[00:02] sergiusens: how do your lp:phablet-tools branch approvals turn into actual merges? [00:08] slangasek: with the silo request I made [00:08] slangasek: that's #ubuntu-ci-eng [00:08] sergiusens: ok - thanks [00:09] * slangasek wonders how one is supposed to know which packages are associated with silos, looking from the archive out [00:22] slangasek: one req didrocks set upon us is that lp:[project] == [project]_.*changes ... extending; any package that is build out of branches (bzr bd) can have silos; well I asked for udd but I thinks that's out of scope [00:23] that said; there is no tool; just the format of the changelog versioning can be used as an indication (and signing) [00:24] right, but that's a fuzzy indicator [00:24] would love something in debian/control that screams "hey, silo" if I accidentally try to skip the process :) [01:13] About link to download ubuntu nexus 7 version | http://askubuntu.com/q/441738 [02:03] why can't I upgrade to 269 image? It looks like pass rate is 100%. === vying is now known as Guest57074 [03:15] If anyone knows... does the Ubuntu Touch keyboard currently support multitouch? It didn't when I tried it a few months back. === jono is now known as Guest60170 [05:08] can anyone help me, i need to install ubuntu touch on my tablet, my tablet is skyworth s7.thanks [06:55] good morning [07:35] hi [07:36] hello? [07:36] hello [08:32] Hee people [08:32] Does anybody know if there is an nexus 5 version of ubuntu touch? not a port [08:32] Good morning all; happy Batman Day! :-D === chihchun_afk is now known as chihchun [08:38] does anybody know if there is an official nexus 5 release? === chihchun is now known as chihchun_afk === chriadam is now known as chriadam|away [10:14] Morning all === MacSlow is now known as MacSlow|lunch [11:07] how can i get apt-get to work on the phablet image? [11:07] i get E: Unable to write to /var/cache/apt/, even as root [11:08] brendand: the image is read-only by default. [11:09] phablet-config writable-image [11:09] that command will make it rw [11:49] is it normal that I can't reach upstart as root on the device? [11:49] root@ubuntu-phablet:~# initctl status powerd [11:49] initctl: Unable to connect to Upstart: Empty address '' [11:51] Saviq, your env is not correctly set I guess? [11:51] seb128, all three of my devices... [11:51] seb128, under image 27`1 [11:51] -` === alan_g is now known as alan_g|lunch [11:53] Saviq, try asking xnox since you have it on the other side of the keyboard in some discussion ;-) [11:55] Saviq, hmm, xnox changed adbd [11:56] ogra_, I tried sudo-ing from phablet, too, didn't help [11:57] well, it processes /etc/profile now when you do adb shell [11:57] due to that change [11:57] which sets the locale etc ... but might also set other variables you dont want set as root [12:02] ogra_, well, but apparently it doesn't set UPSTART_SESSION any more? === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === psivaa is now known as psivaa-lunch === alan_g|lunch is now known as alan_g === _salem is now known as salem_ === psivaa-lunch is now known as psivaa [13:29] Elleo, i just noticed that the qml docs generated in content-hub now have a importStatement at the top that doesn't have a version [13:29] Elleo, i guess that is new with 5.2 [13:30] maybe we need to set the API version somewhere for qdoc [13:30] Import Statement: import Ubuntu.Content . [13:30] kenvandine: ah, okay [13:30] kenvandine: I'll look in to that now and add it to the qdoc fix branch [13:31] thx [13:31] i don't recall even seeing that importStatement before, must be new [13:42] doanac, hello! So you mention system-settle in the "unlock-script" MP. Is that meant to wait until system is ready? [13:43] it checks if the system goes idle [13:47] ogra_, ok, thanks [13:47] doanac, so why do your jenkins scripts need to run it before unlocking? (you fear system being too loaded to unlock?) We have the sleep and retries for that. I'm just trying to avoid making this unlock script anything more than a black box, because the split version of this script won't work the same way. I don't want to build in arguments/features based