is current looking such a mess? Bits are the 16th and others the 12? | 10:32 |
| ogra_ | ? | 10:33 |
| ogra_ | ah, i see what you mean, probably because the android build from the 12th was good enough :) | 10:34 |
| davmor2 | okay | 10:34 |
| ogra_ | theoretically they should ahve been from the 15th for the 16th image, might be that there was a jenkins issue | 10:34 |
| ogra_ | (weekend etc ...) | 10:35 |
| * popey flashes again | 10:38 | |
| * ogra_ wonders when the dashboard will notice there is a new build | 10:40 | |
| davmor2 | ogra_, popey: My pip for the photos I took got wiped is the data source not stored in home? | 10:41 |
| timp | Downloading http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/20130718/saucy-preinstalled-touch-armhf.zip | 10:41 |
| timp | let's see if that works :) | 10:41 |
| popey | davmor2: bug 1202559 | 10:42 |
| ubot5 | bug 1202559 in touch-preview-images "Welcome screen doesn't show history beyond today" [Undecided,New] https://launchpad.net/bugs/1202559 | 10:42 |
| ogra_ | davmor2, no idea where thats stored ... file a bug, we definitely should preserve that data, no matter where it lives | 10:42 |
| popey | davmor2: tap it to refresh it | 10:42 |
| * popey filed that this morning | 10:42 | |
| ogra_ | :) | 10:42 |
| davmor2 | popey: nothing | 10:42 |
| popey | taken any photos today? | 10:43 |
| davmor2 | popey: nope | 10:43 |
| popey | well thats my bug then | 10:43 |
| popey | it only shows what you did today | 10:43 |
| popey | confirm pls | 10:43 |
| === dandrader is now known as dandrader|afk | ||
| davmor2 | popey: done | 10:46 |
| popey | ogra_: 20130718 doesn't look good. nothing on screen but top shows unity8 running... power button doesn't wake display | 10:46 |
| ogra_ | i'll take a look once my rsync finished | 10:47 |
| popey | http://paste.ubuntu.com/5887142/ is my dmesg | 10:49 |
| davmor2 | popey, ogra_ tis working here on maguro | 10:49 |
| davmor2 | oh hang on though I seemed to of regained the music artwork | 10:49 |
| * popey reboots it | 10:49 | |
| davmor2 | ogra_: is the demo music stuff meant to be gone? | 10:50 |
| popey | works on second boot | 10:50 |
| popey | davmor2: takes a while to appear (well did yesterday's image) | 10:50 |
| popey | i see it here on 20130718 on mako | 10:50 |
| davmor2 | popey: ah hang on my phablet-flash seems to of gotten 20130716 that would be why | 10:51 |
| popey | plum | 10:51 |
| davmor2 | popey: is there a command to get latest? | 10:51 |
| popey | phablet-flash --pending | 10:51 |
| davmor2 | popey: ta | 10:51 |
| popey | gets from...... /pending | 10:51 |
| asac | ogra_: so 18 is not good? | 10:52 |
| popey | ok, 20130718 is good for me on mako | 10:54 |
| popey | odd that it's not showing up at http://reports.qa.ubuntu.com/smokeng/ | 10:55 |
| kalikiana | not sure battery ran out, I'm suspecting because when I wanted to shut it down yesterday it was again stuck in "chargin" mode without usb cable | 10:57 |
| popey | kalikiana: i found unplugging/replugging the usb cable a few times "woke" the phone up | 10:58 |
| kalikiana | popey, I saw the it pretended to charge after unplugging, and even when I long-pressed power - I thought it might just still be shutting down and left it over night, now it's empty | 11:00 |
| kalikiana | I get it also stuck in "charging" sometimes after plugging in USB with it being off before | 11:01 |
| kalikiana | so something about that charge logic is fishy | 11:01 |
| kalikiana | only adb then can fix it | 11:01 |
| === dandrader|afk is now known as dandrader | ||
| ogra_ | asac, dunno, still 10min to rsync for me, popey seems to have no luck though | 11:03 |
| === MacSlow is now known as MacSlow|lunch | ||
| popey | ogra_: it's fine. | 11:05 |
| davmor2 | ogra_, popey: it looks fine here on the Magoru | 11:05 |
| ogra_ | just took long at first boot as usual ? | 11:06 |
| === schwuk is now known as schwuk_away | ||
| popey | yes | 11:07 |
| ogra_ | asac, ^^^ | 11:07 |
| ogra_ | seems it is good | 11:07 |
| ogra_ | ah, an dthere is 18 on http://reports.qa.ubuntu.com/smokeng/saucy/ | 11:08 |
| popey | yeah, just appeared | 11:09 |
| davmor2 | hmmm that'll be a bug then, my conversations list seems to of magically change chronological order | 11:09 |
| Saviq | ogra_, q: any idea how to remotely call `stop unity8`? `ssh stop unity8` as well as `ssh bash -ics 'stop unity8'` come back with "unknown job: unity8" | 11:12 |
| ogra_ | initctl stop unity8 | 11:12 |
| ogra_ | as phablet user | 11:12 |
| popey | will try that next time | 11:13 |
| davmor2 | I can foresees a time when initctl stop unity8 will become ogra_ nick | 11:13 |
| ogra_ | popey, thats a step the automatic tests do anyway | 11:13 |
| ogra_ | if that fails, the suite fails | 11:13 |
| Saviq | ogra_, http://pastebin.ubuntu.com/5887190/ :/ | 11:13 |
| Saviq | ogra_, it works when I *log in*, but not if I run a command | 11:14 |
| Saviq | ogra_, bash -ic doesn't help either | 11:14 |
| ogra_ | Saviq, try with ssh phablet@maguro sh -c 'initctl stop unity8' | 11:14 |
| Saviq | ogra_, been there, done that - same thing http://pastebin.ubuntu.com/5887202/ | 11:17 |
| Saviq | ogra_, we're even using "bash -ic" so that we get all the env vars | 11:18 |
| === greyback is now known as greyback|lunch | ||
| ogra_ | why did you add tehse quotes | 11:19 |
| davmor2 | hmm unity8 17.8 % cpu and 20.7%mem, with init not far behind at 12.2 and mem .3 and surfacflinger 3rd with 7.9 and 5.9 | 11:19 |
| ogra_ | Saviq, also really, use initctl | 11:19 |
| Saviq | ogra_, ok, otherwise it complained that it was missing the name | 11:20 |
| ogra_ | which name ? unity8 ? | 11:20 |
| ogra_ | gimme a sec, my phone is just flashing | 11:21 |
| Saviq | ogra_, http://pastebin.ubuntu.com/5887214/ | 11:21 |
| Saviq | ogra_, thanks | 11:21 |
| ogra_ | i know it works with adb ... since thats what we use in the testsuite | 11:22 |
| ogra_ | but there is no reason it shouldnt work the same via ssh | 11:22 |
| ogra_ | adb shell sudo -u phablet -i sh -c 'initctl stop unity8' | 11:23 |
| ogra_ | iirc | 11:23 |
| Saviq | ogra_, thanks, that will work, too | 11:24 |
| ogra_ | ogra@chromebook:~$ adb shell sudo -u phablet -i sh -c 'initctl stop unity8' | 11:25 |
| ogra_ | unity8 stop/waiting | 11:25 |
| ogra_ | ogra@chromebook:~$ adb shell sudo -u phablet -i sh -c 'initctl start unity8' | 11:25 |
| ogra_ | unity8 start/running, process 1227 | 11:25 |
| ogra_ | yeah, that works here | 11:25 |
| * ogra_ doesnt get why ssh doesnt | 11:25 | |
| timp | why does image 20130718 have an old UITK version? | 11:29 |
| timp | old == of some days ago | 11:29 |
| diwic | uhm, ueventd is eating 100% CPU here | 11:30 |
| mallorquin | hello everybody i have one question! | 11:31 |
| ogra_ | diwic, reboot | 11:32 |
| ogra_ | its a race | 11:32 |
| diwic | ok | 11:32 |
| ogra_ | Saviq, hmm, so if you dont log in to ssh the dbus address isnt exported ... seems /etc/profile.d/ isnt processed then | 11:32 |
| ogra_ | not sure why that would affect initctl though | 11:33 |
| cjwatson | timp: It seems to have the current version in saucy | 11:33 |
| cjwatson | i.e. 0.1.46+13.10.20130716-0ubuntu1 | 11:33 |
| cjwatson | timp: 0.1.46+13.10.20130718-0ubuntu1 only *just* got copied to saucy-proposed, so will presumably land in saucy soon | 11:34 |
| mallorquin | I installing on nexus7 grouper, but afther install Sourcer version with adb tools " version -l " say me " Nexus 4 mako occam " its correct? | 11:34 |
| timp | cjwatson: ok, thanks. So it is still automatic and all I need is a little patience. | 11:35 |
| ogra_ | Saviq, ssh phablet@maguro "sudo -u phablet -i initctl stop unity8" | 11:35 |
| ogra_ | Saviq, that one will work | 11:35 |
| ogra_ | (will force /etc/profile sourcing) | 11:35 |
| ogra_ | cjwatson, any idea why ssh doesnt do that if you dont actually log in ? | 11:36 |
| davmor2 | popey: if you do top on your phone and get the keyboard out of the way is the last entry cut in too by the bottom to the terminal? | 11:36 |
| Saviq | ogra_, interesting | 11:36 |
| didrocks | ogra_: once the apps migrates to the release pocket, you can maybe trigger a new image build to check the smoke tests are done | 11:37 |
| user82 | what happened to "whosthere". is there an alternative? | 11:37 |
| didrocks | FYI, all AP tests we run on desktop as part of dailies pass now | 11:37 |
| === chihchun is now known as chihchun_afk | ||
| user82 | whatsapp is a killer criteria for many people to use ubuntu touch | 11:37 |
| popey | davmor2: http://popey.com/~alan/device-2013-07-18-123812.png | 11:38 |
| davmor2 | popey: http://ubuntuone.com/0AhxMGyB95SGqxJkFa8MOQ p.s. thanks for your screenshot script :) | 11:38 |
| davmor2 | popey: now slide the keyboard down | 11:39 |
| popey | ☻ | 11:39 |
| davmor2 | popey: it used to be the same for the keyboard but that looks like it got fixed | 11:39 |
| ogra_ | didrocks, oh, which apps ? | 11:39 |
| didrocks | ogra_: the gallery_app in particular | 11:39 |
| cjwatson | ogra_: /etc/profile is read for me when I "ssh localhost" in a container | 11:39 |
| ogra_ | ok | 11:39 |
| popey | davmor2: http://popey.com/~alan/device-2013-07-18-124002.png | 11:40 |
| * popey stabs rild eating the cpu alive | 11:40 | |
| davmor2 | popey: wow that is really high | 11:40 |
| mallorquin | the nexus 7 have de SO of nexus 4 ? | 11:41 |
| === greyback|lunch is now known as greyback | ||
| popey | bug 1188404 | 11:41 |
| ubot5 | bug 1188404 in touch-preview-images "ofono between 30% and 40% CPU usage when roaming" [Critical,In progress] https://launchpad.net/bugs/1188404 | 11:41 |
| ogra_ | cjwatson, ssh phablet@192.168.2.114 "env"|grep DBUS doesnt get me the DBUS address ... logging in gives it to me ... the address is exported from an /etc/profile.d ssnippet | 11:41 |
| davmor2 | popey: thanks so it look like it is a magoru issue only then although I might try on the n7 too | 11:41 |
| popey | I can do that now | 11:41 |
| cjwatson | ogra_: It doesn't source /etc/profile if you supply a command | 11:41 |
| ogra_ | ah | 11:42 |
| cjwatson | ogra_: Because then you aren't in an interactive login shell | 11:42 |
| ogra_ | right | 11:42 |
| popey | davmor2: http://popey.com/~alan/device-2013-07-18-124240.png | 11:42 |
| davmor2 | popey: ta so it look like it is just the lower res on the maguro then :) | 11:43 |
| popey | yeah | 11:44 |
| ogra_ | popey, i think the branch for the ofono fix is already approved, should land today | 11:44 |
| ogra_ | come on dashboard ... do something | 11:44 |
| popey | great | 11:45 |
| popey | I am worried about going to the US next week. | 11:45 |
| popey | Don't want my phone to go nuts over 3g | 11:45 |
| ogra_ | yeah, ask for a salary raise in advance | 11:45 |
| popey | need to be able to reliably kill 3g roaming | 11:45 |
| popey | heh | 11:45 |
| davmor2 | popey: you know the power switch on the side, that'll do it ;) | 11:49 |
| om26er | phablet-flash -b is just not working :/ | 11:56 |
| ogra_ | fastboot -w complaining ? | 11:57 |
| om26er | it downloads everything, reboots to the bootloader mode but after that it keeps on < waiting for device > | 11:57 |
| ogra_ | i think there was a fix that didnt land yet | 11:57 |
| ogra_ | wait for sergiuiens | 11:57 |
| om26er | and we just killed a phone :p | 11:57 |
| ogra_ | you cant kill them :) | 11:58 |
| om26er | ogra_, fastboot -w erases user data | 11:58 |
| ogra_ | yup, but has issues doing so at times | 11:58 |
| om26er | ogra_, it kind of for this case, since its in the lab ;) | 11:58 |
| ogra_ | which means that sergio changed -b completely | 11:58 |
| om26er | killed for the day is appropriate | 11:58 |
| ogra_ | ah | 11:58 |
| ogra_ | you need a remote power switch :) | 11:59 |
| om26er | well I am still in the bootloader mode, though I don't know how to flash from here | 11:59 |
| ogra_ | do you have console access to the machine that it is attached to ? | 12:00 |
| om26er | ogra_, yes | 12:00 |
| ogra_ | sudo as well ? | 12:00 |
| om26er | ogra_, yep | 12:00 |
| om26er | and very high speed internet as well, we anything needs to get downloaded | 12:00 |
| ogra_ | well, it should work to sudo fastboot flash *-recovery.img and sudo fastboot reboot | 12:01 |
| om26er | *if | 12:01 |
| ogra_ | err | 12:01 |
| ogra_ | sudo afstboot reboot recovery that is | 12:01 |
| ogra_ | *fast | 12:01 |
| om26er | ogra_, we have adb access in recovery mode? | 12:01 |
| ogra_ | ... unless the device isnt actually in fastboot mode indeed | 12:02 |
| ogra_ | om26er, indeed :) | 12:02 |
| ogra_ | not in bootloader mode, but in recovery | 12:02 |
| om26er | ogra_, it seems we cannot boot into recovery from commandline (while in fastboot mode) | 12:03 |
| ogra_ | hmm, you should be able to | 12:03 |
| * ogra_ tries locally | 12:03 | |
| fginther | sil2100, ping | 12:04 |
| ogra_ | om26er, oh, right, it can only reboot to bootloader or normal system ... odd | 12:04 |
| === alan_g is now known as alan_g|lunch | ||
| davmor2 | popey: daft question time, how do I add a feed to the rss reader, there doesn't seem to be an options page like there was before | 12:08 |
| * om26er could flash back to android but there is still 'android debugging' disabled on first run :/ | 12:08 | |
| ogra_ | om26er, there is a slightly fiddly way that could work ... pull the boot.img modify its cmdline and add "break=bottom", flash with fastboot to boot, trigger a normal boot and you should get an adb shell inside the initrd ... from there adb reboot recovery should work | 12:08 |
| om26er | ogra_, I already rebooted :p and it showed google logo, then the battery icon on the screen and screen wen't off. Now no adb or fastboot for that device. | 12:11 |
| om26er | BUT I have a mako as well :p | 12:11 |
| ogra_ | ah, sad | 12:11 |
| ogra_ | gema, are we sure the utah dbus fixes landed yesterday ? the tests seem to still have the same dbus errors | 12:12 |
| ogra_ | asac, ^^^ | 12:12 |
| om26er | ogra_, you means something like this: DBusException: org.freedesktop.DBus.Error.NoServer: Failed to connect to socket /tmp/dbus-i0wV0w6JjV: Connection refused | 12:13 |
| om26er | ? | 12:13 |
| ogra_ | om26er, yeah | 12:13 |
| om26er | ogra_, that happens when we restart ubuntu-touch-session during the autopilot test run | 12:13 |
| ogra_ | the new DBUS_SESSION_BUS_ADDRESS code should have fixed that | 12:13 |
| ogra_ | oh, why would you do that ? | 12:14 |
| om26er | ogra_, we edit /etc/device-services to remove unity from it, then restart ubuntu-touch-session to kill unity | 12:14 |
| ogra_ | i thought you only stop unity8 | 12:14 |
| sergiusens | ogra_: wait, what? | 12:14 |
| ogra_ | om26er, all that was fixed in quite some effort yesterday i thought | 12:14 |
