/srv/irclogs.ubuntu.com/2008/01/24/#ubuntu-devel.txt

articpenguin3800will ext4 ever be in ubuntu00:01
RAOFYes.  Your actual question is "will ext4 be in Ubuntu before it's in the mainline kernel".00:01
RAOFTo that question I don't have an answer.00:02
articpenguin3800it got merged in 2.6.1900:02
articpenguin3800and ty tso said the stable version will be in 2.6.2500:03
RAOFWell, that suggests that Hardy+1 should have ext4 enabled, at least.00:03
articpenguin3800prob wont be default though00:05
RAOFIndeed.00:05
RAOFI'm personally very happy for not-hugely-tested-for-many-years filesystems to not be default.00:05
articpenguin3800to me its just ext3 with extents00:06
RAOFSo it'll need a few less years testing before it's well tested.00:07
=== BenderUn1t22 is now known as BenderUnit22
=== kyle__ is now known as kylem
BlackDiamondsis there a dev roadmap somewhere so once can see the features planned for hardy ?03:16
RAOFblueprints.launchpad.net/ubuntu is about it.03:17
BlackDiamondsthanks again RAOF03:20
BlackDiamondsalso, while you are here, how do I find out which packages are in main and which ones are in restricted ?03:20
RAOFapt-cache policy <pkgname> is a good tool03:21
BlackDiamondsthank you once again03:21
* emgent night.03:37
=== bigon` is now known as bigon
=== \sh_away is now known as \sh
\shmoins07:51
\shpitti, ping seahorse, did you ever try it with a gpg smartcard reader?08:50
\shpitti, seahorse can't cache the keys passphrase...so it can't be used with a smartcard reader, or you set e.g. your keyid explicitly with (debuild syntax) -k'0x<yourkeyid>!' <- the exclamation mark is important08:54
\shpitti, the set keyid is not the keyid of the signing key on the card FYI08:56
pitti\sh: smartcard> no, I don't have such a thing09:03
\shpitti, see :) and I'm hitted by this phenomenon...actually "don't cache passphrase" helps or setting -k'0xC098EFA8!' in my case..this will override gpgs default checking for a signature key09:04
slangaseklamont: yes, it was too late then because I had left for dinner; it's not too late now because I'm in the office09:25
=== \sh is now known as \sh_away
ogracanonical, the company with the time shifting office ....09:27
StevenKIt's a feature09:27
ion_I want a controllable time dilation field generator.09:27
RAOFion_: I'll go you halves!09:28
ion_-EPARSE09:29
=== \sh_away is now known as \sh
ion_sh: Thanks for the information!09:30
\shpitti, I debugged a bit further...09:31
slangasektjaalton: can you clarify for me why linux-restricted-modules has introduced a new "dummy package for easy transition" that has never existed in Ubuntu before? (fglrx-amdcccle)09:31
\shpitti, it looks like that you need restart the pcscd (the daemon for accessing at least this reader)09:31
\shpitti, after that you get this output http://paste.ubuntu.com/3825/09:32
ion_slangasek: IIRC the changelog said it’s for transition from the AMD debs.09:32
\shpitti, and then it works...but with pinentry and with no seahorse dialog09:32
tjaaltonslangasek: people are using the ATI installer generated debs, and upgrading to a new release fails09:32
tjaaltonslangasek: and debian does it too09:32
tjaaltonslangasek: I mean they have that package as well09:33
tjaaltonfor the same purpose09:33
\shrestarting session09:35
=== \sh is now known as \sh_away
ion_Away nicks ftw.09:35
tjaaltonslangasek: actually, I haven't checked what version numbers the ati debs use..09:36
tjaaltonbut we have an epoch now, so it should upgrade nicely09:38
=== \sh_away is now known as \sh
slangasektjaalton: ok; that was the speculation, but I wanted to confirm since the changelog/package description seemed ambiguous to me, thanks09:40
tjaaltonslangasek: yeah, mostly copied from debian :/09:41
\shpitti, it looks like that only gpg-agent works together with the pcscd nicely including caching of the smartcard pins09:41
\shpitti, with only seahorse-agent in this session it doesn't work09:41
davmor2just wanted to say gvfs buggy but much better nice work :)09:41
=== gouki_ is now known as gouki
=== \sh is now known as \sh_away
=== \sh_away is now known as \sh
sorencjwatson_: Would you object in any way if I changed openssh to generate the host keys from the init script, if they don't exist instead of doing it at postinst time?10:01
=== fabbione is now known as thegodfather
