[00:10] Guest42302: the 2013 N7 isn't supported yet, no [00:11] I'm not sure if anybody at XDA has built an image for it yet [00:15] ok, thanks [00:33] Hello is there a way to slide in from the side to reveal the launcher with regular ubuntu? does ubuntu touch have an iso i can make a live image with. Using an dell inspiron 15 7000 [00:39] slide-guy: you can auto-hide the launcher on the desktop,then pressing your mouse pointer against the edge will reveal it [00:40] but I want to use my touch screen :( [00:40] slide-guy: Unity 8 (the Ubuntu touch shell) will have an testable session in 14.04's desktop ISO, but it's not there yet [00:40] cool :^) [00:40] you can run Unity 8 in a windows right now, on your desktop, but not as the actual shell [00:41] is there a guide? [00:41] http://unity.ubuntu.com/getinvolved/development/unity8/ has a guide to getting the code,compiling it and running it [00:41] cool thanks [00:41] but if you just want to run it, I think there are packages to install in 13.10 [00:41] true [00:42] apt-get install unity8, I think that's all you need [00:42] might need unity8-private and/or unity8-fake-env [00:42] can I do that in a live environment? I have a 16gb flash drive [00:42] join #ubuntu-unity and ask in there for additional help [00:42] k === hakr is now known as h[a]kr [00:45] running it I'm pretty sure [00:45] don't know how compiling it would work out though === Namidairo`bnc is now known as Namidairo === _salem is now known as salem_ === chriadam|away is now known as chriadam === JamesTait is now known as Guest62879 [02:01] yop === shuduo_afk is now known as shuduo [02:50] hello, where is right place to report bug for system-settings-touch? there is no 'report a bug' in lp:~system-settings-touch page === ricardodrosales is now known as ricdros === salem_ is now known as _salem === shuduo is now known as shuduo_afk [04:06] Will I see an ubuntu phone or tablet by April 2014? === shuduo_afk is now known as shuduo === shuduo is now known as shuduo_afk === shuduo_afk is now known as shuduo === chihchun_afk is now known as chihchun === iahmad is now known as iahmad|brb [07:59] good morning === iahmad|brb is now known as iahmad === iahmad is now known as iahmad|afk [08:52] didrocks, salut mon ami - do you know when the new ubuntu-system-image is going to land? [08:53] dholbach: it should have landed, right? (2.0) [08:53] I saw it was stalled in proposed for a long time [08:54] didrocks, I couldn't find it in http://people.canonical.com/~j-lallement/touch/changes/ [08:54] dholbach: still in proposed indeed: system-image-dbus | 2.0.2-0ubuntu1 | trusty-proposed/universe | all [08:55] dholbach: I think we need barry to look at his packages to enter the release pocket :p [08:55] ah ok :) [08:55] thanks [08:55] I'd like to help debug https://bugs.launchpad.net/ubuntu-system-image/+bug/1256229 - that's why I was asking [08:55] Ubuntu bug 1256229 in Ubuntu system image ""system-image-cli -i" reports old version" [Undecided,Incomplete] === iahmad|afk is now known as iahmad [08:56] dholbach: yeah, he reuploaded 2.0.2 apparently on Tuesday to fix a FTBFS (and that's why sytem-image was stalled for a week or two in proposed) [08:56] it built yesterday, so can it be pulled in for the release now? [08:56] autopkgtest for system-image 2.0.2-0ubuntu1: FAIL (Jenkins: public, private) [08:56] oh ok [08:56] http://people.canonical.com/~ubuntu-archive/proposed-migration/update_excuses.html [08:57] dholbach: we really need to sensibilize people with upload rights to look at -proposed [08:57] stuff are stalled for days and days sometimes [08:58] didrocks, I don't know ... a short article on packaging.u.c? or is it on a wiki page already? [08:59] dholbach: there is this excellent page: https://wiki.ubuntu.com/ProposedMigration [09:01] didrocks, are mails sent out if something gets stuck in proposed? [09:01] dholbach: I don't think so [09:01] maybe it should be treated like a case of FTBFS [09:02] dholbach: for autopkgtests failures, maybe yeah. For the other cases, like "no ready yet on ", maybe that should be after a threshold time [09:02] but I'm not a specialist ;) [09:03] ok, I hope it's easy to fix, the test suite passed during the build [09:04] I'll file a bug about the packaging guide anyway [09:04] it'll be good to mention it [09:04] and maybe we could have a session at the next UDW [09:04] just so new developers know already know about it [09:12] yeah ;) === shuduo is now known as shuduo_afk === shuduo_afk is now known as shuduo === chihchun is now known as chihchun_afk === shuduo is now known as shuduo_afk === chihchun_afk is now known as chihchun === _salem is now known as salem_ [10:25] Bonjour ? === sil2100_ is now known as sil2100 [10:26] hi guy's, I'd like to up a problem on my mako. If you have a pin code with more than 4 digits it does'nt work. [10:30] Oria, we only support 4 [10:30] Oria, oh [10:30] Oria, you mean with SIM? [10:30] yes mterry with sim [10:30] * mterry doesn't know much about how we handle SIM [10:31] Oria, sorry, I'm stuck in a world where PIN means login password, not SIM [10:31] ok :) [10:32] Someone know if I can try it on my computer ? [10:33] mterry: why to limite [10:33] Oria, I don't know, I didn't do the SIM pin stuff === chihchun is now known as chihchun_afk [10:35] mterry: Is it possible to make a request to dev? Many people think it's a shame to limit to 4. Make a limit to 6 or more would be better? [10:36] mzanetti, did you do SIM pin stuff? ^ === chriadam is now known as chriadam|away [10:38] mterry: yes [10:38] Oria : maybe it's just a bug ; have you try with another SIM/phone ? [10:40] Oria: for SIM pin there is a fix coming in. for the phone lockscreen (not SIM PIN) it's a design decision to keep it to 4. [10:40] I'm not particularly happy with that [10:40] Luclis: I don't have another compatible phone or another SIM :) === chihchun_afk is now known as chihchun [10:46] ok mzanetti [10:46] To design, before security ... I find it illogical finally good in my opinion :) [10:47] Oria: well, if you want more you can use the passphrase one that gives you a full keyboard. but if you want to keep it to simple numbers, it's going to be limited to 4 [10:52] hi [10:53] I would like to have a LINUX phone, I was the home page of ubuntu touch, and it is insanely complicated to install ubuntu. [10:53] Haven't you a cdrom to install Linux directly from Windows 7? [10:53] (on the phone via usb cable) [10:54] a CD-ROM on a phone!? [10:54] Furthermore, the new phone arent compatible with ubuntu touch. [10:54] raptor67782: I'm afraid that doesn't work; unfortunately phones are quite a bit harder to install than desktops, but the phablet-flash tool makes this really easy and a single-command issue already? [10:54] no, you dont ujdersndatnd. I plug my phone on my windows ,and put in my cdrom ubuntutouch, and it installs it automatically. [10:54] I havent linux at all. [10:55] raptor67782: I think the honest answer is "nobody on our side is interested in doing it", and it would be loooots of work for very little gain [10:55] raptor67782: livecd [10:55] and phablet-flash [10:56] ah come on. too complicated [10:56] I regret but if users have to go through complicated processes, it will never work to get more interests. [10:56] It should be simple [10:56] it is simple [10:57] $ phablet-flash ubuntu-system [10:57] done [10:57] raptor67782, why do you want a linux on your phone? [10:57] You have windows xp or 7, you put your cdrom, and you install: SETUP NEXT... [10:57] dadel: linux is the future. Look Steam... [10:57] raptor67782: I'm afraid the people you are talking to here don't even have windows [10:58] * pitti hasn't had a windows box for 16 years [10:58] raptor67782, if you want the future now you may do some efforts [10:58] ha ha, really, you havent windows? [10:58] about 80 pct of machines are windows today. [10:58] you have a computer, you plug you liveCD, you run it, you launch the pitti command and it's done [10:58] I honestly tried to co-install it on my laptop, but it's way too big, and installing it is a royal pain, so I just gave up [10:58] world wide, about 5 - 10pct are linux [10:59] raptor67782: sure, I know the numbers; but that doesn't change the reason :) [10:59] it's not interesting to troll about that raptor67782 if you are happy with your win just stay with it [10:59] ^ fully agreed [11:00] i'm sure there are some #windows irc where you will find your happiness [11:00] thank you for your non-help :( [11:00] lol [11:01] well, we did tell you what the reason was and how to easily install an Ubuntu phone [11:01] *shrug* [11:01] Look https://wiki.ubuntu.com/Touch/Install I have asked two shops [11:02] selling phones in my town, I printed this page, and they have looked and studied, and even them, have no single idea how to install ubuntutouch. so ...I mean. it means everth [11:02] g [11:03] Morning all [11:03] i think that if you considere too complicated to run a liveCD to do so, you may wait that UbuntuTouch project is more advanced and propose windows installer [11:03] well, frankly this stuff is meant to be installed by OEMs (i. e. phone producers) [11:04] I would not ever expect a phone buyer or a shop to install and configure it [11:04] actually it is intended to linux users and courageous people not everybody [11:04] at this point this is very much a developer preview [11:04] the other users will be addressed when the project will be more mature [11:05] I would like a Linux phone. Today, which solution is available? [11:05] Android [11:05] is based on a Linux kernel [11:05] Android isnt a real linux [11:05] jolla [11:05] it's a Linux kernel [11:05] there is no real or false Linux [11:06] there are many distibutions [11:06] for every need [11:06] just take the one for you [11:06] Come on. Ubuntu is meant to be the best, so, why still no ubuntu on phones? [11:06] I think you should see android [11:06] there is no best [11:07] i use ubuntu/android/debian/raspbian/redhat [11:07] it depends on my needs and this is what I like in Linux worldf [11:09] good luck raptor67782 in your life that seems to be really complicated out of easy ways [11:09] I have an android tablet archos, but I have no idea how to install linux ubuntu on it [11:10] I am not working in informatic. Anytime [11:11] dadel: I do physics. I do regularly schrodinger equations. For me, it is simple, I could laugh trying to solve on. Furhtermore, physics pay well, probably better than a developer ;) [11:11] there is currently no ubuntu for the archos tablet [11:11] (only for the n7 and n10, and even these are only developer previews) [11:12] I didn't know, I shall try to sell it on ebay, to buy one compatible with ubuntu. Maybe for next year... [11:20] can i install ubuntu touch on my HCL ME U-1 [11:23] MacSlow: https://code.launchpad.net/~mzanetti/unity8/lockscreen-default-variable-pinlength/+merge/197848 [11:24] MacSlow: #ubuntu-unity btw? [11:24] mzanetti, on it [11:37] I'm about to write the first-ever automatic test for platform-api [11:38] any hints as to what our preferred C++ project test framework is? which of our projects would be a good example? [11:38] tvoss: ^ I think you might be able to answer this? [11:39] (it's also my first-ever C++ test, so bear with me..) [11:39] unless it's ok to write it using python3 unittest, which would make things a lot easier :) [11:40] pitti, we use google test all over our projects [11:40] pitti, let me find you a good example [11:40] * pitti looks at libgtest-dev [11:41] pitti, http://bazaar.launchpad.net/~phablet-team/process-cpp/trunk/files/head:/tests/ has got a good overview, including the required CMake stuff [11:43] tvoss: nice and small, thanks! [11:43] pitti, yup :) just let me know if you need anything else [11:43] tvoss: I guess gtest supports setup/teardown too, but I'll RTFM [11:43] tvoss: so "gtest" and "these examples" should be fine [11:44] pitti, great. For setup/teardown: you probably want TEST_F(Fixture, ...), with you defining the Fixture and per-suite/per-case setup and teardown [11:48] right [11:52] popey: did you say unlocking crashes your phone? [11:52] it did earlier [11:52] popey: does that mean it is not recommended for me to upgrade to the latest image now? === chihchun is now known as chihchun_afk [11:52] uh, no, it was a one off [11:52] and that wasn't latest, it was #47 [11:52] just updated to #48 [11:53] popey: ah, okay. Then I proceed with installing the new image :) thanks. [11:57] is there a way to know when the device was asleep? [11:57] https://wiki.ubuntu.com/Power#Phone asks for it [12:02] Laney, powerd communicates through dbus ... might be possible to capture that from there [12:05] I don't think capturing would work; if necessary, powerd should store it and provide that information [12:06] * Laney will file a bug [12:06] no, unity8 should [12:06] why? [12:07] powerd runs as system service ... unity8 needs to handle its dconf settings already (i was told) ... so that info should go there as well i think [12:10] It sounds weird to do that, but someone can reassign if that is true [12:14] Laney, ideally that would be a feature we would add to upower? [12:14] ++ [12:14] Laney, seb128: upower doesn't do suspend any more [12:15] pitti, well, it records charge history with state, I'm sure it would be easy to listen to some suspend/resume signals and add those events to the record? [12:16] upower would be better and could listen to logind's signals [12:16] yes, you can listen to logind's PrepareForSleep signal, but I doubt that everything on the phone actually uses that [12:16] but I don't think that works with powerd [12:16] k [12:16] so we just can't do that doted line in the graph [12:16] not yet [12:17] seems another of those "too much work for the benefit it brings" [12:17] upower records the charge history, yes, but only if you actually have a battery [12:17] You just get events either side of the suspend afaict [12:17] we have more important things to do that make the charge graph indicate sleep times [12:17] nothing to say 'you were suspended in this period' [12:17] yeah, it's surprisingly difficult to find out when a machine went to sleep and came back :/ [12:17] yeah, well, just filing bugs for things that we can't do [12:17] the kernel log obviously has it, but it doesn't send notifications [12:17] if they get fixed then hey presto we can do them :-) [12:18] pitti, if you don't have a battery you don't have a charge graph anyway ;-) [12:18] right [12:18] dont we need something that monitors the power button anyway (for a shutdown menu) [12:18] Laney, right === boiko_ is now known as boiko [12:18] ogra_, well, the device suspends by itself if you don't touch it [12:18] no need to press a button [12:19] oh, roght ... forgot about that [12:19] seb128, Laney: the charge log already has real-time timestamps, doesn't it? this should have gaps while the device is suspended [12:19] yes [12:20] is there a reliable way to work that out from those? [12:20] for a gap, there are several possibilities: off, suspend, or upower crash :) [12:20] but on a graph they'd all show as the same time gap? [12:22] Laney, pitti, upower doesn't do regular recording, it just records changes [12:22] I seem to get it at the resolution of 1% changes [12:22] yeah, it writes datas when there is a change [12:23] let me file a bug for hadess to close ;-) [12:23] about? [12:23] I filed half a dozen [12:23] what pitti said, tracking sleeps [12:23] anyone get Ubuntu to worm on the note II yet? [12:23] * Laney slithers [12:24] Laney, k, I didn't file that one ;-) [12:24] I filed other about stuff like "if you are charging and unplug, you are getting the new status only on next % change" [12:24] so you might in a wrong state until that 1% is used [12:26] !deviced | sticker [12:26] nod [12:27] !devices | sticker [12:27] sticker: You can find the full list of devices, official images, community images, and works in progress at https://wiki.ubuntu.com/Touch/Devices [12:27] (sorry for the typo) [12:36] are there specific times when new images are created? [12:36] t1mp, not atm, but we're working on that [12:37] if I need to get something in an image before tomorrow afternoon (london time), when should it be in a UITK release? [12:37] t1mp, if i trigger a build i announce it in #ubuntu-ci-eng (and if it is done as well) [12:38] t1mp, best ask in the above channel, the guys there do the testing [12:38] (before anything lands) [12:39] ogra_: ok, thanks === chihchun_afk is now known as chihchun === chihchun is now known as chihchun_afk === mhr3__ is now known as mhr3 [14:21] stgraber, didnt you add a fix for the "not able to make image writable" issue recently ? seems ricmm sees it on his grouper [14:24] ogra_: yes, the initrd was changed for that [14:24] stgraber, yeah, just found it [14:25] dholbach, didrocks that's right. there's no indication at all that things are stuck in proposed. you only find out if you look, and it does not scale to look for everything you upload. that really needs to be fixed. an email when migration fails would *really* help [14:26] stgraber, so it seems the install ricmm has is actually from the 3rd ... but still seems to have a readonly