=== chriadam|away is now known as chriadam [00:28] hello :) [00:29] is here any czech? === duflu_ is now known as duflu === jono is now known as Guest13828 [02:12] can any one help me to install Ubuntu touch in my SONY TABLET S [02:12] Though I don't see it in UBUNTU device list [02:16] can any one help me to install Ubuntu touch in my SONY TABLET S === dlan is now known as dlan^ === vying is now known as Guest21095 === vying is now known as Guest23399 [08:02] pitti, good morning :) [08:02] hey tvoss, wie gehts? [08:02] pitti, hey :) gut geht's und selber? [08:03] tvoss: gut, danke! wrestling with upgrade testing.. [08:03] pitti, sounds like fun :) [08:03] tvoss, what is already written in Go in the current touch images? I saw ogra mentioning that yesterday [08:04] janimo, usensord is written in go, I *think* some tooling as well [08:04] tvoss, ok, I knew about the tooling but not about the daemon [08:04] thanks [08:04] janimo, cool [08:18] guys i have a nexus 4 and i wanna try ubuntu on my phone, any one can say about battery life ? === LarrySteeze is now known as LarrySteeze|Away [08:32] Ciao ragazzi volevo sapere se era possibile installare ubuntu touch su note 2 [08:34] guys any one can say about ubuntu phone battery life ? [08:34] good morning [08:35] good morning [08:43] dholbach: do you know if/how it's possible to get information about the click package/application/version from QML? [08:43] dholbach: or C++ Qt, even [08:43] mardy, good question - I don't know [08:43] Kaleo: maybe you know? ^ [08:44] it'd surprise me if there was a way, but it looks like worth having [08:44] dholbach: I think that QCoreApplication::applicationName() holds the application part of it, but not the package or version [08:45] maybe worth bringing up on ubuntu-phone@? [08:47] dholbach: nope, I just found some info in http://developer.ubuntu.com/publish/apps/security-policy-for-click-packages/ [08:47] dholbach: see at the end of the page: there is an APP_ID environment variable which contains the full app ID [08:47] aha! [09:09] janimo, what tvoss said [09:09] ogra_, ack [09:10] i think there are plans for powerd to go go as well at some point [09:10] gogo powerd ;) [09:12] do we have a mediaservice bzr branch yet? [09:15] mpt: do we have a set order for the indicators on touch? I couldn't find anything specific other than https://wiki.ubuntu.com/StatusBar [09:16] asac, do you mean mediascanner ? https://code.launchpad.net/ubuntu/+source/mediascanner [09:17] ogra_: no i am talking about mediaservice that jhodapp|afk is working on [09:18] asac, ah, not sure about that one (i thinnk he is still debugging video issues on 4.4) [09:18] asac, https://code.launchpad.net/~jhodapp [09:18] there is a media-hub === JamesTait is now known as Guest81255 === Guest81255 is now known as JamesTait [09:22] tvoss: ping [09:24] ogra_: yeah. that sounds right [09:41] tsdgeos, pong [09:41] tvoss: seen the comment in https://code.launchpad.net/~thomas-voss/unity-mir/refactor-oom-score-adj-to-rely-on-process-cpp/+merge/201145 ? [09:41] tsdgeos, nope [09:42] tvoss: it's crashing, we were planning with sil2100 and mzanetti to rebuild the silo without that one [09:42] tsdgeos, fixing [09:42] tvoss: but i'm pretty sure it's just the missing initiaization [09:42] so if you can get it done [09:42] tsdgeos, yup, branching [09:42] maybe we can get sil2100 not to remove the MR but to just rebuild it [09:42] sil2100: ok, don't remove it then :D sorry ^_^ [09:43] tsdgeos: sil2100: I already removed it from the spreadsheet [09:43] wait wait [09:43] https://bugs.launchpad.net/ubuntu/+source/unity8/+bug/1262711 has been around for some time but is "Incomplete" ☹ [09:43] we're getting tvoss to fix it [09:43] Launchpad bug 1262711 in unity8 (Ubuntu) "Unity crashes with lots of music displayed in expanded music category in home scope" [Undecided,Incomplete] [09:43] hi === James is now known as Guest61737 [09:43] mzanetti: add it back :D [09:44] can i install ubuntu Phone on my nokia lumia 1020 [09:44] tsdgeos, all of the Application* should be initialized to nullptrs if I understand it correctly [09:45] tsdgeos, cancel that, that's done in the ctor [09:45] tvoss: except for the one i mention [09:45] it was [09:45] but then it was removed [09:45] in your MR [09:45] , m_focusedApplication(nullptr) [09:45] tsdgeos, yup, I merged in the sidestage and mainstage changes, dropped it while doing that [09:47] tvoss: we need it back :-) [09:48] tsdgeos, pushed [09:51] mzanetti: sil2100: ok, i've re-approved it. Can we get it back to the silo and get a rebuild? [09:59] tsdgeos: I added it back to the spreadsheet... not sure how to rebuild [09:59] mzanetti: let's wait for sil2100 [09:59] he was in a meeting [10:00] Back [10:00] tvoss, tsdgeos: ok, so that I get the right picture - is it not required to remove that MR now? [10:00] sil2100: no, the MR has been updated to fix the problem [10:00] sil2100: we just need a rebuild [10:00] s/no/yes [10:00] depending to which question i'm answering :D [10:01] yes, you get the picture right, no, the MR should not be removed [10:01] tvoss, tsdgeos: if you just updated, you can rebuild it now without any problems - if only unity-mir changed you can press 'Rebuild' and set unity-mir in the list of components to build ;) [10:01] It will then not waste the time with rebuilding unity8 as well [10:02] sil2100: never done that, can you be a bit more specific? [10:02] * sil2100 wonders if tsdgeos has the power to do that [10:02] probably not [10:02] sil2100: mzanetti should have i guess [10:03] he went through all the teaching and stuff [10:03] mzanetti <-> sil2100 [10:03] Right ;) And I guess he'll be filling in for Saviq as the lander this week [10:03] that's the plan [10:04] mzanetti: can you go to silo landing-004, click Build and enter unity-mir in the "PREPARE_ONLY" field? [10:04] mzanetti: this way you'll rebuild only unity-mir and things will start working again [10:04] tsdgeos, tvoss, mzanetti: thanks guys! [10:05] Just remember that, sadly, you'll have to re-test everything again according to your test plans [10:05] popey, it was implied by the list being a numbered one, but I’ve made it explicit now. [10:06] sil2100: sure [10:07] thanks mpt [10:12] sil2100: ok. done. thanks [10:12] dammit. no [10:12] * mzanetti is missing job/build permission [10:12] sil2100: ^ [10:13] ouch! That should not happen, let me add you [10:13] * sil2100 thought didrocks did that [10:14] hm, wait, I see you're added [10:15] mzanetti: make sure that when you're in the jenkins job and log-in you check the checkbox around ci-train-users [10:15] mzanetti: try maybe re-logging [10:15] sil2100: that's it. thanks [10:15] building now [10:16] tsdgeos: ^ [10:16] coo [10:16] mzanetti: do you know how to watch it build? [10:17] \o/ [10:17] I think the spreadsheet would update the status from Preparing to "ready" or something [10:17] besides obviously watching the jenkins job === chriadam is now known as chriadam|away [10:30] mzanetti: ah the building link points to http://162.213.34.102/job/landing-004-1-build/15/console :) [10:31] hey sil2100, how are you doing? I'm just helping out here and keep on forgetting what the right process is to get something autolanded... can you help me understand what I need to do to get ubuntu-html5-theme and unity-webapps-qml into trusty and subsequently on the image? O:-) [10:34] dholbach: hello! I'm fine, thanks, how about you? :) As for landing things into the archive and image, currently we are using something that's called CITrain - it's a new approach, but most of our upstream projects use it [10:34] dholbach: it requires some 'basic training' on the concepts, I'm sure all of our upstreams are supposed to be trained and switched to it soon [10:35] sil2100, doing well - either the jetlag hasn't caught up with me yet or I'm actually good for now :) [10:38] sil2100, do we have some docs about it somewhere? I didn't receive training, but would like to help the webapps guys - it also might help, so I don't need to ask every time about a process I don't know :) [10:55] Morning all [10:58] robru, do you know what needs to get done to land unity-webapps-qml and ubuntu-html5-theme? the former seems to had the last upload reverted, but to me none of the changes from last time look like they could have broken anything (at least from a packaging side of things) [10:58] dbarth, ^ [10:59] dholbach, i think it pulled new packages into the desktop [11:00] (which have not been MIRed before or some such) [11:05] greyback, hey there [11:06] MacSlow: heya [11:20] hi [11:20] i have a question