| sergiusens | ogra_: why would we restart everything now? | 12:15 |
| om26er | ogra_, there is a merge proposal for phablet-tools pending which switches to initctl stop/start unity8 | 12:15 |
| ogra_ | om26er, it should only restart unity and should have the proper session bus address | 12:15 |
| ogra_ | restarting the whole sessuon will kill the session bus indeed and the exported address will be wrong until you re-login | 12:16 |
| sergiusens | om26er: utah imports the DBUS_SESSION_BUS_ADDRESS from the file directly, not a good idea because it's one more point of failure | 12:16 |
| ogra_ | om26er, pending ? i thought that was in already | 12:16 |
| sergiusens | better to login | 12:16 |
| om26er | ogra_, it was working a few days (weeks) ago due to some reason | 12:16 |
| gema | ogra_: I don't know | 12:17 |
| === dandrader is now known as dandrader|lunch | ||
| ogra_ | it hasnt worked since we switched to upstart sessions | 12:17 |
| om26er | ogra_, I am reviewing and I was flashing the device with 'phablet-flash -b' for that | 12:17 |
| ogra_ | and yesterday at least 4 people worked on getting these fixes into utah | 12:17 |
| ogra_ | i dont get how they could not land there | 12:17 |
| om26er | sergiusens, phablet-flash -b does not work. tried on two devices | 12:17 |
| sergiusens | om26er: ? you may have a broken bootloader | 12:18 |
| popey | om26er: what type of device, i can test here | 12:18 |
| om26er | popey, both were maguro | 12:18 |
| sergiusens | broken as incorrect | 12:18 |
| popey | ah, only have mako here | 12:18 |
| ogra_ | om26er, when does utah run phablet-flash -b | 12:18 |
| om26er | sergiusens, the device was already running ubuntu touch | 12:18 |
| ogra_ | i dont see that in the code | 12:18 |
| sergiusens | om26er: can you be more specific on how it doesn't work? | 12:19 |
| om26er | ogra_, I don't think it does but I am talking about this MR https://code.launchpad.net/~sergiusens/phablet-tools/test_update/+merge/175355 (where I was asked to phablet-flash -b) | 12:19 |
| om26er | sergiusens, http://paste.ubuntu.com/5887349/ | 12:20 |
| ogra_ | om26er, i'm talking about http://reports.qa.ubuntu.com/smokeng/saucy/ which blocks us since days to release an image | 12:20 |
| ogra_ | and the code in there is supposed to have been fixed yesterday | 12:20 |
| gema | ogra_: om26er is talking about the same thing, I have asked him to look into some failures | 12:20 |
| gema | ogra_: didn't you and doanac got that fixed yesterday? | 12:20 |
| sergiusens | om26er: your fastboot udev rules are the cause there | 12:20 |
| gema | ogra_: or who were you dealing with? | 12:21 |
| ogra_ | gema, doanac told me it worked for him in clocal tests and he was waiting for it to land in the utah main branch when i went to bed | 12:21 |
| om26er | sergiusens, what should I change? | 12:21 |
| sergiusens | om26er: gema I'm logging into that server and checking | 12:21 |
| gema | ogra_: ok, same he told me | 12:21 |
| sergiusens | om26er: need to see the id's | 12:21 |
| gema | sergiusens: thanks | 12:21 |
| ogra_ | so i was expecting it to be there by now | 12:21 |
| gema | ogra_: me too | 12:21 |
| ogra_ | but apparently it still fails with dbus errors | 12:22 |
| ogra_ | and now om26er tells me there are other fixes needed than utah | 12:22 |
| om26er | ogra_, I think in UTAH we use "phablet-test-run -n" which kills and restarts ubuntu-touch-session | 12:22 |
| * ogra_ is confused | 12:22 | |
| gema | doanac: ping | 12:23 |
| ogra_ | ogra@chromebook:~/branches/utah$ grep phablet-test examples/run_utah_phablet.py | 12:23 |
| ogra_ | ogra@chromebook:~/branches/utah$ | 12:23 |
| ogra_ | doesnt look like | 12:23 |
| sergiusens | ogra_: there are two different issues being discussed here... | 12:23 |
| ogra_ | and i also dont see code for phablet-flash -b | 12:23 |
| ogra_ | sergiusens, phew, thanks | 12:24 |
| === MacSlow|lunch is now known as MacSlow | ||
| sergiusens | ogra_: and two different infrastructures | 12:24 |
| sergiusens | not sure why they are being mixed up | 12:24 |
| om26er | well actually both are interlinked. | 12:24 |
| ogra_ | yeah, ok | 12:24 |
| ogra_ | the logs i see only use run_utah_phablet.py | 12:25 |
| om26er | ogra_, I don't see there is any way tests can run in UTAH without phablet-test-run they might be forking it in their scripts | 12:25 |
| sergiusens | om26er: not really, it fails on ps-android-sandybridge for a reason totally unrelated to why it failed on utah | 12:25 |
| ogra_ | and in these logs the dbus error occurs | 12:25 |
| sergiusens | om26er: they don't use phablet-test-run!!!! | 12:25 |
| om26er | sergiusens, i am flashing the device because I want to get that MR tested, phablet-tools need to be fixed for autopilot tests to run successfully on the device | 12:26 |
| sergiusens | om26er: yeah, but it is not related to utah | 12:26 |
| sergiusens | people don't know everything about all the infrastructure diffs and the case of discussion was utah | 12:26 |
| gema | sergiusens: every time something fails it is either "jenkins" or "utah" | 12:27 |
| gema | sergiusens: I agree we need to start being more specific | 12:27 |
| kalikiana | hm is it a bug that phablet-flash apparently re-flashes even if there's no new image? | 12:28 |
| kalikiana | it seems a little pointless | 12:28 |
| ogra_ | gema, is there any way to know which utah version is currently running ? | 12:29 |
| kalikiana | just noticed in the log afterwards that it didn't use any new files but still went ahead | 12:29 |
| ogra_ | 0.14.2ubuntu1 seems to have all fixes | 12:29 |
| timp | kalikiana: yes it seems to re-flash for me also | 12:29 |
| popey | i think that's how it's designed | 12:29 |
| timp | would be nice if phablet-flash asks you what to do in that case | 12:30 |
| popey | but it can't tell what version is on the device, or doesn't test that | 12:30 |
| === schwuk_away is now known as schwuk | ||
| sergiusens | timp: with image based upgrades this won't be a problem anymore | 12:30 |
| sergiusens | om26er: this is your problem on ps-android-sandybridge http://pastebin.ubuntu.com/5887385/ | 12:31 |
| kalikiana | sergiusens, what does that mean exactly? I thought it's already an image and a separate boot loader image | 12:31 |
| sergiusens | om26er: if you use sudo, it works, which means missing udev rules | 12:32 |
| timp | sergiusens: ETA for image-based upgrades? :) | 12:32 |
| sergiusens | timp: stgraber will send a call for testing on Friday (tomorrow). I'm making the mods to phablet-flash as soon as firefighting stops | 12:33 |
| om26er | sergiusens, so we need the udev rules updates? but how :p | 12:33 |
| timp | sergiusens: so soon. awesome :) | 12:34 |
| ogra_ | popey, | 12:34 |
| ogra_ | ogra@chromebook:~/branches/livecd-rootfs$ adb shell cat /var/log/installer/media-info | 12:34 |
| ogra_ | Ubuntu Saucy Salamander (development branch) - armhf (20130718) | 12:34 |
| ogra_ | popey, it can :) | 12:34 |
| sergiusens | kalikiana: https://wiki.ubuntu.com/ImageBasedUpgrades | 12:34 |
| popey | pmcgowan: can you give davmor2 access to the status spreadsheet as he has a maguro and can update for that | 12:36 |
| pmcgowan | popey, sure | 12:36 |
| davmor2 | popey: I requested it via gdocs :) but thanks for the +1 :) | 12:37 |
| sergiusens | ogra_: do you see any typo in this udev rules? http://pastebin.ubuntu.com/5887398/ | 12:37 |
| popey | thanks | 12:37 |
| davmor2 | pmcgowan: thanks | 12:37 |
| pmcgowan | popey, actually you could do it, but done | 12:37 |
| popey | oh, didnt think i had ownership | 12:38 |
| om26er | ogra_, sergiusens so UTAH was indeed using the restart ubuntu-touch-session till yesterday, http://bazaar.launchpad.net/~utah/utah/dev/revision/968 | 12:39 |
| ogra_ | sergiusens, not on first sight, no | 12:40 |
| pmcgowan | popey, as long as you have edit access you can share | 12:40 |
| popey | oh | 12:40 |
| sergiusens | ogra_: what happens with ACTION if the device is always connected? | 12:40 |
| ogra_ | sergiusens, note that i'm not sure if /var/run/ConsoleKit/database exists on logind machines | 12:40 |
| sergiusens | ogra_: this is precise :-) | 12:40 |
| ogra_ | sergiusens, it will change and add once you reboot it | 12:41 |
| sergiusens | saucy is using TAG+=uaccess | 12:41 |
| ogra_ | sergiusens, perfect | 12:45 |
| pmcgowan | ogra_, sergiusens whats the situation with tests not running? sorry if this is redundant | 12:48 |
| sergiusens | pmcgowan: seems utah isn't using an updated version | 12:48 |
| ogra_ | pmcgowan, breaking on dbus still, even though the fixes should be in | 12:48 |
| sergiusens | ... of itself | 12:49 |
| ogra_ | gema, who could find out which version is running on the machine (it would really help if the currently used version would be dumped into the logs) | 12:53 |
| asac | sergiusens: here :) ... who can review/land https://code.launchpad.net/~sergiusens/phablet-tools/test_update/+merge/175355 ? | 12:53 |
| asac | seems utah would need that | 12:53 |
| ogra_ | asac, utah doesnt use the pahblet tests | 12:54 |
| ogra_ | it has its own | 12:54 |
| asac | ogra_: well they use phablet-flash :) | 12:54 |
| asac | anyway, gema said they need it in order to make stuff better asap | 12:54 |
| ogra_ | asac, the redundant equivalent is examples/run_utah_phablet.py in the utah branch | 12:54 |
| asac | so i trust here :) | 12:54 |
| asac | her | 12:54 |
| ogra_ | they use phablet-flash for flashing | 12:55 |
| asac | maybe they want to move to use that instead of their own? don't know the background | 12:55 |
| ogra_ | not for the tests ... read the logs :) | 12:55 |
| asac | she said they need it | 12:55 |
| sergiusens | asac: the blocking on that MR is udev rules | 12:56 |
| ogra_ | + /usr/share/utah/examples/run_utah_phablet.py -s R32D102RPZL --results-dir /home/ubuntu/jenkins/workspace/saucy-touch-manta-smoke-gallery-app-autopilot/clientlogs --skip-install --skip-network --skip-utah -N -l /tmp/gallery-app-autopilot/master.run | 12:56 |
| ogra_ | from https://jenkins.qa.ubuntu.com/job/saucy-touch-manta-smoke-gallery-app-autopilot/10/consoleText | 12:56 |
| asac | sergiusens: whats that? | 12:56 |
| sergiusens | asac: the workstation being tested on doesn't seem to be processing the udev rules correctly | 12:57 |
| ogra_ | asac, /usr/share/utah/examples/run_utah_phablet.py is fixed, but the machine running the tests doesnt seem to use the latest version of it | 12:57 |
| ogra_ | thats the only issue we have with the dashboard atm | 12:57 |
| ogra_ | phablet-tools is totally unrelated | 12:57 |
| ogra_ | phablet-test-run is a redundant copy of run_utah_phablet.py (or the other way round, no idea which one was first) to replicate the same tests locally | 12:59 |
| sergiusens | om26er: gema ogra_ fixes the fastboot issue on sandybridge, there was a custom udev rule overriding the one from android-tools-fastboot | 12:59 |
| ogra_ | oh my | 12:59 |
| * sergiusens thanks udevadm monitor | 12:59 | |
| ogra_ | could we stop hacking up oour production machines ? | 13:00 |
| sergiusens | ogra_: this predates any android-tools-fastboot udev rule existence | 13:01 |
| ogra_ | oh, probably from nx7 desktop tests | 13:01 |
| sergiusens | ogra_: prior ;-) | 13:02 |
| ogra_ | prior we never had a need for such rules :) | 13:02 |
| pmcgowan | sergiusens, ogra_ whats the bottom line, someone is fixing the test box? | 13:05 |
| pmcgowan | dont leave me hangin | 13:05 |
| ogra_ | pmcgowan, i'm waiting for info from gema ... we dont know which utah version is running on the machine | 13:05 |
| sergiusens | pmcgowan: well someone with access to it should | 13:05 |
| pmcgowan | who is that someone? | 13:06 |
| ogra_ | pmcgowan, the fixes were merged last night and it was verified they work in a local test | 13:06 |
| pmcgowan | yes | 13:06 |
| ogra_ | but it doesnt seem that they reached the utah machine | 13:06 |
| pmcgowan | so maybe two issues? utah update and the udev rule being wrong? | 13:07 |
| sergiusens | pmcgowan: doanac I guess, gema would know | 13:07 |
| ogra_ | nope | 13:07 |
| sergiusens | pmcgowan: udev rule being wrong is not a blocker | 13:07 |
| pmcgowan | ok | 13:07 |
| ogra_ | the udev rule is related to phablet-test-run, not related | 13:07 |
| sergiusens | pmcgowan: and fixed btw (I did have access to the conflicting box) | 13:07 |
| pmcgowan | vg | 13:07 |
| pmcgowan | sergiusens, you have access? so are you checking on utah? | 13:08 |
| ogra_ | all we need it to make sure the right utah version is used for the tests | 13:08 |
| pmcgowan | we should have a test for that ;) | 13:09 |
| ogra_ | well, we should have the script just run "dpkg -l utah" as the first thing and dump that to the log | 13:09 |
| ogra_ | we do that on pretty much every other build machine where such info is essential | 13:09 |
| pmcgowan | makes sense | 13:10 |
| sergiusens | pmcgowan: no, no access to utah | 13:11 |
| sergiusens | only logs and launchpad code | 13:12 |
| ogra_ | pmcgowan, full in the hands of QA | 13:13 |
| === chihchun_afk is now known as chihchun | ||
| === dandrader|lunch is now known as dandrader | ||
| === alan_g|lunch is now known as alan_g | ||
| === chihchun is now known as chihchun_afk | ||
| pmcgowan | fginther, can you assist us with this test setup | 13:26 |
| oSoMoN | didrocks: hey, can you remind me again what was the problem with assets under /usr/lib*, apart from lintian complaining about it? | 13:26 |
| didrocks | oSoMoN: it's just duplicated data that aren't arch-specific | 13:27 |
| didrocks | oSoMoN: once things calm down and as I'm seeing everyone coming with their own solution, I would like to have a meeting with you, Kaleo and some other folks to ensure we decide on standard on where to install things | 13:27 |
| didrocks | oSoMoN: just not as long as Mir didn't land to distro please :) | 13:27 |
| fginther | pmcgowan, checking | 13:28 |
| oSoMoN | didrocks: ok, I’m asking because I’m refactoring the browser’s architecture, moving everything that shouldn’t be public out of the plugin and into the app (and consequently a number of QML files and assets are going to be installed under /usr/share/), and in that branch I have only one asset left for the plugin, so I was wondering if it would be acceptable to ship it under /usr/lib*? it’s only one tiny and cute png file ;) | 13:28 |
| ogra_ | fginther, oh, i wonder if its related to your mail :) | 13:28 |
| ogra_ | (that utah didnt land) | 13:29 |
| * ogra_ only saw the mail now | 13:29 | |
| didrocks | oSoMoN: depends on how cute it is… :p more seriously, what's blocking it for not being in usr/share? | 13:29 |
| oSoMoN | didrocks: I guess I could use robru’s trick with symlinks, but it feels overkill for one asset | 13:30 |
