/srv/irclogs.ubuntu.com/2008/05/30/#ubuntu-devel.txt

cjwatsonglad we (and Debian) aren't the only ones with that kind of horrible packaging hack ;)00:00
Keybukgod00:02
Keybuknow I have to figure out how to do this with cdbs00:02
KeybukI don't suppose you know off the top of your head?00:02
ion_binary-post-install/foo ::00:03
cjwatsonsed -i in ... what ion_ said00:03
ion_        sed -ire '...' debian/$(cdbs_curpkg)/foo/bar.pc00:03
Keybukdoesn't work00:09
Keybukdebian/tmp has been wiped by then00:09
cjwatsondebian/libdbus-1-dev/...00:10
cjwatsonit's been dh_install-ed00:10
Keybukit's not referenced by dh_install00:10
Keybukliterally the whole of debian/tmp is gone, including the directory00:10
KeybukI hate CDBS00:10
cjwatson$ fgrep .pc debian/libdbus-1-dev.install00:10
cjwatsondebian/tmp/lib/pkgconfig/dbus-1.pc usr/lib/pkgconfig00:10
Keybukyeah I removed that in an attempt to do the sed ;)00:10
cjwatsonput it back and do 'sed -i blah debian/libdbus-1-dev/usr/lib/pkgconfig/dbus-1.pc' in binary-post-install/libdbus-1-dev00:11
cjwatsoni.e. let it install and then sed in-place00:11
Keybuksed: can't read debian/libdbus-1-dev/usr/lib/pkgconfig/dbus-1.pc: No such file or directory00:13
Keybuknope00:13
Keybukstill fail00:13
* Keybuk finds a place that works00:16
sorenI would just put it in install/libdbus-1-dev00:17
sorenWhen's binary-post-install?00:17
=== no0tic is now known as maru0tic
Keybuknow I ended up with a .pce file ?!00:18
soreno_O00:19
ion_Oh, -i must be separated from -e etc.00:20
ion_As in, sed -i.backup -re '...' something00:20
sorenOh. Heh.. :)00:20
=== maru0tic is now known as no0uscia
=== no0uscia is now known as no0tic
cjwatsonwhy bother with -e?00:25
cjwatsonyou don't need it00:25
cjwatsonnor do you need -r in the example Keybuk gave00:25
ion_True. I just have -e as a habit. And -r is kind of a default for me. :-)00:26
ion_I didn’t really take a good look at the actual sed expression.00:26
cjwatsonit has $ where it should have \$00:27
ion_Actually, \$$ since it’s in a Makefile.00:27
cjwatsonoh yes00:27
Keybuk?00:29
cjwatsonthis looks about right00:31
cjwatsonbinary-post-install/libdbus-1-dev::00:31
cjwatson        sed -i 's@-I\$${libdir}@-I$${prefix}/lib@' debian/libdbus-1-dev/usr/lib/pkgconfig/dbus-1.pc00:31
Keybukwhy use ${prefix} there at all?00:31
Keybukand why \$ ?00:32
cjwatson\$ because sed thinks $ means end of line00:33
cjwatson${prefix} because the RH code did ;-) and seems neater00:33
Keybukinstall/libdbus-1-dev::00:37
Keybukmkdir -p debian/libdbus-1-dev/usr/lib/pkgconfig00:37
Keybuksed -e 's@-I\$${libdir}@-I$${prefix}/lib@' debian/tmp/lib/pkgconfig/dbus-1.pc > debian/libdbus-1-dev/usr/lib/pkgconfig/dbus-1.pc00:37
Keybukwas what I went with in the end00:37
cjwatsonthe version I had was sufficient to get openssh building, at any rate00:40
cjwatsonif yours produces the same .deb contents (as it looks like it should) then we ought to be fine00:41
Keybukopenssh depends on dbus now? :)00:41
cjwatsonconsolekit00:42
Keybuksshkit00:42
cjwatsonmy Bad Idea light is going off00:43
ion_At this rate, /bin/sh, init and finally the kernel will use dbus. ;-)00:43
Keybukah, you've screwed it back in after spending a week in the same building as me? :)00:43
Keybukion_: well, two out of three00:44
cjwatsonKeybuk: new filament in the bulb00:44
ion_keybuk: dbus being actually useful for some of them was actually supposed to be part of the joke. :-)00:45
cjwatsonopenssh upstream aren't too happy about the fact that dbus is GPL and openssh is rather aggressively not, which unfortunately I'd completely failed to consider00:45
cjwatsonso I think I need to figure out how to split that stuff out into a helper00:45
Keybukdbus is attempting to be MIT00:45
cjwatsonthat would help00:45
Keybukit's also AFL00:45
cjwatsonthe AFL is a bit impenetrable00:46
Keybukthe MIT thing is held up by one of the companies that contributed having vanished00:46
Keybukapparently this is a major stumbling block, because they can't ok it00:46
Keybukto my mind, if they've vanished hard enough, they can't take legal action either :p00:46
ion_Heh00:47
cjwatsonI think (on reading it) that the AFL is MIT-like enough for it not to be a legal problem for Ubuntu; it's literally 10 times the line length though00:49
cjwatson(which tends to trip openssh upstream's WHATEVER meter)00:49
Keybukyes, but openssh is not on the rational side of the world00:49
Keybukit belongs in that special camp with other upstreams like glibc00:50
cjwatsonhmm. on re-reading, describing the AFL as MIT-like is definitely false - it requires source code provision00:53
cjwatson(not that I object but it's one thing the MIT/BSD licences determinedly don't do)00:53
Keybukerr00:55
Keybuknot entirely true00:55
Keybukit only requires source code provision of the original work when distributing it00:55
Keybukit doesn't require source code provision _at_all_ for derivative works00:55
cjwatsonoh, true00:56
cjwatsonugh, though, clause 9 is horrible00:56
cjwatsonyou have to make a "reasonable effort" to obtain explicit assent to the licence00:56
Keybukyeah it's a sucky licence00:56
Keybukwhich is why they want to change it to MIT/X1100:56
cjwatsonis there anything I can quote on that, so openssh upstream don't just get fed up and close the bug? :-)00:57
Keybukon the relicense?00:57
cjwatsonyeah00:58
Keybuklots and lots of ML posts00:58
Keybukbut the conclusion is that it's stuck still00:58
cjwatsonoh, it's on the website, I can quote that01:00
Keybukhttp://lists.freedesktop.org/archives/dbus/2008-February/009410.html01:00
Keybukthe current status is that for almost all of the code, the authors have agreed to an MIT/X11 licence01:01
Keybuk(including us! :p)01:01
cjwatsonta01:02
Keybukrandom use of bzr #1501:13
Keybukimport two version of a package01:13
Keybukand use bzr blame to figure out which lines haven't been changed01:13
cjwatsonlike inverse diff?01:18
Keybukyeah01:18
Keybukpatchutils doesn't have one ;)01:18
Keybuk/usr/bin/same :p01:19
ion_Heh01:19
alex1hello. I've submitted a patch upstream (linux-kernel) for detection of macbook pro 4,1 and macbook air keyboards. It'd be nice if it made it into ubuntu sooner. Should I resubmit it for ubuntu, and if so, where?01:22
Keybukubuntu-kernel01:27
alex1is that a mailing list? what's the full address? Is there a page that details the patch format for ubuntu etc?01:30
Keybukmailing list01:32
Keybuksame general method as lkml01:32
alex1hm https://lists.ubuntu.com/mailman/listinfo/ubuntu-kernel says no such list01:33
Keybukgrep for kernel ;)01:33
alex1kernel-bugs?01:34
cjwatsonkernel-team01:34
alex1ok. thanks guys01:34
alex1new to this stuff... heh01:35
cjwatsonhttps://wiki.ubuntu.com/KernelGitGuide01:35
alex1can I just resend the patch I sent upstream?01:37
Keybukprobably01:44
alex1done. thanks again :)01:48
=== asac_ is now known as asac
=== asac_ is now known as asac
cody-somervilleWhat does it mean when kill -9 won't kill an process? :P03:14
RAOFIt's probably in D state?  It'll be waiting in the kernel, and can't recieve signals at all.03:15
cody-somervilledoh, it is in D state03:16
cody-somervilleStupid firefox :/03:16
RAOFYup.  Unkillable.03:17
cody-somervilleWeird... new applications won't start :/03:17
RAOFD state makes the baby jesus cry.03:17
cody-somervilleNow all my applications are going into D state :/03:18
RAOFRight.  Something has died horribly, and the world is ending.03:18
cody-somervilleReboot is only option?03:20
RAOFIf you can get /sbin/shutdown to not D state, yeah :)03:21
kirklandslangasek: thanks for merging ;-)03:22
slangasekn/p03:22
cody-somervillenooo!!! :(03:23
* cody-somerville just lost the Xubuntu Strategy document draft 2.03:23
=== asac_ is now known as asac
=== pabluk_ is now known as pabluk
DrVagueGood Afternoon.05:52
DrVagueI was sitting briefly, aghast 'n pondrance, and it became notable to one that there is no cron gui management tool, that one of course being me; if you could so explain some burdened parly to me in the form of a hairdresser in silk stockings, I'd be very delighted, or possibly intrigued to find a solution of some sort.05:54
DrVague?05:56
lamontDrVague: that's probably because any gui-driven job scheduler does far more than cron, and the gui comes after the decision that cron is lacking06:04
DrVaguelamont: what?  you're suggest a replacement backend for cron?06:04
DrVagueyou're kidding?06:05
lamontno.  I'm saying that a gui for cron is the least of its issues06:05
lamontif you expose cron to a gui-requiring user, then you also have to teach him shell scripting...06:05
DrVaguehurrrrr?06:07
DrVaguethat doesnt follow for me06:07
lamontwell, what shell commands is he going to put in his crontab?06:07
DrVaguei was thinking he would just use the gui to schedule actual sh scripts to run and set the time with a mouse06:08
DrVague'add new job' ::file selection dialog:: 'time'  ::calendar::06:08
lamontand then the first thing that anyone who's used an actual job scheduler on (say) a mainframe runs into is "how do I tell it to release job B when job A finishes??"  (answer? run a real job scheduler instead of cron, or hack together your own scheduler in the jobs)06:08
DrVague'save' 'cancel'06:08
lamontright.  and someone has to write those shell scripts...06:09
lamontand whoever writes those generally drops something to run them into /etc/cron.*06:09
DrVaguethats like saying the team who maintains geany should be responsible for teaching everyone perl06:09
lamontmy point is more that anyone who is satisfied with cron in the current world doesn't use a gui, and there is zero motivation to provide one for the developers06:10
DrVaguetbh, cron is used for the same, probably 2 dozen unique job scripts anywhere, a pulldown menu would cover most peoples' needs06:11
lamontrather, no motivation for the developers to produce one, since the user community divides neatly into the group that doesn't see the need, and the group that would ask lots of questions of whoever wrote the gui.06:11
DrVagueoh, yeah, well that is def true06:11