coolbhavinot able to change USB mouse to ps/2 mouse.. system is not recognising the mouse10:14
coolbhaviplease help10:14
TheMusocoolbhavi: #ubuntu for support.10:15
=== lifeless_ is now known as lifeless
\shcjwatson, ping console-data, you added with the last merge the decision-tree* targets in debian/rules right?11:28
* pitti defeats grub and does the hardy dance12:26
looldoko: I think you were proposing using -Bsymbolic by default; a blog post on today's planet gnome might interest you: http://www.gnome.org/~michael/activity.html#2008-01-2312:27
loolHmm was mentionned yesterday as well12:27
slangasekpitti: is the hardy dance less or more embarrassing than the warty dance? :)12:34
* Hobbsee waves to pitti. finally caught up, have you?12:35
pittislangasek: I think it gets worse with every release12:35
pittihey Hobbsee12:35
=== DrKranz is now known as DktrKranz
\shinfinity2, bug #183495 please share the crack pipe you smoked to realise your idea ;)12:45
ubotuLaunchpad bug 183495 in openbios-sparc "[FTBFS] openbios-sparc (1.0~alpha2+20070816-1) fails to build in hardy" [Medium,New] https://launchpad.net/bugs/18349512:45
dokolool: yes, but it is -Bsymbolic-functions12:46
lamont\sh: the only "workable" idea I've seen for something like 183495 is what palo does... and it's broken-ish13:04
\shlamont, for me it looks like that every idea we had to deal with those packages, is somehow broken-ish...and imho involves all the time a buildd admin to install/remove/or whatever it needs13:05
lamontpalo builds the offensive binary bit in the clean targetr.13:06
\shlamont, well this doesn't help when you don't have <insert your arch here you need to build <arch>-bios>13:06
\shlamont, well, adams idea would work if I would knew how to fetch an alien build-dep, e.g. i386 would build an Arch: all package, and the source pkg depends on the _sparc.deb...who do you do this in an automated environment?13:08
\shs/who/how/13:09
lamont\sh: you have the requisite pieces of said alien-arch deb in the source package (since fetching the deb isn't reliably doable)13:10
\shlamont, so you upload first the source-package which builds the e.g. bios-sparc.deb on the sparc buildd and then you fetch it from sparc archive and push it into a new source package which build Arch: all packages...13:12
lamontbinary builds can't create source packages.13:13
\shlamont, the other way around...13:14
\shyou need to create a new source package first, which uses the already available alien arch .deb to produce an Arch:all package13:14
StevenKbryce:13:26
StevenKOoops13:26
ScottKI would appreciate it if whoever is arvchive admin for today would please have a look at libetpan in binary New for dapper-backports.  It's the last manual step needed in the clamav backport to Dapper.  It's holding up the sylpheed-claws-gtk2 backport (depwait) and so the currently available dapper sylpheed-claws-gtk2 is not compatible with the current dapper-backports clamav.13:39
ScottKThanks.13:39
* lamont notes that (at least in gutsy) pciutils depends on a lib in oldlibs14:02
\shseb128, mr. ubuntu-gnome :) what can I say to gnome not always opening a window when an LVM device is generated by sbuild ?14:05
\shseb128, I just deselected all hotplug and other media to be appearing on my desktop...but LVM devices are just popping up like mad14:06
seb128\sh: no idea, might be a gvfs thing14:12
\shseb128, any idea where I can plug my brain in and find a solution to stop this annoyance? in some hidden gconf settings eventually?14:13
\shseb128, it's the gvfs mounttracker who receives the dbus messages..and act on it...where can I adjust those settings?14:20
seb128\sh: not sure you can, gvfs is really new14:56
seb128\sh: can you describe how to trigger the bug?14:57
\shseb128, well, use sbuild with LVM and start building a package...sbuild will create a snapshot lvm mount it and voila...you have a window with the contents of this mount14:58
ion_Yeah, Nautilus seems to open a new window for everything that is mounted now.14:58
ion_If i go to sftp://foo and gvfs mounts it, in addition to the active Nautilus opening it, a new window is opened as well.14:59
\shseb128, which is in fact a normal behaviour for normal mounted devices, but using snapshots this can be annoying14:59
\shseb128, forget my statement about "normal behaviour" Re: ion_15:00
* lamont stabs at cdbs15:00