| oSoMoN | didrocks: but nothing blocking | 13:30 |
| didrocks | oSoMoN: yeah, I think we should come with a better solution, keep it there for now and we'll discuss afterward how we can handle it | 13:30 |
| oSoMoN | didrocks: ok | 13:30 |
| fginther | ogra_, I doubt utah was impacted (I don't even think we build utah in our jenkins) | 13:31 |
| ogra_ | fginther, ok | 13:31 |
| ogra_ | sad, that would have been so easy :) | 13:32 |
| xnox | oSoMoN: there is nothing wrong with libraries shipping it's data files under /usr/lib/$package/, even at times preffered. E.g. i am sure some gnome components ship gtkbuilder (plugable) files there, similarish story with like icu transliteration tables. | 13:32 |
| sergiusens | oSoMoN: if it's really cute, no one would mnd | 13:33 |
| oSoMoN | sergiusens: that’s what I thought :) | 13:33 |
| xnox | oSoMoN: also all of *.py files are under /usr/lib/ as they are python-runtime dependant, despite being mostly "arch:all" packages. Thus if the plugin is tied with ui description file, than it is preffered to ship them together in /usr/lib. | 13:33 |
| fginther | ogra_, sergiusens pmcgowan, can you restate the problem for jcollado? | 13:33 |
| sergiusens | fginther: seems the utah infra isn't updated with the version that has the fixes for touch | 13:34 |
| ogra_ | jcollado,, fginther, the fixes to utah dont seem to have made it into production | 13:35 |
| davmor2 | popey, pmcgowan: I've updated it and added my initials in brackets to the end I figure it is a clean way to distinguish who has done the update but feel free to remove it if you don't want that in there | 13:35 |
| oSoMoN | xnox: even under arch-specific dirs, like e.g. /usr/lib/x86_64-linux-gnu/ ? | 13:35 |
| xnox | oSoMoN: yes. | 13:35 |
| oSoMoN | xnox: cool, I’m all good then :) | 13:35 |
| oSoMoN | didrocks: ^ | 13:35 |
| jcollado | ogra_: So you mean that the utah package hasn't been updated in a machine? Which one? | 13:35 |
| popey | davmor2: did consider that, but figured it's google docs, with history, so we can see who edited what | 13:36 |
| ogra_ | jcollado, whatever is behind http://reports.qa.ubuntu.com/smokeng/saucy/ | 13:36 |
| xnox | oSoMoN: arguably, since I doubt that file is different between architectures you can ship it in /usr/lib/webpluginubuntuqml/plugin.qml | 13:36 |
| xnox | oSoMoN: as the packages are co-installable if "clashing" files are identical md5sum =) | 13:36 |
| ogra_ | jcollado, see the touch tests there, they still expose the same dbus errors which should be fixed | 13:36 |
| xnox | oSoMoN: thus avoiding duplication if you have both armhf & amd64 plugin installed =) | 13:37 |
| didrocks | xnox: qml files are the same between archs | 13:37 |
| didrocks | xnox: and I doubt we want to ship all assets in multiple packages | 13:37 |
| didrocks | xnox: assets should be in usr/share/ | 13:37 |
| didrocks | which is the discussion here | 13:37 |
| xnox | didrocks: in general, I agree. but there can be exception. | 13:37 |
| oSoMoN | xnox: the issue is that the assets are expected to be in the same dir as the QML files, and the QML files should be in the same dir as the QML plugin, which is an architecture-dependent *.so | 13:38 |
| xnox | oSoMoN: what are you shipping again? if it's an icon, shouldn't it be provided by the theme and/or installed into the theme? | 13:38 |
| didrocks | xnox: it's icons, images | 13:38 |
| didrocks | xnox: so I really think they should be in usr/share | 13:38 |
| davmor2 | popey: I was thinking more that the general public could tell, but also that devs could tell so they could ask if I could reproduce something with a certain build :) | 13:39 |
| oSoMoN | xnox: not an icon, it’s a PNG used as a UI element | 13:39 |
| didrocks | xnox: btw, sil2100 is pending on your on your branch with the solution you came for include dirs yesterday :) | 13:39 |
| jcollado | ogra_: You're looking at results for build 20130718 for mako, maguro and manta, right? | 13:39 |
| popey | davmor2: anyone can see the history | 13:39 |
| didrocks | oSoMoN: on other apps, there are icons :) | 13:39 |
| xnox | oSoMoN: and how exactly is a PNG resolution independant? =) surely it should be SVG? | 13:39 |
| ogra_ | jcollado, exactly,. the fixes for the failures went into utah yesterday and the errors should not occur anymore | 13:39 |
| davmor2 | popey: ah okay, again I only use gdocs to add info so never looked into it :) | 13:39 |
| davmor2 | removed the dm :) | 13:40 |
| xnox | didrocks: i thought, it's resolved now. one sec. | 13:40 |
| jcollado | ogra_: This morning om26er asked me about the version installed in the lab machine that runs those jobs and the package seems to be up to date. | 13:40 |
| didrocks | xnox: oh? nice :) | 13:40 |
| ogra_ | jcollado, hmm, that is weird | 13:40 |
| ogra_ | jcollado, 0.14.2ubuntu1 is the version we need | 13:40 |
| ogra_ | according to the bzr branch | 13:41 |
| om26er | the one on phoenix is: 0.14.2ubuntu1-r971~precise1 | 13:41 |
| xnox | didrocks: https://code.launchpad.net/~thomas-voss/platform-api/fix-1202309/+merge/175482 | 13:41 |
| oSoMoN | xnox: the UITK has a mechanism to scale assets automatically depending on the value of a grid unit, and the PNG provided should cater for all our supported resolutions, despite not being resolution-independent (but I think that’s irrelevant to the current discussion) | 13:41 |
| didrocks | xnox: so, it's not the big "fix" and solution you wanted yesterday? | 13:42 |
| ogra_ | om26er, hmm, the branch only has raring in the changelog, but f871 should be the good one according to bzr | 13:42 |
| xnox | didrocks: no, but it's the correct one for now. see comments https://bugs.launchpad.net/platform-api/+bug/1202309 | 13:42 |
| ubot5 | Launchpad bug 1202309 in platform-api "libplatform-api-headers needs to make its headers accessible through #include <ubuntu/*.h>" [High,New] | 13:42 |
| didrocks | you talked for 10 minutes how things should be versioned, I'm disappointed you didn't provide the implementation of it :p | 13:42 |
| jcollado | ogra_: Maybe is something related to the daily/stable PPA (0.14.2ubuntu1-r971~precise1 vs 0.14.2ubuntu1~precise1) | 13:42 |
| * jcollado checks | 13:43 | |
| ogra_ | jcollado, well, r971 in bzr should be the right version | 13:43 |
| ogra_ | do if we can trust that versioning it should be good (which doesnt explain why the identical error persists in the logs though) | 13:43 |
| xnox | didrocks: i can provide implementation for it, but the default headers will still be accesible from /usr/include/ubuntu/*.h And i'd want two versions to play with, otherwise i'd have a metapackage pointing to default implementatin out of a set of multiple implementations, where only single implementation exists at the moment. | 13:43 |
| ogra_ | s/do/so/ | 13:43 |
| jcollado | ogra_: Yes, indeed tag "0.14.2" was created for rev.971, so they should be the same pacakge. | 13:44 |
| jcollado | package | 13:44 |
| xnox | didrocks: so post 13.10 release, I can provide/implement the versioning. | 13:44 |
| * ogra_ doesnt get that | 13:44 | |
| didrocks | xnox: you can be sure we'll have another evolution of the API before 13.10 ;) | 13:44 |
| xnox | didrocks: oh, ok. | 13:44 |
| didrocks | xnox: will keep you posted then! | 13:45 |
| jcollado | ogra_: What I mean is that we have two different PPAs, the daily one and the stable one, but the packages each PPA right now are the same. | 13:45 |
| lool | cjwatson dholbach: Hey, the PK changes for Click: are these in the images already, or what's the best way to get them there? Seems Sebastian and Richard are still discussing longer terms plans which are in a too distant future | 13:45 |
| xnox | didrocks: and upstream will continue to build both, and not remove the old one / change inline? | 13:45 |
| ogra_ | jcollado, right, i just dont get why the tests would still fail | 13:45 |
| pmcgowan | jcollado, asac informs me some of your qa brothers are looking into this | 13:45 |
| pmcgowan | jcollado, can you try to confirm that so I am not wasting your time | 13:45 |
| lool | cjwatson: we were just chatting with Roberto on landing his scope to search + install clicks, and he lacks download service (Manuel on it with Barry) and the PK changes | 13:45 |
| xnox | didrocks: (or like imply ifdef magic to be able to build "old.so" and "new.so")? or use versioned symbols to include both apis, in single .so? | 13:46 |
| didrocks | xnox: I think that's the goal for platform-api | 13:46 |
| ogra_ | asac, who is looking into it ? | 13:46 |
| asac | rfowler, waiting also for doanac to confirm the exact thing afaiui | 13:46 |
| xnox | didrocks: ok. | 13:47 |
| jcollado | ogra_, pmcgowan, asac: Ok, let me check with rfowler and doanac | 13:47 |
| mardy | seb128: hi! I need to leave soon, I won't be able to attend the SystemSettings meeting | 13:51 |
| mardy | seb128: OTOH, I don't have anything to report | 13:51 |
| seb128 | mardy, hey, ok | 13:51 |
| seb128 | mardy, I was going to ask if you had anything | 13:51 |
| seb128 | mardy, do you plan to look at the unitymenumodel support or is that not on your list? I'm unsure who owns that one | 13:52 |
| ogra_ | jcollado, do you have an idea where /usr/local/bin/utah-autopilot comes from ? it seems to be a wrapper around the code in question and i cant find such a file in the utah branch | 13:52 |
| asac | jcollado: coordinate with gema and your folks to avoid directors sending you on an overloaded road :)... just help folks like ogra etc. for now | 13:52 |
| asac | i guess | 13:52 |
| asac | but really coordinate with yhour team | 13:53 |
| jcollado | ogra_: That file is created by run_utah_phablet.py to run autopilot tests just after booting | 13:53 |
| jcollado | asac: Ok | 13:55 |
| === jono is now known as Guest30113 | ||
| jcollado | ogra_: I'm going to run some tests with the device to check if I get the same problem and troubleshoot it. | 14:00 |
| tvoss_ | sil2100, didrocks in case you haven't seen https://code.launchpad.net/~thomas-voss/platform-api/fix-1202309 | 14:01 |
| ogra_ | jcollado, thanks ... btwm, theoretically all that exporting of the dbus address shouldnt be needed, "sudo -u phablet -i" should automatically have the address in the environment | 14:02 |
| didrocks | sil2100: do you mind taking care of this? ^ (had a quick look, I think that's good) | 14:02 |
| ogra_ | exporting it is kind of redundant | 14:02 |
| ogra_ | (but shouldnt do any harm) | 14:02 |
| jcollado | ogra_: Ok | 14:03 |
| om26er | can we have bash-completion in the image, atleast till the final release. | 14:10 |
| ogra_ | om26er, thats quite some overhead | 14:11 |
| om26er | ogra_, its 173k only | 14:11 |
| ogra_ | no, i mean running it | 14:12 |
| sil2100 | Looking! | 14:14 |
| popey | if om26er gets bash completion i want man pages! | 14:14 |
| popey | ☻ | 14:14 |
| ogra_ | haha | 14:14 |
| om26er | oh man | 14:14 |
| * ogra_ wants changelogs | 14:14 | |
| * sergiusens wants everything to just work | 14:15 | |
| bobweaver | http://qt.gitorious.org/qt/qtdeclarative/blobs/b9ccb579c4c93f23e6ceeea26b07d418ad4e5562/examples/qml/script/shell/main.cpp | 14:15 |
| bobweaver | woops wrong window sorry about that | 14:15 |
| mhall119 | ogra_: sergiusens: can one of you answer http://forum.xda-developers.com/showthread.php?p=43700742#post43700742 | 14:17 |
| stgraber | ogra_: hey, so you mentioned some fixes to phablet-test-run, does that mean you have a version which doesn't attempt to push stuff through /data/ubuntu on flipped? | 14:18 |
| ogra_ | stgraber, you wish ... no | 14:19 |
| ogra_ | stgraber, currently we try to get tests working at all | 14:19 |
| om26er | sergiusens, you can simplify fetch_artifact() in the MR if you want to: adb -s $ANDROID_SERIAL pull /tmp/test_results.xml $RESULTDIR | 14:19 |
| sergiusens | stgraber: wait, is that all you need? that's changeable in phablet-test-run | 14:19 |
| === balloons_ is now known as balloons | ||
| om26er | else I'll just approve now and do a MR to change that later | 14:19 |
| sergiusens | om26er: I'll change with stgraber's comment, I must of missed that one | 14:20 |
| om26er | sergiusens, in this MR or another ? | 14:20 |
| sergiusens | om26er: same MR | 14:21 |
| om26er | ack | 14:21 |
| sergiusens | om26er: done | 14:21 |
| ogra_ | mhall119, i think rsalveti was working on merging 4.2.2 | 14:21 |
| stgraber | sergiusens: well, that's the first problem I saw ;) After that, I'll at least need to add some code to put the device in developer mode and reboot it (so that phablet-test-run can actually write files to memory) | 14:21 |
| sergiusens | ogra_: he is | 14:21 |
| ogra_ | yeah | 14:22 |
| sergiusens | mhall119: look for the phablet-saucy in phablet.ubuntu.com | 14:23 |
| ogra_ | stgraber, note that you likely need to do the same changes in utah | 14:23 |
| annerajb | hello | 14:23 |
| stgraber | sergiusens: I also suspect all that SSH code could go away now that adbd is running on the ubuntu side but it may not be worth the effort | 14:23 |
| ogra_ | stgraber, examples/run_utah_phablet.py in lp:utah is the equivalent to phablet-test-run | 14:24 |
| ogra_ | (no idea why it doesnt just use phablet-test-run, i guess that would be easier) | 14:24 |
| sergiusens | ogra_: I'm going to make one out of these | 14:24 |
| stgraber | ogra_: ok, thanks. I'll try to get phablet-test-run working first as I can actually test that one, then I'll propose some changess against UTAH and QA can take care of it | 14:24 |
| mhall119 | sergiusens: what the phablet-saucy? | 14:25 |
| sergiusens | mhall119: the git branch with the 4.2.2 changes rsalveti is working on | 14:27 |
| mhall119 | ah, ok | 14:28 |
| mathrb | Hello, i would like to know if there is a Windows tutorial on how to install ubuntu touch on a nexus ? | 14:32 |
| seb128 | is sound supposed to work on grouper? | 14:32 |
| seb128 | things hand when I try to play a sound from qtmultimedia there | 14:32 |
| cjwatson | lool: packagekit-plugin-click is in the archive, but not quite on the image yet. Before it can land I need to upload some changes to PackageKit itself. I hope to do this today or tomorrow (I'm trying to focus on the hook changes today). | 14:33 |
| popey | seb128: no | 14:34 |
| popey | seb128: bug 1187865 | 14:34 |
| ubot5 | bug 1187865 in touch-preview-images "Audio doesn't work on Nexus 7" [Medium,Confirmed] https://launchpad.net/bugs/1187865 | 14:34 |
| seb128 | popey, hum, k, thanks | 14:34 |
| cjwatson | lool: Oh, there's also the problem that you currently need a .pkla file for it to be useful, since apparently PolicyKit doesn't work on touch yet. I might upload that in click as a temporary workaround | 14:34 |