lamontwhich tends to discourage people from bothering with that, when there are far more interesting/significant things to work on that don't immediately become a support nightmare.06:12
DrVaguehrm.  i hate that community gap between devs and recreational users.06:12
dholbachgood morning06:12
lamontmorning dholbach06:12
DrVaguemorn'06:12
Mithrandirmorning, lamont, Daniel06:12
lamonthrm... I think that means it's about bedtime. :-)06:13
dholbachhi lamont DrVague Mithrandir :)06:13
Mithrandirlamont: hehe.06:14
lamontwell, given that the workday starts in just under 7 hours...06:14
lamonthrm.. importing a 300MB, 5900 file cvs repo into a real VCS?   not so fast.06:15
Mithrandirare you surprised?06:15
lamontnope06:16
lamontI just hope it finishes06:16
pittiGood morning06:32
TheMusoMorning pitti.06:34
Hobbseepitti!06:35
dholbachhi pitti, TheMuso, Hobbsee06:35
TheMusoHeya dholbach.06:35
ajmitchhello pitti06:38
geserHi pitti, Hobbsee, TheMuso, ajmitch06:51
wgrantAnybody got any idea why /dev/mapper contains all of my LVM volumes, but /dev/<VGName> doesn't?06:52
pittiwgrant: /dev/vgname isn't even supposed to exist, AFAIK06:58
wgrantpitti: sbuild seems to want it.06:59
wgrantWell, schroot.06:59
slangasekdon't schroot, I'm unarmed07:06
pittiRiddell: done now07:16
sdhhmmm07:42
sdhif you install slocate and run update-alternatives to select it as the preferred locate (over mlocate)07:43
sdhthen the slocate cron.daily script still doesn't fire because it has the following guard07:43
sdhif [ -x /usr/bin/slocate ] && [ ! -x /usr/bin/mlocate ]07:43
pittisdh: that was done to avoid running slocate on gutsy->hardy (or dapper->hardy) upgrades07:43
pittisince we don't want/can't automatically uninstall slocate completely07:44
sdhpitti: interesting... why not consult alternatives and act based on that?07:44
sdhi did a gutsy -> hardy upgrade, and got mails from cron like this:07:44
sdh/usr/bin/updatedb.mlocate:/etc/updatedb.conf:4: unknown variable `FINDOPTIONS'07:44
sdh/usr/bin/updatedb.mlocate:/etc/updatedb.conf:5: unknown variable `export'07:45
sdhetc... which i assume is mlocate trying to parse slocate configuration07:45
pittinoone went through that much trouble, I guess, since slocate is basically deprecated07:45
pittiright, that would be it07:45
sdhhmm07:45
pittisdh: can you please file a bug about it? it's worth it, at least07:45
sdhpitti: i still have to work out exactly what happened ;-)07:45
sdhbut yeah, sure07:45
sdhnow i'm confused:07:46
sdh% dpkg -S /etc/updatedb.conf07:46
sdhmlocate: /etc/updatedb.conf07:46
sdhi won't file a bug on that until i can understand what is going on07:48
sdhbut i note that somebody said "8.04 uses mlocate instead of slocate on desktops now" -- but it seems it is used on servers too, which imho negates the reasoning behind deprecating slocate in favour of mlocate07:49
sdh(i.e. affecting user experience especially on laptops)07:49
pittisdh: both slocate and mlocate ship this configuration file07:49
pittion hindsight, we should have probably made the two conflict to each other07:50
sdhsounds like a plan07:50
pitti(we can still do that in -updates, at least)07:50
sdhcool07:51
Mithrandirsdh: may I ask why you prefer sloacte?07:51
Mithrandirslocate, even07:51
sdhMithrandir: it's not that i prefer slocate - though i do think, in my case "if it's not broken, don't fix it" - it's that i dislike getting error messages from crond when the locate implementation is changed07:52
Mithrandirsdh: oh, that I can agree with.07:52
sdhi see that i've been using mlocate unknowingly on my hardy laptops for the last few months anyway and i have no problem with it :)07:52
Mithrandiryeah, I can't see anything slocate gives us over mlocate so I was wondering why anybody would want to continue using slocate.07:54
sdhsentimental reasons ;-)07:54
sdhoh god the whole mlocate/slocate thing is a mess07:55
sdhi have /etc/cron.daily/{mlocate,slocate,find}07:55
sdhi think that's https://bugs.launchpad.net/ubuntu/+source/mlocate/+bug/19717007:56
ubottuLaunchpad bug 197170 in mlocate "cron daily runs updatedb twice" [Undecided,Confirmed]07:56
* sdh wonders why findutils has cron.daily/find doing an updatedb07:58
liwsdh, so that the locate command is useful; there's been discussions of getting rid of that07:58
sdhliw: i thought locate (1) was just an alternatives link into mlocate/slocate08:00
sdhman i am confused now, i have mlocate installed and seem to have no locate (1)08:00
liwsdh, locate is part of find, and the cron job runs regardless of what the alternatives say (which, as it happens, could be considered a bug: please file one :)08:00
sdhliw: trying to get my box working first :) any chance you could dpkg -S bin/locate for me please? i appear to have lost mine08:01
sdhfindutils, apparently. odd that i don't have it, then.08:03
liwI don't seem to have it either08:03
liwhmm08:03
liwthis seems like a bit of a mess08:03
sdhyes! b0rkage!08:03
sdhdid you do any mlocate/slocate manual tweaking?08:03
liwno08:03
sdhsteve@ace:~% dpkg -l '*locate*'08:04
sdhNo packages found matching *locate*.08:04
sdhsteve@ace:~% dpkg -S bin/locate08:04
sdhfindutils: /usr/bin/locate08:04
sdhactually ignore that, that's a gutsy box! :)08:04
sdhbut the problems i've been discussing are on a hardy (from gutsy) box08:04
sdhright, having reinstalled findutils it definitely contains no locate(1)08:05
Mithrandir> dpkg -S bin/locate08:05
Mithrandirlocate: /usr/bin/locate.findutils08:05
Mithrandiron hardy08:05
sdh*boggle*08:05
Mithrandir/usr/bin/locate is handled as an alternative08:05
sdhah yes08:06
sdh% update-alternatives --list locate08:06
sdh/usr/bin/mlocate08:06
sdh% locate08:06
sdhzsh: command not found: locate08:06
sdhsorry for the pasting btw, is it excessive yet?08:07
Mithrandirnah, no problem when it's just a line or two at a time.08:07
sdh;>08:07
Mithrandirhm, I think this machine was installed with hardy, not gutsy.08:08
sdh% dpkg -L mlocate | grep 'bin.*locate'08:08
sdh/usr/bin/mlocate08:08
sdh/usr/bin/updatedb.mlocate08:08
sdh% sudo update-alternatives --config locate08:08
sdhThere is only 1 program which provides locate08:08
sdh(/usr/bin/mlocate). Nothing to configure.08:08
sdh% locate foo08:08
sdhzsh: command not found: locate08:08
Mithrandiryou might need to rehash08:08
sdhthat path is wrong08:08
sdhsteve@nemesis:~% rehash08:08
sdhsteve@nemesis:~% locate foo08:08
sdhzsh: command not found: locate08:08
sdh;)08:08
sdhit's the path in the alternatives08:08
Mithrandirwhat does ls -l /etc/alternatives/locate say?08:09
sdhor, hang on08:09
Mithrandirlrwxrwxrwx 1 root root 16 2008-04-03 15:21 /etc/alternatives/locate -> /usr/bin/mlocate*08:09
Mithrandiris what I have08:09
sdherk!08:09
sdh% ls -l /etc/alternatives/*locate*08:09
sdhlrwxrwxrwx 1 root root 16 2008-05-30 07:42 /etc/alternatives/locate -> /usr/bin/slocate08:09
Mithrandirtry update-alternatives --auto locate ?08:10
sdhthat seems to have fixed it08:10
sdhbut why is u-a telling me it points to mlocate when it points to slocate, i wonder?08:10
Mithrandirit's not, it's just telling you there's nothing to configure08:10
sdhgood point >08:11
MithrandirI wonder if this is really an update-alternatives bug.08:11
sdhi am inclined to agree08:11
Mithrandircan you try installing slocate, changing the alternative to point to slocate, verify it's in manual mode pointing to slocate, then purge slocate?08:11
sdhit shouldn't say "only 1 program, nothing to configure" if it is in fact handling it, badly :)08:11
Mithrandirand then see what it points to?08:11
sdhthough i'm reluctant to keep screwing with my production boxes... sure! :)08:12
sdhMithrandir: step 2 using u-a --config locate, right?08:12
Mithrandiryes08:12
Mithrandirno manual hacking in /etc/alternatives. ;-)08:12
sdhUsing '/usr/bin/slocate' to provide 'locate'.08:13
sdhlrwxrwxrwx 1 root root 16 2008-05-30 08:12 /etc/alternatives/locate -> /usr/bin/slocate08:13
sdhnow to purge08:13
sorenMithrandir: Why not?08:13
sdh% sudo update-alternatives --list locate08:13
sdh/usr/bin/mlocate08:13
sdhlrwxrwxrwx 1 root root 16 2008-05-30 08:13 /etc/alternatives/locate -> /usr/bin/mlocate08:13
sdhseems good now08:14
Mithrandirsoren: in this case, because I want to see if u-a DTRT.08:14
sorenMithrandir: Oh :)08:14
Mithrandirsdh: hmm, ok.08:14
sdhMithrandir: strange though...08:14
sdhMithrandir: would you expect an apt-get remove, without --purge, to update alternatives?08:14
sdhMithrandir: let me try that again without purge.08:14
Mithrandirsdh: sure, try.  I think it should.08:14
Mithrandirsince you should never be left with a dangling alternatives symlink.08:15
Mithrandirbut you said you were left without /usr/bin/locate at all, or was it just dangling?08:15
sdhhmm that seemed to work too, odd.08:15
sdhMithrandir: hard to tell, now that i've messed about! it gave me no such f/d, but that could be either08:16
sdhlooks to me like it's just a gutsy/hardy upgrade quirk08:16
sdhbut it does feel like a can of worms, including *locate and even u-a08:17
Mithrandiryou don't happen to have a backup of the system you could take a look at a directory listing from?08:17
sdhand i definitely think that mlocate/slocate should be mutex08:17
Mithrandiryeah, it looks like it might be ordering dependent.08:17
sdhMithrandir: not in any kind of realistic timescale, afraid not08:17
Mithrandirwhich is.. ugh.08:17
Mithrandirok.08:17
sdhthanks for helping out08:19
Mithrandiroh, happy to help; I still wonder how we can get it properly fixed though.08:19
sdhyeah but it's at that awkward stage where i'm not sure how to reproduce the problem08:20
Mithrandirit might be related to /usr/bin/locate being a proper file in gutsy.08:20
sdhi'm not even 100% sure what the problem was08:20
sdhi have to get to work soon... but i suppose i have time to break out a few VMs :)08:21
Mithrandirif it's present on a stock install + upgrade, it should be fairly easy to see what's happening.08:21
sdhi conveniently have a gutsy template vm08:22
sdhright Mithrandir... i have a gutsy box ready for upgrade to hardy08:31
sdhi'll install slocate then do the upgrade08:32
