[00:03]  * TheMuso shkes his head. So many people filing bugs for audio using lucid's proposed 2.6.32-25 kernel series...
[00:53] <SpamapS> is now a bad time to try and upload packages to a ppa for maverick? I am getting a lot of uninstallable stuff.. like.. cdbs and debhelper..
[00:53] <SpamapS>  cdbs : Depends: intltool but it is not going to be installed
[01:00] <slangasek> SpamapS: dunno what that is, things shouldn't be problematic right now given that we just froze.  How long ago was it, can you reproduce the problem in a chroot?
[01:00] <slangasek> (I can't, but the chroot I'm in isn't a clean one)
[01:11] <SpamapS> slangasek: I'm unable to update my chroot or pbuilder right now because my internet connection shuts itself down if I sustain downloads. :(
[01:12] <SpamapS> technician coming tomorrow with a new DSL modem.. hopefully that fixes it. :-P
[01:12] <SpamapS> slangasek: http://launchpadlibrarian.net/55773453/buildlog_ubuntu-maverick-i386.cassandra_0.6.5ubuntu1_FAILEDTOBUILD.txt.gz
[01:12] <SpamapS> there's the full failed to build log
[01:25] <avi_> Hello, can anyone tell me if there are any signals for appindicator menus themselves? Like not clicking a menu item, but clicking the indicator icon itself.
[02:06] <persia> Could someone give-back all the failed langpacks on i386 ( http://qa.ubuntuwire.com/ftbfs/ ) gettext seems less broken now.
[02:20] <TheMuso> persia: no shell access to mass give back here, but there are not a lot, so I'll give them back manually.
[02:23] <persia> TheMuso, Thanks.  I was hoping a buildd-admin was around, but I figured it would be unpleasant for the Europeans to wake up to those.
[02:24] <TheMuso> np should all be back in the queue now.
[02:25] <persia> Also, thanks for cleaning up after me on 623242: I missed the flat volumes somewhere.
[02:31] <TheMuso> np
[03:17] <slangasek> SpamapS: ^^ would have been the gettext issue mentioned above, believed solved now
[05:39] <UbuXubu> i would like to report a bug
[05:40] <UbuXubu> i did a clean install of 10.04 a couple weeks ago and i let ubuntu have the entire disk.
[05:40] <UbuXubu> it updated and has since updated a few more times.
[05:41] <UbuXubu> the last update was small..maybe 5-6 of them. i tokk them all as i always do. after that when i log in the panels are gone?
[05:41] <UbuXubu> so i did the terminal command and it worked but the panels are gone again everytime i log in.
[05:42] <UbuXubu> soi was given a second termal command and it worked too, but anytime i log in the problem comes back.
[05:43] <UbuXubu> so just for the heck of it i right clicked on the panels and installed one of those little widgets offered by ubuntu. i put a couple of them on each panel AND THAT FIXED IT?
[05:44] <UbuXubu>  i am hoping this fix somehow leads the coders to what is wrong because i really dont want the widgets, but for now it keeps my panels frpm disappearing.
[05:45] <UbuXubu> from*
[05:46] <TheMuso> UbuXubu: If you wish to file a bug, you should file the bug on launchpad.net. I suggest you go to https://launchpad.net/ubuntu/+source/gnome-panel/+filebug to file your bug. The link I gave you is to file a bug against the component, otherwise known as package, that you are having problems with.
[05:46] <TheMuso> !filebug | UbuXubu
[07:27] <MIH1406> Is there a program that check my sources against standards coding?
[07:55] <pitti> Good morning
[07:56] <poolie> pitti: hi, i wanted to see about getting bzr in to https://wiki.ubuntu.com/StableReleaseUpdates/MicroReleaseExceptions
[07:56] <poolie> this is basically the process we follow now, but i think it would be worth fixing any accidental divergences
[07:56] <poolie> and making it formally recognized
[07:56] <pitti> poolie: this sounds reasonable to me, given its history, test coverage, and regression potential
[07:57] <pitti> poolie: can you please mail technical-board@ about this, so that we can discuss it officially?
[07:57] <poolie> we have put some SRUs up but in the last one someone (quite reasonably) questioned if it was ok to upload a whole minor release rather than individual patches
[07:57] <pitti> then we can discuss it next Tuesday, or on the ML
[07:57] <poolie> sure, just though i'd sound you out first
[07:57] <poolie> np
[07:57] <pitti> poolie: minor releases have always been okay as long as the individual changes all match SRU criteria
[08:01] <pitti> hm, did someone already giveback all langpacks after the libunicode bug?
[08:02] <pitti> libunistring0, actually
[08:02] <pitti> ah, seems so; thanks
[08:06] <ari-tczew> pitti: thanks for reject wrong file :)
[08:09] <poolie> thanks
[08:24] <poolie> pitti, sent, it's currently in the moderation queue
[08:34] <pitti> poolie: I discarded the first post and accepted the second
[08:34] <poolie> was it sent twice?
[08:35] <poolie> how strange
[08:36] <pitti> apparently so
[09:02] <Foobar_> Hi, I want to start programming an open source project for learning purpose and for helping the open source community
[09:03] <Foobar_> But I am confused in what type of project?
[09:04] <Foobar_> Who can suggest a project for me that is not a duplicate of any other project?
[09:05] <primes2h> pitti: I can't build apport. pastebin.ubuntu.com/494806 svg icon error.
[09:05] <primes2h> ?
[09:06] <primes2h> pitti: first of all, hello! :-)
[09:08] <primes2h> pitti: it seems that symlink point at the wrong dir.
[09:10] <cjwatson> TheMuso: shell access for give-backs> you don't need it - there's an ubuntu-build program in ubuntu-dev-tools, with a --batch option
[09:40] <pitti> hello primes2h
[09:41] <pitti> primes2h: no, the dir is correct; it rather looks like debian/tmp/usr/share/icons/hicolor/scalable/mimetypes
[09:41] <pitti> doesn't exist?
[09:43] <primes2h> pitti: text-x-apport.svg points to ../apps/apport.svg but apps dir doesn't exist.
[09:43] <primes2h> text-x-apport.svg is into mimetypes
[09:43] <primes2h> dir
[09:44] <pitti> lrwxrwxrwx 1 martin martin 18 2010-08-25 12:51 ./data/icons/scalable/mimetypes/text-x-apport.svg -> ../apps/apport.svg
[09:44] <pitti> -rw-r--r-- 1 martin martin 16628 2010-08-05 17:55 ./data/icons/scalable/apps/apport.svg
[09:44] <pitti> seems fine here
[09:46] <primes2h> pitti: it's present in ./data/icons/scalable/etc... but not in debian/tmp/usr/share/icons/hicolor/scalable/
[09:47] <primes2h> pitti: just mimetype dir is present  in debian/tmp/usr/share/icons/hicolor/scalable/
[09:52] <pitti> hm, they build just fine here
[09:53] <primes2h> pitti: tried with pbuilder and debuild. Am I doing something wrong maybe?
[09:54] <pitti> primes2h: not sure; perhaps put the complete build log somehwere?
[09:54] <pitti> I did apt-get source apport; cd apport-*; debuild -b
[09:59] <primes2h> pitti: same here, I've just modified source code a bit (two line added). dch etc..
[10:00] <primes2h> pitti: I'll try to have a deeper look into
[10:00] <pitti> primes2h: http://people.canonical.com/~pitti/tmp/apport_1.14.1-0ubuntu7_amd64.build if you want to diff the build dlog
[10:01] <primes2h> pitti: Thanks!
[10:38] <didrocks> cjwatson: do you know if there will be a script to close all ubuntu bugs from the last week upload or should we care about that ourself?
[10:39] <cjwatson> didrocks: there is such a script which I am currently running ;-)
[10:39] <didrocks> cjwatson: great! thanks :-) is everything working again now?
[10:39] <cjwatson> didrocks: (you may have noticed me closing a number of your quickly bugs ...)
[10:39] <cjwatson> I believe so
[10:39] <cjwatson> if it isn't, let me know and I'll escalate
[10:39] <didrocks> cjwatson: right, but as there were some minutes between emails, I was hoping you wasn't going crazy closing them manually :)
[10:39] <cjwatson> well, not manually via a browser
[10:40] <cjwatson> I have to pay some attention to it, as there are some cases where the bug would have been closed via the upload, but then somebody noticed it wasn't fixed after all
[10:41] <cjwatson> so I have a script that looks for all the bugs that should be closed, then presents me all the bug comments and the proposed close message and lets me say close, skip, or open in browser
[10:41] <cjwatson> and I apply judgement from there
[10:41] <cjwatson> still much faster than having to open them all up in firefox or whatever
[10:41] <didrocks> cjwatson: oh ok, some pretty touchy case in it and so, long task. good luck with that! Hope it won't take too long for you :-) thanks!
