/srv/irclogs.ubuntu.com/2010/07/07/#ubuntu-devel.txt

lifelessjames_w: are you still up?00:40
highvoltage#ubuntu-devel is quite different when you're in a US timezone00:46
=== bjf is now known as bjf[afk]
=== Chex changed the topic of #ubuntu-devel to: Alpha-2 released | Development of Ubuntu (not support, not app development) | #ubuntu for support and general discussion for dapper-lucid | #ubuntu-app-devel for application development on Ubuntu | http://wiki.ubuntu.com/UbuntuDevelopment See #ubuntu-bugs for http://wiki.ubuntu.com/HelpingWithBugs
=== Chex changed the topic of #ubuntu-devel to: Alpha-2 released | Development of Ubuntu (not support, not app development) | #ubuntu for support and general discussion for dapper-lucid | #ubuntu-app-devel for application development on Ubuntu | http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://wiki.ubuntu.com/HelpingWithBugs
keesis initramfs-tools stuck in binary NEW?01:43
keeshm, no...01:46
keesis the publisher just being slow?01:46
ScottKMight have been affected by the LP downtime.01:49
=== rootusr is now known as tutux
=== tutux is now known as tutuxg
=== jjohansen is now known as jjohansen-afk
=== bjf[afk] is now known as bjf
=== j1mc is now known as j1mc|away
=== bjf is now known as bjf[afk]
=== bjf[afk] is now known as bjf
=== rootusr is now known as tutuxg
=== un214_ is now known as un214
=== v is now known as vorian
pittiachiang: no, debian.tar.gz is created automatically, no need to fuss with that; you just need to apply the patch (quilt push -a) before building binaries07:02
ricotzpitti, good morning07:09
=== freeflyi1g is now known as freeflying
ronchi devs08:03
roncI just wanted to say that some man pages are not being updated since the early days of ubuntu08:04
dholbachgood morning08:04
roncI was trying to browse in w3m the other day, but couldn't find the items that these websites told me... then wham! I found this bug report: https://bugs.launchpad.net/w3m/+bug/52536008:06
ubottuLaunchpad bug 525360 in w3m (Ubuntu) "Man page isn't synchronized" [Undecided,Confirmed]08:06
roncwhich is like crazy, man08:06
pitticking: good morning08:20
pitticking: wrt. bug 599352, does linux' suspend-to-disk use compression at all?08:21
ubottuLaunchpad bug 599352 in upower (Ubuntu) "[Maverick] Need to not display hibernation in UI if do not have enough swap space" [Low,Confirmed] https://launchpad.net/bugs/59935208:21
ckingpitti, as far as I know, there is no compression used in hibernation08:24
ckingit just dumps the used pages + some meta data08:25
pitticking: thanks08:25
=== fddfoo is now known as fdd
=== almaisan-away is now known as al-maisan
=== DrPepperKid is now known as MacSlow
evAm I going insane or should md5sum /dev/sdb /dev/sdb produce identical results (all partitions are unmounted)?10:26
sorenev: Sounds like a reasonable assumption.10:30
evyeah, I've got three usb disks that continuously produce different results.  I guess it must be a bad lot.10:31
evsoren: thanks10:31
pittiev: do you see read errors in dmesg?10:31
sorenev: Have you had anything from sdb mounted since your last reboot? Perhaps there's some cache flushing going on that's massing things up.10:31
evpitti: nope10:32
evhmm10:32
=== amitk is now known as amitk-afk
ricotzpitti, could you please accept the docky sru upload? or is there something wrong? https://bugs.edge.launchpad.net/docky/+bug/57904910:56
ubottuLaunchpad bug 579049 in docky (Ubuntu Lucid) "Dragging files to folders causes data "corruption"" [Undecided,New]10:56
pittiricotz: I'm not doing SRU every day, just once a week or so (I'm not in platform this cycle)10:57
cjwatsonI can have a look10:57
pittiricotz: but I'll get to it this week, promised10:57
ricotzpitti, sorry, ok10:57
ricotzcjwatson, that would be great10:57
cjwatsonI've been doing them intermittently, just haven't had time to completely keep up10:57
cjwatsonoh, right, it's the one with an infinite number of bugs10:58
* cjwatson watches queuediff DoS his firefox10:58
ricotzcjwatson, yes, there are quite some fixes10:59
cjwatsonricotz: is anyone working on getting this into maverick?  in general, fixes should go to maverick first10:59
ricotzcjwatson, yes, i requested an sync with debian/unstable11:00
* cjwatson looks at the sync queue11:00
ricotzthe packaging in debian/unstable isnt compatible with lucid anymore11:00
cjwatsonok, I'll process that11:01
Laneywould be nice if someone could accept lhs2tex too11:01
ricotzLaney, you are always behind my back :P11:02
LaneyI have my own SRUs in addition to yours!11:02
ricotzLaney, i know11:03
Laney:)11:03
BlackZcjwatson: could you look at bug #600121 when you can ? we're waiting this sync for the ekiga merge, thanks!11:04
ubottuLaunchpad bug 600121 in opal (Ubuntu) "Sync opal 3.6.8~dfsg-2 (universe) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/60012111:04
cjwatsonoh dear, I should go back to hiding, now everyone wants a piece11:06
BlackZheh11:07
Laneythe joys of archive adminship11:07
Laneydid something happen wrt sparc very recently? I see that glib2.0 has no build record there11:08