| sergiusens | mhall119: trying to reply after 10 captcha failures | 14:35 |
| ogra_ | write a sript for the captcha login :P | 14:35 |
| popey | some people would pay good money for that | 14:36 |
| ogra_ | :) | 14:36 |
| * sergiusens gives up | 14:36 | |
| ogra_ | forums are so annoying | 14:36 |
| popey | Yes. Yes they are. | 14:37 |
| mhall119 | sergiusens: I can reply, just let them know that it's being worked on | 14:37 |
| mhall119 | sergiusens: I don't see a phablet-saucy on http://phablet.ubuntu.com/gitweb though, can you give me a link to pass along? | 14:37 |
| sergiusens | mhall119: this is my reply to the thread http://pastebin.ubuntu.com/5887740/ | 14:38 |
| sergiusens | mhall119: so phablet-saucy is a branch inside each of the git repos, e.g.; look at the heads in http://phablet.ubuntu.com/gitweb?p=CyanogenMod/android_system_core.git;a=summary | 14:39 |
| mhall119 | cool, I'll just copy/paste and sound way smarter than I am :) | 14:39 |
| mhall119 | thanks sergiusens | 14:40 |
| kalikiana | meh, lucky I don't need to do too much in the phone, bash is such a pain | 14:40 |
| kenvandine | ogra_, would it be a bad idea to flash with pending right now? | 14:40 |
| ogra_ | kenvandine, runs fine for me | 14:41 |
| kalikiana | not even colors and it cannot properly handle line breaks | 14:41 |
| kenvandine | ogra_, thanks | 14:41 |
| ogra_ | kalikiana, through ssh it can :) | 14:41 |
| ogra_ | thats not bashs fault | 14:41 |
| mhall119 | kalikiana: yeah, I think it's adb that messes it up | 14:41 |
| mhall119 | the Terminal app in the phone works fine :) | 14:42 |
| kalikiana | ogra_, I wasted too much time re-configuring ssh after flashing… | 14:42 |
| kalikiana | so I use adb shell | 14:42 |
| ogra_ | yeah, well, if i use adb i do 8% of the stuff through pipes on my lappie | 14:42 |
| ogra_ | 80% | 14:42 |
| mhall119 | I wonder if adb isn't setting a good $TERM variable | 14:43 |
| kalikiana | what does that mean? | 14:43 |
| ogra_ | adb shell /system/bin/logcat -d | less | 14:43 |
| ogra_ | for example | 14:43 |
| kalikiana | ah. but then you get no completion | 14:43 |
| mterry | katie, poke about pin unlocking | 14:43 |
| ogra_ | that gives me the data from the phone in my local shell | 14:43 |
| kalikiana | I don't exactly know all the folders by heart | 14:43 |
| ogra_ | ah, k, yeah that makes it more complicated indeed | 14:44 |
| seb128 | popey, should the touch-preview-images bug reassigned to their correct component or should the project be kept, to have an overview, with the correct source added to the list? | 14:47 |
| seb128 | popey, e.g should I reassign https://bugs.launchpad.net/touch-preview-images/+bug/1202279 or just do "also affect system settings" | 14:47 |
| ubot5 | Launchpad bug 1202279 in touch-preview-images "Carrier is "Aubergine" should be "Orange"" [Undecided,New] | 14:47 |
| ogra_ | seb128, for packages in the archive we should use the proper distro bugflow | 14:48 |
| sergiusens | seb128: from what I recall we were only going to log stuff there that was not in ubuntu archives yet | 14:48 |
| seb128 | ogra_, well, I do, but I don't mind listed the project as well for tracking | 14:48 |
| ogra_ | we need to come up with something here anyway | 14:48 |
| seb128 | ogra_, sergiusens: thanks, I'm going to reassign | 14:48 |
| ogra_ | like a touch tag or so | 14:48 |
| lool | cjwatson: ah *policy*kit, right, I was heavily confused for a sec; alright, thanks for the updates, I was just double-checking the PackageKit changes had someone looking after them, thanks a lot | 14:49 |
| annerajb | anybody see anything weird on this lgso of my cellphone it's not showing anything it just have a black screen. logcat > http://pastebin.com/izdcBy7X dmesg > http://pastebin.com/D5HCaBmD ps > http://pastebin.com/3jgqndLs | 14:51 |
| rsalveti | morning | 14:52 |
| annerajb | rsalveti: morning | 14:53 |
| katie | mterry, hi | 14:53 |
| rsalveti | pitti: thanks for testing the ofono mr | 14:54 |
| rsalveti | annerajb: hey, able to flash the ubuntu zip after all? | 14:54 |
| annerajb | yeah and with 400mb to spare on the 918gb partition | 14:54 |
| rsalveti | seems it didn't start the ubuntu container from your ps list | 14:55 |
| rsalveti | can you use 'ubuntu_chroot shell'? | 14:55 |
| mterry | katie, hi! Are there mockups or design guidance for how the "SIM card unlock" interaction should happen? | 14:55 |
| rsalveti | mhall119: yeah, we hope to announce the phablet-saucy branch soon | 14:56 |
| popey | seb128: thanks for re-assigning. I wasn't sure where to put that one. | 14:56 |
| annerajb | rsalveti: it says su cannot determine your username | 14:57 |
| seb128 | popey, yw, good news is that we got it working using qtsystems and ofono | 14:57 |
| seb128 | popey, so that should be fixed on tomorrow's image | 14:57 |
| popey | nice! | 14:57 |
| popey | do the roaming switches work seb128 ? | 14:57 |
| popey | (going to america on sunday, don't want to run up a huge bill) :D | 14:58 |
| seb128 | popey, no, none of that panel works, we are adding a warning UI as well :p | 14:58 |
| popey | ah okay | 14:58 |
| popey | i should find some way to block 3g | 14:58 |
| katie | mterry, yes | 14:58 |
| katie | mterry, I'll ping you a link | 14:58 |
| cjwatson | lool: Do you know if there's a plan anywhere to get PolicyKit working, BTW (well, I think it kind of works but the session isn't considered active)? I know other people have had to work around this - I saw a patch to network-manager as well | 14:58 |
| rsalveti | xnox: please don't disable icu, it's used by some binary blobs | 14:59 |
| xnox | rsalveti: sqlite binary as well or no? | 15:00 |
| xnox | rsalveti: ok. I guess then i'm done and can prepare an upload for security team to review. | 15:00 |
| rsalveti | broadcom/grouper/proprietary/glgps uses libicuuc.so it seems | 15:01 |
| rsalveti | probably not | 15:01 |
| ogra_ | cjwatson, once we use lightdm that will magically start working | 15:01 |
| ogra_ | cjwatson, the hacked up session we have atm simply doesnt register with logind | 15:01 |
| lool | cjwatson: I know we want to get rid of the patches so it's on the radar, but I'm completely unable to tell you who's after it | 15:02 |
| rsalveti | xnox: binaries is usually fine, just concerned about libs as it might be required by the blobs | 15:03 |
| cjwatson | ogra_: Ah, cool | 15:04 |
| katie | mterry, here is the SIM PIN stuff https://docs.google.com/a/canonical.com/document/d/1VajNkWbBH61iVixXJAmOvNGiG__GWQTMXGNOZijXWJw/edit#heading=h.y8mp9oy60bxd | 15:05 |
| katie | mterry, and then I also made this doc: https://docs.google.com/a/canonical.com/document/d/1FyQXYlVoGJxIVAr_2ofvZl0au1F1Xv2ek5zydr2ziaM/edit# | 15:05 |
| katie | which is an attempt to point to all SIM related UI designs | 15:06 |
| mterry | katie, aha, I had seen that first link after all, just couldn't find it again | 15:06 |
| === ckpringle_ is now known as ckpringle | ||
| mterry | katie, where you talk about installing the SIM, you say a non clickable notification. But what if the SIM needs to be unlocked, but is inserted during the session? | 15:07 |
| annerajb | rsalveti: this is the line failling in the ubuntu_chroot exec $CHROOT $ROOT /bin/bash -c su - | 15:08 |
| katie | mterry, good catch | 15:09 |
| katie | mterry, I guess we should show the same SIM PIN prompt at this point. It can be dismissed, in which case the notification will be shown | 15:10 |
| mterry | katie, so an interrupting prompt? I guess not so bad, since user just inserted it. not like we are interrupting a movie | 15:10 |
| mterry | katie, if dismissed, they can unlock how? remove/insert again? | 15:11 |
| katie | mterry, just discussing this with mpt | 15:12 |
| rsalveti | annerajb: did you enable all the needed options in your kernel? | 15:21 |
| katie | mterry, on talking to mpt it makes sense to have an indicator that says SIM locked (that appears in the same space as SIM error) | 15:21 |
| jdstrand | jcollado: hey, can you restart http://reports.qa.ubuntu.com/smokeng/saucy/image/3053/security/? | 15:21 |
| jdstrand | jcollado: (Total count is 0, the tests didn't run) | 15:22 |
| annerajb | rsalveti: yes | 15:23 |
| === dandrader is now known as dandrader|afk | ||
| jcollado | jdstrand: Yes, I see some SSH problem in the log. I've rescheduled it. | 15:24 |
| jdstrand | jcollado: thanks | 15:24 |
| jcollado | jdstrand: No problem | 15:25 |
| rsalveti | annerajb: because that's the line that starts the container itself | 15:26 |
| flo__ | annerajb, I once had the same issue, re-flashing helped in my case (probably flashing didn't work because /data didn't have enough free storage). Not sure what the error message means or where it comes from though | 15:32 |
| annerajb | flo__: data shows around 300mb free but ill give it another try thanks | 15:34 |
| flo__ | the flash process is like this: *copy tar.gz to /data * extract tar.gz * delete tar.gz from /data. so you will always have about ~300mb free after the flash | 15:35 |
| flo__ | at least last time I checked :) | 15:35 |
| mpt | katie, mterry: https://wiki.ubuntu.com/Networking?action=diff&rev2=115&rev1=114 | 15:35 |
| annerajb | flo__: all right i was having issues earlier by running out of space on /data while extracting but this last time i flashed it didnt ran out of memory so it may have failed extracting or something like that | 15:36 |
| doanac | rsalveti, ogra_: I think I may see the root issue of the qa-smoke issue: https://jenkins.qa.ubuntu.com/job/smoke-saucy-touch-apps-mako/24/console | 15:38 |
| doanac | that was our most recent run and it somehow flashed the 20130716 build | 15:38 |
| doanac | it should have grabbed "pending" but didn't (I think) | 15:39 |
| ogra_ | ah | 15:39 |
| rsalveti | right | 15:39 |
| ogra_ | doanac, i know there are proxy issues in lexington, pmcgowan has them if he works from the office | 15:39 |
| ogra_ | i wonder if thats the same | 15:39 |
| sergiusens | doanac: \o/ | 15:39 |
| doanac | ogra_: ah - that's interesting. because that script grabs the right image from my house | 15:40 |
| === dandrader|afk is now known as dandrader | ||
| doanac | let me poke around in the lab | 15:40 |
| sergiusens | doanac: can you print out the phablet-flash command output in the console? | 15:40 |
| ogra_ | sergiusens, we use still wget, right ? | 15:40 |
| sergiusens | doanac: or in a different jenkins job | 15:40 |
| doanac | sergiusens: yea - was going to add that as a debug today | 15:40 |
| sergiusens | ogra_: rsync to find out what pending is and then wget | 15:41 |
| ogra_ | i wonder if it has an option to force the proxy to update when it requests the download | 15:41 |
| ogra_ | oh | 15:41 |
| ogra_ | rsync | 15:41 |
| ogra_ | i wonder if rsync even gets through | 15:41 |
| sergiusens | ogra_: well the build would fail if not | 15:41 |
| sergiusens | the flash* | 15:41 |
| sergiusens | wget --no-cache ? | 15:41 |
| ogra_ | if the http traffic is already proxied i dont want to know what rsync does | 15:41 |
| pitti | rsalveti: yw, I'm happy that it stopped draining power like made :) | 15:42 |
| pmcgowan | yeah I could never get past that once its confused | 15:42 |
| doanac | hmm, just ran phablet-flash by hand in the lab and it started downloading todays image. let me poke more | 15:42 |
| ogra_ | sergiusens, that might help, i need to read the manpage :) | 15:42 |
| sergiusens | ogra_: sends no-cache in the headers, I know that proxy servers are free to ignore it though | 15:42 |
| ogra_ | yeah, that looks like it could help | 15:43 |
| ogra_ | according to the description in man | 15:43 |
| ogra_ | if the proxy IS maintaines ignores it we can probably ask IS to change that :) | 15:43 |
| sergiusens | rsync seems to go straight through | 15:44 |
| ogra_ | good | 15:45 |
| doanac | ogra_, sergiusens, rsalveti: i've got the issue. the jenkins job was using a branch to run utah. I need to update it to use the released utah | 15:47 |
| ogra_ | ++ | 15:48 |
| doanac | that's why it wasn't using --pending | 15:48 |
| rsalveti | doanac: awesome | 15:48 |
| doanac | its a simple fix and i'll kick everything off | 15:48 |
| doanac | see what happens | 15:48 |
| ogra_ | so i wasnt actually wrong, good ... i wonder what jcollado saw then when looking on the machine for the utah version | 15:48 |
| ogra_ | there seems to be a discrepancy | 15:48 |
| doanac | i think he didn't read the logs close enough to see I was executing utah from a branch and not the system pkg | 15:48 |
| doanac | semi-amusing patch for the fix: http://paste.ubuntu.com/5887950/ | 15:49 |
| jcollado | ogra_, doanac: No, I didn't see that. Which job was affected by that? | 15:50 |
| rsalveti | doanac: :-) | 15:52 |
| sergiusens | didrocks: question, if I have commit #1 without manual changelog entry, then commit #2 the same way, and commit #3 with a changelog entry, will daily release just use what's in commit #3 or will it aggregate? | 15:56 |
| mattyw | has anyone else seent this when trying to build the music-app? music-app.qml:287 Cannot assign to non-existent property "filterDirectories" | 15:57 |
| didrocks | sergiusens: daily release just care about what's in trunk mainline (what you see with bzr log), if the commit has a change in debian/changelog, it won't list it, if the commit hasn't one, it will collect | 15:57 |
| sergiusens | didrocks: if I read that correctly, it never uses it: "if the commit has a change in debian/changelog, it WON'T list it" | 16:00 |
| didrocks | sergiusens: right | 16:01 |
| sergiusens | didrocks: so the only purpose is to bump the version then, right? | 16:01 |
| didrocks | sergiusens: or having a personal changelog if you don't like your commit message :) | 16:02 |
| sergiusens | didrocks: ack | 16:02 |
| didrocks | sergiusens: that can be useful to blame someone else for your changes (discret attribution :p) | 16:03 |
| sergiusens | didrocks: yeah, my goal was to write a changelog and include what was in the previous commits in between daily releases to build up something nice, I guess that won't work | 16:04 |
| didrocks | sergiusens: no, you needed to use #nochangelog in your previous commit message | 16:04 |
| didrocks | sergiusens: you can cheat with adding: * Automatic snapshot from revision <rev> | 16:05 |
| didrocks | sergiusens: where rev is the rev after the commit you don't want to list | 16:05 |
| sergiusens | didrocks: cheating sounds good! | 16:05 |
| didrocks | are you so ashame of your commit message? :p | 16:05 |
| sergiusens | but I will leave it for next time | 16:05 |
| didrocks | ok! | 16:05 |
| sergiusens | didrocks: no, but I wanted to rewrite with bullets and sub bullets | 16:06 |
