| pitti | Good moring | 05:17 |
|---|---|---|
| darkxst | hey pitti | 06:16 |
| darkxst | pitti, could you take a look at https://code.launchpad.net/~darkxst/ubuntu/trusty/gnome-settings-daemon/lp1265127 | 06:19 |
| pitti | darkxst: queueing for today | 06:21 |
| darkxst | pitti, thanks | 06:21 |
| mlankhorst | early morning hacking is best hacking :) | 07:25 |
| pitti | mlankhorst: +1 | 07:32 |
| ritz | is it just me, or is gnome-shell staeel like unity ? | 08:33 |
| ritz | stating* | 08:33 |
| ritz | starting* to | 08:33 |
| hyperair | uh no, there were accusations of gnome-shell copying unity from some time back. | 08:35 |
| darkxst | hyperair, or the other way around! | 08:37 |
| hyperair | well, considering unity had its roots in the ubuntu network remix, i'm inclined to think that unity started first. | 08:38 |
| hyperair | er netbook | 08:38 |
| hyperair | what i particularly find funny are the comments that compiz is bloated and slow, and people should use mutter, when compiz sans unity used to run at 20MB RSZ. | 08:39 |
| hyperair | if anything, it was unity that slapped on all the bloat onto compiz. | 08:40 |
| hyperair | from RSZ went from 20MB to 100+MB | 08:40 |
| darkxst | hyperair, pretty sure the overall design concept of gnome-shell predates the netbook stuff | 08:42 |
| hyperair | bah | 08:43 |
| hyperair | this war never ends | 08:43 |
| darkxst | no war! | 08:44 |
| darkxst | I refuse to participate in the "I dislike feature X, so therefore nobody likes it" rubbish | 08:45 |
| hyperair | isn't that why we're getting a spanking new file manager? | 08:46 |
| hyperair | what comes next anyway? a new evince? | 08:46 |
| ali1234 | hyperair: not compiz sans unity. compiz sans c++ | 08:46 |
| hyperair | ali1234: nope. | 08:47 |
| ali1234 | yes. | 08:47 |
| hyperair | ali1234: compiz with c++ was still at 20MB RSZ | 08:47 |
| hyperair | i was compiling daily compiz++ builds on my machine | 08:48 |
| hyperair | and using that before unity came along | 08:48 |
| hyperair | so i know this for a fact. | 08:48 |
| darkxst | hyperair, not real sure, I just cheer everytime a new component is announced, since that another we (ubuntu GNOME) can move to tracking upstream ;) | 08:48 |
| hyperair | lol | 08:48 |
| hyperair | the current situation between indicator and tray icon is a bit crappy | 08:49 |
| darkxst | with mir/unity8 the only remaining painpoint right now is gnome-desktop3 | 08:49 |
| darkxst | hyperair, gnome-shell doesnt really have tray icons in the traditional sense | 08:49 |
| hyperair | pretty much all of the tray icon apps have been patched to use indicators, but the UX for indicators suck in gnome shell | 08:49 |
| hyperair | er well, the notification bar thing at the bottom right. | 08:49 |
| hyperair | they work much better as tray icons than as indicators | 08:50 |
| hyperair | indicators are menus, whereas those things in the gnome shell notification bar don't really fit in as menus | 08:50 |
| hyperair | which just makes the situation pretty damn crappy when attempting to use gnome-shell | 08:51 |
| darkxst | hyperair, each to their own, I would rather not have 100 icons in my tray | 08:52 |
| ali1234 | notification icons also cannot support multimonitor | 08:52 |
| darkxst | and menus on tray icons seem kinda pointless | 08:52 |
| hyperair | darkxst: you could hardly fit 100 of them in the tray | 08:52 |
| * hyperair shrugs. i'm using unity happily despite my gripes | 08:53 | |
| hyperair | my solution was to just bump my RAM up as high as i could go | 08:53 |
| hyperair | because apparently nobody working on desktop unity had low-ram constraints. | 08:53 |
| hyperair | which is why the indicators can take several hundred megabytes of memory at a time until they're killed | 08:54 |