cjwatsonnot that I know of ...11:08
Laneyoh, no it hasn't in the previous uploads either11:10
=== smb is now known as smb-afk
cjwatsonISTR lamont killed glib2.0 on sparc because it had a habit of killing the buildd11:10
Laneyok then, I won't worry about knock-on failures11:11
ricotzdo built packages need to be accepted manually like https://edge.launchpad.net/ubuntu/+source/qt4-x11/4:4.7.0~beta1+git20100706-0ubuntu1 ?11:12
Laneynew binaries or source packages do11:13
cjwatsonricotz: docky done (maverick and lucid-proposed)11:13
ricotzcjwatson, thank you!!!11:13
cjwatsonBlackZ: opal done11:14
BlackZcjwatson: thanks11:14
cjwatsonricotz: qt> I'd like to accept all architectures at once if possible, to minimise the chance of build failures due to them being out of sync11:18
ricotzcjwatson, ok, so this will take some time then, currently it breaks my ia32lib build in xorgedgers11:21
cjwatsonLaney: lhs2tex done11:23
cjwatsonricotz: hopefully not too long - I've scored it up on the other architectures11:23
cjwatsonthey should start very shortly11:23
ricotzcjwatson, ok11:25
Laneythanks11:25
Laneythe FP lab will appreciate it11:25
cjwatsonFP lab?11:25
Laneymy research group11:26
cjwatsonah, right11:27
mdzmvo: is there a way to test a lucid->maverick upgrade of a real system in a sandbox?11:58
mdzI remember you'd worked on some tools to do that in the past11:59
sorenmdz: update-manager has a -s option: "-s, --sandbox         Test upgrade with a sandbox aufs overlay"11:59
mdzsoren: do you know if it works in lucid?12:00
dholbachor just check out http://people.canonical.com/~mvo/automatic-upgrade-testing/2010-07-06-15:03:01/ ;-)12:01
sorenmdz: Nope, never used it. The option is available on Lucid, though.12:01
* mdz notices --sandbox isn't documented on the man page, and prepares a branch12:01
mdzsoren, mvo: it seems to have hosed my system pretty badly ("command not found: ls" and such)12:12
mdzseems to be getting EPERM on everything12:13
mvomdz: a reboot should cure it12:19
mdzmvo: it has12:19
mvomdz: all changes are in /tmp, but it can take a bit until tmp is cleared12:19
mdzmvo: my /tmp is tmpfs12:20
mvomdz: hrm, hrm, sorry for the trouble, the code should check that12:20
mdzwhich is why the upgrade failed early (lack of disk space in /tmp)12:20
mdzis it possible to set an alternative location for the sandbox?12:20
mvomdz: it uses tempfile.mkdtemp(), so it should honor TMPDIR in the environment12:22
mdzmvo: ok, thanks12:23
mvomdz let me know how it goes12:23
mdzmvo: having a non-functional $PATH will make it difficult to monitor the upgrade, which is a pain because I'm trying to reproduce a kernel bug12:23
mdzmvo: (the disk free space was correctly checked, that worked fine. it's just that my system was still non-functional after the upgrade was canceled)12:24
mvoa kernel bug that happens on upgrade12:24
mvo?12:24
mdzmvo: yes12:25
mvomdz: the --sandbox option did not got a lot of love because aufs is not really suitable for the workload. this is why its a bit bumpy12:25
mdzbug 60226112:25
ubottuLaunchpad bug 602261 in linux (Ubuntu) "Thrashing and OOM during upgrade from 10.04 to Maverick" [High,Confirmed] https://launchpad.net/bugs/60226112:25
mdzrickspencer3 saw this as well12:26
* mvo looks12:26
mdzI thought it was related to apparmor, but apw suspects it may be ext412:26
mvomdz: is it only reproducable on real HW?12:26
mvootherwise the auto-upgrade-tester may be helpful12:26
mdzmvo: I don't know12:27
apwcertainly there are reports of jdb2 being stuck running for over 2 minutes which is abnormal12:27
mvomdz: http://people.canonical.com/~mvo/automatic-upgrade-testing/current/ shows that a server upgrade (minimal server) is about 10min, ubuntu 30min, so total runtime seems to be not a good indicator12:28
mdzmvo: I suspect it may be related to any/all of a) running processes (i.e. a normal user session), b) replacing apparmor profiles, or c) the idle I/O scheduler (ionice -c3)12:29
mdzmaybe even a combination of them12:30
* mvo nods12:30
mvoI can create a auto-login profile so that there is a normal session12:31
mdzapw: interestingly, doing a big dd on this system doesn't bog it down like it does my laptop12:33
mdzthough both are ext412:33
mdzthis one has a faster disk and less memory12:33
mdzand is 32-bit12:33
=== amitk-afk is now known as amitk
apwmdz i wonder why that would be, less memory?12:33
mdz(while the other is 64)12:33
apw32bit does have an effect on the kernel memory size12:34
mdzapw: I get very high CPU utilization (system) and very low I/O wait12:34
mdzflush-8:16 is the active thread12:34
mdzon the other system I believe I get high I/O wait12:35
apwhrm ... i use 64 bit quite a bit and have done an upgrade there without issue (lucid -> maverick)12:37
=== al-maisan is now known as almaisan-away
=== smb-afk is now known as smb
=== MacSlow is now known as MacSlow|lunch
psurbhii wanted to know where I could pick the latest unreleased (or in development) mdadm source package for maverick. Is there a bzr repository for mdadm? where would that be?13:06
Keybukpsurbhi: we don't really have unreleased/in development repos like that13:07
