/srv/irclogs.ubuntu.com/2011/11/16/#ubuntu-devel.txt

SpamapSI guess so many patches have been applied.. most of upstream's changes are already there. :-P00:00
slangasekSpamapS: heh, no clue I'm afraid00:11
SpamapSslangasek: this is.. the worst merge ever00:12
slangasekmy condolences ;)00:13
SpamapShalf the patches have been upstreamed, half haven't00:13
SpamapSall centered around the same 4 or 5 files. :-P00:13
TheMusoDoes anybody know how often status.ubuntu.com refreshes?00:15
Riddellmterry: libbison-dev and python-scikits.statsmodels accepted, sorry for the delay I'm on holiday this week00:36
SpamapSrmdir: failed to remove `/var/lib/nfs/rpc_pipefs': Device or resource busy00:42
SpamapSslangasek: seen that on trying to upgrade nfs-common?00:42
cjwatsonDaviey: could you please merge libapache2-mod-perl2 from unstable (or tell me I can do it)?  I think http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=636651 is what's causing current build failures with perl 5.1401:27
ubottuDebian bug 636651 in src:libapache2-mod-perl2 "libapache2-mod-perl2: FTBFS with perl 5.14: -D_FILE_OFFSET_BITS stripped from CFLAGS" [Serious,Fixed]01:27
slangasekSpamapS: gnar; yes and no02:00
slangasekSpamapS: I thought I had that upgrade path stable02:00
slangasekSpamapS: do you have nfs-kernel-server installed?02:01
TheMusoAnybody using irssi in gnome terminal running precise? If so, are you able to use alt + 1-0 and alt q-o to switch between channels?02:50
StevenKTheMuso: I've been using Esc-1-0 and Esc-q-o for a long while02:51
StevenKBut I'm on Oneiric02:52
TheMusoStevenK: Right, it worked for me in oneiric. :)03:05
StevenKTheMuso: alt-1 and so on switch terminal tabs for me03:06
TheMusoStevenK: Right, but I have irssi in a separate terminal window on a separate workspace, and since there are no other tabs, irssi's keyboard shortcuts worked.03:07
TheMusoThat was oneiric.03:07
TheMusoSame setup in precise, but no shortcuts to move between channels.03:07
TheMusoStevenK: Turns out something in gtk 3.2.2 in precise changed, requiring a patch to vte.03:43
TheMusoSo back to normality for me at least, and hopefully for other users of vte if the patch gets committed upstream.03:43
TheMusoAt which point I'll upload to precise.03:43
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
SpamapSslangasek: yes04:19
SpamapSslangasek: I have nfs-kernel-server installed, and I'm still unable to rmdir even with it stopped and portmap and idmapd and basically everything stopped04:19
pittiGood morning05:13
pittislangasek: no, none of cups etc. should have increased; the 20111114 images increased by 8 MB compressed due to some duplicated fonts, though; we added fonts-nanum and removed ttf-unfonts-core, but that's 7.5 MB delta only05:15
pittislangasek: yesterday's daily images were back to 702 MB05:15
pittislangasek, micahg: no, I don't have a magic script to check and fix for kernel overrides; in the PPA they are all in "main", and when copying they sometimes randomly go to universe05:17
pittithe kernel team has a cron job which yells at us if there are uninstallable packages, but apparently it doesn't pick this up (it looks at universe, too?)05:17
micahgpitti: I realized after I said that that it was fixed at one point copying from native PPAs to apply overrides, but it seems to selectively break05:26
slangasekpitti: hmm, ok; maybe it was part of something I have here that's not part of the default install then05:38
slangasekguess I'll investigate further :)05:38
pittislangasek: perhaps you can salvage your /var/log/dpkg.log before it gets rotated?05:38
slangasekSpamapS: should only be nfs-kernel-server, gssd, and idmapd that might need to be stopped; nfs-common's postinst not handling the nfs-kernel-server part of that however05:39
pittiI'm not tracking installed size here really, it's well possible that some package grew a lot uncompressed, or some postinst is creating lots of data05:39
slangasekpitti: I certainly have plenty of logs, it's sorting through them to figure out what changed in size that's arduous05:39
slangasekpitti: revert the workaround> yep, my thinking exactly :)06:19
pittislangasek: haven't had tea yet :)06:20
pittihm, seems I lost my ability to target bugs to a release06:57
slangasek!06:57
infinity!!06:57
nigelbNeeds coffee.06:57