| didrocks | ah ok ;) | 16:06 |
| popey | can someone with ubuntu touch on a device and some music confirm https://bugs.launchpad.net/music-app/+bug/1202730 ? | 16:09 |
| ubot5 | Launchpad bug 1202730 in Ubuntu Music App "Music plays too fast on device" [Undecided,New] | 16:09 |
| cjwatson | lool: I've uploaded the various pieces to get the PK plugin into the image and hopefully working by default; should be in tomorrow's images | 16:12 |
| lool | cjwatson: \o/ thanks | 16:13 |
| lool | ralsina: ^ | 16:13 |
| lool | cjwatson: I guess it should work equally on the desktop? | 16:13 |
| === rtg is now known as rtg-afk | ||
| cjwatson | lool: Not yet, due to incompatibility with aptdaemon | 16:15 |
| cjwatson | lool: It works in an LXC container | 16:15 |
| cjwatson | lool: I'm not worried about this yet because convergence is not this month's target ;-) | 16:15 |
| om26er | popey, could be specific to some codecs your music use, My test song plays fine | 16:23 |
| ogra_ | or device specific | 16:25 |
| WebbyIT | Another countdown? :D http://www.ubuntu.com/ | 16:26 |
| iBelieve | WebbyIT, maybe a demo of convergence??? | 16:28 |
| popey | om26er: mp3s? | 16:28 |
| om26er | popey, yeah | 16:28 |
| om26er | WebbyIT, wow | 16:28 |
| om26er | iBelieve, do you plan to have an icon for 'Home' in file manager? | 16:30 |
| iBelieve | om26er, right now it lives at the top of the Places popover. Do think it should be more obvious, like in the toolbar? | 16:32 |
| om26er | iBelieve, the Home in places having an icon will suffice. | 16:35 |
| om26er | iBelieve, i don't think we need to go to home that frequently to have it in the toolbar | 16:35 |
| === Jim_Lahey is now known as Guest67902 | ||
| iBelieve | om26er, okay, thanks for the input. The toolbar is already rather full, especially once I add a settings sheet and rename the the settings popover to options. | 16:37 |
| === Guest67902 is now known as bobweaver | ||
| om26er | iBelieve, do you have any guess, why our icons in the file manager toolbar are blurry but the one's in other apps look fine ? | 16:39 |
| oSoMoN | fginther: ping | 16:39 |
| iBelieve | om26er, the icons where? In the toolbar, or the file/folder icons, or the icons in the places popover? | 16:40 |
| === datachomper is now known as rmeyerriecks | ||
| om26er | iBelieve, in the toolbar | 16:40 |
| iBelieve | om26er, do you have a screenshot? Are you on a real device? | 16:41 |
| iBelieve | om26er, here is what I see: http://i.imgur.com/3KgxODX.png | 16:41 |
| om26er | iBelieve, they look fine in your screenshot. let me upload mine | 16:42 |
| iBelieve | om26er, are you testing it on a phone? If so, I think I can guess what the problem is. | 16:43 |
| om26er | iBelieve, yep, I am on a phone | 16:43 |
| om26er | iBelieve, http://imagebin.org/index.php?mode=image&id=264942 | 16:45 |
| iBelieve | om26er, my guess is that it is because of the difference in resolution. The icons I'm using are copied from the ubuntu-mobile icon theme so they don't look pixilated, but I rendered them at 16 px. | 16:46 |
| om26er | iBelieve, understood | 16:47 |
| iBelieve | om26er, if you want to file a bug (or I can), I'll try and fix it. | 16:47 |
| om26er | iBelieve, so you would need new icons with better resolutions ? | 16:47 |
| om26er | or pixels if you must | 16:47 |
| iBelieve | om26er, the toolkit has the great icons, but I just need to not scale them down. I'll take a look at what the other apps do. | 16:48 |
| iBelieve | om26er, the toolkit's icons are the right size. | 16:48 |
| mhall119 | Calculator team meeting is starting in #ubuntu-touch-meeting | 16:55 |
| === schwuk is now known as schwuk_away | ||
| === alan_g is now known as alan_g|EOD | ||
| ogra_ | asac, http://reports.qa.ubuntu.com/smokeng/saucy/image/3052/camera-app-autopilot/ | 17:08 |
| ogra_ | \o/ | 17:08 |
| ogra_ | doanac, seemss to work :D | 17:09 |
| doanac | ogra_: yeah. the mako webbrowser test seems to have some test cases that hang. I'm killing them by hand now, to get the test to continue | 17:09 |
| ogra_ | k | 17:10 |
| asac | ogra_: you rock! | 17:12 |
| ogra_ | asac, well, doanac does all the work now :) | 17:12 |
| asac | ogra_: but you are on it ... and that is worth gold | 17:12 |
| ogra_ | heh | 17:12 |
| asac | ogra_: http://reports.qa.ubuntu.com/smokeng/saucy/image/3001/ | 17:12 |
| asac | ogra_: that list is basically all the autopilts taht should run | 17:13 |
| asac | so dont give in untless all those report something | 17:13 |
| asac | once stuff starts to improve :) | 17:13 |
| ogra_ | indeed | 17:13 |
| === joseph_thegreat is now known as bobweaver | ||
| doanac | asac, ogra_: a couple of tests seemed to have gotten skipped for maguro. looked like adb didn't find the device at the time of the test | 17:14 |
| doanac | i'm still trying to analyze it all though | 17:14 |
| ogra_ | sounds a bit racy | 17:14 |
| asac | doanac: black box is next? :) | 17:16 |
| asac | just setup once and kick off basically :) | 17:16 |
| asac | for each boot | 17:16 |
| asac | ok later | 17:16 |
| doanac | ogra_: yeah. I think i need to put together some sort of retry logic in our jobs for adb | 17:16 |
| om26er | how to check the build number ? | 17:21 |
| om26er | from the phone | 17:22 |
| stgraber | sergiusens: hey there, so how are things going? do you think you'll have the new --bootstrap-image code landed upstream by EOD today? | 17:26 |
| rsalveti | doanac: adb usually works fine, but you shouldn't trust it 100% :-) | 17:26 |
| stgraber | om26er: not sure on the current flipped but with the read-only images we're working on, you'll get /etc/ubuntu-build for that | 17:26 |
| om26er | stgraber, thanks, just figured on flipped its /system/ubuntu_stamp | 17:27 |
| stgraber | om26er: is that the version of the rootfs or the android bits? both are built separately so it's hard to really have a single version number for the device | 17:28 |
| om26er | stgraber, right, not sure about that | 17:28 |
| === Jim_Lahey is now known as Guest39471 | ||
| === Guest39471 is now known as bobweaver | ||
| sergiusens | stgraber: yes I will | 17:32 |
| sergiusens | stgraber: om26er ubuntu_stamp is going away | 17:32 |
| sergiusens | om26er: stgraber /var/log/installer/media-info | 17:34 |
| === rtg-afk is now known as rtg | ||
| rsalveti | doanac: http://reports.qa.ubuntu.com/smokeng/saucy/image/3052/ | 17:35 |
| om26er | cool | 17:35 |
| rsalveti | http://reports.qa.ubuntu.com/smokeng/saucy/image/3053/ | 17:35 |
| DJJeff | does ubuntu touch REALLY need to be making HTTPS connections to productsearch.ubuntu.com | 17:35 |
| ogra_ | om26er, root@ubuntu-phablet:/# cat /var/log/installer/media-info | 17:36 |
| ogra_ | Ubuntu Saucy Salamander (development branch) - armhf (20130718) | 17:36 |
| doanac | rsalveti: we need to sync the dashboard with jenkins. there's some more up-to-date results now | 17:36 |
| rsalveti | doanac: why do we have different test cases there? mako seems to have more than maguro | 17:36 |
| rsalveti | doanac: oh, cool | 17:36 |
| doanac | rsalveti: the other thing - maguro had some tests fail (adb couldn't connect) | 17:36 |
| doanac | i'm going to kick those off again | 17:36 |
| sergiusens | doanac: ogra_ rsalveti at least it looks like we are finally on track again :-) | 17:37 |
| rsalveti | doanac: do you have logs for one of those? | 17:37 |
| rsalveti | sergiusens: right | 17:37 |
| ogra_ | stgraber, ^^^ we already have that ... :) livecd-rootfs crates it, should i make it move to /etc/ubuntu-build ? (i thought usingg something standard that tools like apport already use would be cleverer when i implemented it) | 17:37 |
| doanac | rsalveti: http://10.97.0.1:8080/job/saucy-touch-maguro-smoke-mediaplayer-app-autopilot/8/console | 17:38 |
| doanac | i *think* i may know what's happening there. | 17:38 |
| doanac | i need to test, but sometimes "adb reboot" doesn't reboot fast enough, so wait-for-device comes back immediately | 17:38 |
| sergiusens | doanac: yeah, that's why I have ugly sleeps in the phablet-tools code | 17:39 |
| rsalveti | doanac: right, guess I had similar issues in the past as well | 17:39 |
| sergiusens | doanac: can also fail right after an adb root too | 17:39 |
| === sfeole` is now known as sfeole | ||
| ogra_ | sergiusens, adb root ? | 17:39 |
| ogra_ | on flipped ? | 17:39 |
| doanac | sergiusens: i'll look at your sleeps and copy them to my jobs | 17:40 |
| sergiusens | ogra_: in general | 17:40 |
| ogra_ | ah | 17:40 |
| rsalveti | right, we don't need adb root anymore | 17:40 |
| rsalveti | so it might be with adb reboot | 17:40 |
| sergiusens | ogra_: if adb is restarted server side it creates havoc | 17:40 |
| stgraber | ogra_: no, ubuntu-build is something entirely different, it's a build number for the whole image (so rootfs + android + firmware + customizations), the format is also slightly different as it's YYYYMMXX instead of YYYYMMDD.X | 17:40 |
| ogra_ | rsalveti, well, if we want to keep adb for release we should probably think about starting it unprivileged | 17:41 |
| === dpm_ is now known as dpm | ||
| sergiusens | ogra_: yeah, true | 17:41 |
| sergiusens | ogra_: although developer mode should enable adb and nothing else | 17:41 |
| doanac | rsalveti: mako results are now up-to-date: http://reports.qa.ubuntu.com/smokeng/saucy/image/3053/ | 17:41 |
| stgraber | ogra_: so currently the latest system-image build number for mako is 20130714 which is made of 20130718 (rootfs) and 20130718 (android), but those two may differ if some bits aren't updated in an image | 17:42 |
| ogra_ | stgraber, hmm, so thats magic you do when repacking images ? in fact the pan was to have all these bits from one build anyway | 17:42 |
| ogra_ | *plan | 17:42 |
| stgraber | ogra_: yeah, basically we generate a new version per device every time we import something, /etc/ubuntu-build is then applied ontop of that so it doesn't come from the rootfs and doesn't come from android | 17:43 |
| stgraber | ogra_: if you look at an image in http://system-image.ubuntu.com/daily/mako/index.json ubuntu-* is the rootfs <device>-* is system.img, boot.img, recovery.img, ... and version-* is /etc/ubuntu-build | 17:43 |
| sergiusens | doanac: Jul 18 16:28:52 ubuntu-phablet utah: #012RuntimeError: Unable to instantiate any backends#012X11: ImportError('No module named Xlib',)#012#012#012Ran 1 test in 6.313s#012FAILED (failures=1) | 17:43 |
| sergiusens | wonder why that happened | 17:43 |
| ogra_ | stgraber, so wouldnt it be better to use media-info and i leave a placeholder in ther that you can replace ? | 17:43 |
| rsalveti | xlib? | 17:44 |
| sergiusens | rsalveti: yeah, autopilot might be on _desktop_ mode | 17:44 |
| ogra_ | stgraber, just thinking about tools like apport that use this location by default already | 17:44 |
| rsalveti | doanac: so use wait-for-device; sleep 3; wait-for-device again | 17:44 |
| stgraber | ogra_: well, media-info is fine as it's as it'll give you the rootfs version which is what apport usually wants | 17:44 |
| rsalveti | doanac: just to be sure | 17:44 |
| ogra_ | stgraber, ah, so we leve that part as is, ok | 17:45 |
| doanac | rsalveti: okay. | 17:45 |
| stgraber | ogra_: yep, they have different goals, there's definitely value in having the rootfs version being stored in a standard location on the fs | 17:46 |
| ogra_ | okay | 17:46 |
| sergiusens | om26er: can you give this a go? https://code.launchpad.net/~sergiusens/phablet-tools/flipped_only/+merge/175625 | 17:46 |
| sergiusens | after that's in I want to do a daily-release | 17:47 |
| === bfiller is now known as bfiller_afk | ||
| * om26er looking | 17:48 | |
| rsalveti | sergiusens: why are you disabling support for unflipped? | 17:53 |
| rsalveti | sergiusens: we don't support it for daily anymore but people might still need them | 17:53 |
| sergiusens | rsalveti: I already did it in the tests | 17:53 |
| sergiusens | rsalveti: want it to stay for network? | 17:53 |
| rsalveti | right, but I think phablet-network-setup is still useful for the tagged images | 17:54 |
| rsalveti | remember we don't yet have any tagged image for flipped :-) | 17:54 |
| sergiusens | rsalveti: ok :-/ | 17:54 |
| rsalveti | sergiusens: I know it's a pain but we might still need to keep it a bit more | 17:55 |
| rsalveti | at least until we have a tagged image for flipped | 17:55 |
| ralsina | cjwatson: problem! For demo purposes we want it to work on the desktop because it's what we can hook to a projector :-) | 17:56 |
| ogra_ | rsalveti, per asac definition once the dashboard is all gree we have a tagged image every day ;) | 17:56 |
| rsalveti | ogra_: yeah yeah | 17:56 |
| rsalveti | haha :-) | 17:56 |
| ogra_ | *grin* | 17:56 |
| rsalveti | ogra_: I know they should all have enough quality, but it's still useful to tag once a month at least | 17:57 |
| ogra_ | yeah | 17:57 |
| rsalveti | we all love the automated tests, but that's not yet covering everything we need :-( | 17:57 |
| asac | rsalveti: it is quite good if all the autopilots land | 18:01 |
| rsalveti | asac: sure, but that's not yet covering everything we need | 18:01 |
| asac | at least from a system perspective we have a bunch of building blocks then that are testing quite a lot | 18:01 |
| asac | apps have to go deeper; we have to chek how we can carefully craft good integration tests that test our whole featureset etc. :) | 18:01 |
| asac | so yeah ... once we go green we have another big batch of pilots and then platform testing etc. in the pipeline somewhere | 18:02 |
| rsalveti | right | 18:02 |
| pmcgowan | yeah be great to get tests hitting the platform apis | 18:02 |
| rsalveti | yeah, that's the next big step | 18:02 |
| asac | and once all is good we just go and ensure that our daily-release testing gives more accurate confidence about what happens on our real images | 18:02 |
| sergiusens | rsalveti: updated https://code.launchpad.net/~sergiusens/phablet-tools/flipped_only/+merge/175625 | 18:05 |
| rsalveti | sergiusens: got approved already by om26er | 18:07 |
| === joseph_thegreat is now known as bobweaver | ||
| om26er | sergiusens, I was testing ;) | 18:07 |
| sergiusens | om26er: thanks | 18:08 |
| rsalveti | doanac: hm, a lot of RLIMIT_NICE not set in the camera app autotest, let me check that | 18:08 |
| rsalveti | not an issue, but hard to track the logs with them there | 18:08 |
| doanac | rsalveti: yeah. its been a little annoying | 18:09 |
| === joseph_thegreat is now known as bobweaver | ||
| pmcgowan | rsalveti, is the smoke dashboard up to date now or still in process? | 18:10 |
| rsalveti | pmcgowan: it's for maguro and mako | 18:11 |
| pmcgowan | rsalveti, so those are real failures now? | 18:11 |