on how the current version works [13:49] mterry, hey, so what format do we supply that spinner boot animation thingy [13:50] cwayne_, what format? Currently it's an executable that you specify to USC with --spinner=/path [13:50] so it's not like a video file or something? [13:55] cwayne: why don't you look at the code find out where the file lives and see what format it is ;) [13:55] cwayne: how's life anyway? === VDVsx_ is now known as VDVsx [14:00] davmor2, i would, but i didn't think it's actually in an image yet [14:01] cwayne: Man I've heard some excuses before ;) [14:02] cwayne: I think it was MacSlow working on the animation but I could be wrong [14:02] cwayne_: and why are there 2 of you [14:02] bfiller: hi! So, did your team maybe have an additional look at the telepathy-ofono crash during messaging-app tests? [14:03] davmor2, not sure [14:04] mhall119, pingaling [14:06] sil2100: I will ask, in meeting now [14:07] bfiller: thanks! Since the last smoketesting runs should have better crash-files, maybe something more can be squeezed out of those [14:12] sil2100: do you have a link to this crash? I don't remember seeing any telepathy-ofono crash on tests [14:12] sil2100: on the other hand, the telephony-service-indicator ones are alraedy fixed [14:13] boiko: ok, one moment [14:13] boiko: http://ci.ubuntu.com/smokeng/trusty/touch/mako/270:20140331.1:20140331/7488/messaging_app/ <- [14:13] sil2100: thanks, let me take this one a look [14:13] boiko: thanks :) [14:17] cwayne_: pongalong [14:18] mhall119, hey, was wondering if youd have a moment today to talk about the importer scripts to get the customization docs up on the api site? [14:19] cwayne_: I should, yes [14:22] cwayne_, davmor2: the logo of the current boot-animation is a PNG-file animated via OpenGL in with a dedicated binary that's passed to u-s-c using the --spinner command-line option [14:23] cwayne: see I knew MacSlow was something to do with it :) Thanks MacSlow [14:24] davmor2, just curious or do you want to mess/play with it? [14:24] davmor2, fyi... it's not finished :) [14:24] MacSlow: I don't want it at all it was cwayne [14:25] MacSlow: I'm assuming for customisation for oem but could be wrong [14:25] davmor2, it's not delaying the boot-process in case you wonder === alan_g is now known as alan_g|tea [14:29] yeah, it was about how to customize it [14:34] arg, the scopes API is not great for subclassing is it? [14:35] kyleN, dpm, I have so far 2 categories that use an XML feed, so a lot of duplicate code [14:35] I was planning to subclass category, like XMLFeedCategory, and then specialize it for my 2 feeds [14:36] but I see that you don't instantiate a category directly, you get one like this: [14:36] auto planet_cat = reply->register_category("planet", "Planet Ubuntu", "", rdr); [14:36] thoughts? [14:38] just a moment.. [14:41] mhall119, would a hangout at 11:30 work for you? (i think we're in the same time zone...) === alan_g|tea is now known as alan_g [14:44] rickspencer3, so yes, the API docs say: To create a Category, use ReplyProxy::register_category. So the API does not appear to be designed right now to directly subclass Category [14:44] kyleN, any suggestions for good OO design here to avoid code duplication? [14:45] I mean, it seems like many queries are doing to construct a url, fetch a file, and the parse it [14:45] so it seems like we really need a class with a place to specify the url and then get the file back [14:45] I guess I could just make a private method and add it to the class [14:45] the query class, I mean [14:45] or a couple of private helpers [14:45] thoughts? [14:47] rickspencer3, full disclosure ;) I have only been thinking about scopes for one week and only since then have I revisted c++ (a 10+ year absence on my part). that being said: [14:47] rickspencer3, in general I think we should provide a much simpler API for run-of-the-mill web api handling [14:47] kyleN, ok [14:48] kyleN, let me try adding helper functions for fetching and parsing the XML files [14:48] cool [14:48] I'd prefer subclassing for my own code [14:48] but, I guess the API isn't designed that way [14:51] mterry: just responded to your MP comment. Basically "system-settle" is treated like a real test case in daily image testing. and my worry is that if it takes a while to settle, the screen may wind up locked before we run the autopilot test [14:55] doanac, isn't that why jenkins puts a powerd freeze on timing out? [14:55] mterry: in the past that didn't prevent the screen from locking [14:56] doanac, oh curious. I think it would today. But anyway. I added a -w argument to unlock-device [14:56] doanac, defaults to sleep 20, but you can put any command(s) in there [14:56] mterry: yeah, lots of things have been fixed, so you are probably right [14:57] but the -w is probably best in case we need workarounds and such in the future === dandrader is now known as dandrader|afk [15:03] Saviq, can you retry USC in silo 002? It crashes with a "symbol lookup error" for Mir. Probably got built too early [15:03] Saviq, I'm guessing you had to synchronize the builds in the train and we have to resync in PPA? [15:04] mterry, it's just that the whole thing got rebuilt from scratch [15:04] Saviq, right, just saying any sync jiggering we had to do we have to re-do [15:05] mterry, hmm? https://launchpad.net/~ci-train-ppa-service/+archive/landing-002/+sourcepub/4058089/+listing-archive-extra [15:06] Saviq, I'm getting a runtime crash [15:06] mterry, ah, got it [15:07] Saviq, but I'm not sure how... USC shouldn't have built with old Mir... And if it did with new Mir why would I see the error [15:07] mterry, it built with "old" mir as in yesterday's [15:07] or whenever it was last built [15:07] mterry, and devel changed between then and an hour ago [15:07] k [15:13] cwayne_: can we go an hour later? I need to pickup my car from the shop [15:14] mhall119, hm, i have an appt at 1 [15:16] is there a way to actually upgrade lxc-android-config on the device? [15:17] umounting the udev rules file worked before, not any more apparently, getting device or resource busy [15:30] cwayne_: 2pm? [15:30] tedg, how do i make it dep on dbus? [15:31] that's the click-user-hook that needs to dep right? [15:31] kenvandine, Yeah, it should probably be start on started dbus [15:31] mhall119, not sure how long my appt will last, shall i just ping you when i'm back from it? [15:31] kenvandine, Though guessing cjwatson had a reason to put it before xsession. [15:31] mhall119, and can i grab that bzr branch of importer scripts from you first? maybe i can make some progress first [15:31] kenvandine, Careful to avoid unsolvable dependencies :-) [15:32] cwayne_: just ping me when you're back [15:32] cjwatson, i'm seeing some problems with the content-hub hook failing to change settings in dconf, tedg is suggesting that it's probably click-user-hooks not depending on dbus [15:32] cjwatson, thoughts? [15:33] I didn't think about it that closely, but I wanted to make sure that it ran before any desktop files were read [15:34] understandable === gatox is now known as gatox_lunch [15:34] content-hub hook modifies gsettings === dandrader|afk is now known as dandrader [15:38] cwayne_: https://launchpad.net/ubuntu-api-website/importers has the import scripts [15:39] kenvandine: I wonder if "start on starting xsession-init and started dbus" would do it? [15:39] is suspect it would [15:39] i can test it locally [15:39] mhall119, thanks [15:39] i just worry that might cause other problems [15:40] cjwatson, not sure what else to look for there... besides my hook not failing :) [15:43] kenvandine: checking for log output from the other hooks would be a good plan [15:44] cjwatson, ok, that fixed it [15:45] cjwatson, nothing unusual in click-user-hooks.log, some glib criticals [15:45] but those were there before too [15:46] kenvandine: ok, is there already a bug for this? (don't worry if not) [15:46] nope [15:49] kenvandine: ok. committed that change for the next upload [15:49] cjwatson, thanks! [15:50] that might well explain that racy-looking hang in the emulator .. [15:50] boiko: any luck with that crash file? === howefield is now known as howefield_afk [16:05] cwayne_: if you have time to chat now, I'm back [16:07] mhall119, sorry, jsut