[00:00] <IntuitiveNipple> To do this remote support, I've got the user in S.A. on a Windows box with putty, ssh to my laptop here in the U.K, then a screen Multiuser session joining us, then ssh back to the user's site and port-forwarded to the faulty machine :)
[00:06] <digistyl3> hi everyone, i think there's a problem with todays acpi-support update
[00:07] <digistyl3> it's changelog mentiones adding 90-hdparm.sh to the ac.d folder
[00:08] <digistyl3> while it is added to battery.d, resume.d, start.d and suspend.d (which isn't mentioned in the changelog), i can't find it in ac.d
[00:08] <digistyl3> eventhough it states it in the changelog
[00:09] <digistyl3> and so it didn't change hdparm-s B parameter to 254 as it is described in the changelog
[00:09] <digistyl3> i'm using version 0.111
[00:09] <slangasek> bryce: ^^
[00:10] <bryce> slangasek: ok I'll take a look
[00:13] <bryce> $ ls */90-hdparm.sh
[00:13] <bryce> ac.d/90-hdparm.sh  battery.d/90-hdparm.sh  resume.d/90-hdparm.sh  start.d/90-hdparm.sh
[00:13] <bryce> hmm, appears to be there for me
[00:14] <slangasek> I don't see ac.d/90-hdparm.sh in the binary package, perhaps there's an install rule that needs updated
[00:14] <digistyl3> ls */90-hdparm.sh
[00:14]  * bryce tests install
[00:14] <digistyl3> resume.d/90-hdparm.sh  start.d/90-hdparm.sh
[00:14] <digistyl3> wierd :|
[00:14] <cjwatson> bryce: milestones> ah, thanks
[00:14] <wgrant> william@irranat:/etc/acpi$ ls */90-hdparm.sh
[00:14] <wgrant> resume.d/90-hdparm.sh  start.d/90-hdparm.sh
[00:14] <wgrant> I upgraded about 5 minutes ago.
[00:15] <bryce> yeah debian has their own completely replaced debian/ dir, and I may have missed tweaks to .install files
[00:16] <wgrant> Hmm, why are they not symlinks?
[00:16] <bryce> dunno
[00:16] <bryce> feel free to send me a patch
[00:17] <bryce> aha, ok so rules installs stuff in suspend.d, start.d, resume.d, events, and lib, but no other dirs.  Should be easy to fix.
[00:18] <wgrant> Hmm, is that script meant to actually do anything in the default installation?
[00:19] <wgrant> laptop-mode seems to default to controlling it, but it isn't doing a very good job.
[00:20] <digistyl3> before this i used laptop-mode
[00:20] <digistyl3> and it did it's job for me
[00:20] <digistyl3> setting hdparm -B to 254 while on AC and to 128 while on battery (as i told him to do)
[00:20] <digistyl3> but before i installed the update i disabled laptop-mode, as it is disabled in a default instalation
[00:21] <wgrant> What triggers laptop-mode? I can hear my disk slowly killing itself a couple of times a minute on AC.
[00:22] <digistyl3> wgrant: i had to enable laptop-mode in /etc/default/acpi-support
[00:23] <wgrant> Umm, that's bad.
[00:23] <digistyl3> then in /etc/laptop-mode/laptop-mode.conf i changed some values (255 to 254 when on ac, and 1 to 128 when on battery)
[00:23] <wgrant> But true.
[00:24] <slangasek> I thought that was supposed to be fixed at a different level now, so that -B 254 was the default
[00:24] <wgrant> 90-hdparm.sh checks if laptop-mode is installed, and if so it checks its config file. If laptop-mode's config file says that it will manage HDD power settings, 90-hdparm.sh does nothing.
[00:24] <wgrant> Even if laptop-mode is disabled in /etc/default/acpi-support.
[00:25] <slangasek> i.e., without any intervention from laptop-mode or acpi-support
[00:25] <bryce> digistyl3, wgrant: is there a LP# for this missing 90-hdparm.sh issue?
[00:25] <bryce> digistyl3, wgrant: also, I'm curious if the 90-hdparm.sh is useful for you?
[00:25] <wgrant> bryce: I haven't looked, and only discovered it 10 minutes ago.
[00:25] <digistyl3> bryce: i'll check, if there isn't one, should i create one?
[00:26] <wgrant> bryce: I was previously running hdparm manually after resume after I noticed things going bad... it would be useful if it worked, yes.
[00:26] <wgrant> But in the default installation it doesn't do anything.
[00:26] <ogra> acpi-support isnt used at all anymore on suspend/resume (at least its not supposed to as it interferes with the new default now)
[00:27] <bryce> digistyl3: I'm sure slangasek would appreciate having the change associated with a bug number
[00:27] <ogra> look in pm-utils for stuff being executed
[00:27] <slangasek> bryce: at all times, yes :-)
[00:27] <wgrant> How is laptop-mode meant to be run now?
[00:27] <slangasek> I thought it was meant to be obsolete
[00:28] <wgrant> Right.
[00:28] <ogra>  /usr/lib/pm-utils/sleep.d/96laptop-mode
[00:28] <ogra> from there
[00:28] <wgrant> But 90-hdparm.sh is a noop if it's installed and meant to be managing your HD power settings.
[00:28] <slangasek> well, acpi-support is also meant to be obsolete
[00:29] <ogra> right
[00:29] <ogra> its only there for backwards compatibility with user written scripts
[00:30] <slangasek> which one? :)
[00:30] <ogra> no idea, but that was the reason to keep it in hardy
[00:30] <ogra> :)
[00:31] <slangasek> I mean: which one, laptop-mode or acpi-support?
[00:31] <ogra> acpi-support
[00:31]  * cjwatson unbreaks the mobile seeds a bit to avoid breaking the publisher
[00:31] <ogra> cjwatson, what did i break ?
[00:31] <cjwatson> hopefully that's all that's needed
[00:32] <cjwatson> you didn't, persia and StevenK did ;-)
[00:32] <ogra> the meta upload wasnt approved yet
[00:32] <cjwatson> the blacklist file is still needed
[00:32] <ogra> phew
[00:32] <cjwatson> doesn't affect mobile-meta
[00:32] <ogra> yeah lool wondered as well about all that dropping
[00:32] <cjwatson> clearly nobody tried running germinate on it as a test case
[00:33] <cjwatson> * Downloading http://people.ubuntu.com/~ubuntu-archive/seeds/mobile.intrepid/blacklist
[00:33] <cjwatson> ! Could not open (any of):
[00:33] <cjwatson> !   http://people.ubuntu.com/~ubuntu-archive/seeds/mobile.intrepid/blacklist
[00:33] <digistyl3> bryce: i reported it: https://bugs.launchpad.net/ubuntu/+source/acpi-support/+bug/274599
[00:34] <bryce> thanks
[00:36] <digistyl3> yw
[00:37] <wgrant> If laptop-mode-tools is obsolete, why is it recommended by ubuntu-desktop?
[00:37] <bryce> digistyl3: btw it doesn't install into suspend.d
[00:37] <digistyl3> what the... i have no sound after the latest updates :D
[00:37] <slangasek> perhaps I'm wrong about that part; but I thought g-p-m was taking over most of the duties of laptop-mode
[00:38] <ogra> well, pm-utils
[00:38] <ogra> g-p-m is a dumb frontend :)
[00:38] <slangasek> sure
[00:40] <digistyl3> bryce: sorry, my bad :)
[00:41] <ogra> wgrant, see /usr/lib/pm-utils/sleep.d/96laptop-mode
[00:41] <bryce> slangasek:  btw it was brought to my attention that while I'd indicated in the changelog that thinkpad.modprobe was reenabled, that was not actually the case.  I'll drop mention of that in the previous changelog entry.
[00:41] <slangasek> ok
[00:42] <wgrant> ogra: So it still respects /etc/default/acpi-support?
[00:42] <ogra> it only reads the ENABLE_LAPTOP_MODE value from either there or /etc/default/laptop-mode
[00:42] <ogra> to then call /usr/sbin/laptop_mode auto
[00:42] <wgrant> The latter of which doesn't exist.
[00:43] <wgrant> So by default it's all pretty broken.
[00:43] <bryce> any other acpi-support changes to put in while I'm at it?
[00:43] <ogra> why is it broken ?
[00:43]  * slangasek glares at python-wxgtk2.6
[00:45] <slangasek> >>> import wx
[00:45] <slangasek> Traceback (most recent call last):
[00:45] <slangasek>   File "<stdin>", line 1, in <module>
[00:45] <slangasek> ImportError: No module named wx
[00:45] <wgrant> ogra: Because laptop-mode-tools is installed, its CONTROL_HD_POWERMGMT defaults to 1, and /etc/default/acpi-support defaults to disabling laptop-mode. Thus 90-hdparm.sh does nothing.
[00:45] <digistyl3> wgrant: have you restarted your computer? do you have sound? :\
[00:45] <wgrant> digistyl3: I do. There are lots of sound drivers.
[00:46] <ogra> wgrant, right, on purpose .. please read the various mailthreads on -devel and -devel-discuss that piled up over the last yeras about that :) we never enabled it by default
[00:46] <TheMuso> digistyl3: WHat problem are you seeing exactly?
[00:46] <digistyl3> no sound at all, amarok won't play mp3s
[00:46] <digistyl3> when i hit test sound in gnome
[00:46] <digistyl3> i get this
[00:46] <digistyl3> audiotestsrc wave=sine freq=512 ! audioconvert ! audioresample ! gconfaudiosink: Failed to connect stream: Invalid argument
[00:46] <wgrant> ogra: So how is Joe User meant to know how to avoid killing his HDD?
[00:47] <ogra> he isnt ... its not an OS problem
[00:47] <digistyl3> TheMuso: it worked a few minutes ago, i just rebooted after the updates
[00:47] <TheMuso> digistyl3: Right. Can you please try the package for your architecture located here: http://www.themuso.id.au/ubuntu/alsa-plugins and tell me if anything has changed.
[00:47] <ogra> (as you can see from the mail threads i mentioned)
[00:47] <TheMuso> digistyl3: What updates did you download?
[00:47] <wgrant> ogra: In Intrepid's default configuration, my laptop's HDD parks multiple times a minute. This is a probably, and an easily fixable on.
[00:48] <wgrant> Er, this is a *problem*.
[00:48] <digistyl3> TheMuso: the ones from update manager, yes, it contained the respective package
[00:48] <digistyl3> TheMuso: can i restart only the alsa server or do i need a full reboot?
[00:48] <TheMuso> digistyl3: ok please try the package for your architecture at the url I gave above and tell me if anything has changed.
[00:48] <ogra> wgrant, really i'm not after discussing that issue every three months, there are mails and explanations of the issue a bug and even a lwn.net article
[00:49] <TheMuso> digistyl3: You shouldn't need to reboot.
[00:49] <bryce> fwiw, debian dropped the ENABLE_LAPTOP_MODE=false  bit in their version of acpi-support
[00:49] <wgrant> ogra: What are people recommended to do?
[00:50] <ogra> please read up about it ...
[00:50] <wgrant> Do you happen to recall the subject line of one of the threads?
[00:50] <wgrant> There's an awful lot of mail on -devel and -devel-discuss.
[00:50] <JontheEchidna> pitti: ahoy, wanna picture of jockey-kde from a native kde environment for your blog?
[00:50] <digistyl3> TheMuso: i installed the package, but i get the same error
[00:50] <digistyl3> maybe i do need a reboot?
[00:51] <TheMuso> digistyl3: yes please try a reboot.
[00:51] <ogra> wgrant, bug 59695 .... enjoy ...
[00:52] <wgrant> ogra: I was hoping for something with a slightly higher signal-to-noise ratio, but thanks.
[00:53] <slangasek> ogra: so how do we fix that bug, once for all?
[00:53] <ogra> wgrant, http://lwn.net/Articles/257426/
[00:53] <ogra> slangasek, teach HW vendors ?
[00:53] <wgrant> AFAICT this is a regression in Intrepid, at least for me.
[00:53] <ogra> that might be something else then
[00:54] <slangasek> ogra: er, TTBOMK all hardware DTRT with -B 254 and the problem has been that we weren't consistently using -B 254 in software
[00:54] <slangasek> if we /are/ doing -B254 everywhere now, then the bug should be closed
[00:54] <ogra> slangasek, well, then we should fix it elsewhere
[00:54] <ogra> but surely not in acpi-support
[00:54] <wgrant> The code that sets -B254 doesn't run in the default installation, AFAICT.
[00:54] <wgrant> Or is there another place that does it?
[00:54] <ogra> /etc/hdparm.conf
[00:55] <digistyl3> TheMuso: i'm back
[00:55] <digistyl3> i heard sound when starting gnome
[00:55] <TheMuso> digistyl3: how did it go?
[00:55] <digistyl3> but after that
[00:55] <digistyl3> it just dies
[00:55] <digistyl3> i get the same problems
[00:55] <wgrant> ogra: 'quiet' is set, and that's all.
[00:55] <TheMuso> digistyl3: so if you run the test you ran earlier, you get the same error?
[00:55] <wgrant> ogra: I installed this laptop with alpha 5, FWIW.