[10:41] <didrocks> right
[10:41] <cjwatson> it's not too bad, I'm up to r
[10:42] <cjwatson> been poking it on and off since yesterday afternoon
[10:42] <didrocks> (sweet, soon unity them ;))
[10:42] <didrocks> then*
[10:42] <cjwatson> my thought exactly.  I assume there are several bugs on the RC list that this will catch
[10:43] <didrocks> right, there are also some others with yesterday's evening release. I'm still catching up on emails on beeing away from the last 2 days and will package the new stack then
[10:43] <didrocks> it fixes also a bunch of bugs
[10:43] <didrocks> the RC list is getting smaller for it, that's a good news :)
[10:45] <zyga> what is bug watch updater (on launchpad)
[10:45] <zyga> it has just changed the importance of a bug to critical
[10:46] <cjwatson> zyga: see ubuntu-devel
[10:46] <cjwatson> zyga: and note that it's changing the importance of the upstream task
[10:46]  * zyga chcecks
[10:47] <zyga> cjwatson, thanks, that explains it
[10:54] <ttx> pitti: what's the status of bug 494141, now that we reverted the cups/samba stuff ?
[10:54] <pitti> ttx: ah, that should be fixed now, let me close the bug
[10:54] <ttx> pitti: good, thanks :)
[10:55] <pitti> ttx: bug updated
[10:55] <soren> The conclusion was that Samba would reload the printers on SIGHUP, right?
[10:56] <pitti> right
[10:57] <soren> Cool beans.
[10:58] <cjwatson> didrocks: ubiquity 2.3.19 had its bugs auto-closed, at least
[11:01] <didrocks> cjwatson: ok, thanks for the notice :)
[11:30] <cjwatson> didrocks: should I close bug 631446 with the unity 0.2.38-0ubuntu1 upload?  there are several comments from after the upload, so I'm not clear on this
[11:31] <didrocks> cjwatson: yes please. It will be tracked in a separate bug report.
[11:31] <cjwatson> ok, done
[11:33] <cjwatson> didrocks: same question for bug 632991
[11:35] <didrocks> cjwatson: I'm asking to gord about that one, will get back to you then
[11:35] <didrocks> cjwatson: don't close if for now. It will be closed by the upload I'll do in a couple of hours
[11:36] <cjwatson> ok, skippin
[11:36] <cjwatson> g
[11:36] <didrocks> thanks :)
[11:38] <cjwatson> didrocks: bug 622146, you said "let the unity opened" - does that mean that you want the unity (Ubuntu) task to stay open, despite closing it in the 0.2.36-0ubuntu1 upload?
[11:38] <ari-tczew> pitti: could you take a look on bug 557290 ? I think that this patch is useless.
[11:39] <didrocks> cjwatson: no, both should be stayed open, there is finally the need of a last minute patch by Jason which should be there in today's upload as well.
[11:39] <cjwatson> didrocks: ok, skipping that too
[11:40] <cjwatson> gah, API timeouts
[11:44] <pitti> ari-tczew: I didn't look at maverick build logs at all -- they could have failed for completely different reasons
[11:45] <ari-tczew> pitti: so, I'm not convinced to patching lucid, if this patch couldn't fix FTBFS for maverick. I'll unsubscribe ubuntu-sponsors, do you agree with it?
[11:47] <pitti> ari-tczew: what's wrong with the lucid one?
[11:48] <ari-tczew> pitti: we are not sure whether it patch will be buildable for amd64, powerpc and sparc
[11:48] <ari-tczew> and current source package built fine on all archs
[11:48] <pitti> but that's easy enough to find out?
[11:48] <pitti> https://edge.launchpad.net/ubuntu/+source/gnu-smalltalk/3.0.3-2
[11:48] <pitti> it built fine in jaunty, but lucid has all binaries removed
[11:49] <pitti> ari-tczew: and maverick amd64 indeed fails in a test case, but other arches succeeded
[11:49] <Wubbbi> Hey :) I wanna change the source-Code of Firefox. I want to change the default settings in about:config of firefox. Where can I find it?
[11:52] <ari-tczew> pitti: launchpad is not clear for me. 2009-10-29 18:02:30 CET  	Published  	 Lucid   	release  	universe
[11:52] <ari-tczew> pitti: no words saying that it was deleted from lucid
[11:52] <pitti> ari-tczew: the source isn't
[11:52] <pitti> ari-tczew: but the binaries were removed obviously
[11:52] <ari-tczew> but packages.ubuntu.com doesn't show lucid packages
[11:52] <pitti> gnu-smalltalk |    3.0.3-2 | karmic/universe | source, amd64, i386
[11:52] <pitti> gnu-smalltalk |    3.0.3-2 | lucid/universe | source
[11:54] <ari-tczew> pitti: well, if my pbuilder will build package for lucid in i386, do  I should upload SRU for lucid-proposed?
[11:54] <pitti> ari-tczew: sounds good
[11:55] <ari-tczew> pitti: I'm not convinced, but I'll do it to your command.
[11:56] <ari-tczew> pitti: and what about maverick status?
[11:56] <diwic> Wubbbi, apt-src install firefox ?
[11:57] <pitti> ari-tczew: I don't know; it looks like a separate problem, or the patch is already applied
[11:57] <Wubbbi> diwic: I know how to get the SC. I just want to know in which file ( in the source-Code ) I can change the default about:config. So when I build and install it, the default config is the way I did it in the c
[11:57] <ari-tczew> pitti: patch is already applied, right,
[11:57] <Wubbbi> source-Code
[11:58] <pitti> ari-tczew: so I guess the maverick task can be closed then
[11:58] <ari-tczew> pitti: as fixed or invalid?
[11:58] <pitti> fixed, if the patch is applied
[11:58] <diwic> Wubbbi, okay. Sorry, I don't know that. You could try searching for a string you know exist in that file
[11:59] <Wubbbi> diwic: ok thx
[12:19] <ogra> GRRR samba
[12:19]  * ogra starts getting annoyed by it 
[12:20] <ogra> can a buildd admin bump https://edge.launchpad.net/ubuntu/+source/samba/2:3.5.4~dfsg-1ubuntu7/+build/1963227 ?
[12:20] <ogra> its sitting there since two days now blocking all image builds
[12:33] <zyga> has anoyne been getting lots of rendering artefacts on lucid with compiz and GMA945 recently?
[12:33] <zyga> I managed to get a screenshot of one just now
[12:39] <pitti> ogra: kicked
[12:40] <ogra_cmpc> pitti, TA
[12:55] <elmo> is there any official ubuntu stance/policy on folks following up to bugs and blindly saying "is this still a problem in $latest_version?" and setting the bug to incomplete?
[13:00] <cjwatson> elmo: I wish there were.  I complain when I see people doing it
[13:52] <ari-tczew> pitti: could I upload a SRU which fix 2 different bugs?
[13:53] <pitti> ari-tczew: sure, the policy has a section about that
[13:53] <ari-tczew> policy, policy, policy...
[13:59] <hallyn> GAH  after about an hour of reading email etc, i just realized my mouse isn't working this morning.
[14:07] <apw> elmo, JFo spamming you  ?
[14:08] <JFo> ?
[14:08] <JFo> apw, I shouldn't be... or should I?
[14:09] <elmo> haha
[14:09] <elmo> apw: nah, it wasn't JFo
[14:09] <JFo> oh heh, I see what you asked about now elmo :-)
[14:37] <pitti> hallyn: what does "isn't working" mean?
[14:38] <hallyn> pitti: can't move the mouse with my touchpad
[14:38] <hallyn> just got done with a few reboots to older kernels to compare
[14:40]  * hallyn goes to file a bug
[14:40] <pitti> hallyn: I think this could really do with a proper ubuntu-bug filed bug report, to get all the details
[14:40] <pitti> we had a couple of -evdev and touchpad updates recently for multitouch
[14:40] <hallyn> pitti: hm, ok, i'll reboot back into that kernel and try.  hopefully it doesn't present me a too crappy of a gui that i can't navigate with kbd
[14:40] <hallyn> (it makes sense, though, i agree)
[14:41] <hallyn> biab
[14:44] <pitti> hallyn: you don't have an USB mouse?
[14:49] <hallyn> pitti: not handy, no
[14:49] <hallyn> it's ok, it did fine.  apart from trying to pull up chrome to do the filing (for which i don't have the equivalent of vimperator plugin installed)
[14:53] <didrocks> hallyn: can you check on #ubuntu-touch? those are the people who made the changes
[14:55] <hallyn> didrocks: didn't know about that one, thanks
[14:55] <didrocks> hallyn: yw :)
[15:27] <Sarvatt> hallyn: do you have an alps touchpad?
