[00:01] d- as in cd-rom? it's just another acronym nobody understands [00:03] disk-bus [00:05] desktop bus. [00:05] or probably just the D bus these days [00:07] james_w - do the reporters get a dialog with what looks like a D-Bus error in them? [00:08] cjwatson: bug 442480 closed [00:08] Launchpad bug 442480 in console-setup "Recent update to console-setup is causing packages to FTBFS" [Undecided,New] https://launchpad.net/bugs/442480 [00:13] lamont: thanks [00:14] cjwatson: you OK with a mass-giveback? [00:14] mass give-back as long as it's zero-score would be fine, I think. The queue was kinda long last I checked === Riddelll is now known as Riddell [00:25] what package should rfkill switch bugs be filed on? my D430 kill switch kills the bluetooth & wifi, but when the normal position is restored it doesn't re-enable them [00:26] so, one needs to do "echo 1 | sudo tee /sys/devices/virtual/rfkill/rfkill1/state" which is a little kludgy [00:28] lifeless: Works here (also D430) [00:28] lifeless: 32 bit or 64 bit? [00:28] ScottK: 64 [00:28] lifeless, there is a bug already opened on that [00:28] Hmmm. I'm on 32 [00:28] superm1: ok, cool [00:28] it's a kernel bug against the dell-laptop kernel module [00:28] superm1: for the record, where should I have looked? [00:28] superm1: thanks [00:31] lifeless, bug 430809 [00:31] Launchpad bug 430809 in linux "[Dell Latitude D430, iwl3945] Wireless can't be activated after disabling kill switch" [High,Triaged] https://launchpad.net/bugs/430809 [01:13] Preparing to replace texlive-latex-base 2007.dfsg.2-4ubuntu1 (using .../texlive-latex-base_2007.dfsg.2-4ubuntu1_all.deb) ... [01:13] Reinstalling deleted mandatory conffile color.cfg [01:13] cp: cannot stat `/usr/share/texlive-base/color.cfg': No such file or directory [01:13] dpkg: error processing /srv/royal.buildd/home/buildd/build-karmic-autotest/chroot-karmic-autotest/var/cache/apt/archives/texlive-latex-base_2007.dfsg.2-4ubuntu1_all.deb (--unpack): [01:13] subprocess pre-installation script returned error exit status 1 [01:13] dear texlive. DIEDIEDIE [01:14] was that my outside voice? sorry [01:26] meh. nm [02:06] Gnometris 2.28.0 is a PIECE OF SHIT! [02:07] it doesn't do SHIT [02:08] !ohmy [02:08] Please remember that all Ubuntu IRC channels share the same attitude of providing friendly and polite interaction with all users of all ages and cultures. Basically, this means no foul language and no abuse towards others. [02:09] the game area [02:09] doesn't do [02:09] SHIT [02:10] heh what's this about? [02:10] in the terminal I get [02:10] (gnometris:8489): ClutterGLX-CRITICAL **: Unable to make the stage window 0x4400042 the current GLX drawable [02:11] Does anyone know what this shit means? [02:15] '-' [02:16] OH BOB SAGET! [02:21] O_O === statik` is now known as statik [02:50] so gwibber seems quite messed up these day :/ [02:50] is it going to make it into the final release as default? === Hellow_ is now known as Hellow [03:00] and I'm not understanding the gwibber-daemon thing === Yasumoto_ is now known as Yasumoto [04:09] dtchen, when you get a min can you pop in #mythtv to chat a little about the current situation with pulseaudio w/ mythtv and karmic. see http://svn.mythtv.org/trac/browser/trunk/mythtv/libs/libmyth/audiopulseutil.cpp for what's currently being done about it === Whoopie_ is now known as Whoopie [05:10] any ~ubuntu-sru people here? I'm waiting for an SRU for... almost two months? :-( [05:21] hi. im in the process of compiling an older cross compilation toolchain. several things fail with gcc-4, so i installed gcc-3.4 and have been passing it via CC=gcc-3.4 [05:22] i'm wondering though, if there is a way to manage symlinks in /usr/bin, eg, to have gcc point to 3.4 then be able to switch it back to 4.x [05:22] update-alternatives doesn't seem to recognize 3.4 as a provider of 'cc' [05:23] Mine is a symlink to gcc-4.3 directly; it doesn't use update-alternatives (/etc/alternatives/gcc) [05:23] so I'd just change