about to get lunch and then make way to my appt, mind if i ping you when I return? [16:07] cwayne_: sure,whenever you're back [16:21] * t1mp flashing 272. Looks good on the dashboard so far === GreySynt- is now known as GreySyntax [16:32] sil2100: salem_ and I are investigating that still === gatox_lunch is now known as gatox [16:42] kgunn, greyback, tvoss, one thing I was thinking about the decoupling of ui and render threads, we lose throttling on framedrops - _not sure_ it's a problem, but we need to keep that in mind [16:43] Saviq, I think we would still block eglSwapBuffers if the screen is on [16:43] Saviq, or am I missing something? [16:43] tvoss, yes, but the UI thread will go "oh, rendering busy, let's go without it" [16:44] Saviq, I don't think so, as I understand it, we leave the sync logic in place for eglSwapBuffers blocking the render thread for longer than one iteration of the UI thread [16:45] Saviq, we just return early with false/EGL_CONTEXT_LOST if the screen is off [16:48] tvoss, huh? [16:48] tvoss, I missed something, since when did we say returning straight away is fine? [16:48] tvoss, and I'm sure that's not what greyback is implementing [16:50] What I was implementing was if eglSwapBuffers returned false, with the EGL_CONTEXT_LOST, then Qt would stop its render loop properly, leaving GUI thread to continue [16:53] oO [16:53] greyback, tvoss, ok, means I missed some discussion, when I left it last it was "let's decouple ui thread from render thread so it ignores blocked render thread" [16:54] but I'm actually happier with what you're doing, although I was under the impression that's not something we wanted to do.. [16:54] Saviq, mind joining here real quick: https://plus.google.com/hangouts/_/72cpiupvbrqbqgrd44ubq129gs?hl=en [16:54] Saviq: I see. Yeah, we moved away from that approach for a bit. I had code to do it, but it has regressions and I suspect for that very reason === bfiller is now known as bfiller_afk [17:03] fginther: ping === alan_g is now known as alan_g|EOD [17:05] In ubuntu 12.04LTS Two finger vertical scrolling works for sometimes and then doesn't work for sometime | http://askubuntu.com/q/442026 === vying is now known as Guest62054 [17:16] kgunn, some of the packages in silo 002 need to be rebuilt in the right order. Can you rekick platform-api? (it also needs a new version number to be higher than distro) [17:16] mterry: ok.... [17:17] kgunn, (they all got put in the PPA at the same time and some built with older Mir versions) [17:18] mterry: ok...i think we can just delete the platform-api out of the ppa right ? [17:18] (solves the prob) [17:18] kgunn, but we'd need to put one back in that builds against the mir in the PPA [17:19] kgunn, oh, Saviq apparently was already working on this [17:20] Saviq, I didn't realize you had to start with the train again (instead of just restarting some builds in the PPA) [17:20] mterry, new version number will happen automagically [17:20] mterry, I don't have the perms to restart in ppa [17:20] Saviq, because you redid it from train so we get a new source. OK [17:20] Saviq, ah [17:20] mterry, we could've restarted usc indeed, but whatever [17:21] (asked landing team to, that is) [17:21] Saviq, yeah, but we needed platform-api to get a version bump too [17:28] mterry: Saviq so there's already a build going? [17:28] https://ci-train.ubuntu.com/job/landing-002-1-build/10/console [17:28] kgunn, yes [17:28] ok...i'll assume you got it [17:28] kgunn, thanks anyway! :) [17:29] mterry, it just uploaded, should be building in a few minutes [17:29] now [17:29] Saviq, so which things got updated? Are we going to have to do them one by one (papi first, then unity-mir + usc)? [17:30] mterry, papi now, not sure we'll need um, usc, will we? [17:30] mterry, papi and usc now, I mean [17:31] Saviq, possibly not [17:31] mterry, um probably, I can upload it now already I think, since papi ABI won't change [17:31] mterry, let's see [17:31] I'm a little confused about what built with what in that PPA at this point [17:32] mterry, and anyway might not be a problem depending on which symbols changed [17:33] Hello [17:34] Which