sdhsound like a reasonable test to you?08:32
Mithrandirsounds good08:33
sdhjust to confirm, on gutsy /usr/bin/locate is a symlink to /usr/bin/slocate08:33
sdh(with slocate installed, obviously)08:33
Mithrandir*nod*08:34
sdhbefore that it was provided by findutils08:34
sdhit's chugging away :)08:40
dolphinhow would i go about messing with [remapping] keys08:41
sdhMithrandir: well it came back with mlocate installed and slocate uninstalled, which i guess is right08:53
sdhand the alternatives link points to mlocate08:53
sdh...so it all seems ok.08:53
Mithrandirhmm08:57
Mithrandirso it's.. something else or harder.08:57
pittiseb128: WDYT about bug 182945?08:58
ubottuLaunchpad bug 182945 in gio-standalone "gio-standalone should be remove, gio is in glib now" [Undecided,Confirmed] https://launchpad.net/bugs/18294508:58
pittiseb128: no reverse dependencies08:58
seb128pitti: it should have been removed before hardy, that was a temporary package for gio before having it in glib09:01
pittiseb128: merci; removing then09:01
seb128thanks09:01
seb128pitti: I just uploaded a new rhythmbox revision, I attached the debdiff to one of the zillion bugs it closes, is that good enough or should it be attached to all of those?09:02
pittiseb128: one is enough09:02
dolphinwhat's the best way to learn how to program?  i've taken a basic college course in C & C++, but I wanna look at real programs... even if they're mildly confusing at first09:03
dolphinany suggestions?09:03
dolphinwhat program has easy to read source code for a n00b??09:11
sdhdolphin: there's a good book... let me find it09:12
sdhhttp://www.spinellis.gr/codereading/09:12
dolphinsdh:  aww, man... you gotta pay for the whole thing?!?  =-(09:16
sdhdolphin: it's worth it09:17
liwdolphin, pick a program in Ubuntu that interests you, then do "apt-get source foo" to get the source, and have at it; it's probably best to pick a simple command line utility to start with, they tend to be simpler09:18
sdhfileutils or something is interesting09:18
sdhand easy to see how code relates to runtime09:18
dolphinyea, i was thinking graphical... but text is definitely a better start09:18
sdhdolphin: if you want graphical then you have to get familiar with the intricacies of whatever graphical toolkit is being used, be it gtk/qt/etc09:19
liwdolphin, the coreutils package might be a start, for example: cat, rm, cp, and such, although they have a lot of complexity by having to deal with all sorts of quirks in various Unix flavors09:19
sdhdolphin: that is hardy, so i'd start with text09:19
sdherr, s/hardy/harder/09:19
liwfindutils is a good choice09:19
dolphin$51 isn't that bad, i guess i'll order that book to keep me from getting too bored this summer09:20
sdhdolphin: it seems there is a sequel but i haven't read that09:20
dolphinwhat is fileutils?09:21
sdhdolphin: but code reading is a good book because it talks you through various idioms in real software (e.g. apache, netbsd) that means you can make the leap from understanding, say, C syntax, to being able to understand large chunks of code09:21
sdhdolphin: sorry, i meant coreutils i guess09:21
dolphinahh... k09:21
sdhthe package with head/tail/wc etc09:21
dolphinthis is definitely the starting point i'm looking for... i'm tired of simple labs09:22
sdhright09:22
sdhso between apt-get source and that book, i'd say you could spend a lot of useful time understanding and then modifying code09:22
dolphinthanks for the recommendations, ya'll!09:22
sdhnp09:22
dolphingonna catch some ZZZzz's for now.... lata09:23
dolphinthanks again!09:23
liwthe nice part of going to apt-get source way is that you can easily modify code too, and try out your modifications for real09:23
liw(running, say, your modified corutils in a virtual machine may give additional confidence)09:23
=== thekorn__ is now known as thekorn
james_wpitti: there's a packagekit discussion going on about offering to install relevant drivers when a piece of hardware is plugged in, does jockey take care of that for us?09:38
james_wI mean does it get triggered by hal, or does it just scan at startup?09:39
pittijames_w: not quite yet, but getting there09:39
pittijames_w: ATM it just scans at session start09:39
james_wso it's planned, and I can ignore trying to implement it for packagekit?09:39
pittijames_w: but for 0.5 I plan to hook it into hal09:39
james_wthanks09:39
pittijames_w: yes, pretty much; we'll do it the other way round, we'll use PK to install the drivers from jockey :)09:39
james_wheh, works for me :-)09:40
pittijames_w: and jockey will go into Fedora soon, too09:40
pittiso I'm not even sure whether they should do that in PK itself (one tool for one purpose, etc.)09:40
pittibut *shrug*09:40
james_woh, cool. Apport as well, they're loving your code.09:40
james_wpitti: <hughsie> james_w: atm, gnome packagekit has a GpkFirmware GObject that watches udev for missing firmware requests and prompts to install it if finds then in a source09:52
james_wsorry to keep bothering you09:52
pittijames_w: (no problem)09:52
pittijames_w: that sounds nice; however, our packages don't support that ATM09:52
pittijames_w: most firmware should already be present in l-r-m09:52
pittiand if someone uninstalls l-r-m, he won't want the non-free firmware IMHO09:53
pittiso that doesn't really work for/apply to us09:53
pittiATM, at least09:53
|DuReX|https://bugs.launchpad.net/ubuntu/+bug/23588909:56
ubottuLaunchpad bug 235889 in linux "BUG: scheduling while atomic: archhttp64/7146/0x1000000001" [Undecided,New]09:56
|DuReX|if somebody wants to look @ it :)09:56
james_wpitti: sorry, last question I hope, hughsie is interested to speak to the Fedora folks that you have been in contact with to find out what their plans are, could I pass on a name to him?10:08
pittijames_w: sure, that's Jon Masters (RedHat employee)10:08
james_wthanks10:09
pittijames_w: he and I are members of the LinuxFoundation driver backports workgroup, we work together on the tools in that context10:09
=== iceman is now known as iceman-away
sorenwtf..10:36
sorenAre moves of packages between components logged?10:37
cjwatsonnot really10:38
cjwatsonyou can see that something happened10:38
cjwatsone.g. https://launchpad.net/ubuntu/+source/db4.4 has both "Published" and "Superseded" for the same version (4.4.20-11)10:39
sorenAh, interesting.10:39
cjwatson(I think that indicates a publishing record change)10:39
sorenNo such luck on https://edge.launchpad.net/ubuntu/+source/nagios210:40
sorenin spite of https://bugs.edge.launchpad.net/ubuntu/+source/nagios2/+bug/21132310:40
ubottuLaunchpad bug 211323 in nagios2 "MIR for nagios" [Medium,Fix released]10:40
pittibryce: x-intel SRU upload does not have a LP # for tracking verification; can you please reupload and mention it in the changelog?10:44
brycedone10:44
pittiwow, that was fast :)10:44
brycepitti: yeah I realized I'd forgotten it right after uploading10:44
brycebtw, the other quirk is a fix for bug 235155, which is a private bug10:45
ubottubryce: Bug 235155 on http://launchpad.net/bugs/235155 is private10:45
brycedunno if it's appropriate to list private bugs in srus or not10:45
pittibryce: bug 235643 is public10:46
ubottuLaunchpad bug 235643 in xserver-xorg-video-intel "Need a Pipe-A quirk for 1028:0163" [Undecided,Fix committed] https://launchpad.net/bugs/23564310:46
pittibryce: the diff itself looks fine, so if there's at least one bug to say "yes, this .deb still works for me", that's enough for SRU purposes10:46
pittibryce: the quirks themselves are probably fine10:46
bryceok great10:46
pittiogra: heh, ltsp fix-o-rama :)10:48
wgrantsoren: It was never promoted. https://edge.launchpad.net/ubuntu/+source/nagios2/+publishinghistory shows all publishing records ever.10:49
=== ogra_ is now known as ogra
pittiogra_: can you please add hardy/intrepid tasks and pointers/attachments to the patches?10:49
ograpitti, ?10:49
ograoh ltsp you mean ?10:50
pittiyes10:50
ograwill do, one sec10:50
pittiogra: danke; please also say whether they are fixed in intrepid, etc (intrepid task status)10:50
sorenwgrant: Oh, that's handy. Thanks.10:50
wgrantsoren: Only nagios-plugins was promoted - I guess pitti misread something...10:51
sorenwgrant: They were handlede completely separately.10:51
wgrantnagios-plugins was mentioned a comment or two before the end, and I've had this sort of thing happen once or twice before.10:52
pittiwgrant: context?10:52
wgrantpitti: nagios2 was never promoted, though you said you did it. Bug #21132310:53
ubottuLaunchpad bug 211323 in nagios2 "MIR for nagios" [Medium,Fix released] https://launchpad.net/bugs/21132310:53
liwhrmph, reportsync isn't working for me10:53
liwer, requestsync, rather10:54
sorenliw: Maybe because it's called..10:54
sorenoh.10:54
pittiwgrant: hm; "oops"10:54
liwI edit the message template. I save it. requestsync doesn't notice and signs the unedited file.10:54
pittiwgrant: promoted in intrepid now; that doesn't really help hardy, of course :/10:54
wgrantsoren: ^^10:56
sorenpitti: I think the main problem (har har) is that component-mismatches is full of noise, so it's hard to notice when stuff like this pops up in there.10:56
pittisoren: yeah :/10:56
pittiseb128: voila, all gnome SRU stuff done11:00
* seb128 hugs pitti, you rock!11:01
liwhm, it works with nvi, but not with my own editor... weird11:02
seb128pitti: is that normal that the sru page has a language-pack-en entry without a bug number?11:05
liwor at least it presented me the right file having been signed, but then it seems stuck after I press enter to actually file the bug11:05
liwsocket.error: (110, 'Connection timed out')11:05
liwbah11:05
* liw goes to file bugs manually11:06
liwhmm... actually, it's probably because outgoing port 25 is disabled11:09
liwif requestsync would use my local MTA, it should work11:11
liwand there is a way to do that, which is not documented in the manpage11:15
ograpitti, they all got hardy tasks ... i cant do much about the intrepid nomination yet as there was no ltsp upload to intrepid yet11:15
