/srv/irclogs.ubuntu.com/2011/01/24/#ubuntu-devel.txt

LaurenKi found some bugs in ubuntu 10.04 lts do i report them here?01:36
ari-tczewLaurenK: launchpad is better place01:36
KeybukLaurenK: no, report them using the "Report a Problem" option in the appropriate application's Help menu01:37
LaurenKthere just small bug problems that take away from the polish of the system and make it look unprofessional, like x butons and cancel buttons that arre dummy buttons and don't actualy do anything01:38
LaurenKwhen installing propretary drivers the cancel buttons don't do anything01:38
LaurenKor warn you that can't be canceld01:38
LaurenKi don't know if situtations like that are important01:39
Keybukthey are, look up the "papercuts" project01:39
LaurenKlet me as one oquestion, this may be the wrong spot but on the regular ubuntu channel they were extreamly confused with it, do I need to install a different kernel besides the i386 default one to take advantage of the core duo processor01:43
Keybukcore duo or core 2 duo?01:43
LaurenKjust core duo01:44
Keybukthen no, only the i386 kernel will work for you01:44
Keybuk(which is the "generic" kernel in Ubuntu)01:44
Keybukif you have >~3GB of RAM, you will want the "generic-pae" kernel01:45
macono01:45
macothe default one has SMP enabled01:45
macooops sorry Keybuk. i have high latency01:45
LaurenKexelent thank you01:46
kklimondahas anyone investigated packaging libevent 2.0.10?02:00
kklimondaah, it's already in experimental02:05
jorjosohello people :D03:05
jorjosoI have a little question03:06
jorjoso I would like to  create a background for ubuntu narwhal do you know the deadline ?03:06
broderjorjoso: http://design.canonical.com/2011/01/bright-light-and-beautiful/03:07
jorjosothankss :D03:09
=== Amaranth_ is now known as Amaranth
AbsintheSyringedashua, would you mind pushing this patch (http://bit.ly/fVqvBl ) upstream? Apparently that's the fastest way I'll get it in Debian05:13
kklimondaAbsintheSyringe: it was sent upstream afaict, gnome bug 60851105:33
ubottuGnome bug 608511 in general "Theme, button background, left_middle drawn for left_right" [Normal,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=60851105:33
AbsintheSyringekklimonda, ok awesome, tnx05:33
kvahello, anybody is available to give me a hand?05:36
kklimondakva: shoot, the worst that could happen is that no one is going to answer you.05:38
kvawell, guess I am in the right channel, I am not here for support05:38
kvaI got an emachines e527 and debian is running right on it05:39
kvabut ubuntu doesn't. even boot screen is fade and almost dark05:40
kvaone of live cd's permits to boot but it says that hardware isn't supported by unity and after install it still blinks05:41
kvaso, any ideas?05:41
jmarsdenkva: Try using a Ubuntu 10.04 LTS CD rather than a very recent one?05:42
kvatried it05:42
kvaright now I am on debian squeeze05:43
jmarsdenIf this is just for one (older) PC... and you have Debian installed and running... why not just keep using Debian? :)05:43
kvawell, problem like this were reported, but not for this model of note05:43
kvait's not that old, that's already 64bit :-)05:44
RAOFkva: Hm.  By ‘blinks’, do you mean something like bug #681054 ?05:44
ubottuLaunchpad bug 681054 in xserver-xorg-video-intel (Ubuntu) "screen repeatedly goes black for a fraction of a second" [High,Confirmed] https://launchpad.net/bugs/68105405:44
kvayup, near that05:45
jmarsdenSince it perhaps/probably relates to the video chipset...   what does      lspci -v | grep -i vga     output (in Debian is fine)?05:45
RAOFkva: Interesting information would be - does your machine have the same GPU, and does it also mis-detect a connected VGA output?05:46
kvano, not for a fraction for a second, it's lightens for a fraction for a second, after it it's black05:46
kva00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 09)05:47
RAOFThat particular chip is well supported, but it's always possible for manufacturers to screw things up.05:48
kva00:02.0 VGA compatible controller: Intel Corporation Mobile 4 Series Chipset Integrated Graphics Controller (rev 09) (prog-if 00 [VGA controller])05:48
kvaSubsystem: Acer Incorporated [ALI] Device 045905:48
kvaKernel driver in use: i91505:49
kvathat's from lspci -v05:49
RAOFPerhaps you'd like to continue in #ubuntu-x?05:49
kvano idea05:49
kvaproblem is before X11 starting05:49
kvaif you want I can continue there :-)05:50
RAOFYes please :)05:50
=== Amaranth_ is now known as Amaranth
=== pitti is now known as pitti_
pitti_Good morning05:58
pitti_tkamppeter_: can you please have a look what happened here?06:39
pitti_ghostscript (Δ 3.7 MB - 8.71.dfsg.2-0ubuntu7: 2.8 MB   9.01~dfsg~svn12047-0ubuntu1: 6.5 MB)06:39
pitti_tkamppeter_: i. e. ghostscript ballooned from 2.8 to 6.5 MB06:40
pitti_doko__: good morning!06:52
pitti_doko__: given that we haven't shipped oo.o-filter-binfilter in the past (for legacy document formats like staroffice 5.2), do you think we can drop the -writer recommends: libreoffice-filter-binfilter to suggests? that'd save 8 MB again06:53
AnAntHello07:07
=== tkamppeter_ is now known as tkamppeter
tkamppeterpitti, about ghostscript, the maintainer at Debian has changed and he has introduced a new packaging scheme. Seems that I have to compare file by file now where the wrong thing leaked in ...07:20
pitti_tkamppeter: my hunch is that /usr/share/ghostscript/9.01/Resource/CMap/ got a lot bigger07:20
pitti_tkamppeter: and we additionally got a new 2.1 MB libgs9 package; apparenlty that was integrated into ghostscript earlier on?07:21
pitti_ah, sorry, that was libgs8, nevermind07:22
tkamppeterpitti_, with which version of GS are you comparing? Does your old version have files in CMap at all?07:22
pitti_./usr/share/ghostscript/8.71/Resource/CMap -> /var/lib/defoma/gs.d/dirs/CMap07:23
pitti_tkamppeter: no, it only had above symlink07:23
pitti_not that this would exist..07:23