[00:55] <ogra> right, i was just pointing out a more proper place to add the fix
[00:55] <JontheEchidna> pitti: http://i31.photobucket.com/albums/c355/Woremar/jockey_05beta1.png
[00:55] <wgrant> ogra: Ah, right.
[00:56] <digistyl3> TheMuso: yes, exactly the same error
[00:56] <ogra> acpi-support is considered obsolete since a while ... doesnt appear right to me to add fixes there we want to persist if its dropped completely one day
[00:56] <TheMuso> digistyl3: Ok, thanks. I think that rules out the package I got you to test.
[00:57] <wgrant> ogra: One cannot actually remove laptop-mode-tools - is this deliberate?
[00:57] <TheMuso> digistyl3: Now I suggest running the following command to revert back to the originally stalled version: "sudo apt-get --reinstall install libasound2-plugins=1.0.17-0ubuntu3"
[00:58] <ogra> wgrant, i can remove it here
[00:58] <wgrant> ogra: It takes acpi-support and thus ubuntu-desktop with it here.
[00:58] <digistyl3> TheMuso: ok, i downgraded
[00:58] <ogra> and ?
[00:58] <wgrant> ogra: ... removing ubuntu-desktop isn't a solution.
[00:58] <slangasek> heh
[00:59] <bryce> should I add a Suggests for laptop-mode-tools?
[00:59] <wgrant> It isn't a solution to removing something that is meant to be obsolete and is turned off.
[00:59] <slangasek> bryce: I don't imagine it matters
[00:59] <bryce> ok; I see debian added it in their version
[00:59] <slangasek> I wouldn't balk at seeing it added in a freeze diff :)
[01:00] <ogra> wgrant, why do you *want* to remove it ?
[01:00] <wgrant> ogra: So 90-hdparm.sh does something.
[01:00] <bryce> slangasek: okies
[01:02] <TheMuso> digistyl3: and you said amarok doesn't work? Could you possibly try running a sound applicatino from the terminal, and pastebinning the terminal output?
[01:02] <wgrant> ogra: And while I have no problems with hacking the script or strange config files to get it working, users shouldn't have to.
[01:03] <ogra> wgrant, laptop-mode-tools isnt obsolete, acpi-support is
[01:03] <ogra> as well as pmi
[01:04] <digistyl3> TheMuso: ok
[01:04] <wgrant> ogra: Oh, I took your statement earlier as meaning that laptop-mode-tools was being replaced by pmutils and g-p-m. You meant acpi-support?
[01:04] <ogra> laptop-mode-tools is executed by pm-utils if its enabled
[01:04] <ogra> as i showed you before
[01:04] <wgrant> Right.
[01:05] <wgrant> So is 90-hdparm.sh's behaviour a bug? (it doesn't care if laptop-mode-tools is switched on or not)
[01:05] <ogra> anyway, 2am here .... bed time ... if -B254 is needed we should fix it in hdparm.conf and make it a default ... or even patch the hdparm binary to use that value by default
[01:06] <wgrant> Night ogra.
[01:06] <ogra> but honestly i dont think your disk access is realted to that, i would look for apps that do this or a kernel change
[01:06] <slangasek> eh, the issue was that things were using -B255 instead of -B254
[01:07] <slangasek> but that may have only been the acpi-support angle
[01:07] <digistyl3> TheMuso: only mplayer was able to produce sound
[01:07] <wgrant> Setting -B254 manually works fine.
[01:07] <digistyl3> i opened it with totem too
[01:07] <wgrant> So something isn't setting it.
[01:07] <digistyl3> here's the pastebin: http://pastebin.com/m79255621
[01:07] <slangasek> I would say that everything isn't setting it :)
[01:07] <ogra> wgrant, and you are 100% sure it was set in i.e. hardy ?
[01:07] <wgrant> slangasek: Probably, but I'd hope that only a subset of those were meant to be setting it.
[01:08] <wgrant> ogra: I'm not 100% sure, but I think it was.
[01:08] <TheMuso> digistyl3: thanks, looking.
[01:08] <wgrant> It's quite noticable when it does it.
[01:08] <slangasek> I'm not sure that -B254 was getting set in hardy; if it was, somebody forgot to close that darned bug
[01:08] <wgrant> I suppose it's possible that I tweaked a config file somewhere years back.
[01:09] <TheMuso> digistyl3: ah thats because it switched over to OSS output.
[01:09] <TheMuso> digistyl3: mplayer that is. Alsa still borked.
[01:09] <ogra> sladen, right, so wgrant's constant disk accesses that didnt happen in hardy are very unlikely related to hdparm at all
[01:09] <ogra> err
[01:09] <ogra> slangasek, ^^^
[01:09] <slangasek> seems that way
[01:09] <wgrant> ogra: DIsk accesses? Head parkings, you mean?
[01:09] <calc> cough, i just noticed i need to update a package and the beta freeze is started, if i get it updated in the next couple hours would it be acceptable?
[01:09] <TheMuso> digistyl3: thanks, I will have to do some further digging. Whats annoying is that I can't yet reproduce it here.
[01:09] <calc> its lucene2
[01:10] <TheMuso> digistyl3: What sound card are you using BTW?
[01:10] <slangasek> calc: if it's freeze-appropriate, yes...
[01:10] <digistyl3> TheMuso: yw, i'm using hda-intel on a dell studio 15
[01:10] <digistyl3> TheMuso: 00:1b.0 Audio device: Intel Corporation 82801H (ICH8 Family) HD Audio Controller (rev 03)
[01:11] <calc> slangasek: ok
[01:11] <TheMuso> digistyl3: Ok thanks. I don't think it is card related, but its goot to know.
[01:12] <bryce> slangasek: ha, I found this for thinkpad sleep issues (X60) - http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=407188
[01:12] <digistyl3> TheMuso: if i can debug it somehow, i'll help :)
[01:13] <bryce> slangasek: I wonder if that might help your sleep issue?
[01:13] <TheMuso> digistyl3: Thanks but I am not sure where to go from here, still trying to gather data.
[01:13] <slangasek> bryce: I don't have a reproducible sleep issue
[01:13] <slangasek> just the sleep key issue
[01:14] <TheMuso> digistyl3: what version of pulseaudio are you running BTW?
[01:14] <bryce> slangasek: hmm ok
[01:14] <ogra> wgrant, feel free to repoen bug 250938 ... seems thats related to your prob
[01:14] <digistyl3> TheMuso: 0.9.10-2ubuntu6
[01:14] <TheMuso> digistyl3: Right, thanks.
[01:14] <ogra> anyway, really bed time now
[01:17] <wgrant> ogra: Thanks, I'll have a look.
[01:22] <digistyl3> TheMuso: the funy part of this is that i hear those drums on the login window and also the login theme (the new one)
[01:22] <digistyl3> is that OSS too?
[01:22] <digistyl3> or alsa jams later?
[01:29] <TheMuso> digistyl3: I am not sure yet, but I don't think it is anything specifically related to pulseaudio or alsa, but I don't know any more than that yet.
[01:36]  * slangasek squints at libvlc2 Conflicts: libvlc1
[01:37] <digistyl3> TheMuso: if you need me, just say my name, i'll be watching a film
[01:39] <TheMuso> digistyl3: Sure. I can reproduce the problem with the sound test in the sound properties window, but not with mplayer.
[01:39] <crimsun> TheMuso: it's an alsa-lib issue.
[01:40] <TheMuso> crimsun: argh!
[01:40] <crimsun> TheMuso: you should be able to reproduce using the following test cases:
[01:40] <crimsun> 1) aplay -Dplughw:0 /usr/share/sounds/purple/send.wav
[01:40] <crimsun> 2) aplay -Dhw:0 /usr/share/sounds/purple/send.wav
[01:40] <crimsun> 3) aplay /usr/share/sounds/purple/send.wav
[01:41] <TheMuso> crimsun: 1 works for me
[01:41] <crimsun> I have pulse pcm+ctl active
[01:41] <TheMuso> crimsun: 2 works for me
[01:41] <TheMuso> crimsun: 3 eworks.
[01:42] <crimsun> digistyl3: can you test the above 3 cases?
[01:42] <TheMuso> in the case of 3, it is going through pulse
[01:43] <crimsun> TheMuso: for me, (1) results in nondeterministic clipping, (2) results in no clipping but wrong sampling rate, (3) results in expected (no clipping, correct sampling rate)
[01:43] <digistyl3> TheMuso: only case 3 isn't working for me
[01:43] <crimsun> wow, ok, so it's not /just/ alsa-lib
[01:43] <TheMuso> crimsun: oh right, yes the second gave me a sample rate warning
[01:43] <digistyl3> but the ouput of 1 and 2 are different
[01:43] <crimsun> digistyl3: (1) should be correct sampling rate, at least
[01:44] <digistyl3> crimsun: yes, (1) works correctly without errors
[01:44] <TheMuso> crimsun: yeah 1` is fine for me
[01:44] <TheMuso> crimsun: confirmed on two machines to be the same for #1
[01:44] <digistyl3> TheMuso: do you need the output of (3)?
[01:45] <TheMuso> digistyl3: it may be useful
[01:46] <crimsun> ok, away for dinner, but will be back in 10-15 minutes
[01:46] <digistyl3> TheMuso: here it is http://pastebin.com/d8cf747
[01:46] <ArneGoetje> Someone here familiar with libtool? libtoolize in intrepid seems to delete config.sub and config.guess from the working directory... a following ./configure can't find them anymore and the package FTBFS... libtoolize in hardy seems to leave them in place... regression?
[01:46] <TheMuso> digistyl3: thanks
[01:46] <digistyl3> np
[01:47] <TheMuso> Ok to deepen the mystery, setting to alsa in gnome-sound-properties fails to play a test sound, but totem works no problems.
[01:49] <bryce> slangasek: ok, fixed acpi-support is ready for upload
[01:50] <bryce> slangasek: I went through the dozen or so patches listed for it in launchpad, and found a couple that looked safe enough to include
[01:50] <slangasek> "looks safe, ship it"
[01:51] <slangasek> upload, please :)
[01:51] <bryce> okie doke
[01:51] <bryce> its up
[01:55] <digistyl3> bryce: can i test it? or will it be in the updates?
[01:58] <StevenK> slangasek: So, I'd like to upload a new xterm that contains a .desktop file for Ubuntu MID. Would you like to see a debdiff?
[01:59] <slangasek> StevenK: I'd like my eyes to uncross after having read the words "xterm" and ".desktop file" in the same sentence
[01:59] <bryce> digistyl3: it'll be in the updates once slangasek is done reviewing it and scolding me for including too many whitespace changes ;-)
[01:59] <slangasek> StevenK: are you serious?  A .desktop file for xterm?
[02:00] <bryce> slangasek: fwiw I've reviewed StevenK's change and it looks ok to me
[02:00] <calc> slangasek: it claims to be waiting for approval for lucene2
[02:01] <StevenK> slangasek: Okay, so we use xterm as the terminal application for Ubuntu MID. mobile-basic-flash included a .desktop file for it which is the wrong place, and it went away after we dropped m-b-f for kourou.
[02:01] <slangasek> calc: yep, I'll grab it in a bit, thanks
[02:01] <calc> slangasek: thanks :)
[02:02] <bryce> digistyl3: what sort of acpi stuff would you be interested in testing?
[02:02] <bryce> there's more changes in the debian acpi-support package, but that I wouldn't want to include without some additional testing/review, so if you're keen to test acpi-support, I could roll a package with those specific things in for you to play with
[02:03] <bryce> ooh, this fills one with confidence:
 ...
[02:03] <bryce>  hal is fucked.
[02:04] <digistyl3> bryce: i wanted to see if those 90-hdparm.sh files work, but i could help testing other things as long as they don't damage my new laptop :D
[02:05] <slangasek> bryce: this is... recent?
[02:05] <bryce> slangasek: just now on #xorg-devel ;-)
[02:05] <slangasek> awesome
[02:06] <bryce> digistyl3: ok no worries.  I think we aim to eliminate acpi-support anyway in jaunty.
[02:06] <digistyl3> bryce: why? is it old?
[02:06] <digistyl3> what would replace it?
[02:07] <bryce> digistyl3: yeah it's a ubuntu/debian specific thingee
[02:07] <bryce> digistyl3: well that's the big question.  We'll have to get that answered in jaunty, hopefully at the next UDS
[02:07] <bryce> probably some combination of pm-utils + hal + ??
[02:08] <bryce> I'm thinking that maybe we start by just focusing on sleep/hibernate functionality, get any relevant quirks migrated to pm-utils/hal, but leave the remainder of acpi-support in place until we find better homes for them
[02:09] <slangasek> sadly, I think this is the third release cycle now that I've heard that was supposed to be happening
[02:09] <bryce> mm
[02:09] <slangasek> i.e., sleep/hibernate was supposed to have been taken over long ago
[02:12] <bryce> slangasek: well if we can come up with a plan, I don't mind putting a little time into it for jaunty.
