/srv/irclogs.ubuntu.com/2010/07/23/#ubuntu-devel.txt

=== dendro-afk is now known as dendrobates
=== yofel_ is now known as yofel
=== dendrobates is now known as dendro-afk
JoshStroblHello, I have packaged a folder into a .deb and using the "data" folder I directed the files to install in the /usr/share/themes and /usr/share/icons folders, however I realized that they need to be installed in the home directory. How would I be able to make the .deb install in the user's .themes and .icons folders?03:03
micahgJoshStrobl: try #ubuntu-packaging03:05
JoshStroblOk03:05
micahgkees: ping04:41
hyperairhuh, dpkg no longer has perl? then what'll happen to debhelper?05:15
wgranthyperair: dpkg, not dpkg-dev, is become Perl-less.06:51
hyperairwgrant: aah i see.06:51
wgrantdpkg-dev still has an awful lot of it :(06:52
bilalakhtarwgrant: are you freeeee ?06:52
wgrantHm?06:53
bilalakhtarwgrant: can you sponsor bugfix #55095506:53
bilalakhtarbug #55095506:53
ubottuLaunchpad bug 550955 in software-center (Ubuntu) "About window is modal and doesn't look like it" [Low,In progress] https://launchpad.net/bugs/55095506:53
wgrantbilalakhtar: I can't upload to main.06:53
bilalakhtarwgrant: you are not core-dev ?06:54
wgrantNo.06:54
* bilalakhtar mistook wgrant to be a core-dev06:54
wgrantI'm a Launchpad dev, and formerly an active MOTU, but never a core-dev.06:54
=== dendro-afk is now known as dendrobates
=== nenolod is now known as moogle
=== nobuto is now known as nobuto1
pittiGood morning07:27
ogracan anyone from the MIR team process/pre-promote the shotwell MIR (bug 607758) it breaks all images currently08:32
ubottuLaunchpad bug 607758 in shotwell (Ubuntu) "[MIR] shotwell" [Wishlist,New] https://launchpad.net/bugs/60775808:32
falehello08:56
falehow does ubuntu server creates ISOs? Does they use a script?08:57
=== almaisan-away is now known as al-maisan
=== fta_ is now known as fta
=== ion_ is now known as ion
=== Amaranth_ is now known as Amaranth
* ogra wonders why "buildlive base" also fails with gtk/shotwell errors09:20
ograthere shouldnt be *any* graphical stuff in a base livefs09:21
ogracjwatson, ^^^ any idea ?09:21
slangasekEtienneG: no profile exists for pam_group because pam_group is blecherous and not to be encouraged.  A separate package outside of pam, which depends on the correct bits of pam, could provide such a profile.  The naming convention is s/^pam_// + .* on the module name.09:22
=== oubiwann is now known as oubiwann-away
ograhmm, seems i should wak up before reading logs09:40
* ogra looked at the wrong log09:40
cjwatsonogra: I am on top of shotwell etc.; you can skip it09:52
ogracjwatson, well, i'm resorting to base for my testbuilds now09:55
* ogra doesn really care for the image content atm09:55
=== ogasawara_ is now known as ogasawara
=== fta_ is now known as fta
=== BlackZ_ is now known as BlackZ
cjwatsonzul: I checked with Keybuk; the problem with echoing a timeout message is that it basically doesn't work very well right now.  If you do 'start mysql' in X then the message will go to /dev/console, not your terminal!  On the whole, at the moment we reckon it would be better to simply remove both the timeout message and 'console output' - although 'console output' is safe in itself (as opposed to 'console owner', which is ...10:25
cjwatson... what I was remembering)10:25
zulack..in a meeting right now but ill get to it first thing10:25
cjwatsonzul: thanks10:28
glickhi guys, im not sure if there is a serious bug in ubuntu, or if im simply using it wrong10:41
glickI got folder sharing to work, using System->prefs->personal file sharing, and then right clicking on my public folder, and selecting share this folder10:42
glickhowever, when i un check share this folder, and reboot10:42
glickthe folder is still shared10:42
glickis that a bug?10:46
glickor am i not doing something right?10:46
=== fta_ is now known as fta
bigonis there an issue with amd64 buildd?10:48
glickis there an open bug report on not being able to unshare folders in ubuntu 10.04?10:49
glickfor some reason i cant unshare a folder that i once shared10:53
glicki unclick share folder10:54
glickeven reboot10:54
glickyet its still shared10:54
glickif i type net usershare info10:54
glickit doesnt show up, but its still shared10:54
=== fta_ is now known as fta
glickdoes anyone know how i unshare a foldeR?11:00
glicki unclick the share button, yet its still shared11:00
glickno output is generated from the comman net usershare info11:01
glickyet its still shared11:01
=== fta_ is now known as fta
LucidFoxEep, what happened to Launchpad?11:34
=== hunger_ is now known as hunger
bilalakhtarSomeone, please sponsor fix for bug #15593011:48