pittimmmm cdbs :)15:00
* Hobbsee stabs at lamont15:01
* ion_ stabs at Hobbsee’s knife15:01
\shpitti, saw my comment about seahorse? (filed a bug btw) I removed seahorse from Xsession.d now and back to gpg-agent15:01
pitti\sh: no, I didn't saw it15:02
pittiand I'm not really a seahorse expert unfortunately15:02
lamontpitti: I mean, what's so hard about writing an actual makefile that people find the need to bury everything under magic callback hooks that are burried 4 levels of included-makefiles away?15:02
pittilamont: because we rather change build policy at one place rather than in 500 source packages15:03
Hobbseeion_: i don't have a knife.  i don't *need* a knife.15:03
\shpitti, well, this regression will hit everybody who is using a gpg smartcard (e.g. some of FSF people do) (bug #185563)15:03
ubotuLaunchpad bug 185563 in seahorse "[hardy] seahorse agent doesn't work with gnupg smartcard readers" [Undecided,New] https://launchpad.net/bugs/18556315:03
lamontpitti: never mind reviewing those packages to make sure it's the right change, etc. :-)15:03
pittilamont: stuff like fixing translation domains, producing .pot files, symlinking gnome help files, etc.15:03
pittilamont: with your argument you can also argue that using debhelper is bad because you don't see the raw commands that are executed15:04
pittialso, what's the point of writing 50 lines of boilerplate every time?15:06
lamontpitti: the best thing I can say about cdbs is "it's not DBS"15:07
lamontor rather, "It's a vast improvement  over DBS"15:08
slangasekgar, is there any way at all to remove an update-alternatives slave without removing the master?15:08
lamontslangasek: vi?15:08
lamontor were you meaning in a postinst?15:08
slangaseklamont: you mean /usr/share/cdbs/1/class/vi.mk?15:08
slangasekyes, in a postinst15:08
lamonthrm.. that would tend to lead one to really wanting to use update-alternatives, wouldn't it.15:09
lamontsed would be bad. :-)15:09
slangasekdon't tempt me15:09
slangaseksed is my weakness15:09
pittiuse perl then :-P15:10
lamontor patch update-alternatives?15:10
Mithrandirlamont: cdbs solves a completely different problem than DBS does.15:11
Mithrandirslangasek: remove both, reinstate the master?15:11
slangasekpitti: "it's my weakness" as in "I can't resist it"15:11
slangasekMithrandir: that would kill the user's selections15:11
Mithrandirslangasek: you could save it..15:12
pittisoren: hm, building kvm-source now changed kvm from hanging in X to immediately segfaulting15:13
=== Amaranth_ is now known as Amaranth
=== beuno_ is now known as beuno
slangasekMithrandir: apparently I can get away without doing that, I can just take over the slave name with a new path and it DWIM15:24
sorenpitti: Ping me when you're back at your seat.15:29
seb128\sh: you can unactivated the new mounts window opening in the media tab in the preferences dialog15:48
\shseb128, I did already...doesn't help15:48
seb128weird15:48
\shseb128, yepp15:49
\shseb128, I can see some hal rules which are triggering that...15:51
\shseb128, and I don't know if they are handled like removable or hotplug media15:52
seb128no idea15:52
asac_the_2ndANNOUCE: network manager 0.7 for hardy in my ppa ... it could deserve some moderate testing. remember to get the packages it would upgrade to your cache so you can downgrade without network in case it breaks :)15:52
asac_the_2ndhttps://edge.launchpad.net/~asac/+archive15:52
seb128require debugging and I've no similar setup nor willing to configure sbuild only to look at that we have a lot of other issue which impact on normal desktop use15:52
Hobbseeasac_the_2nd: are they working for at least 1 user already?15:52
seb128asac_the_2nd: that one has the dbus configuration?15:52
seb128Hobbsee: I'm running it since this morning without issue15:53
Hobbseecool, OK15:53
tjaaltonsweet, fedora will use upstart15:59
asac_the_2ndseb128: yes ... it should be fine by now16:12
asac_the_2ndHobbsee: sure :) ... i guess a single person can live without network :)16:13
Hobbseehehe :P16:13
asac_the_2ndsoren: ready?16:14
sorenasac_the_2nd: I just finished the un-cancellable stuff,yes.16:15
sorenasac_the_2nd: And you want me to make sure that I have packages for my current existing network-manager, libnm-util, and wpasupplicant around?16:18
sorenasac_the_2nd: Ok, here goes nothing.16:20
=== bryce_ is now known as bryyce