pitti"You do not have permission to nominate this bug."06:57
JackyAlcinenigelb: +1 aye06:57
nigelbExpired from some team?06:57
pittinot that I know of06:57
slangasekpitti: what URL?06:58
infinitypitti: Which bug?06:58
pittibug 82841506:58
ubottuLaunchpad bug 828415 in gnome-themes-standard (Ubuntu Precise) "package gnome-accessibility-themes 2.32.1-0ubuntu1 failed to install/upgrade: trying to overwrite '/usr/share/icons/HighContrast/index.theme', which is also in package gnome-accessibility-themes-extras 2.32.1-0ubuntu1" [Medium,In progress] https://launchpad.net/bugs/82841506:58
slangasekpitti: weird06:58
pittiI could target it to precise just fine, though06:58
slangasekpitti: and you didn't accidentally get logged out?06:58
pittinope06:58
slangasekoh06:58
slangasekyou can target but not nominate?06:59
slangasekthat's not new then06:59
infinity"The Ubuntu release manager is Ubuntu Drivers"... I can't wait for all that to get fixed.06:59
pittijust tried to target bug 865199 to lucid, which works06:59
slangasekor do you mean that you can target to precise but not to other releases?06:59
ubottuLaunchpad bug 865199 in apport (Ubuntu Lucid) "apport crashes when /etc/apport/native-origins.d contains any files." [Undecided,New] https://launchpad.net/bugs/86519906:59
slangasekok weird06:59
pittiso either it's special to oneiric or to that bug06:59
slangasekpitti: worked for me07:00
pittislangasek: ok, thanks; at least I have a task to play with now :007:00
micahgpitti: did your original target to release leave you with lucid in the URL?07:00
pittihttps://bugs.launchpad.net/ubuntu/precise/+source/gnome-themes-standard/+bug/82841507:01
ubottuLaunchpad bug 828415 in gnome-themes-standard (Ubuntu Precise) "package gnome-accessibility-themes 2.32.1-0ubuntu1 failed to install/upgrade: trying to overwrite '/usr/share/icons/HighContrast/index.theme', which is also in package gnome-accessibility-themes-extras 2.32.1-0ubuntu1" [Medium,In progress]07:01
pittimicahg: yes, indeed; that might be it?07:01
micahgpitti: yes, I think there's a bug for it07:01
pittiI targetted it to precise before07:01
pittiah, good07:01
infinitypitti: Yeah, need to be on the "raw" bug URL.07:01
infinity(Or the parent task, or whatever you want to look at it as)07:02
pittiat least next time I know where to look07:02
lathiat<broken record risk> security.ubuntu.com is having problems .. is this known / where is the right place to report it?07:03
micahgpitti: I can't seem to find the bug ATM07:04
slangaseklathiat: what's the problem?07:06
lathiatslangasek: well i was having no connection.. now it seems to be working but going slowly.. another guy said the same in #ubuntu so wasnt just me07:06
lathiatnot responding again now for me07:07
slangasek(this is a reasonable place to report it)07:08
lathiatthought that might be the case07:08
* micahg is also getting a 500 error07:09
slangasekmicahg: are you rousing IS or do you want me to?07:09
micahgoops, not a 500 error07:09
micahgslangasek: if you could please, I'm trying to finish some stuff up07:10
micahgit's a connection timed out error07:10
slangaseklathiat: believed to be sorted now07:15
lathiatslangasek: ok thanks07:15
pittiwow, xz FTW; I recompressed langpack-locales orig.tar.gz (3.6 MB), now it's 1.2 MB07:33
slangasekpitti: oh, I found my 200MB; it was being eaten by deleted files belonging to running processes ;)07:44
pittihah!07:44
pitti"Reboot now and win an extra 200 MB"07:44
slangaseknever seen it quite that extreme on a daily update07:45
pittiwho could resist that07:45
pittisabdfl: reducing ccsm unity settings and put the most common one into a "real" settings dialog> +1000!07:51
mvodidn't we had "simple-ccsm" at some point for exactly tihs?07:52
mvothis07:52
didrocksmvo: indeed, but it's not compatible with 0.9 and still, it has too many options07:59
mvodidrocks: aha, yeah, I noticed it got removed08:04
dholbachgood morning08:07
brodercould i get an archive admin to accept the pdns-recursor packages in hardy bin NEW? (it's just a -dbg package that wasn't there before)08:15
* broder is going to have to take a backports break after this to earn his archive admin brownie points back08:16
infinitybroder: Waiting on the PPC build.08:29
* broder looks again08:29