ubottuLaunchpad bug 155930 in synaptic (Ubuntu) ""Unmark all" clears the package list (!)" [Low,In progress] https://launchpad.net/bugs/15593011:48
=== al-maisan is now known as almaisan-away
hyperairLucidFox: looks fine to me.11:57
LucidFoxHmm, now it works... Used to give a "connection time out" error11:58
hyperairhow strange.11:58
=== fta_ is now known as fta
=== fta_ is now known as fta
ogracjwatson, hmm, for-project doesnt know about base/livecd-base ?13:09
cjwatsonogra: it's a special case for some reason I do not recall; the crontab line is build-livecd-base13:27
ogracjwatson, yeah, i looked at that, it doesnt actualy roll an image it seems, just copies the livefs around13:30
ograjdstrand, could you let linux-ti-omap4 out of NEW so we can build images again ?13:41
jdstrandogra: k13:41
ograthanks :)13:41
jdstrandsure13:41
cjwatsonogra: 'buildlive base', according to the crontab13:49
cjwatsonnot for-project13:49
ogracjwatson, yes, but how do i roll an image out of that without for-project ? :)13:49
ograsure i get a base filesystem, that worked just fine13:50
ograbut i somehow need to call cron.ports_daily-preinstalled13:50
cjwatsonogra: dunno, it wasn't designed for rolling images, it was designed for dumping out a squashfs that lamont could customise13:50
ograah13:51
lamontbase isn't shipped13:51
lamontor even generally built, afaik13:52
ogralamont, right, i'd like to be able to use the base fs with for-project13:52
ograi think its in the crontab13:52
lamontfor-project?13:52
ogra6 4 * * *buildlive base; build-livecd-base13:52
ografor-project is the script from debian-cd that rolls the actual images13:53
ograi guess i need to add it then13:53
lamontah, ok.13:53
lamontbase comes from a time when ubuntu-base was a valid meta-package.  not sure it even lets one login13:54
lamontthen again, I haven't done anything with base in at least 4 years.13:54
lamontwow.  time flies13:54
ograheh13:55
ograwell, livecd-rootfs installs ubuntu-standard if base is selected13:55
lamontmuch more likely to be usable then13:56
ograthrough the prreinstalled path it also pulls in jasper and oem-config ... so it creates a perfect cmdline image13:56
lamontpart of me wants to speculate that the target was non-desktop machines and having a livecd for them13:56
ogralike a 256M beagleboard ;)13:56
lamontogra: I think the original target was maybe hppa13:57
lamontbut the beagle board is much  more cool13:57
ograwell, so be beagle the new hppa then :)13:57
jpds.13:57
lamontogra: no.  no it is not.13:58
smosercjwatson, so you think it would be ok for eucalyputus node controller to run-time depened on grub-rescue-pc, and mount grub-rescue-floppy.img, replace config, add user supplied multiboot image and then boot it14:28
=== almaisan-away is now known as al-maisan
pittiev: new jockey uploaded, BTW; please let me know if you have trouble with it14:33
evpitti: yay!  thanks!14:34
apwcjwatson, i wonder if it might make sense to rescore the remaining gcc-4.4 builds in maverick14:46
dokoapw: I don't think so, because Linaro is enabled only on i386 and amd6414:47
pittiI thought powerpc as well?14:47
apwdoko, not on ports ?14:47
dokopitti: we will, but only the generic parts, not the powerpc specific backports. these won't be part of linaro gcc-4.5. if these changes would go into 4.4, they would be seen as an regression compared to 4.514:49
dokoapw: no, it was never the plan to enable these for sparc and ia64 (which are likely to go away in maverick)14:50
apwdoko, though we need to upload the kernel after a compiler version number change, so we need to wait for them all to finish either way14:57
cjwatsonsmoser: I *think* so14:58
cjwatsonsmoser: in a "twitch" kind of way :-)14:58
cjwatsonapw: I've scored them up anyway14:59
smoserwe had said build time depend, but that seems more difficult to fix as it might require a no source change rebuild of eucalyptus-node controller to deliver the new grub rescue image.14:59
cjwatsondoko: ... and armel, according to the changelog15:00
smoserthe thing I *dont* like is that in dynamically creating it, a grub update could break running of guest instances.15:00
dokocjwatson: armel had the linaro changeset already turned on with the last upload15:00
cjwatsonsmoser: that's why I suggested grub-rescue-pc rather than grub-pc, since grub-rescue-pc doesn't do any bootloader installation (if I'm understanding you correctly)15:00
cjwatsondoko: ah, ok15:00
smosercjwatson, right.15:01
smoseroh, cjwatson, grub-rescue-pc will get into main, right ?15:02
smoserdo i need to MIR that ?15:02
cjwatsonsmoser: binaries from a source already in main don't need an MIR15:06
smoserthats what i had thought, but i had become confused.15:07