[02:13] <slangasek> yes; I understand this is an item on the UDS Jaunty agenda
[02:13] <bryce> I seem to get tasked with fixing sleep and hotkey problems now and then, and I would love to get things cleaned up enough that the responsibility gets more appropriately targeted elsewhere ;-)
[02:19] <slangasek> pitti: why does pykdeuic4 generate python files with a blank line ahead of the interpreter line? :)
[02:19] <slangasek> pitti: (kdeui/ui_ManagerWindowKDE4.py in jockey)
[02:21] <slangasek> james_w: bug #263888> <giggle>
[02:27] <StevenK> slangasek: I've just uploaded a new linux-lpia-meta, this will give lpia installable kernels via the meta-packages.
[02:29] <NCommander> hey StevenK
[02:29]  * StevenK waves
[02:30] <NCommander> How goes it StevenK
[02:31] <StevenK> NCommander: Trying to get stuff sorted out for beta
[02:31] <NCommander> StevenK, anything I can do to help?
[02:31] <StevenK> NCommander: Not really, you're not a core-dev
[02:31] <NCommander> :-/
[02:43] <TheMuso> crimsun: I'm playing with alsa-lib, and gone so far back as alsa-lib 1.0.17 with no patches from git, and still get the same issue, so its nothing that has been introduced in the last little while, unless there is something in the pulse plugin thats causing issues.
[02:45] <crimsun> TheMuso: are you using intrepid's pulseaudio?  I'm using your ppa's.
[02:46] <TheMuso> crimsun: yes I am using pulse 0.9.10.
[02:49] <TheMuso> crimsun: just tried alsa-plugins 1.0.17 without any of Lennart's fixes, and the gstreamer audio test in sound properties works again. Just going to try rhythmbox through pulse again, but it seems that one of Lennart's fixes breaks it somehow.
[02:50] <TheMuso> crimsun: ok rhythmbox is fine, so gstreamer is behaving now, at least for me.
[03:03] <TheMuso> git bisect time.
[03:16] <Hobbsee> TheMuso: ping?
[03:17] <Hobbsee> TheMuso: are my audio settings supposed to be set as alsa, now, and not pulse?
[03:21] <TheMuso> Hobbsee: No, leave as pulse.
[03:21] <TheMuso> Hobbsee: the issue that is being debugged is to do with the alsa pulse plugin, which happens to cause gstreamer apps to die. I'm looking into it. The gstreamer pulse plugin is fine though, but nevertheless this needs to be sorted.
[03:21] <Hobbsee> TheMuso: well, they were magically set as alsa yesterday.  Maybe the gnome update, i'm not usre.  Have you had similar issues reported?
[03:22] <TheMuso> Hobbsee: Yes, look at the backscroll for one, and a private message as well as a bug report comment, so I'm on it.
[03:22] <TheMuso> Hobbsee: as for changing to alsa thats weird.
[03:22] <TheMuso> it should be set to autodetect.
[03:22] <Hobbsee> TheMuso: right.
[03:22]  * Hobbsee hasn't checked backscroll adn such yet, sorry.
[03:23] <TheMuso> thats ok.
[03:53] <Hobbsee> ah, we are frozen now.
[03:53]  * beuno stays very still
[03:53] <Hobbsee> just when my thunderbird explodes.
[04:05] <persia> Hobbsee: freeze doesn't mean "Don't fix anything", it means "only fix *really* broken stuff, and test *lots*".
[04:05] <Hobbsee> persia: well, it means accepting uploads, too.
[04:05] <Hobbsee> persia: but that's true.
[04:05] <Hobbsee> persia: id' have been interesting in seeing what the freeze mail says, but it seems that'll take longer :)
[04:06] <persia> Hobbsee: https://lists.ubuntu.com/archives/ubuntu-devel-announce/2008-September/000491.html
[04:06] <Hobbsee> ah, rihgt, cool.
[04:07] <crimsun> TheMuso: I still can't reproduce that symptom
[04:07]  * Hobbsee was going with the lazyness option.
[04:07] <Hobbsee> (ie, not looking up webmail)
[04:07] <TheMuso> crimsun: Hrm.
[04:08] <TheMuso> crimsun: Bisecting Lennart's tree is giving me some unusual behavior, compared to the actual alsa-=plugins ackage in the archive as well. I think I need to start with a clean slate and go from there.
[04:09] <crimsun> TheMuso: I noticed yesterday that there seemed to be missing commits between intrepid's base alsa-plugins and what lennart's tree has
[04:09] <crimsun> I didn't chase it, because it was getting late
[04:12] <TheMuso> crimsun: I think he may have merged them in.
[04:12] <TheMuso> as of yesterrday.
[04:27] <TheMuso> crimsun: I think I have found the commit that breaks things. The reason why you can't reproduce is because you are using 0.9.12, which supports S32 sample types. 0.9.10 doesn't, and the latest code for the pulse plugin allows using S32 sample types. For some reason, this is breaking gstreamer's access to alsa via the pulse plugin.
[04:27] <TheMuso> So I've tested the reversion of that commit, and all seems to work again.
[04:34] <crimsun> TheMuso: I'm using 0.9.10 now, though.
[04:34] <TheMuso> crimsun: ah.
[04:34] <crimsun> however, I certainly agree with reverting that commit
[04:39] <TheMuso> crimsun: so you are able to get a test sound when attempting to test audio inside gnome-sound-properties when the setting is set to the alsa sink
[04:39] <TheMuso> ?
[04:40] <crimsun> let me test that.
[04:41] <TheMuso> crimsun: For me the problem manifests itself in different ways depending on the machine. On one machine with intel8x0, rhythmbox and the sound test in gnome-sound-properties both don't work. On a machine with hda-intel, only the sound test doesn't work.
[04:41] <crimsun> no, I get a GSt pipeline error on an HDA Conexant
[04:41] <TheMuso> In my above tests, with intel8x0, totem also works.
[04:41] <TheMuso> and with hda-intel
[04:50] <TheMuso> digistyl3: Do you happen to still be around?
[04:57] <TheMuso> crimsun: Going to get some further feedback from others who experienced this before settling for that as the problem, but I am confident the issue has been found.
[04:59] <wgrant> What's the issue being debugged here?
[04:59] <wgrant> I just downgraded to PA 0.9.10, and things hang...
[04:59] <TheMuso> wgrant: Basically when gstreamer apps tried to use pulseaudio via the alsa pulseaudio polugin, they would bomb with an error.
[04:59] <TheMuso> wgrant: try this package here: http://www.themuso.id.au/ubuntu/alsa-plugins
[05:00] <wgrant> (0.9.12 seemed to work fine apart from some SDL strangeness)
[05:00]  * wgrant tries.
[05:01] <crimsun> TheMuso: confirmed to fix the GSt pipeline error in sound properties testing when set to ALSA w/ pulse pcm+ctl plugin active
[05:04] <TheMuso> crimsun: Ok thats good. Does that solve any of those aplay test cases you mentioned earlier?
[05:07] <wgrant> TheMuso: gnome-sound-properties and Rhythmbox still hang, not sure about the others.
[05:07] <crimsun> TheMuso: no, but they shouldn't be considered regressions here, as they're fragment-specific for HDA
[05:07] <TheMuso> crimsun: Right.
[05:08] <TheMuso> wgrant: Did you log out/back in after you downgraded? I think you might still have to do that for some reason.
[05:08] <wgrant> TheMuso: I did, but that didn't work so I rebooted.
[05:08] <TheMuso> Hrm. What do you have set as your sink in gnome-sound-properties?
[05:10] <wgrant> TheMuso: When testing in g-s-p: OSS works, ALSA makes a click and then aborts, PulseAudio looks like it's working but doesn't play anything, and hangs when you click OK.
[05:10] <wgrant> Hm, bad stuff in syslog.
[05:11] <wgrant> Sep 26 14:10:05 irranat pulseaudio[6655]: shm.c: Invalid shared memory segment size
[05:11] <wgrant> Sep 26 14:10:05 irranat pulseaudio[6655]: pstream.c: Failed to import memory block.
[05:11] <TheMuso> hrm.
[05:11] <TheMuso> wgrant: Try killing pulse, clearing any .pulse* files from your home dir, and restarting pulse. Easiest way to restart pulse is "pulseaudio -D --log-target=syslog"
[05:12] <TheMuso> There are differences between 0.9.11/12 and 0.9.10 in terms of what gets put in the home dir.
[05:13] <wgrant> TheMuso: Same thing...
[05:13] <TheMuso> crimsun: any ideas?
[05:13] <TheMuso> because I'm fresh out currently.
[05:15] <crimsun> wgrant: with 0.9.10-2ubuntu6?
[05:15] <crimsun> of pulseaudio* and libpulse*, sorry
[05:16] <wgrant> crimsun: Looks that way.
[05:16] <wgrant> Hmm. paplay works.
[05:17]  * wgrant restarts pulseaudio a few more times for good measure.
[05:17] <crimsun> in g-s-p, what behaviour when you set playback to Autodetect?
[05:18] <wgrant> It... works.
[05:18] <wgrant> Hrnnnng. ALSA and Pulse work directly too.
[05:18] <TheMuso> The only thing I can think of is that a pulse package is out of version sync, maybe one of the modules, or one of the libs.
[05:18] <wgrant> So something has unbroken itself.
[05:19] <wgrant> Maybe there was some lib loaded that still knew about the 0.9.12 ~/.pulse* stuff.
[05:20] <digistyl3> TheMuso: yes i am
[05:21] <TheMuso> digistyl3: Could you please try http://www.themuso.id.au/ubuntu/alsa-plugins again and let me know if mplayer and gstreamer stuff works again?
[05:21] <digistyl3> sure
[05:23] <digistyl3> TheMuso: do i need a reboot? because it isn't working yet
[05:24] <crimsun> you shouldn't need to reboot; closing all audio apps should suffice.  However, rebooting tends to be a "safer bet".
[05:24] <digistyl3> ok, i'll brb
[05:25]  * Hobbsee declares victory over evil email.
[05:27] <Hobbsee> bah.  or not
[05:28] <TheMuso> heh
[05:29] <digistyl3> TheMuso: it isn't workinh
[05:29] <TheMuso> digistyl3: Are you trying mplayer, or a gstreamer app? What do you have set as your output in gnome-sound-properties?
[05:29] <digistyl3> TheMuso: audiotestsrc wave=sine freq=512 ! audioconvert ! audioresample ! gconfaudiosink: Failed to connect stream: Invalid argument
[05:29] <digistyl3> mplayer works as it falls back to oss
[05:30] <digistyl3> totem doesn't
[05:30] <TheMuso> digistyl3: Did you download the package for your arch from that URL?
[05:30] <digistyl3> TheMuso: yes i did
[05:30] <digistyl3> the one last modified on: 26-Sep-2008 11:52
[05:32] <TheMuso> digistyl3: And that is with alsa as the output?
[05:33] <digistyl3> TheMuso: yes, i get the same errors
[05:33] <TheMuso> digistyl3: What about set to pulse?
[05:33] <digistyl3> the same
[05:34] <TheMuso> ok.
[05:34] <digistyl3> totem sais:
[05:34] <digistyl3> ** Message: Error: Failed to connect stream: Invalid argument
[05:34] <digistyl3> pulsesink.c(487): gst_pulsesink_prepare (): /play/visbin/abin/audiosinkbin/audio-sink/bin6/autoaudiosink1/autoaudiosink1-actual-sink-pulse
[05:34] <digistyl3> isn't this related to the pulseaudio plugin?
[05:34] <TheMuso> yse it is so its something else.
[05:34] <TheMuso> digistyl3: go to a terminal, and kill pulseaudio, "pulseaudio -i"
[05:34] <TheMuso> them delete .pulse* from your home directory.
[05:34] <TheMuso> re-run pulseaudio with "pulseaudio -D --log-target=syslog"
[05:35] <TheMuso> then try again.
[05:35] <digistyl3> TheMuso: pulseaudio: invalid option -- 'i'
[05:35] <TheMuso> sorry meant -k
[05:37] <digistyl3> TheMuso: it's working now :)
[05:37] <TheMuso> Great.
[05:37] <TheMuso> Tr with alsa in gnome-sound-properties, just to be sure.
[05:37] <TheMuso> try
[05:37] <TheMuso> goodness my typing is bad this afternoon.
[05:38] <digistyl3> yep, it's working alright :)
[05:38] <TheMuso> Ok great, thanks for your help. I'll get that fix into intrepid as soon as I have a bit more feedback.
[05:38] <TheMuso> digistyl3: Oh if you could pastebin mplayer output again, that would be appreciated, just want to make sure it works as it should.
[05:38] <digistyl3> i'm glad i could help :)
[05:39] <digistyl3> TheMuso: the output with the fixed version installed?
[05:40] <TheMuso> digistyl3: Yes please.
[05:41] <digistyl3> TheMuso: here it is http://pastebin.com/d38e61579
[05:41] <TheMuso> digistyl3: Great, thats all. Thanks again.
[05:42] <digistyl3> np :)
[05:43]  * digistyl3 back to watching movies :)
[05:46] <dholbach> good morning
[06:04] <elmo> bryce/cjwatson: for the record, that's not what I asked.  I asked that you not give automated processes full access to your canonical accounts.  I suggested several alternatives; there was no need for you to migrate scripts off, it was your choice
[06:05] <dholbach> slangasek (or any other release team member): I uploaded gnome-system-tools (new upstream release, new translations, added a patch we carried in our old version before) - can you accept it?