Keybukthe latest version is generally just in the archive13:07
Keybukif there's a bzr branch, it's possibly lp:ubuntu/mdadm13:08
psurbhiok, i checked that.. no bzr branch then. So does it mean that apt-get source mdadm is the latest mdadm package?13:08
amitkpsurbhi: 'rmadison mdadm'13:09
psurbhiamitk, Keybuk, thanks13:10
Keybukpsurbhi: yup13:10
psurbhiok13:10
=== cking is now known as cking-afk
=== yofel_ is now known as yofel
=== almaisan-away is now known as al-maisan
=== MacSlow|lunch is now known as MacSlow
LucidFoxI think for the next year's April Fools, I'm going to blog about Ubuntu Genuine Advantage.14:27
LucidFoxWith fake activation screens and everything.14:27
mdzmvo_: /var/log/dist-upgrade/term.log suddenly stopped (in the middle of a line), but the upgrade is continuing to run14:30
mdzplenty of disk space, no obvious problem14:30
mvo_oh14:31
mvo_odd, anything from strace?14:31
mdzmvo_: it's as if a buffer isn't getting flushed somewhere14:32
mvo_that is insndbox mode? or nomal?14:32
sorenLucidFox: http://www.linuxgenuineadvantage.org/14:32
mdzmvo_: normal14:32
mdzmvo_: strace of the maverick process shows write(11, "pmstatus:ure:70.6563:Installed u"..., 3514:32
mvo_mdz: ok, that looks normal14:33
LucidFox"According to an independent study conducted by some scientists, many users of Linux are running non-Genuine versions of their operating system. This puts them at the disadvantage of having their computers work normally, without periodically phoning home unannounced to see if it's OK for their computer to continue functioning."14:33
LucidFoxYeah, that's WGA in a nutshell14:33
mvo_mdz: do you have apt-term as well? and if so, is that keep getting updated?14:33
mdzmvo_: but there is no activity in strace even though dpkg is completing more operations14:33
mdzmvo_: I have a tail -f running on /var/log/dist-upgrade/*.log and nothing is changing14:33
mdzmvo_: the terminal window is not updating either14:33
mdz(the one inside the upgrader)14:34
sorenLucidFox: LGA doesn't work with upstart, though.14:34
LucidFoxHmmmm *scribble scribble*14:34
LucidFoxsoren> You know...14:34
LucidFoxNow I'm actually tempted to write an actual GUI system for Ubuntu mimicking WGA. Deliberately easily circumventable, of course14:35
sorenLucidFox: sorry :)14:35
mdzmvo_: hmm, it just dumped about 2k of text into term.log and apt-term.log, all at once, and then stopped again14:35
LucidFoxSorry? :)14:35
mdzdefinitely some buffering issue14:35
sorenLucidFox: Yeah, didn't mean to get you motivated on.. um.. slightly less than useful things :)14:35
mdzmvo_: I've emailed you the logs14:36
LucidFoxHah14:36
mvo_mdz: thanks, I'm working on the code currently (to port to the latest python-apt 0.8 api), that is going to be interessting :)14:36
mvo_mdz: it looks like there are some issues in p-apt itself, nothing too scary so far fortunately, all easily work-aroundable14:37
mdzmvo_: there seem to be two upgrader processes, and the child is blocking writing to the pipe to the parent14:40
mdzmvo_: two things look weird to me: 1) the parent still has both ends of the pipe open, and 2) the parent is polling but it doesn't seem to be polling this FD14:41
mvo_mdz: thanks, got the logs.14:41
mvo_mdz: which frontend did you use? text? gtk? I will try to reproduce in a vm14:42
=== cking-afk is now known as cking
mdzmvo_: GTK14:42
=== rgreening_ is now known as rgreening
=== rgreening is now known as rgreening_
=== rgreening_ is now known as rgreening
tseliotcjwatson: do you have news on bug #258486?15:12
ubottuError: Could not parse data returned by Launchpad: The read operation timed out (https://launchpad.net/bugs/258486)15:12
cjwatsontseliot: no, sorry - please check with ev though15:13
cjwatsonI've not been doing much on ubiquity this cycle15:13
cjwatsonand he's reworking jockey handling anyway, iirc15:14
tseliotcjwatson: yes, that's why I asked but I was looking for "evand" not "ev". Thanks15:14
kirklandpitti: howdy, around?15:19
pittikirkland: hey, how are you? I'm in the OEM meeting, but otherwise here, yes15:19
kirklandpitti: awesome, i'm okay ...  i just realized a libvirt SRU that has slipped through the cracks for me for ~1 month: https://bugs.edge.launchpad.net/ubuntu/+source/multipath-tools/+bug/57109315:20
ubottuLaunchpad bug 571093 in libvirt (Ubuntu Lucid) "[SRU] multipath + libvirtd eats away more memory over time" [Medium,In progress]15:20
kirklandpitti: it needed a minor version change in the upload;  i just re-uploaded that to lucid-proposed15:20
kirklandpitti: i just wanted to make sure the other end didn't slip through the cracks too :-)15:21
pittikirkland: ah, thanks; I'll have a look at it15:21
kirklandpitti: cheers, mate15:22
evtseliot: you're seeing something like this on Maverick?15:23
evIntrepid was a long time ago :)15:24
tseliotev: I haven't tried with Maverick yet. The problem can be reproduced in Lucid though15:25
tseliotand this makes it more recent ;)15:25
evinteresting15:25
tseliotev: basically xorg.conf remains there but the packages don't15:26