the symlink by hand. [05:23] actually you're right about it not using that [05:23] i forgot about the symlink chain that would point to an intermediate link in /etc [05:26] Hi, is it possible to catch N-key rollovers and have them passed to an application? [05:26] Please [05:26] By the way [05:26] Hello [05:32] somebody__: wrong channel for that question, I would find a programming related channel [05:35] They said in #perl that it depends on the OS's handling [05:35] I've been to programming and back [05:36] cool wel the toolchain built [05:36] thanks [05:36] somebody__: Ubuntu OS related questions still don't go here... [05:37] this is for development OF Ubuntu, not ON Ubuntu or FOR Ubuntu [05:40] I'm trying to add this application to my official stack for Ubuntu [05:40] It will help the many that use Ubuntu [05:40] well once you are ready to begin packaging of the software, #ubuntu-motu would be glad to help [05:41] but this place is not the right place to ask Ubuntu related programming questions; the topic is for coordination of the development of ubuntu itself [05:41] developers tend to read everything that's said in here and would appreciate a better signal to noise ratio [05:43] For goodness sakes. The other channel is noisy and being of no help [05:43] besides, people in here probably don't know as much about GUI stuff with perl as they do about systems programming :) [05:43] Do you have any idea what kind of noise a screen reader can generate when in an IRC room [05:44] ? [05:44] I am trying to improve Ubuntu as a whole for those in my comunity [05:45] that isn't the topic of this channel either. [05:45] Is that really a crime [05:45] ? [05:45] somebody__: It's not a crime, but trouble getting help elsewhere doesn't magically put your question on topic for this channel. [05:47] Which channel is it on topic for then? [05:47] I'm not here to cause trouble [05:48] I've been Googling for some time now and this is my last resort [05:48] If you have a lead for me to follow, by all means post it here [05:52] somebody__: I would look at handling via xorg input for the keyboard..also find the source for 'stickey support' in KDE as it has rollover stuff in there but I am sure is probably used via the Solid library for KDE...I don't know of any floss apps that support it to even look at...maybe find some old games that have since been open sourced that might have supported it...but ya, I have to agree a bit, this chan isn't best suited for this === fabrice_sp__ is now known as fabrice_sp [06:00] Thanks fo your help === highvolt1ge is now known as highvoltage [07:09] good morning [07:13] slangasek: did you have a chance to check out the acpi-support sponsoring bugs? [07:15] 425155 and 432578 === porthose is now known as porthose|afk [07:26] Good morning [07:35] pitti: good morning [07:36] hey stvo [07:41] kees: xsplash> yes, both you and lool ignored Vcs-Bzr:; I'm committing both of your changes now, but please do respect it in the future, otherwise we'll continue to drop patches [07:47] kees: why is setresuid() any better than setuid()? I find the latter easier to use, and it should do the same for a root process [07:48] what does the release team say about bug 440203 [07:48] Launchpad bug 440203 in onboard "Freeze Exc. Req. - New release - diff.gz supplied" [Undecided,New] https://launchpad.net/bugs/440203 [07:48] pitti: Yeah sorry, I realized there was one when kees mentionned it in a random other bug; I think I was too stressed by the beta critical issue and just uploaded [07:49] don't worry, committed now [07:50] thanks === dendrobates is now known as dendro-afk [08:23] Anybody from ubuntu-sru around who could ACK bug #330766? It's affecting our site deployment, and we've already tested the fix [08:23] Launchpad bug 330766 in pulseaudio "pulseaudio hangs, prevents login, home as ntfs" [Unknown,Fix released] https://launchpad.net/bugs/330766 [08:37] hmm lyx seems to fail miserably with ibus [08:37] at least, with