[15:27] <Sarvatt> there's discussion about it right now in #ubuntu-kernel if so
[15:31] <hallyn> Sarvatt: yes, i do, i'll head there, thanks.  (the ppl in #ubuntu-touchpad have been very helpful, they probably brought it up there)
[15:36] <cjwatson> dpm: is the language-selector part of bug 630924 still release-critical?
[15:37] <cjwatson> dpm: (it's been fixed in ubiquity)
[15:42] <dpm> cjwatson, otp, will answer in a bit
[15:42] <cjwatson> ok
[16:25] <doko> slangasek: you touched libxml-twig-perl last. b-d on libunicode-map8-perl (universe), but I can't see when this was introduced
[16:34] <cnd> how do I handle a bug where all that's needed is a package recompile due to a change in /usr/include/linux/input.h?
[16:35] <cnd> the evdev protocol version bumped, which broke lsinput, but it should be backwards compatible as far as lsinput is concerned
[16:35] <cnd> so a rebuild of input-utils would fix it
[16:35] <cjwatson> bump the version by either one ubuntu revision (if it's already ubuntuN or buildN) or by appending build1 (if it isn't) and reupload
[16:35] <cjwatson> so version 0.0.20051128-4ubuntu2
[16:35] <cjwatson> er, input-utils has its own copy of linux-input.h though
[16:35] <cjwatson> you'd need to refresh it
[16:36] <cjwatson> this is often the way for userspace software that uses kernel headers
[16:36] <cnd> cjwatson, there's a patch that already replaces linux-input.h with linux/input.h
[16:36] <cnd> so it should just need a rebuild
[16:36] <cjwatson> so there is.  what I said above, then
[16:36] <cnd> I'll test in one minute
[16:37] <cnd> cjwatson, is this something I should pursue for the RC freeze?
[16:37] <cjwatson> sounds reasonable to me
[16:37] <cnd> ok
[16:37] <cjwatson> I mean it's completely broken right now
[16:37] <cjwatson> you aren't likely to make it any worse
[16:38] <cnd> yeah
[16:39] <cnd> but there's more than one util that makes up input-utils :)
[16:39] <cnd> there's three!
[16:39] <cnd> but yeah, a rebuild isn't even like adding a patch one would think
[16:40] <cjwatson> cnd: yes, and all three are broken
[16:40] <cjwatson> I checked before I said anything :-)
[16:40] <cnd> ahh, I didn't :)
[16:42] <pitti> james_w: still online? I'm afraid I need some more help with bzr dh-make
[16:42] <james_w> hi
[16:43] <pitti> james_w: so, I just created https://edge.launchpad.net/udisks-automounter and want to package it the new way
[16:43] <pitti> james_w: so I go into my ubuntu dir, and do
[16:43] <pitti> bzr branch lp:udisks-automounter
[16:43] <pitti> cd udisks-automounter
[16:44] <pitti> bzr dh-make -v udisks-automounter 0.0.1 ~/upstream/udisks-automounter-0.0.1.tar.gz
[16:44] <pitti> but this wreaks havoc
[16:45] <pitti> james_w: this gives me http://paste.ubuntu.com/495363/
[16:45] <pitti> james_w: i. e. it removes most of the files in my upstream tarball
[16:46] <pitti> oh, perhaps it's stumbling over the fact that my upstream tarball doesn't have a single toplevel dir
[16:47] <dpm> cjwatson, I think the language-selector task could be marked as fixed, as it happened to be another similar issue, but fixed in bug 612825
[16:47] <dpm>  (regarding your question on bug 630924 earlier on)
[16:47] <cjwatson> dpm: ok, please go ahead and do that if you think it's appropriate then
[16:48] <primes2h> pitti: hello, no luck. Tried to download source and build it only but same error http://pastebin.ubuntu.com/495312
[16:48] <dpm> cjwatson, ok, done that then, marked as Fix Released
[16:49] <primes2h> pitti: I'm trying to fix the link advise issue, but I can't test it if I can't build it ;-)
[16:50] <james_w> pitti: wow
[16:50] <cnd> is the process for RC freeze exception the same as FFE?
[16:50] <james_w> pitti: that's a good one. It might be the toplevel dir thing, but I thought it should cope with that. Are you willing to try it with a toplevel dir?
[16:50] <cnd> looks like there's a bullet point about milestone freeze exceptions on the wiki page
[16:50] <cjwatson> cnd: upload, we'll review.  if it's non-trivial, make sure there's a bug filed and that ubuntu-release is subscribed
[16:50] <pitti> james_w: yes, I just fixed the upstream Makefile to produce a proper tarball
[16:51] <cjwatson> cnd: not really much point doing the bug dance for a rebuild-only upload though
[16:51] <cnd> cjwatson, except that I need a bug dance to upload (ubuntu-sponsors subscription)
[16:51] <pitti> james_w: looks much better already: http://paste.ubuntu.com/495367/
[16:51] <cnd> unless you want to grant me core-dev status right now :)
[16:51] <pitti> james_w: i. e. it didn't remove files, it just still complains about a "No such file or directory"
[16:52] <pitti> james_w: the two additions are fine; make dist ships the pregenerated .c files from vala, but they aren't in upstream bzr
[16:53] <SpamapS> slangasek: thanks for letting me know about the gettext issue, I'll retry the build
[16:53] <pitti> james_w: oh - that might just be because I don't actually have dh-make installed?
[16:54] <pitti> james_w: anyway, it's looking fine now, so sorry for the noise
[16:54] <james_w> pitti: yeah, that will be it. Could you please file two bugs?
[16:54] <pitti> can do
[16:54] <james_w> thanks
[16:56] <cjwatson> cnd: not so sure I'm allowed to ;-)
[16:56] <cnd> darn :)
[16:59] <apw> davidm, i hear you have a busted touchpad since you upgraded ...
[17:03] <cnd> cjwatson, thanks for the help
[17:03] <cnd> got everything filed as bug 628392
[17:03] <smb> pitti, Would you be able to let the latest lucid kernel upload in. (its just a re-upload with security of the last proposed)
[17:07] <pitti> smb: done
[17:07] <smb> pitti, thanks
[17:12] <apw> davidm, if so could you test the kernels here and report back, pretty urgent we get confirmation: https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/641320
[17:22] <slangasek> doko: libunicode-map8-perl> but apparently that was in main before?
[17:22] <slangasek> doko: because the package built
[17:23] <slangasek> doko: if I didn't document it in the changelog, it's not an intentional delta from Debian on my part :)
[17:23] <doko> ok
[17:35] <davidm> apw, pulling down now, do I need the headers too?
[17:35] <apw> only if you have like broadcom crap
[17:35] <davidm> Nope no broadcom nvidia gpu however
[17:38] <davidm> apw back in a bit, rebooting now
[17:40] <smoser> i don't really understand what 'FinalFreeze' means, and how hard exceptions are.
[17:40] <smoser> i have a complete understanding of bug 509562 and have a patch for it.  If i upload, will it get accepted ?
[17:41] <cjwatson> if you think it should be in the release and is worth the risk, upload it and we will review it
[17:41] <cjwatson> as usual, exceptions get harder the closer to release we are
[17:42] <smoser> thanks.
[17:52] <apw> davidm, ... did you make it back
[17:53] <sistpoty|work> didrocks: are you up to handle FFe delegations for netbook again for this cycle?
[17:53] <davidm> apw, that was "fun" I found out the hard way, first you remove the nvidia driver then you install kernel reboot
[17:53] <davidm> And no that did not work for me, the pad was still dead
[17:53] <sistpoty|work> superm1: are you up to handle FFe delegations for mythbuntu again for this cycle?
[17:54] <superm1> yup
[17:54] <sistpoty|work> great, thanks! superm1
[17:55] <apw> davidm, can i get the dmesg off that machine with test kernel no pls.  pastebin perhaps
[17:55] <ScottL> sistpoty|work,  ping
[17:55] <sistpoty|work> cody-somerville, mr_pouit: are you up to handle delegations for xubuntu again for this cycle?
[17:55] <sistpoty|work> ScottL: pong
[17:55] <didrocks> sistpoty|work: some argued that delegations email wasn't sent, so, are not active. Consequently, I think that's a "no" :)
[17:55] <ScottL> sistpoty|work, persia tells me that you might be looking for me for ubuntu studio release
[17:56] <cjwatson> didrocks: he means, were the delegation mail to be sent properly, would you be willing to accept the delegation
[17:56] <sistpoty|work> didrocks: I'm about to arrange delegations ;)
[17:56] <davidm> apw, bug amended with my comments
[17:56] <didrocks> sistpoty|work: cjwatson: oh ok. Of course, I'll be pleased to help there :)
[17:56] <davidm> apw I'll grab you a dmesg
[17:56] <sistpoty|work> ScottL: does that mean you'll volunteer? that would be excellent!