tseliotbut I guess that's already in the report15:26
evindeed, I'm quite surprised we haven't heard more people screaming about this15:27
evI've bumped the priority and assigned it to myself15:27
tseliotev: thanks. Let me know if you need any help for debugging, etc.15:29
evsure thing15:30
mdzmvo_: I sent you a merge request with a  couple of small update-manager fixes I did while looking at this problem15:36
mvo_thanks mdz15:38
mvo_merged15:39
mdzmvo_: one is a documentation fix, the other is a one-line code change (untested but trivial)15:39
apwcjwatson, the grub boot sexification, are we keeping X on VT7 ?15:47
=== sconklin-gone is now known as sconklin
bankixHello.15:50
bankixI'm currently customizing Lucid, but however I'm unable to find out which file is responsible for the little ubuntu logo in the upper left corner of the gnome panel. Could you give me a pointer?15:51
pittiwgrant: hey, how are you?15:57
pittilamont, wgrant: do you happen to know the current state of native soyuz ddeb support?15:57
=== thekorn_ is now known as thekorn
lamontpitti: nfc15:59
a_okI still seem to have the problem that looks a lot like https://bugs.launchpad.net/ubuntu/+source/initramfs-tools/+bug/546964 . I am using an Intel ET Nic and I have confirmed that the igb.ko driver is in the initrd image15:59
ubottuLaunchpad bug 546964 in initramfs-tools (Ubuntu) "10.04 beta1 iSCSI-on-root fails on boot" [High,Fix released]15:59
cjwatsonapw: yes16:01
a_okI am using the stable 10.4 release by the way and the latest initramfs-tools seems to be the one that comes with it (0.92ubuntu78)16:01
cjwatson(to the best of my knowledge)16:01
ograapw, sexification ? will it be male or female after that ?16:02
apwcjwatson, the kernel patches as were break VTs basically for good.  we really want the special functionality to remain only till we are on VT7 and settled in, i am thinking a VT switch may be the correct trigger to return normal behviour16:03
apwthen if you fiddle manually mid-boot you get the normal behaviour too16:03
a_okthis is the exact error I get when trying to boot http://pastebin.com/3UXhaPKu16:03
cjwatsonapw: breaking VT switching definitely isn't on16:04
cjwatsonapw: remember that server boots won't have a VT switch but will have plymouth and the grub splash, though16:04
cjwatsonapw: so I'm not sure that's the right trigger16:04
apwcjwatson, hrm, then i suspect i will need to be told you are clear16:05
cjwatsonapw: how about plymouth's switch to KD_TEXT?16:05
cjwatsonthat seems like a good indicator16:05
apwcjwatson, i presume it only does that on non-desktop16:05
apwoh perhaps also on timeout ?16:05
cjwatsonwe always switch back to KD_TEXT at some point, even if it's only when people do ctrl-alt-f1 fromX16:05
a_okis there a way I can get some more information on what goes wrong?16:06
apwcjwatson, well no you don't change mode then you change which vt is displayed which is in a different mode16:06
apwcjwatson, anyhow i am getting ahead.  i suspect a switch from graphics mode to text within a VT or switching VT should trigger normality16:07
apwi thin16:07
apwi think16:07
cjwatsonthat seems fair16:07
cjwatsonapw: vt1 is switched to KD_TEXT even on desktop boots, I believe16:07
cjwatsonbloody well ought to be anyway16:07
apwonce i figure out why vt2 ->6 are all full of vomit of course i can smarten things up16:07
apwcjwatson, so can you tell me when we switch to vt7 and which thing plymouth is on, i thought it was on vt716:08
cjwatsonplymouth is on vt116:09
cjwatsonwe switch to vt7 when X starts16:10
apwcjwatson, so if we vt switch to the same contents that is scemeless ?16:10
apwvisually ?16:10
apwie. plymouth maks vt1 purple, x starts on vt7 and copies the contents over, then vt switches ?16:11
cjwatsonX is *supposed* to take care of that; it didn't seem to quite work in my video but I am deliberately not caring about that16:11
apwcjwatson, i suspect that as you said you don't have KMS therefore you cannot do a seemless cut16:12
apwi will get this heap installed on some kms h/w and see what happens there16:12
pittikirkland: do you know the status of bug 566792 in maverick: the SRU is ready for -updates, but I won't as long as it's not yet fixed in maverick16:12
YokoZarpitti: you probably don't need to copy the SRUed ia32-libs to maverick, as it needs to be freshly generated there to get the maverick versions of the libs16:12
ubottuLaunchpad bug 566792 in eucalyptus (Ubuntu) "metadata service returns empty data with 200 OK" [High,Confirmed] https://launchpad.net/bugs/56679216:12
pittikirkland: (or at least committed to bzr)16:12
apwcjwatson, what do you use to take the video of the VM ?16:12
cjwatsonapw: gtk-recordmydesktop16:12
pittiYokoZar: right, but as long as nobody does that, copying the SRU is better than nothing16:12
cjwatsonapw: by way of scoping I'm intentionally only caring about the period until plymouth comes up16:13
kirklandDaviey: you've been managing that one ... can you tell us if that code has been applied to maverick?16:13
YokoZarpitti: I suppose that somebody is supposed to be me.  Is there any movement on multiarch at all in maverick (and thus things that can be dropped from ia32-libs?)16:13