pitti_but as we didn't need them in the past, perhaps we can split them out into a new ghostscript-cmaps package?07:23
tkamppeterpitti_, now we have a real CMap directory, as once defoma got deprecated and second, the license of the CMap files which come from Adobe has changed, allowing to ship them.07:24
tkamppeterSee debian/changelog07:24
pitti_tkamppeter: given that we have never really supported anything but UTF-8, what do you think about a splitout to a seprate package?07:27
pitti_we can then install this package only for CJK locales (if it's any help there)07:28
tkamppeterpitti, the re-introduction of CMap fixes several bugs, and therefore it should be part of the core distro (Desktop CD): bug #321932, ghostscript bug 691212, ghostscript bug 691345.07:28
ubottuLaunchpad bug 321932 in msttcorefonts (Ubuntu) "Ghostscript does not render when ttf-mscorefonts-installer is installed" [High,In progress] https://launchpad.net/bugs/32193207:28
ubottuBug 691212 on http://launchpad.net/bugs/691212 is private07:28
ubottuLaunchpad bug 691345 in tftp-hpa (Ubuntu Natty) "buffer overflow in tftp" [Medium,Fix released] https://launchpad.net/bugs/69134507:29
pitti_tkamppeter: and these need all of them?07:30
tkamppeteraccording to http://bugs.ghostscript.com/show_bug.cgi?id=691345 CMap files are needed also for non-CJK.07:35
tkamppeterpitti_, ^^07:35
=== almaisan-away is now known as al-maisan
tkamppeterpitti_, it would be a maintenance nightmare to sort the CMap files by languages and make them installed only if the appropriate locale is installed. Also it can happen that a PDF contains some characters of a foreign language and these should also get rendered correctly. even if the user does not understand this language.07:38
pitti_tkamppeter: ok; I just wondered why it by and large worked fine so far07:38
pitti_tkamppeter: thanks for the heads-up!07:38
tkamppeterpitti_, note also that the re-introduction of CMap already happened in Maverick, still with the old packaging scheme of Ghostscript.07:41
pitti_tkamppeter: so in maverick it was broken because of replacing the CMap dir with the dangling symlink?07:42
tkamppeterpitti, now I discovered it, it seems that I have changed the source tarball, but somewhere in the installation process it must have been "rm -rf"ed in favor of Defoma.07:45
=== hunger_ is now known as hunger
tkamppeterpitti_ is defoma really deprecated in Debian (= not maintained any more)?07:45
pitti_tkamppeter: yes07:46
pitti_it's supposed to get removed after squeeze07:46
tkamppeterpitti_, does Maverick still contain defoma?07:48
pitti_yes, and natty does as well, as it still has a couple of rdepends07:48
tkamppeterpitti, so in Maverick ghostscript was still working defoma-based then. The changes in Debian's Ghostscript 8.71, still done by the old maintainer were incomplete then.07:50
tkamppeterpitti, this natty's ghostscript should then really fix the Ghgostscript part of bug 321932.07:53
ubottuLaunchpad bug 321932 in msttcorefonts (Ubuntu) "Ghostscript does not render when ttf-mscorefonts-installer is installed" [High,In progress] https://launchpad.net/bugs/32193207:53
tkamppeterpitti_ ^^07:53
dholbachgood morning08:01
AnAntHello08:06
doko__pitti_: yes, will do08:06
pitti_doko__: danke08:06
amitkdholbach: pitti_: can I ask for help in getting 'powerdebug' into the archive. It is pending in the NEW queue for a week now.08:06
dholbachamitk, I'm not an archive admin - somebody in https://launchpad.net/~ubuntu-archive/+members should be able to help08:07
AnAntwhy didn't I get this FTBFS http://launchpadlibrarian.net/62702256/buildlog_ubuntu-natty-i386.libgwenhywfar_4.0.3-1_FAILEDTOBUILD.txt.gz on maverick ?08:08
amitkdholbach: thx08:08
didrocksgood morning08:15
=== al-maisan is now known as almaisan-away
pitti_amitk: I'm not a regular AA, but I can have a look09:16
AnAntdoko__: Hello, can you help with the verilator question I asked yesterday ?09:23
pitti_amitk: you have mail09:24
pitti_bbl09:24
doko__AnAnt: ?09:28
AnAntdoko__: this FTBFS http://launchpadlibrarian.net/62644475/buildlog_ubuntu-natty-powerpc.verilator_3.810-1_FAILEDTOBUILD.txt.gz09:29
AnAntI am getting the same FTBFS on Debian sparc, I compared the amd64 & powerpc buildlogs for natty, I found that they differ in the binutils revision used. binutils 2.21-4ubuntu1 in other archs (with successful build), while 2.21-3ubuntu1 on powerpc (also I see that binutils 2.21-4ubuntu1 failed to build on powerpc)09:30
doko__AnAnt: does the package use linker scripts?09:30
brendandhi09:31
brendandi've made a change to update-manager and want to test it out09:31
AnAntdoko__: you mean something like libtool & ltmain.sh ?09:31
brendandwhat do i need to do?09:31
doko__no, linker scripts09:32
AnAntdoko__: I'm not sure what is meant by "linker scripts"09:32
amitkpitti_: thanks09:33
doko__AnAnt: ld called with -T <script>09:33
doko__or --script=scriptfile09:33
AnAntdoko__: please note that FTBFS is not happening during linking09:33
=== almaisan-away is now known as al-maisan
AnAntdoko__: anyways, I don't see any -T nor --script in the buildlog09:34
doko__AnAnt: I'm not that interested in debugging this ...09:34
AnAntdoko__: I was advised that you know something about different arch quirks, that's why I asked you09:34
janimoAnAnt,  indeed that does not look like a toolchain error09:37
janimoor not obviously at least09:38
smb@pilot in09:51
=== udevbot changed the topic of #ubuntu-devel to: Archive: Open | Development of Ubuntu (not support, not app development) | #ubuntu for support and general discussion for dapper -> maverick | #ubuntu-app-devel for application development on Ubuntu | http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://wiki.ubuntu.com/HelpingWithBugs | Current Friendly Patch Pilots: smb
=== doko__ is now known as doko
dokoScottK: about #684703, did the object layout of one of the parent class change?10:19
evfoolhi everyone10:22
janimodoko, should the lo33 tagged bugs be moved to the libo package now that is it in the archive?10:34