ibus active, the open/save dialogs don't work [08:42] Good morning! [08:46] hey dholbach - about bug 406077: [08:46] Launchpad bug 406077 in gnome-python-extras "Should build gda module" [Wishlist,Confirmed] https://launchpad.net/bugs/406077 [08:46] i've done a MIR already for this one - bug 432715 [08:46] Launchpad bug 432715 in libgda4 "[MIR] libgda4" [Wishlist,Fix committed] https://launchpad.net/bugs/432715 [09:15] hi === doko__ is now known as doko === dholbach_ is now known as dholbach [09:42] chrisccoulson: ah ok [09:42] chrisccoulson: I'll let seb128 decide :) [09:42] or somebody else desktop-y [09:42] dholbach, what? [09:42] dholbach, ETOOMUCHTODO [09:42] dholbach - thanks:) [09:42] seb128: python-gda built from gnome-python-extras [09:43] * seb128 fighting over 1000 bug emails from the weekend [09:43] dholbach, I let you decide ;-) [09:43] so glom can be built [09:43] bah [09:43] what is to decide? [09:43] doit? [09:43] yeah, doit probably [09:43] I didn't review too closely [09:46] did anybody see bug #442348 before? [09:46] Launchpad bug 442348 in gdm "GDM doesn't let live user (casper) entering live system" [Undecided,New] https://launchpad.net/bugs/442348 [09:46] dholbach, the changes looked fine to me when I reviewed those some days ago === Tonio__ is now known as Tonio_ [09:49] please change the splash... [09:50] i know you can do MUCH better... [09:50] :) [09:56] darkham: Unless you have an alternative you want us to look at this is the wrong place [09:57] bug-report-via-irc is not supported yet :) [09:59] Amaranth, http://www.youtube.com/user/madsrosendahl#play/uploads/3/XlCVrtgxVcI [10:00] please contact him, they've many good ideas [10:00] darkham: We don't know how long the boot will take so that animation is impossible [10:00] and way too over the top [10:01] the animation can't be in syncro withn boot? [10:02] it can't be clocked? [10:23] Hello, I have a question about texlive [10:23] I see ubuntu did a change to texlive-base [10:23] http://changelogs.ubuntu.com/changelogs/pool/main/t/texlive-base/texlive-base_2007.dfsg.2-4ubuntu1/changelog [10:23] I suspect that Debian recently fixed this bug in texlive-bin instead: http://packages.debian.org/changelogs/pool/main/t/texlive-bin/texlive-bin_2007.dfsg.2-7/changelog [10:25] the bug is something to do with "5-years is too old" [10:26] presumably it's not that important exactly where it's fixed, and we can just resync in lucid? [10:27] ok [10:28] another thing, I filed a bug LP 438031 [10:28] Launchpad bug 438031 in texlive-bin "texlive-bin won't compile against libpoppler 0.12" [Undecided,New] https://launchpad.net/bugs/438031 [10:28] I suspect that luatex too won't compile against libpoppler 0.12, since it has the same problematic code portion that is in texlive-bin [10:29] but I didn't try yet [10:43] Riddell: can you ack/nack bug 442571? [10:43] Launchpad bug 442571 in kipi-plugins "FFe for kipi-plugins 0.7.0" [Wishlist,New] https://launchpad.net/bugs/442571 [10:44] I can, he's got another one coming up too [10:47] Riddell: So how many bug reports did you get about amarok not being installable? :) [10:48] Keybuk, cjwatson: does https://code.edge.launchpad.net/~tormodvolden/casper/inspect-dm/+merge/11423 look sensible to you? I'm afraid I don't know too much about the boot mount process to see possible regressions [10:48] pitti: I was already in the middle of reviewing that [10:48] nice timing, thanks [10:49] Can someone from ubuntu-release look at bug #429445? [10:49] Launchpad bug 429445 in zephyr "[Karmic FFe] Sync zephyr 3.0~rc.2544-1 (main) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/429445 [10:58] siretart: gah, I meant ruby1.9, not 1.9.1... [10:59] could someone give-back ruby1.9 on i386? it failed in an unexpected way, and built fine everywhere else === tkamppeter__ is now known as tkamppeter [11:10] lucas: done [11:10] hi tkamppeter [11:15] Amaranth: havn't noticed any, what's wrong with it? [11:16] Riddell: Last time I checked