[06:18] <digistyl3> dholbach: does it contain a fix for remembering the current session in gnome?
[06:18] <dholbach> digistyl3: not AFAIK
[06:19] <bryce> elmo, sorry I didn't mean to misrepresent what you'd told me...
[06:21] <bryce> elmo: but the alternative you suggested was to not use scp, but rather to fetch from rookery via http
[06:23] <bryce> so I took that to mean that you do not want automated scripts scp'ing stuff to people.ubuntu.com
[06:28] <bryce> elmo: also, fetching stuff via http implies I have a webserver up on my end with the files already hosted there, so it ends up being simpler to just move the whole kit there
[06:29] <bryce> elmo: if you offered other alternatives, sorry, I must have missed them
[07:05] <slangasek> Riddell: hmm, your upload of kdebase-workspace is rather opaque to review; no bug referenced in the changelog, and the only diff is to disable a patch, but it's not obvious what that patch does :)
[07:07] <slangasek> bryce: scp isn't really possible to restrict, IIRC; but there are ways to push files up not using scp, e.g., streaming tar + a helper script on the remote side + a restricted ssh key
[07:09] <slangasek> dholbach: will get it, but it's first-come, first-served :)
[07:09] <dholbach> no problem
[07:09] <dholbach> I'm going to be around a little bit longer ;-)
[07:11] <bryce> slangasek: streaming in what way?
[07:11] <slangasek> bryce: tar zcvf - | ssh host [...]
[07:11] <slangasek> well, probably without the v, sorry for my reflex :)
[07:13] <bryce> hmm, I dunno, it sounds a bit hackish
[07:17] <bryce> slangasek: however thanks, I didn't realize you could copy files across like that
[07:17]  * StevenK throws more packages at unapproved
[07:18] <slangasek> it's not as simple as just doing scp, no, but it's e.g., what I use to avoid giving unsupervised scripts access to Debian servers for doing debian-installe daily builds
[07:31] <siretart> morning
[07:33] <siretart> pitti: tjaalton: I remember you've been talking about hotkey cleanups the last days. since yesterday some hotkeys on my X60s have stopped working. are you aware of a bug or issue that might cause this?
[07:46] <sbeattie> slangasek: is there some problem with the live cd builds? The newest on cdimages is 3 days old/
[07:49]  * persia broke it.  It was fixed earlier today
[07:50] <persia> (or last night, depending on your current longitude)
[07:53] <tjaalton> siretart: only thing that might have something to do with it is the new acpi-support
[07:53] <tjaalton> siretart: what laptop?
[07:54] <siretart> tjaalton: thinkpad X60s
[07:56] <bryce> tjaalton: any reason other than the timing of the update?
[07:59] <tjaalton> bryce: right, it's the only component that has been updated recently
[08:00] <tjaalton> but i'm still sceptical about how much a-s matters on thinkpads
[08:01] <bryce> siretart: well you can try downgrading to 0.110 or earlier and see if it makes any difference
[08:01] <tjaalton> bbiab ->
[08:07] <pitti> Good morning
[08:07] <pitti> kwwii: thanks
[08:08] <pitti> slangasek: consolekit crasher bug? any #?
[08:09] <pitti> Jonbo: sure, if you have one
[08:09] <slangasek> pitti: bug #263245
[08:10] <pitti> slangasek: pykdeuic4> no idea TBH; I mean, for those .py files a shebang line doesn't even make sense...
[08:10] <pitti> siretart: I think the cause might be the recent acpi-support merge
[08:11] <pitti> siretart: for reference maybe you could downgrade this again and test again?
[08:12] <bryce> pitti, any particular reason you think it might be acip-support, aside from the timing of the upload?
[08:12] <pitti> siretart: there is another proposed fix in bug 267682 (the .fdi file)
[08:12] <pitti> bryce: no, just by the time of the upload; I don't think that tjaalton uploaded the -evdev fdi file yet? at least not yesterday?
[08:13] <pitti> slangasek: haven't seen it so far, but I'll take this one; CK crashes are eww
[08:15] <sbeh> hi, "localedef --no-archive --magic=20051014 -i en_US -c -f UTF-8  en_US.UTF-8" hangs at 100% cpu reproducable and "strace -p `pidof  localedef`" gives no output, what can i do to determine whats  causing this problem?
[08:18] <mdke> dholbach: thanks for the gnome-user-docs upload yesterday, I've merged your branch. Could you perhaps do another one at some stage? I've merged from upstream this morning: some new material and translations
[08:19] <tjaalton> pitti: what about adding/merging the fdi file to hal-info?
[08:20] <dholbach> mdke: sure
[08:22] <pitti> tjaalton: we can do that as well, but if it's a workaround and not something to be committed upstream, and the knowledge about changing it is mostly in -evdev, why not ship it in -evdev?
[08:22] <dholbach> hi mvo, hi pitti!
[08:22] <mdke> dholbach: thanks!
[08:22] <tjaalton> pitti: evdev has no fdi file
[08:22] <tjaalton> pitti: the magic is in hal
[08:22] <pitti> tjaalton: if you don't want it there, I'd rather put it into hal; hal-info is the wrong place fo rit
[08:23] <tjaalton> pitti: ok, hal it is then :)
[08:23] <siretart> pitti: I have a meeting now, can only test this afternoon. besides, is there an deb version 110 somewhere available or shall I rebuild the package myself?
[08:23] <tjaalton> siretart: I have an X61 and things work like before even with all the updates
[08:24] <siretart> tjaalton: on my x60s, xev doesn't show my anything when I press fn-f4 e.g.
[08:24] <pitti> tjaalton: "has no fdi file"> yes, it could just ship it; but as I said, if you don't like that for some reason, we can put it into hal, too
[08:24] <tjaalton> siretart: that's normal
[08:24] <pitti> tjaalton: my concern about this is that it would break backportability of the hal package
[08:24] <siretart> well, gnome hotkey settings dialog doesn't detect the keystroke either
[08:24] <mvo> hey dholbach!
[08:24] <tjaalton> pitti: oh, ok
[08:24] <pitti> tjaalton: (same reason for hal-info, which is why I really don't like it there)
[08:25] <tjaalton> siretart: same thing, it shouldn't (and never has) :)
[08:25] <mdke> hi mvo: I sent you an email a while about whether there were any disadvantage to using apturl instructions in the Ubuntu help wiki, have you had a chance to consider it?
[08:25] <pitti> siretart: 0.109 is in hardy, that should work well enough, too
[08:26] <pitti> siretart: otherwise 0.110 should be a quick build (but it only fixes laptop-mode, no keys)
[08:27] <siretart> kk
[08:27] <mvo> mdke: oh, right. sorry that I have not ansered yet. the only disadvantage that I can see is a the "do not teach people to install software from the web" argument. but apturl itself is fine (as it just uses the stuff from the repository)
[08:29] <mdke> mvo: is it possible to use it to give an instruction to install a non-repository package (like a ppa, or external repository)?
[08:30] <mvo> mdke: not eaisly, the feature to add repos is disabled because it seems to risky
[08:30] <mvo> mdke: so the user needs to add the repository first in some way, than apturl will work with the packages in that repo too
[08:30] <mdke> mvo: ok, good. So it seems pretty safe to use it on a user-editable wiki
[08:30] <mvo> yes
[08:30] <mdke> I'd really like to use it in yelp too, I have to see if that works
[08:32] <mdke> mvo: ok, thanks for your help
[08:32] <mvo> I added a gnome url handler for it, let me know if there are issues, then I have a second look
[08:32] <mvo> thank you!
[08:32] <mdke> ah, cool
[08:40] <StevenK> pitti, slangasek: Could either one of you be tempted to prod through my uploads? :-)
[08:46] <pitti> StevenK: slangasek will regularly do that, or ask me for assistance; don't worry, they won't go ignored :)
[08:47] <StevenK> pitti: I could probably do it, but I didn't want to get my hand slapped :-)
[08:52] <dholbach> mdke: uploading, it will sit in the queue for a bit
[09:07] <mvo> Riddell: thanks for your test with kubuntu-kde4-desktop, I look into it
[09:11] <amitk> asac: do you already know about the erratic scrolling behaviour in firefox?
[09:11] <asac> erratic?
[09:11] <asac> amitk: ?
[09:12]  * directhex finds it ironic that he experienced erratic scrolling behaviour in xchat when reading the last 3 lines
[09:13] <asac> amitk: ah yeah. thats cairo but also boils down - at least a good part of it - to the driver situation we have
[09:13] <PecisDarbs> I am sorry to say that, but NM is still terribly broken when doing manual configuration of *anything*
[09:14] <amitk> asac: driver?
[09:14] <asac> PecisDarbs: what exactly do you mean by manual configuration? static ips?
[09:14] <mvo> Riddell: do we have a open bug about the kde->kde4 upgrade behavior?
[09:14] <asac> amitk: i think X driver is the better description
[09:14] <PecisDarbs> asac: yep
[09:16] <asac> amitk: at least there appear to be quite a substantial performance difference depending on what kind of driver and accellmethod you are using
[09:16] <asac> PecisDarbs: i never had problems witht hat
[09:16] <PecisDarbs> asac: https://bugs.edge.launchpad.net/ubuntu/+source/network-manager/+bug/256054
[09:17] <PecisDarbs> when I do manual configuration via NM, it doesn't activate interface
[09:17] <asac> PecisDarbs: what has that to do with that bug?
[09:18] <PecisDarbs> ok, maybe not directly
[09:18] <asac> (sorry i should get a copy ;))
[09:18] <asac> err coffee
[09:18] <PecisDarbs> :)
[09:18] <amitk> asac: PgDn scrolling not working is also part of this problem?
[09:18] <asac> amitk: not working completely?
[09:18] <PecisDarbs> asac: problem is that leaving NM as only opinion to configure net on Ibex is huge regression so far. I hope situation will improve.
[09:19] <asac> amitk: or slow?
[09:19] <mvo> tkamppeter: I just got this here in a hardy chroot: http://paste.ubuntu.com/50774/ <- (line 6 is the ps output that seem to have triggered it). why is it doing that?
[09:19] <StevenK> Mithrandir: Launchpad keeps bugging me about my memership in bluetooth and telepathy expiring, can you fix it or make me an admin?
[09:19] <StevenK> s/memership/membership/
[09:19] <asac> PecisDarbs: please describe me the exact steps you take
[09:20] <asac> PecisDarbs: most likely you right click on applet and select "connection editor ..." ... what next?
[09:20] <PecisDarbs> asac: 1) boot Ibex (fresh install, sunday/monday) 2) login 3) click on nm-applet icon with right button 4) Edit Connections...
[09:20] <amitk> asac: it works for 1 page down and then doesn't work at all. Sometimes it goes back and forth on current and next page producing a cool effect but no scrolling :-/
[09:21] <asac> amitk: that started with latest update?
[09:21] <Mithrandir> StevenK: I'll fix it.
[09:21] <PecisDarbs> asac: 5) Wired tab 6) Auto eth0, Edit 7) put in manual mode, provide IP and press OK 8) open Terminal and ifconfig - nothing has changed
[09:21] <asac> PecisDarbs: yes. dont do that ;) ...
[09:21] <asac> PecisDarbs: create a "new" connection for now
[09:21] <PecisDarbs> asac: wtf?
[09:21] <asac> PecisDarbs: editing auto eth0 isnt workding
[09:22] <PecisDarbs> why?
[09:22] <amitk> asac: No, with 3.0.2+build6+nobinonly-0ubuntu2. I have to upgrade to 3.0.3 now.
[09:23] <asac> amitk: ok. but it started with 3.0.1?
[09:23] <PecisDarbs> NM configuration is very lame so far, and is step backwards from network-admin. I tried to love it. We need new network-admin, for sure, but NM so far is no cut.
[09:23] <asac> err .2
[09:23] <tkamppeter> mvo, this is really strange. A package install process always runs as root, so lpadmin always should be able to create print queues without asking for a password. It even recognizes that it runs as root, as it asks for the root password.
[09:23] <asac> PecisDarbs: i didnt hear productive suggestions from you so far. only that nothing works
[09:23] <amitk> asac: yes
[09:23] <asac> PecisDarbs: try how it works with "new" connection
[09:24] <tkamppeter> mvo, pitti, what is the state of cups-pdf, did we not drop it from the seeds?
[09:24] <pitti> tkamppeter: we did, it's in universe now
[09:24] <mvo> tkamppeter: it happens in a hardy chroot, I was just pretty puzzled when I saw the prompt
[09:24] <mvo> (with kubuntu-kde4)
[09:24] <asac> PecisDarbs: and comment based on that experience. there are still bugs, yes. but most should be sorted for final
[09:25] <PecisDarbs> asac: productive suggestion would be to put network-admin back in default, but you guys already have decided on that. No offense, but this really have ticked me off a little bit off.