tjaaltonasac_the_2nd: the new n-m failed on my laptop, doesn't recognize the devices at all16:45
emgentkeescook, ping16:53
emgenthard idle :°16:54
keescookemgent: pong, hi!17:03
emgentheya :)17:04
emgentall good? :)17:04
emgentmalone 17315317:05
ubotuLaunchpad bug 173153 in audacity "[CVE-2007-6061] Denial of service and deletion of an arbitrary directory tree via symlink attack" [Undecided,Confirmed] https://launchpad.net/bugs/17315317:05
emgentgutsy feisty e dapper ready, when you have time to upload, rembmer :P17:05
=== infinity2 is now known as infinity
ion_sh: Yay for ruby1.9! Thanks for merging. :-)17:16
\shion_, hope it works17:17
loolslomo: Do you have the libsoup2.4 2.3.0~r1050 tarball around?  I'd like to push it to hardy17:58
simon360anyone here able to give me some help? I want to replace grub with gfxboot on the live CD, so the installer won't install grub, but will install gfxboot instead18:03
simon360I also need a bit of help making a gfxboot theme :\18:03
simon360and am somewhat noobish at developing :P , although I've been a Linux user for a few years18:04
_MMA_simon360: Reconstructor is the easiest way to do this.18:05
pittisoren: almost, but not quite...18:30
sorenpitti: Erm... I'm sorry to hear that.18:31
sorenNo biscuit?18:31
pittisoren: I see the desktop, but no panel, etc.18:31
simon360_MMA_: I'm using reconstructor, but it doesn't replace grub with gfxboot afaik18:31
sorenpitti: -ESOMEBODYELSESPROBLEM18:31
sorenpitti: I blame seb :)18:32
pittisoren: I don't think so; the screen is cut off at the top18:32
pittisoren: the window is too small to see the entire screen, and I can't scroll18:32
simon360I have a simple gfxboot theme already, but I had something fancier in mind...18:33
simon360just a sec18:33
simon360http://img.skitch.com/20080124-g184as8q142wb4b82kq66jnhag.png18:37
simon360would something like that be easy/possible?18:37
simon360I know how to change the text, colors, backgtround, etc.18:38
simon360but moving the menu to the right is something I haven't figured out18:38
_MMA_simon360: The Ubuntu Forums #ubuntu-derivative or maybe #ubuntu-installer might be better places for this.18:39
simon360alright, thanks18:39
calc_MMA_: pitti said that we can demote ooo language-support-* depends once apt autoinstalls recommends18:40
calc_MMA_: so not yet but soon hopefully18:40
_MMA_calc: Awesome. Should shave a bit off Xubuntu and Ubuntu Studio disks. :)18:41
somerville32calc: Moving to Universe is not a demotion :P18:42
* somerville32 makes posters to help spread the new paradigm.18:42
_MMA_somerville32: He's just talking about moving deps. Not actual packages.18:43
somerville32haha, sorry, hair trigger ;]18:44
_MMA_Yeah. Holster that thing. ;)18:45
\shseb128_, this is really annoying...we need to fix that behaviour...18:52
seb128_you are welcome to send a patch, nobody else ran into the issue yet18:53
seb128_need to go to dinner now18:53
Kanohttps://bugs.launchpad.net/ubuntu/+source/xserver-xorg-video-ati/+bug/15197418:58
ubotuLaunchpad bug 151974 in xserver-xorg-video-ati "[RV410] X700SE [1002:5e4f] - DRI images corrupted" [Undecided,Confirmed]18:58
Kanodoes anybody like to compile a new mesa for gutsy?18:58
\shpitti, can you check this: http://pastebin.ubuntu.com/3838/ (this is send out from dbus when a snapshot lvm device comes up, and I don't find anything in lshal about it)19:06
calcsomerville32: not moving to universe, moving from Depends to Recommends19:22
slomolool: sure... one moment19:25
slomolool: http://slomosnail.de/~slomo/temp/libsoup2.4_2.3.0~r1050.orig.tar.gz19:26
slomolool: i hope it gets accepted in debian soonish19:28
=== \sh is now known as \sh_away
=== \sh_away is now known as \sh
=== bigon is now known as bigon`
\shI wonder if it's a sane setting for nautilus to set preferences->media_automount_open to false or to fix nautilus not to open local mounted devices, especially lvm snapshots20:53
=== \sh is now known as \sh_away
=== Ubulette_ is now known as Ubulette
=== jamesh__ is now known as jamesh
keisangihi there23:58
keisangii have a problem with ubuntu 7.10 on my lenovo thinkpad x60,23:58
Seveaskeisangi, support in #ubuntu23:59
keisangiinside eclipse, i have choppy sound and no keyboard responses23:59
keisangii'm developping a game in java using lwjgl/slick2d23:59

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