broderoh, huh. i thought everything had finished, but i must have been distracted by the sparc/hppa failures08:30
brodersorry about that, then08:30
infinitybroder: Although, the sparc failure is interesting, since the same version built on lucid/sparc...08:31
infinitybroder: Oh well.  Backports are a big "who cares" in that arena, I suppose.08:31
broderthey were certainly much weird failures than i care about as a backporter08:32
broder*weirder08:32
infinitybroder: Without really looking (and now being unable too, thanks LP), I'm going to guess from parisc/sparc failing that it's endian-related, and ppc will also fail.08:35
infinitybroder: We'll find out, I guess.08:35
infinitybroder: (I scored up the ppc build to see)08:35
infinitybroder: But sparc and parisc also have other oddities in common that aren't shared with powerpc...08:36
* broder nods08:36
infinitybroder: PPC good, accepting all 5.08:49
mvoev: I haven't used usb-creator in a long time, but the test-disk feature with kvm is just such a awsome idea!09:38
Davieycjwatson: Erm, libapache2-mod-perl2 - slangasek touched it last in Precise.  For future, unless i have a "please merge foo" bug opened and assigned to me indicating it's a complex merge and something is in-progress.. i'll never object to a hijack.09:44
DavieyNot quite sure why slangasek did the merge TBH, it was a no-change merge.. :)09:45
DavieyOh no it wasn't, my bad.09:46
=== dholbach_ is now known as dholbach
DavieyI'm a plum.09:46
Daviey(Doing the merge now.)09:47
cjwatsonDaviey: thanks.  I ask because I sometimes get annoyed when people don't ask me so it would be hypocritical not to. :-)09:49
Davieycjwatson: Yeah, you are on my "must ask list" :)09:49
cjwatsonI just end up with a lot of merges I guess, so higher chance ...09:50
evmvo: you can thank kirkland for the idea :)10:31
evbut cheers!10:31
* Daviey looks at component-mismatches, and cries.10:33
cjwatsonDaviey: most of it's due to (AIUI) one package pulling in maven; jamespage said he was going to clean that up10:37
Davieycjwatson: Yeah, i imagine it won't be looked at until next week - he's sprinting.10:38
cjwatsonah10:38
Davieycjwatson: BTW, the testsuite failed for libapache2-mod-perl2 failing the build.  Annoyingly, it worked in local pbuilder before uploading10:38
DavieyAnother sample of pbuilder env not matching buildd :(10:39
* ogra_ wonders what cjwatson did ... i get build failures for x86 images etc but apparently all armel images built just fine tonight .... thats a true novum \o/10:43
cjwatsonDaviey: I noticed the failure, although have no particular ideas to offer about it :-/10:45
cjwatsonogra_: heh10:45
cjwatsonI assure you it wasn't deliberate10:45
ogra_i wonder why though ... armel shouldnt differ in content from x86 and x86 failed on perl stuff10:45
cjwatsonUbuntu desktop x86 images built10:45
cjwatsonthere was an x86 build attempt from last night when the perl transition was at a point when it couldn't possibly have worked; ignore that10:46
ogra_yesterdays didnt ... i got failure mails from 23:48 here10:46
cjwatsonyes, that was somebody unidentified doing a manual build for some reason10:46
ogra_ah, k10:46
cjwatsonmaybe I should put $SUDO_USER in the log :-)10:46
ogra_good idea :)10:46
ogra_though its over anyway, who cares :)10:46
Davieycjwatson: i gave back and it worked, non-determinisitic builds \o/10:47
cjwatsonjoy10:48
=== _salem is now known as salem_
DrakesonAlt no longer sends "Meta" in gnome-terminal since two days ago. Which is frustrating. Alt-f, Alt-b, etc. no longer work.  Is it the same for you?11:42
brendandDrakeson - in precise?11:44
infinityDrakeson: Fix already uploaded.11:47
infinityDrakeson: https://launchpad.net/bugs/89055511:48
ubottuLaunchpad bug 890555 in gtk+3.0 (Ubuntu) "Alt stopped working as Meta in gnome-terminal with gtk+3.0 3.2.2-0ubuntu1" [Undecided,Confirmed]11:48
=== MacSlow is now known as MacSlow|lunch
kirklandmvo: ev: hey guys, thanks for revisiting that idea ;-)13:31
Drakesoninfinity: Thanks. I looked into gnome-terminal and gtk-3.0 in bugs.launchpad.net, and couldn't find anything related. Guess I should have searched more before complaining here.13:45
cjwatson$ change-override.py -c universe -y libperl5.1214:04
cjwatsonwhee14:04
* ogra_ applauds14:06
pitticjwatson: ah, we can't move everything to 5.14 in precise?14:07
