UbuntuIRC / 2012 /02 /20 /#ubuntu-devel.txt
niansa
Initial commit
4aa5fce
=== Torpedo_Skyline is now known as TorpedoSkyline
[00:45] <TheMuso> superm1: You're a core-dev, so you should hae direct commit access to lp:~ubuntu-audio-dev/pulseaudio/ubuntu.precise
[01:03] <slangasek> lamont: did you file a bug about the nm-applet memory usage problem? I'm seeing the same problem here (though I noticed it at < 3GB, I didn't wait for it to hit 6GB)
[01:04] <RAOF> slangasek: lifeless was also seeing that; I believe he filed a bug.
[01:06] <slangasek> RAOF: right, bug #930491 then - thanks
[01:06] <ubottu> Launchpad bug 930491 in network-manager-applet (Ubuntu) "Large memory leak in nm-applet" [High,Confirmed] https://launchpad.net/bugs/930491
[01:16] <micahg> slangasek: I think cyphermox said he fixed some of that with the latest nm upload, but there's more to do
[01:18] <slangasek> micahg: ah, well, it's possible that fixing that bug is why nm-applet now won't *restart* for me ;)
[01:18] <slangasek> (glibc free() assert)
=== JackyAlcine_ is now known as JackyAlcine
[02:21] <Sarvatt> 31mb usage after 2 days uptime here, nm-applet is definitely much better that it was here
[02:24] <cyphermox> slangasek: ah, will look into it. could you get me a backtrace (things work here, so it might be that you have some device I couldn't test and messed up)
[02:24] <slangasek> cyphermox: bug #936644
[02:24] <ubottu> Launchpad bug 936644 in network-manager-applet (Ubuntu Precise) "nm-applet assert failure: *** glibc detected *** nm-applet: free(): invalid pointer: 0x0000000001eb1370 ***" [High,Confirmed] https://launchpad.net/bugs/936644
[02:24] <cyphermox> thanks
[02:24] <slangasek> so it's related to the new patch changes
[02:25] <cyphermox> yeah
[02:26] <cyphermox> I know what causes this, weird that this doesn't affect me too though
=== webjadmin is now known as JackyAlcine
[02:29] <lifeless> cyphermox: slangasek:
[02:29] <lifeless> 2386 robertc 20 0 3014m 1.8g 12m S 0 24.1 27:53.58 nm-applet
[02:30] <lifeless> Sarvatt: ^ not that much better :P
[02:30] <cyphermox> indeed
[02:30] <cyphermox> actually, it depends a lot on how much noise is around -- if there's a lot of changes in scan results, it tends to go up faster
[02:31] <lifeless> well, i have an N network
[02:31] <lifeless> and iwlwifi
[02:31] <cyphermox> something is still broken in the menu updates but I'm having a hard time tracking it down
[02:31] <lifeless> so my network reconnects like crazy
[02:31] <cyphermox> lifeless: ah
[02:31] <cyphermox> guess that would make sense yeah :/
[02:31] <lifeless> combined with it not remembering my credentials, thats pretty annoying (which is presumably related to the lack of themeing ...)
[02:31] <Sarvatt> wow, yeah the networks barely ever change here
[02:32] <cyphermox> Sarvatt: you're seeing it behaving better because before it was updating on every signal level change... a little too agressivel
[03:30] <superm1> TheMuso: yeah I realized that shortly later. I pushed it through
[03:42] <TheMuso> superm1: ok thanks.
[03:56] <wzssyqa> gpg --export-secret-subkey xxxxxxxx > sub.gpg give an error: gpg: key yyyyyyyy: not protected - skipped why?
=== webjadmin_ is now known as JackyAlcine
=== fenris is now known as Guest47832
=== Guest47832 is now known as ejat
=== Ursinha` is now known as Ursinha
=== Ursinha is now known as Guest37626
[07:43] <superm1> could someone clarify a seed question for me? mythbuntu isos are fairly oversized right now. if something that is only pulled in via recommends of something seeded is added to the mythbuntu.precise/blacklist file, will that keep it from getting pulled in or have some other undesirable effect?