loop device [14:26] mount: cannot remount block device /dev/loop1 read-write, is write-protected [14:27] barry, maybe somebody should bring it up on u-devel? [14:30] or just in #ubuntu-release ... [14:31] tvoss: is it some kind of policy to put all tests into /tests/, or could the simulated-sensor test live in src/ubuntu/testbackend/ right next to the implementation? [14:31] (wheer the maintainers of that bit of infrastructure live) [14:31] tvoss: the latter would avoid having to figure out an appropriate LD_LIBRARY_PATH, as cmake already sets an rpath (but that doesn't help us when calling it from another directory) [14:31] pitti, not a strict policy, I would call it best practice [14:31] tvoss: I have my first test working BTW, so gtest/cmake are fine [14:32] tvoss: ok, thanks [14:32] tvoss: I can set LD_LIBRARY_PATH it in the CMakeLists.txt for the add_test definition, then calling the test program directly will fail; but not such a big deal, I think [14:32] pitti, oh that's great [14:33] tvoss: http://paste.ubuntu.com/6524971/ [14:33] pitti, I think that's fine then for setting the LD_LIBRARY_PATH there. Perhaps put a comment in the CMakeLists.txt [14:33] tvoss: and that first test found the first bug :) [14:34] tvoss: yes, of course [14:34] tvoss: I guess usually people should call "ctest" or "make test", which will deal with that [14:34] pitti, usually, you just call make test, yeah [14:35] pitti, you can try make NightlyMemCheck, too [14:35] $ make -C obj && (cd obj; ctest --verbose) [14:35] pitti, to run everything under valgrind automatically [14:35] that's my invocation for a turnaround [14:35] tvoss: ah, nice [14:35] pitti, usually I do make && make test or make && make NightlyMemCheck [14:35] yeah, "make test" is horribly un-verbose [14:36] it doesn't show me the actual failures [14:36] tvoss: I guess I need some additional magic for NightlyMemCheck, doesn't exist here [14:36] but, that's for later [14:37] let's get these tests written [14:37] pitti, +1 [14:38] pitti, you want include include(CTest) as opposed to enable_testing() in your CMakeLists.txt [14:39] tvoss: ok; I was following some stackexchange advice [14:40] tvoss: works fine [14:41] tvoss: http://paste.ubuntu.com/6525001/ with above change and the LD_LIBRARY_PATH setting [14:42] * pitti wonders if set() will set the environment var, too; would look nicer for multiple tests [14:44] no, it doesn't; and CTEST_ENVIRONMENT doesn't help either, so go ugliness [14:47] pitti, looks good albeit the ugliness. I remember a lengthy cmake mailing list discussion about that specific part [14:47] tvoss: I was hoping for something like set(ENV{LD_LIBRARY_PATH} ...), but that doesn't work [14:48] I live too much in a make world.. [14:49] pitti, the above actually works, but not for ctest iirc [14:49] pitti, let me quickly check [14:49] tvoss: right, only for make [14:50] tvoss: the other bit of ugliness is that it currently hardcodes the gtest path; but the other alternatives of including huge amounts of boilerplate don't look any better to me.. [14:50] pitti, yup, agreed. And it will at least fail early and with a sensible error message === Namidairo is now known as Namidairo`bnc [14:55] zsombi, ping [14:55] cwayne_: pong [14:56] zsombi, is there a plan to let the user choose their app theme from system settings? [14:56] rather than editing an ini file [14:56] cwayne_: that's mostly a question to design I think :) [14:57] tvoss: thanks for the review; I'll commit that to my branch then, and flesh out the tests [14:57] cwayne_: settings will anyway edit the ini file :) [14:57] tvoss: we can discuss the finer details in the MP (I'm a C++ n00b and probably do a lot of things wrong :) ) [14:57] zsombi, ah, so there's no plan to move to a gsettings key or something like that? [14:58] cwayne_: so if design thinks that it should be there, it's up to seb128 to have it implemented :) [14:58] zsombi, cwayne_: there is no design for a such feature [14:59] cwayne_: yet we do not have such plans, however changing the system theme would also have to have a setting UI element... and as sebq128 says, so far there's no UI element in Settings app [14:59] pitti, happy to help :) [14:59] cwayne_: however UITK does not provide API yet for that purpose [15:00] zsombi, seb128 ACK, thanks [15:00] cwayne_: yw === codinho_ is now known as codinho [15:20] tvoss: argh, not enough; I need to run each test case in a forked subprocess, as with the current sensor API there is no way to "reset" everything (unload the library, free all existing sensors, etc.) [15:21] dholbach: hope my email captured the problem without being too negative ;) [15:21] * barry now goes to fix the bug [15:25] ogra_: http://paste.ubuntu.com/6525209/ [15:25] robotfuel, hey got your mail. Here's the MP: https://code.launchpad.net/~nicolas-doffay/ubuntu-ui-toolkit/fix-1231939/+merge/197899 [15:25] literally just proposed it now. [15:26] stgraber, yeah, weird, not sure why that doesnt work for ricmm [15:26] though he tries to remount /system [15:26] to put some changed android binaries in place [15:26] barry, yep, thanks! [15:30] ogra_: ah yeah, /system is another block device and may suffer from the same problem. I don't think I changed that one in the initrd. [15:30] ogra_: I believe I didn't bother since the system.img file has 0 byte free anyway, so it's not like you could write anything to it [15:30] ah [15:30] ricmm, ^^^ [15:31] I've been trying to get a,'My Notebook', to work on a Ubuntu after using it on the window's, without any success, does anyboay know how to...? [15:32] is there anyway i can compile ubuntu touch image myself? [15:32] ricmm: so your best bet is probably to copy system.img somewhere, dd some free space to it, then resize2fs it to get the new extra space, then loop mount it, add your stuff and finally overwrite the original system.img + reboot [15:32] asd [15:33] stgraber, he went for a cdimage image now :) [15:34] stgraber, oh, btw, i was told you will fix the isotracker so didrocks' team can trigger image builds, any ETA for that ? [15:37] nic-doffay: re-flashing my phone to test now. [15:37] robotfuel, cool ta [15:38] glad to hear it was ok too :P [15:45] ogra_: yeah, asac poked me about it, should be done at some point today [15:45] perfect, thanks [15:45] ogra_: except that even if you have access it apparently won't work [15:46] ugh, why [15:46] because they're not showing up on the tracker at the moment [15:46] whats the way to make them show up ? [15:47] there's probably some mismatch between nusakan and the tracker, looking for errors now in the logs [15:48] No iso.qa.ubuntu.com product found for ubuntu-touch/daily-preinstalled/trusty-preinstalled-touch-armhf; skipping. [15:48] right, so we'll need to tweak that a bit :) [15:49] yeah, thats a weird name :) [15:49] ogra_: so I'll disable all of the current Ubuntu Touch Preinstalled products on the tracker and add a single "Ubuntu Touch armhf" one and link it with nusakan [15:50] yeah [15:52] ogra_: so I'll need at least two touch builds to confirm that this works once I've done the changes [15:52] stgraber, ok, we'll likely do one in a few hours ... the next probably only tomorrow morning though [15:53] zsombi, so would it be a possibility to switch the theme to at least a gsetting? It just seems the more 'correct way' of doing things [15:55] cwayne_: most probably would... we could have this as future task for theming [15:56] did anyone get the emulator running on ubuntu 13.10? i just downloaded the "android-emulator_20131202-2236-0ubuntu1_i386.deb" package and ran the commands from the wiki. however the screen was black in the emulator [15:56] zsombi, if we could add that as a task that would be perfect, we're trying to minimize copying files into ~ for customization in favor of gsettings keys, this would be a big help for us :) [15:56] user82, the first boot takes over 10min [15:57] ogra_: so the change I just landed "should" (completely untested and not really testable short of spinning an image) at lesat push the build to the tracker [15:57] ogra_, oh...that might be it ;) [15:57] ogra_: once that's done, I'll look at the ACL and trigger of the build from the tracker (that's where we'll need a second build) [15:57] user82, subsequent ones are faster [15:57] the terminal did not move and i got impatient..but it sounds like a clue [15:58] stgraber, ok, i'll bring that up in the landing team meeting, we can probably spare a build [15:58] ogra_, even after the login already appeared? [15:58] "ubuntu phablet login" [15:58] yeah [15:58] allright [15:58] so i log in as "phablet" and