pittiI had thought it was NBS now14:08
pittior was that to say "main is done now", and a milestone?14:08
pitti(congrats!)14:08
cjwatsonpitti: precise will be entirely 5.1414:11
cjwatsonpitti: and yes, that was just to say that all the libperl5.12 users in main are now on libperl5.14 - still universe to go14:11
pittinice14:11
cjwatsonlibperl5.12 is uninstallable so it has to go :-)14:11
didrockscongrats cjwatson (and yeah, no more spam in precise-changes ;))14:13
cjwatsonI wouldn't go that far14:13
didrocks:)14:13
ogra_didrocks, wail until he does the next haskell run :P14:14
didrocksogra_: heh ;)14:15
=== MacSlow|lunch is now known as MacSlow
=== tkamppeter_ is now known as tkamppeter
=== Ursinha is now known as Ursinha-lunch
pittiev: just updated https://code.launchpad.net/~ev/apport/disable-core-removal/+merge/81974 again with a question, FYI14:31
stokachuwhat package needs to be installed in order to have /usr/lib/pymodules available15:11
stokachuin my package build ive enabled dh_pysupport -- does that not put the python libraries in their appropriate directories?15:12
stokachui see them listed in /usr/share/pyshared which is odd to me15:13
tumbleweedstokachu: that's how pysupport shares modules between python versions15:14
tumbleweedstokachu: you should use dh_python2 these days, pysupport is deprecated15:14
stokachuwill dh_python2 create /usr/lib/pymodules?15:14
stokachuin my chroot that directory doesn't exist and is not in the python sys.path15:14
stokachuim using pbuilder to attempt to build this package that has python libraries15:14
stokachuso far everything is placed in /usr/share/pyshared but then fails because its not listed in python path15:15
tumbleweedit shouldn't be in sys.path. It's a python helper implementation detail15:15
tumbleweedstokachu: the maintainer scripts, should provide symlinks into the python path15:16
stokachuand dh_python2 will do that?15:16
tumbleweedso does dh_pysupport15:17
stokachuso dh_pysupport places files in /usr/share/pyshared but where does /usr/lib/pymodules come into play15:17
pittistokachu: just forget that /usr/lib/pymodules/ exists :)15:18
stokachuok15:18
stokachuso at what point does sys.path get updated to look in /usr/share/pyshared?15:19
tumbleweedstokachu: it doesn't15:19
tumbleweeddoes your package have maintainer scripts, which don't have #DEBHELPER# in them?15:19
stokachuyea in my rules file15:19
=== _iron is now known as i_ron
tumbleweedstokachu: no I meant postinst. Can I see the source package? Also maybe this should move to #ubuntu-packaging15:21
stokachuok15:21
=== Tm_K is now known as Tm_T
evpitti: ah, cheers. I've been knee deep in porting the crash reporter to C.  Will have a look after my team meeting.15:49
pittiev: oh, you are porting the client side apport parts?16:13
broderbarry: if you want a citation for your port, dbus strings are required to be utf-8 as per the table at http://dbus.freedesktop.org/doc/dbus-specification.html#message-protocol-signatures16:17
broder(the dbus-daemon does verification and will replace code points that don't validate)16:18
barrybroder: thanks!16:18
evpitti: I had to write a parser for the RFC822 stuff, but this is just the daemon that takes the report and sends it over the wire to the crash database frontend16:24
evyou had requested that if it must live forever, that it not be written in python :)16:24
pittiev: right, but I thought we could start this on demand16:24
pittiby upstart or cron16:25
evthat doesn't appear to be the case16:25
evupstart doesn't have support for inotify watches yet16:25
evor an Internet connectivity event16:25
evI'm happy to spawn it from that with a --single-shot or whatever flag once we have those bits in place16:25
sladenon default-route ?16:26
evbut I didn't want to block on them16:26
pittiev: or just have a small daemon doing the inotify and spawning the python process16:26
evon actual path to crashdb.ubuntu.com or what-have-you16:26
pittiev: but either way, your call16:26
evpitti: no need to do it that way around - I've already written it (except for the n-m dbus bits, but that's a cakewalk)16:27
evthe apport crash format isn't sufficiently complex that I'm worried about having to constantly tune a C parser16:28
=== dendrobates is now known as dendro-afk
=== zyga is now known as zyga-afk