dokojanimo: I already did. if you find more, yes, go ahead10:34
janimodoko, ah ok. I just saw it only show two bugs for natty but the full ubuntu list invludes them, thanks10:35
=== al-maisan is now known as almaisan-away
sladeniain!11:32
=== MacSlow is now known as MacSlow|lunch
=== brendand is now known as brendand|lunch
pitticjwatson: http://people.canonical.com/~pitti/tmp/nbs.html is the current output of the "better NBS report" script I mentioned at the sprint; do you have suggestions for improvement?12:21
pitticjwatson: it doesn't have cycle detection yet12:21
pittiI'll add the "command to remove empty packages" still12:22
cjwatsonpitti: sorting within each package (maybe by section first and then by depended-upon package name).  Perhaps an inverse view as well - packages that depend on NBS packages, sorted by the number of such packages they depend on?12:28
pitticjwatson: ah, do you actually care much for the component when doing NBS cleanup?12:31
pitticjwatson: inverse view> nice idea, will add that12:31
cjwatsonpitti: component> not desperately I suppose, I do have a slight preference for fixing main first since that sometimes means that the NBS thing can at least be demoted12:35
dmartcjwatson: hi there, do you how I can see all the console output from startup jobs using upstart/plymouth?  Ideally, I'd like to get a log over serial, as can be achieved on Debian with console=12:39
=== almaisan-away is now known as al-maisan
=== csurbhi is now known as csurbhi-afk
geserpitti: your updated NBS page looks great12:45
cjwatsondmart: upstart generally doesn't give jobs a console12:45
cjwatson       console output|owner12:45
cjwatson              By default the standard input, output and error file descriptors of jobs are connected to /dev/null12:45
cjwatson              If this stanza is specified, they are connected to /dev/console instead.12:45
cjwatson(init(5))12:45
cjwatsonI think ideally we'd like them all to be at least logged12:45
pitticjwatson: http://people.canonical.com/~pitti/tmp/nbs.html -> now with more sorting and reverse map (first by number of NBS deps, then by name)12:46
cjwatsonthat looks pretty good12:47
geserpitti: would it possible to have the component tags ("main", "universe") in different colors?12:48
Chipzzcjwatson: shouldn't stdout/stderr be redirected to syslog?12:48
pittigeser: sure, suggestions?12:49
Chipzzpitti: you're missing a fancy launchpad template though ;)12:50
pitticjwatson: hm, didn't cron.NBS use to be in lp_archive@cocoplum cron?12:51
ScottKdoko: I don't know (and I'll be offline most of the next two days - so no time to check).  Maybe Riddell could looke into it if you would ask him.12:52
Chipzzpitti: is it normal that libaqbanking29 and libaqbanking-plugins-libgwenhywfar47 have a circular dependency?12:53
pittiI think yes12:53
cjwatsonChipzz: 12:45 <cjwatson> I think ideally we'd like them all to be at least logged12:53
pittiwell, "normal" as in "the NBS report", not "it should be like that"12:53
geserpitti: perhaps "main" and "restricted" in red/orange and "universe"/"multiverse" in darkyellow/gold12:53
cjwatsonpitti: ~lp_archive/dak/cron.sync12:54
cjwatsoncircular dependencies are not unusual between libraries and plugins, and they're not intrinsically a bug12:54
=== csurbhi-afk is now known as csurbhi
Chipzzcjwatson: that's what I was referring to; as a solution to "we want to all logged", log to syslog, as opposed to for example letting upstart log to a file :)12:55
cjwatsonI think it would need to be more sophisticated than that, to allow for things like "start this job right now and show me its console output"12:56
cjwatsoncompared to the current approach, I'd say syslog would just be differently inadequate12:56
dmartcjwatson: for a quick hack, would it be adequate to paste the line "console output" into /etc/init/*.conf 9~/12:58
cjwatsondmart: you could do that locally and it would probably have roughly the kind of effect you want, but watch out for jobs that already have a (different?) console line13:02
Chipzzcjwatson: ok, just chipping in some cents13:02
\shguys, does someone know why we didn't ship any live-initramfs packages in maverick? (live-initramfs as in srcpkg:live-initramfs or srcpkg:live-boot)?13:05
dmartcjwatson: understood ... I that solves the problem enough for me for now.  Thanks!13:07
pittigeser: http://people.canonical.com/~ubuntu-archive/nbs.html13:10
pittialso added component to reverse list13:11
apwcjwatson, btrfs was a tech-preview in maverick wasn't it?  as such its not really something we put much effort into supporting13:11
geserpitti: looks great with the coloured and aligned components13:12
cjwatsonapw: yes13:13
=== diwic is now known as diwic_afk
=== MacSlow|lunch is now known as MacSlow
pittihttps://wiki.ubuntu.com/ArchiveAdministration updated accordingly13:21
=== sforshee is now known as sforshee-afk
maxxies<Fuchs> maxxies: Du verstoesst gerade gegen die Regeln.  Bitte diskutiert das im OT Kanal, wenn ihr es diskutieren wollt.13:49
maxxies<maxxies> Fuchs: nenne mal die regel, bitte.13:49
maxxies<maxxies> was zum  henker ist OT an dem wunsch nach einem aktuelleren kernel?13:49
maxxies* tux-flo1 hat die Verbindung getrennt (Client Quit)13:49
maxxies* nexx hat die Verbindung getrennt (Quit: quit)13:49
maxxies<Fuchs> die Diskussion aktuell ist OT, und die Regeln waeren erstens das Topic und zweitens  "Befolge die Anweisungen der OPs - nimm Verwarnungen ernst." die.13:49
maxxies<maxxies> Fuchs: wieviel bist du jünger als 35 jahre um so ein affiges verhalten gegenüber mir an den tag zu legen?13:49
evilvishpitti: hi.. i was reading the keymap README , and it instructs to file bugs in https://bugs.launchpad.net/udev/+bugs < which does not exits yet..13:52
evilvishI guess we can file bugs in udev(ubuntu)..13:52
=== csurbhi is now known as csurbhi-afk
evilvishexist*13:52
=== Quintasan_ is now known as Quintasan