pitticjwatson: I'm about to move the kernel to -updates; do you want to handle d-i?15:10
=== fta_ is now known as fta
ScottKpitti: Would you please rescore qt4-x11.  It finally built on armel last night, but now there's a new upload so archive skew is about to kill our ability to fix FTBFS if it doesn't get going (this would just affect powerpc and armel).15:28
cjwatsonpitti: ok, let me know when I can15:28
pitticjwatson: you can now, I moved it15:28
cjwatsonI'll wait 'til publish-distro finishes15:29
gesercould someone please give-back "bsh" on i386. It's in depwait on "default-jdk-builddep" which is provided by gcj-native-helper which is now in main on i386 too.15:37
sorengeser: done15:38
smosercjwatson, maybe i was confusing above.  the grub-rescue-floppy.img is a iso9660 fs, so i can't write to it.15:41
smoser(i was thinking it was a floppy disk image -- ext2 or something -- that i could mount, replace config and unmount)15:41
rsalvetikirkland: where is the qemu page that you were talking about?15:46
rsalvetiwant to look just to see what is tested and "supported" :-)15:46
cjwatsonzul: regarding bug 608423, were you planning to do anything about point 2 in the bug report?16:09
ubottuLaunchpad bug 608423 in mysql-dfsg-5.1 (Ubuntu) "[lucid-proposed] post-start script broken" [Undecided,New] https://launchpad.net/bugs/60842316:09
zulcjwatson: not yet...i need to find a better way to address it16:10
cjwatsonzul: upstart script fragments are always run with 'set -e', so the reporter's right16:10
cjwatsonzul: I think you want:16:10
cjwatsonret=016:10
cjwatson/usr/bin/mysqladmin --defaults-file=$HOME/debian.cnf ping || ret=$?16:10
zulcjwatson: good idea16:11
cjwatsonthat's my standard construct for these things anyway16:11
zulcjwatson: uploaded16:15
cjwatsonzul: thanks, will look when it generates the diff16:17
zulcjwatson: no probs16:17
cjwatsonpitti: could you look at ubiquity/lucid-proposed?16:20
pitticjwatson: sure16:23
pittiERROR: queue does not have a debdiff16:23
pittimeh16:23
pittiold-style then16:23
pitticjwatson: if I may ask for gparted in return?16:24
* pitti rejects the older mysql upload16:24
pittizul: ^ FYI16:24
zulpitti: ty16:24
pitticjwatson: done16:27
cjwatsonpitti: ah yes, will look16:31
cjwatsondone16:33
=== dendrobates is now known as dendro-afk
=== fta_ is now known as fta
=== beuno is now known as beuno-lunch
micahgif I have a segfault in an armel build failure, could that be a transient error?17:34
=== al-maisan is now known as almaisan-away
=== beuno-lunch is now known as beuno
=== fta_ is now known as fta
=== beuno_ is now known as beuno
ebroderCan anybody give me some pointers on figuring out how the live CD build works?20:34
=== fta_ is now known as fta
=== sebner_ is now known as sebner
achiangebroder: what's your question"20:52
achiang?20:52
ebroderI want to roll a custom live CD and I'd like to start from the same process that the official CDs use20:52
ebroderAnd I have no idea where to start looking20:52
achiangoh, i don't know what process the official CDs use, but you could probably start with apt-cache show live-helper21:00
lamontebroder: it's not pretty. nor does it port well21:06
lamonthttps://help.ubuntu.com/community/LiveCDCustomization <-- ebroter21:07
lamontebroder: even21:07
lamontthe livecd-rootfs package builds the ugly mess that is the root filesystem.  while nearly unreadable, that bit is pretty straight forward.  replacing that rootfs on the CD is the easiest way to go for mass changes.  the scriptage to take the livecd rootfs and wrap an iso around it is, um, yeah.  I've avoided learning anymore about that than I have had to.21:08
ebroderOk, that's helpful. Do you know where the code to transform the rootfs into an iso is?21:10
ebroder(is it in livecd-rootfs?)21:11
knittlhi. does anybody know what "status 5" of ureadahead is?21:49
knittlbooting works fine, but ureadahead terminates instantly upon boot21:50
knittland i can't find proper documentation of ureadahead termination status codes either21:54
=== fta_ is now known as fta
chughgaurav_I know Java , can I contribute  in the development of Ubuntu ?22:06
=== moogle is now known as nenolod
=== almaisan-away is now known as al-maisan
jbebelIs there anything that can be done to accelerate an archive.ubuntu.com packages file update?22:59
jbebelLucid isn't installable using d-i right now.22:59
jbebelbecause linux-meta depends on a kernel that doesn't exist in the Packages file, though it is in the pool.22:59
jbebelI'd also like to know why linux-meta was published 6 hours before the kernel it depended on.23:00
jbebelLooks like i made it in now.23:08
=== fta_ is now known as fta

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