[09:26] <tkamppeter> mvo, the lpadmin in post-install worked for me all the time without problems (but I have tested only on the GNOMEy standard edition)
[09:26] <Gast_314_> http://www.meine-privaten-nacktvideos.net?id=4812552
[09:26] <asac> PecisDarbs: network-admin never worked together with NM ... it just pretended that it did
[09:26] <PecisDarbs> asac: thanks for a tip
[09:26] <PecisDarbs> asac: and? that is not a reason to drop one app or another.
[09:26] <asac> PecisDarbs: those applciations conflict.
[09:26] <PecisDarbs> asac: NM configuration is a very different is very anti-GNOME-ish
[09:27] <tkamppeter> Who does not understand german, the URL posted by the guest is spam
[09:27] <PecisDarbs> asac: yes, then someone should make them work together :)
[09:27] <PecisDarbs> ok
[09:27] <asac> PecisDarbs: no. NM should be improved
[09:27] <mvo> tkamppeter: ok, might be some fluke on my local system or something
[09:27] <asac> PecisDarbs: but btw, in some way we are working on making that happen (by implementing the ifupdown plugin)
[09:28] <Mithrandir> StevenK: you're an admin on both teams already AFAICT?
[09:28] <asac> PecisDarbs: but its not the primary motivation for that to get network-admin back
[09:29] <PecisDarbs> asac: another reason why I want network-admin because it was so much easier to use. Simple as that.
[09:29] <PecisDarbs> ok
[09:29] <StevenK> Mithrandir: Then why is LP telling me to bug you? It's odd ...
[09:29] <asac> PecisDarbs: i would appreciate if you could give detailed info on what aspects dont work as you want
[09:29] <asac> PecisDarbs: easier to use -> make suggestions how to improve that in NM
[09:30] <PecisDarbs> asac: there is no profiles
[09:30] <Mithrandir> StevenK: no idea.  Can you edit yourself in the list?  I
[09:30] <Mithrandir> 'm happy to extend your membership, obviously
[09:30] <asac> PecisDarbs: the idea is that you dont need profiles anymore
[09:30] <StevenK> Mithrandir: I'll fiddle in a sec
[09:31] <PecisDarbs> asac: really, what do you use then? :)
[09:31] <asac> PecisDarbs: we probably cannot introduce profiles, but we could look into what exact use-cases you previously covered with profiles and see if we can make those coevered
[09:31]  * PecisDarbs sights
[09:31] <asac> PecisDarbs: what use-cases did you covere with profiles?
[09:31] <PecisDarbs> asac: it is reason why people hate changing functionality - if it works, don't fix!
[09:32] <PecisDarbs> asac: laptop, changing networks three times a day
[09:32] <PecisDarbs> everytime totally different configuration
[09:32] <PecisDarbs> network-admin did it with ease
[09:32] <PecisDarbs> three clicks and I was set
[09:32] <asac> PecisDarbs: in which sense?
[09:32] <asac> PecisDarbs: what is network config 1 and what is network config 2?
[09:33] <asac> give me an example ;)
[09:33] <PecisDarbs> ok
[09:33] <PecisDarbs> dhcp and manual with different DNS settings
[09:34] <asac> PecisDarbs: that should be at least as simple as with network-admin. you create two connections: one with dhcp the other with manual
[09:34] <asac> when you are at place with dhcp you click on that connection in the applet ... otherwise on the other place
[09:34] <PecisDarbs> asac: and then?
[09:35] <asac> there is even some magic going on if you have both set to "connect automatically"
[09:36] <asac> PecisDarbs: how is that worse than network-admin?
[09:37] <PecisDarbs> asac: ok, I will evaluate that. This is very different, I must say, but ok, maybe I am too harsh without knowing it better
[09:37] <mvo> Riddell: update-manager (in bzr) should be fixed for kubuntu-kde4-desktop
[09:38] <asac> PecisDarbs: well. i think with some luck and the magic you might not even need to click anything anymore.
[09:38] <asac> PecisDarbs: so in best case everything will happen automatically when you plugin your system at different places. in worst case you have to do one click
[09:39] <PecisDarbs> ok :)
[09:39] <PecisDarbs> asac: anyway, this change from network-admin to NM should be noted, because user will expect old app to be in tradional place.
[09:40] <asac> PecisDarbs: we could put "connection editor" in the system -> administration menu
[09:40] <PecisDarbs> asac: that would do the trick
[09:40] <asac> would that help?
[09:40] <asac> ok. noted.
[09:41] <PecisDarbs> asac: definitely
[09:42] <asac> amitk: sorry. i forgot about you ;) ... do you have any extensions installed?
[09:46] <PecisDarbs> asac: thanks for listening, I will give NM another shot, maybe it is really good as you tell. Maybe there only should be way for users to avoid editing "Auto eth0" profile as I did. And is there any way to configure simple modem or bluetooth modem on NM? I couldn't find it.
[09:46] <amitk> asac: np. Yes I have extensions - Adblock plus, Google notebook, firegpg, ubuntu firefox modifications
[10:01] <asac> amitk: please try to disable a bunch and see if things improve
[10:02] <asac> PecisDarbs: yeah. nm-applet didnt have any UI mean to present read-only connections. this has been fixed on trunk from what i can tell
[10:02] <PecisDarbs> ok, it is being worked on. Nice.
[10:05] <PecisDarbs> asac: about modem connections?
[10:09] <tjaalton> pitti: I've got a new xorg-server waiting which fixes bug 261977
[10:10] <asac> PecisDarbs: you are actually the first asking that.
[10:10] <asac> PecisDarbs: which is strange when thinking about it ;)
[10:11] <asac> PecisDarbs: NM supports Mobile Broadband modems and PPPOE/DSL
[10:11] <directhex> will intrepid's kernel contain any of the latest git stuff to support the latest intel wifi cards?
[10:12] <asac> PecisDarbs: i cannot tell out of the head why modems dont work. but there probably is a good (even though technical) reason
[10:12] <asac> PecisDarbs: i will find out
[10:17] <PecisDarbs> asac: thanks in advance
[10:21]  * mvo is supprised that python3.0 is priority required
[10:21] <mvo> well, python3.0-minimal
[10:22] <lool> Sounds a bit early for that
[10:22] <lool> mvo: Do you know what the python3 meta is for?
[10:22] <mvo> no
[10:22] <lool> It's to point at the latest python3.x when python it's a python3?
[10:23] <mvo> update-manager will pull in python3.0-minimal on upgrades because of the priority
[10:42] <DktrKranz> pitti, when you have time to put your ubuntu-sru hat on, mind reviewing bug 207406? It causes a couple of segfaults in xvidcap in Hardy, I tested it and it no longer fails with that patch applied.
[10:45] <cjwatson> mvo: I'll fix that
[10:46] <mvo> thanks Chipzz
[10:46] <mvo> thanks cjwatson
[10:46] <cjwatson> done for next publisher run
[10:46] <Chipzz> mvo: typo? :)
[10:47] <mvo> Chipzz: yes :) tab-completion to quick
[10:47] <Chipzz> hehe :)
[10:47] <Chipzz> happens :)
[10:53] <cjwatson> persia: I don't think you broke the live CD builds - that's a separate problem
[10:54] <persia> cjwatson: Oh.  I thought me breaking germinate was what broke the live CD builds.  Thanks for easing my guilt.
[10:54] <ogra> eeek, whats up with vitrualbox ?
[10:55] <ogra> it asks me for registration ... and now hangs
[10:55] <ogra> oh, doesnt hang
[10:55] <ogra> fun
[10:55] <ogra> a popunder behind the registration window
[11:03] <seb128> slangasek: I've uploaded a bug-buddy update which make bug-buddy not run by default so apport can catch the gtk applications crashes again if you want to accept the upload (not sure if you look through the queue or want to be pinged on IRC)
[11:33] <shingara> Hi
[11:34] <shingara> I search a guide where the policy of package nomming is write ?
[11:34] <mpt> nomming?
[11:34]  * wgrant takes a bite out of mpt.
[11:35] <mpt> nom nom nom
[11:35] <mpt> oh, naming?
[11:35] <wgrant> Ahh, maybe.
[11:35] <shingara> namming
[11:35] <shingara> sorry for my english
[11:35] <wgrant> It really depends on the type of package.
[11:36] <shingara> yes, but there are a policy
[11:36] <shingara> a guide about that ?
[11:37]  * mpt finds http://oreilly.com/catalog/debian/chapter/book/appc_01.html#DAPPC-PGFID-945614
[11:37] <shingara> because I want separate the package name, original version name and the ubuntu specific name
[11:37] <wgrant> Oh.
[11:38] <wgrant> Not the kind of naming that I was thinking of, but mpt has it right.
[11:38] <shingara> yes mpt but I have example like rdesktop : rdesktop-1.5.0-3+cvs20071006ubuntu0.1
[11:38] <wgrant> That's a very strange (and likely incorrect) version string.
[11:39] <shingara> yes but exist
[11:39] <wgrant> rdesktop_1.5.0+cvs20071006-3ubuntu0.1, perhaps.
[11:39] <wgrant> Where?
[11:39] <wgrant> Erm.
[11:39] <wgrant> Indeed.
[11:39] <wgrant> Ummm.
[11:39] <shingara> USN 646-1
[11:39] <shingara> Ubuntu 8.04
[11:39] <wgrant> Right, I see it.
[11:39] <wgrant> Now to work out how such an abomination came about.
[11:40] <shingara> So I suppose that all origin version and patchlevel ubuntu are separate by -
[11:40] <shingara> but I don't know if there are - in patchlevel too :(
[11:41] <shingara> /^.+-(.+)-(.+)/
[11:41] <shingara> send me $1 original version and $2 patchlevel specific unbutu ?
[11:41] <wgrant> Versions generally only have one -
[11:42] <shingara> /^.+-(.+)-(.+)$/
[11:42] <wgrant> Anything before the final - is the original upstream version number.
[11:42] <wgrant> Anything after is the Debian/Ubuntu bit.
[11:42] <shingara> ok, thanks :)
[11:44] <Riddell> mvo: dist upgrade tool seems to do the job now for kubuntu-kde4-desktop
[11:44] <Riddell> mvo: are you going to upload?
[11:47] <mvo> Riddell: I'm in the middle of adding the fglrx detection/removal code currently, but after that I will
[11:47] <Riddell> groovy
[11:48] <mvo> Riddell: if you want it in asap, then just upload the current version with my fixes, but I should get a new version uploaded before the weekend
[11:49] <asac> hmmm ... anyone has a interfaces configuration with _mutliple_ _wep_ keys?
[11:49] <Riddell> that's fine
[11:50] <directhex> asac, can you do that?
[11:51] <asac> directhex: i think you can do everything that iwconfig key ... offers: http://paste.ubuntu.com/50819/
[11:51] <asac> wonder if a) thats really the case and b) if anyone uses such kind of setup
[11:54]  * directhex deploys double-rot13 encryption
[11:57]  * Spads deploys quadruple-rot13 encryption
[11:58] <directhex> damn, beaten by Spads
[11:59] <shingara> I found another exception :(
[12:00] <shingara> ia32-sun-java5-bin-1.5.0-15-0ubuntu1
[12:02] <shingara> there are no really convention :(
[12:06] <persia> shingara: the exact regex used can be found in the dpkg source.  Generally it's of the form (.*)-([^-]*)
[12:07] <cjwatson> shingara: I'm struggling to work out what your point is, or what you're actually trying to find out
[12:08] <cjwatson> shingara: version numbers are to some extent arbitrary, but there *are* rules that govern their use
[12:08] <cjwatson> however you aren't going to work them out just by staring at several examples
[12:08] <directhex> fail. ia32-sun-java5-bin_1.5.0-15-0ubuntu1 - note the all-important underscore
[12:09] <directhex> package name ia32-sun-java5-bin, version 1.5.0-15, debian version 0 (i.e. non-existent), ubuntu version 1
[12:09] <directhex> blame sun for the unhelpful version number for update15
[12:40] <persia> A question about seeding: in ubuntu-mid there's a recommends chain that is unwanted, but it is sensible to keep the package recommends (in this case, ubuntu-mid is the exceptional case mentioned in policy).  Would adding a blacklist for the recommended package starting the chain be the correct way to resolve this?
[12:42] <cjwatson> never use blacklists, to a first approximation
[12:42] <cjwatson> I have only ever seen one legitimate use
[12:42] <cjwatson> persia: what's the package in question?
[12:43] <persia> OK.  Here's the specifics.  ubuntu-mid wants to use thunar.  thunar recommends xfce4-panel.  The thunar trash applet needs xfce4-panel, but isn't displayed in the hildon environment.
[12:43] <persia> So, while thunar *should* recommend xfce4-panel (until some future time when the package is split), ubuntu-mid doesn't want xfce4-panel.
[12:43] <cjwatson> we really don't have a particularly good way to express that
[12:44] <cjwatson> a blacklist isn't *correct*, although in this case it wouldn't be too harmful
[12:44] <persia> That's what I thought from the documentation.
[12:44] <cjwatson> you see, the problem with blacklists is that if something depends on that package in the future, then germinate will just leave it uninstallable rather than doing anything sensible
[12:45] <cjwatson> the correct solution here sounds like splitting the package, but go ahead and use a blacklist if that isn't an option
[12:45] <persia> Ah, right, because it generates an unresolved conflict.