pittievilvish: right, already fixed that upstream, to now point to the mailing list14:04
evilvishneat! :)14:05
evilvishpitti: so, no need to file a bug in udev(ubuntu)? just need to send the keycodes to ML?14:06
* evilvish looks for ML link..14:06
pittievilvish: as you prefer, either works; I'm just not very good at seeing new incoming ubuntu bugs14:06
evilvishok.. will send to ML then. :)14:07
hyperairDktrKranz: ping14:15
mterrykees, I asked a question about how to ask the security team about UPnP in bug 682404, but wanted to ping you in case your filters ate it14:17
ubottuLaunchpad bug 682404 in hupnp (Ubuntu) "MIR hupnp" [Undecided,Incomplete] https://launchpad.net/bugs/68240414:17
mterrydoko, also, what's the protocol for something that was pre-promoted but when reviewed was not yet ready?  I mean, the packaging issue for hupnp is easily overcome, but the security one may be trickier.14:18
dokomterry: well, if kees thinks the issue should be solved, then milestone it to alpha3 and open a natty task.14:20
mterrydoko, sure14:20
dokokees: $ DEB_BUILD_OPTIONS=hardening dpkg-buildflags --get CFLAGS14:25
doko-g -O214:25
doko^^^: shouldn't the hardening options be appended?14:25
bdrungmterry: thanks for approving libkibi's MIR. now can i build main package with libkibi?14:32
mterrybdrung, heh, you got that notification quick.  :)  Not yet, an archive admin now has to actually promote the package14:32
mterrybdrung, if you want that to go faster, poke one14:33
bdrungmterry: i saw that you assigned yourself, then i looked at the bug in lp and saw that you approved it. :)14:36
mterrypitti, does python-distutils-extra have support for finding/running a test suite?14:37
mterrypitti, (or python-distutils for that matter)14:37
* bdrung pokes doko and pitti ^.14:37
pittimterry: p-d-e doesn't, as I don't know about a de-facto standard to invoke a test suite with setup.py14:39
pittibdrung: sure14:39
ogracould an archive admin please promote x-loader-omap3-beagle and x-loader-omap4-panda binaries to main ? the package names of the binary packages have changed14:39
bdrungpitti: look at ubuntu-dev-tools how we use build tests.14:40
ogra(that currently blocks us from building armel packages)14:40
pittiogra: yep14:40
ograpitti, TA!14:40
pittibdrung: romoted14:40
pitti'p'14:40
pittiogra: done14:41
ogramerci !14:41
pittiogra: de rien14:41
bdrungpitti: my brain matched the pattern "romoted" with "removed" :D14:41
bdrungthanks14:42
RoAkSoAxCan somebody please take a look to bug #52528714:46
ubottuLaunchpad bug 525287 in lvm2 (Ubuntu) "Add support for corosync based clusters in clvm" [High,Confirmed] https://launchpad.net/bugs/52528714:46
DktrKranzhyperair: pong14:46
=== al-maisan is now known as almaisan-away
seb128kees, there? do you have a minute to discuss remmina, rdp and what choices we have?14:49
hyperairDktrKranz: could you look at zeitgeist# in debian NEW? =)14:50
hyperairpretty please?14:50
DktrKranzhyperair: this evening (CET time), probably14:50
=== smb` is now known as smb
hyperairDktrKranz: okay, thanks.14:52
hyperairDktrKranz: oh clutter# as well, please.14:52
DktrKranz# => sharp?14:53
pitticjwatson: WDYT about a lucid freeze exception for xubuntu-docs? we don't even do xubuntu .2 point releases, do we?14:53
hyperairDktrKranz: yes14:57
cjwatsonpitti: it seems unlikely to cause a problem14:57
pitticjwatson: that's what I thought14:57
cjwatsonI think in fact I already said to micahg that it was fine14:57
micahgpitti: I think xubuntu wanted to have disks spun for 10.04.214:58
pittimicahg: ok; reviewed and accepted now, I just wanted some peer review here14:58
micahgpitti: np, thanks14:58
=== sforshee-afk is now known as sforshee
janimocjwatson, on preinstalled images (such as those for armel) which is the best way to have something run on first install? There's a WI that requires we present tasksel on the headelss images15:11
=== mbarnett` is now known as mbarnett
sconklinpitti: I uploaded maverick kernel 2.6.35-25.44 friday to revert the regression we had for Radeon. Could you copy that to -proposed at your convenience, please?15:13
=== dendro-afk is now known as dendrobates
james_wcjwatson, hi. When installing on btrfs is UUID still used to in fstab? If so, can you point me to the code that sets or finds the UUID?15:15
ograjames_w, i think thats ROOT="/dev/disk/by-uuid/${ROOT#UUID=}" .... in /usr/share/initramfs-tools/init15:22
ogra(for finding it)15:22
ograand i think its one of the partman udebs that sets it in fstab15:22
ogradunno which off the top of my head15:23
james_wpartman-btrfs sets it, but it gets passed the first element (UUID/path/whatever) by something else15:23
ogras7fstab/fstab and cmdline/15:23
james_wthe issue we are seeing is the mkfs.btrfs doesn't have a -U option, so we're not sure how to set the UUID, or if it's even possible15:24
ograit likely calls blkid <device>15:24
cjwatsonyes, we just use blkid15:24
cjwatsonwe have no need to force the UUID to a specific value in d-i15:24
cjwatsonpartman-btrfs just lets mkfs.btrfs pick something15:25
ogratune2fs has an option to change UUID for extX based filesystems, if you need to set it there is probably a similar tool for btrfs15:25
cjwatsonjanimo: the preinstalled images are nothing to do with me and I don't really know the intricacies of how they work; ogra can probably help you15:26
james_wthanks ogra, cjwatson15:26
ogracjwatson, well, the WI for the minimal images requires to enable tasksel in oem-config, i guess thats why janimo pinged you about it15:26
ograi guess that needs some ubiquity magic15:27
cjwatsonthe oem-config debconf frontend has some stuff for handling tasksel already15:27
ograah, sweet15:27
ograjanimo, ^^^15:27
cjwatsonI forget how you enable it, but there's a tasks plugin15:28
ograthats all we should need15:28
cjwatsonoem-config      oem-config/steps        multiselect language, timezone, keyboard, user, network, tasks15:28
cjwatsonlooks plausible15:28
ograyeah15:28