for ubuntu phone developers [11:22] do you consider releasing "ubuntu for phone" for samsung galaxy s3 SHV-E210K? [11:25] Cihan, thats something the community needs to do [11:25] !devices | Cihan [11:25] Cihan: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [11:25] there might be a port on that page [11:25] (more or less working) [11:28] Thank you all. I had seen the list before already and i was wondering is there an idea to expand the device list with SHV-E210K? [11:29] But Thanks For replies. Have a great day to you all. === _salem is now known as salem_ [11:47] ogra_, hum like which? [11:47] dholbach, dunno, ask stgraber [11:48] he did the revert iirc [11:48] ubuntun-html5-theme -> ubuntu-html5-ui-toolkit maybe? [11:48] could be [11:50] anyone know why my /etc/fstab is being overwritten when i reboot device? ie. how do i add a auto mount on touch device? [11:50] ogra_: ^ ? [11:50] your fstab is coming from android iirc [11:52] well, from android an generation of the writable paths [11:52] hm [11:52] what do you want to automount ... and how ? [11:53] * ogra_ isnt even sure we run the USB port in host mode at all [11:54] i need to do something very dodgey to get some more space available on my root part :) eg. mount /usr/share at ~/usr/share [11:54] you could use my roostock script and hack up the size of the image in the script [11:55] in fact your mail got me thinking about adding a --size option to it for such cases :) [11:55] that sounds promising! [11:55] jdstrand: ping [11:56] ogra_: yeah, that would be really useful [12:00] ogra_: [34465.164750] usb 1-1: Product: DataTraveler 3.0 ;-) [12:00] usb works as host, I would say [12:01] ogra_: if I connect a usb-stick. it will be /dev/sda it works on the Nexus 5 ootb :) [12:02] dednick, grab the system.img files for your device from http://cdimage.ubuntu.com/ubuntu-touch/daily-preinstalled/current/ ... also grab the tar.gz rootfs file ... then grab rootstock-touch-installer and change the line: [12:02] do_shell "dd if=/dev/zero of=/data/system.img seek=500K bs=4096 count=0 >/dev/null 2>&1" [12:02] to read: [12:02] do_shell "dd if=/dev/zero of=/data/system.img seek=1000K bs=4096 count=0 >/dev/null 2>&1" [12:02] then run rootsotck-touch-install as described ... [12:03] that should make you end up with a 4G image [12:03] ogra_: cool. thanks, i'll give it a try [12:11] FuLgOrE_, i doubt dednick is on a nexus5 and i dont think all devices have the HW or the feature in the kernel [12:12] iirc galaxy nexus cant doe host mode at all and n4 needs an intrusive kernel patch which we might or might not ship [12:12] ogra_: as fas as I know for example the N4 should also be able to use usb host [12:12] oh [12:12] i'm on N10 [12:13] I thought you just need to connect 5V additionally to supply USB devices on N4 [12:15] sil2100: mzanetti and me have run autopilot and manual tests and all is green, what do we do now? [12:15] well, he ran manual tests, me only autopilot [12:17] yeah. I did manual testplans for unity-mir and unity8. everything working as expected [12:17] tsdgeos: ok, so now I'll try landing it, but we need to wait around 1h since we want to kick a new image with the nested mir thing [12:18] mzanetti, tsdgeos: thanks! [12:18] ok [12:18] no problem [12:18] It will land today for sure [12:18] \o/ [12:18] thanks sil2100 === MacSlow is now known as MacSlow|lunch [12:19] np :) Nice to see everything resolved === rachelliu_ is now known as rachelliu === boiko_ is now known as boiko [12:34] mpt, you just swamped my inbox :) [12:34] you and everyone else ☻ [12:35] tvoss: you make that sound like it doesn't happen daily ;) [12:37] popey, :) === jhodapp|afk is now known as jhodapp === tito_ is now known as tito [12:51] tvoss, don’t worry, just 181 to go [12:52] popey, you might know this, in the bug you reported about a missing file in an update, are we already using push notifications for the updates? [12:56] mandel: i dont think we have any kind of push notification.. it's just polling when you open settings (I believe) [12:56] mandel, no push notifications yet [12:57] popey, ok, 'cause that could be a change.. I'll ping barry, I have already pinged gatox to provide info about the changes in system settings [12:57] beuno, ack, thx for the info === Sleepitay is now known as Shiggitay [13:02] ogra_: getting an error from script. "adding android system image to installation ... mount: could not find any free loop device" [13:02] you only changed the one line ? [13:02] switching from 500K to 1000K ? [13:08] stgraber, if you have a bit of time, could you tell me what needs to be done to unblock the landing of lp:unity-webapps-qml? [13:10] ogra_: sorry, i was using all loop devices :/ [13:10] mardy: hey [13:11] dednick, ah, phew, so not the scripts fault then :) [13:12] jdstrand: hi, it's about bug 1278859 [13:12] bug 1278859 in apparmor-easyprof-ubuntu (Ubuntu) "Click apps need access to their own Online Accounts files" [Undecided,New] https://launchpad.net/bugs/1278859 [13:12] mpt, seriously? [13:14] mardy: so, the files in @{HOME}/.local/share/accounts/ are symlinks to the actual file? [13:15] files* [13:15] jdstrand: yes [13:16] ciao [13:16] jdstrand: without the apparmor rule to allow access to /.local/share/accounts/**, the access to the real file is denied, even if I think the app should be able to read them (because it came with its click package) [13:17] ok, that means that the account providers can't run/use the others, so that is good, however it does mean that the providers can see what other providers are installed [13:18] mardy: the symlink has to be dereferenced and for that you need read on the symlink [13:18] jdstrand: seeing what providers are installed is not a problem, but I disagree with the first part of yoru sentence: "the account providers can't run/use the others," [13:18] mardy: who is the one dereferencing the symlink? [13:18] jdstrand: this is indeed what we'd like, but it doesn't seem to happen [13:18] ogra_: unfortunately that error was with 500K, with 1000K i get: http://pastebin.ubuntu.com/6914945/ [13:18] mardy: ok, well lets talk through this [13:19] jdstrand: the application reads the symlinks, via libaccounts (which is just a library) [13:19] mardy: ~/.local/share/accounts/foo -> ~/.local/share/foo/provider [13:19] ~/.local/share/accounts/bar -> ~/.local/share/bar/provider [13:19] mardy: is it setup something like that? [13:19] jdstrand: yes [13:20] so, 'bar' should not have read access to ~/.local/share/foo/provider [13:20] jdstrand: I did a quick test (maybe wrong?), and it seems to me that if an app has access to ~/.local/share/accounts/foo, then it can read also ~/.local/share/foo/provider, wherever that is [13:20] so while 'bar' can see that ~/.local/share/accounts/foo points to ~/.local/share/foo/provider, it shouldn't be able to actually use ~/.local/share/foo/provider [13:20] jdstrand: ideally, yes [13:21] jdstrand: but it doesn't seem to work that way [13:21] that would be a grave bug in apparmor, but I really doubt that is the case [13:22] apparmor is supposed to dereference symlinks [13:22] jdstrand: does apparmor have a cache? [13:23] there are times when it doesn't have to revalidate [13:23] but I don't think that is it [13:23] jdstrand: I ran the app, which had access to the symlink and the target file, then I changed the target file to be somewhere the app shouldn't be able to read it, then I started the app again [13:24] can you show me the app's policy? [13:24] jdstrand: well, let me reboot :-) === MacSlow|lunch is now known as MacSlow [13:24] jdstrand: OK, I'll upload it somewhere [13:25] jdstrand: http://paste.ubuntu.com/6914983/ [13:25] dednick, hmm [13:26] dednick, is that a clean system ? or did you hack up anything on it before [13:27] mardy: can you paste 'ls -l ~/.local/share/accounts/' [13:30] jdstrand: lrwxrwxrwx 1 phablet phablet 86 Feb 11 13:48 .local/share/accounts/services/com.ubuntu.developer.webapps.webapp-facebook_webapp-facebook.service -> /home/phablet/tmp/com.ubuntu.developer.webapps.webapp-facebook_webapp-facebook.service [13:30] jdstrand: I moved the target file under ~/tmp/; I guess the app shouldn't be able to read that [13:31] no, it shouldn't [13:31] mardy: can you give the output of 'sudo aa-status' [13:32] jdstrand: http://paste.ubuntu.com/6915012/ [13:32] mardy: is there a way you can run that command when the app is accessing the provider? [13:34] ogra_: it was clean [13:34] jdstrand: ooops, I just tried now with aa-exec and strace, and it seems like that symlink is not being read now [13:35] jdstrand: I might have some caching