[17:56] <sistpoty|work> didrocks: cool, great!
[17:56] <didrocks> thanks sistpoty|work :)
[17:57] <apw> davidm, have you tried the -19 kernel to confirm it works there ?
[17:57] <ScottL> sistpoty|work, i'm not sure what i'm volunteering for actually :P
[17:57] <ScottL> sistpoty|work, but as ubuntu studio lead i'll do what is necessary
[17:58] <apw> davidm, for -19 read 'the previous one'
[17:58] <davidm> apw dmesg dump attached to bug now
[17:58] <sistpoty|work> ScottL: if anyone wants to upload a package that falls under the ubuntu studio set, and contains features instead of just bug fixes, you'll be in charge to decide if it should be accepted or not
[17:58] <davidm> apw nope, I just got this machine yesterday first install had .21
[17:59] <apw> davidm, could you try and older kerenel please, hallyn is reporting the revert working for him
[17:59] <ScottL> sistpoty|work, i accept that charge
[17:59] <davidm> apw if you can aim me at a URL where that kernel is at I'm happy to try
[17:59] <sistpoty|work> ScottL: excellent, thanks!
[18:00] <apw> hallyn, was it -22 which broke you?  ie -21 was a good kernel ?
[18:00] <hallyn> apw: hm, i don't see a -21 in /boot
[18:00] <hallyn> so 20 was good, and i'm not sure about -21
[18:00] <sistpoty|work> highvoltage, stgraber: are you up to handle edubuntu FFe delegations for this cycle again?
[18:00] <apw> hallyn, thats fine, any known good one is fine
[18:01] <apw> davidm, https://edge.launchpad.net/ubuntu/maverick/+source/linux/2.6.35-20.29 ... click the appropriate arch on the right ....
[18:02] <sistpoty|work> chrisccoulson: would you be willing to accept the FFe delegation for mozilla releated packages?
[18:02] <chrisccoulson> sistpoty|work, sure, that's ok
[18:02] <sistpoty|work> chrisccoulson: cool, thanks!
[18:03] <sistpoty|work> micahg: would you like to act as additional delegate for mozilla related packages for this cycle again?
[18:04] <micahg> sistpoty|work: sure :)
[18:04] <sistpoty|work> micahg: excellent, thanks!
[18:04] <Sarvatt> davidm: you have a vaio F? I just came across another bug where that machine is requiring i8042.nopnp for the touchpad to work
[18:04] <apw> davidm, that dmesg is not from the test kerenl, its from an official -21 kerenl
[18:05] <micahg> sistpoty|work: regarding my own requests, if need be, should I get chrisccoulson to ACK?
[18:05] <cjwatson> the delegates are additions - the release team can always ack
[18:05] <sistpoty|work> micahg: yes, please
[18:05] <micahg> sistpoty|work: k
[18:13] <highvoltage> sistpoty|work: yes, at this stage things are looking good so I'm happy to report that we probably won't need Edubuntu FFe's this release :)
[18:13] <davidm> apw whoops, I had already rebooted, I'm about to test 20 will capture dmesg from that when I do
[18:13] <sistpoty|work> highvoltage: even better, thanks!
[18:13] <Darxus> Are there any stats on the number of cpu cores in machines running Ubuntu?
[18:13] <persia> Darxus, No reliable ones: we can't even count installs, let alone be sure of the HW they run on.
[18:14] <Darxus> Thanks.
[18:14] <apw> davidm, and if -20 doesn't work either, then does booting the debug kernel with  i8042.nopnp on the kernel command line work
[18:15] <Darxus> Do you think it would be reasonable to expect people with more than 16 cpu cores to compile their own kernel for maximum efficiency?
[18:15] <sistpoty|work> oh, gotta run, bbl
[18:15] <apw> Darxus, i would think its reasonable to expect they might be doing their own tweaking if they are needing a machine that size
[18:16] <Darxus> apw: Yeah that's what I was thinking, thanks.
[18:16] <davidm> apw, no love, with 2.6.35-20-generic my mousepad still dead
[18:17] <davidm> apw so my system apparently is not regressed it's just borked.... :-(
[18:17] <apw> davidm, and if -20 doesn't work either, then does booting the debug kernel with  i8042.nopnp on the kernel command line work
[18:17] <apw> could you try either kernel actually with that
[18:18] <Darxus> My questions were loaded.  The BFS CPU scheduler is faster than the mainline kernel's CFS scheduler for machines with no more than 16 CPUs.  But mainline won't accept BFS because 1) It's slower on machines with more than 16 cores, and 2) They're unwilling to allow a compile time option to select a cpu scheduler because it might devide developement time between multiple schedulers.
[18:18] <Sarvatt> davidm: i'm seeing multiple bugs from people with vaio F series machines that need to boot with i8042.nopnp for the touchpad to work if you could try that
[18:18] <davidm> apw tried that last night with a regular kernel and it did not work, nor did the longer version of that help
[18:19] <apw> i would suggest trying it with the test kernel as well in case you are affected by both
[18:19] <davidm> did kill my on-screen FN audio volume keys though
[18:19] <Darxus> The relevant bug is bug #424927.  And the last comment is that BFS is unlikely to be accepted into Ubuntu before mainline.  And mainline isn't going to accept it for agravating reasons.
[18:20] <davidm> apw the -22?
[18:21] <apw> yeah the one i gave you, try booting that with the i8042.nopnp option
[18:21] <davidm> OK I'll install that again (with headers this time) and add the command line info
[18:21] <davidm> back in a bit
[18:21] <apw> thanks
[18:36] <davidm> apw with -22 and i8042.nopnp mousepad works :-)  And I still have my FN key for audio working too.
[18:38] <davidm> Sarvatt, where are you seeing folks talking about the vaio F? I still have one more bug to sort my mic does not work.  Thanks
[18:38] <Sarvatt> davidm: what does sudo dmidecode -s system-product-name give for your machine?
[18:38] <Sarvatt> davidm: https://bugs.edge.launchpad.net/ubuntu/+source/xserver-xorg-input-synaptics/+bug/638219 is one of them
[18:38] <apw> davidm, ok great ... Sarvatt do you have a bug number for the i8042.nopnp jobbie ?
[18:39] <davidm> Sarvatt, VPCF1290X
[18:39] <Sarvatt> there are a ton of VPC-F series and it looks like not all of them are affected because hallyn has one too and doesn't need it
[18:40] <Sarvatt> which is going to making quirking fun :)
[18:42] <apw> Sarvatt, is hallyn ok if he does supply it, that would make the quirk easier
[18:42] <davidm> apw am I ok to keep using the test kernel for now?  I really like having my mousepad working ;-P
[18:43] <Sarvatt> apw: here is another with the same model davidm has from JFo - https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bug/641324
[18:45] <Sarvatt> https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bugs?field.searchtext=vaio+touchpad&orderby=-importance&search=Search&field.status:list=CONFIRMED&field.status:list=FIXCOMMITTED&field.status:list=INCOMPLETE_WITHOUT_RESPONSE&field.status:list=INCOMPLETE_WITH_RESPONSE&field.status:list=INPROGRESS&field.status:list=NEW&field.status:list=TRIAGED&field.assignee=&field.bug_reporter=&field.omit_dupes=on&field.has_patch=&field.has_no_package=
[18:45] <Sarvatt> almost all vaio F series
[18:52] <davidm> the bug JFo filed the bug for me, LP was timing out
[18:53] <Sarvatt> ah
[18:53] <ogasawara> davidm: care to post a quick comment to bug 641320 that the test kernel indeed does work for you
[18:55] <davidm> ogasawara, done
[18:56] <davidm> now if I can find something on my internal mic and external mic in not working I'm gold
[18:56] <davidm> can't use mumble without a mic of some sort
[19:06] <mathiaz> rmcbride: hi - re bug 639768
[19:06] <mathiaz> rmcbride: where does it hang?
[19:06] <rmcbride> mathiaz: at "Generating /etc/default/samba" near as I can see after 3 attempts
[19:07] <rmcbride> mathiaz: how can I help get more detail?
[19:07] <mathiaz> rmcbride: is there a "start smbd" process on the system?
[19:08] <rmcbride> mathiaz: Hmm, actually not in this case (it was doing that before I think) seems to be samba.postinst configure in two processes
[19:08] <mathiaz> rmcbride: ok - so you're probably seeing a different bug/issue
[19:08] <rmcbride> mathiaz: oh goodie :)
[19:08] <davidm> apw thanks for the help, and I think I found a fix for my mic but it won't be built for about 6 hours yet
[19:09] <rmcbride> mathiaz: I'll amend my comments in that bug and check for/file a new one as needed.