[12:45] <cjwatson> and add a comment about why it's there
[12:45] <persia> At this point in the cycle, I'm not tempted to ask Xubuntu to split such a critical package, but it may be possible for jaunty.
[12:46] <persia> cjwatson: Thank you.
[12:46] <pitti> tjaalton: ah, I'll have a look at the queue again soon
[12:47] <pitti> DktrKranz: looks good to me
[12:54] <pitti> mvo: did you see the verification-failed in bug 220890?
[12:55] <mvo> pitti: yes, I haven't looked deeper into it though (no time yet)
[12:55] <mvo> pitti: do you happen to know if we have a porters team or something that I can get quick feedback from for patches?
[12:55] <pitti> mvo: ok, I'm just not sure whether it is a regression which should block -updates propagation, or just "not fixed yet" and thus just keeping open
[12:56] <pitti> mvo: lamont perhaps?
[12:59] <mvo> pitti: the upload was mainly about the powerpc regression (but it fixes another crash too) so I would say keep out of -updates for now and get a new version for -proposed
[13:07] <pitti> bryce, tjaalton: argh, it did it again; since Tuesday or so I suddently get a completely white or black screen (X and apps keep running, though), which can only be fixed with a reboot; happens about three times a day; I never had that before, did you get similar reports about that?
[13:07] <tjaalton> pitti: no.. what hardware?
[13:07] <pitti> bryce, tjaalton: Xorg.0.log has tons of "pipe A underflow" messages, could that be relevant? I didn't see anything other log-ish
[13:08] <pitti> tjaalton: Intel GMA945
[13:08] <tjaalton> pitti: and that's with compiz?
[13:08] <pitti> tjaalton: yes, it is
[13:08] <pitti> but I have run compiz for some weeks again now
[13:08] <tjaalton> sounds mesa-ish
[13:09] <tjaalton> damn stable releases ;)
[13:09] <Treenaks> tjaalton: At least it's not logging a DDC probe every few minutes, like my radeon driver :)
[13:09] <pitti> tjaalton: hm, last mesa was uploaded Wednesday evening; the problems started before that
[13:09] <tjaalton> Treenaks: that shouldn't happen anymore with a recent libgtk ;)
[13:10] <tjaalton> pitti: oh, phew
[13:10] <Treenaks> tjaalton: ah, so that's the problem :)
[13:11] <pitti> tjaalton: latest -intel changelog looks unrelated to my hardware, too, but I could try downgrading
[13:12] <tjaalton> pitti: nah, that can't be it.. the newer compiz maybe?
[13:12] <pitti> tjaalton: well, hm, maaaybe it was Wednesday, too, I'm not really sure TBH
[13:13] <pitti> tjaalton: last upload was last week, I extensively used that box with that version
[13:13] <pitti> tjaalton: ^ compiz
[13:13] <pitti> tjaalton: I think I'll try to downgrade mesa and test
[13:15] <tjaalton> pitti: maybe a good idea
[13:15] <tjaalton> lunch->
[13:16] <pitti> tjaalton: do you know if the pipe-A log could be related?
[13:16] <pitti> tjaalton: ah, enjoy
[13:25] <tjaalton> damn Friday.. forgot that restaurants close at 1500
[13:26] <tjaalton> pitti: it might. if you can monitor the log remotely (via ssh) and check if the messages correlate with the crash
[13:27] <pitti> tjaalton: VT1 still works, so I can look at it
[13:27] <pitti> tjaalton: I have it tail -f'ed right now
[13:27] <pitti> tjaalton: wow, they *close* at 15:00? I mean, if people go out to restaurants, then certainly on Fri/Sat evenings?
[13:28] <Koon> pitti: it's evening at 15:00 in Finland in winter :)
[13:29] <liw> I suspect tjaalton is talking about lunch restaurants, 11-14 is the official lunchtime in Finland
[13:29] <pitti> tjaalton: is there a way to enable timestamps in Xorg.0.log, btw?
[13:30] <tjaalton> pitti: well, "restaurant" was maybe an exxageration, "canteen" might be more accurate :)
[13:31] <tjaalton> s/exxa/exag/
[13:32] <tjaalton> liw: right, Alvari ftl
[13:32] <mvo> Riddell: I think the current bzr is good for upload of update-manager, if its not too much hazzle for you, I would appricate if you could give it a test-run. I'm running tests here as well and if all is well I upload in a bit
[13:32] <liw> tjaalton, indeed; I don't have fond memories of it from my time in Otaniemi :)
[13:32] <tjaalton> pitti: ok, so it doesn't really crash? you can shut down X from the VT?
[13:33] <pitti> tjaalton: yes, apps keep running, X keeps running; the screen just goes black (I still see the cursor) or completely white (nothing visible any more)
[13:33] <tjaalton> liw: heh, well it has become quite decent actually, but can't understand why they (too) close so early. mon-thu they are open ->1700
[13:33] <pitti> tjaalton: it's not the screensaver, I already killed that; and killing everything and restarting X doesn't help, I need to reboot
[13:33] <pitti> tjaalton: it seems like it puts and leaves the graphics card in a bad state?
[13:34] <tjaalton> pitti: ok then.. try an older kernel :)
[13:35] <tjaalton> although the changelog doesn't seem to contain anything related..
[13:36] <pitti> tjaalton: I could try -3, yes
[13:36] <pitti> tjaalton: I'll watch this step by step, watching the log for pipe-A, trying older mesa, trying older kernel; thanks so far!
[13:36] <pitti> I guess each step takes a day or two :/
[13:36] <tjaalton> pitti: great, np
[13:45] <pitti> seb128: do the system monitor panel applets still work for you? I just restarted my session, and now I just have a thin horizontal bar instead of the gauges
[13:49] <seb128> pitti: that's happens every now and then, the bug is not new
[13:49] <seb128> pitti: bug #24243
[13:49] <pitti> ugh, 5 digit bug :)
[13:49] <seb128> yeah ;-)
[13:49] <pitti> seb128: ok, nevermind then, thanks :)
[13:49] <seb128> np
[13:50] <seb128> pitti: btw all the pending hardy bugs have been retraced, many dups closed and the intrepid retracers have been restarted using the intrepid python-launchpad-bugs so daily dist-upgrade work again
[13:50] <pitti> rocking
[13:50] <seb128> pitti: and there is a bug-buddy waiting for approval which don't hijack apport, dunno if you do beta approvals
[13:51] <seb128> s/don't/doesn't
[13:52] <pitti> seb128: I'm not sure whether I'm supposed to; ATM I "just do" them if they fix a beta-RC bug
[13:54] <seb128> pitti: well, that's a gconf default value change to not have bug-buddy running, would be nice to have but no hurry either
[13:54] <pitti> seb128: oh, indeed; that was accidentally enabled again?
[13:54] <pitti> (I thought it was supposed to be off for ages)
[13:55] <kwwii> seb128: the only folder_copy and oflder_move we have are at 16x16 and already have a stock_ prefix
[13:55] <seb128> pitti: yes, new gnome-session changed the way bug-buddy is started, that was broken for the whole cycle, they fixed in for the stable tarball so bug-buddy is running again now
[13:55] <pitti> seb128: ok, I'll ack that and take the bullets
[13:56] <seb128> pitti: thanks
[13:56] <kwwii> seb128: if you know which size is missing I can make them
[13:56] <exarkun> How can I learn of proposed package removals (eg, much of python-xml from Hardy) before they are actually part of a release?
[13:57] <seb128> pitti: I'll ping slangasek for the next uploads, he should be around soon
[14:00] <kwwii> pitti: sent those icons off as well as a new sounds package last night
[14:00] <pitti> kwwii: saw it, thanks
[14:00] <kwwii> excellent :-)
[14:00] <seb128> kwwii: there is something wrong there, let me have a look to those
[14:00] <liw> kwwii, you might know this: if a program written for Ubuntu requires an icon, is there a process for getting one?
[14:01] <kwwii> seb128: they are in Human/16x16/stock/generic/ in case you are looking for them
[14:03] <seb128> kwwii: ok, that's the other way around, evolution uses folder-copy and folder-move
[14:03] <seb128> kwwii: do you use evolution?
[14:04] <pitti> seb128: do you happen to know if there's a reference bug for the broken libgphoto camera handling already?
[14:04] <pitti> seb128: if not, I'd like to create one, to track it for beta/final
[14:05] <pitti> seb128: (release team meeting prep)
[14:05] <pitti> \o/ hardy SRU queue empty again
[14:05] <seb128> pitti: bug #258083
[14:05] <pitti> seb128: merci
[14:05] <seb128> pitti: thanks for approving my uploads there ;-)
[14:07] <seb128> pitti: btw alex just fixed gnome bug #530654
[14:07] <pitti> seb128: I added a comment to 258083, which I think is the current state and approach
[14:07] <kwwii> seb128: nope, I do not use evolution
[14:07] <seb128> pitti: that should fix the issue we were speaking about yesterday
[14:07] <pitti> seb128: oh, that sounds related
[14:08] <kwwii> seb128: ok, I will take care of the renaming and then give you a new human-icon-theme package with those fixes and the new icons for the logout stuff
[14:08] <seb128> kwwii: ok, anyway if you could just do symlinks folder-copy -> stock_folder-copy and folder-move->stock_folder-move in the 16x16 variants that would be nice
[14:08] <seb128> kwwii: thanks!
[14:10] <pitti> seb128: ah, and f-spot would accept that converted URL?
[14:12] <seb128> pitti: gvfs should automatically convert those to the correct url, so that should solve the cases where you try to open a recent files on a fuse which not yet mounted for example
[14:12] <tedg> kwwii: We should probably add some way to set a list of symbolic links in the Breathe script.
[14:13] <kwwii> tedg: I would just do it in the debian/links file as we currently do
[14:13] <kwwii> keep that stuff away from artists, they will just bork things
[14:14] <tedg> kwwii: Well, while I don't want to discourage your dislike of artists, I figured there were some things that they may know better.  Like if an icon covers several mimetypes or something.
[14:16] <kwwii> they need the information, sure
[14:16] <pitti> hey tedg
[14:16] <pitti> tedg: can you please give me a quick status update of the f-u-s-a "wrong suspend/hibernate" options bug? (for our release meeting today)
[14:17] <pitti> tedg: bug 274140 I mean
[14:17] <pitti> kwwii: and I guess bug 269504 is underway, as far as I understood it?
[14:19] <tedg> pitti: I'm working on it, I hope to have it done this morning, but definitely before I finish work today.
[14:19] <pitti> tedg: right, shouldn't be a rush; I'm primarily interested in whether the bug is understood, our you need help (from me with hal, or from upstream, etc.)
[14:19] <pitti> tedg: it seems that fusa and g-p-m currently use differnet methods to determine available options then?
[14:20] <tedg> pitti: No, it's that FUSA is not asking GPM currently.
[14:21] <tedg> pitti: GPM provides a DBus command to query what it's thinking, so it'll use that so that they are assured to agree.
[14:25] <kwwii> seb128: just sent an email with the package details, let me know if there is a problem
[14:25] <seb128> kwwii: ok, I'll have a look, thank you
[14:26] <pitti> tedg: that sounds good, thanks; so this can be chalked off as "in progress"?
[14:27] <tedg> pitti: Yes.
[14:27] <pitti> tedg: thanks, I updated the bug (for the records)
[14:27] <pitti> seb128: the f-spot bug: my or your plate?
[14:28] <seb128> pitti: we both have lot on our plates, the first to have some free slot to look at it?
[14:28] <seb128> pitti: I'll not do it today for sure but I can have a look next week
[14:28] <pitti> seb128: right, same here
[14:33] <kwwii> seb128: wait on that update, I found a bug
[14:37] <pitti> kwwii: human-theme uploaded; I'll upload ubuntu-sounds as well, but they won't make beta (unless you can convince slangasek to approve it with a good reason why we need them)
[14:38] <seb128> pitti: what version did you upload?
[14:38] <pitti> kwwii: oh, it seems -sounds is already it
[14:38] <pitti> in
[14:38] <seb128> pitti: I think -sounds is a conflicting update, themuso did some updates too
[14:38] <shingara> How I can know the number of patchlevel of ubuntu specific in a package ?
[14:38] <pitti> seb128: ah, indeed
[14:38] <kwwii> pitti: excellent, thanks!
[14:38] <pitti> sending mail back
[14:39] <seb128> ok, trying some changes, be back in a bit
[14:40] <directhex> shingara, sorry, the question isn't 100% clear. you don't just want the number after the word "ubuntu" in a package version?
[14:42] <shingara> no
[14:42] <shingara> ubuntu create a package specific
[14:43] <shingara> maybe several time on only one version upstream
[14:43] <Hobbsee> Hm, linux identity wants an article on the 8.10 release.  I wonder who to ask about doing that
[14:43] <shingara> I want separate the twice information
[14:43] <shingara> the version upstream and the version ubuntu specific
[14:44] <Hobbsee> shingara: the left and right hand sides of the - will tell you that
[14:44] <RainCT> (Does someone here have -or had- a Samsung SGH-ZV60?)