* Daviey reads16:13
kirklandDaviey: i suspect it has, in the course of your regular weekly Eucalyptus merges16:13
apwcjwatson, yeah, but sadly i have to care thorugh KMS to make sure i am not breaking that by munching the kernel16:13
kirklandDaviey: can you check?16:13
cjwatsonapw: ah, fair enough16:13
pittiYokoZar: no, I didn't mean "you" in particular16:14
YokoZarpitti: I know that, but I usually do the ia32-libs uploads ;)16:14
pittiYokoZar: I don't know about the multiarch plan, I'm afraid16:14
Davieypitti, Ah.. yes.. -> PM16:14
pittiYokoZar: it's much faster to do in the DC, though16:14
YokoZarslangasek might know about multiarch16:16
slangasekstep 1) fix dpkg16:17
pittiis it just me, or is LP very slow today? (including launchpadlib)16:29
=== Amto_res_ is now known as Amto`OFF
apwpitti, it was upgraded to a nice shiney version last night i think16:30
sebnerapw: shiny = b0rken? :P16:30
pittithey forgot to take out the sleep() debugging statements?16:30
sebnerpitti: agreed, slowwww16:30
apwpitti, it does feel slow, but i am not sure if its more excruciatingly slow than normal16:31
apwi already want to poke my own eyes out16:31
YokoZardoes anyone know if the new SFTP for PPAs can be done for normal uploads?16:31
YokoZarto the main repository16:31
Laneyyes16:31
Laneyyou have to set incoming to ubuntu, or so geser told me16:32
pittiapw: it's even slow with the CLI tools on cocoplum, so I guess it's real16:32
YokoZarLaney: and presumably modify /etc/dput.cf manually16:32
Laneyright16:33
apwpitti, i am usng edge of course, so my milage is different16:33
pittiapw: same here16:33
pittithere, all SRU queues emtpy (except for some disputed uploads)16:33
geserYokoZar: see http://paste.ubuntu.com/460266/ for my ~/.dput.cf for SFTP uploads16:33
apwpitti, heh the kernel per16:34
apwperhaps ?16:34
pittithat (karmic), and sane16:34
YokoZargeser: thanks16:34
apwyep, thats the one16:34
YokoZargeser: (I need sftp to upload ia32-libs from home otherwise my internet connection hangs forever)16:34
arulalanDear All, i have a doubt. How can i write graphics program in GCC. I googled , but cant get good link for this16:38
arulalani cant post this query in #gcc. it says ' Cannot send to channel'16:39
=== amitk is now known as amitk-afk
=== MacSlow is now known as MacSlow|afk
Chipzzarulalan: this is not a good question for this channel; but either way, you don't specify what kind of graphics program you want to write (ie plain OpenGL, Gtk+, Qt...)16:52
Chipzzarulalan: and if you can't send to the channel, there's probably a very good reason for it (like the channel being moderated); read the topic, it most likely has further instructions (and this goes for any channel, including this one; if you would have read the topic, you would have known your question wasn't appropriate here)16:54
arulalanChipzz: Ok sir. i need to write a program using graphics.h header file like turboc. is opengl right choice ?16:56
Chipzzarulalan: again, your question is not appropriate here. also, a quick google for graphics.h shows a lot of different projects (quite expectable with such a filename) so your question is ambigous17:01
Chipzzbut again, this is not the right place to ask.17:01
apwcjwatson, http://people.canonical.com/~apw/misc/fbcon-handoff.ogv <-- video from my latest test kernel17:02
Chipzzbut if graphics.h refers to sth like VGA/SVGA drawing, the closest thing on linux is SDL I would say17:02
apwarulalan, normally you can't send to a channel if you arn't registered as owning that nick17:04
cjwatsonapw: not bad17:16
apwcjwatson, note the cursor only appears when the debugging printk's come out, and they are at KERN_EMERG17:16
cjwatsonincidentally, with this, does the console subsystem start in KD_GRAPHICS or KD_TEXT?17:19
dupondjeRiddell: I did build imagemagick from debian, and those are the depds:  Depends: libbz2-1.0, libc6 (>= 2.11), libfontconfig1 (>= 2.8.0), libfreetype6 (>= 2.2.1), libglib2.0-0 (>= 2.12.0), libgomp1 (>= 4.2.1), libice6 (>= 1:1.0.0), libjasper1 (>= 1.900.1), libjpeg62, liblcms1 (>= 1.15-1), liblqr-1-0 (>= 0.1.0), libltdl7 (>= 2.2.6b), libpng12-0 (>= 1.2.13-4), libsm6, libtiff4, libx11-6, libxext6, libxml2 (>= 2.7.4), libxt6, zlib1g (>= 1:1.2.3.3.dfsg)17:26
Riddelldupondje: no librsvg?17:26
dupondjeRiddell: nope17:26
dupondje Package: libmagickcore317:27
Riddelldupondje: something must depend on it though surely?17:27
dupondjeRiddell: libmagickcore3-extra and libmagickcore-dev depend on librsvg2-dev,17:35
dupondjelibmagickcore3-extra on librsvg2-2, libmagickcore-dev depend on librsvg2-dev17:35
Riddelldupondje: that seems fine then17:35
dupondjetrue :) can you give it a sync ? :)17:36
Riddelldupondje: what about liblqr-1-0-dev ?17:38
dupondjeRiddell: it was dropped because it wasn't in main, but now it is17:40
Riddelldupondje: groovy17:41
Riddelldupondje: synced17:41
dupondjesweet :)17:44
LucidFoxAny chance I could get the Liferea messaging indicator patch reviewed before the heat death of the universe? It's been waiting on LP for two months now and none of the core-devs even acknowledged its existence in the bugreport17:59