ograjanimo, so we can just add a pressed line for it15:29
ograi guess that needs some debian-cd adjustment to add it to the cmdline15:29
=== bjf[afk] is now known as bjf
loolcjwatson: (Poking your brain on binfmt-support  :-)  we're discussing qemu-user support in schroot, and currently we have a shell script which captures output of `file $chroot/bin/true` and maps it to QEMU architecture, then copies qemu-$arch-static into the chroot's usr/bin; instead of doing that, we considered asking the question to binfmt-support, or parsing binfmt_misc's data to figure out the name of the interpreter15:36
cjwatsoncall, be with you in 20mins15:36
loolack15:37
janimoogra, cjwatson  thanks.15:37
pittisconklin: done15:37
janimoogra, I was under the impression preseed only worked with d-i and once the installation phase is done it has no effect15:38
sconklinpitti: thanks !15:38
ograjanimo, it also works with ubiquity15:38
cjwatsonjanimo: not quite correct no15:38
ograjanimo, and oem-config is essentially ubiquity15:38
ograwe run oem-config on second boot in the preinstalled images15:39
ogra(and cjwatson will likely correct me the preseeding works in tons of other cases at package installation time for packges using it)15:39
ogras/the/that15:39
ogra:)15:40
cjwatsonthat's right15:40
cjwatsonfor example it's common to preseed away debconf licence questions15:41
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
=== Riddelll is now known as Riddell
cjwatsonlool: so you'd like to have a way to ask binfmt-support which interpreter would be used for a given executable, presumably going through the detector system too - a sort of dry-run mode?16:03
loolcjwatson: Yes; basically goal is to have the interpreter work in the chroot16:04
cjwatsonlool: it sounds much the same as what run-detectors is already doing, so we could pull most of that out into common code and add a new query-binfmts command, say16:04
loolcjwatson: We considered other options, like running outside of the chroot, but doesn't seem too good an idea16:04
loolcjwatson: That would be ideal16:05
cjwatsonI'm happy to have a go at that16:05
loolcjwatson: If you can think of a way where we're not involved in any copy, that's even better, but I think query interpreter + copy it from schroot seems like the best split of responsibility16:05
cjwatsonwell, a query option would be a good thing to have anyway, and doesn't preclude other approaches16:06
cjwatsonI suspect you need significant kernel support for a better approach; basically a per-chroot /proc/sys/fs/binfmt_misc16:06
=== beuno is now known as beuno-lunch
cjwatsonmaybe update-binfmts --query rather than query-binfmts16:07
cjwatsonafter all, --display isn't an "update" either16:07
=== deryck is now known as deryck[lunch]
ari-tczewzul: I'm very disappointed that you didn't check bug 701182 before merging. IIRC this is not first time.16:26
ubottuLaunchpad bug 701182 in nut (Ubuntu) "Merge nut 2.4.3-2 (main) from Debian unstable (main)" [Wishlist,New] https://launchpad.net/bugs/70118216:26
zulari-tczew: sorry..16:26
Riddellslomo: how would you feel about dropping gstreamer0.10-plugins-base's recommends on gvfs to a suggests?16:41
slomoRiddell: not good as long as there's no alternative for kde applications using gstreamer (someone with kde knowledge should really write a kio plugin... or how it's called today)16:42
pittislomo: but as KDE doesn't actually have a gvfs integration, what does the gvfs recommends help?16:43
slomopitti: it still gives you sftp support for example (if you have a no-password ssh key or gnome-keyring is running)16:45
Riddellslomo: why does gstreamer need sftp support?16:54
slomoRiddell: maybe you want to play stuff from some sftp location, it was just an example. you'll need it for samba too, to give another example16:55
cjwatsonlool: want to give http://bzr.debian.org/binfmt-support/trunk a try?  sudo apt-get build-dep binfmt-support && ./configure && make && src/update-binfmts --find $chroot/bin/true16:59
SpamapSjhunt: still around?16:59
jhuntSpamapS: hi16:59
cjwatsonlool: note it may print multiple interpreters, one per line17:00
SpamapSjhunt: had a chance to look at the umountroot I sent?17:01
keesseb128: hi! just starting work now.17:04
seb128hey kees17:04
keesseb128: yeah, not sure about rdp :(17:04
seb128kees, so you n-acked the rdp thing remmina is using17:05
keesseb128: but that bit of freerdp is really horrible :(  "return True" on a security validation? ewww17:05
seb128kees, it's not clear that tsclient that we have is any better, it's non maintained and doesn't seem to have any certificate code at all17:05
keesseb128: that's not even an iffy call. :)17:05
seb128options are to17:06
keesseb128: well, if nothing does that cert check, I guess it's a wash, but looking at what rdesktop does, it at least tries something in that area17:06
seb128- stay on tsclient which is buggy and unmaintained (and never got a security review I think)17:06
seb128- try to fix the freerdp code17:06
keesI think asking upstream to actually fix their code would be the best all around option.17:06
seb128- drop rdp from the default install17:06
keesdropping rdp seems like a bit of a shame, but afaik, everyone was using rdesktop out of universe when they needed rdp anyway17:07
seb128I'm not even sure how much the standard user care about rdp17:07
jhuntSpamapS: yeah - like you, not crazy about the timeout, but which is better worst-case - sleep forever or corrupt FS? :|17:07
seb128and if that should not be featured in s-c17:07
seb128kees, would you fell comfortable with freerdp is they fix that issue17:08
=== cking is now known as cking-afk
SpamapSjhunt: I think given that we cannot *guarantee* that init is even doing the re-exec, we have to have a fail-safe which doesn't leave shutdown waiting forever.17:08
seb128or do you think it shows that they don't care much about security in their code and would still prefer to not use freerdp by default?17:08