in libaccounts-glib which I forgot about :-) [13:35] mardy: how are you reloading the profile? [13:35] weird, looks to me like either your /data partition doesnt have 4G free space or there was something mounted on /cache/system already [13:35] jdstrand: I rebooted the device; do I have to reload the profile, or are the profiles reloaded at boot? [13:35] dednick, your device is in recovery mode when running the installer script ? [13:36] mardy: the profiles are loaded on boot, but if you are modifying click package policy, you have to be aware of a couple of things [13:36] ogra_: yeah. it works with 500K, but not 1000K [13:36] that is strange [13:36] mardy: if you modify /var/lib/apparmor/profiles/click_...foo, then you can use apparmor_parser -r /var/lib/apparmor/profiles/click_...foo like normal [13:37] mardy: but, if you modify the policy group in /usr/share/apparmor/easyprof/ubuntu/policygroups/1.*/accounts, you'll need to regenerate the profile in /var/lib/apparmor/profiles/click_...foo by running 'sudo aa-clickhook -f' [13:39] mardy: I don't know if that was the case, but maybe that had something to do with it? [13:40] mardy: anyhoo, you see that apparmor will allow the dereference but not the read, which leaves us with leaking click installed providers [13:40] jdstrand: no, sorry, forget about it, I checked and the application is not reading that file at all [13:41] jdstrand: that's fine [13:41] mardy: you say that is not a problem-- why is that? I can see that argument for system installed providers, but this is a way to see if the device has facebook installed, for example [13:41] jdstrand: app can read the account DB anyway [13:41] oh right [13:41] ok, then when I update the policy, I'll add that note so I don't forget again [13:41] :) [13:42] mardy: do you mind if I change the description in the bug? [13:42] jdstrand: OK, thanks; do you know when you'll update it? [13:42] dednick, can you check from recovery if your /data partition has enough space ? [13:42] jdstrand: feel free to [13:43] mardy: I should be able to do it soon-- I have another update I need to push for alex-abreu [13:43] jdstrand: excellent, thanks! [13:44] mardy, yup === alex__ is now known as alex-abreu [13:44] alex-abreu: ah, you were under disguise :-) [13:44] yeah [13:44] not sure where the name cae from [13:44] alex-abreu: ah, my tab complete saw that, but I didn't know it was you :) [13:45] jdstrand, :) I hide from security [13:45] ogra_: /data has 12.7G free [13:45] alex-abreu: so, I finally got everything working, once bug 1278859 will be fixed [13:45] alex-abreu: I'm up to date as of yesterday, does that mean I can test the click on the desktop and it should work? [13:45] bug 1278859 in apparmor-easyprof-ubuntu (Ubuntu) "Click apps need access to their own Online Accounts files" [Undecided,New] https://launchpad.net/bugs/1278859 [13:45] dednick, hmm, then i dont get why it breaks ... 1000K shouldnt make a difference in the code ... just produce a bigger loop img [13:45] jdstrand, I have to check if anything was released ... [13:45] alex-abreu: except one big issue: apparmor blocks the call from webapp-container to signon-ui, to get the cookies [13:46] jdstrand, its been on the release sheet for 5days now ... [13:46] jdstrand: there is also https://bugs.launchpad.net/ubuntu-html5-theme/+bug/1278818 which i filed this morning [13:46] Launchpad bug 1278818 in apparmor-easyprof-ubuntu (Ubuntu) "HTML5 container needs apparmor authorization" [Undecided,New] [13:46] ogra_: i'll see if i can run the commands manually and see if i get any more meaningful errors. [13:46] alex-abreu: i couldn't find the reference to one for the container in the hmtl5-blocker's list [13:47] mardy, why does it? ... didn't you have any policy update for that [13:47] dednick, just remove all the redirection to /dev/null at the end of the dd and mke2fs [13:47] alex-abreu: no, because I thought the call would come from an unconfined process [13:47] dbarth, didn't we put unity-webapps-qml in it along w/ ubuntu-html5-theme ? [13:47] alex-abreu: and I definitely don't want to create an apparmor rule for that, or everyone will be able to read any cookies [13:48] jdstrand, dbarth, what is failing is the webapps-qml package not being up to date, while the html5 one has been release prematurely [13:48] mardy, ah! and the webapps container is confined ... [13:50] ogra_: http://pastebin.ubuntu.com/6915091/ [13:50] mardy, well ... let me check ... 