tested Ubuntu-touch? === dandrader is now known as dandrader|lunch [17:37] Hello dandrader|lunch [17:40] Hello [17:40] Which tested Ubuntu-touch? [17:46] mterry, built, check it out? [17:46] mterry, let me know if um needs a kick [17:46] Saviq, my device's wifi stopped working. :( [17:47] nmcli d wifi list is empty [17:47] hello [17:47] Which tested Ubuntu-touch? [17:47] Maybe a reflash is needed.... [17:47] creaprog, that's wrong English [17:48] creaprog, but start from https://wiki.ubuntu.com/Touch [17:48] Who use ubuntu-touch ? [17:48] ubuntu-touch is problem driver for Samsung Galaxy Note 2 ? [17:48] most of us [17:49] ok [17:49] I am the novice for smartphone [17:50] ubuntu-touch is bêta or alpha ? [17:50] creaprog: see Saviq 's comment the builds we make are for the Nexus 4, 7 and 10 [17:50] ok :( [17:50] Why ? [17:51] smartphones with ubuntu are comming on the market creaprog [17:51] creaprog: https://wiki.ubuntu.com/Touch should tell you but basically because we can't create it for every device [17:52] ok [17:52] Ubuntu-touch use kernel Linux ? Ubuntu-touch is new distribution for smartphone ? [17:55] davmor2: im gonna buy a nexus7 2013 soon and try touch on it :p [18:03] who is working on gallery-app? [18:04] can ubuntu touch disable an nfc chip? [18:04] Saviq, I'm getting crashes in unity8 with "Library unity-mir not found/loaded" -- guessing a symbol failure on load of unity-mir [18:04] Saviq, can we kick um too then? [18:04] mterry, k, building [18:04] thanks [18:05] lotuspsychje: no but it currently isn't supported, i'm assuming you can turn it off the same way you can on android though [18:06] davmor2: well i trust ubuntu more as android in future anyways :p [18:06] many tnx to all ubuntu-touch devs for all the good work! [18:07] bringing security to mobile lifes will be amazing [18:26] mterry, built [18:27] Saviq, hrm. greeter still crashing, but not clear why yet [18:27] mterry, might be unity8 needs rebuilding, too? [18:27] mterry, does it link to mir directly? [18:27] Saviq, it shouldn't after mir update should it? [18:28] not directly, it dlopens unity-mir [18:28] mterry, ok then no [18:28] mterry, you sure you got the newest unity-mir? [18:28] 0.3+14.04.20140401.1-0ubuntu1 [18:28] yup [18:28] mterry, ok, let me know if you find something / need me [18:29] Saviq, k thanks [18:31] Ubuntu Touch keyboard problem | http://askubuntu.com/q/442059 [18:34] Saviq, "/usr/lib/arm-linux-gnueabihf/mir/platformgraphics/android/libmirplatformgraphics.so: undefined symbol: _ZTIN5boost15program_options29value_semantic_codecvt_helperIcEE" [18:34] Saviq, I thought we rebuilt papi? [18:35] mterry, yeah, we just did [18:35] why would it be a boost symbol anyway... [18:43] mterry, what do you think we should try and rebuild? [18:44] Saviq, I guess? I'm trying to rebuild locally and see if that helps. But might as well kick it in PPA too [18:44] mterry, but what? papi? [18:44] that'd be really weird, /me looks at the build log [18:44] Saviq, oh sorry. I was thinking papi yah. Even though it doesn't make sense [18:45] mterry, https://launchpadlibrarian.net/171482447/buildlog_ubuntu-trusty-armhf.platform-api_0.20%2B14.04.20140401.1-0ubuntu1_UPLOADING.txt.gz has everything from the ppa :/ [18:46] mterry, and program options is a distro-sourced thing [18:46] Saviq, OK, good to know. It's a missing boost symbol anyway though, not even something from the PPA [18:46] mterry, yeah exactly... [18:46] Saviq, did boost just get updated in distro... /me checks [18:47] mterry, last one on New Year's Eve https://launchpad.net/ubuntu/+source/boost1.54 [18:47] no idea, /me kicks it anyway [18:48] * mterry scratches head [18:49] tedg, sorry it's taking me a while to confirm your sound fixes, I'm having a hard time getting split up and running [18:49] tedg, I can test manually poking AS though... let me try that [18:53] mterry, papi uploaded === bfiller_afk is now known as bfiller [19:11] tedg, I have a minor patch to fix indicator-sound racing with unity-greeter [19:11] tedg, on startup, it may not get active user from the greeter [19:12] mterry, Yeah, patched that I thought, no? [19:12] You