[19:09] <mathiaz> rmcbride: great - thanks
[20:19] <Willyant> I recently upgraded to Maverick and these are the bugs ive found when installing: Samba's "nmbd" process cannot be killed so i had to manually kill it using gnome-system-monitor.
[20:19] <Willyant> The postinst script therefore fails.
[20:21] <Willyant> My updatemanager crashed when i clicked on it as well.
[20:21] <Willyant> (The icon)
[20:21] <Willyant> Do i have to login to send the apport bugreports or is that only if i want to discuss them ?
[20:24] <micahg> Willyant: bug 639768
[20:24] <Willyant> Thanks micahg
[20:27] <Willyant> The last poster in that thread has narrowed it down somewhat "Actually it seems to be hanging during 'samba.postinst configure' which is a different thing apparently."
[20:29] <Willyant> The bug is in that package "samba" for ubuntu must kill "nmbd" but fails since itll not die. Maybe the smbd process keeps it alive and so smbd must be killed first ?
[20:29] <micahg> Willyant: well, there could be more than one issue, IIRC the upgrade to ubuntu7 fixed it for me
[20:29] <Willyant> Hehe, theres an ubuntu7 ?
[20:29] <micahg> Willyant: 2:3.5.4~dfsg-1ubuntu7
[20:30] <Willyant> ah, the samba package you mean ?
[20:30] <micahg> yes
[20:31] <Willyant> Ok, but i also tried updating to the new samba version and it also hung at postinst
[20:31] <Willyant> Ill see what version i have now
[20:31] <micahg> Willyant: did you upgrade cups as well?
[20:32] <Willyant> I upgraded everything, or tried to. Theres nothing more to install now.
[20:32] <Willyant> I think samba came before cups though...
[20:33] <Willyant> Cups should come before samba because its a dep of samba (if it matters)
[20:33] <Willyant> But thats not the problem.
[20:34] <Willyant> try to kill "nmbd", youll have to do it quite a few times before it dies
[20:35] <Willyant> Mine wont die if i issue a "killall -9 nmbd" for atleast 20 tries
[20:35] <micahg> Willyant: why not just stop it?
[20:35] <micahg> wfm
[20:35] <Willyant> Something is restarting it. Yep, ill have a look at the init script to see whats going on
[20:38] <Willyant> Aha, its also handeled by upstart
[20:39] <Willyant> So the postinst script could be designed only for sysinit
[20:39] <Willyant> Possible reason No1 :)
[20:41] <Willyant> micha: "stop nmbd" in the postinst for samba kills it right away.
[20:42] <Willyant> micha: Maybe you can add that in the thread ? -> "stop nmbd" in the postinst for samba kills it right away. Ive got to get coding.
[20:43] <Willyant> Also put that "Using killall -9 or kill -9 pidof nmbd" wont kill it anymore.
[20:44] <Willyant> Seems to also be an upstart bug to me. It should honour kill commands.
[20:48] <mathiaz> Willyant: if you have issue with nmbd it's not bug 639768
[20:49] <mathiaz> Willyant: please open a new bug rather than tagging along exisint Fixed Release bug
[20:49] <Willyant> The normal PowerOff button has changed to a wall-eletricity switch. Hard to find and is different from all other Off-Switches. (Minor detail though)
[20:50] <mathiaz> rmcbride: ^^ was Willyant report similar to what you've experienced with the samba stalled upgrade?
[20:50] <rmcbride> mathiaz: reading scrollback
[20:52] <rmcbride> mathiaz: yes, and in fact the comment referred to regarding "actually it seems to be..." was by me.
[20:52] <Willyant> Nice work!
[20:52] <Willyant> mathiaz: Thanks, but i had hoped someone could pitch in my experiences and atleast a working solution to the problem.
[20:52] <rmcbride> mathiaz: I had written #641519 on the new behavior
[20:53] <mathiaz> rmcbride: great thanks
[20:53] <Willyant> It seems to be upstart that refuses to let it be killed by signals.
[20:53] <rmcbride> ah
[20:53] <rmcbride> that fits
[20:53] <Willyant> Yep.
[20:54] <mathiaz> Willyant: is bug 641519 what you're experiencing too?
[20:54] <mathiaz> is there any messages in /var/log/samba/log.nmbd?
[20:54] <mathiaz> (as it seems that it is nmbd that blocks)
[20:54] <rmcbride> checking
[20:55] <Willyant> Nope, the issues i had where all releated to a perl/dpkg command that hung because nmbd refused to be killed.
[20:56] <rmcbride> nothing here either
[20:56] <mathiaz> rmcbride: hm ok.
[20:56] <Willyant> It hung the entire install process so thats not so good.
[20:56] <mathiaz> rmcbride: are you able to reproduce it?
[20:56] <Willyant> Otherwise everything was fine at install
[20:56] <mathiaz> rmcbride: if so, could you add a 'set -x' in samba postinst script?
[20:57] <mathiaz> rmcbride: it should be in /var/lib/dpkg/info/samba.postinst
[20:57] <rmcbride> mathiaz: Yep. every time. I'll do that now
[20:57] <Willyant> He can check that nmbd refuses to be killed
[20:57] <Willyant> :)
[20:57] <Willyant> Thanks!
[20:57] <mathiaz> rmcbride: that way we should be able to see where the postinst script hangs
[20:58] <Willyant> I saw that in gnome-system-monitor (enable the command view)
[21:00] <rmcbride> mathiaz: well... in order to get back to a state where I can install, I have to apt-get purge samba, which nukes the samba.postinst that I edited :/
[21:00] <rmcbride> ah Willyant I'll look at that
[21:01] <Willyant> Good.
[21:01] <mathiaz> rmcbride: hm right.
[21:02] <rmcbride> well this is frustrating. nmbd is not active on my system.
[21:03] <Willyant> nmbd -D
[21:03] <Willyant> cjwatsons comments about it beeing a cups error seems entirely wrong. If you kill something with a signal of "-9" it should just die, always.
[21:03] <mathiaz> rmcbride: you may wanna look into /var/log.samba/log.nmbd
[21:04] <mathiaz> Willyant: you're seeing a different bug
[21:04] <rmcbride> mathiaz: no such animal in this case
[21:04] <Willyant> Checking...
[21:04] <rmcbride> I have /var/log/samba/log.everylastmachineI'vesmbconnectedto
[21:04] <rmcbride> but no nmbd
[21:05] <Willyant> #639768 (Sorry cjwatson) ... wrong bug
[21:05] <mathiaz> rmcbride: hm - I can properly install samba on a test vm here
[21:05] <persia> mathiaz, Can you upgrade it?
[21:06] <mathiaz> persia: I'm gonna try that now
[21:06] <Willyant> can anyone paste the output of that samba postinst ?
[21:06] <Willyant> Or tell me how i can obtain it...
[21:06] <mathiaz> Willyant: this is what I see on the terminal: http://paste.ubuntu.com/495508/
[21:07] <mathiaz> Willyant: it's a successfull install though
[21:07] <Willyant> Im very sure it tries to kill the nmbd process by a signal but that doesnt work because upstart refuses that and or just restarts it again
[21:07] <mathiaz> rmcbride: from which version were you trying to upgrade from?
[21:07] <rmcbride> Willyant: here's my output with the hung install http://paste.ubuntu.com/495509/
[21:07] <mathiaz> rmcbride: you should be able to get the information from /var/log/dpkg.log
[21:07] <rmcbride> mathiaz: at first I was trying to upgrade yesterday
[21:08] <rmcbride> mathiaz: looking
[21:08] <rmcbride> mathiaz: today I'm just trying to install it after purging
[21:08] <mathiaz> rmcbride: I'd like to know the last two versions of the samba package that have been installed
[21:08] <mathiaz> rmcbride: so that I can reproduce an upgrade
[21:08] <Willyant> Hmm, are old apport things stored somewhere ?