=== beuno is now known as beuno-lunch
stokachu_jmp_, :X17:34
igetrootwhat up all17:45
igetrootanyone alive in here?17:46
slangasekfrequently so17:47
igetrooti got a quick question..not exactly a development question, for of a glitch i found17:47
igetrootatleast i think it's a glitch lol17:48
igetrootlets say im an unprivileged user, working on the terminal. if i enter this specific string and hit enter, the console freezes for a second and then gives it a /bin/sh shell w/ root17:48
igetrootbut it wont respond to commands afterwards17:48
igetrootany cmd i enter at the prompt after, just gives me another blank prompt17:49
igetrootlooks like it only works on ubuntu thus far17:49
mdeslaurigetroot: what string?17:51
igetrootit'll flood the channel bad if i post it17:52
igetrootill put it on a pastie.org post in 1 sec17:52
mdeslaurigetroot: please send an email to security@ubuntu.com17:52
mdeslaurigetroot: and this is the wrong channel, please join #ubuntu-hardened if you want to continue talking about it17:53
igetrootokay definitely17:53
igetrootyeah it will give me a /bin/sh root prompt but no commands seem to do anything17:53
igetrootokay thank you :D17:53
=== Ursinha-lunch is now known as Ursinha
=== beuno-lunch is now known as beuno
=== dendro-afk is now known as dendrobates
=== adam_g_ is now known as adam_g
=== dendrobates is now known as dendro-afk
bjsniderdoko_, can we discuss binutils-gold on lucid for a minute?20:14
dobeycan someone approve my ubuntuone-client upload for oneiric-proposed please?20:39
mbieblbjsnider: binutils-gold can have very negative effects, if the autotools* files aren't very recent20:47
mbieblsee http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=55482120:48
ubottuDebian bug 554821 in libtool "[libtool] Says that binutils-gold doesn't support -version-script" [Normal,Fixed]20:48
bjsnidermbiebl, i have a potential fix for a longstanding bug20:48
Randolphhi all21:09
FusioniteHey21:09
=== salem_ is now known as _salem
angelo-cHi guys, I'm trying to solve bug 773841, I wrote the unity part22:12
ubottuLaunchpad bug 773841 in unity-place-files (Ubuntu) "\\192.168.1.x opens http:\\192.168.1.x in firefox as opposed to smb://192.168.1.x in nautilus" [Low,Confirmed] https://launchpad.net/bugs/77384122:12
angelo-cbut when I try to open the smb:// or ssh:// I have this error: Error showing url: The specified location is not mounted22:13
angelo-cthe same error could be easly reprodeced whit xdg-open smb://192.168.1.222:13
angelo-cbith Unity and Xdg open uses the g_app_info_launch_default_for_uri22:14
angelo-chave you any hunts?22:14
angelo-chints?22:14
angelo-cthere are many bug in lauchpad involving xdg-open22:16
angelo-cI tought that xdg-open sould open nautilus, but nothing happens22:16
angelo-canybody out there?22:25
angelo-canybody out there?22:53
=== dendro-afk is now known as dendrobates
broderangelo-c: i think most of the people that work on unity hang out in #ayatana, so that might be a better place to ask23:29
broderoh, hmm...i missed that xdg-open does the same thing as well23:30
angelo-cit's not a bug reletad to ayatana, but xdg-open23:33
broderright, i see that23:33
angelo-cor better, gnome-open which is called by xdg-open23:33
broderactually, it sounds like an issue with gvfs, which is what gnome uses for its "virtual" filesystem stuff23:34
angelo-cyes, I think so, I'm digging hard the source but I'm stuck23:36
broderi don't know gvfs very well, but i am able to access an smb server if i run "gvfs-mount smb://my/share", and then xdg-open23:37
broderi don't know whether you're supposed to have to explicitly mount it first like that23:37
angelo-cI think the bug is here, xdg-open should open nautilus for smb or ssh, or mount in advance the23:38
broderit seems to me that somebody had a design in their head when they set this up. i would personally try and figure out what that design was before trying to make any changes23:41
broderit might be a good idea to ask around in, say, #gtk+ on irc.gnome.org23:41
angelo-cthe bug is located in this function g_file_query_default_handler in gfile.c, it should mount the path if not mounted23:41
angelo-cgood idea!23:42
broderi think you're making assumptions about somebody else's design23:42

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