| hyperair | and notify-osd leaks til today | 08:54 |
| darkxst | ^nobody full stop! | 08:54 |
| ali1234 | i like to fix memory leaks - which ones are leaky? | 08:54 |
| darkxst | I running 32GB on desktop, 8GB on laptop | 08:54 |
| hyperair | darkxst: hey 4GB of RAM isn't what i call low-RAM, but it's painful to run unity on even that kind of spec. | 08:54 |
| darkxst | hyperair, gnome-shell runs well on 4GB :) | 08:55 |
| hyperair | darkxst: it's only when i hit 8GB that i stopped caring so much for the memleaks | 08:55 |
| hyperair | yeah well, it also sucks more | 08:55 |
| hyperair | i really like the super+1,2,3,4,5,6,7,8,9,0 for the first 10 windows in the launcher that you get in unity | 08:55 |
| hyperair | which gnome shell doesn't seem to be picking up for whatever reason | 08:55 |
| hyperair | so screw that, i'm sticking with unity | 08:56 |
| darkxst | hyperair, most of my VM's get 2GB, but they don't really run the extra RAM blood suckers like firefox, libreoffice | 08:56 |
| hyperair | darkxst: yeah then that doesn't count. | 08:56 |
| hyperair | ever tried running chromium and thunderbird together? | 08:56 |
| hyperair | i had 8GB of RAM, and i was so frustrated with the overall memory usage i went to mutt | 08:56 |
| mlankhorst | it does count.. | 08:56 |
| hyperair | it was a good move though | 08:56 |
| darkxst | hyperair, sure, but that makes the whole leaky DE thing, null and void! | 08:57 |
| hyperair | pfft. | 08:57 |
| hyperair | when you can't run a browser and an email client concurrently on 4GB of RAM, the situation is fscked up. | 08:57 |
| ali1234 | xubuntu discussed increasing the minimum requirement to 1GB this cycle, almost entirely because of firefox | 08:58 |
| hyperair | derp 1GB. | 08:58 |
| darkxst | firefox leaks like a sieve | 08:58 |
| hyperair | hyper derp. | 08:58 |
| hyperair | not as bad as thunderbird. | 08:59 |
| hyperair | no actually firefox didn't leak much for me | 08:59 |
| hyperair | it just had consistently high memory consumption | 08:59 |
| mlankhorst | try firefox without javascript | 08:59 |
| darkxst | sure its the 3rd party firefox that leaks! | 08:59 |
| hyperair | i hate you. | 08:59 |
| darkxst | and the rubbish GC, but that is getting fixed ;) | 09:00 |
| ali1234 | "try not going on any websites, it will really reduce the memory use" | 09:00 |
| hyperair | lol | 09:00 |
| hyperair | thunderbird's awesome. you should see it trying to fetch lkml from gmane. | 09:00 |
| darkxst | ali1234, switching off computer elimates memory usage! | 09:00 |
| hyperair | it goes past 4GB of RAM usage, and grinds the CPU for over half a day | 09:00 |
| ali1234 | that's exactly what happens when i try to connect it to my gmail account | 09:01 |
| hyperair | and oh yeah chromium actually takes more RAM than firefox | 09:01 |
| darkxst | I've never seen thunderbird get past 1GB | 09:01 |
| hyperair | but firefox operates a hell lot slower | 09:01 |
| hyperair | darkxst: you haven't opened big enough folders | 09:01 |
| hyperair | my gmail's all mail folder made claws-mail rise to 500+MB | 09:01 |
| darkxst | I only use it for IMAP accounts | 09:01 |
| hyperair | yeah so was i | 09:01 |
| hyperair | except for nntp | 09:01 |
| darkxst | probably 10GB all up | 09:02 |
| hyperair | anyway it was thunderbid that leaked like a sieve. firefox was mostly decent. | 09:02 |
| tjaalton | sigh, so which package trumps the user settings like focus-follows-mouse on upgrades? | 09:11 |
| tjaalton | it gets reset every now and then | 09:12 |
| pitti | tjaalton: bug 1063617 | 09:13 |
| ubot2` | Launchpad bug 1063617 in Compiz 0.9.9 "1:0.9.8+bzr3319-0ubuntu1 regression: keeps setting gsettings keys to wrong values" [High,Fix committed] https://launchpad.net/bugs/1063617 | 09:13 |
| pitti | it's ancient | 09:13 |