[07:46] <dholbach> good morning
[08:20] <micahg> doko: would it be nice to drop gnat-4.4 for precise
=== lag` is now known as lag
[08:58] <Sweetshark> Hi there, looking for an package sponsor as piiti is on vacation.
[09:04] <tumbleweed> !sponsorship|Sweetshark
[09:05] <ubottu> Sweetshark: You can find out about the package sponsorship process here http://wiki.ubuntu.com/SponsorshipProcess - For !UDS sponsorship see http://uds.ubuntu.com/participate/sponsorship/
[09:10] <Sweetshark> tumbleweed: ok thx, let put that different: as pitti in on vacation, i need somebody else daring to upload libreoffice-3.5.0-1ubuntu1 to precise. its already parked on chinstrap and waiting to be picked up.
[09:11] <tumbleweed> ah, libreoffice. I can see why you may not want the usual queue
[09:12] <Sweetshark> tumbleweed: ;)
=== hughhalf is now known as hugh_afk
[09:41] <geser> Sweetshark: have you considered to apply for PPU rights for libreoffice so you can upload yourself in future?
[09:43] <dholbach> geser, https://wiki.ubuntu.com/BjoernMichaelsen/YourDeveloperApplication :)
[09:44] <Sweetshark> dholbach: ;)
[09:45] <Sweetshark> i found a victim on #ubuntu-desktop it seems.
[09:46] <tumbleweed> more victims are also useful for endorsements on that PPU application. We like people to have worked with multiple sponsors
[09:46] <micahg> good, I was worried I was going to have nightmares worrying that it'll still need sponsoring in the morning my time :)
[09:51] <Sweetshark> micahg: the Linux philosophy is "laugh in the face of danger" http://groups.google.com/group/linux.dev.kernel/msg/74ee0b1d6420cd08
[09:59] <hrw> doko: what do you think about http://u.42.pl/2HXy patch? it is needed to get cross compiler built for multilib arm - I am keeping it in gcc-4.6/cross package
=== ttx` is now known as ttx
[10:30] <doko> hrw: why is this necessary?
=== chrisccoulson_ is now known as chrisccoulson
[10:34] <hrw> doko: dpkg-shlibdeps otherwise is not able to find libc6 for armhf
[10:35] <hrw> doko: and vice versa if build for armhf
=== shadeslayer_ is now known as shadeslayer
[12:08] <cortexuvula> Hi all. I have written an Quickly app and started a project on Launchpad. I have updated setup.py with the url of the project. Clicking on Help-> Report a problem produce an error that this is not an ubuntu package
[12:09] <cortexuvula> my question is: How do I link the app and the project
[12:10] <kelemengabor> cyphermox: hi, could you take a look at the merge proposal for bug #875017 ?
[12:10] <ubottu> Launchpad bug 875017 in Ubuntu Translations "Translation for "Wired Connection %s" not displayed" [Medium,Triaged] https://launchpad.net/bugs/875017
=== tsimpson_ is now known as tsimpson
=== MacSlow is now known as MacSlow|lunch
=== om26er__ is now known as om26er
=== yofel_ is now known as yofel
=== MacSlow|lunch is now known as MacSlow
[14:26] <cyphermox> kelemengabor: got my comments?
[14:27] <kelemengabor> cyphermox: yes
[14:28] <kelemengabor> I'll resubmit it, thanks
=== ochosi_ is now known as ochosi
[14:32] <peaceZ> hello all
[14:33] <peaceZ> i wanted to burn de cd of pangolin to make some test but it is 715mb and i could not overburn it using brasero and k3b on a 700mb cd :d
[14:33] <peaceZ> what to do please except than buy 800mb cd ? is it some files i can delete without problems ? as languages files or anything low importance ?
[14:38] <apw> jodh, is this one actually upstart? https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/936256
[14:38] <ubottu> Launchpad bug 936256 in linux (Ubuntu) "kernel panic on boot" [Undecided,Confirmed]
[14:40] <jodh> apw: Just marked it as dup of 935585.