nigelbvish: poke!18:00
vishnigelb: here? me? o.018:00
nigelbvish: ^^ I think thats something ayatana could look at18:00
nigelbLucidFox wrote this awesome patch to get indicator working with liferea, which we use via ppa for now18:01
vishLucidFox: tried #ayatana  ?18:01
LucidFoxI did, nobody replied there either18:01
vishi think tedg already had a patch for that.18:01
LucidFoxWhere? It's not on LP18:02
LucidFoxMy patch was all over the news, got talked about on Omg! Ubuntu, and I've never seen any other one in a public place. Definitely not on LP, at least18:02
vishLucidFox: bug #  , tedg/kenvandine are probably the best people to review it?18:04
vishLucidFox: its attached to a bug? or just a branch?18:04
LucidFoxbug #54049018:04
ubottuLaunchpad bug 540490 in liferea (Ubuntu) "liferea should be added to the indicator applet" [Undecided,Confirmed] https://launchpad.net/bugs/54049018:04
nigelbg6318:05
nigelbgrr18:05
nigelbwindow fail again18:05
vishLucidFox: tedg seems absent today .. :(18:06
nigelbisn't there a tag for it?18:07
vishyeah , there is18:07
vishLucidFox: if no response , jcastro is the next best person who can be poked :)18:07
kenvandinetedhave you looked at that?18:10
kenvandinewhoops18:11
kenvandinetedg isn't here :)18:11
kenvandineLucidFox, ping tedg in #ayatana18:11
vishkenvandine: thanks  :)18:14
vishkenvandine: seems LucidFox has been asking about that for a while and no one has responded..18:15
JontheEchidnaSo I've added a few PPAs via software-properties, but they do not seem to be showing up in software-center, and the origin as provided by libapt-pkg is empty. Any APT expert have any idea what could cause that?18:28
JontheEchidnaOne hint that I have is that while adding the PPA, the lookup for keyserver.ubuntu.com fails, but I don't know if that would make apt not see the origin...18:29
LucidFoxI wonder, how does bash tab completion know when to unwrap to .dsc, when to .changes, and the like?18:31
LucidFoxI noticed that it always tab-completes to .dsc with pbuilder-dist, and to .changes with dput18:31
pittiLucidFox: you can't upload (dput) a .dsc18:31
pittiLucidFox: and likewise you can't unpack (dpkg-source -x) a .changes, only a .dsc18:31
pittiso it's quite obvious18:31
JontheEchidnadput installs a file for the bash autocomplete to use: /etc/bash_completion.d/dput18:31
pittilikewise dpkg -i expands in *.deb18:31
geserLucidFox: it's context-sensitive (i.e. command specific)18:31
LucidFoxI know, but how does bash know which file to tab-complete t-- oh18:31
pittiah, yes; completion.d/ rules18:32
JontheEchidnaI imagine the other CLI tools do similar things18:32
mvo_JontheEchidna: the missing authentication keys for the ppa are most likely the problem18:33
mvo_JontheEchidna: if you add them the origins should become visible18:33
JontheEchidnamvo_: ok, neat. thanks18:33
JontheEchidnaIs there a way to manually add the keys without the keyserver.ubuntu.com lookup?18:34
JontheEchidnaoh, the PPA page gives the key, nevermind :)18:34
mvo_you can manually grab them from launchpad18:34
mvo_but that is a bit cumbersome18:35
mvo_is keyserver.u.c timeing out for you?18:35
JontheEchidnafor both software-properties and add-apt-repository, yes18:35
JontheEchidnathe web interface seems to be working though18:36
JontheEchidnawell, the import seems to be working now :)18:38
JontheEchidnamust have fixed itself while I was at lunch18:38
bigondhcp 4 is now in debian unstable, any plan to switch to this version for maverick?18:39
=== Lutin is now known as Guest14959
=== MacSlow|afk is now known as MacSlow
JontheEchidnamvo_: yeah, it's all fine now. Origin and origin labels showing up as expected :)18:41
mvo_cool18:45
shadeslayerjcastro: thanks :D18:47
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
=== al-maisan is now known as almaisan-away
=== fta_ is now known as fta
=== fta_ is now known as fta
td123does anyone know where I can get the source to the mentioned fixed package all the way at the bottom of https://bugs.launchpad.net/do/+bug/591998 ? I'm mainly interested in the patch they applied to fix this bug20:22
ubottuLaunchpad bug 591998 in Do "Can't not recive any command" [Undecided,New]20:22
Laneytd123: you better talk to the gnome do developers20:23
td123Laney: it's an ibus issue btw :)20:24
=== fta__ is now known as fta
=== tkamppeter_ is now known as tkamppeter
=== oubiwann is now known as oubiwann-away
=== oubiwann-away is now known as oubiwann
apacheloggerlex79: you might want to consider backporting my superior desktop file patch in kde4libs to the PPA, should improve performance quite a bit, especially combined with latest pkg-kde-tools22:50
=== sconklin is now known as sconklin-gone
apacheloggerneversfelde: bug 528259 is on your todo?22:55
ubottuLaunchpad bug 528259 in kdepim (Ubuntu) "Blogilo "QSqlDatabase: QSQLITE driver not loaded" " [Undecided,New] https://launchpad.net/bugs/52825922:55
bdrungto which mailing list can i send legal questions?22:57