james_wliw: there's a --lp option that files over http if you have python-launchpad-bugs installed.11:15
liwjames_w, yeah, but that's scary :)11:16
ograPici, (and especially jwz's toy bug #199675 is a bit tricky here)11:16
ubottuLaunchpad bug 199675 in ltsp "configure-x.sh generates broken xorg.conf files from lts.conf" [Medium,Fix committed] https://launchpad.net/bugs/19967511:16
liwjames_w, I'll try it next time, though11:16
liwand am putting fixing the manpage on my todo list11:17
* |DuReX| looks sweet to the devs -- https://bugs.launchpad.net/ubuntu/+bug/23588911:20
ubottuLaunchpad bug 235889 in linux "BUG: scheduling while atomic: archhttp64/7146/0x1000000001" [Undecided,New]11:20
pittiogra: I added the missing one to 224259, rest is fine; but still no patches...11:24
pittiogra: I'll read the debdiff first, and get back to you with questions if I can't figure it out11:25
pittibut having patches/bazaar.lp.net URLs are a great help for fast processing11:25
ograah well ... i was suppposed to hop on a train now ... but since i missed that anyway i can as well split the patches ...11:26
pittiogra: ah, debdiff is relatively small, so don't bother for now11:26
ograthey are all separately in ltsp upstream bzr anyway11:29
ogra(or will go there for the ones tat arent yet)11:29
* ogra hugs pitti 11:30
Mirvcalc: are you aware that the OOo in hardy-proposed breaks (deinstalls) a) OOo localizations b) openoffice.org-voikko extension (part of language-support-fi)11:38
Mirvbug 236010 has been seemingly filed11:40
ubottuLaunchpad bug 236010 in openoffice.org "Broken packages with -l10n, -help and language-support" [Undecided,Confirmed] https://launchpad.net/bugs/23601011:40
norsettopitti: do the archive admins use a tool to check to which section (universe/multiverse/etc.) a package should pertain?11:40
cjwatsonnorsetto: yes11:43
cjwatsonnorsetto: packages are seeded, seed dependencies are expanded, discrepancies are reported in the component-mismatches.txt output11:43
norsettocjwatson: ah thanks! I'm asking because of bug 214727, I'm wondering if I should do something more (or less ...)11:44
ubottuLaunchpad bug 214727 in tovid "mplayer should be in universe" [Wishlist,Confirmed] https://launchpad.net/bugs/21472711:44
cjwatsonoh, we don't have any automatic tracking of universe vs. multiverse11:45
cjwatsonI'm afraid that's just done by bug reports11:45
cjwatsonnothing more needs to be done for that bug though11:45
norsettocjwatson: ok, thx11:45
Mithrandircjwatson: in hardy, the highlighting done when searching in man pages seems off; do you have a bug about this or should I report it?11:50
cjwatsonbug in whatever pager you're using11:50
cjwatsonI've seen that with less too, but not got round to reporting it11:50
Mithrandirare you sure it's not man-db?11:50
cjwatsonyes11:50
Mithrandirok11:50
cjwatsonwell, reasonably11:50
Mithrandirit's less, yes.11:50
cjwatsonnothing relevant in man-db has changed11:50
Mithrandirgrep seems to highlight the right bit, so I suspect you're right.11:51
cjwatsonman does pass some funky options to less, so you might not see it in other documents11:51
cjwatsonthose options last changed in 200211:52
Mithrandirheh, ok.11:52
emgentmorning11:55
brycehttp://www.oooninja.com/2008/05/openofficeorg-getting-faster-benchmark.html11:56
bryce"In conclusion, OpenOffice.org is generally getting slower with each release. However, startup performance has made great improvements, the performance losses are relatively small, advances in new computer hardware are more than making up the loses, and OpenOffice.org continues to mature with new features."11:56
* cjwatson vomits all over http://bazaar.launchpad.net/~ubuntu-core-dev/net-retriever/ubuntu/revision/34911:58
cjwatsonhow many impediments could there possibly have been to a simple bug-fix11:58
Mithrandircjwatson: why not just add --compare-versions to udpkg?12:01
cjwatsondon't feel like doing that for hardy-proposed ...12:01
Mithrandiroh, good point.12:01
MithrandirI thought it was for i*12:01
Mithrandirwhy vergt, btw?12:02
cjwatsondon't need a generalised vercmp there and I don't want to encourage anything else to rely on it12:06
cjwatsonI'll look at adding it to udpkg though12:06
loolpersia, mvo: update-manager-hildon needs python-vte; would one of you please confirm this is the only missing bit and add it to the u-m trunk?12:23
persialool: Looking now...12:25
loolFor some reason, sudo is borken in one of my envs, can't test directly12:27
ograhostname issues ?12:28
pittiseb128: ah, seems I'm not the only one with this problem: bug 22799412:28
ubottuLaunchpad bug 227994 in gnome-screensaver "password not recognized after suspend" [High,In progress] https://launchpad.net/bugs/22799412:28
loologra: Could be, not quite sure what the rules are; u-v-b seems to create a 127.0.1.1 ubuntu.$mydomain ubuntu in /etc/hosts and hostname reports ubuntu12:31
ogralool, try dropping ubuntu.$mydomain from that line and just keep ubuntu12:31
ograthough sudo tells you if the prob is caused by gethostbyname() if its realy a resolving issue12:32
loolCan one loopmount qemu qcow2 files?12:34
* lool lunch &12:35
ograpitti, i saw some complaints about keymaps not being properly re-initialized after resume on the ubuntu-users ML since we switched to pm-utils ... that might be related12:37
pittiogra: I tried both en and de layouts, that's not it12:37
ograeven though its weird that they say "leave message" works as expected12:37
pittiogra: the same happens with src/test-passwd12:38
pittiand I verified in gdb that the password is correct12:38
pitticurrently gdb'ing it12:38
ograweird12:38
pittiogra: you don't need a password for that12:38
pittiogra: 'login as another user' works, too (but that's gdm again)12:38
ograright, not gss12:38
pittiogra: so the g-s lock dialog works for you?12:38
ogralet me try12:38
pittignome-screensaver-command -l12:39
pittior ctrl+alt+l or so12:39
pittiMay 30 13:30:58 donald gnome-screensaver-dialog: pam_unix(gnome-screensaver:auth): conversation failed12:39
pittiMay 30 13:30:58 donald gnome-screensaver-dialog: pam_unix(gnome-screensaver:auth): auth could not identify password for [martin]12:39
pittiis what I get12:39
ograyes, works fine after suspend12:39
pittiogra: ok, thanks for trying12:39
ogrado you have an upgraded system ?12:40
ograthis laptop was a new hardy install12:41
pittiogra: both my desktop and my laptop are fresh hardy installs, and it happens for both12:41
ogrameh12:41
ograoh12:41
ogramy auth log looks itrestingly different12:42
ograMay 30 13:39:31 osiris gnome-screensaver-dialog: gkr-pam: unlocked 'login' keyring12:42
ograsmells like a communication error between gss and gkr12:43
pittihm, it doesn't use unix_chkpwd??12:44
pittihow the heck is it supposed to be able to verify my password then?12:44
pittiit calls PAM directly, as my user12:44
pittiwithout shadow privileges12:44
pittiogra: is /usr/bin/gnome-screensaver root:root 0755 for you?12:45
seb128pitti: does it happen every time for you?12:45
pittiseb128: yes, it has never worked in hardy12:45
pittifirst thing I do after login is 'killall gnome-screensaver'12:46
seb128urg12:46
seb128we would have received lot of complain if that was happening to everybody for sure12:46
seb128do you configure something special in your boxes?12:46
pittiright, that's why I didn't treat it as OMGPONIES12:46
pittiseb128: not that I know of (pam-wise)12:47
pittistill debugging12:47
seb128pitti: I can get debug informations on my working systems if you need something to compare12:47
pitti/* #undef PASSWD_HELPER_PROGRAM */12:47
pitti^ in my config.h12:47
pittiunix_chkpwd can be passed as a configure argument12:48
ograogra@osiris:~/Devel/packages/ltsp-5.0.40~bzr20080212$ ls -l /usr/bin/gnome-screensaver12:48
ogra-rwxr-xr-x 1 root root 134532 2008-04-09 17:06 /usr/bin/gnome-screensaver12:48
pittibut isn't12:48
mantienahello all12:48
pittiseb128: I'm still stunned how g-s-s is supposed to verify passwords without using unix_chkpwd and without being sgid shadow12:50
pittiseb128: do you happen to have a built gnome-screensaver tree? does src/test-passwd work for you?12:50
seb128pitti: trying12:50
mantienacjwatson: I have some questions about ubiquity's netboot support - do you have some time to answer ?12:51
mantienaI found, that blueprint https://wiki.ubuntu.com/Ubiquity/Automation is implemented already12:52
pittiseb128: ah, nevermind; it does run execve("/sbin/unix_chkpwd", ["/sbin/unix_chkpwd", "martin", "nullok"], [/* 0 vars */])12:52
cjwatsonmantiena: depends on the questions, I'm trying to meet 8.04.1 deadlines at the moment12:53
seb128pitti: the test-passwd thing works fine here12:54
mantienacjwatson: So, I'm asking: there is one important (for me) sentence in that blueprint: "To support netbooted installations, we will ensure that the existing casper support for netbooting works properly (it is believed to have some bugs)..."12:54
pittiseb128: ok, thanks; mind to try something else?12:54
cjwatsonmantiena: I'm afraid I'm not up to date on that; I suggest asking evand12:54
seb128pitti: I'm happy to debug it with you12:54
pittiseb128: echo 'yourpasswd' | /sbin/unix_chkpwd yourlogin nullok12:54
cjwatsonhe was responsible for that specification12:54
pittiseb128: while watchign tail -f /var/log/auth.log12:54
pittiMay 30 13:54:15 donald unix_chkpwd[29021]: check pass; user unknown12:54
pittiMay 30 13:54:15 donald unix_chkpwd[29021]: password check failed for user (martin)12:54
pittiso, unix_chkpwd itself seems to be broken12:55
mantienacjwatson: ok, so, you don't know if casper and ubiquity would work if booting from network (on computers without CD/DVD drivers) ?12:55
pittiaaargh *headdesk*12:56
pitti-rw-r----- 1 root root 1470 2008-05-30 13:55 /etc/shadow12:56
seb128pitti: same error12:56
cjwatsonmantiena: not without trying it, no12:56
mvolool: added to trunk and uploaded to intrepid already :)12:56
seb128pitti: it's root shadow 640 on my install12:56
pittiright, as it should be12:56
seb128pitti: any idea why it's different for you?12:57
elmohmm12:57
mantienacjwatson: thanks for info, I will talk with evand12:57