1. you validate that the app is what it prentends to be, 2. the app has to have a .application file installed w/ the proper provider [13:50] mardy, alex-abreu: then what if we did a merge proposal for jdstrand instead? [13:50] alex-abreu: of course, signon-ui could check who the caller is and see if it's in the ACL for the account, but that's growing into something big [13:50] dednick, woah [13:50] cause i'd rather test the changes in a ppa before we presse the big landing button [13:50] dbarth, alex-abreu: aren't bug #1274640 and bug #1278818 the same bug? [13:50] bug 1274640 in apparmor-easyprof-ubuntu (Ubuntu) "[Html5 apps] Add the new HTML5 launcher /usr/bin/ubuntu-html5-app-launcher to the ubuntu-sdk profile" [High,Confirmed] https://launchpad.net/bugs/1274640 [13:51] bug 1278818 in apparmor-easyprof-ubuntu (Ubuntu) "HTML5 container needs apparmor authorization" [Undecided,New] https://launchpad.net/bugs/1278818 [13:51] jdstrand: ah correct, i'll mark as a duplicate [13:51] dednick, can you pastebin the while script ? [13:51] dbarth, jdstrand mmh yes, they seem to be [13:51] *whole [13:51] dbarth: no way :-) That would mean that any application can read the cookies of any account [13:51] dednick, i.e. ... pastebinit rootsotck-touch-install [13:51] ogra_: http://pastebin.ubuntu.com/6915099/ [13:52] hm. need 64bit dd? [13:52] dbarth: right, so alex-abreu filed 1274640 last week and I am aware of it. I'm working with alex-abreu on how to test it [13:53] actually, that was more like a week and a half ago [13:53] anyhoo, yes, on it [13:53] dednick, hmm, shouldnt really matter ... i dont really get it [13:53] jdstrand: ok [13:53] jdstrand: can you ping us a branch we can make test packages from? [13:53] jdstrand: once you're done with the changes [13:54] mardy, dbarth not sure if there is way to mitigate it ... dont you in addition to validating that the app is who it is supposed to be, also validate that the is listed as an application w/ a "webapp" .application file [13:54] mardy, so its not *any* app really [13:54] mardy, you could also check that the app has a webapp apparmor profile [13:54] dednick, testing the dd line on maguro just works here ... [13:54] mardy, those 3 checks might be "enough" and acceptable [13:55] alex-abreu, mardy: we can finish that during that call, i'll just open the line [13:55] alex-abreu: well, any app can install an .application file [13:55] dbarth, what jdstrand was missing is the updated unity-webapps-qml (which hasn't been released yet :/) ... [13:55] ogra@styx:~/Devel/branches/rootstock-ng$ adb shell "dd if=/dev/zero of=/userdata/test.img seek=1000K bs=4096 count=0" [13:55] jdstrand, there is one updated version on our staging ppa though that you can work from [13:55] ogra@styx:~/Devel/branches/rootstock-ng$ adb shell "ls -lh /userdata/test.img" [13:55] -rw-rw-rw- 1 root root 4.0G Feb 11 14:55 /userdata/test.img [13:55] dednick, ^^^ [13:56] hm [13:56] * ogra_ tries from recovery mode ... probably the dd there is different [13:56] mardy, sure, but the app need to have "ubuntu-webapp" app armor profile ... whicch I assume you can check & validate & brings us closer to what we want no? [13:56] alex-abreu: cool, can you point me at that url? [13:57] ogra@styx:~/Devel/branches/rootstock-ng$ adb shell "dd if=/dev/zero of=/data/test.img seek=1000K bs=4096 count=0"dd: /data/test.img: Invalid argument [13:57] aha [13:57] dednick, looks like the dd in recovery has some limitations [13:57] :( [13:57] ogra_: ah. how did you manage it then? [13:57] the first one was in a normally booted setup [13:57] alex-abreu: but any app can declare it, can't it? [13:58] the second one in recovery [13:58] so it cant process more than 2G with the android dd [13:58] alex-abreu: nm, I found them [13:58] mardy, well ... sure but the app will be confined according to this profile ... [13:58] ah, i c [13:58] not the sdk regular profile [13:58] dednick, might be a bionic vs libc thing [13:59] the recovery dd is from androids busybox ... linked against bionic [13:59] stgraber, did you knwo that ? [13:59] *know === greyback is now known as greyback|lunch [14:02] mardy: ok, the only change I am making is I am prepending 'owner ' to your rule [14:02] jdstrand: what does it do? [14:03] mardy: makes sure the uid of the process matches the owner of the file [14:04] jdstrand: sounds good [14:04] that's important on multiuser devices [14:04] (and what we use for accounts.db*, etc [14:04] ) [14:05] dednick, so the only thing that comes to mind is that you can probably remove the dd and mkfs.ext2 from the code and replace it with an adb push for a system.img file you create on the PC ... [14:05] so you puch it and mount it ... but create it in advance on a machine with a sane dd === JamesTait is now known as Guest59020 [14:06] ogra_: heh, ok. i might give that a go later. === Guest59020 is now known as JamesTait [14:13] dholbach, yes, unity-webapps-qml got reverted because it pulled click into the desktop, which is verboten. That should be fixed in trunk but I've held off on releasing it because I got the impression that there was more that needed to be fixed as well, it was unclear to me that it was ready. [14:14] robru, ahhh ok [14:15] dholbach, so please confer with dbarth / alex-abreu to make sure that trunk is in a releasable state and then I can release it. [14:15] robru, I looked at the diff earlier and was a bit confused as to what exactly might have broken things - but that was probably including your good fixes already :) [14:16] robru, I'll take another look and then we can propose it and get an ACK from stgraber and move forward [14:16] dholbach, ok, great === om26er_ is now known as om26er [14:22] alex-abreu: ok, I confirmed it worked. thanks! [14:23] jdstrand, oh my! thanks a lot! [14:23] really [14:23] np, sorry it took a while [14:24] I'll test on a device too for good measure [14:27] ogra_: WOO! [14:27] works ? [14:28] yeah [14:28] adb push takes about 20 minutes with a 4GB file though :) [14:28] lol, yeah [14:35] hey mterry, how are you doing? regarding bug 1157732 - do you know what still needs to be done to get qtdeclarative5-accounts-plugin into main? [14:35] bug 1157732 in qml-friends (Ubuntu) "[MIR] circle of friends" [Undecided,Fix committed] https://launchpad.net/bugs/1157732 [14:38] dholbach, it is approved to enter main, so something just needs to pull it in. Either a Depends, Build-Depends, or add it to a seed directly [14:38] mterry, great! [14:38] robru, ^ :) [14:40] dholbach, oh, ok. i guess we should seed it then? [14:40] robru, AFAICS unity-webapps-qml is going to pull it in directly [14:42] dholbach, oh ok [14:43] dholbach, so did you do a manual upload of webapps-qml then? [14:43] rsalveti: hi, how is it going? any news regarding the hammerhead image? :-D [14:44] robru, not yet - I'm writing a changelog entry summing up the changes since last time right now - if you want, you can take a second look and upload it... just want to make sure the version number trumps what's in the PPAs [14:44] dholbach, no, I won't do an upload until I can get everything under citrain (just waiting for this hangout to end so I can concentrate on doing the transition) === dandrader is now known as dandrader|afk [14:46] robru, sorry, then I must have misunderstood you - I thought we'd do one upload beforehand and then go on citrain - sure... works for me to use citrain directly [14:46] dholbach, I like citrain a lot so i'd prefer to just go straight there. i think I can do it today [14:46] sure sure [14:46] robru, and ubuntu-html5-theme too? (I think there was nothing blocking an upload right now) [14:48] dholbach, yeah, they should be released together [14:48] robru, fantastic - thanks so much! === greyback|lunch is now known as greyback [14:53] barry, I'm going to try and land asap a new udm with better error handling etc.. I've spoken with sil2100 to try and do this and then get udm in the CI train [14:53] barry, maybe that way we get more info regarding the image updates [14:55] mandel: sounds great. i am going to work up a branch of s-i with better debugging, and while i won't try to land that just yet, at least i'll have a ppa folks can install from and try to gather more information [14:57] barry, you can use https://launchpad.net/~mandel/+archive/mandel because it has the latests udm in trunk (but nor for armhrf :-/ ) [14:58] mandel, ask for armhf builds on #webops ;-) [14:58]