[14:44] <directhex> Hobbsee, "the nice brown ubuntu you know and love, but moar new!". there, that was easy
[14:44] <kwwii> TheMuso: dude, you uploaded an ubuntu-sounds_0.9?
[14:45] <Hobbsee> directhex: heh
[14:45] <shingara> in case of 0.5.2~alpha+svn-r248-1
[14:45] <shingara> it's -1 the ubuntu specific version ?
[14:45] <directhex> shingara, that's a debian version
[14:45] <shingara> package rest2web-doc
[14:45] <Hobbsee> !info rest2web-doc
[14:45] <directhex> shingara, there is no ubuntu specific patching to that
[14:46] <Hobbsee> 0.5.2~alpha+svn-r248 <-- upstream version.  -1 <-- debian version (no 'ubuntu' in it)
[14:46] <shingara> ok, so ubuntu add an ubuntu version to debian
[14:46] <Hobbsee> shingara: when we change it, yes
[14:46] <directhex> unless the number after the last - is 0, in which case it's an ubuntu original
[14:47] <directhex> -1 is a debian package, -1ubuntu1 is an ubuntu package based on debian -1, -0ubuntu1 is an ubuntu package with no debian basis
[14:47] <shingara> ok
[14:53] <shingara> gnome-dbg version  Version: 1:2.20.2.2
[14:54] <shingara> what is the 1: ?
[14:54] <shingara> because it's the version 2.20.2.2 no ?
[14:55] <dholbach> shingara: http://www.debian.org/doc/debian-policy/ch-controlfields.html#s-f-Version
[14:56] <shingara> tahnks
[14:56] <shingara> I searched this doc
[15:02] <Ng> pitti: re your comments on bug 258083, has the fix been uploaded? Should I be testing anything?
[15:02] <seb128> pitti: could you also accept gnome-python-extras?
[15:03] <seb128> pitti: it fixes the package not being installable due to gdl changes (which leads to all applications depending on it not being installable too)
[15:03] <pitti> Ng: no, nothing uploaded to intrepid yet
[15:03] <seb128> pitti: it's basically a small tweak to build using the new gdl and a rebuild
[15:03] <pitti> seb128: ah, I already noticed that, dist-upgrade wants to remove some packages due to that; accepted
[15:04] <seb128> pitti: thanks!
[15:04] <seb128> ok, I've away for around an hour, will catch back on other things after that
[15:04] <seb128> I'm away rather ;-)
[15:05] <Ng> pitti: ok. I'll keep my eyes open for something then. thanks :)
[15:06] <pitti> asac, Riddell: on whose plate is bug 259278? it's marked for beta, but doesn't have an assignee; also, any idea whether we can actually expect a fix in time for beta?
[15:07] <pitti> mvo: what's the status of bug 274303? you were working on that yesterday, right?
[15:07] <asac> pitti: i will ping the upstream guy who claimed to have something last week
[15:08] <pitti> it sounds like it'd entirely break networking on Kubuntu?
[15:11] <Hobbsee> pitti: based on #kubuntu-devel, it looks like Riddell has a fix, and that will be coming thru soon.
[15:11] <pitti> nice
[15:12] <Riddell> yes, give me a bit and I'll have something for people to test
[15:13] <pitti> Riddell: so it's not entirely hopeless then? :-)
[15:13] <pitti> Riddell: thanks
[15:13] <superm1> TheMuso, i just saw a ping from you many days ago regarding BCM4328.  have you tried it yet?
[15:14] <superm1> w.r.g to airport
[15:15] <pitti> seb128: you said that consolidate-spell-checkers was pretty much done, right? can we set it to implemented, or is tere still something to do for it?
[15:23] <Riddell> asac: please try http://www.kubuntu.org/~jriddell/tmp/knetworkmanager/network-manager-kde_0.7svn864988-0ubuntu1_i386.deb
[15:28] <asac> Riddell: is that the latest svn?
[15:28] <Riddell> asac: yes
[15:29] <asac> Riddell: i think you can just uploade to network-manager PPA :) ... current state is broken and it cant get worse :)
[15:30] <asac> Riddell: could you upload there? i will get feedback from all the folks waiting for knetworkmanager over the weekend then
[15:33] <Riddell> ok
[15:38] <Hobbsee> seb128: were you planning to deal with python-gnome2-extras?  it's not installable again
[15:38] <pitti> Riddell: do you think bug 270423 is important enough to milestone it for beta? seems mostly cosmetic to me, but it might hide a bigger problem
[15:40] <Riddell> pitti: I don't especially mind, although it should be fixed for final.  evand or cjwatson might be able to say if it's more important than just cosmetic
[15:41] <pitti> Riddell: right, since it might be a crash near to the end, or so
[15:43] <cjwatson> it's just very confusing; I don't think it's beta-critical
[15:45] <slangasek> seb128: pinging on IRC is fine, though so far I haven't gotten the queue back down to zero which puts it up on my todo list :)
[15:45] <StevenK> slangasek: Us nasty developers keep uploading stuff? :-)
[15:46] <slangasek> pretty much
[15:53] <mathiaz> cjwatson: now that partman-auto-raid has been promoted, can it be added to the installer seed or should it wait for after Beta ?
[15:56] <slangasek> argh, desktop CDs oversized at 711MB, and not even a previous daily to compare with :/
[15:57] <Hobbsee> ouch.  Try throwing the All New and Shiny We Love Pitti (tm) merchandise off.
[15:57] <pitti> slangasek: just alpha 6 for comparison? cronjobs were disabled for quite some time, right?
[15:57] <pitti> Hobbsee++
[15:57] <Hobbsee> pitti: :
[15:57] <Hobbsee> * :)
[15:58] <slangasek> they weren't disabled for that long, I don't know why there's not a daily from yesterday on here
[15:58] <directhex> how much does that make in excessive on-disk package use inside the image, roughly?
[16:00] <Hobbsee> pitti: you have 40404 karma.  I find that disturbing.
[16:00] <pitti> nice!
[16:03] <directhex> how exactly is LP karma earnt?
[16:04] <geser> directhex: with bugs, answers, branches, translations, specs and I guess some more
[16:05] <mvo> pitti: re 274303> that is fixed in bzr, I was doing final testing and then I will upload
[16:05] <directhex> so my stuff on altioth earns me no respect. *sniff*
[16:05] <StevenK> By doing stuff on LP
[16:05] <StevenK> Answering questions, fiddling with bugs, ...
[16:05] <directhex> closing 30 mono bugs probably helped.
[16:06] <geser> directhex: do that every day and your LP karma will go up (and mono get bug free :)
[16:07] <mvo> pitti: I added flglrx and the two old nvidia drivers to the current code
[16:07] <directhex> geser, i have my eye on all open mono bugs. half are incomplete, half are upstream bugs, 1 is fixed in debian NEW, one i can't reproduce but seems to affect old user accounts (but not fresh user accounts)
[16:08] <directhex> and by fixed in debian NEW i mean by me, of course ;)
[16:08] <cjwatson> mathiaz: I think it's OK to add it
[16:09] <directhex> geser, now, if i wasn't terrified of going near it, beagle is a treasure trove of bug merging opportunities.
[16:09] <cjwatson> slangasek: the livefs build got stuck on a lock
[16:09] <cjwatson> slangasek: I fixed it with #is's help this morning
[16:11] <slangasek> grr, not again
[16:12] <pitti> mvo: nice, thanks
[16:20] <seb128> Hobbsee: what is the issue? what version are you speaking about?
[16:24] <sbeattie> cjwatson: I'd hoped to see if the modprobe -Os changes helped with bug 258432, but lack of live cds prevented that.
[16:25] <cjwatson> sbeattie: I think we probably should just crank up the timeout; that should be straightforward given a few spare minutes
[16:27] <sbeattie> cjwatson: yeah, that's pretty simple to do and will hopefully paper over issues for most problematic hardware.
[16:28] <IntuitiveNipple> Can someone remind me of the security team name for bug reports? I just accidentally submitted a bug without ticking the security warning box, and searching for subscribers with "security" in the name gives me ten's of pages.
[16:29] <seb128> grrrr
[16:29] <jdstrand> IntuitiveNipple: ubuntu-security
[16:29] <seb128> pitti: python-gnome-extras failed to build, "Xvfb failed to start"
[16:29] <IntuitiveNipple> jdong: many thanks
[16:29] <IntuitiveNipple> oops...
[16:29] <IntuitiveNipple> jdstrand: many thanks
[16:29] <jdstrand> :)
[16:29] <pitti> seb128: I think my pygtk sponsored upload FTBFSed for exactly the same problem, but didn't have time yet to look at it
[16:30] <seb128> let me ask on #ubuntu-x
[16:33] <seb128> kwwii: looking at your icon theme update, there is no sleep icon, is that expected?
[16:33] <kwwii> seb128: you mean suspend?
[16:33] <seb128> kwwii:
[16:33] <seb128> +#define GSM_ICON_HIBERNATE "drive-harddisk"
[16:33] <seb128> +#define GSM_ICON_SLEEP     "sleep"
[16:34] <kwwii> seb128: I made one for that...lol, I named it wrong
[16:34] <seb128> kwwii: I mean "sleep" which is the icon for the suspend action
[16:34] <kwwii> I looked at the wrong line in the #define
[16:34] <seb128> kwwii: the update adds system-shutdown.svg and view-refresh.svg and computer.svg
[16:34] <seb128> kwwii: none of that seems to be suspend or sleep icons?
[16:34] <kwwii> seb128: yes, system-shutdown should be sleep
[16:34] <seb128> oh
[16:34] <seb128> the name is misleading ;-)
[16:34] <kwwii> it is a monitor with a moon on it
[16:35] <kwwii> because I screwed up the naming :-)
[16:35] <seb128> system-shutdown is what is used for halt action
[16:35] <kwwii> I will move it and make another package
[16:35] <seb128> kwwii: ok, can you fix that? no need to update the debian/changelog version again
[16:35] <kwwii> sorry, thanks for noticing
[16:35] <kwwii> yes, I'll get that to you in a minute
[16:35] <seb128> kwwii: no problem, just use the same version, fix the name, upload and ping me again ;-)
[16:42] <Riddell> Keybuk: you have ops in #ubuntu-meeting, hokmen needs a +q
[16:42] <Riddell> oh, sorted
[16:42] <Keybuk> Riddell: he's not in the channel
[16:42] <Keybuk> #ubuntu-meeting is owned by the irc council cabal
[16:42] <Keybuk> so I have no authoritah there
[16:43] <digistyl3> TheMuso: hi, are you around?
[16:43] <kwwii> seb128: I updated the packages on my server, so the links in that email now point to the new stuff
[16:43] <cjwatson> Keybuk: access #ubuntu-meeting list says you do, FWIW
[16:43] <Keybuk> cjwatson: chanserv says I don't ;)
[16:43] <Keybuk> -ChanServ- You are not authorized to perform this operation.
[16:43] <seb128> kwwii: your update doesn't fix the "switch user" issue right?
[16:44] <seb128> kwwii: #define GSM_ICON_SWITCH    "system-users"
[16:44] <cjwatson> Keybuk: oh, oh well
[16:44] <kwwii> seb128: I did not realize that there was a problem with that :-)
[16:44] <seb128> kwwii: open the logout dialog ;-)
[16:45] <kwwii> seb128: I am not running intrepid yet :-)
[16:45] <seb128> oh
[16:45] <seb128> kwwii: ok, so "system-users" has a similar issue
[16:45] <kwwii> all I have is my thinkpad and the disk is full
[16:46] <kwwii> can someone make a screenshot of the problem so I understand that context?
[16:46] <kwwii> or is there a bug about this?
[16:46] <seb128> kwwii: https://bugs.edge.launchpad.net/ubuntu/+source/human-icon-theme/+bug/269500
[16:47] <seb128> kwwii: the first screenshot
[16:47] <seb128> kwwii: what icon did you change to fix this one?
[16:47] <seb128> kwwii: is that your computer.svg change?
[16:48] <kwwii> seb128: erm, I think I marked the wrong bug as fixed :-)
[16:49] <seb128> kwwii: ok, is that something you can, want fix today too?
[16:50] <kwwii> seb128: yes, that I will make an icon for logout and then it will be fixed :-) I'll make another package after that and let you know
[16:53] <seb128> kwwii: ok, still no need to update the changelog, just add the icon that will be good enough ;-)
[16:54] <kwwii> seb128: ok, I'll ping you later when done..thanks for the help
[16:55] <seb128> ok, no problem, I didn't do a lot ;-)
[17:10] <mdz> pitti: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=467374#34 hints that perhaps the change is due to the ibm-acpi module
[17:10] <mdz> pitti: remember, the kernel started printing that warning, saying we shouldn't set the mask anymore, and we changed it?
[17:11] <pitti> mdz: I just wonder why it breaks my dell as well then? I don't even have that module
[17:11] <pitti> but again, that could be yet another bug, of course
[17:11] <mdz> pitti: I bet acpi-fakekey is doing the job on 8.04
[17:13] <mdz> pitti: does acpi_fakekey 142 do anything for you?