Riddellslomo: any KDE programme would just read those with kio though17:08
keesseb128: that was the first place I looked, so I stopped the review there. I can look more closely at it again17:08
seb128kees, well I would appreciate to know if it's worth working with them on that one issue or not17:09
keesseb128: well, the code seems to indicate an eye for modularization, so that's an improvement over rdesktop.17:09
keesseb128: give me 30 minutes, let me look through their network code and if I run away screaming, we can drop rdp, and if it's sane, we can ask them to fix their crypto?17:10
cjwatsonSpamapS,jhunt: I guess a contributing factor is that wait forever -> user hits power button isn't really especially better than timing out too early17:10
slomoRiddell: how would it do that without a gstreamer plugin that uses kio? :)17:10
seb128kees, seems great, no hurry though just do that whenever you can this week and that will do17:10
slomoRiddell: but sure, that would be the best solution... if existant17:10
seb128kees, thanks17:10
smb@pilot out17:10
=== udevbot changed the topic of #ubuntu-devel to: Archive: Open | Development of Ubuntu (not support, not app development) | #ubuntu for support and general discussion for dapper -> maverick | #ubuntu-app-devel for application development on Ubuntu | http://wiki.ubuntu.com/UbuntuDevelopment | See #ubuntu-bugs for http://wiki.ubuntu.com/HelpingWithBugs | Current Friendly Patch Pilots:
seb128kees, we can start working with them to fix the certificate issue in any case since that would be nice to fix anyway17:11
jhuntSpamapS,cjwatson: maybe a last-ditch call to sync before we potentially rip the carpet out from under upstart?17:11
slomoRiddell: if you know someone who could be interested in this... i'd be happy to help on the gstreamer side of things17:11
cjwatsonjhunt: Keybuk investigated this a while back and it turns out that the kernel's reboot() syncs anyway17:12
cjwatsonoh, wait17:12
cjwatson        * util/reboot.c (main): Restore the sync() system call before17:12
cjwatson        calling reboot(); the Linux kernel says we have to do this, and I17:12
cjwatson        suspect that ext4 is no longer forcing this before power off.17:12
SpamapSI wonder if thats why when init exits it says "Panic -- not syncing"17:13
cjwatsonbut anyway, the reboot *program* does it so it probably isn't necessary to do it in umountroot17:13
smoserwhat is the "correct" way to configure timezone on ubuntu ? is it dpkg-reconfigure?17:15
SpamapSRight.. I think with that.. and waiting 5 seconds for a re-exec which *should* only be harvesting at most 9-10 inodes.. that seems to count as a best effort.17:15
smoserI'd like to be able to take as input something like "US/Eastern" and "make it so"17:15
=== yofel_ is now known as yofel
cjwatson'dpkg-reconfigure tzdata' but it's also perfectly OK to write to /etc/timezone and copy /usr/share/timezone/$zone to /etc/localtime17:17
Riddellslomo: apachelogger knows all and says phonon will use kio streaming17:17
slomoRiddell: ah that's interesting, i guess i should talk to him17:17
slomoRiddell: then feel free to make it a suggests17:18
LaneyI accidently uploaded haskell-utf8-string thrice to NEW, please reject two of them if that will cause problems (and I don't know if it needs the original source or not as it was previously in the archive but removed — it wasn't rejected either way)17:18
bigonzul: hi any reason you didn't upload my debdiff for nut?17:19
Riddellslomo: apparantly we want to keep it in sync with debian17:26
Riddellwhere I can't upload17:26
slomoRiddell: i'll change it with the next upload17:27
Riddellthanks slomo17:27
seb128Riddell, do you want to get in this week? feel free to get the current debian version and change that over it17:32
=== beuno-lunch is now known as beuno
seb128Riddell, we can sync gstreamer0.10 and gst-plugins-base0.1017:32
=== deryck[lunch] is now known as deryck
=== sebner_ is now known as sebner
udienzari-tczew: what happen if a packages has already removed from debian and still in Ubuntu? should be removed too?17:43
ari-tczewudienz: depends on the reason17:44
ari-tczewif something like 'buggy' unmaintained', then yes17:44
ari-tczewrc bugs17:45
udienzhm..17:45
Laneyusually we follow Debian in such removals17:45
=== dendrobates is now known as dendro-afk
geserthe archive admins "sync" removals from Debian to Ubuntu now and then17:45
ari-tczewudienz: could you tell us about which package do you thinking17:46
cjwatsonat the moment I'd suggest filing a bug about any specific cases you care about, and subscribing ubuntu-archive17:47
udienzari-tczew, wait.. i'll find it, i found a package (during try to fix ftbfs) which removed in debian and still in ubuntu17:47
ari-tczewcjwatson: without ACK from ubuntu-sponsors?17:48
geserari-tczew: I assume the usual sponsoring if needed17:49
cjwatsonari-tczew: whatever process is necessary17:50
cjwatsonthough TBH the archive admins carefully review all removal requests so I don't think sponsoring is necessarily required.17:50
cjwatsonIOW I'd think fairly hard about a removal request from anyone, even if they could upload17:50
udienzari-tczew, https://launchpad.net/ubuntu/+source/fluxconf17:55
udienzremoved from debian17:55
udienzdebian bug https://launchpad.net/ubuntu/+source/fluxconf17:56
udienzrrr17:56
udienzdebian bug 59664017:56
ubottuDebian bug 596640 in ftp.debian.org "RM: fluxconf/0.9.9.2-3" [Normal,Open] http://bugs.debian.org/59664017:56
cjwatsonFYI: if you're seeing empty apt term.log files in apport dumps, it's probably bug 68032817:56
ubottuLaunchpad bug 680328 in qapt (Ubuntu Natty) "Many postinst scripts fail using either AptDaemon, PackageKit, or QApt" [Critical,Confirmed] https://launchpad.net/bugs/68032817:56
cjwatsonpackagekit fixes sent upstream (phew).  echidnaman will need to deal with copying some bits of that over to qapt17:57