apacheloggerJontheEchidna: in case I forget, I booted cia off kubuntu-devel for now, to revert that just set the basic filtering to custom again :)22:59
JontheEchidnaapachelogger: kk, I was about to message you about noise pollution in that regard, actually :P22:59
=== bjf is now known as bjf[afk]
geserbdrung: I'd probably try debian-legal as both Debian and Ubuntu share a great part of a common understanding in licenses23:01
geserbdrung: and if the questions is very Ubuntu specific, I'd probably mail the TB23:02
bdrunggeser: it's not a license problem. it's a problem with laws.23:02
geserin what way? can you be more specific?23:02
bdrunggeser: then i mail the TB. i want an answer to the question why libdvdcss2 is not in the archive. i found no answer on the web.23:03
maxwellianOh man, I forgot about Kubuntu tutorials day!23:03
apacheloggerinteresting, apparently one can only decline nominations for all packages, despite affected status -.-23:03
Riddellmaxwellian: still going on23:03
apacheloggeryeah, you hijacked my doctor who fan talk channel -.-23:04
maxwellianAwesome. :)  I wanted to see the packaging one, but I think I missed it.  Anyway, thanks.23:04
apacheloggerJontheEchidna: don't blink! :P23:04
JontheEchidnanot the angels!23:04
apacheloggerin that case...23:04
apacheloggercan anyone here the drums? ;)23:04
apacheloggers/here/hear23:05
apacheloggerJontheEchidna: muon build fails horribly if debconf stuff is not there :/23:12
lex79apachelogger: iirc the latest pkg-kde-tools needs the latest dpkg :)23:12
JontheEchidnaapachelogger: don't I have a cmake check for that now?23:12
apacheloggerI really think a find script would be good there :/23:12
apacheloggerJontheEchidna: yeah, just that it will fall because the file is not there23:12
lex79but I will backport your patch23:12
apacheloggerwhich is not a good error message :/23:12
apacheloggerlex79: well, you can backport my changes to pkg-kde-tools ;)23:13
apacheloggerlex79: you only need to take the cdbs bits I suppose23:13
lex79kk23:13
lex79I have to redo kdelibs, they uploaded the wrong tarballs23:13
lex79bah23:13
lex794.5.60 instead of 4.4.9223:13
apacheloggerthat is because dirk is using a custom rip off of markey's amarok release script, instead of using my superior rewrite23:14
Riddelldirk messed up?23:14
JontheEchidnaapachelogger: I will also have to talk to dantti about libdebconf-kde, it currently does not installed a versioned .so, so dh_shlibdeps isn't picking it up as a dependency23:14
lex79Riddell: he already fixed in ktown, me not yet :)23:15
apacheloggerRiddell: it rarely happens I have been told :)23:15
geserbdrung: I guess mailing the TB is the best plan. They can forward this to the Canonical lawyer if necessary. I've only found a thread from 2006 but with no clear answer.23:16
apacheloggerJontheEchidna: updating muon is quite the adventure with all those deps ^^23:16
apacheloggerJontheEchidna: search is the broken23:16
JontheEchidnaorly?23:16
apacheloggerJontheEchidna: search for linux -> does not yield linux-* stuff23:16
bdrunggeser: the mail is sent to the TB. let's wait for the results.23:16
Riddellbdrung: the letterws DCMA would be the reason23:16
Riddellalthough ordered as DMCA :)23:17
apacheloggerJontheEchidna: search for linux-image works though23:17
JontheEchidnaapachelogger: it might be that linux-* is below the quality cutoff for "linux" search term23:18
* apachelogger finds that unhandy23:18
apacheloggeroh, while I am at it23:18
JontheEchidnae.g. other package names/descriptions contain "linux" to place "linux-" below the threshold23:18
apacheloggerwhat would be cool is to have a button "nuke all old kernels" ^^23:18
JontheEchidnasince "linux" is a more exact match23:18
apacheloggerwith such a button I would not mind linux not matching linux-image ;)23:19
geserRiddell: but wouldn't this only prevent it from entering main/universe? e.g. the DMCA doesn't apply (yet) to Europe (but they might be other laws in Europe that prevent the legal use there)23:19
JontheEchidnaI usually search for "2.6.35" or whatever23:19
apacheloggereven though that is the most appropraite package really :P23:19
apacheloggerJontheEchidna: well, I had a billion billion kernels installed, as noticed earlier23:19
JontheEchidnalol23:20
apacheloggerspanning across various versions of course23:20
* apachelogger missed mass marking for action on that occasion too :)23:20
Riddellgeser: main and universe are mirrored in the US, and in Europe the Copyright Directive 2001/29/EC makes it illegal here too23:20
bdrungRiddell: it's in the gray zone in europe.23:21
JontheEchidnaapachelogger: oh oh, try marking the package "gnome" for install :D23:21
JontheEchidnaapachelogger: are you in maverick?23:21
apacheloggerno23:22
* apachelogger needs to do serious development for master Riddell :P23:22
JontheEchidnaI'll have to show you then ^.^23:22
apacheloggermust not have my system broken by lex79 uploading something funny :P23:22
* apachelogger hugs lex7923:22
apacheloggerJontheEchidna: pretty please :)23:22
Riddellbdrung: Copyright Directive 2001/29/EC is pretty clear23:22