| rsalveti | pmcgowan: yeah, investigating still if there's any infra related issue | 18:11 |
| pmcgowan | rsalveti, ok trying to test locally, takes a hile though | 18:11 |
| pmcgowan | and om26er is as well | 18:12 |
| pmcgowan | om26er, are you seeing the same failures as the smoke test? | 18:12 |
| om26er | pmcgowan, gallery-app worked fine for me. more apps tests running | 18:12 |
| pmcgowan | om26er, where does the output get logged, can I check it during the test? | 18:13 |
| om26er | pmcgowan, they are running outside of any jenkins job so there won't be any output to look at | 18:14 |
| DJJeff | NetworkManager[3566]: <warn> could not mark modem as powered: (null) The name org.ofono was not provided by any .service files | 18:15 |
| pmcgowan | om26er, so I ran gallery suite and got 22 fails out of 23 | 18:15 |
| pmcgowan | what am I doing wrong | 18:15 |
| om26er | pmcgowan, how did you run ? phablet-test-run -n ? | 18:15 |
| DJJeff | this error is cauing NetworkManager 100% CPU | 18:16 |
| pmcgowan | om26er, yes phablet-test-run -n gallery_app | 18:16 |
| pmcgowan | with latest from bzr | 18:16 |
| pmcgowan | latest phablet-tools | 18:16 |
| om26er | pmcgowan, run without -n 'phablet-test-run gallery_app' and before make sure you unlock the greeter | 18:16 |
| pmcgowan | om26er, trying again | 18:17 |
| doanac | pmcgowan: rsalveti: there's a couple of more maguro tests i'm re-running. the ones from 12:07 i think are bad | 18:20 |
| rsalveti | doanac: indeed | 18:20 |
| doanac | manta is in-flight. but the mediaplayer job is hung. i'm kicking that by hand | 18:20 |
| pmcgowan | doanac, mako tests for notes-app all failed at 12:07 as well | 18:22 |
| doanac | pmcgowan: okay. i'll kick that off by hand as well | 18:22 |
| om26er | for camera-app 1 test failed | 18:23 |
| om26er | (for me) | 18:23 |
| pmcgowan | om26er, when the tests run, do you see the UI get events? | 18:23 |
| pmcgowan | I just see it start and stop the app | 18:23 |
| om26er | pmcgowan, that's bad, It runs fine for me. I you can see the live video as tests run | 18:24 |
| pmcgowan | om26er, must be missing something here | 18:24 |
| pmcgowan | om26er, did you initialize with the test-run -i? | 18:25 |
| om26er | pmcgowan, make sure you have python-ubuntu-platform-api installed on the phone | 18:25 |
| om26er | pmcgowan, I installed things manually | 18:25 |
| pmcgowan | I suspected | 18:25 |
| pmcgowan | om26er, I have that | 18:26 |
| pmcgowan | om26er, same thing bfiller_afk was seeing | 18:27 |
| pmcgowan | om26er, MismatchError: After 10.0 seconds test on AbstractButton.opacity failed: 0.35 is not < dbus.Double(0.0, variant_level=1) | 18:27 |
| pmcgowan | they are all failing with mismatch cause nothing actually happened | 18:28 |
| === Namidairo is now known as Namidairo`bnc | ||
| davmor2 | pmcgowan, ogra_, popey: I'm going to try and reproduce this on .18 but I could the last few nights https://bugs.launchpad.net/touch-preview-images/+bug/1202794 | 19:08 |
| ubot5 | Launchpad bug 1202794 in touch-preview-images "Maguro: Apps lens sometimes keeps scrolling" [Undecided,Confirmed] | 19:08 |
| popey | not seen that | 19:08 |
| popey | will try | 19:08 |
| davmor2 | popey: you just have to keep opening and closing apps and eventually it gets stuck or scroll | 19:09 |
| popey | k | 19:09 |
| ogra_ | i havent been able to reproduce, i only had it on the completely broken shell | 19:14 |
| rsalveti | ogra_: I'll add back 45-binder.conf so we don't pollute the syslog much when running the autopilot tests | 19:16 |
| ogra_ | rsalveti, ++ | 19:16 |
| rsalveti | we fixed the issue when running the apps with the shell, but without the shell the limits will not be set correctly | 19:16 |
| rsalveti | not automatically | 19:16 |
| rsalveti | so that message will show again | 19:16 |
| ogra_ | well, cant we set the limits in the session job ? | 19:16 |
| rsalveti | we could fix in the jobs itself, but I'd prefer to just ignore it in the image by default | 19:17 |
| ogra_ | instead of the shell job | 19:17 |
| rsalveti | right | 19:17 |
| rsalveti | that might indeed be better | 19:17 |
| rsalveti | forgot we have a proper session now | 19:17 |
| rsalveti | let me give that a try | 19:17 |
| ogra_ | :) | 19:17 |
| mhall119 | ogra_: rsalveti: any ETA on when we'll have a working HUD again? | 19:18 |
| mfisch | pmcgowan: who is working on or thinking about desktop backgrounds? I'm specifically curious if we have a way to change it currently | 19:18 |
| popey | mfisch: hasn't Laney done something there... | 19:18 |
| ogra_ | mhall119, probably a question for ted | 19:18 |
| pmcgowan | mfisch, I think thats in settings somewhere, seb128? | 19:19 |
| ogra_ | mhall119, i have never seen the HUD from the inside :) | 19:19 |
| pmcgowan | unless its a gallery option | 19:19 |
| mhall119 | ogra_: hmm, looks like ted is out al week | 19:20 |
| davmor2 | pmcgowan, mfisch: there is something already added to settings but it is only demo by the look of it currently | 19:20 |
| pmcgowan | I expect thats right | 19:20 |
| mfisch | Yeah I'm planning on manipulating gsettings | 19:20 |
| mfisch | my manipulations this far have not produced a change | 19:20 |
| rsalveti | pmcgowan: bfiller_afk: which team manages the hud? | 19:21 |
| rsalveti | I know ted works with it, but it's broken for ages | 19:21 |
| pmcgowan | rsalveti, oh thought it was fixed? | 19:21 |
| davmor2 | n | 19:21 |
| mhall119 | pmcgowan: it was briefly, now it's broken again | 19:21 |
| pmcgowan | how can that be? | 19:22 |
| mhall119 | I think it was working on the pre-flipped images | 19:22 |
| mhall119 | not sure if it was on the flipped or not, I can't remember | 19:22 |
| rsalveti | don't we have autopilot tests for it? | 19:22 |
| davmor2 | pmcgowan: I saw buttons a couple of weeks ago not seen them since ref hud | 19:22 |
| kenvandine | mhall119, the background isn't changeable yet... but there is a MP in review that supports that | 19:22 |
| kenvandine | whoops | 19:23 |
| kenvandine | mfisch, | 19:23 |
| kenvandine | :) | 19:23 |
| mhall119 | kenvandine: that's okay, I was interested in the answer too :) | 19:23 |
| kenvandine | https://code.launchpad.net/~unity-team/unity8/unity8.background_gsettings/+merge/174958 | 19:23 |
| mfisch | glad I'm not the only one who does that | 19:23 |
| === bfiller_afk is now known as bfiller | ||
| rsalveti | doanac: we run the utah autopilot tests as phablet user, right? | 19:26 |
| ogra_ | rsalveti, yep | 19:27 |
| rsalveti | great | 19:27 |
| cjwatson | ralsina: I already flagged in the call that that was going to be a problem. But you can probably manage it if you temporarily install packagekit and remove aptdaemon. | 19:27 |
| doanac | rsalveti: yes | 19:27 |
| ralsina | cjwatson: ok | 19:29 |
| mfisch | pmcgowan: interestingly I can set the wallpaper and the settings app shows it, but not the desktop | 19:39 |
| pmcgowan | mfisch, makes sense, waiting on a shell fix prolly | 19:40 |
| nik90 | om26er: ping | 19:46 |
| om26er | nik90, pong | 19:46 |
| nik90 | om26er: you were discussing previously with spencer from the file manager regarding the toolbar icons | 19:46 |
| nik90 | om26er: he scaled it to 16x16 which caused it to be blurry | 19:46 |
| nik90 | om26er: would 32x32 be okay? | 19:47 |
| nik90 | We plan on doing something similar for the clock app to use the official icons rather than our own. | 19:47 |
| om26er | I am not sure about that since things that would look fine on a 720p phone may look blurred on high res tablets like Nexus 10 | 19:48 |
| om26er | all in all its better than what we have | 19:48 |
| nik90 | om26er: the problem is that the official icons are in svg format. This does not render well with qml | 19:48 |
| nik90 | so we are forced to convert to svg. | 19:48 |
| nik90 | should I convert them to 64x64 just to make sure? | 19:48 |
| om26er | png you mean ;) | 19:48 |
| nik90 | png yes | 19:49 |
| om26er | 32x32 is fine I guess, till people develop a better solution | 19:49 |
| nik90 | om26er: okay. thnx. Just cautious before I commit to trunk | 19:49 |
| om26er | does anyone know where are the sample videos that we have in phone images ? | 19:49 |
| ogra_ | om26er, dpkg -L demo-assets-videos | 19:50 |
| ogra_ | (on the phone obviously) | 19:51 |
| mfisch | pmcgowan: sorry to bug you again, but who can tell me how those "launcher" icons get on the left hand side, how can I add one? | 19:51 |
| seb128 | mfisch, once https://code.launchpad.net/~unity-team/unity8/unity8.background_gsettings/+merge/174958 merged you should be able to see the background change working | 19:52 |
| mfisch | seb128: thanks | 19:52 |
| seb128 | mfisch, I just approved it again, it failed merging today, which seems to be due to the jenkins issues | 19:52 |
| pmcgowan | mfisch, thats not in yet either afaik | 19:53 |
| pmcgowan | mfisch, need to bug Saviq | 19:53 |
| mfisch | seb128: perhaps you know, will the "launchers" (if that's what they're still called) be controlled by the old launcher gsettings key? | 19:54 |
| seb128 | mfisch, no idea about that, it's a question for Saviq I guess | 19:54 |
| seb128 | but seems like it should, no reason to break compat | 19:54 |
| mfisch | seb128: merci, I'll ping him | 19:55 |
| Saviq | mfisch, atm they're hardcoded, I'm afraid, we're getting closer and closer to actually reading the list from gsettings | 19:57 |
| mfisch | oh hey Saviq | 19:57 |
| ajalkane | Saviq: gsettings? No scanning of .desktop files as customary? | 19:58 |
| mfisch | Saviq: that's all I need to know, just the mechanism, take your time ;) | 19:58 |
| Saviq | ajalkane, not for launcher | 19:58 |
| mfisch | ajalkane: we always use gsettings for launchers | 19:58 |
| mfisch | ajalkane: you're thinking of the dash | 19:58 |
| ajalkane | aah... right of course | 19:58 |
| Saviq | ajalkane, "Installed" apps already use .desktop files - same app scope that on the desktop | 19:58 |
| kentb | is there a way to add a new user to my nexus 10? All I've got right now is guest | 19:59 |
| ajalkane | Saviq: that's great | 19:59 |
| Saviq | seb128, sorry, unapproved as we never approved it ourselves :) | 19:59 |
| Saviq | seb128, someone from our side needs to review | 19:59 |
| seb128 | Saviq, oh, sorry, I got confused by the jenkins crazyness | 20:00 |
| seb128 | Saviq, I saw that kenvandine comment approved it | 20:00 |
| mhall119 | wait a minute, do google contacts sync when using Online Accounts? | 20:00 |
| seb128 | Saviq, so I though it was approved and got bounced back due to broken panda boards | 20:01 |
| mhall119 | I didn't do anything, but all my contacts are in the address book | 20:01 |
| seb128 | mhall119, they don't "sync" but they are fetched online (afaik, I'm not sure but I think it's what kenvandine said) | 20:02 |
| mhall119 | that's cool, I wasn't expecting that to work | 20:03 |
| mhall119 | always nice to be pleasantly surprised like that | 20:03 |
| seb128 | ;-) | 20:03 |
| kenvandine | :) | 20:05 |
| mhall119 | Saviq: we also have the bug again where closing the first app thumbnail puts the next app into focus instead of staying on the Apps lens | 20:10 |
| ajalkane | why does this happen:The following packages have been kept back: linux-generic linux-headers-generic linux-headers-generic-pae linux-image-generic qtcreator qtcreator-plugin-ubuntu qtdeclarative5-ubuntu-ui-toolkit-plugin ubuntu-sdk | 20:11 |
| ajalkane | 0 upgraded, 0 newly installed, 0 to remove and 8 not upgraded. | 20:12 |
| ajalkane | i can't run some apps, and perhaps it's because of this (raring if that makes difference) | 20:12 |
| ogra_ | because they want to add or remove another package | 20:12 |
| ogra_ | but you only run apt-get upgrade instead of apt-get dist-upgrade | 20:13 |
| ajalkane | ogra_: any good hint how to resolve this conundrum? | 20:13 |
| Saviq | mhall119, TBH I don't know it ever landed | 20:13 |
| Saviq | ricmm, https://bugs.launchpad.net/touch-preview-images/+bug/1178288 ? | 20:13 |
| nhaines | Hooray for `apt-get dist-upgrade`! | 20:13 |
| ubot5 | Launchpad bug 1178288 in touch-preview-images "Closing application in dash causes focus on another application" [High,In progress] | 20:13 |
| Saviq | mhall119, ah, yeah it did, but is back | 20:13 |
| Saviq | mhall119, let me reopen | 20:13 |
| ajalkane | ogra_: i thought running dist-upgrade does more devious things | 20:13 |
| ogra_ | ajalkane, apt-get upgrade will never add or remove packages, only upgrade ... if a dependency changed it will hold the depending packages back | 20:13 |
| ajalkane | i should run dist-upgrade? i thought that dist-upgrade would change from raring to whatever comes next | 20:14 |
| ogra_ | so in that case you want dist-upgrade which allows removals/additions | 20:14 |
| ogra_ | no | 20:14 |
| ajalkane | and sorry about lack of capital i's, keyboard does not work | 20:14 |
| ajalkane | ok i'll have a go | 20:14 |
| ogra_ | the dist-upgrade will not upgrade you to the next release, no worries | 20:14 |
| ogra_ | (unless you manually edit your sources.list to actually point to a new distro) | 20:15 |
| Saviq | mhall119, actually, what I think happened, is I reverted that change... | 20:15 |
| Saviq | or maybe that wasn't it... | 20:15 |
| ajalkane | thanks, sounds safe :) | 20:16 |
| Saviq | mhall119, yeah https://bugs.launchpad.net/unity8/+bug/1178288/comments/5 | 20:16 |
| ubot5 | Launchpad bug 1178288 in touch-preview-images "Closing application in dash causes focus on another application" [High,In progress] | 20:16 |
| Saviq | ricmm, you remember bug #1178288 ? | 20:17 |
| Saviq | ricmm, I think you had a fix, even | 20:17 |
| ricmm | Saviq: I think I did, didnt I ? | 20:21 |
| Saviq | ricmm, I think you did indeed | 20:21 |
| ricmm | lemme check | 20:21 |
| mfisch | sergiusens: ping | 20:25 |
| sergiusens | mfisch: pong | 20:26 |
| mfisch | sergiusens: hey I have some odd zip files on my phone, 2 copies of some zip files in /data/media. Is that normal? | 20:27 |
| mfisch | sergiusens: saucy-preinstalled-touch x2 and 2 others | 20:27 |
| mfisch | so not copies | 20:27 |
| sergiusens | mfisch: might be flipped vs unflipped | 20:28 |
| mfisch | sergiusens: ah these are dated June 19 | 20:28 |
| mfisch | I should probably do a bootstrap install ;) | 20:29 |
| sergiusens | mfisch: one is touch and the other phablet perhaps | 20:29 |
| mfisch | the new files are dated yestefday | 20:29 |
| mfisch | yep | 20:29 |
| mfisch | so just from pre-flip | 20:29 |
| Saviq | mfisch, btw, re: launcher, there's one tricky thing to keep in mind and solve at some point - if you dock your phone and get desktop experience, should changes to the launcher propagate between form factors? ;) | 20:31 |