cjwatsonI think the aptdaemon bit of that subject is mistaken, as I can't reproduce the problem there17:57
smbcjwatson, Hi Colin believe to remember that you had to make some quirks for me and apw to give us nomination rights in lp (because the groups stuff does not work). Could you do the same for sconklin ?18:00
sconklinbjf will need this also, and possibly others18:01
sconklinunless upload rights are a prerequsite18:02
apwcjwatson, you had to add linux to the people directly the package set doesn't enbue nomination rights18:02
apwsconklin, it would be those with upload rights18:02
sconklinok, it's time for Brad to work on that anyway . . .18:03
chrisccoulson_anyone know what keeps swt-gtk in main? when i check it's rdepends, i don't see anything obvious depending on it18:07
=== `Alessio` is now known as quadrispro
bcurtiswx_can i get pbuilder-dist to give me a config.log ?18:12
seb128it's in the build dir18:13
seb128the same way you got the makefiles the other day18:13
=== dendro-afk is now known as dendrobates
seb128don't clean it after build18:13
kklimondapbuilder-dist saves the last log as ~/pbuilder/$release-result/last_operation.log18:14
cjwatsonapw,sconklin: done, and for smb too.  yes, upload rights are a prereq18:14
smbcjwatson, Err I thought I already had those...?18:15
smbAt least I could nominate. :)18:15
cjwatsonoh, who knows, I'm not going to try to debug launchpad18:15
=== sforshee is now known as sforshee-lunch
bcurtiswx_kklimonda, but is that what config.log would output (noob question warning) :P18:15
smbHeh, understandable18:15
sconklincjwatson: thanks!18:15
seb128bcurtiswx_, read what I just wrote?18:16
kklimondabcurtiswx_: ah, ignore what I say and listen to seb128 :)18:16
bcurtiswx_seb128,  I did read it.  I am trying to find a way to automatically stop pbuilder from quitting after a failed build18:24
bcurtiswx_oh, quit and not back18:24
kklimondabcurtiswx_: https://wiki.ubuntu.com/PbuilderHowto#Running%20a%20Shell%20When%20Build%20Fails%20%28Intro%20to%20Hook%20Scripts%2918:25
kklimondabcurtiswx_: it will launch a shell when the build fail18:25
bcurtiswx_kklimonda, i was going to tell seb i just found that and am in there :P much thanks though :D18:26
bcurtiswx_kklimonda, http://paste.ubuntu.com/557772/18:27
bcurtiswx_seen that before possibly?18:27
kklimondabcurtiswx_: and that's it?18:28
kklimondait's a little short :)18:28
bcurtiswx_thats all it gave me18:28
kklimondacopy a log outside of the chroot, and use pastebinit to paste it18:29
kklimonda(or use cat config.log and then select entire log using mouse - but it's going to be long)18:30
bcurtiswx_copy the config.log from the pbuilder to somewhere outside of it.. then pastebinit /18:30
kklimondamore config.log just displayed the first screen18:30
kklimondabcurtiswx_: you can even install pastebinit inside of the chroot - the idea is to paste the whole log somewhere, and pastebinit is an easy way to do that18:31
bcurtiswx_kklimonda, oh duh me.. i see what stupidity i have18:32
bcurtiswx_kklimonda, http://pastebin.com/646TdDjg18:33
kklimondabcurtiswx_: I guess "error: Could not find gtk-update-icon-cache" doesn't help you much? :)18:34
kklimondais it in your PATH?18:34
bcurtiswx_kklimonda, nope but if i had to guess i may need to do something with configure.ac because gtk-update-icon-cache is gtk-update-icon-cache-3.018:34
kklimondabcurtiswx_: then, the last resort is to browse through the configure.ac (or even configure script itself) to see what's exactly happening18:35
kklimondaah, it was renamed?18:35
bcurtiswx_kklimonda, yes18:35
kklimondayes, it makes sense that the script can't find it then18:35
kklimondayou'll have to edit configure.ac then18:36
bcurtiswx_kklimonda, was include /usr/share/cdbs/1/rules/simple-patchsys.mk made something else for quilt 3?18:36
kklimondabcurtiswx_: I don't understand, but the warning you have pasted in #ubuntu-desktop (simple-patchsys.mk is deprecated - please use source format 3.0 (quilt) instead) means that you should update your package to the new source format, and stop using simple-patchsys. I think it's just a friendly reminder18:37
bcurtiswx_kklimonda, it is 3.0 already.. so idk why i c that warning18:38
kklimondabcurtiswx_: if it is 3.0 (quilt) then you should remove /usr/share/cdbs/1/rules/simple-patchsys.mk from d/rules18:38
ari-tczewzul: could you sponsor bug 707050 ?18:48
ubottuLaunchpad bug 707050 in nut (Ubuntu) "Lost patches0004-netvision-improvements-lp-600950.patch" [Medium,New] https://launchpad.net/bugs/70705018:48
ari-tczewas you're last merger18:48
zulari-tczew: yeah gime a few minutes18:48
=== hunger__ is now known as hunger
ari-tczewdoko: around?19:01
=== sforshee-lunch is now known as sforshee
=== dendrobates is now known as dendro-afk
siretartdoko: xine-lib is fixed19:11
micahgudienz: bug 703718, it depends on the reason it was removed from Debian19:12
ubottuLaunchpad bug 703718 in fluxconf (Ubuntu) "Requesting removal of source package `fluxconf' from Ubuntu" [Undecided,Confirmed] https://launchpad.net/bugs/70371819:12
=== dendro-afk is now known as dendrobates
udienzmicahg, aha.. ok. i have another questions. package rsbac-admin has been removed base on this (debian bug 364685)19:20
ubottuDebian bug 364685 in ftp.debian.org "RM: rsbac-admin -- RoM; RC-buggy; unmaintained" [Wishlist,Open] http://bugs.debian.org/36468519:20
udienzbut upstream still exits19:21
micahgpitti: we've got a new bug in xubuntu-docs for lucid (not sure if it was there before or not), xubuntu.org links are incorrect, I'm getting it fixed, but I just wanted to check on procedure, I should reupload with -v on the version in the release pocket?19:21
udienzrsbac-admin will be removed too?19:22
micahgudienz: ok, so without someone looking after the package it ends up being buggy, it should be removed from Ubuntu unless someone wants to look after it19:22
ari-tczewshould make sure whether there is b-d and depends on binary packages included in package which you want to remove19:25