=== tlyu_ is now known as tlyu
[14:48] <cyphermox> peaceZ: no, you pretty much need all the files (or what you could remove isn't likely to give you enough space). as an alternative you could burn the ISO image to a USB key if you have one large enough
[14:49] <apw> bug #935585
[14:49] <ubottu> Launchpad bug 935585 in upstart (Ubuntu) "[kernel panic] init: log.c:786: Assertion failed in log_clear_unflushed: log->remote_closed" [Undecided,Confirmed] https://launchpad.net/bugs/935585
[14:52] <apw> is the unity-video lens nailing the crap out of their disk for anyone else?
[14:53] <seb128> apw, not that I noticed but I'm on a ssd and I don't watch ios closely, you should open a bug if that happens to you in any case ;-)
[14:53] <apw> seb128, it seems to be reading everything in my home directory
[14:54] <seb128> urg
[14:54] <apw> its been humping my disk for 10m now
[14:54] <apw> -o /home/apw/.cache/unity-lens-video/videos.db -l 0 -U /home/apw
[14:54] <apw> it doesn't seem to be limiting itself to my videos directory
=== dendro-afk is now known as dendrobates
[14:55] <seb128> apw, report a bug please, what value has XDG_VIDEOS_DIR in .config/user-dirs.dirs?
[14:56] <apw> $ grep XDG_VIDEOS_DIR ~/.config/user-dirs.dirs
[14:56] <apw> XDG_VIDEOS_DIR="$HOME/"
[14:56] <apw> this is one of those where if you remove empty directories which you don't use the desktop loses its mind, isn't it
[14:56] <seb128> apw, that might be part of the issue
[14:56] <seb128> apw, yes ;-)
[14:57] <apw> that happened with pictures as well, as they were empty i don't have them
[14:57] <seb128> well it fallback to the dir under the one your deleted
[14:57] <seb128> so Videos -> user dir
[14:57] <seb128> it's a bad idea to delete the standard xdg dirs
[14:57] <seb128> not sure how we can message that to users though
[14:57] <peaceZ> cyphermox, my laptop does not enable usb boot, unluckily, thank you for your reply
[14:58] <seb128> one way would be to recreate them for you at every login :p
[14:58] <apw> or perhaps we could just behave sensibly in the face of resonable user behaviour
[14:58] <apw> seb128, if i remake it will the world recover ?
[14:58] <seb128> apw, prompt a dialog asking if you are sure to want to remove those directory with a warning of the side effects?
[14:58] <apw> seb128, and against what do i file a bug, i'd not like to blame unity when its not
[14:59] <seb128> apw, you can just tweak .config/user-dirs.dirs to whatever dir you want the video dir to be
[15:00] <apw> seb128, ok i'll do that, i should file a bug, even if it gets closed 'Don't do that', so that others can find the symptoms
[15:00] <seb128> apw, right, I'm trying to figure what they do, if they just index the video dir it's technically not a bug on their side, it's not especially new
[15:01] <seb128> we keep getting bugs about nautilus getting really slow because people delete "Templates"
[15:01] <apw> the video lens is brand spanking new though isn't it
[15:01] <seb128> and the dir fallback to the user dir
[15:01] <seb128> then nautilus tries to build templates with your whole userdir content
[15:01] <apw> perhaps we should fall back to '.empty' and make one thats empty
[15:01] <seb128> apw, right, the issues due to xdg dirs deletions are not new though
[15:01] <apw> seb128, oh i probabally have that too
[15:02] <seb128> apw, ideally we would lock down the dirs on the fs and don't let users delete them or something
[15:02] <seb128> or at least display a warning dialog when somebody does saying they are asking for issues
[15:02] <apw> its a bit of a strange paradimn though, as it makes sense to keep everything say to do with ones 'wedding' in one place
[15:03] <seb128> apw, well, we could fallback to an empty dir yes, but then you would get i.e no video indexed and not understand why
[15:03] <apw> so forceing me to put my videos in one place and pictures in another for an event ... is going to hurt heads
[15:03] <seb128> apw, until we get something that doesn't suck to replace inotify we can't do much better
[15:04] <apw> what is inotify missing which prevents this from working ?