| mfisch | Saviq: interesting question | 20:32 |
| * mhall119 is so happy to have working camera and audio :) | 20:35 | |
| mhall119 | Saviq: why wouldn't it propagate? | 20:36 |
| mfisch | Has anyone else noticed that the clock in the top area and the one on the main screen don't match? Mine are off by 1 min | 20:39 |
| mfisch | with the screen one being 1 min slow | 20:39 |
| Saviq | mhall119, because you use different things on your destkop than on your phone | 20:40 |
| Saviq | mhall119, I'm not saying that or the other, I can see benefits to both | 20:41 |
| Saviq | mhall119, and disadvantages to both as well | 20:41 |
| Saviq | mhall119, and this is just one example | 20:41 |
| Saviq | mhall119, but really every single setting needs some thinking: "will it at all be useful in situation a, b, c, d" | 20:42 |
| Saviq | "should it be a per-device or per-screen setting" | 20:42 |
| Saviq | I can think of many more questions like that, that we need to answer in time for convergence | 20:42 |
| mhall119 | Saviq: should you be able to use the same things on phone and desktop? | 20:48 |
| Saviq | mhall119, well, that's convergence, isn't it? | 20:50 |
| Saviq | mhall119, the apps should adapt, is all | 20:50 |
| mhall119 | yeah | 20:50 |
| mhall119 | that's what I meant | 20:51 |
| mhall119 | I suppose you could hide any Launcher items that done have X-Ubuntu-Touch=true | 20:51 |
| Saviq | mhall119, sure, but that's not even that - you *can* use the same | 20:51 |
| Saviq | mhall119, but you rarely do | 20:52 |
| ryukafalz | Ubuntu Touch's messaging is based on Telepathy, right? | 20:52 |
| ryukafalz | If so, you may want to watch out for this bug right here: https://bugs.freedesktop.org/show_bug.cgi?id=47647 | 20:53 |
| ubot5 | Freedesktop bug 47647 in tp-qt "Tp-qt fails to remove temp avatar file" [Normal,Resolved: fixed] | 20:53 |
| ryukafalz | It's fixed, but not in a released version | 20:53 |
| mhr3 | Saviq, https://code.launchpad.net/~mhr3/unity8/add-to-empty-lvwhp/+merge/175682 | 20:55 |
| DJJeff | with ofonod stopped this error goes away https://bugs.launchpad.net/touch-preview-images/+bug/1202787 | 20:56 |
| mhall119 | Saviq: you rarely do right now, but we do hope to change that don't we :) | 20:56 |
| ubot5 | Launchpad bug 1202787 in touch-preview-images "[NetworkManager] High CPU Usage" [Undecided,New] | 20:56 |
| Saviq | mhall119, still, I can easily imagine people asking to have them separate | 20:57 |
| Saviq | mhall119, like, say, an office suite | 20:57 |
| Saviq | which is the obvious overused example | 20:57 |
| Saviq | mhr3, thanks! | 20:57 |
| Saviq | mhr3, I'd rather you move the check just before the call to .first(), though | 20:58 |
| Saviq | to group logically, and in most cases not even reach the isEmpty() | 20:59 |
| user82 | is there a new whatsapp client in progress? or does someone resume whosthere work | 20:59 |
| mhr3 | Saviq, damn, you're demanding... :P pushed | 21:00 |
| mhall119 | user82: there was a whatsapp client being written, but the company behind whatsapp asked us to stop | 21:00 |
| mhr3 | crap | 21:00 |
| Saviq | mhr3, crap what? | 21:00 |
| mhr3 | nothing :) pushed again :) | 21:01 |
| Saviq | mhr3, k;) | 21:01 |
| ajalkane | mhall119: there's an open source whatsapp client. Will whatsapp's asking to stop mean it can't be available in ubuntu phone store? | 21:06 |
| ajalkane | i'm asking because BB got that app too, but pulled it out after initially supporting it after whatsapp complained | 21:07 |
| popey | ajalkane: we're not promoting it | 21:07 |
| === boiko__ is now known as boiko | ||
| Andy__ | Hey can some one help me I erased my origibnal android 4.2.2 I installed the ubuntu touch it loaded but said error Make sure the device is connected and viewable by running 'adb devices' Ensure you have a root device, one which running 'adb root' does not return an error | 21:09 |
| Andy__ | when I sid run abd my device didn't show | 21:09 |
| Andy__ | Then since it didn't work I tried restarting the phone and it sticks at the google logo | 21:10 |
| Andy__ | any solutions I just need my phoine back I don't care about the data I lost | 21:11 |
| Andy__ | any help here? | 21:15 |
| user82 | mhall119, what does that mean. is that legally valid? | 21:18 |
| user82 | without whatsapp any mobile os is missing a big target audience. | 21:19 |
| mhall119 | user82: the app was having to extract auth tokens that kept changing, without official support from WhatsApp it would have been a constant battle to keep it working | 21:24 |
| user82 | mhall119, sucks. what did they say? "we give you more paying customers and make a free client you can make profit of" "nah stop that stuff"? | 21:24 |
| user82 | i assume the fee they want could've been implementend, since it is for their service, and not their software? | 21:25 |
| Saviq | mhr3, how about http://pastebin.ubuntu.com/5888875/ ? | 21:25 |
| k1l | hmm, that whatsapp point is very tricky imho. my girlfriend got a webos palm device that doesnt have a whatsapp client and that is a big negative. she would not buy any device now that doesnt have whatsapp :/ | 21:28 |
| nhaines | Andy__: I suspect a first step would be to let us know what device you're running and what happens when you run 'tail -f /var/log/dmesg' and then plug it in. | 21:28 |
| Andy__ | Well it's a android galaxy nexus | 21:29 |
| mhr3 | Saviq, oh the header tests don't actually test just the header? | 21:30 |
| mhr3 | anyway, pushed | 21:30 |
| Saviq | mhr3, header tests? there's one set of tests for a listview with section headers, another without section headers | 21:30 |
| AlanBell | k1l: someone was telling me about whatsapp the other day (I hadn't heard of it) it is free SMS messages or something right? | 21:31 |
| mhr3 | :/ saucy kernel doesn't support bumblebee | 21:31 |
| k1l | AlanBell: yes, its a messenger for Ios and android that works over the internet | 21:31 |
| mhr3 | Saviq, i read it as lvwph header test and lvwph section test | 21:31 |
| Saviq | mhr3, got it | 21:31 |
| AlanBell | why would I want that if I have thousands of texts a month and I send maybe 2/month? | 21:32 |
| Saviq | mhr3, so that lets you bring back diffing? | 21:32 |
| k1l | AlanBell: most people use that instead of sms now. because there is no fee per message but a very small fee per year for the app and most users already got a internet flat on their mobiles anyway | 21:32 |
| mhr3 | Saviq, no that makes unity not crash when you go from empty result set to some items :) | 21:32 |
| Saviq | mhr3, yeah, that too ;) | 21:33 |
| mhall119 | user82: they were most concerned about their brand and making sure any WhatsApp app looks like WhatsApp | 21:33 |
| nhaines | mhall119: I'd be rather happy to have LibreOffice on my Ubuntu phone. :) Not sure it's a good idea, of course. | 21:33 |
| k1l | AlanBell: that sms-flatrates are not that common. | 21:33 |
| mhall119 | nhaines: probably not LibreOffice, but Calligra could make a small-touch-screen interface | 21:34 |
| AlanBell | huh, most contracts have 5000 texts/month or something like that now | 21:34 |
| nhaines | mhall119: LibreOffice is already going to be installed anyway, so it makes sense to... erm... promote synergy. | 21:34 |
| k1l | AlanBell: no in here :/ | 21:34 |
| k1l | *not | 21:35 |
| k1l | AlanBell: most people i know swapped their sms-flat for a internet-flat | 21:35 |
| AlanBell | so is it a walled garden, all users have to have it to use it? | 21:36 |
| k1l | AlanBell: yes | 21:36 |
| AlanBell | well that sucks | 21:36 |
| k1l | same with the google hangouts | 21:36 |
| user82 | mhall119, okay. interesting. so a control issue, they feared what could happen with the client....no interest in a contract that they could influnce development? | 21:36 |
| k1l | AlanBell: i dont like that, too. whatsapp was free in the first place (for android) and then raised a fee for it. | 21:37 |
| user82 | 99cents/year. for the service they provide that is quite ok. security issues aside | 21:37 |
| k1l | AlanBell: but as i told you, people make their decision to buy a device (or more a mobile OS) on that fact. got it whatsapp? | 21:38 |
| AlanBell | so it is icky closed then | 21:38 |
| * AlanBell installs the thing anyway to have a look | 21:38 | |
| k1l | i know someone who got a no-touchscreen-nokia but that got whatsapp and facebook :/ | 21:38 |
| user82 | AlanBell, wahtsapp itself is not that interesting. it's the social stuff that comes with it. everybody you know uses it->you need to have it to stay in touch | 21:39 |
| user82 | plus sending pics/location etc is sometimes handy | 21:39 |
| AlanBell | heh, I had never heard of it until last week, I doubt I know anyone who has it | 21:40 |
| user82 | hahaha. talk to you later when whatsapp grabbed your contacts and scanned who has it | 21:40 |
| user82 | i'd say 30% at least | 21:40 |
| mhall119 | user82: they might have been interested in a contract if money was involved, but not to help an open source project no | 21:40 |
| k1l | user82: yep, thought the same | 21:40 |
| mhall119 | user82: I've just been telling anybody interested in having a WhatsApp client for Ubuntu Touch to contact them and let them know, if they see enough user interest they will either work with us and open source developers or port their own | 21:41 |
| user82 | mhall119, allright. thanks for the information! i still think there would not have been a big deal with payment in the open source software, since it is for their service | 21:41 |
| sergiusens | stgraber: I think I am almost done with the tool, just getting a slow download now | 21:41 |
| AlanBell | user82: ok, so maybe I do know some people who have it, including my daughter! | 21:42 |
| user82 | mhall119, funyy. i did that 5 minutes ago. very polite email, just saying i plan on using it. | 21:42 |
| mhall119 | user82: I meant them getting paid, not them paying the developer | 21:42 |
| k1l | AlanBell: your daughter will either use the facebook chat or whatsapp. or both :) | 21:42 |
| === MrDHat is now known as MrDHat|afk | ||
| stgraber | sergiusens: cool | 21:43 |
| stgraber | sergiusens: I have a big transparent squid proxy here to deal with that kind of things ;) | 21:43 |
| user82 | mhall119, you could just have said "yearly fee will be charged as in android and 100% of it trasnferred to you". free client with paying customers. anyway do not want to keep you from anything...just wanted to have some information | 21:43 |
| sergiusens | stgraber: probably need to set that up, but I rarely stay at one place for too long | 21:44 |
| k1l | AlanBell: i would like to see some open (encrypted) messenger spread all around the world, too. but its not the case yet | 21:44 |
| user82 | AlanBell, so you do know some, see. It is kind of peer preasure, i know some people would not sms me for fees, even some better friends. (broke students...i understand kinda). or not that often | 21:45 |
| user82 | i restircted the location and call log access for them..do not really want then to know everything | 21:45 |
| k1l | i only send sms to my mother who still doesnt have mobile internet. but all of the rest i send emails, hangout-chat or whatsapp. | 21:45 |
| AlanBell | yeah, I know some people use texts quite a bit, I tend to use my phone for the browser, google maps and ssh/irssi for IRC | 21:46 |
| AlanBell | I try to avoid talking to people (I turned off voicemail because it is annoying, and I often leave my phone off for days at a time) | 21:47 |
| user82 | ha even my parents started using hangouts and whatsapp... | 21:47 |
| k1l | AlanBell: the problem behind that is: webos showed (besides their mad pricing plans) what a new mobile OS needs to grab users: apps! | 21:47 |
| user82 | either apps or apples | 21:48 |
| AlanBell | I am quite looking forward to the Ubuntu touch stuff though, it probably does about all I need from a phoen | 21:48 |
| k1l | nokia did pay alot to make their windows phones cheaper than the android ones to get any customers. | 21:48 |
| AskUbuntu | Can I install Ubuntu desktop programs on my Nexus 4? | http://askubuntu.com/q/321640 | 21:48 |
| AlanBell | are there any details on the javascript APIs for phone services yet? like accessing the GPS position or making calls? | 21:49 |
| Andy___ | ho=i | 21:51 |
| Andy___ | hi* | 21:51 |
| AlanBell | I assume that there will be extensions to the API that does the laucher/notification/indicator/hud integration at the moment for phone related stuff | 21:51 |
| === Andy___ is now known as Guest15160 | ||
| AlanBell | the documentation for which appears to have gone AWOL http://developer.ubuntu.com/api/ubuntu-12.10/javascript/uwapi-ref.html | 21:52 |
| DJJeff | after that last "apt-get upgrade | 21:53 |
| DJJeff | it crashes after joining the Wireless AP | 21:53 |
| DJJeff | w00h00 progress!!!!!! | 21:54 |
| Saviq | AlanBell, http://developer.ubuntu.com/api/ubuntu-12.10/javascript/unity-webapps/ | 21:57 |
| Saviq | AlanBell, all of the current APIs should work on touch indeed | 21:57 |
| Saviq | AlanBell, as for extensions - probable, nothing planned at the moment I think | 21:57 |
| AlanBell | Saviq: cool, how did you find that link from http://developer.ubuntu.com? | 21:58 |
| Saviq | AlanBell, tbh I just removed the last part of your link | 21:58 |
| Saviq | AlanBell, but if it can't be found - mhall119 something for you maybe? | 21:58 |
| mhall119 | AlanBell: we're working on fixing the API links | 21:59 |
| AlanBell | http://developer.ubuntu.com/resources/platform/api/12-10/ link busted on that page then | 21:59 |
| mhall119 | actually we're going to be doing some major renovations of developer.u.c to make it easier to find the right info | 21:59 |
| Saviq | indeed | 21:59 |
| AlanBell | great | 21:59 |
| AlanBell | will I be able to create webapps with gu spacing and ubuntushape objects? | 22:00 |
| AlanBell | it was mentioned that HTML5 apps would be first class citizens (and that is a *much* better thing for me to generate) so I am curious what I will be able to do | 22:02 |
| mhall119 | AlanBell: ideally yes, but there's not really any way to add grid units without making non-standard changes to our browser | 22:02 |
| mhall119 | in QML we use a javascript function that returns an actual pixel value at runtime | 22:03 |
| DJJeff | WIFI only tablets need OFONO removed | 22:03 |
| DJJeff | to work :) | 22:03 |
| mhall119 | but you can'd add Javascript function calls to CSS | 22:03 |
| AlanBell | hmm, would need sass or less or something | 22:04 |
| AlanBell | there is a calc() function in css that can do some stuff | 22:05 |