mantienaevand: hi, are you online ?12:57
pittiseb128: yes, indeed I have; I blame my 'restore my configuration from bzr after install' script, which probably doesn't restore the permissions12:57
pittiseb128: bzr doesn't save group membership12:57
pittiseb128: so, sorry for the noise12:57
seb128ah ok12:58
seb128iz pitti bug12:58
seb128;-)12:58
pittiworks fine now12:58
persiamvo: The other thing I found looking it over is that it needs the versioned dependency on update-manager-core to get the more flexible DistUpgrader12:58
elmopitti: thanks very much, that just helped me fix one of the admin pcs12:59
pittielmo: hah, /etc in bzr, too? :-)12:59
pittiin fact it's not due to bzr, it's due to my 'postinst-setup' script which only adds the real users from teh backup to shadow, but keeps the system bits13:00
elmopitti: no, someone messed up group ownership when they were r00ting it13:01
charliecbi want to download my photos from an canon ixus  80 with nautilus. since gnome 2.22, there should be an option for gphoto2. enter in nautilus gphoto2:// . but that doesn't work. i can import photos, but i can't use nautilus to see all photos from the camera. see http://library.gnome.org/misc/release-notes/2.22/index.html.de. is gphoto2 not compiled into nautilus?13:10
pittidpkg: ../../src/packages.c:265: process_queue: Assertion `!queue.length' failed.13:11
pittiAborted (core dumped)13:11
pittiyay intrepid's dpkg13:11
seb128charliecb: it's not compiled in the hardy gvfs no13:12
cjwatsonpitti: test case?13:12
cjwatsonthough, er, maybe after 8.04.1 deadline13:12
charliecbseb128: do you know why not?13:12
pitticjwatson: trying to construct one13:13
seb128charliecb: the current applications doesn't understand the gphoto uris, so you can't open an image by clicking on it and the gvfs mount lock the access to the device so you can't use an another application to download those13:13
charliecbseb128: ok. thx13:14
seb128you are welcome13:14
pitticjwatson: ah, seems to happen on "dpkg: too many errors, stopping"13:15
ograAmaranth, hey13:17
Amaranthhey13:17
pitticjwatson: filed as bug 236047; I try it in sid now, too13:19
ubottuLaunchpad bug 236047 in dpkg "dpkg: ../../src/packages.c:265: process_queue: Assertion `!queue.length' failed." [Undecided,New] https://launchpad.net/bugs/23604713:19
pittiyup, it's in sid, too13:20
ograAmaranth, do you mind me taking over packaging etc for wilow during intrepid or do you prefer to be my upload bitch ? i added a branch to the upstream project which already carries a good bunch of changes13:21
cjwatsonpitti: that's a relief13:21
cjwatsonof sorts13:21
Amaranthogra: go ahead, do whatever you want with it13:21
pitticjwatson: ah, debian bug 483655; I'll send more information to that13:21
ubottuDebian bug 483655 in dpkg "../../packages.c:265: process_queue: Assertion 'queue.length' failed." [Important,Open] http://bugs.debian.org/48365513:21
ograAmaranth, thanks, i thought so, just didnt want to just steal it away without asking :)13:21
=== fta_ is now known as fta
loolpersia: Hmm not sure but did you push update-manager to ppa?  I don't see it13:53
=== evalles_ is now known as effie_jayx
Riddellpitti, doko: whole bunch more MIRs for you there14:20
pittiRiddell: yeah, just saw them14:22
pitticalc: is bug 220911 actually a problem in writer2latex? If not, we should remove the milestone from the task and set it to invalid14:22
ubottuLaunchpad bug 220911 in openoffice.org "Maintainer scripts of openoffice.org-writer2latex fail" [High,In progress] https://launchpad.net/bugs/22091114:22
pittimvo: while I agree that bug 231805 is primarily a j2se bug, u-m should handle this a little more gracefully ideally?14:26
ubottuLaunchpad bug 231805 in j2se1.4-i586 "error reported during Hardy Heron update" [High,New] https://launchpad.net/bugs/23180514:26
pittimvo: oh, it doesn't actually crash u-m, it just reports the error14:27
pittiso that's fine14:27
pittidoko, mvo: bug 225927 is milestoned for .1, but has no assignee; is it important enough to deserve the 8.04.1 milestone?14:29
ubottuLaunchpad bug 225927 in python-central "python2.4-minimal fails on removal ;  chillispot upgrade fails to upgrade" [High,New] https://launchpad.net/bugs/22592714:29
pittiseb128: do you think bug 206583 is really 8.04.1-worthy? I can reproduce it, but tricky to debug14:51
ubottuLaunchpad bug 206583 in gnome-system-monitor "System Monitor crashes when changing nice value of process" [Medium,Confirmed] https://launchpad.net/bugs/20658314:52
pittibut it doesn't seem terribly important to me14:52
seb128pitti: no, I just milestoned it as a target of opportunity in case there was an easy fix available, slangasek did open the hardy task14:53
pittiseb128: so you're ok with unmilestoning?14:53
seb128pitti: yes14:53
seb128pitti: or delay to 8.04.214:54
cpufreakwin 2714:54
cpufreakmeh sorry :)14:54
seb128pitti: we should get a 8.04.2 milestone ;-)14:54
pitticjwatson: ^ can you create one?14:54
cjwatsonpitti: done, with a rather notional target date14:55
seb128cjwatson: thanks14:55
pitticjwatson: thanks14:55
pittiseb128: bug 196277 sucks, though15:02
ubottuLaunchpad bug 196277 in xserver-xorg-input-keyboard "[hardy] keyboard layout switching shortcut doesn't work after reboot" [High,Confirmed] https://launchpad.net/bugs/19627715:02
* pitti grabs two more milestoned bugs and sits in a corner for hacking15:03
* Hobbsee throws pitti a gummy bear so he hacks faster.15:04
pittiyummy15:04
seb128pitti: right, iz xorg bog though15:04
seb128pitti: which ones are you working on?15:04
pittiseb128: iz task inflation, too15:04
pittibug 22900015:04
ubottuLaunchpad bug 229000 in ntfs-3g "random file corruptions in ntfs-3g < 1.2506" [Undecided,In progress] https://launchpad.net/bugs/22900015:04
pittiand bug 20679015:04
ubottuLaunchpad bug 206790 in dpkg "Deprecated call in dpkg-dev to init method in Dpkg::Changelog::Entry" [Medium,In progress] https://launchpad.net/bugs/20679015:04
seb128pitti: I've decided that I hate bug tasks now15:04
pittiseb128: lol15:04
Hobbseehaha15:04
pittiseb128: s/ task//, certainly? :-)15:05
Hobbseeseb128: i'm sure you can write a greasemonkey script to rename them.15:05
seb128thanks to people who open bugs having a zillion tasks15:05
seb128ie, "not using correct ubuntu maintainer information" or "need rebuild for perl transition"15:05
ograyeah15:05
seb128you get a mail every time somebody add a comment or change a task15:05
seb128when your task as already been fixed and you don't care about the bug15:05
Hobbseeseb128: you can actually get out of those.  you can change the project to 'ubuntu', rather than the package that you're getting mail for.15:06
seb128Hobbsee: I did that but that's an ugly workaround15:06
seb128Hobbsee: and can have several tasks on "ubuntu"?15:06
Hobbseeseb128: i know.  there isn't a good solution, short of not filing bugs that way15:06
Hobbseeseb128: unsure15:06
* ogra wonders why his bugs have suddenly a ton of "also notified" entries15:06
ogramakes me feel observed somehow15:07
seb128pitti: no, I've nothing against bugs, I don't get flooded by other people bugs activity ;-)15:07
cjwatsonyes, people just shouldn't file bugs that way with a zillion different tasks15:07
cjwatsontasks are for multiple bits of the same problem15:07
seb128maybe a motu workflow issue15:08
cjwatsonthe way these are being used are different problems that happen to have basically identical fixes15:08
seb128dholbach: ^15:08
cjwatsonI should have objected to this use when I first saw it, rather than now that it seems to be entrenched among a small subset of people15:09
pittiI still prefer them for things like "these 5 packages need a rebuild for libfoo transition"15:15
pittiit makes it so much easier to see the status15:15
geserI need some help with the perl 5.10 transition, I found a build-dependency loop :( how to proceed?15:19
Amaranthgeser: ?15:20
geserlibmodule-build-perl build-depends on libextutils-cbuilder-perl which build-depends on libmodule-build-perl15:20
geserand neither can currently be installed15:20
Amaranthfun15:20
pitticr3, cjwatson: any idea how to reproduce bug 206790? alien works fine for me, and the example POD in /usr/share/perl5/Dpkg/Changelog/Debian.pm is absolutely, totally useless and wrong15:28
ubottuLaunchpad bug 206790 in dpkg "Deprecated call in dpkg-dev to init method in Dpkg::Changelog::Entry" [Medium,In progress] https://launchpad.net/bugs/20679015:28
cr3pitti: err, didn't soren write patch that code?15:29
cr3pitti: I initially encountered that problem while attempting to install the lsb suite, need a url?15:29
cjwatsoncr3: the bug log might help resolve confusion here15:30
pitticr3: anything that demonstrates the problem; I can't reproduce it15:30
pittiand for a hardy SRU for dpkg I insist on something I can test :)15:30
cjwatsonpitti: joeyh's comment implied that asking it to parse a changelog that had old-format entries at the end would do it15:31
cjwatsonoff the top of my head, man-db is one such15:31
pittiright, finding that shouldn't be a problem15:31
pittiso now I just need to find documentation how to actually use that beast15:31
pitti/usr/share/perl5/Dpkg/Changelog/Debian.pm is from dpkg-dev, while the POD in that file uses use Parse::DebianChangelog, which is libparse-debianchangelog-perl15:32
cjwatsonaha15:32
pitti(separate implementation)15:32
cjwatsonpitti: unpack man-db15:32
cjwatsonpitti: run 'dpkg-parsechangelog --all'15:32
pittiaaaaah15:32
pitticjwatson: thanks a lot15:32
cr3pitti: I'll append the steps to reproduce to the bug report, one moment...15:34
pitticr3: I did that already now; thanks15:34
dholbachseb128: no, there's no MOTU process that says something like that15:37
cr3pitti: excellent, need anything else from me then?15:38
pitticr3: that's fine, no15:38
seb128dholbach: ok, some have just decided that's a good idea apparently then and decided to use it to organize transitions, etc15:39
dholbach*nod*15:39
pittiwow, http://ntfs-3g.org/pjd-fstest.html is really nice for testing patches to packages like ntfs-3g15:42