| tjaalton | ahh, thanks | 09:13 |
| tjaalton | yeah it is | 09:14 |
| pitti | I have a couple of "gsettings set" commands in my "start my session" script | 09:14 |
| tjaalton | huh, now I lost my desktop grid | 09:14 |
| tjaalton | marked fix released | 09:16 |
| tjaalton | but I'm on trusty | 09:16 |
| seb128 | larsu, bug #1277370 | 09:21 |
| ubot2` | Launchpad bug 1277370 in evince (Ubuntu) "Now used Aiatana design blocks Evince accessibility usage if the current session is not Unity" [Undecided,New] https://launchpad.net/bugs/1277370 | 09:21 |
| Laney | HMM! | 09:21 |
| seb128 | Laney, HMM? | 09:21 |
| Laney | getting a partial upgrade that I think is mir/xorg | 09:21 |
| larsu | Laney: morning! | 09:21 |
| Laney | did you see that? | 09:21 |
| Laney | o hai | 09:21 |
| seb128 | yes | 09:21 |
| Laney | also | 09:23 |
| Laney | happyaron: E: Unable to locate package libreoffice-help-en-us | 09:23 |
| Laney | E: config/hooks/100-ubuntukylin.chroot failed (exit non-zero). You should check for errors. | 09:23 |
| seb128 | it wants to uninstall the old soname version | 09:23 |
| Laney | aaaaaaand | 09:23 |
| Laney | The following packages have unmet dependencies: xserver-xorg-input-synaptics : Breaks: kde-config-touchpad (< 0.8.1-2~) but 0.8.1-1ubuntu6 is to be installed | 09:24 |
| Laney | kubuntu | 09:24 |
| Laney | today's image build failures | 09:24 |
| tjaalton | uh | 09:24 |
| seb128 | ok, so all good | 09:24 |
| Laney | :P | 09:24 |
| seb128 | ;-) | 09:24 |
| seb128 | happy friday desktopers btw! | 09:24 |
| Laney | oh yeah! | 09:24 |
| tjaalton | tgif | 09:25 |
| pitti | bonjour seb128, ça va ? | 09:25 |
| pitti | seb128: you too! | 09:26 |
| seb128 | pitti, hey, wie gehts? | 09:26 |
| pitti | seb128: gut, danke! | 09:26 |
| happyaron | Laney: got it, added to my list... | 09:26 |
| Laney | happyaron: I think they got renamed or removed or something? | 09:26 |
| pitti | much better than today's live CD, anyway -- compiz/unity are b0rked | 09:26 |
| Laney | hey pitti | 09:26 |
| pitti | no panel, no launcher, no window decorations | 09:26 |
| pitti | ctrl+alt+t (terminal) and nautilus work, though | 09:26 |
| Laney | sounds like the protobuf problem | 09:27 |
| happyaron | Laney: maybe, first day back from holiday and still catching up, :) | 09:28 |
| Laney | pitti: does it have the newest libprotobuf8? | 09:28 |
| pitti | ah no, -7ubuntu1 | 09:29 |
| Laney | nod | 09:30 |
| pitti | http://cdimage.ubuntu.com/cdimage/daily-live/current/ is from today | 09:30 |
| pitti | and yet http://cdimage.ubuntu.com/cdimage/daily-live/current/trusty-desktop-amd64.manifest is old | 09:30 |
| Laney | the isos look old | 09:30 |
| pitti | oh right, amd64+mac is from today, but not the others | 09:30 |
| Laney | I wonder why this is | 09:31 |
| Laney | perhaps something was disabled for the point release | 09:32 |
| pitti | http://cdimage.ubuntu.com/cdimage/daily-live/pending/ are current | 09:32 |
| pitti | yep, failures on https://jenkins.qa.ubuntu.com/view/Trusty/view/Smoke%20Testing/ | 09:32 |
| Laney | oh, I always forget about that | 09:32 |
| Laney | i don't understand the failure | 09:33 |
| pitti | I don't know exactly which tests are being looked at to promote an image to /current | 09:34 |
| pitti | but it seems the health-check ones didn't even run for the 06 and 07 build? | 09:34 |
| pitti | and they never succeeded anyway | 09:34 |
| seb128 | xnox, if (g_strcmp0 (g_getenv ("XDG_CURRENT_DESKTOP"), "GNOME") == 0 | 09:55 |
| seb128 | ... upstream code | 09:56 |
| seb128 | else | 09:56 |
| seb128 | ... ubuntu change | 09:56 |
| mpt | MacSlow, would it bother you at all if we switched to tracking Notify OSD bugs solely on the package <https://bugs.launchpad.net/ubuntu/+source/notify-osd>, no longer on the project? <https://bugs.launchpad.net/notify-osd> | 09:58 |