| AlanBell | you can do width:calc(100% - 20px) for example | 22:05 |
| mhall119 | can you have variables in CSS? | 22:07 |
| mhall119 | if so, we can maybe do | 22:07 |
| mhall119 | width: calc(10 * gu) | 22:07 |
| AlanBell | nope, not without sass or less as a preprocessor | 22:07 |
| mhall119 | hmmm | 22:07 |
| AlanBell | less will run client side, so that could work | 22:08 |
| AlanBell | http://www.lesscss.org/ | 22:08 |
| AlanBell | basically it would be nice to be able to have a framework of javascript and CSS that would make a simple web page look precisely like it fits with the native app, so titles, drag up footer thing, typography, spacing pixel perfect | 22:14 |
| AlanBell | yeah, I think if you can put the size of a gu in the unity context object, so readable from javascript then the rest can be sorted out with less.js | 22:18 |
| AlanBell | and the chameleon colour reference would be nice too :) | 22:19 |
| AlanBell | presume that is a thing in Ubuntu Touch? | 22:19 |
| AlanBell | oh and are the APIs for pushing data to the blob sizes on the home screen exposed yet/going to be web-app accessible? | 22:23 |
| mfisch | stgraber: ping | 22:26 |
| stgraber | mfisch: pong | 22:26 |
| mhall119 | AlanBell: have you seen the HTML5 theme work that daker did? It has almost everything you just described | 22:30 |
| mhall119 | in terms of UI anyway | 22:30 |
| mhall119 | not APIs | 22:30 |
| rsalveti | jjohansen: hey, question for you, the android image (and container) sets up the nice rlimit to 40 by default, but in ubuntu the default max nice priority is set to 0 | 22:34 |
| rsalveti | jjohansen: binder contains a logic which increases the nice value depending on the message, so it can have a higher scheduler priority | 22:35 |
| rsalveti | jjohansen: in our case, if we don't set rlimit to something higher than 0, binder will work but complain that it could not set a custom nice value inside the binder driver | 22:36 |
| rsalveti | printing 'RLIMIT_NICE not set' | 22:36 |
| rsalveti | jjohansen: as we want our compat layer (which uses binder) to have the same rlimit as set by android, I added a 'limit nice 40 40' inside the ubuntu-touch-session upstart job | 22:37 |
| rsalveti | so any app started by the session/shell would have a max priority compatible with binder's logic | 22:37 |
| rsalveti | problem is that the same doesn't apply when you're connected via ssh, which is how we're running our autopilot tests | 22:38 |
| rsalveti | you can see the amount of noise (errors) in https://jenkins.qa.ubuntu.com/job/saucy-touch-mako-smoke-camera-app-autopilot/9/consoleFull, for example | 22:39 |
| rsalveti | so first question is if it's fine for us (from ubuntu's perspective) to also set rlimit nice to 40 40 by default (soft hard), at least for the touch session | 22:41 |
| rsalveti | and second is if you know a way we can set a default limit for the 'phablet' user, or even generically for the system | 22:41 |
| rsalveti | seems /etc/security/limits.conf is not actually working when connecting via ssh | 22:42 |
| rsalveti | mdeslaur: maybe something you can help as well ^ | 22:42 |
| jjohansen | rsalveti: /etc/security/limits.conf | 22:45 |
| jjohansen | rsalveti: oh hrmm, thats interesting | 22:46 |
| rsalveti | maybe I'm missing something, let me try forcing it to '*' | 22:47 |
| rsalveti | so it can be the default | 22:47 |
| jjohansen | rsalveti: Added explicit 'UsePAM yes' to /etc/ssh/sshd_config. | 22:48 |
| jjohansen | as for defaulting to 40 40 I don't like that | 22:48 |
| rsalveti | right, that's the default value used in android | 22:49 |
| jjohansen | yes I get that, binders running in the container right? | 22:49 |
| jjohansen | we could possibly default it for just the container | 22:49 |
| rsalveti | we don't necessarily need to set default to 40, seems anything higher than 0 would already make binder "happy", as it'll then use whatever max value it can | 22:49 |
| jjohansen | so android bits get 40 40 | 22:49 |
| rsalveti | right, but we have stuff running in ubuntu which uses binder | 22:50 |
| jjohansen | so why is binder complaining if its 0, what is the logic to that? | 22:51 |
| rsalveti | let me get the link to the code | 22:52 |
| rsalveti | jjohansen: http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-saucy.git;a=blob;f=drivers/staging/android/binder.c;h=85ed6cfd18038341ca607afe46e8c3a8b3cf8759;hb=refs/heads/mako | 22:53 |
| rsalveti | line 516 is the function that sets up the nice values | 22:53 |
| sergiusens | stgraber: just noticed if ubuntu_command has a typo we go into infinite loop | 22:55 |
| sergiusens | reboot loop | 22:55 |
| stgraber | sergiusens: yep, I noticed that :) | 22:55 |
| annerajb | which repository is ubuntu_deploy.sh rsalveti ? | 22:56 |
| rsalveti | jjohansen: so it tries to set up to a desired nice value (seems it can even be the priority used by the other end, which can come from android), if it fails, it then tries to set up a min_nice value | 22:56 |
| stgraber | sergiusens: easy enough to fix, I'll just have the script moved the command file to ubuntu_command.applying as the first thing it does | 22:56 |
| stgraber | sergiusens: will send you a patch in a minute | 22:56 |
| jjohansen | rsalveti: so part of this seems to be the difference between user nice and knice values. The knice value range from 40..1 | 22:58 |
| jjohansen | user nice is -20..19 | 22:58 |
| jjohansen | that is IIRC | 22:58 |
| rsalveti | right | 22:58 |
| rsalveti | rlimit's nice priority is from 1..40 | 22:58 |
| jjohansen | setting 40 for the system as a default is a no go | 22:59 |
| jjohansen | setting binder to 40 by default is fine | 22:59 |
| rsalveti | sure | 22:59 |
| rsalveti | in android 40 is default because any process can use binder | 22:59 |
| rsalveti | and I don't think that this is exported to the java stack | 22:59 |
| rsalveti | this I mean I don't think any app can take up to 40 | 23:00 |
| AlanBell | mhall119: nope, not seen the theme stuff from daker, got a link? | 23:00 |
| thomi | hmm, running phablet-test-run this morning seems to hang and do nothing. I've re-flashed the device, and updated phablet-tools. Any other ideas? | 23:00 |
| sergiusens | thomi: there's an unreleased update from today | 23:01 |
| sergiusens | cyphermox_: kenvandine can you do a daily release for phablet-tools? | 23:01 |
| thomi | sergiusens: I'm not sure what that means, sorry | 23:01 |
| mhall119 | AlanBell: not sure how up to date it is, but http://people.ubuntu.com/~daker/ubuntu-html5-theme/ | 23:01 |
| thomi | oh, I see | 23:01 |
| sergiusens | thomi: bzr branch lp:phablet-tools and use that ;-) | 23:01 |
| thomi | phablet-tools won't work until something gets released | 23:01 |
| mhall119 | if you enable touch events in Chromium, you'll see more of how they work | 23:01 |
| thomi | sergiusens: gotchya, thansk | 23:01 |
| thomi | *thanks even | 23:02 |
| thomi | sergiusens: I see you're joining us in Boston? | 23:02 |
| stgraber | sergiusens: didn't test it, but looks right: http://paste.ubuntu.com/5889077/ | 23:02 |
| rsalveti | jjohansen: 'phablet soft/hard nice -20' works in limits.conf (via ssh), | 23:02 |
| daker | mhall119: it does also work with a mouse now, not perfect but it works | 23:02 |
| rsalveti | jjohansen: but don't think we want that either | 23:02 |
| mhall119 | daker: ah, thanks, is the link I gave relatively up to date? | 23:02 |
| jjohansen | rsalveti: nope | 23:03 |
| mhall119 | or is there somewhere better to point people | 23:03 |
| AlanBell | that is *great* | 23:03 |
| AlanBell | unfinished, but exactly what I was hoping for | 23:03 |
| mhall119 | AlanBell: that's part of the Ubuntu SDK now :) | 23:03 |
| jjohansen | rsalveti: so are we seeing a 0 in the kernel code? | 23:04 |
| rsalveti | jjohansen: yes | 23:04 |
| daker | mhall119: i'll try to put the last update | 23:04 |
| rsalveti | jjohansen: when debugging here 'binder_set_nice' had 0 and -8 as values | 23:05 |
| daker | mhall119: there is a MR for suru-dark http://people.ubuntu.com/~daker/ubuntu-html5-theme/examples/widgets/Toggles.html | 23:05 |
| mhall119 | daker: nice | 23:05 |
| mhall119 | daker: how about suru-gradient? | 23:05 |
| mhall119 | can we do that with css? | 23:06 |
| daker | in the pipeline ;) | 23:06 |
| mhall119 | you rock man! | 23:06 |
| jjohansen | okay, so the binder driver is broken and needs to be fixed | 23:06 |
| jjohansen | rsalveti: 0 is how the kernel represents no limit | 23:07 |
| daker | mhall119: http://i.imgur.com/DJhD8jM.png | 23:08 |
| rsalveti | jjohansen: don't know if there's any easy solution here, we could also just ignore the error message, as the rest of the system shold have a max limit of 0 | 23:08 |
| rsalveti | just the android container that has 40 by default (entire container) | 23:08 |
| rsalveti | but I believe in the android model that was allowed so binder could have a higher priority than the rest of the system | 23:09 |
| jjohansen | rsalveti: well its odd that we are seeing 0 as a nice value, we shouldn't, hrmm unless maybe realtime I am not sure how realtime priority is represented atm | 23:09 |
| jjohansen | right binder wants to be highest priority | 23:10 |
| jjohansen | but why are we seeing 0 else where | 23:11 |
| === _salem` is now known as _salem | ||
| rsalveti | jjohansen: but it seems 0 it's not 'no limit' here | 23:13 |
| rsalveti | it calls set_user_nice with 0 | 23:13 |
| rsalveti | which can be from -20 to 19 | 23:13 |
| rsalveti | isn't 0 just the default we use? | 23:13 |
| jjohansen | right, but 0 should be a value we see in kernel code | 23:14 |
| sergiusens | thomi: seems so | 23:15 |
| jjohansen | a 0 in userspace is 20 in the kernel | 23:15 |
| sergiusens | stgraber: also, the checksum in the json index, is that sha256? | 23:15 |
| rsalveti | jjohansen: right, just looking at http://lxr.free-electrons.com/source/kernel/sched/core.c?a=mips#L3650 | 23:16 |
| rsalveti | and in our case rlim_cur is 0 by default, which will make the binder code to call set_user_nice with 20, which just makes it to return without changing anything | 23:17 |
| stgraber | sergiusens: yep | 23:17 |
| sergiusens | good :-) | 23:19 |
| jjohansen | so, I see it as binder needs patched | 23:20 |
| rsalveti | jjohansen: right, but what would you change there, just removing the error? | 23:21 |
| rsalveti | otherwise we could change that logic to set up the nice anyway (adjusting the limits), but still not safe | 23:25 |
| jjohansen | rsalveti: I don't see how why the error should happen | 23:25 |
| jjohansen | min_nice = 20 - current->signal->rlim[RLIMIT_NICE].rlim_cur; | 23:25 |
| jjohansen | is converting for kpri to user pri, so from 1..40 to -20..19, we shouldn't get a value >= 20 | 23:25 |
| jjohansen | hrmm, no I guess actually do get 20 | 23:26 |
| rsalveti | jjohansen: rlim_cur is 40 | 23:26 |
| rsalveti | yeah | 23:26 |
| rsalveti | ops, 0 actually | 23:26 |
| rsalveti | binder: nice value 0 not allowed use 20 instead | 23:27 |
| jjohansen | rsalveti: what is 0? rlim_cur? | 23:27 |
| rsalveti | right | 23:27 |
| Noize | Does Ubuntu for Android have it's own channel? If so, could you direct me? | 23:28 |
| jjohansen | ah no, that message is post conversion, thats the min_nice value | 23:28 |
| popey | Noize: no, it doesnt | 23:28 |
| rsalveti | yeah, but for it to be 20 rlim_cur needs to be 0 | 23:28 |
| Noize | popey, how do i contact devs for that? for like help and such? | 23:29 |
| jjohansen | rsalveti: okay, right I was looking at the wrong value in the message. | 23:30 |
| jjohansen | rsalveti: so I need to poke more as to the rlim values | 23:31 |
| rsalveti | jjohansen: right | 23:32 |
| rsalveti | jjohansen: want me to open a bug for you? | 23:32 |
| jjohansen | rsalveti: sure | 23:32 |
| mdeslaur | rsalveti: what are you setting the rlimit in limits.conf? | 23:37 |
| rsalveti | mdeslaur: not setting yet, was just testing if that would work in our tests | 23:37 |
| mdeslaur | rsalveti: nice at 1 or more should work | 23:38 |
| rsalveti | yeah | 23:39 |
| mdeslaur | so you need to set limits.conf to -20, for example to get the full range | 23:39 |
| mdeslaur | the conversion is pretty stupid | 23:40 |
| rsalveti | right, in our case -1 would already get rid of that error, but I also wanted to discuss what could be the proper solution | 23:40 |
| rsalveti | :-) | 23:40 |
| mdeslaur | stgraber: can you set an rlimit for a container? | 23:44 |
| rsalveti | mdeslaur: the android one has 40 40 by default | 23:44 |
| rsalveti | mdeslaur: http://paste.ubuntu.com/5889171/ | 23:45 |
| mdeslaur | rsalveti: the android container, or android itself? | 23:45 |
| rsalveti | this is from android's init | 23:45 |
| rsalveti | container | 23:45 |
| === Aww is now known as [[Aww]] | ||
| mdeslaur | and you're still getting the error message? | 23:45 |
| mdeslaur | that's odd | 23:45 |
| stgraber | mdeslaur: it may be kept per pidns, if so, yes, but I'm really not sure. Pinging hallyn about it. | 23:45 |
| rsalveti | mdeslaur: I'm getting that error because I'm using an app that needs binder from the ubuntu side | 23:46 |
| mdeslaur | rsalveti: whats PID 337 in your example, is it binder? | 23:46 |
| rsalveti | to talk with the service available in the android side | 23:46 |
| rsalveti | tjat | 23:46 |
| mdeslaur | rsalveti: oh, hrm | 23:46 |
| rsalveti | that's android's init service | 23:46 |
| mdeslaur | rsalveti: are we compiling our own binder in our images? | 23:46 |
| rsalveti | bug 1202887 | 23:47 |
| ubot5 | bug 1202887 in linux-mako (Ubuntu) "'binder: RLIMIT_NICE not set' when using binder from the ubuntu side" [Undecided,New] https://launchpad.net/bugs/1202887 | 23:47 |
| mdeslaur | rsalveti: *cough*killitwithfire*cough* | 23:47 |
| rsalveti | mdeslaur: binder is a kernel driver, we just use it from the ubuntu side | 23:47 |
| rsalveti | hahah | 23:47 |
| sergiusens | stgraber: we certainly want that patch, even after a successful update the ubuntu_command just stays there | 23:51 |
| rsalveti | jjohansen: the bug ^ | 23:51 |
| rsalveti | also added affects to all android kernels we use | 23:51 |
| rsalveti | mdeslaur: binder is heavily used by android when you need to talk with most services | 23:54 |
| rsalveti | most jni code uses binder in the end, to talk with a service done in c++, for example | 23:54 |
| rsalveti | currently that's used by us to play any kind of media, use camera, communicate with the app manager and so on | 23:55 |
| sergiusens | stgraber: rsalveti https://code.launchpad.net/~sergiusens/phablet-tools/image_updates/+merge/175708 | 23:55 |
| DJJeff | any fix for this http://paste.ubuntu.com/5888986/ | 23:58 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