evandnice!15:44
cjwatsonnice15:44
pittithat's exactly what I was looking for for testing fakechroot15:47
loolI wonder whether it would have catched the unionfs regression with hard links, probably15:47
ograi bet it would expose 20 new bugs in unionfs ... DONT RUN IT ON THERE, OMG ! :)15:48
geserpitti: who do I need to ask to get libextutils-cbuilder-perl and libmodule-build-perl manually rebuild on the buildds? infinity?15:52
sistpoty|workseb128: well, I did use tasks for a transition once (though that sucked, since LP rendered it very slowly). what's your problem with that?15:52
pittigeser: infinity or lamont, yes; what's up with them? a no-change upload won't help?15:52
geserlibmodule-build-perl build-depends on libextutils-cbuilder-perl which build-depends on libmodule-build-perl15:53
geserpitti: ^^15:53
lamontgeser: you ask infinity15:53
lamontpitti: no clue...15:53
* norsetto now understands why infinity nick is infinity15:53
lamontnorsetto: to be fair, his work day starts in just over an hour..15:54
emgenttseliot: o/15:55
tseliotemgent: hey15:55
lamontgeser: if there isn't a way to do multiple uploads and have it happy (please wait for all 7 architectures if you do that...), then you're stuck asking the ubuntu OSA (infinity) to "halp make better"15:55
martiihi guys I got no answer on #ubuntu15:55
martiiso I hope I can ask one question15:56
martiican I?15:56
norsetto!ask | martii15:56
ubottumartii: Please don't ask to ask a question, ask the question (all on ONE line, so others can read and follow it easily). If anyone knows the answer they will most likely answer. :-)15:56
lamontmartii: that was a very simple question, and yes, you could. :-)15:56
martiihttp://library.gnome.org/users/user-guide/2.22/user-guide.html#nautilus-accessnetwork15:56
martiihardy uses 2.22 right?15:56
martiiso where is NFS gone?15:56
lamontdid you install nfs?15:57
martiias well nautilus doesn't recognize nfs://15:57
geserlamont: thanks, will try that first15:57
martiilamont: installed libs for nfs-client15:57
martiilamont: don't tell me I need to become nfs server to access nfs shares :)15:57
lamontmartii: well, that expends my knowledge on the subject. :-(15:57
seb128sistpoty|work: my issue is that your mail every people who are watching a component which has a task open there15:57
lamontmartii: you need to have /sbin/mount.nfs on the system15:58
lamontwhich should be in nfs-common, iirc15:58
martiilamont: I installed nfs-common and portmap15:58
seb128sistpoty|work: which means if I fix the task I'm watching I get zillions of bugs about things I've no interest in every time somebody touch one of the zillion other tasks I don't care about15:58
martiilamont: all according to docs15:58
seb128zillion mails rather15:58
martiiwhich mount.nfs15:59
martii/sbin/mount.nfs15:59
martiilamont: so it's there15:59
sistpoty|workseb128: ah, I see. I guess it shouldn't be called "task" then *g* (and maybe mails shouldn't be sent to people subscribed via packages, which are already listed fixed in a task, though I'm not entirely sure about that)15:59
cjwatsonsistpoty|work: tasks should be used when it's a single problem that just needs coordinated fixes in multiple places16:00
seb128sistpoty|work: you might want get comments about a bug you fixed in case the fix is not correct, etc16:00
cjwatsonsistpoty|work: they aren't for when you've seen the same bug in 20 different packages16:00
cjwatson(IMO anyway)16:00
seb128sistpoty|work: there is no interest to list tasks in the transitions scenario, it's easier to file bugs and tag those16:00
lamontcjwatson++16:01
cjwatsonto put it a different way: if the fixes required would be independent, they should be separate bugs16:01
martiilamont: https://bugs.launchpad.net/ubuntu/+source/nautilus/+bug/4336316:01
ubottuLaunchpad bug 43363 in gnome-user-docs "network: should list NFS shares too" [Low,New]16:01
martiilamont: looks like this bug to me16:01
sistpoty|workhm... good point. (though I guess a transition would equally qualify as a problem, that needs coordinated fixing in multiple packages)16:01
martiilamont: but as well nobody interested to fix it?16:02
seb128martii: nautilus doesn't do nfs16:02
cjwatsonfor the mailbomb reason alone I think multiple bugs are a more practicably reasonable approach16:02
sistpoty|workhowever, as I wrote already, the ui just renders really slowly for many tasks, so I definitely won't be abusing it for this :)16:02
martiiseb128: yep but It did16:05
martiiseb128: as well as gnome 2.22 docs show there is NFS support16:05
seb128martii: not in ubuntu16:05
martiiseb128: I noticed ;)16:06
lamontseb128: why'd we drop that?16:06
seb128lamont: we didn't, I'm not sure any GNOME 2 version ever did that16:06
lamontah, o16:06
lamontk16:06
martiiseb128: so why is it in docs?16:06
martiiseb128: I'm sure it worked16:07
seb128because not a lot of people are interested in updating documentation16:07
ograwishful thinking of doc writers ?16:07
martiiseb128: as I have used it alot16:07
seb128and some very old GNOME versions might have had crackish nfs support16:07
martiiseb128: nfs:// worked the same as smd:// in nautilus16:07
martiiseb128: nfs:// worked the same as smb:// in nautilus16:07
seb128martii: that was GNOME1?16:07
martiiseb128: when ? year ago?16:08
martiiseb128: maybe that was gnome-vfs16:08
seb128martii: http://bugzilla.gnome.org/show_bug.cgi?id=32810716:08
martiiseb128: and now you got gnome-vfs216:08
ubottuGnome bug 328107 in Module: (other) "No nfs:// support" [Enhancement,Unconfirmed]16:08
ogramartii, its called gvfs, gnome-vfs2 is the older one16:08
seb128gnome-vfs2 and gnome-vfs are the same thing16:09
martiiogra: whatever16:09
martiiogra: I can't understand why new version is worse :)16:09
martiiogra: you upgrade to net LTS and things not work :)16:09
martiiogra: you upgrade to next LTS and things not work :)16:09
seb128martii: you have one thing that almost nobody used and which was buggy which has been dropped and ton of other improvements16:10
martiiseb128: yep but as you can see16:10
martiiseb128: this bug is still unconfirmed16:10
martiiseb128: so many people confirmed a problem16:11
seb128gnome-vfs is not maintained, they are working on gvfs now16:11
seb128upstream doesn't care about confirming bugs, the doesn't make a difference, they know it's a wishlist but have other things to do16:11
seb128and a nfs gvfs backend is a low priority thing, most people using nfs use system mounts for that16:12
martiiseb128: I understand I could do that as well16:12
martiiseb128: but it's nice to mount stuff only when you need it16:13
martiiseb128: as NFS tends to lock up my machine when I loose connectivity16:13
seb128use automount?16:13
ogrause sftp16:13
martiiseb128: yep that's an option16:13
martiiogra: it's netapp filesever no sftp ssh16:13
ograless server overhead and supported in nautilus16:14
ograsad16:14
seb128anyway the option was to work on improving ssh, ftp, smb supports or write a nfs backend16:14
martiiogra: nfs only (there is smb but of course nautilus is unable to pass my LDAP username and password correctly)16:14
seb128almost everybody prefer to have better support for the common backends16:14
seb128but you are welcome to write a gvfs nfs backend if you think that's needed16:14
martiiseb128: I know and understand that16:14
martiiseb128: if it's C++ foreget it :)16:15
martiiseb128: if it's C++ forget it :)16:15
ograC16:15
martiiseb128: I can only do python :)16:15
seb128it's C16:15
ograplain and simple actually16:15
martiiC is more likely I could remind myself16:15
martiiseb128: any links to docs?16:15
martiiseb128: if developers will be happy to apply my patch or extension16:16
seb128http://library.gnome.org/16:16
martiiseb128: I had problems before16:16
seb128you can mail the gvfs upstream list16:16
martiiseb128: they might start talking like you that people don't need nfs16:16
martii;)16:16
seb128but I don't expect a nfs backend to be easy to write16:16
seb128well, they will not likely put efforts into it no16:16
seb128but if you send a work patch, that's going to be difficult though, that's not trivial coding16:17
martiiseb128: http://library.gnome.org/devel/references16:17
martiino reference to gvfs16:17
lamontjust remember that NFS stands for Not a File System16:18
ogramartii, they wont tell you nfs isnt needed if you send code :)16:18
lamont(per posix, with interpretation)16:18
ograheh16:18
martiiogra: many times developers did that16:18
ograwell, ususally developers are happy if you send code and help you to improve it is what is my experience16:18
martiiogra: moslty they say they don't want to brake their code with untested stuff when priority for extension is not critical16:18
geserpitti: does it needs MIRs to get some perl modules from universe to main which are needed for build-dependencies?16:19
seb128martii: no real documentation out of the source code for now, gvfs is pretty new and it's really a rush, people are busy trying to get it work right now and didn't stop to write documentation16:19
pittigeser: at least MIR bugs; trivial ones don't need a full wiki page16:19
martiiseb128: OK I'll stick with autofs16:19
martiisorry for wasting your time guys16:19
* geser hates the perl 5.10 transition already16:19
ograyou expressed a need16:19
ogranot wasted time, really16:19
martiiogra: :)16:21
martiihttp://www.acis.ufl.edu/~ming/gvfs/ so that's not the gnome thing?16:21
pittigeser: can't say how grateful we all are that you deal with it *hug*16:22
hungerWhich version of the telepathy spec will be supported in intrepid?16:27
geserpitti: http://launchpadlibrarian.net/14685812/buildlog_ubuntu-intrepid-i386.libfile-sync-perl_0.09-4build1_FAILEDTOBUILD.txt.gz Is this a bug in the package or in pkg-create-dbgsym?16:27
pittigeser: it's a bug in the package, since debhelper compat 1 needs debian/tmp; it's a problem with pkg-create-dbgsym in the sense that it is not as forgiving as debhelper about packaging bugs16:29
geserpitti: thanks, will fix the package then16:29