it depended on amarok-common which doesn't seem to exist [11:18] pitti, hi [11:21] Riddell: weird, must have been an archive issue because the source package says it should exist and it does now without any changes to the package [11:21] But all weekend people were popping into #ubuntu+1 asking about it [11:22] maybe it was in universe [11:22] or maybe it was in the NEW queue [11:22] the latter's actually quite plausible - since it's an architecture: all binary package, non-i386 users would have seen amarok being uninstallable until amarok-common passed through NEW [11:24] cjwatson: That would be exactly it [11:24] nothing new about it, probably amd64 buildds were ahead of i386 === akgraner_ is now known as akgraner === soren_ is now known as soren === highvolt1ge is now known as highvoltage [12:27] tkamppeter: did you recently see bug reports about graphics in PDF files not being printed? I just filed bug 443026, and wondered how widespread the regression is [12:27] Launchpad bug 443026 in cups "Does not print graphics in PDF documents; after pdf2ps it works" [Undecided,New] https://launchpad.net/bugs/443026 === luisbg_afk is now known as luisbg [12:43] pitti, I see this for the first time now. The only way to discover what is happening is to run the filters one by one to find the culprit. PDF handling software seems to be EXTREMELY fragile and doing one change seems to introduce 20 regressions. Here a good regression testing infrastructure needs to be introduced. [12:50] pitti: I can't seem to get the fix for bug 395079 working. I'm trying http://paste.ubuntu.com/286135/, but gnome-mount -o locale=en_GB.UTF-8 still seems to be just ignored; and I don't understand how volume.fstype.alternative gets used, given that in Ubuntu we want ntfs-3g to be the default NTFS implementation (indeed, it clearly is, by virtue of being /sbin/mount.ntfs) [12:50] Launchpad bug 395079 in ntfs-3g "[KDE4][Karmic] Error mounting ntfs volume from dolphin's resources panel" [Undecided,Triaged] https://launchpad.net/bugs/395079 [12:55] hum, bug #442742 [12:55] Launchpad bug 442742 in language-pack-kde-es-base "Update of 20091003 have 15.4% of the translations from 20090926" [Critical,Confirmed] https://launchpad.net/bugs/442742 [12:55] cjwatson - that won't work in karmic, as HAL doesn't do the mounting anymore [12:56] * seb128 kicks launchpad which timeout on bug reassign [12:57] grrr, stupid 3g connection again [13:01] cjwatson - there is also already a fdi file in the source, which i wrote last cycle - debian/25-ntfs-3g-policy.fdi [13:01] it just isn't installed at the moment [13:04] Hi zul. Online? [13:04] ion: yes will get to your debdiff today [13:04] zul: Alright :-) [13:05] chrisccoulson1: whoops, so there is, but it doesn't use the new scheme pitti linked to [13:05] oh, hmm [13:05] I hate fdi files [13:06] chrisccoulson1: so what's ntfs-3g supposed to do now? I'm thoroughly lost ... === porthose|afk is now known as porthose [13:07] cjwatson - me too. i'm not sure what happens in KDE now [13:07] is it still using HAL for mounting? [13:07] search me [13:07] if all ubuntu flavours are using DK-disks, then the fdi file should not be needed [13:08] it's not needed for ubuntu, but i really don't know about kubuntu [13:08] where does dk-disks get its mount option lists? [13:08] i hate fdi files too. HAL mounting is horrendously over-complicated ;) [13:09] cjwatson - the mount options are currently hard-coded in dk-disks [13:09] ok [13:09] Riddell: above conversation - can you shed any light? [13:09] they're not configurable yet, so if the mount options are wrong, you'll need to patch dk-disks [13:11] locale= will deal with 90% of the complaints [13:11] KDE isn't using devicekit [13:11] it's still using HAL [13:11] ah, so it defaintely needs a fdi file then [13:12] s/defaintely/definately [13:13] chrisccoulson1: so if I just revert your ntfs-3g change that stopped installing the fdi file, that should be good enough and won't break gnome? === azeem_ is now known as azeem [13:14] cjwatson - possibly. it shouldn't have any effect in gnome, as we're not using HAL for that anymore [13:17] I think at this point that stands a good chance of making it Somebody Else's Problem which will make me happy. :) [13:18] re [13:19] cjwatson, chrisccoulson1: KDE still uses hal for mounting, through solid [13:21] cjwatson: GNOME does not use hal at all any more, so changing FDI files is fine from Ubuntu's perspective [13:21] chrisccoulson1 already said that, sorry [13:37] james_w: Hi, is there anything I can do to get bug 428254 sponsored? [13:37] Launchpad bug 428254 in powertop "powertop crashed with SIGSEGV in strstr()" [Medium,Confirmed] https://launchpad.net/bugs/428254 [13:38] I'm worried about two seemingly random crashes within the Python interpreter with computer-janitor, bug 420307 and bug 435580 -- any suggestions? ideas? [13:38] Launchpad bug 420307 in computer-janitor "computer-janitor-gtk crashed with SIGSEGV in gdk_window_set_geometry_hints()" [Medium,In progress] https://launchpad.net/bugs/420307 [13:38] Launchpad bug 435580 in computer-janitor "computer-janitor-gtk crashed with SIGSEGV in realloc()" [Medium,New] https://launchpad.net/bugs/435580 [14:10] Keybuk: any ideas on bug #439138 ? [14:10] Launchpad bug 439138 in xorg-server "[karmic] Xorg 100% CPU utilization -- only after first login" [High,Confirmed] https://launchpad.net/bugs/439138 [14:10] Keybuk: strace reports repeated (failing) calls to ioctl(5, TCFLSH, 0x2) = -1 EIO (Input/output error) where fd 5 is /dev/tty7. [14:13] simon-o: I'll get to it [14:13] simon-o: sorry for the delay [14:14] Can anybody from ubuntu-release look at bug #429445? [14:14] Launchpad bug 429445 in zephyr "[Karmic FFe] Sync zephyr 3.0~rc.2544-1 (main) from Debian unstable (main)" [Wishlist,Confirmed] https://launchpad.net/bugs/429445 [14:14] james_w: thanks. I wasn't sure if there was something missing from me. [14:19] ebroder: you're better off with just subscribing u-r to the bug [14:32] pitti, did you see my last message and my comment on bug 443026? [14:32] Launchpad bug 443026 in cups "Does not print graphics in PDF documents; after pdf2ps it works" [Undecided,New] https://launchpad.net/bugs/443026 [14:39] tkamppeter: is there a way to display a cups raster file? [14:39] tkamppeter: would it help if I revert cups/evince/splix/gs to the jaunty version and see in which particular package the regression is? [14:43] seb128: did you see this python2.6 error "error writing to ..." yourself? [14:43] doko_, no [14:51] kirkland, nurmi: ok, I'll file bugs for the issues I found and comment on the already-filed [14:51] ttx: cool [14:51] great [14:53] pitti, CUPS Raster files can be visualized with RasterView: http://www.easysw.com/~mike/rasterview/index.html [14:54] nice, thanks [14:54] tkamppeter: that saves a lot of paper :-) [14:55] I cannot imagine that SpliX is the culprit, if the rasterization by pdftoraster conserves the drawing, SpliX, which converts CUPS Raster to the printer's raster format should not loose the drawing. [14:56] tkamppeter: so it seems it's most likely an evince bug? [14:58] Only filters which can loose the drawing are pdftopdf and pdftoraster. [14:59] dustin,ttx: can we work with revno 911 today (i.e. use a package based on r911)? [14:59] pitti, for me it really looks more like an evince problem, as Ghostscript and my PDF printers have problems wityh the file. [14:59] nurmi: would that one fix the DB borkage issue ? [15:00] dustin,ttx: there was a problem with database startup < 911, introduced by an SSL database connection fix [15:00] ttx: i believe so, yes [15:00] kirkland: could you update your merge to that version, and publish in PPA ? [15:00] nurmi: are you working on this credentials lost fix? [15:01] ttx: i can do that; i'll also try to fix the orig.tar.gz too [15:01] trying to reproduce [15:01] pitti, another problem is that from the PDF of evince the filters are not able to produce a