| seb128 | robert_ancell, sudo apt-get -o Dpkg::Options::="--force-confmiss" install --reinstall gnome-settings-daemon | 09:59 |
| xnox | seb128: ack. | 09:59 |
| xnox | robert_ancell: where is the ubuntu-settings-daemon branch? | 09:59 |
| xnox | robert_ancell: such that i can use it. | 09:59 |
| robert_ancell | xnox, lp:~ubuntu-desktop/gnome-settings-daemon/unity and lp:~robert-ancell/gnome-settings-daemon/unity | 10:03 |
| darkxst | Hey Laney | 10:11 |
| Laney | hellllllllllllllloooooooooooooooooooooooooo | 10:11 |
| darkxst | hey seb128 | 10:11 |
| seb128 | hey darkxst | 10:12 |
| darkxst | seb128, can we land gnome-desktop transition next week? cant really wait on g-s-d fork, because we will just run out of time? | 10:13 |
| pitti | seb128: https://code.launchpad.net/~darkxst/ubuntu/trusty/gnome-settings-daemon/lp1265127/+merge/201995 sounds ok to me, any objection? darkxst replied to your question | 10:17 |
| seb128 | darkxst, that involves that new dbus service for screen resolution handling? | 10:18 |
| seb128 | pitti, looks fine to me, do you want to sponsor it? | 10:18 |
| pitti | seb128: yes | 10:18 |
| * pitti resolves conflicts as another versino got uploaded in between, but it's trivial | 10:19 | |
| darkxst | seb128, yes, but its largely the same code moved to a different place | 10:19 |
| seb128 | pitti, danke | 10:19 |
| darkxst | pitti, that MP is several weeks old now | 10:20 |
| seb128 | darkxst, well, it's an infrastructure change, replacing a library by a dbus service | 10:20 |
| pitti | darkxst: yes, I didn't blame you :) | 10:20 |
| seb128 | darkxst, did you sort out of the config migration question? | 10:20 |
| darkxst | seb128, I dont think that is worth the effort | 10:21 |
| darkxst | it would quite some work to make configs migrate and most people don't even use them | 10:22 |
| darkxst | the ones that do, have to reconfigure just once | 10:23 |
| seb128 | I'm not convinced that transition is something we should do before the LTS... | 10:25 |
| darkxst | seb128, it has too happen on way or the other | 10:34 |
| seb128 | "has to" | 10:35 |
| darkxst | obviously you guys don't need it but we do | 10:35 |
| seb128 | not sure I agree with that, at least not this cycle | 10:35 |
| seb128 | darkxst, what do you need in the new version? | 10:37 |
| pitti | I'm still baffled that upstream didn't provide a migration for configs -- how are distros supposed to cope with that? | 10:39 |
| seb128 | pitti, yeah, that's annoying | 10:41 |
| pitti | well, it's more than that | 10:42 |
| darkxst | seb128, well nothing specifically, but its more a case not spending the rest of the cycle backporting upstream fixes | 10:44 |
| seb128 | the new version changes quite some apis, it's going to force us to update code in g-s-d u-s-d gnome-screensaver colord | 10:45 |
| seb128 | there is the config migration issue | 10:45 |
| seb128 | then it's a new architecture/dbus service, it's basically untested for us | 10:45 |
| seb128 | that sounds like a risky change mid-cycle of a LTS | 10:45 |
| darkxst | seb128, right I can deal with the api changes | 10:46 |
| seb128 | thanks, but that doesn't resolve the other issues though | 10:47 |
| darkxst | seb128, and really, its seems to becoming, wait on this, wait on that, oh you have now run out of time! | 10:47 |
| darkxst | seb128, as much as I know you are trying to unblock these things for us, its just kinda frustrating at this point. | 10:50 |
| seb128 | darkxst, there are conflicting needs, and we are working on solving that, that's why we are doing unity-control-center/unity-settings-center | 10:51 |
| seb128 | but resources are what they are and it takes some time to get there | 10:51 |
| seb128 | sorry about that, but I just don't see a way around | 10:51 |