pittiideally we'd fix pkg-create-dbgsym to deal with those16:30
pittiif it builds locally16:30
=== zsoilworker is now known as zSoilworker
=== Zic is now known as Zic__
=== Zic__ is now known as Zic
* pitti kicks rothera again to get back to work16:54
pittiinfinity, cprov: ^ FYI16:54
=== macd_ is now known as macd
Kanohi, is anybody using ICH9R with raid 5 successuflyl?17:22
Keybukok, that's weird17:33
Keybuk"People nearby"17:33
Keybuk O  Keybuk17:33
evandheh17:39
ion_keybuk: What says that? :-)17:39
Robot101Keybuk: you can see yourself in Empathy?17:40
KeybukI can indeed17:41
KeybukI suspect the machine upstairs just woke up for something and logged in17:41
Robot101ah ok, i was about to say sjoerd's here now so you can badger him on #tp if you've found a bug :)17:42
sjoerdI usually use sjoerd on X as my nick in salut to prevent that kind of confusing :)17:43
Keybukif only we could set the status usefully ;)17:44
Robot101Keybuk: what do you mean?17:56
Keybukdoesn't jabber have an extended status bit?17:56
Robot101like the "custom messages" thing in the status dropdown?17:57
ion_Is intrepid beginning to be usable? That is, are there still some huge things like a new major release of glibc coming?18:04
ion_I don’t mind running an unstable distro on this box, but rather wouldn’t upgrade yet if it’s expected to break badly. :-)18:05
geserion_: we are still in the middle of the perl 5.10 transition, but you could have luck18:09
ion_That sounds like something i might even be able to help with.18:11
fdeHello, several people in #ubuntu are having issues due to hardy-proposed currently... it appears that gnome-about depends an explicit version of gnome-desktop-data (2.22.2-0ubuntu1) but today it was upgraded... this removed gnome-panel for many... and I don't see this package in http://archive.ubuntu.com/ubuntu/pool/main/g/gnome-desktop/ and don't have it locally to upload to a ppa...18:46
fdeWhat can be done about this situation?18:46
cjwatsonsounds to me like architecture desync18:47
cjwatsongnome-desktop-data is built on i386 and the same package is used on all architectures18:48
cjwatsongnome-about is built on each architecture separately18:48
fdecjwatson: It is people that have hardy-proposed enabled due to virtualbox I think... seems to be unfortunate timing on their part.18:48
cjwatsonbut they come from the same source package18:48
cjwatsonso when i386 builds out of step with the others, you'll get this happening transiently18:48
cjwatsontell them to wait for a while and try later, and it will go away18:48
fdecjwatson: Some have already lost their panels... just apologize to them?18:49
pochuthe question is, why people who don't check what is going to be removed use -proposed?18:49
cjwatsonhttps://launchpad.net/ubuntu/+source/gnome-desktop/1:2.22.2-0ubuntu2 shows that the new version hasn't quite built everywhere yet, though it should be OK for i386 and amd64 users18:49
cjwatsonfde: why did they allow the upgrader to remove packages in the first place?18:49
pochufde: first ask them to disable -proposed! :)18:49
cjwatsonfde: advise them to reinstall ubuntu-desktop18:49
fdepochu: it is a documented work around for virtualbox, although apparently that documentation didn't mention disabling it when they're done.18:49
fdecjwatson: alright, thanks. pochu definitely will do  :)18:50
pochufde: right, there should be some easy way to install packages from -proposed without enabling it completely...18:50
fdeThanks for the assistance, wanted official word  :)18:50
cjwatsonfde: sounds like they used an option designed for advanced testers and got burned by the consequences; the documentation should definitely be fixed18:50
fdepochu: maybe something like experimental on debian should be enacted?18:50
cjwatsonfde: hardy-proposed *is* a bit like experimental18:51
cjwatsonalthough not so much18:51
cjwatsonwe have plenty of things that fill similar niches; however if people get told to use them then what can we do?18:51
fdecjwatson: still, even when enabled, people don't need to be using it always unless they're aware of what that entails.18:51
pochucjwatson: yesterday we had the same problem with people removing firefox via update-manager, due to the "Partial upgrade" dialog in it18:51
pochuso while I think they shouldn't be running -proposed at all, update-manager could be more verbose there18:52
cjwatsonhardy-proposed contains non-QAed updates that may break your system, because they have not yet been verified18:52
cjwatsonit's that simple18:52
pochuof course18:52
cjwatsonso while we welcome people helping us with the QA process, we can't really also have them coming to us and complaining that things were set up wrong because their system broke :-)18:53
fdecjwatson: I think it's safer to make people type 'sudo aptitude -t hardy-proposed install whatever' and document how to enable it always _no_where_... there's been a few issues like this, and people inevitably want things like firefox and virtualbox to work.18:53
cjwatsonI agree that there is room for improvement in update-manager18:53
cjwatsonI know that mvo knows about this, since he mentioned it yesterday (I think)18:53
ion_After this operation, 2179MB disk space will be freed.18:53
fdecjwatson: Even just having it deselect everything originating from hardy-proposed, and forcing users to select what they actually want would be better... if that is even possible?18:54
cjwatson-> mvo18:54
cjwatsonthough I was under the impression it already did that; could be wrong18:54
cjwatsonoh, actually, no, I'm thinking of something else18:54
cjwatsonremember that -proposed was instituted in response to a problem that slipped through QA that caused X not to start for some people18:55
pochuI reported bug 152335 some time ago to try to reduce the number of users running -proposed18:55
ubottuLaunchpad bug 152335 in software-properties "Remove -proposed checkbox in Updates, or warn of 'not stable' updates" [Undecided,New] https://launchpad.net/bugs/15233518:55
fdecjwatson: Yes, I actually discussed that at a LUG a couple nights ago  ;)18:56
cjwatsonregardless of whether you deselect things by default, people will try selecting them out of curiosity - isn't X not starting much worse than firefox going away?18:56
pochualthough that may not be a good solution at all :)18:56
ograthe software sources dialog doesnt really tell you that its dangerous to enable it18:56
cjwatsonthus the most important thing is not to advise people to use -proposed who can't cope18:56
ograthe wroding should be better18:56
ogra*wording even18:56
cjwatsoneven if that's the only location for a bug fix in progress, it's more important to have it go through QA before wide deployment18:56
fdecjwatson: It's hard to ensure that in private mediums though... so maybe a technical solution is better...18:56
fdeShould I file a bug with some propositions real quick just to remind?18:57
cjwatsonsure, better than talking to me about it since I'm not the relevant developer ;-)18:57
cjwatsonshould go on apt I think18:57
fdecjwatson: alright, thank you for your recommendations... if that doesn't fix it for users (reinstalling ubuntu-desktop) then I'll likely be back... or maybe I'll break it and try to fix it locally  :P18:58
fdeOn a more positive note: Thanks for all the hard work, you guys are awesome  :D18:59
cjwatsonit will depend on the architecture they're using18:59
cjwatsonpowerpc doesn't seem to have caught up yet18:59
fdecjwatson: I'll try to look into that too... luckily qemu can emulate a ppc system - although it'll be slow as molasses.19:00
fde(only from a user perspective though... see if I can come up with another solution if that doesn't fix things)19:01
mvocjwatson: we could disable partial upgrades at all in stable, they are only required when things need to be removed and Ithat should never be the case for stable19:05
romhi19:27
alex1hi guys. what does it mean when a bug's status is triaged in launchpad?19:28
MethoxypropanHello :)19:33
MethoxypropanI've got a BCM94311 rev 2 wireless network card and read about a patch to get it working with the native bcm43xx-fwcutter. Where can I download the patched driver?19:35
Kopfgeldjaegerbug #21438619:35
ubottuLaunchpad bug 214386 in linux "bcm94311 rev 02 not detected at boot time" [Undecided,New] https://launchpad.net/bugs/21438619:35
Methoxypropanthanks Kopfgeldjaeger19:35
romhi, I reported a bug in compiz with a patch this morning : https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/235982 (my first patch \o/ )19:37
ubottuLaunchpad bug 235982 in compiz "/usr/bin/compiz : 1 bug and 1 missing setting" [Undecided,New]19:37
romdo you think it could be applied19:37
Methoxypropanwhere can i download the patch from bug #21438619:41
ubottuLaunchpad bug 214386 in linux "bcm94311 rev 02 not detected at boot time" [Undecided,New] https://launchpad.net/bugs/21438619:41
elmopitti: can I turn off the python apport stuff19:48
MethoxypropanHi dev's ;)19:50
spearthi, can init.d rules, or udev entries prevent computer shutdown?19:52
mvorom: the fist part of the patch looks good, I'm not sure about the "-l" bits19:54
rommvo: why that?19:54
romnvidia-settings -l19:54
mvorom: it seems like this is something that we shouldn't enforce in the compiz script, I don't need it and it gives me a ugly flickering19:55
romto load the settings (if you don't, nvidia settings are not applied)19:55
romsince a long time19:55
Methoxypropancan anybody help me with bug #21438619:55
rom(antialiasing, anisotropic, sync to vblank)19:55
ubottuMethoxypropan: Error: Could not parse data returned by Launchpad: The read operation timed out (https://launchpad.net/bugs/214386/+text)19:55
mvorom: isn't that only required if the nvidia-settings panel is used and shouldn't it be used in a place that is more generic that the compiz start script? like .xinitrc or something similar?19:56
Methoxypropani need that patch to get my wireless working :(19:56
romhmmm...19:56
mvorom: actually, if its something that should always be run, I think it should be part of the nvidia-settings package and go into /etc/X11/Xsession.d maybe19:56