lex79my stuff is always good23:22
JontheEchidnaapachelogger: http://simplest-image-hosting.net/i0-plasma-desktopv18355-jpg.jpg <- apt-get has the same error, so it's not muon. I suppose I should file a bug23:22
geserRiddell: does it imply that if a package is illegal in one country with an Ubuntu mirror, it can't be included in main/universe at all?23:23
Riddellgeser: in a country where canonical has an office yes23:23
Riddellwhere there's a serious likelyhood of canonical getting sued it won't go in the archive23:24
apacheloggerJontheEchidna: yeah sure, like you couldnt sell this as killer feature :P23:24
JontheEchidna:P23:24
* apachelogger would buy a product that prevented him from having a billion billion kernels ;)23:24
=== rsalveti` is now known as rsalveti
apacheloggeroh, and associated headers \o/23:24
apacheloggerJontheEchidna: marking foo as purge, where bar depends on foo, will mark bar as remove - IMHO logical assumption should be that bar ought to be purged too (though I suppose a dialog presenting additional changes with a tick box for purging would be best either way)23:26
JontheEchidnahrm, recursive purge23:27
JontheEchidnaI'd probably have to set an apt setting or something23:27
JontheEchidnabecause when you hit that button, you're really only telling that package to purge, then apt takes care of dependencies and you have no say in what it does23:27
JontheEchidnakinda sucks23:28
apacheloggermeh23:28
apacheloggermore interesting: http://imagebin.ca/view/sbgYh6V.html23:28
apacheloggercan not help but noticing that it orders 19 < 22 < 23 < 20 < 21 < 1623:29
JontheEchidnathat is a lot of headers23:29
apacheloggerthat is an interesting way of seeing the world really :)23:29
JontheEchidnathat is some compression: http://simplest-image-hosting.net/i0-plasma-desktopa18355-jpg.jpg23:32
JontheEchidna(download size vs install size)23:32
JontheEchidnaapt-get agrees, but I had to check to make sure it wasn't a bug :P23:33
apacheloggerimagine if that was all lzma ^^23:33
apachelogger3mib vs. 1.2 gib :P23:34
JontheEchidna:P23:34
* JontheEchidna uses the fancy undo button23:34
JontheEchidnaundo only goes back 20 actions, tho.23:35
JontheEchidnato save ur RAMz23:35
apacheloggerhm23:36
apacheloggermanually swap it23:36
apacheloggerthat way you can go across sessions23:36
apacheloggerimagine that!23:36
apacheloggeryou go install foo23:37
apacheloggertest foo23:37
apacheloggerdont like it23:37
apacheloggeropen muon again23:37
apacheloggerpress undo23:37
apacheloggerand voila!23:37
JontheEchidnaundo gets lost after commit23:37
apacheloggerthat would be brilliant I think23:37
apacheloggerJontheEchidna: well, manually rebuild it ;23:37
apachelogger)23:37
JontheEchidnabut a history log is definitely a good idea23:37
=== tkamppeter_ is now known as tkamppeter
apacheloggerif you log it you can go the extra mile and store it in a format that can be reused for cross session undo ;)23:38
* apachelogger wishes he had a timey wimey detector just to be a bit cooler :/23:39
apacheloggerJontheEchidna: suggestions on bug 554514 ?23:45
ubottuLaunchpad bug 554514 in akonadi (Ubuntu) "cant find resource agents" [Medium,New] https://launchpad.net/bugs/55451423:45
JontheEchidnaabandon all hope -> close23:46
apacheloggerother than setting it on fire and run while we can23:46
JontheEchidnaheh23:46
apachelogger:/23:46
apacheloggerakonadi trunk is a whole different code base -.-23:46
apacheloggerthere is not even hope to get that stuff fixed23:46
JontheEchidnawe could start a role playing game where we pose as various different animals and the reportees have to guess which animal we are23:47
JontheEchidna*reporters23:47
apacheloggerrofl23:47
keescan an archive admin take a look at the -7 kernel binaries in NEW?23:47
apacheloggerJontheEchidna: so, how do I bring this news to the reports without getting shot? there are 41 people affected, which is mostly because that bug is as inprecise as they get :/23:48
JontheEchidnaapachelogger: bribe an LP admin to silently delete it? :P (just kidding)23:49
ScottKJontheEchidna: appmenu in action: http://kitterman.com/kubuntu/appmenu1.jpeg23:50
JontheEchidnaoh lovely, it has icons and shortcuts now. (It didn't at UDS)23:50
apachelogger"due to ethic believes I must not do certain things to dead people, animals or inspects"23:50
apacheloggers/inspects/insects23:51
* apachelogger really should get more sleep23:51
apacheloggerScottK: I wonder how that goes under load23:52
ScottKapachelogger: That's on my netbook, so everythin is always under load it seems.  Not too bad so far.23:52
apacheloggerk23:52
ScottKJust started playing with it today though.23:52
apacheloggerdbusmenu has some oddish issues when it comes to high load situations, seeing that menus can have loads and loads of entries that is a bit of concern of mine23:53
apacheloggerhow kubuntu.org will look on fluffy http://imagebin.ca/view/V-aeXjgi.html \o/23:56
lex79lol23:57
Riddellwe should just do that for stock kubuntu.org, would give us a new market niche23:58
apacheloggeragreed23:59

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!