| seb128 | doing risky changes just before the LTS is not benefiting anyone I think | 10:52 |
| darkxst | seb128, I am well aware of all that, but really until gnome-desktop3 is in a state it can be forked we still have a big pain point right ther] | 10:52 |
| seb128 | right, we have, I just don't see a way to resolve that in a non risky way before the LTS | 10:53 |
| seb128 | risky or costy | 10:53 |
| pitti | well, if there is no migration path, how would you deal with the migration in ubuntu-gnome? | 10:53 |
| darkxst | hmm, is display config really that critical? 90% probably don't even use it | 10:54 |
| pitti | you mean the resoltion/monitor config? why would people not use that? | 10:55 |
| darkxst | the 10% that do, can spend ~3secs reconfiguring | 10:55 |
| MacSlow | mpt, that would be ok with me | 10:56 |
| darkxst | pitti, because it just works perhaps? how many go and mess with config in that case? | 10:56 |
| pitti | darkxst: I don't know numbers, I just know that on conferences pretty much everyone has to | 10:56 |
| MacSlow | mpt, would this change only for NotifyOSD or for all packages in general? | 10:57 |
| mpt | MacSlow, just NotifyOSD. It would match what the Touch-specific packages already do, for example. | 10:57 |
| seb128 | darkxst, what "just works"? configuration is a custom thing, my docked config turns off my laptop screen when I dock it for example | 10:58 |
| pitti | darkxst: but yes, it certainly isn't an insurmountable pain, except that you have to re-configure with every device you already saw (monitors.xml is by device set, not a single configuration) | 10:58 |
| MacSlow | mpt, only funny is that NotifyOSD isn't really touch-specific at all :) | 10:58 |
| mpt | MacSlow, sure, that’s why I said “for example”. Ubiquity is another example. | 10:58 |
| mpt | MacSlow, anyway, thanks, I’ll do the refiling on Tuesday. | 10:59 |
| darkxst | seb128, that is a gsettings key atleast in shell, no reason that would change | 10:59 |
| MacSlow | mpt, ok | 10:59 |
| pitti | but really, why can't they leave in the old xml reading code for a cycle or two and do a live migration upstream? | 10:59 |
| seb128 | darkxst, are you sure? the config is currently in ~/.config/monitors.xml | 10:59 |
| darkxst | seb128, was | 11:00 |
| seb128 | well, currently in Ubuntu | 11:00 |
| seb128 | we are talking about config migration | 11:00 |
| seb128 | so what we have and how we keep those configs working on upgrade | 11:00 |
| darkxst | right, I get that | 11:01 |
| robert_ancell | fginther, I'm confused by the Jenkins failures for u-c-c (https://jenkins.qa.ubuntu.com/job/unity-control-center-trusty-amd64-ci/3/console). It's complaining about missing dependencies but I'm not sure if it's downloaded any of them | 11:06 |
| Laney | seb128: http://ci.ubuntu.com/smokeng/trusty/touch/mako/169:20140207:20140115.1/6492/ubuntu-system-settings-autopilot/746509/ | 11:09 |
| Laney | :((((((((((((( | 11:09 |
| seb128 | Laney, what did you do?! | 11:10 |
| Laney | bah! | 11:12 |
| darkxst | seb128, go ahead and feel free to fork gnome-desktop3 as well ;) | 11:13 |
| seb128 | darkxst, it's less easy with a library... | 11:13 |
| seb128 | darkxst, with what library would we build e.g colord | 11:14 |
| darkxst | seb128, I am working with upstream to move the important bits into gtk | 11:14 |
| seb128 | great | 11:14 |
| darkxst | no idea what will happen with the bits that you guys are hanging on to though | 11:14 |
| darkxst | seb128, upstream wants gnome-desktop to only be used gnome-shell essentially | 11:15 |
| darkxst | which right now is really just the thumbnail module | 11:17 |
| darkxst | (excluding the others that are reverted in ubuntu like background etc) | 11:18 |
| seb128 | darkxst, "only be used in" you mean? | 11:21 |