[15:04] <seb128> apw, you can't watch your whole user dir
[15:04] <seb128> it's limited to 32k watches or something
[15:04] <seb128> so basically you have to watch the main dir and go through all the subdir every time something change
[15:04] <seb128> you can't tell inotify "just let me know when something in the user dir change"
[15:05] <seb128> or rather "let me know what changed"
[15:08] <seb128> apw, well in any case yes please open a bug on "unity-lens-video", I will comment on it
[15:08] <apw> seb128, i can see why, that would be vastly expensive for the kernel to do
[15:08] <seb128> apw, I think it should be easy to just not index if the Video dir is the user dir
[15:08] <apw> it would have to know that someone was watching somewhere higher in the hierachy
[15:08] <seb128> apw, well the bottom line is that there is no way to monitor files through an user dir which is not io expensive
[15:09] <apw> whcih would be a lock heavy and extremly time consuming operation on every single io
[15:09] <apw> seb128, how about we have a Media directory instead
[15:09] <seb128> apw, well we have music, pictures and videos
[15:09] <seb128> they could be subdir of a media though
[15:10] <seb128> apw, but yeah, those could do with some design thinking
[15:10] <apw> i wonder what happens if i point them all to Media/ which is what i mean really
[15:10] <seb128> thing is that you don't want shotwell to index your music for example
[15:10] <apw> so i can have Media/wedding/*
[15:10] <seb128> or rhythmbox to index your photos
[15:10] <apw> i would hope shotwell would know that a tarball isn't something it can index
[15:10] <apw> and same with rhythmbox
[15:11] <seb128> I guess design will said that the solution would be to have one application doing music, video and photos
[15:11] <seb128> apw, right, they still watch their library though
[15:11] <apw> yeah probabally
[15:11] <seb128> the more you watch, the more you wake up looking at stuff to see that you don't care about those
[15:11] <seb128> anyway
[15:12] <seb128> going back to the issue
[15:12] <seb128> please open a bug on unity-lens-video saying that it gets confused when the video xdg got deleted
[15:12] <seb128> and please subscribe me to the bug ;-)
[15:12] <seb128> I doubt we will address the different directories for different medias this cycle
[15:12] <seb128> but we can probably handle the cases where some xdg dir needed got deleted
[15:13] <apw> no probabally not as i am not sure design agree my concern is valid anyhow
[15:13] <apw> in their model i mean
[15:13] <seb128> would it be by displaying a dialog at login telling about the issue and the side effect it has an asking you if you want to pick another dir for $MEDIA
[15:13] * apw will file something, and go remake these directories
[15:13] <seb128> login or whenever it detects that the directory is being deleted
[15:16] <apw> does anyone know what the offical matrix key binding is? as i have it appearing at random
[15:19] <seb128> apw, alt
[15:20] <apw> seb128, doing what with alt? single press?
[15:20] <seb128> apw, that's a known bug, should be fixed this week, it appear if you alt-tab sometimes or if you alt-screenshot
[15:20] <seb128> apw, yes, a simple tap, i.e just press and relax without using other keys
[15:20] <seb128> apw, though it's buggy and i.e alt-printscreen or sometimes alt-tab done quickly will trigger it
[15:21] <seb128> bug #925293
[15:21] <ubottu> Launchpad bug 925293 in Unity Distro Priority "Plugins can't tell the difference between a modifier key-tap, and a modifier key-release (after being used to modify other keys)" [High,Fix committed] https://launchpad.net/bugs/925293
[15:21] <seb128> apw, ^
[15:21] <apw> ahh ok, thats odd as i didn't think it was single-tap from playing with it, but double, strange
[15:22] <apw> seb128, are cursors mean to work once you have typed a term ?
[15:23] <seb128> apw, cursors?