PostScript which PostScript printers understand, see bug 419143. The PostScript files get embedded fonts which the printers cannot parse. [15:01] Launchpad bug 419143 in evince "Printing from evince (and perhaps other GTK apps) to PostScript printers is broken" [Low,New] https://launchpad.net/bugs/419143 [15:01] nurmi: you're trying to reproduce this credentials issue? [15:01] kirkland: yes [15:01] nurmi: cool [15:02] nurmi: i'll cherrypick that one, once you have something for me [15:02] nurmi: ttx: i need to reboot and make a pot of coffee before i tackle this new merge; give me 15 minutes [15:02] sure [15:02] * kirkland reboots [15:02] kirkland: same here (coffe, at least :) === stefanlsd1 is now known as stefanlsd [15:13] nurmi, kirkland: DB borkage on upgrade issue is bug 443125 [15:13] Launchpad bug 443125 in eucalyptus "Upgrade to r908 loses admin credentials" [High,Triaged] https://launchpad.net/bugs/443125 [15:13] nurmi, kirkland: No autoregistration if first startup is at boot-time is bug 443118 [15:13] Launchpad bug 443118 in eucalyptus "No autoregistration on first startup after ISO install" [High,Triaged] https://launchpad.net/bugs/443118 [15:14] https://wiki.ubuntu.com/UbuntuOpenWeek/Prep <-- still looking for speakers for openweek! [15:14] + I added a comment about the 16-minute timeout on bug 439251 [15:14] Launchpad bug 439251 in eucalyptus "Eucalyptus restart is needed after autoregistration of components" [High,Triaged] https://launchpad.net/bugs/439251 [15:15] nurmi: which ones of these should a merge at level 911 fix ? just 439251 ? or also 443125 ? [15:16] ttx: i havn't been able to reproduce 443125 yet, and so i'm not sure what is going wrong [15:17] ttx: however, it should fix 439251 [15:17] anyone: what would be the right way to specify that an upstart script should start when all networking is up and running ? I don't mind if it starts after everything else. === Tonio__ is now known as Tonio_ [15:22] lamont: Was libmaven-clean-plugin-java on your maven bootstrapping list? I have users screaming the world as we know it will end if maven doesn't get updated and we need that one. === dendro-afk is now known as dendrobates [15:28] uery pitti [15:30] ScottK: If you need help on the Maven update, just ping me. I use Maven fairly often. [15:30] hello, against which package do I need to report a bug for packages installed by the minimal installation CD? [15:31] simon-o: The biggest problem is the manual bootstrapping. Because we don't allow binary uploads like Debian does, it's more complex for us. [15:31] /msg dholbach thanks for that s3kr1t information! [15:31] Mostly just waiting for lamont [15:32] nurmi: ttx: okay, i'm redoing the merge now [15:32] nurmi: ttx: let's make sure that we resolve the two conflicts properly [15:32] http://paste.ubuntu.com/286252/ [15:32] http://paste.ubuntu.com/286253/ [15:33] nurmi: ttx: for http://paste.ubuntu.com/286252/, i'm going to take upstream's change [15:33] nurmi: ttx: ie, the second stanza in the conflict [15:33] nurmi: ttx: agreed? [15:33] mdke, hey [15:33] mdke, got a minute to discuss menu items naming? [15:34] kirkland: sounds sane [15:34] kirkland: nod [15:35] nurmi: ttx: and http://paste.ubuntu.com/286253/ ? [15:36] kirkland: in the eucalyptus-cc.in diff, we want the 'sed' line that has 'ThreadsPerChild 1', but the preseed stuff should also remain [15:36] first line of MERGESOURCE and the rest of TREE [15:36] ttx: nod [15:36] ttx: nurmi: ack [15:37] kirkland: also see my remark about the boot-order patch [15:37] nurmi: would upgrading from 854 need to rebuild to wsdl stubs ? [15:37] ttx: nurmi: http://paste.ubuntu.com/286256/ [15:39] hey all, any tips for a gnome-session that does not run metacity, gnome-panel, or fail in any obvious way? I'm wondering if the root cause is perhaps gnome-settings-daemon? [15:39] nurmi: can you take this difference and apply upstream, to remove the conflict for the next time we merge? [15:40] dist-upgraded to karmic at the