| darkxst | either way surely you can see why I get frustated? my g-c-c patches from last cycle are still waiting to land, although now half are unnecessary | 11:21 |
| seb128 | yes, and as said I'm sorry about that | 11:21 |
| seb128 | I just don't have a magical solution for you | 11:21 |
| darkxst | seb128, as in gnome-desktop should become a shared (private) library for gnome-shell, gdm, etc | 11:22 |
| darkxst | as in no apps should depend on it | 11:22 |
| darkxst | in which case it would be very easy for 'legacy' users to fork | 11:23 |
| darkxst | for their needs | 11:23 |
| robert_ancell | xnox, lp:~robert-ancell/indicator-network/network-manager | 11:24 |
| xnox | robert_ancell: thanks! | 11:27 |
| * darkxst is gone for the night ;) | 11:34 | |
| robert_ancell | fginther, could you set up Jenkins for lp:unity-settings-daemon | 11:41 |
| robert_ancell | darkxst, bye | 11:47 |
| seb128 | Laney, $ LANG= LANGUAGE= LC_ALL=bu_GY.UTF-8 autopilot run ubuntu_system_settings | 11:57 |
| === MacSlow is now known as MacSlow|lunch | ||
| qengho | didrocks: i think it's https://bugs.launchpad.net/ubuntu/+source/chromium-browser/+bug/1250174 | 12:12 |
| ubot2` | Launchpad bug 1250174 in libdrm-2.4.23 (Ubuntu) "chromium-browser crashed with SIGSEGV in content::GpuWatchdogThread::DeliberatelyTerminateToRecoverFromHang()" [Undecided,Confirmed] | 12:12 |
| robert_ancell | seb128, Laney, darkxst, bug 1277485 | 12:19 |
| ubot2` | Launchpad bug 1277485 in unity-settings-daemon (Ubuntu) " [MIR] unity-settings-daemon" [Wishlist,New] https://launchpad.net/bugs/1277485 | 12:19 |
| seb128 | \o/ | 12:19 |
| Laney | neat | 12:24 |
| robert_ancell | seb128, Laney, please add any packages that need migration to bug 1277487 | 12:37 |
| ubot2` | Launchpad bug 1277487 in unity-greeter (Ubuntu) "Create Unity Settings Daemon so can remain on old GNOME Settings Daemon version" [Undecided,New] https://launchpad.net/bugs/1277487 | 12:37 |
| === MacSlow|lunch is now known as MacSlow | ||
| fginther | robert_ancell, I've added that to task list and should have it updated today. In the future, please contact the vanguard in #ubuntu-ci-eng as we are trying to share responsibility for these things among the team | 13:54 |
| Laney | desrt: http://people.canonical.com/~laney/0001-Fix-tests-for-builddir-srcdir-by-setting-G_TEST_-SRC.patch | 14:02 |
| Laney | seb128: just tried, it passed for me here | 14:16 |
| * seb128 shakes fist at Laney | 14:16 | |
| Laney | :(((((( | 14:16 |
| seb128 | you want to make me debug that one, don't you! | 14:16 |
| desrt | Laney: pushed | 14:17 |
| desrt | but there are more distcheck issues now.... msvc stuff :( | 14:17 |
| Laney | messy buckets | 14:17 |
| Laney | :( | 14:17 |
| desrt | srsly. | 14:17 |
| Laney | larsu: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=706065 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=712208 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=712628 http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=675008 | 14:22 |
| ubot2` | Debian bug 706065 in libvte-2.90-common "libvte-2.90-common: /etc/profile.d/vte.sh is not sourced by interactive shells" [Normal,Open] | 14:22 |
| ubot2` | Debian bug 712208 in libvte-2.90-common "gnome-terminal: New tab is always opened in home directory (instead of working directory)" [Normal,Open] | 14:22 |
| ubot2` | Debian bug 712628 in libvte-2.90-common "gnome-terminal: ctrl-shift-n doesn't keep directory" [Normal,Open] | 14:22 |
| ubot2` | Debian bug 675008 in bash "bash: should handle /etc/bashrc.d (or similar) for non-login interactive shell" [Wishlist,Open] | 14:22 |
| mterry | jdstrand, if my app asks policykit about a certain permission, but apparmor denies the request, what does that look like to my app? (would apparmor ever deny a policykit authentication check?) | 14:23 |