[15:23] <apw> seb128, i am looking for something to select anything other than the first entry in the suggestions list
[15:23] <seb128> oh, you mean in the hud? down arrow should work
[15:23] <seb128> it works for me
[15:23] <apw> it seems to highlight the first one, nothing seems to work here
[15:24] <seb128> keypad arrows don't work though
[15:24] <seb128> if that's what you try
[15:24] <seb128> but the normal "down" arrow should go down
[15:25] <apw> seb128, what keysym is it bound to ?
[15:25] <apw> doesn't work with any of the down arrows on any of my keyboards
[15:26] <seb128> dunno, it works for me with
[15:26] <seb128> state 0x10, keycode 116 (keysym 0xff54, Down), same_screen YES,
[15:26] <seb128> i.e "Down"
[15:26] <apw> seb128, ok mine is producing Down, and no dice
[15:26] <seb128> you might want to ask on #ubuntu-unity about it
[15:27] <apw> i'll file a bug
[15:27] <apw> any idea what one files matrix things in ?
[15:28] <seb128> apw, https://code.launchpad.net/~gordallott/unity/hud/+merge/90085
[15:28] <seb128> that was the feature merge
[15:28] <seb128> it might have code clues
=== Torpedo_Skyline is now known as TorpedoSkyline
[15:29] <seb128> gord, ^
[15:29] <seb128> apw, gord might be able to help you, he did most of the unity side of the hud
[15:29] <apw> seb128, Videos thingy filed and you sub'd
[15:29] <seb128> gord, is there any known issue where down doesn't work to navigate through the hud suggestions?
[15:29] <seb128> apw, thanks
[15:30] <apw> seb128, interesting the keys which produce Left and Right work to move in the entry box ok
[15:31] <gord> seb128, saw the same issue with the dash earlier, not looked at it yet though
[15:31] <seb128> unity bug then I guess...
[15:32] <apw> gord, xev says my up and down arrows produce Up and Down as keys, so i am a bit confused
[15:34] <apw> seb128, i've got those numbers stuck on again, having fiddled with hud for a bit, perplexing
[15:35] <seb128> apw, what numbers?
[15:35] <apw> seb128, the black numbers in the launcher, when you hold for a while on Super
[15:35] <apw> the hints for the Super+<n> functionaolity
[15:35] <apw> have one of my launchers with them stuck on
[15:36] <seb128> oh, weird
[15:36] <apw> seb128, so is the hud bug to be filed against unity?
[15:36] <seb128> apw, yes
[15:36] <seb128> gord, should that down arrow bug assigned to you?
[15:37] <apw> how odd, when yo uhave the launchers on the screen a
[15:37] <apw> and hit alt, all your launchers go away even though only one needs to
=== dendrobates is now known as dendro-afk
[15:37] <apw> is that intended
[15:38] <gord> seb128, i guess, I don't really know enough about it to say where the issue is really, i'll farm it off wherever once i know more so can't hurt
[15:38] <seb128> gord, ok, thanks
[15:38] <seb128> apw, not sure, doesn't hurt to open a bug and get design to look at it
=== dendro-afk is now known as dendrobates
[15:46] <apw> https://bugs.launchpad.net/ubuntu/+source/unity/+bug/937025
[15:46] <ubottu> Launchpad bug 937025 in unity (Ubuntu) "HUD: cursor up/down keys are not working with hud searches" [Undecided,New]
[15:46] <apw> seb128, ^^ for the HUD cursor issue
[15:46] <seb128> apw, thanks
[15:54] <apw> gord, ok i just logged out and back in to fix my launcher icons, and i seem to have hud up/down working now
[15:54] <apw> so it may be random whether they work or not
[15:56] <apw> seb128, ^^
[15:56] <seb128> seems so, gord said he got it once as well
[16:10] <m4n1sh> ev: ping
[16:10] <ev> m4n1sh: pong
[16:11] <m4n1sh> ev: do you have time to refactor the whoopsie code
[16:11] <ev> m4n1sh: yup, been working on it all day
[16:11] <m4n1sh> and split it out of ccpanel_foo_init method
[16:11] <m4n1sh> cool
[16:11] <m4n1sh> I cant seem to get my way around it
[16:11] <m4n1sh> i am bad at C
[16:11] <ev> no worries
[16:11] <ev> I'm on top of it
[16:11] <m4n1sh> yeah!!