beta, and then started having problems [15:41] kirkland: i'll give it a try; i want to see what happens when that file doesn't exist (node-preseed.conf) as it will only be there on ubuntu installs [15:42] ** (gnome-settings-daemon:3019): WARNING **: Failed to acquire org.gnome.SettingsDaemon [15:42] ** (gnome-settings-daemon:3019): WARNING **: Could not acquire name [15:42] * rcaskey is a sad boy [15:42] nurmi: okay, one more thing ... [15:43] rcaskey, you have one already running? [15:43] nurmi: could you go through https://bugs.edge.launchpad.net/ubuntu/+source/eucalyptus, and update any bugs that you expect to be solved in this r911 upload [15:44] nurmi: please assign them to me (kirkland) and mark them "in progress" [15:44] seb128, yes, hrmm, wasn't before the reboot though [15:44] gnome-session is still failing silently though [15:44] kirkland: will do, looking now [15:45] Keybuk: about upstart not reloading /etc/init/*.conf on "restart" (see https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/430883/comments/1): was that behavior changed by the 0.6.3-6 update ? [15:45] Launchpad bug 430883 in upstart "Can start, but not restart, a stopped Upstart job" [Low,Fix released] [15:45] kirkland: if the bug is already 'in progress' and assigned to you, leave it be? [15:46] nurmi: yes; i'm doing this same thing in parallel :-) [15:46] nurmi: i want double coverage on it, though; as I'm documenting this in the changelog [15:47] seb128, killing that old copy and rerunning gnome-session produces the same results [15:47] (polkit-gnome-authentication-agent-1:3236): GLib-GObject-WARNING **: cannot register existing type `_PolkitError' [15:51] or more to the point, if you killal gnome-settings-daemon and then rerun gnome-session you get (gnome-settings-daemon:3449): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed [15:52] kirkland: done [15:53] nurmi: rock, thanks [15:58] pitti: actually, you're right -- setuid() when called by root is the same as setresuid(). [15:59] nurmi: glad i asked; you had a few on your list that i didn't [15:59] pitti: I'm used to using it on set-uid-root processes, though, where that does not happen. (i.e. real-root setuid==setresuid, effective-root setuid!=setresuid) [16:02] nurmi: https://bugs.edge.launchpad.net/ubuntu/+source/eucalyptus/+bug/439251 [16:02] Launchpad bug 439251 in eucalyptus "Eucalyptus restart is needed after autoregistration of components" [High,In progress] [16:02] nurmi: any idea what revno fixes that? [16:03] if I'm correct, it should be 892 [16:03] kirkland: i believe this was a symptom of the database race problem [16:03] nurmi: cool [16:04] kirkland: i have a proposed general solution to the preseed stanza: http://paste.ubuntu.com/286274/ [16:04] kirkland: line 128 [16:05] if both the module (mod_alias) and the node-preseed.conf files are there, it will add the configuration to httpd-cc.conf [16:05] otherwise, it will skip [16:05] nurmi: is this already in this merge? [16:06] nurmi: or are you asking me to add that? [16:06] kirkland: no, i'm proposing it for the next, unless you would like me to commit this now as 912 [16:07] nurmi: if you think it's safe, please go ahead and commit [16:07] nurmi: i'll just merge again [16:09] kirkland: it is safe (tested) from a eucalyptus init script perspective, but I don't know how to test the upstart conversion part [16:09] kirkland: at some point, the init scripts must be getting converted to upstart scripts [16:11] kirkland: if you think that adding this bit to the regular init script will translate properly to /etc/init/eucalyptus-cc.conf, then I'll commit [16:12] nurmi: let me think on it for a minute [16:14] ScottK: yeah - I wanted to get the chroots happy and such - then I had intarwebs issues at home this weekend [16:14] so... today/tomorrow for maven - the downside is that it means a crowbar for i386 buildds (--> manual) [16:15] ScottK: otoh, i386 is down to almost nothing left in the queue [16:15] lamont: I understand. I'd