IKnowWhoIAmguys i must tell you19:34
IKnowWhoIAmi know who i am19:34
IKnowWhoIAmi finally figured out why all tehse years i been missing the larger picture19:34
IKnowWhoIAmi am my own emitter19:34
IKnowWhoIAmi am whatever i choose to be.19:34
IKnowWhoIAmthis is incredible19:34
evilvish!offtopic ! IKnowWhoIAm19:34
ubottuError: I am only a bot, please don't think I'm intelligent :)19:34
IKnowWhoIAmthankyou for givin me this opportunity19:34
evilvish!offtopic | IKnowWhoIAm19:35
IKnowWhoIAmreal life is real19:35
ubottuIKnowWhoIAm: #ubuntu is the Ubuntu support channel, for all Ubuntu-related support questions. Please use #ubuntu-offtopic for other topics (though our !guidelines apply there too). Thanks!19:35
ari-tczewIKnowWhoIAm: http://t3.gstatic.com/images?q=tbn:ANd9GcQYW2037uD8kePM64SlDYLm-g0J0-CtSCB84LAZWECQLxY61dhi&t=119:36
=== sconklin is now known as sconklin-lunch
_UsUrPeR_hey all. I am having a heck of a time installing software RAID with LVM in Ubuntu 10.04 Server x64. Check out this pastebin: http://pastebin.com/4rDmyjyM . Is software RAID + LVM capable of being installed? If so, is having a /boot partition on a software RAID possible?20:06
broder_UsUrPeR_: this channel is for development, not support. please ask support questions in #ubuntu or #ubuntu-server20:08
ari-tczewwhen I should change build-depends from libwebkit-dev to libwebkitgtk-dev?20:09
=== sconklin-lunch is now known as sconklin
_UsUrPeR_will do. my apologies.20:09
* _UsUrPeR_ tips his hat20:09
ari-tczewtumbleweed: ^^ I saw you have an upload like mentioned above20:09
tumbleweedari-tczew: libwebkit-dev didn't exist for a while, but there's a dummy transitional package now, so you probably don't need to20:10
ari-tczewtumbleweed: look, bug 706982 I'll upload next revision of python-jswebkit so I can change it20:11
ubottuLaunchpad bug 706982 in python-jswebkit (Ubuntu) "No change rebuild of python-jswebkt 0.0.3-1build1" [Undecided,In progress] https://launchpad.net/bugs/70698220:11
tumbleweedari-tczew: I'd stick with a no-change rebuild20:12
kklimondaari-tczew: it would make more sense to fix that on the Debian level, and not introduce unnecessary delta. I think the libwebkit-dev package is going to stay for a while.20:13
ari-tczewok20:13
ari-tczewI hope that toolchain of Ubuntu 11.10 won't be too much different. :(20:25
Kanohi, will somebody update ntfs-3g to 2011.1.15?20:31
ari-tczewKano: file a bug about it20:33
ari-tczewand add tag 'upgrade'20:33
micahgKano: actually upgrade-software-version20:47
broderhmm...has anybody tried to duplicate snapshot.debian.org for ubuntu? presumably the awesome version would use lp/the librarian as its data store, though i don't know where you'd get the Release,etc. files from20:50
tumbleweedcan a core-dev please accept lucid and maverick tasks on bug 683189?21:10
ubottuLaunchpad bug 683189 in xdg-utils (Ubuntu) "xdg-open fails with spaces in filename" [Undecided,Fix released] https://launchpad.net/bugs/68318921:10
=== dendrobates is now known as dendro-afk
=== dendro-afk is now known as dendrobates
kirklandtumbleweed: done21:22
tumbleweedkirkland: thanks21:23
DktrKranzhyperair: packages accepted, with a comment21:34
smosermaybe someone can help. https://launchpad.net/ubuntu/+source/linux-ec2 says that lucid has a linux-ec2 package (version 2.6.32-312.24 ) that was in proposed "13 days ago"21:47
smoserbut as far as I can tell, it is not in the archive21:48
smoseralthough there are linux-ec2-doc and linux-ec2-source packages that match that version21:48
StevenKsmoser: Reading the Packages.gz file manually shows them21:49
JontheEchidnacjwatson: you rock! Thanks for the work on bug 68032821:50
ubottuLaunchpad bug 680328 in qapt (Ubuntu Natty) "Many postinst scripts fail using either AptDaemon, PackageKit, or QApt" [Critical,Confirmed] https://launchpad.net/bugs/68032821:50
smoserStevenK, really ?21:51
smoser wget http://archive.ubuntu.com/ubuntu/dists/lucid-proposed/main/binary-i386/Packages.gz21:51
StevenKsmoser: http://pastebin.ubuntu.com/557858/21:53
smoserright. there is no linux-ec2 metapackage there.21:53
StevenKThat isn't built by the linux source package21:54
StevenKsmoser: https://launchpad.net/ubuntu/+source/linux-meta-ec2 builds that21:55
StevenKAnd a 312 for that hasn't be uploaded to lucid-proposed21:55
smoseryea. so we need a build of that up21:56
smoserthanks StevenK21:56
StevenKsmoser: Welcome21:56
sconklinStevenK: you here? Could you publish that meta-ec2 package for lucid if you haven't already?22:24
StevenKsconklin: Aye, looking22:34
sconklinStevenK: thanks22:34
YokoZarWhoops!  System->About Ubuntu  (on 10.10) shows..."Welcome to 11.04 Natty"22:37
ari-tczewI saw bug today about it22:39
StevenKsconklin: All done.22:41
sconklinThanks!22:41
StevenKsconklin: It's currently pending, it will get published in the next hour or so22:42
=== sconklin is now known as sconklin-away
=== jtechidna is now known as JontheEchidna
jelmerbdrung: thanks for the syncs!23:27
bdrungjelmer: you're welcome. sorry for the fakesync mistake.23:31
kirklandsmoser: what are you looking for, kernel-proposed wise?23:32
bdrungjelmer: that happens if one application (ack-sync) doesn't explain what it does and another (syncpackage) has a bug.23:32
kirklandbdrung: https://launchpad.net/ubuntu/natty/+queue?queue_state=0&queue_text=errno is waiting in the NEW queue23:34
bdrungkirkland: thanks.23:35
bdrunglifeless: ping23:36
* bdrung pokes lifeless to process bug #52468023:36
ubottuLaunchpad bug 524680 in ubuntu-dev-tools (Ubuntu) "[lp-project-upload] not really ubuntu specific" [Wishlist,Triaged] https://launchpad.net/bugs/52468023:36
lifelessbdrung: hmm ?23:40
lifelessbdrung: ah, I see. updated.23:42
bdrunglifeless: whom can i poke to get in pulled into lptools?23:44
smoserkirkland, i guess it was taken care of. thanks.23:45
kirklandsmoser: sweet23:45
lifelessbdrung: dobey, poolie23:47

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