[21:08] <rmcbride> mathiaz: 2010-09-10 16:36:30 status half-installed samba-common 2:3.5.4~dfsg-1ubuntu3
[21:09] <rmcbride> mathiaz: that may not be the last though, still looking
[21:10] <Willyant> Aww! The apport bugreport has been converted to my language so its useless for you
[21:10] <Willyant> (Or... ehem: /var/crash/samba.0.crash)
[21:10] <rmcbride> mathiaz:  2:3.5.4~dfsg-1ubuntu4 is the version fron yesterday
[21:10] <rmcbride> mathiaz: or rather Wednesday
[21:10] <Willyant> Thats me killing the post installation script though
[21:11] <mathiaz> rmcbride: so IIUC you tried to upgrade from ubuntu4 to ubuntu7
[21:11] <mathiaz> rmcbride: ubuntu4 was working correctly
[21:11] <rmcbride> mathiaz: ubuntu5
[21:11] <mathiaz> rmcbride: and the upgrade to ubuntu7 fails
[21:11] <mathiaz> rmcbride: ah ubuntu5
[21:11] <rmcbride> mathiaz: I hadn't gotten close enough int he log
[21:11] <rmcbride> ubuntu4 was back on the 10th
[21:12] <Willyant> Tags: maverick
[21:12] <Willyant> Title: update-manager crashed with TransactionFailed in _run()
[21:12] <mathiaz> rmcbride: so no 2:3.5.4~dfsg-1ubuntu6 ?
[21:12] <Willyant> Thats the update icon i pressed
[21:12] <Willyant> Its a segfault btw
[21:12] <rmcbride> mathiaz: I do see ubuntu6 in the log from yesterday, but hte behavior was no different (in that the install hung)
[21:13] <mvo_> Willyant: could you please report a bug?
[21:13] <mathiaz> rmcbride: ok thanks
[21:13] <mvo_> Willyant: about the u-m crash with the backtrace?
[21:13] <mathiaz> rmcbride: I'll try to reproduce it here
[21:13] <Willyant> mvo_: Yes, i have the backtrace
[21:13] <rmcbride> mathiaz: OK. If I can help in any other way, do re-ping me. I'm highly motivated to help fix this :)
[21:14] <mathiaz> rmcbride: cool - thanks!
[21:14] <Willyant> mvo_: I can ul it so you can get it if you want ?
[21:14] <mvo_> Willyant: please do , the full backtrace is important to fix the bug
[21:14] <Willyant> Will do, hold on
[21:15] <mvo_> something like paste.ubuntu.com is fine as well for now, but a real bugreport is better
[21:16] <Willyant> Do with it as you like: http://mange.dynalias.org/linux/misc/ubuntu_Maverick_Update_manager_Crash/_usr_bin_update-manager.1000.crash
[21:17] <mathiaz> rmcbride: is cups running on your system as well?
[21:18] <rmcbride> mathiaz: it is
[21:18] <mvo_> Willyant: yes, thanks
[21:18] <mathiaz> rmcbride: ok - thanks
[21:18]  * mathiaz installs cups as well
[21:18] <Willyant> Np
[21:25] <sistpoty> hey siretart, around? what do you think about bug #588203?
[21:26] <sistpoty> (I just don't know enough about the state of rdepends nor about emacs in general *g*)
[21:27] <ScottK> Maybe barry knows.  IIRC he's of the church of emacs.
[21:27] <sistpoty> ^^
[21:28]  * barry wakes up
[21:28] <barry> yep, i heart emacs
[21:28] <barry> sistpoty: you want me to look at that bug?
[21:28] <sistpoty> barry: yes, was about to write that :)
[21:29] <sistpoty> barry: please keep a special look out on (build)-rdepends
[21:29] <barry> :) looking...
[21:29] <barry> sistpoty: you mean that differ from the current package?
[21:29] <sistpoty> barry: or could be broken by the update
[21:30] <barry> sistpoty: gotcha
[21:30] <barry> sistpoty: i'll investigate and comment on the bug, okay?
[21:30] <sistpoty> barry: that'd be great :)
[21:30] <Willyant> mvo_: Another suggestion is to recompile gadmin-dhcpd with ./configure .... DHCPD_CONF="/etc/dhcp3/dhcpd.conf" instead of what it has now: DHCPD_CONF="/etc/dhcp3/dhcpd/dhcpd.conf" (Im not sure how this happened)
[21:31] <sistpoty> (if (true) (print "sistpoty no good at emacs-lisp) (print "upgrade rejected"))
[21:31] <Willyant> Hmm, i have won something called The Famous Why award. Is that any cool ? :)
[21:31] <barry> this one is interesting: * Use -O1 rather than -O2 on ia64. Fixes a build failure (looks
[21:31] <barry>      like a broken byte compiler) with newer versions of gcc
[21:31] <barry>      (c.f. #207580).
[21:32] <barry> maybe not related, but we've just gotten reports today of python being broken by some versions of gcc on 10.04.1
[21:32] <sistpoty> barry: I doubt ia64 should be much of a problem :)
[21:32] <Willyant> barry: Optimisations other then those the sources are packaged with is always risky.
[21:33] <sistpoty> (of course if it's a genuine gcc problem not only on ia64, than that's very noteworthy)
[21:33] <barry> Willyant: yep, and it's probably not related, it just caught my eye ;).  we don't know enough about the python problem yet, but we think it's gcc related.  let me see if i can dig up the python issue
[21:33] <Willyant> Okies
[21:34] <barry> http://bugs.python.org/issue9880
[21:34] <barry> gcc 4.6 == bad apparently
[21:35] <barry> doesn't affect mav, but it's still interesting
[21:35] <geser> sistpoty: you sure you used enough brackets? :)
[21:37] <Willyant> barry: Is this the python version used ?:  /usr/local/src/python-2.7... or it it a case of mixed /usr/lib links because its installed in a different location ?
[21:37] <Willyant> Highly probable
[21:38] <barry> Willyant: OP was building python 2.7 from upstream svn head.  there was an email thread on python-dev about it, and i could not reproduce no matter what pythons i had installed locally.
[21:38] <barry> Willyant: i think the mixture isn't the problem.  python should be good about isolation there.  at least, it was not a prob for me
[21:39] <Willyant> I see. Also watch aout for version changes, because pythons installer isnt exactly very good :)
[21:39] <barry> ;)
[21:39] <barry> doko has actually made py2.7 available in mav, which is nice (not officially supported though - wait til natty!)
[21:40] <sistpoty> geser: no idea, I don't have vim bracket highlights in kvirc :P
[21:41] <geser> sistpoty: all I know about lisp is that you need many, many brackets :)
[21:41] <mr_pouit> sistpoty: yeah, sure (about xubuntu delegation)
[21:41] <sistpoty> mr_pouit: cool, thanks!
[21:41] <sistpoty> geser: heh
[21:41] <Willyant> Ive built many B/LFS systems and i see the weaknesses in python. One is that for each new version it installs to different dirs and sometimes its failed to symlink. Not to speak of the addon things for python that must be recompiled to be linked to the new python directory. Seems like a pain.
[21:42] <ScottK> It is.  Python 3 will be better.
[21:42] <Willyant> gtk-clutter was also a pain to be fair
[21:42] <Willyant> Good!
[21:42] <barry> what ScottK said
[21:44] <Willyant> Can the same packages be recompiled again once frozen ?
[21:45] <sistpoty> Willyant: sure
[21:46] <Willyant> Good... See my post on gadmin-dhcpd above^
[21:48] <Willyant> Steps to reproduce the mal configured path: 1. Install and lauch gadmin-dhcpd 2. Watch it suggest the wrong default dhcpd config dir.
[21:48] <Willyant> So the users will need to go into the settings and select a new path. Irritating.
[21:49] <sistpoty> Willyant: I guess "--sysconfdir=/etc/dhcp3" might be at fault?
[21:49] <sistpoty> (together with no additional flags passed)
[21:50] <Willyant> Yep, it should be "/etc" and this should be set to the conf path DHCPD_CONF="/etc/dhcp3/dhcpd.conf"
[21:51] <Willyant> Just change that conf path and set sysconfdir to /etc
[21:52] <Willyant> ### Debian/Ubuntu/etc ###
[21:52] <Willyant> # ./configure --prefix=/usr --sysconfdir=/etc \
[21:52] <Willyant> # --localstatedir=/var/lib/dhcp3 --sbindir=/usr/sbin \
[21:52] <Willyant> # DHCPD_CONF="/etc/dhcp3/dhcpd.conf" \
[21:52] <Willyant> # DHCPD_BINARY="dhcpd3" \
[21:52] <Willyant> # SYSINIT_START_CMD="update-rc.d -f dhcpd defaults" \
[21:52] <Willyant> # SYSINIT_STOP_CMD="update-rc.d -f dhcpd remove" &&
[21:52] <Willyant> # make && make install
[21:53] <Willyant> Sais so in the package as well.
[21:54] <Willyant> Latest is: http://mange.dynalias.org/linux/gadmin-dhcpd/gadmin-dhcpd-0.4.9.tar.gz (Also asks to update remote failover pool configurations:)
[21:54] <sistpoty> Willyant: hm... not too sure if this won't break the upgrade path: configure:default_settings_path=`eval echo $sysconfdir`"/gadmin-dhcpd"
[21:55] <sistpoty> Willyant: how about just adding DHCPD_CONF="/etc/dhcp3/dhcpd.conf"?