did? [19:13] mterry, https://code.launchpad.net/~ted/indicator-sound/null-selected-user/+merge/212656 [19:15] mterry, papi built, btw [19:15] tedg, no this is if indicator-sound starts before the greeter exposes its DBus interface I think [19:15] tedg, needs to watch name [19:16] Saviq, ok, will look in a bit [19:16] mterry, Oh, okay. [19:16] mterry, If the greeter is starting the indicators... how does that happen? [19:17] tedg, I *think* that's what happening. I'm not 100% yet [19:18] * mterry adds more debug comments [19:20] Or it could be because we pass null to a method that doesn't allow it... tsk tsk Mike [19:27] boiko, unset UPSTART_SESSION as root to make upstart work again, it's fixed in silo 015 already, should land soon [19:28] fginther, ↑ [19:28] Saviq: thanks [19:28] fginther, re: bug #1300880 potentially [19:28] bug 1300880 in dialer-app "CI can't install dialer-app debian packages on image 272" [Undecided,New] https://launchpad.net/bugs/1300880 [19:28] Saviq, thanks [19:33] Whatsapp on Ubuntu Mobile | http://askubuntu.com/q/442091 [19:35] jhodapp: hi, media-hub depends on : libhybris-dev (>= 0.1.0+git20131207+e452e83-0ubuntu13), but *12 is the latest in trusty/main so CI is broken. Does it really need 13? [19:35] robotfuel, it does yes [19:36] robotfuel, how is CI broken though if media-hub isn't even in the archive? [19:38] jhodapp: http://s-jenkins.ubuntu-ci:8080/job/media-hub-trusty-armhf-ci/37/console the pbuilder is failing on installing libhybris-dev [19:39] robotfuel, well version 13 of libhyris will land along with media-hub [19:40] robotfuel, they're in landing silo 006 together [19:41] jhodapp: ack, I was going to collect coverage data in jenkins, I'll wait until it lands to finish. [19:41] robotfuel, alright, trying to land it asap which will be in the next few days [19:42] mhall119, ping [19:47] cwayne_: pong [19:51] mhall119, heya, got a few minutes to talk about importer stuff? [19:52] I will in a moment === dandrader|lunch is now known as dandrader === ajalkane_ is now known as ajalkane [20:05] cwayne_: ok, I'm available now [20:06] mhall119, cool, hangout work for you? [20:06] yup [20:08] cwayne_: https://plus.google.com/hangouts/_/7acpjlo8tqb25nc4d6rd842kug?hl=en [20:10] mhall119, http://people.canonical.com/~achiang/ubuntu_savvy/ [20:18] cwayne_: sorry, hit the wrong button [20:18] I wasn't trying to be rude, hoenst :) [20:18] cwayne_: is the stuff on http://people.canonical.com/~achiang/ubuntu_savvy/ all up to date? [20:20] mhall119, it should be, yep [20:22] ok, thanks [20:22] mhall119, and no worries, i figured you hit the wrong button, I do it all the time :D === jhodapp is now known as jhodapp|afk [21:25] Saviq, well as predicted, papi rebuild didn't help [21:25] mterry, want me to kick a full rebuild then? did mir change in the mean time? [21:26] Saviq, not that I know of [21:26] Saviq, but I'm skeptical because this isn't our normal problem with Mir symbols but rather boost symbols [21:26] mterry, indeed [21:27] Saviq, but maybe it's a boost symbol created by mir headers [21:33] Saviq, yeah, it's a templated class... so I get that mir might define a special version of it, but not sure how papi could have screwed that up given the rebuilds [21:34] mterry, so what do we do, blanket rebuild? [21:34] Saviq, can't hurt, might help [21:34] mterry, ok, but looks like we need to do it in order [21:34] mterry, there's new stuff in mir [21:34] Saviq, ah. Well, let's just do mir then to start [21:35] mterry, maybe even not breaking abi, but let's play it safe [21:35] * mterry will be happy when Mir lands and we can take it out of this silo [21:35] :) [21:36] mterry, there's plenty of happiness to be had, I'm putting the right edge through its paces [21:37] mterry, ugh https://ci-train.ubuntu.com/job/landing-002-1-build/14/console [21:38] mterry, conflicts :| [21:38] Saviq, always [21:40] mterry, every time that happens I feel like we should all just push for ~*-team [21:40] mterry, can you merge, push somewhere and resubmit? [21:41] Saviq, yeah give me a minute [21:41] mterry, sure === salem_ is now known as _salem === popey_ is now known as popey