| jdstrand | mterry: hey. so, the app is either going to use pkexec or talk to the service over dbus which then prompts the user | 14:24 |
| mterry | jdstrand, in my case, my app is pkcheck itself | 14:25 |
| jdstrand | mterry: with pkexec, if the policy doesn't allow using it, you'll get the denial | 14:25 |
| jdstrand | mterry: with talking over dbus, if the policy doesn't allow the dbus communication, you'll get a denial (the server won't ever get the message) | 14:26 |
| jdstrand | mterry: in either case, you'll know if apparmor is blocking you by looking in /var/log/syslog (note, not *kern.log or dmesg-- dbus denial necessarily have to go to syslog) | 14:26 |
| mterry | jdstrand, so it will look like a normal polkit denial to my app? | 14:27 |
| jdstrand | mterry: when you say 'pkcheck itself', what are you talking about? | 14:28 |
| mterry | jdstrand, I have an autopilot test that uses pkcheck on a certain NetworkManager permission as an integration check (to make sure that the session is marked active by logind/lightdm). | 14:29 |
| mterry | jdstrand, on Touch it works fine | 14:29 |
| robert_ancell | fginther, will do, thansk | 14:29 |
| mterry | jdstrand, on Desktop, it says permission for that pkcheck is denied, but we can't figure out why | 14:29 |
| mterry | jdstrand, I was wondering if it could be some weird apparmor thing | 14:29 |
| jdstrand | mterry: ewll, with my example, in the former case, (pkexec), the app would get EPERM, in the latter (dbus denial), the app gets DBus AccessDenied | 14:29 |
| jdstrand | mterry: the dbus denial will mention AppArmor in the message (and you'll see the message in syslog) | 14:30 |
| mterry | jdstrand, hmm, so I'd have to dig into pkcheck and see how it handles those error cases, to see if that's likely. But I could check the syslog.. | 14:30 |
| mterry | fginther, ^ | 14:30 |
| jdstrand | mterry: yes, check syslog | 14:30 |
| mterry | fginther, did we look at syslog for apparmor messages yet? | 14:30 |
| jdstrand | if there is no denial there, it isn't apparmor. plus, the app would have to be confined for it to be affected by apparmor any way | 14:31 |
| jdstrand | (well... technically that might not have to be the case, but in Ubuntu with how we confine things it is) | 14:31 |
| attente_ | ChrisTownsend, was your most recent test under a fresh unity session after installing the g-s-d from the ppa? | 14:34 |
| ChrisTownsend | attente_: Yes | 14:40 |
| attente_ | ChrisTownsend, does pressing alt+f work properly? | 14:42 |
| seb128 | larsu, wget https://launchpad.net/ubuntu/+source/file-roller/3.4.1-0ubuntu1/+build/3410817/+files/file-roller_3.4.1-0ubuntu1_amd64.deb | 14:43 |
| ChrisTownsend | attente_: You mean it opens the first menu on the panel and not the decoration, right? | 14:43 |
| attente_ | ChrisTownsend, yes | 14:44 |
| ChrisTownsend | attente_: Yes, that is working correctly. | 14:44 |
| seb128 | larsu, https://launchpad.net/~ubuntu-security/+archive/ppa/+build/4789846/+files/file-roller_3.6.1.1-0ubuntu1.2_amd64.deb | 14:45 |
| attente_ | ChrisTownsend, dbus-send --session --dest=org.gnome.Shell --print-reply /org/gnome/Shell org.gnome.Shell.GrabAccelerator string:'<Control>a' uint32:0 | 14:46 |
| attente_ | ChrisTownsend, then dbus-monitor member=AcceleratorActivated | 14:46 |
| attente_ | then ctrl+a | 14:48 |
| mterry | jdstrand, thanks btw! it probably isn't apparmor, just clutching at straws at this point | 14:48 |
| ChrisTownsend | attente_: http://pastebin.ubuntu.com/6891649 | 14:50 |
| attente_ | ChrisTownsend, have you run the tests in your current session? | 14:50 |
| ChrisTownsend | attente_: Not this exact same session. I can try it now. | 14:51 |
| attente_ | ChrisTownsend, thanks | 14:51 |