[21:55] <Willyant> eval echo $sysconfdir will evaluate to /etc (Or should)
[21:56] <Willyant> sistpoty: Will be much better as itll work out-of-the-box then.
[21:56] <sistpoty> Willyant: sure it *should*, but did it for the last release?
[21:56] <Willyant> :)
[21:56] <Willyant> Show the configure script for gadmin-dhcpd...
[21:57] <Willyant> Or a cmd to let me view it
[21:58] <sistpoty> Willyant: dget -x http://archive.ubuntu.com/ubuntu/pool/universe/g/gadmin-dhcpd/gadmin-dhcpd_0.4.7-1.dsc (for lucid)
[21:59] <Willyant> wget :)
[22:00] <Willyant> sistpoty: E_NO_CONFIGURE_PARAMS :)
[22:00] <sistpoty> Willyant: look at debian/rules
[22:01] <Willyant> Url ?
[22:02] <Willyant> The packager is doing military service atm :(
[22:02] <sistpoty> sistpoty: just call dget (as found in devscripts)... it will download and extract the package
[22:03] <sistpoty> (it's not a typo for wget :P)
[22:03] <Willyant> Aha! :)
[22:03] <Willyant> E_NO_SUCH_PKG
[22:04] <sistpoty> hm? you'll need the url to the .dsc file as argument to dget
[22:04] <Willyant> E_NOT_INSTALLED && E_NO_SUCH_PACKAGE
[22:05] <Willyant> "dget"
[22:05] <sistpoty> Willyant: devscripts
[22:05] <Willyant> Thx
[22:06] <cjwatson> Willyant: FYI, upstart doesn't and can't affect how a process deals with signals sent by kill
[22:07] <Willyant> Something is ignoring the kills signals it recieves, maybe smbd restarts it ?
[22:07] <Willyant> Or its a bug in nmbd
[22:07] <cjwatson> the only way that upstart is involved in that is that it becomes the parent of processes which don't have a parent of their own
[22:07] <barry> interesting.  sbuild is barfing on trying to install mailx package
[22:07] <cjwatson> I have no idea, just letting you know that it's quite inaccurate to point at upstart for this
[22:08] <cjwatson> and regarding cups, that wasn't a guess, we had debugged that problem very precisely.  I assume you're seeing something else.
[22:08] <Willyant> cjwatson: Good to know. Then the nmbd process itself will not accept a kill signal or the smbd will start it up again.
[22:09] <cjwatson> processes are entitled to wait for a while when they receive SIGTERM
[22:09] <Willyant> But what kill signal does upstart emit ? (stop nmbd) ?
[22:10] <Willyant> SIGKILL is instant
[22:10] <Willyant> give or take a few microseconds
[22:10] <cjwatson> assuming that the process isn't in uninterruptible sleep
[22:10] <Willyant> Yeah.
[22:10] <cjwatson> SIGKILL is an unfriendly way to stop a process anyway
[22:11] <cjwatson> postinst scripts shouldn't use it
[22:11] <Willyant> Its there because its a --force'd kill.
[22:11] <Willyant> It should override everything as far as the kernel is conserarned
[22:11] <cjwatson> no
[22:12] <Willyant> It should override everything as far as the kernel is consearned
[22:12] <cjwatson> you need a copy of Stevens :)
[22:12] <cjwatson> anyway, upstart sends SIGTERM, waits a while, then sends SIGKILL if it has to
[22:13] <Willyant> I am a copy of brilliance. I feel no need to know more copies of excellence ;)
[22:13] <Willyant> Thanks!
[22:13] <Willyant> So, then somehow SIGKILL doesnt work on "nmbd"
[22:13] <cjwatson> stop(8) is probably the right way for the postinst to stop nmbd, unless there's some reason to do otherwise
[22:13] <Willyant> Good, we have narrowed it down
[22:14] <cjwatson> but if SIGKILL doesn't work, then the process is in uninterruptible sleep and there is no guaranteed way to stop it, at all
[22:14] <Willyant> Naah, it must adhere to the KILL signal
[22:14] <cjwatson> wrong
[22:14] <cjwatson> guess you've never used NFS :)
[22:14] <Willyant> I have :)
[22:15] <cjwatson> then you've almost certainly encountered situations where the NFS server fails and all the clients go into uninterruptible sleep
[22:15] <Willyant> Somewhat a mess of about 10 services (daemons including portmap)
[22:15] <Willyant> That need to be started in order
[22:15] <cjwatson> unless your experience is recent enough that all the mounts were with intr
[22:16] <Willyant> "If something isnt restarting nmbd, like upstart then its a bug in samba"
[22:17] <Willyant> Agreed ?
[22:17] <Willyant> "Hmm, it could also be smbd that keep-alives the nmbd process"
[22:18] <cjwatson> no, that's not a given either
[22:18] <cjwatson> kernel bugs can leave a process stuck in uninterruptible sleep
[22:18] <Willyant> Tell me what you think then ... Nope, thatd be a nmbd bug as well.
[22:19] <Willyant> Youre talking zombies, this is not the case. Checked that.
[22:19] <cjwatson> uninterruptible sleep != zombie.
[22:19] <cjwatson> please, find a copy of "Advanced Programming in the Unix Environment" before using these terms
[22:20] <Willyant> If i make a program and it goes into what you call "Uniterruptable sleep" itll be an infinite timout value or a spinloop.
[22:20] <Willyant> Dont need one. I have several C books
[22:21] <cjwatson> they're not telling you the right things; this is I/O wait *inside the kernel*
[22:21] <cjwatson> it is a detail of Unix, not a detail of C
[22:21] <Willyant> A linger you mean ?
[22:21] <cjwatson> OK.  So.  Where is the evidence that the samba postinst is trying to kill nmbd?  It's not clear from your bug report.
[22:21] <cjwatson> there's no process tree or anything in there, or strace of what the postinst is doing
[22:22] <Willyant> Well, i tried to do as apport suggested. I opted to send a bugreport for that too, but im not sure it got thru because i was presented with a launchpad login. Forgot the user/pass
[22:23] <cjwatson> the script that emits the last message in your bug report does not even try to stop nmbd, although it does try to start it
[22:23] <cjwatson> I'm reading bug 641519
[22:23] <sistpoty> cjwatson: btw, I think that bug #544139 might affect me too (xdm + fluxbox). Strange thing is that sometime if things get screwed up and I login, start konsole then any command seems to be mirrored to the first getty (and hence executed twice), howerver the first getty captures the plain output (e.g. of ssh-add's password, not a thing I like to have in any log)
[22:23] <rmcbride> cjwatson: 641519 was written by me
[22:24] <cjwatson> sistpoty: right, it *is* on my to-do list but other things keep bumping it off
[22:24] <sistpoty> cjwatson: of course it might be an entire separate bug though
[22:24] <cjwatson> rmcbride: ah
[22:24] <cjwatson> Willyant: ok, so how did you arrive at the conclusion that it's trying to kill nmbd?
[22:24] <sistpoty> cjwatson: if I can help in any way (unless it involves the kernel), feel free to ask :)
[22:25] <cjwatson> sistpoty: an extra four hours per day would be nice
[22:25] <cjwatson> sistpoty: or maybe babysitting
[22:25] <cjwatson> :-)
[22:25]  * sistpoty changes the time zones with his magic hand :P
[22:26]  * Willyant Goes away like Obi Wan - I guess the bug wasnt important :)
[22:26] <cjwatson> Willyant: um?  I'm trying to get to the bottom of what you reported
[22:27] <cjwatson> but it's important to do it step-by-step and logically
[22:27] <cjwatson> a good start would be to at least see a transcript of what's happening in your case
[22:29] <Willyant> Well, a dpkg command was run as seen by a launched instance of gnome-system-monitor with the process command column enabled and i thusley saw: perl something .. dpkg ... nmbd stop/wait something. When i manually killed nmbd the install process continued.
[22:29] <cjwatson> no terminal output from dpkg?
[22:29] <Willyant> this stop/wait comes from upstart
[22:30] <cjwatson> if you mean "stop/waiting", that actually means that as far as upstart is concerned the process isn't running and nothing has told it to start it
[22:30] <Willyant> I think i have submitted 2 apport bug reports, unless you have to have a valid login to actually submit those ?
[22:30] <cjwatson> you need a valid login to submit bugs to Launchpad, yes
[22:31] <Willyant> Crud, ok. Then ive only shown you the bugreport on the crashing update manager
[22:31] <Willyant> I can probably reproduce the thing, hold on...