appreciate that one. I could always boot strap it locally and do an upload with an embedded binary if you'd prefer? [16:18] nurmi: ttx: okay, testing the local build now [16:18] nurmi: let's hold off on merging that hypotetical r912 at the moment [16:19] ttx: oh, wait, there was a patch you told me that I should drop... [16:19] ttx: where was that? email or in irc? [16:19] kirkland: email, ubuntu-devel. boot-order.patch [16:20] applies to old init scripts, iirc [16:23] nurmi: also, do we need to refresh the WSDL stubs? [16:25] kirkland: nope [16:25] nurmi: good [16:25] mdz: around? [16:25] mdz: i have a couple of questions about your eucalyptus upstart scripts [16:25] kirkland, yes, but on the phone. what can I do for you? [16:25] ScottK: the trick is to add the debs to the chroot, manual everything, build them, upload them, and then put the old chroot back and auto-mode [16:25] kirkland, I can answer asynchronously [16:26] mdz: that's fine [16:26] lamont: Yes. Please. It's more than just the one package (I assume you know this already) [16:26] mdz: 1) eucalyptus-cc.upstart:exec apache2 -f /var/run/eucalyptus/httpd-cc.conf -D FOREGROUND [16:27] mdz: nurmi asked why this was being started in the foreground... is this linked to some upstart logging, or some such? [16:27] yeah - I have a lizt [16:27] list, too [16:28] mdz: i' [16:28] kirkland, that's by design [16:28] kirkland, upstart expects that [16:28] mdz: okay, thanks; glad I checked; i was inclined to drop that [16:28] it lets it monitor and stop the process [16:28] mdz: unless you use 'expect fork' or 'expect daemon' or whatever ... [16:29] you don't *have* to run upstart jobs in the foreground [16:29] mdz: 2) ttx and I were wondering if perhaps we're attempting registration too soon, before the tcp/ip stack is functional; wondering if we should push registration back (or have it depend) on networking [16:32] mdz: i was curious if you had an opinion or input on (2) [16:32] mdz: i don't have proposed changes on that one yet [16:35] kirkland: if you are in them, fix /etc/init/eucalyptus-walrus-registration.conf so that it uses ipaddrs.conf as the others do [16:36] kirkland: can't seem to reproduce first-boot registration failure on subsequent boots, maybe some DHCP racing is involved here [16:38] ttx: okay, can do [16:38] kirkland: do you have an ETA for a r911-based package in PPA ? [16:38] ttx: i'm building locally right now, should be done with that in 2-3 minutes [16:38] I'd like to test if the admin credentials lost on upgrade issue is still there [16:38] ttx: i'm pushing to ppa as soon as my local build succeeds [16:38] kirkland: ok [16:39] ttx: though i'll add your ipaddrs.conf change too, if you like [16:39] ttx: that shouldn't affect the build [16:39] kirkland: that's cosmetic, the key to upgrading to r911 is that db upgrade issue [16:40] kirkland: so the sooner I get a r911 package, the better ;) [16:40] ttx: local build done; attending your ipaddrs.conf [16:41] * ttx reinstalls from CD to get prstine state and one more autoregistration test [16:41] slangasek: on upgrades, libpam0g prompts for service restarts. should we adjust that (as done for libc, I think) to not prompt in update-manager ? [16:42] +1 for that [16:42] sorry folks, i'm have a issue with the last 2 livecd, alpha 6 and beta don't boot, make the boot process but freeze... can't even enter to a console (ctrl+f1) [16:43] i try noapic, nolapic, ....etc and nothing [16:44] kirkland, I committed a patch to get rid of the error redirect for euca_conf during registration. that should make it obvious if that's what's happening [16:44] ohh, i forget, in virtualBox work without problems. [16:44] kirkland, my gut reaction is, let's test that hypothesis before we try to fix it that way [16:44] slangasek: oh, nevermind, I'm reading 278117 now; it seems like this is only done if the services are non-default [16:45] mdz: ack, that sounds perfectly reasonable [16:45] ttx: uploaded to my ppa [16:45]