[16:11] <m4n1sh> :)
[16:11] <ev> I'll propose a branch hopefully by the end of the day
[16:11] <m4n1sh> releasing on wed
[16:11] <m4n1sh> yes, All my worries gone
[16:11] <m4n1sh> thanks
[16:13] <lamont> slangasek: nm memory - I was waiting to reproduce it since I had upgraded a bunch of things
[16:22] <noodles775> Hi. Are there known issues with precise upgrades at the moment (nvidia perhaps?). I've just updated an hour ago (hadn't done so for a while), and after reboot can only login to a recovery console.
[16:23] * noodles775 takes a photo.
[16:26] <noodles775> http://people.canonical.com/~michaeln/tmp/IMAG1140.jpg
[16:26] * noodles775 checks if nvidia issue from the other week was resolved.
[16:43] <noodles775> Hrm, checking dpkg.log seems to show that most of y system has been removed (I blindly did `sudo apt-get dist-upgrade` and hit Y without checking the packages) :/
[16:44] <damg> noodles775, I've installed today's daily today and had this issue after login for 30 seconds, then X logged me into unity
[16:44] <damg> after that I've installed proprietary though and didn't experience the problem anymore
[16:47] <noodles775> damg: yeah, as above, I've just checked my dpkg.log and seen that it's unrelated to a driver issue, but rather that many packages were removed when I just updated (including essential stuff like software-center, gnome-panel, unity-2d etc.)
[16:47] <damg> hehe
[16:48] <noodles775> `cat /var/log/dpkg.log | grep 2012-02-20.*remove | wc -l` -> 52
[16:52] <damg> how does unity decide whether an application window will be managed by it. it seems that it ignores blender in alt-tab or on its icon clicks.
[16:53] <damg> hm, it doesn't even get focus. pressing alt calls the menu of the next app below it.
[17:06] <bulldog98> cjwatson: hello we the kubuntu team is working on an active image (aka mobile) could you please build following seed? lp:~kubuntu-active/ubuntu-seeds/kubuntu-active-seed/
[17:06] <bulldog98> thanks
[17:16] <slangasek> lamont: looks like the nm memory bug is well understood now
=== JackyAlcine_ is now known as JackyAlcine
[18:25] <Riddell> bulldog98: it's not quite as easy as just asking cjwatson to build a seed :) I'm looking at script to be fiddled now
[18:26] <bulldog98> Riddell: ok ping me if finished, I’ll test it then
[18:27] <slangasek> hallyn: the AT_EMPTY_PATH issue now also affects qemu-linaro, apparently
[18:27] <Riddell> bulldog98: it's probably not today, needs bits put on servers that I don't have access to
[18:28] <bulldog98> Riddell: ah ok
[18:40] <lamont> slangasek: cool
=== Quintasan_ is now known as Quintasan
=== allison_ is now known as wendar
[19:47] <int_ua> I want to report a bug against renice but I can't find it's upstream bug tracker.
[19:51] <int_ua> Can anyone help me?
[19:53] <int_ua> Since it's bsdutils I don't think that nor Ubuntu neither Debian bug trackers are the best choice
[19:53] <int_ua> but since it's fundamental part
[19:53] <int_ua> Anyone?
[21:03] <pitti> mdz: hey Matt, are you there for TB meeting?
=== dendrobates is now known as dendro-afk
=== Guest65979 is now known as albrigha
[21:44] <jodh> Hi - is there anyone about with build-rescore-superpowers that can bump
[21:44] <jodh> https://launchpad.net/~jamesodhunt/+archive/bug-935585/+packages ?
[21:45] * jodh double-checks to see if emacs has a build-rescore function...
[21:46] <stgraber> jodh: done
[21:47] <jodh> stgraber: thanks very much! emacs drew a blank. Wonder if it's got
[21:47] <jodh> a make-me-a-build-rescore-function function?...