[22:36] <Willyant> apt-get remove samba ; apt get install samba and this is what i see to begin with: "smbd start/running, process 4133
[22:36] <Willyant> start: Job failed to start"
[22:37] <Willyant> smbd isnt running after a reinstall
[22:38] <Willyant> ehm, i mean "nmbd"
[22:38] <Willyant> nmbd -D
[22:38] <Willyant> root      4172  0.0  0.2  12268  1476 ?        Ss   23:38   0:00 nmbd -D
[22:38] <Willyant> root      4174  0.0  0.1   5168   740 pts/1    S+   23:38   0:00 grep nmbd
[22:38] <Willyant> Then its running
[22:39] <cjwatson> did the samba installation complete?
[22:40] <Willyant> Yes, but previously perl and dpkg tries to run postinst and fails. Thats for sure.
[22:40] <cjwatson> if not, was it with a hang, and if not a hang, was there an error message after "start: Job failed to start"?
[22:40] <cjwatson> it's best to post an exact transcript to e.g. paste.ubuntu.com
[22:40] <cjwatson> or pastebin.com or whatever floats your boat
[22:40] <Willyant> Ställer in samba (2:3.5.4~dfsg-1ubuntu7) ...
[22:40] <Willyant> update-alternatives: använder /usr/bin/smbstatus.samba3 för att tillhandahålla /usr/bin/smbstatus (smbstatus) i automatiskt läge.
[22:40] <Willyant> smbd start/running, process 4133
[22:40] <Willyant> start: Job failed to start
[22:40] <cjwatson> please don't ever paraphrase error messages, it's the number one way to get developers not to look :)
[22:41] <Willyant> Dont translate bash then :)
[22:41] <cjwatson> is that the last text on the screen?  is it hung at that point?
[22:41] <Willyant> Nope, it didnt hang
[22:41] <cjwatson> so what do you mean by "perl and dpkg tries to run postinst and fails"?
[22:41] <cjwatson> if the installation completed, then they succeeded
[22:42] <Willyant> So the new version seems to be ok, but not the current Beta of Maverick. Checking to see if samba.conf is the same
[22:42] <Willyant> smb.conf is the same.
[22:42] <Willyant> So the new samba version seems to be ok.
[22:42] <cjwatson> if the new version is working, great, we don't have to worry.  It will supersede the beta
[22:42] <cjwatson> we do know that we've fixed bugs since beta ;-)
[22:43] <Willyant> Good, but why wasnt this picked up before beta release ? /Just wondering
[22:43] <cjwatson> that I can't tell you
[22:44] <Willyant> Ok, can you diff the postinst scripts with current and prev versions for me ?
[22:44] <cjwatson> er, I suppose I could, but it's already quarter-to-eleven on a Friday night for me.  You could do it yourself :-)
[22:45] <Willyant> cjwatson: Okies, go out and party! :)
[22:45] <cjwatson> sadly, I have three more bugs to fix
[22:45] <Willyant> Ok, well we have solved this i think. It would have been a bad one.
[22:46] <Willyant> cjwatson: Saw the crash i posted on update manager ?
[22:46] <cjwatson> no
[22:46] <cjwatson> not really likely to be involved in that
[22:46] <Willyant> Here you go: http://mange.dynalias.org/linux/misc/ubuntu_Maverick_Update_manager_Crash/_usr_bin_update-manager.1000.crash
[22:46] <Willyant> Okies
[22:46] <Willyant> I just clicked the icon
[22:46] <cjwatson> Ubuntu is a big project, not every developer does everything
[22:47] <Willyant> But some can tell others to fix things to make things solid
[22:47] <Willyant> I like all dists. I dont like bugs.
[22:47] <cjwatson> I could, but the developer working on update-manager is already working as hard as he possibly can
[22:47] <Willyant> Ok, good.
[22:54] <Willyant> I like the new icon theme in Maverick, but the "Shutdown-switch" is far worse then the old one that was actually very good.
[22:55] <Willyant> Melts into the panel and becomes almost invisible.
[22:56] <Willyant> In contrast, the old switch had a nice graphical layout, but was a few pixels too high and wide.
[22:57] <Willyant> But waay better
[22:58] <Willyant> Show Icons in applications is back (Very good). I didnt like that.
[23:00] <Willyant> cjwatson: Cant there be an "Anonymous submitter" for the apport bugs ?
[23:01] <cjwatson> no, because then we would be unable to contact submitters for clarification
[23:01] <cjwatson> bugs aren't fire-and-forget kinds of things
[23:01] <cjwatson> we often need to find out more detail about something or other that apport doesn't capture
[23:02] <Willyant> But they are generated by apport and i think it could be good to read if many applies to the same package.
[23:02] <Willyant> Otherwise discarded
[23:03] <Willyant> Make anonymous bugreports, have an evaluation script see if many are the same and then move to anonymous-reports
[23:07] <Willyant> Im making a gui for torque cluster atm. Would this be a welcome addition granted its good ?
[23:13] <Willyant> cjwatson: NFS guis are hard to make because of the many portmapper binaries needing to be started first. What are those used for ?
[23:13] <cjwatson> no idea
[23:13] <Willyant> Its the only application that requires them
[23:13] <Willyant> Okies
[23:13] <cjwatson> I try to avoid knowing about its intricacies
[23:15] <Willyant> xinetd is also useless and i think it should be phazed out by patching the last thing requireing it (tftpd)
[23:16] <Willyant> Or just run: in.tftpd ... There no server requires it.
[23:16] <Willyant> It was meant to save a few kb of ram, but it makes servers suffer because they cant use all the features then.
[23:17] <Willyant> (When not used)
[23:17] <Willyant> And the servers get a bit slower
[23:19] <Willyant> In terms of usability xinetd have done more bad then good.
[23:20] <Willyant> cjwatson: This comes from 15 years of coding with many server developers.
[23:22] <cjwatson> I'm not involved with that
[23:23] <cjwatson> I do installer and bootloader work
[23:23] <cjwatson> the only reason I chipped in on your samba problem was because you mentioned my name in connection with it
[23:23] <Willyant> cjwatson: Do you have an oppinion on the matter ?
[23:23] <cjwatson> no
[23:23] <Willyant> cjwatson: Yes, i noticed that. Ok
[23:24] <Willyant> cjwatson: Who can i speak to for dropping the obsolete xinetd package ?
[23:24] <cjwatson> file a bug
[23:25] <cjwatson> IRC is a hopeless way to report bugs, generally
[23:25] <Willyant> For a 386 computer it could possibly be of use
[23:25] <Willyant> Naah :)
[23:25] <Willyant> Ive done it for many many years. Someone is always listening
[23:25] <cjwatson> if you like having your bugs forgotten because the person who might deal with it happens not to be around, it's great
[23:25] <Willyant> .
[23:26] <cjwatson> please do not use this channel as a bug reporting channel
[23:26] <cjwatson> we don't mind the odd comment alerting us to something serious
[23:26] <cjwatson> it's primarily for discussion among developers
[23:26] <cjwatson> if it became a bug reporting channel we would never get anything done
[23:26] <Willyant> So developers doesnt like to fix bugs ?
[23:26] <Willyant> Crusial bugs
[23:26] <cjwatson> we are entirely happy to fix bugs reported through the system designed to help us track them
[23:27] <Willyant> Aha, so then you think i cant report bugs on irc at all ? /We can talk about other things if you like. What do you suggest ?
[23:28] <Willyant> Camping experiences ?
[23:28] <Willyant> :)
[23:28] <cjwatson> you don't need to talk about anything if you have nothing on-topic to say.
[23:28] <cjwatson> you are dangerously close to being asked to leave.
[23:28] <Willyant> My name is danger in that case. Now get back on topic please
[23:28] <cjwatson> we welcome bug reports via our bug tracking system.
[23:36] <Rankor> "Please see that not everyone going against your beliefs is a troll"
[23:36] <Rankor> cjwatson
[23:37] <cjwatson> this has nothing to do with going against my beliefs; it is to do with retaining the purpose of the IRC channel
[23:37] <Rankor> "Development of Ubuntu"
[23:37] <cjwatson> and I can't say I'm hugely impressed with nick-morphing either
[23:37] <sistpoty> Rankor: that's a development channel nonetheless, we *need* to keep the discusssion more or less focussed on development
[23:38] <cjwatson> you were advised that bug reports should go to the bug tracking system, multiple times
[23:38] <Rankor> Im not sure where i digressed
[23:38] <cjwatson> it's a perfectly simple thing
[23:38] <cjwatson> Ubuntu is far too big a project for bugs to be reported just as a single giant stream of stuff on IRC
[23:38]  * Rankor goes Obi Wan ... And like that... Pfsst