* mvo is not a expert for the binary driver though19:56
romshould not always be run, only before compiz lunch19:57
romI guess19:57
rom Load the configuration file, send the values  specified  therein19:57
rom              to  the X server, and exit.  This mode of operation is useful to19:57
rom              place in your .xinitrc file, for example19:57
romok19:57
mvobefore compiz or anything that uses 3d I guess? like games etc?19:57
mvotjaalton: ---^ do you have a opinion on the right place to put nvidia-settings -l ?19:58
romso, it should be added in nvidia-settings package, no?19:58
romto add it to .xinitrc file19:58
mvoyes, that looks like a better place to me (/etc/X11/Xsession.d probably)19:58
romwhere is .xinitrc?20:00
romI don't find it20:02
romlocate .xinitrc doesn't give any result20:02
roml20:03
romand in which file compiz in launched when starting?20:04
rommvo?20:06
rom?20:09
mvorom: its in the users homedir, but that is not a place where packages can put stuff, this is why I mentioned /etc/X11/Xsession.d20:10
romok, and do you know how compiz starts when system starts?20:11
romin which file20:11
romis it started?20:11
Picirom: It doesnt start when the system is started.20:11
Picirom: It starts after a user logs in.20:11
romin ubuntu, yes20:11
romah20:11
romyes :)20:11
romwhere is it started?20:11
Picirom: Have you looked into mvo's suggestion?20:12
romI have no .xinitrc20:12
Picirom: No one but you mentioned that, see /etc/X11/Xsession.d20:12
romyes I searched20:13
rombut this dir is for putting nvidia-settings -l, I didn't find compiz here20:13
romcat /etc/X11/Xsession.d/* | grep compiz → no results20:13
ion_UUOC ;-)20:14
romno idea?20:16
mvorom: compiz is started via the gnome-wm script by gnome-session. but I it seems to me that the nvidia-settings -l run should be independant of the compiz one20:21
rommvo, yes, I agree nvidia-settings -l should be independant ;)20:22
romwhere is "gnome-wm script by gnome-session"?20:22
romah, gnome-wm is in /usr/bin :)20:25
romthanks20:25
rombut doesn't it slow down the boot time ?20:25
romto load metacity first, then when the user is logged20:25
romto load compiz...20:25
rom?20:25
PiciI dont see that there.20:29
romwhat do you mean (sorry, I'm not english, I don't get it)20:30
rom"I dont see that there"20:30
rom?20:30
Picirom: gmome-wm looks like it calls whatever window manager based on a set of rules, not both wms.20:30
Picianyway, I need to run.20:30
romok, but before running compiz20:31
romwhen you have login screen20:31
romit's metacity, no?20:31
romto run nvidia-settings -l in /etc/X11/Xsession.d, I just have to make a script containing "nvidia-settings -l" with a special filename?20:31
rommvo? could you confirm how to add a script to Xsession.d?20:48
mvorom: that sounds right, it needs to be added to the nvidia-settings package20:49
mvorom: you may talk to the people in #ubuntu-x too20:49
simiramvo: are you still responsible for the update-manager?20:49
rommvo: ok, but do you know how to add a script in Xsession.d? Just put the script "like this", no particular format?20:50
mvosimira: yes20:53
simiramvo: works very well now, I like it!20:56
mvosimira: cool, I like that :)20:56
* mvo hugs simira20:56
calcMirv: yes, i need to make an upload of the openoffice.org-l10n packages that i will be doing in a few minutes20:56
calcpitti: 220911 is a bug in openoffice.org not writer2later afaicr but i haven't added it in yet to the upload since i need to verify the patch works properly (its not a patch in ooo-build)20:58
ion_keybuk: What was this about? :-) * At this rate, I'm going to prove pitti right!21:09
MethoxypropanHello21:38
MethoxypropanWhere do i get a patched bcm43xx driver ?21:39
ion_Note to self: mention http://heh.fi/tmp/recovery-mode-dpkg.patch to mvo21:58
cjwatsonion_: 'if [ -x $python -a -e $script ]' -> 'if [ -x "$python" ] && [ -e "$script" ]' please?22:08
cjwatson(quoting is just good style, and the rules for -a and -o are so twisted and unintuitive they're best avoided)22:09
ion_cjwatson: Will do. I usually do quote pretty much everything, but that case seemed so obvious, since there are ‘2.5 2.6’ right above.22:10
MethoxypropanHello22:10
MethoxypropanWhen will the patched bcm4311xx driver be in the ubuntu repo?22:10
ion_cjwatson: Updated.22:12
Methoxypropanno answer?22:14
=== Amaranth_ is now known as Amaranth
cjwatsonMethoxypropan: which patch are you talking about here?22:18
cjwatsonMethoxypropan: and is there a bug filed about whatever it is?22:19
Gaming4JChi22:19
Methoxypropancjwatson, yes there is a bug report on launchpad...wait..i'm going to search it22:20
Gaming4JCI need some one who can compile a modem driver (I got the source), so I can get online in Ubuntu. :)22:20
Mirvcalc: remember also the rebuild of openoffice.org-voikko, to be published at the same time so as not to force language-support-fi/ooo-voikko deinstallation22:20
Gaming4JCIf anyone could do it, I'd give you much thanks. http://linmodems.technion.ac.il/pctel-linux/pctel-0.9.7-9-rht-6_for_Ubuntu-2.6.15-23-386.tgz22:20
Mirv(really should have automatic testing/stuff about the dependencies etc.)22:21
Methoxypropancjwatson, its bug #21438622:21
ubottuLaunchpad bug 214386 in linux "bcm94311 rev 02 not detected at boot time" [Undecided,New] https://launchpad.net/bugs/21438622:21
Gaming4JC:) ...22:21
* Gaming4JC you can see I cannot get online to compile the tool, so I was wondering if anyone could do it. (would help so much) 22:23
Gaming4JCI'm on Windows now.22:23
Gaming4JC:'(22:24
ChipzzGaming4JC: the reason no-one is answering is because this is not a support channel22:24
Gaming4JCit's development?22:24
Chipzzyes, OF ubuntu22:24
Gaming4JCso not drivers for Ubutnu?...22:24
Gaming4JC:-/22:24
Chipzzwhat you're trying to do is not even development22:24
Chipzzit's compiling22:25
Gaming4JCeww... so where should I go for that? (main support doesn't seem to know)22:25
Chipzzforums or #ubuntu22:25
Methoxypropancjwatson, in the last post the guy sais something about a patch.... but when will the patch be in the repo?22:25
cjwatsonMethoxypropan: "a new patch has been submitted" but no link22:25
Gaming4JCok thanks22:26
Gaming4JCbyes ;)22:26
cjwatsonMethoxypropan: the name in that last comment isn't one I recognise; he's not a member of the Ubuntu kernel team22:26
cjwatson(AFAIK)22:26
cjwatsonso somebody needs to actually provide the patch to the Ubuntu kernel team (or a reference to it or whatever) in order for it to have a hope22:26
cjwatsonI suspect he's perhaps a bcm43xx upstream guy looking at his subscribed bug list22:26
cjwatsonso perhaps what he meant was that a patch has been sent to bcm43xx upstream22:27
Methoxypropancjwatson, http://cantrip.org/bcm43xx-2619.patch22:27
cjwatsona patch from October 2006? surely not22:27
cjwatsonwhatever Larry is describing sounds much more recent than that22:28
Methoxypropanso how long will i have to wait?22:28
cjwatsonthe bug is in far too early a state to be able to say22:28
Methoxypropanok thanks a lot ;)22:29
cjwatsonif you want to help, track down whatever it is that Larry was actually referring to22:29
cjwatsonI do think you will need to look for something that's actually from this year22:30
Methoxypropancjwatson, they said that the patch that made it working has been removed because an othe card didnt work22:31
cjwatsonyes, I read that22:32
cjwatsonah, he sent some patches to kernel-team@22:35
cjwatsonwhich Tim seems to have picked up22:35
Methoxypropancjwatson, so maybe it'll be in the next kernel?22:36
cjwatsonit looks like it22:36
cjwatsonsee bug 19795922:36
ubottuLaunchpad bug 197959 in linux "[Hardy]Recent kernel update to 2.6.24-11 breaks b43 (with bcm4312)" [High,In progress] https://launchpad.net/bugs/19795922:37
cjwatson"There is now a test kernel in my PPA with these 4 patches at http://ppa.launchpad.net/timg-tpi/ubuntu. Please give it a try and report the results."22:37
cjwatsonperhaps you could give that a shot and report results to 197959, assuming that's essentially your bug22:37
Methoxypropanthanks a lot cjwatson22:39
cjwatsonI doubt that's targeted for 8.04.1 at this point (though I could be wrong); it's a bit late - but at least you'll have the PPA kernel to tide you over until then, and if it's working well then it'd be at least in the next round of hardy kernel updates after 8.04.122:39
cjwatson(and, as I suspected above, Larry does indeed seem to be a bcm43xx upstream guy, helping out)22:40
Methoxypropan;)22:43
Methoxypropancjwatson, so i'm going to reboot and hopefully be back again22:44
Methoxypropancjwatson, so its working perfectly ;)23:31
cjwatsonoh good; can you update the bug on that if you haven't done so already?23:32
cjwatson(both the one you originally gave and the one I pointed out)23:32
Methoxypropancjwatson, mhm...can i do it tomorrow? i'm really tired23:32
MethoxypropanWhich bug numbers did the bugs have?23:33
Methoxypropancjwatson, can you remember the bug numbers?23:34
cjwatson22:21 <Methoxypropan> cjwatson, its bug #21438623:34
ubottuLaunchpad bug 214386 in linux "bcm94311 rev 02 not detected at boot time" [Undecided,New] https://launchpad.net/bugs/21438623:34
cjwatson22:36 <cjwatson> see bug 19795923:34
ubottuLaunchpad bug 197959 in linux "[Hardy]Recent kernel update to 2.6.24-11 breaks b43 (with bcm4312)" [High,In progress] https://launchpad.net/bugs/19795923:34
cjwatsonmy IRC client can ;-)23:34
Methoxypropanthanks ;)23:35
Methoxypropanso i should say that it's working on my system and tell the guys from bug #214386 that they can use the kernel from bug #197959 ?23:36
ubottuLaunchpad bug 214386 in linux "bcm94311 rev 02 not detected at boot time" [Undecided,New] https://launchpad.net/bugs/21438623:36
ubottuLaunchpad bug 197959 in linux "[Hardy]Recent kernel update to 2.6.24-11 breaks b43 (with bcm4312)" [High,In progress] https://launchpad.net/bugs/19795923:36
cjwatsonI think so23:39
Methoxypropanbbl [02:00 pm GMT+1] :P23:39
Methoxypropangn823:39

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