| ChrisTownsend | attente_: Same failure | 14:52 |
| jdstrand | np | 14:52 |
| ChrisTownsend | attente_: I need to step away for a little bit. | 14:53 |
| attente_ | ChrisTownsend, ok, np | 14:53 |
| qengho | robru: https://launchpad.net/~canonical-chromium-builds/+archive/stage | 14:56 |
| qengho | robru: https://code.launchpad.net/~chromium-team/chromium-browser/trusty-working debian/patches/\d-.* | 15:00 |
| ChrisTownsend | attente_: Hey, I'm back if you need me to do any more debugging. | 15:35 |
| attente_ | ChrisTownsend, hi | 15:35 |
| attente_ | ChrisTownsend, after running the AP tests, can you do 'dbus-monitor member=AcceleratorActivated' then try shift+ctrl+alt+a? | 15:36 |
| ChrisTownsend | attente_: Sure | 15:36 |
| ChrisTownsend | attente_: dbus-monitor has no output when I hit shift+ctrl+alt+a | 15:37 |
| attente_ | dbus-send --session --dest=org.gnome.Shell --print-reply /org/gnome/Shell org.gnome.Shell.GrabAccelerator string:'<Shift><Control><Alt>a' uint32:0 | 15:38 |
| ChrisTownsend | attente_: Ok, I did that | 15:39 |
| attente_ | 'dbus-monitor member=AcceleratorActivated' again | 15:39 |
| ChrisTownsend | Ok | 15:40 |
| attente_ | any output on shift+control+alt+a? | 15:40 |
| ChrisTownsend | attente_: No, nothing | 15:40 |
| attente_ | ChrisTownsend, what was the output of the dbus-send call? | 15:40 |
| ChrisTownsend | attente_: http://pastebin.ubuntu.com/6891915 | 15:41 |
| attente_ | ChrisTownsend, your compiz is trunk, right? | 15:43 |
| attente_ | ChrisTownsend, nvm i guess it must be if alt+f works for you | 15:43 |
| ChrisTownsend | attente_: It's compiz from the daily-build PPA that was published on 1/28. | 15:44 |
| ChrisTownsend | attente_: That definitely includes your Compiz changes. | 15:45 |
| attente_ | ChrisTownsend, ok | 15:45 |
| attente_ | ChrisTownsend, do you have any shortcuts set for switching keyboard layouts? | 15:45 |
| attente_ | in the text entry panel of g-c-c? | 15:45 |
| ChrisTownsend | attente_: Not that I know of. I'll look though. | 15:46 |
| ChrisTownsend | attente_: Just the Super+Space and Shift+Super+SPace. | 15:46 |
| attente_ | ChrisTownsend, ok | 15:46 |
| attente_ | ChrisTownsend, just to cover all bases: apt-cache policy gnome-settings-daemon | 15:47 |
| ChrisTownsend | attente_: Installed: 3.8.6.1-0ubuntu4ppa1 | 15:48 |
| attente_ | ChrisTownsend, sorry, let me think some more, i'm short on ideas for things we can try | 15:51 |
| ChrisTownsend | attente_: Ok, just ping me when you want me to try something. It's a mystery to me as well since it works for you and Trevinho. | 15:52 |
| === gatox is now known as gatox_lunch | ||
| attente_ | ChrisTownsend, you mentioned you had another application running called Terminator? | 15:53 |
| ChrisTownsend | attente_: It's a terminal emulator. I also ran this with Terminator closed and just using gnome-terminal with the same results. | 15:54 |
| attente_ | ChrisTownsend, ah, ok | 15:54 |
| === alan_g is now known as alan_g|vt | ||
| === gatox_lunch is now known as gatox | ||
| === alan_g|vt is now known as alan_g | ||
| mterry | dobey, so with ubuntu-purchase-service, any app that can talk to the service can buy stuff? What's the process like? | 19:01 |
| dobey | mterry: it's a service that launches the UI to purchase something. it's only used on touch at the moment and you can only buy apps at the moment | 19:03 |
| mterry | dobey, so it's not an insta-buy? Still requires user interaction before final purchase? | 19:03 |
| dobey | mterry: it opens a window/overlay with the web view for confirming the purchase. purchases are not automatable | 19:04 |
| mterry | cool | 19:04 |
| === m_conley_away is now known as m_conley | ||
| === mjohnson15_2 is now known as mjohnson15 | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!