keep on waiting? [15:58] cwayne_: not sure when are we going to have time for that though... [15:59] yey its up :) [15:59] zsombi, well as always, i'm happy to help with any testing if that helps you guys out at all [15:59] cwayne_: thx ;) === jdstrand_ is now known as jdstrand [16:00] zsombi, thank you :) this was one of our last files copied into ~, will be a big win for us, so again, happy to help :) [16:04] cwayne_: any preference on the key? === h[a]kr is now known as hakr [16:06] zsombi, maybe something like com.ubuntu.touch.app-theme? I don't have any strong preference [16:07] cwayne_: ok, we'll come up with sthing then :) [16:08] zsombi, sounds good to me :D [16:15] cwayne_: https://bugs.launchpad.net/ubuntu-ui-toolkit/+bug/1258209 [16:15] Ubuntu bug 1258209 in Ubuntu UI Toolkit "[theming] Store theme settings in gsetting" [Wishlist,New] === rickspencer3_ is now known as rickspencer3 [16:26] stgraber: thanks for checking this out for us [16:26] asac, not done yet :) [16:26] dholbach: s-i 2.0.2-0ubuntu2 just uploaded. it was an out-of-date autopkgtest dep [16:27] ogra_: hence i didnt say "sorting" :) [16:28] lol [16:28] janimo`, you sole my WI ! [16:28] *stole [16:28] (thanks !!) [16:29] ogra_, well, I just want to leave you more time for the hard stuff (cdimage) :) [16:29] heh [16:31] tvoss, Chatting with charles here about whether LTTng in UAL should be enabled by default. Curious what you think. [16:31] barry, *crossing fingers* :) [16:41] tedg, charles I would make it a build option, with that we can easily generate instrumented packages if required [16:42] tvoss, My concern there was that we'd not be able to get data on images then... only custom stuff. [16:42] In theory we'd never need timing data. [16:43] tedg, alternatively, you could make it a command line option [16:43] tvoss, Won't the kernel tell me if it's enabled? [16:43] tedg, I would need to check that [16:43] tedg, a command line option is more explicit and self-documenting though [16:44] tvoss, Sure, but that doesn't really help the can't use an image problem. [16:45] tedg, I think I'm not following :) [16:45] tedg, if the kernel does not support it, the call to updating the probe will just fail [16:45] tvoss, http://twitter.com/tedjgould <- Click on "Follow" ;-) [16:46] tedg, so enable in build by default, but make it optional at runtime [16:47] tvoss, I guess for me, do you see any reason that we'd use the tracing on a production image. Or is that not a useful requirement. I'm undecided on that. [16:48] tedg, hmmm ... I think it's undecided so far. That being said, I would enable it in the build by default [16:49] tvoss, Okay, so we'll enable it and I'll look to see what happens at the kernel level. It might be faster to check there than to do any other command line parsing or environment variables. [16:49] tedg, but still: having it in the kernel might not necessary mean: do the actual tracing [16:50] tvoss, Sure, I'm hoping they'll tell us both. [16:50] tedg, I think a command-line flag would be the best approach [16:51] tvoss, The problem is that we're not running any of the tools directly, but via the upstart jobs. So I think injecting an environment variable into upstart might be easier. [16:52] tedg, cool [16:53] ted, the current MR always enables it; IMO making it optional is a nice middle ground between that and disabling it in production [16:55] I agree about using an env variable even if it's less self-documenting -- it doesn't matter how documenting the command-line args are if we're not always able to get the arg passed in [16:55] greyback: hi! Any news about bug 1254333? [16:55] bug 1254333 in Unity 8 "Sign-on UI makes Unity8 restart" [Undecided,Confirmed] https://launchpad.net/bugs/1254333 [16:56] mardy: hey. I've not been able to look at it this week. I've good idea why it fails, but the fix isn't easy. === hakr is now known as h[a]kr [16:59] tvoss, Looking at man lttng-ust and there's a var LTTNG_UST_REGISTER_TIMEOUT. Seems like we could just set that to zero by default. And then allow overriding it to something more reasonable if we want to trace. [17:01] tedg, makes sense, and people actually use it for that purpose, see http://lists.lttng.org/pipermail/lttng-dev/2013-July/020968.html [17:01] tvoss, Ha, I was just there :-) [17:15]