[17:13] <pitti> mdz: it might be interesting to compare the output of acpi-fakekey and the one from -evdev
[17:14] <mdz> pitti: it would be if acpi_fakekey produced any output ;-)
[17:14] <pitti> mdz: yes, it suspends
[17:14] <mdz> pitti: oh, interesting
[17:14] <mdz> pitti: that doesn't work on my thinkpad
[17:14] <mdz> pitti: perhaps yours is a different bug indeed
[17:15] <pitti> mdz: I think so, too, most of hte bug is pretty thinkpad specific
[17:15] <pitti> mdz: I guess I'll try to bisect that a bit; run with just acpid, run with just hal, run with hardy kernel, etc.
[17:16] <kirkland> mdz: just saw your post to 267682...  i have a thinkpad x61 running hardy, and a thinkpad t61p running intrepid, i can probably gather the required lshal info
[17:16] <pitti> mdz: so acpi_fakekey converts something like the HKEY/whatever_for_Fn_F3 to XF86Standby? or what is it used for?
[17:16] <mdz> kirkland: that would be fabulous; mine is at home so I can't to it right now
[17:17] <mdz> kirkland: could you also try sudo acpi_fakekey 142 ?
[17:17] <pitti> mdz: (xev shows XF86StandBy when I press Fn+F3)
[17:17] <mdz> kirkland: on both systems?
[17:17] <kirkland> sure...  let me power up the wife's x61
[17:17] <mdz> pitti: acpi_fakekey takes an ACPI event and converts it into a simulated keypress
[17:18] <mdz> pitti: this makes it possible to convert non-standard ACPI events to standard keycodes which can then be heard by programs which care about the events
[17:18] <mdz> pitti: (like hal)
[17:18] <pitti> mdz: that sounds pretty close to what evdev now does as well?
[17:18] <lukehasnoname> Do you guys know of an error with installing LVM on encryption in alternate install?
[17:18] <mdz> pitti: where does evdev get acpi events from?
[17:19] <pitti> mdz: quantum flux phase inverter? sorry, no idea
[17:19] <pitti> mdz: but I don't think there's a lot in between it and the kernel
[17:19] <pitti> and, after all, the XEvents it generates don't actually look bad, they just have no effect
[17:20] <mdz> pitti: unless the kernel emits acpi events through evdev, we would still need something to hear them
[17:20] <mdz> though it certainly would be nice if everything just came through as input events
[17:20] <mdz> from the kernel
[17:20] <kirkland> mdz: okay, both systems up...
[17:21] <mdz> kirkland: first check whether the suspend key works on both
[17:21] <mdz> kirkland: should work on 8.04, should be ignored on intrepid
[17:21] <kirkland> mdz: fn-F4 (little moon?)
[17:21] <mdz> kirkland: y
[17:21] <kirkland> mdz: hmm, sadly, no effect on either
[17:22] <pitti> mdz: well, that might very well be a red herring for me, of course; the XF86Standby I see might actually come from acpi_fakekey istead of directly from kernel->evdev
[17:22] <kirkland> mdz: the obvious ones work on the hardy machine
[17:22] <kirkland> mdz: sound up/down, etc.
[17:22] <mdz> kirkland: check your power management preferences
[17:22] <kirkland> mdz: battery, screen lock
[17:22] <kirkland> mdz: k
[17:22] <mdz> kirkland: if battery works, that's a useful data point, it doesn't work for me
[17:22] <mdz> kirkland: (on intrepid)
[17:23] <kirkland> mdz: ah, that was it
[17:23] <kirkland> mdz: suspended fine, once configured
[17:23] <kirkland> mdz: i had it set to "do nothing"
[17:23] <kirkland> mdz: i'll try on my intrepid laptop (might be gone for a moment)
[17:24] <kirkland> mdz: nope, doesn't work on intrepid (confirming bug)
[17:24] <kwwii> seb128: ok, same links as before
[17:24] <kwwii> seb128: I added a system-users vector and 48x48 icon
[17:24] <mdz> kirkland: ok, good
[17:24] <mdz> kirkland: on the hardy one, can you try sudo acpi_fakekey 142 ?
[17:25] <kirkland> mdz: okay, it's asleep
[17:25] <mdz> kirkland: and on intrepid?
[17:25] <seb128> kwwii: looking
[17:25] <kirkland> mdz: on intrepid, nada
[17:25] <pitti> mdz: ok, that confirms that the thinkpad and dell issues are in fact different
[17:25] <mdz> kirkland: ok, please post your findings to the bug and I'll think about what to do from here
[17:25] <kirkland> mdz: did you want some lshal info?
[17:26] <kirkland> mdz: shall i post that to the bug?
[17:26] <mdz> kirkland: ->bug please
[17:26] <kirkland> mdz: k, no problem
[17:26] <cjwatson> mathiaz: BTW, I was talking to Ian Jackson last night about the python-gobject triggers problem, and he reckoned it's fixed in current Debian dpkg
[17:26] <cjwatson> mathiaz: so I'll have a look at that post-beta
[17:26] <mathiaz> cjwatson: great - thanks !
[17:27] <pitti> that would get us rid of those nasty pre-depends?
[17:27] <cjwatson> so I'm told
[17:30] <cjwatson> we think, anyway. Basically any package in triggers-awaited shouldn't be allowed to satisfy a dependency
[17:31] <mdz> kirkland: does sudo acpi_fakekey 236 work on the intrepid one?
[17:31] <seb128> kwwii: there is no 48x48 icon in this update and you included the .bzr directory in the source (I can clean that), is the scalable icon enough?
[17:31] <kirkland> mdz: doesn't look like it did anything....
[17:32] <kirkland> mdz: what would that do?
[17:32] <mdz> kirkland: pop up a battery status dialog (try it on the hardy system)
[17:32] <kwwii> seb128: yes, there should be 48x48/apps/system-users.png
[17:32] <mdz> s/dialog/notification/
[17:32] <mdz> kirkland: same as fn+f2
[17:33] <seb128> kwwii: oh, there is one
[17:33] <kwwii> :-)
[17:33] <kwwii> sorry about the .bzr dir, it was a mistake
[17:33] <seb128> kwwii: debdiff is not listing it, weird
[17:33] <kirkland> mdz: ironically, on hardy, fn+F2 works (battery pops up), but not 236
[17:34] <mdz> kirkland: :-?
[17:35] <mdz> kirkland: fn+f2 should be running /etc/acpi/sleepbtn.sh which runs acpi_fakekey 236 as root
[17:36] <mdz> kirkland: can you watch xev on both sides when doing acpi_fakekey 142 and see what's different?
[17:36] <mdz> (both hardy and intrepid)
[17:36] <Riddell> mvo: upgrade with current dist upgrade tool seems to be going well
[17:37] <slangasek> mvo: does the compiz FFe fix the gconf backend issue? :)
[17:37] <mvo> Riddell: excellent, thanks. I run it over the auto tester now and then it ready to go (just fixed a issue with the nvidia stuff)
[17:37] <mdz> kirkland: I'd hoped 236 would be a simpler test case since it doesn't suspend the machine, but whatever you can get to work...
[17:38] <kirkland> mdz: right
[17:38] <kirkland> mdz: i'm on the phone with IBM, at the moment (regular server call)
[17:38] <kirkland> mdz: let me ping you in a few
[17:38] <mdz> kirkland: once you've done that, please check whether the brightness keys work
[17:38] <mdz> kirkland: ok, I'll queue up some more questions if that's OK
[17:38] <kees> Keybuk: what did the modprobe delay turn out to be?
[17:39] <mdz> kees: it turns out it's a billion times faster when compiled with -Os
[17:39] <mvo> slangasek: sort of, it certainly will work with gconf and has it as default. now the issue that if gconf is not wroking for some reason and compiz switch to flat file but never back is more problematic to fix, I look into it after bug 274303 is done
[17:39] <mdz> rather than -O2
[17:39] <kees> asac: any news on why n-m still keeps thinking it should manage my eth0 when it's configured "static" in my interfaces file?
[17:39] <slangasek> mvo: ok
[17:39] <kees> mdz: wild.
[17:39] <Treenaks> mdz: it makes that much difference? impressive..
[17:39] <slangasek> \sh: ping
[17:39] <mdz> kees: 256054
[17:39] <kees> mdz: how did that get uncovered?
[17:39] <kees> mdz: thanks
[17:40] <mdz> kees: arjan's presentation showed that fedora's was running faster, and scott narrowed it down from there
[17:40] <kees> mdz: so it's just a long-standing debian glitch?
[17:40] <cjwatson> bzr: ERROR: Pack '56b441f9b0f40b61da5873827fa4f86b' already exists in <bzrlib.repofmt.pack_repo.RepositoryPackCollection object at 0x9649d0c>
[17:40] <cjwatson> 'bzr get bzr+ssh://bazaar.launchpad.net/~ubuntustudio-dev/ubuntu-seeds/ubuntustudio.intrepid /tmp/tasksel-Lf9QGJ/checkout-ubuntustudio' failed with exit status 768
[17:40] <cjwatson> blink, what's up with ubuntustudio?
[17:40] <mdz> kees: it's a freak optimization side effect I think
[17:40] <kees> mdz: heh
[17:41] <mdz> kees: probably -O2 causes the code size to blow up somewhere, and it's overflowing cache
[17:41]  * kees imagines "freak optimization"
[17:41] <_MMA_> cjwatson: Hmm... I'll have to have TheMuso take a look.
[17:41] <seb128> kwwii: so, none of those new icons are installed in the deb
[17:41] <mdz> kirkland: what we need to do is map out the path that the event is taking on hardy and intrepid and where they differ
[17:41] <cjwatson> _MMA_: needs help from #bzr probably
[17:41] <cjwatson> note that I fetch all the seeds into a single repository in order to save time
[17:42] <mdz> kirkland: on hardy, what I think happens is kernel -> acpid -> /etc/acpi/*.sh -> acpi_fakekey -> ??? (maybe hal)
[17:42] <seb128> kwwii: you didn't touch the makefiles to install those apparently
[17:42] <_MMA_> cjwatson: Ok. I'll let him know as he would know best.
[17:42] <mdz> kirkland: on intrepid we've confirmed that acpi_fakekey doesn't get heard properly, so we need to see where it diverges
[17:42] <cjwatson> so in order to reproduce this: bzr get bzr+ssh://bazaar.launchpad.net/~ubuntu-core-dev/tasksel/ubuntu tasksel; cd tasksel; rm -rf ubuntu-tasks; make ubuntu-tasks
[17:42] <mdz> kirkland: apparently, changing the X keyboard driver to evdev impacts it
[17:43] <seb128> kwwii: scalable/actions/Makefile.am scalable/apps/Makefile.am scalable/devices/Makefile.am need to be updated to list the new icons
[17:43] <cjwatson> make that 'bzr get -r1373 ...', since I'm probably going to work around this
[17:44] <kwwii> seb128: ouch, I'm used to doing the python build stuff
[17:44] <seb128> kwwii: do you know what changes are required and how to do those?
[17:44] <seb128> kwwii: debdiff is useful to compare deb contents between versions btw
[17:45] <seb128> kwwii: basically edit the files I just listed, add the icons to install, run autogen.sh and that should it
[17:45] <seb128> "should be it"
[17:45] <_MMA_> cjwatson: noted.
[17:45] <mdz> kirkland: also check what hal-addon-input is doing on both systems. is it listening to the same event devices?  is it listening to the one where the event from acpi_fakekey comes out?
[17:46] <kwwii> seb128: yes, I am adding the files to the makefiles now
[17:47] <mdz> kirkland: the most bizarre thing is that it seems to work if the user logs in on a guest session, so there's something at the session level which is impacting this
[17:48] <mdz> kirkland: which I wouldn't expect since hal listens to the /dev/input/event* directly
[17:54] <Treenaks> mdz: can't you replace acpi_fakekey with something that logs getppid() (and possibly the name of that process) before doing the actual key insert?
[17:55] <kirkland> mdz: off the phone now, okay.
[17:56] <kirkland> mdz: do you want me to try the evdev driver next, or the guest session?
[17:59] <mdz> kirkland: I want you to find out what's listening to the event that acpi_fakekey generates on hardy
[17:59] <mdz> kirkland: I think it's hal-addon-input
[18:00] <kwwii> seb128: ok, same place as before :-)
[18:00] <seb128> kwwii: looking
[18:02] <seb128> kwwii: do you try to build your updates? ;-)
[18:03] <seb128> kwwii: you added a system-users.svg in 32x32/actions/Makefile.am but there is no such icon in the 32x32/actions directory
[18:04] <mdz> kirkland: you may find input-utils useful
[18:05] <mdz> kirkland: (lsinput and input-events)
[18:06] <mdz> kirkland: can I call you?
[18:07] <kirkland> mdz: confirmed: suspend does work from an intrepid guest session
[18:08] <kirkland> mdz: what's the lock-screen key code?
[18:09] <mdz> grep LOCK /usr/share/acpi-support/key-constants
[18:09] <mdz> KEY_CAPSLOCK=58
[18:09] <mdz> KEY_NUMLOCK=69
[18:09] <mdz> KEY_SCROLLLOCK=70
[18:09] <mdz> KEY_LOCK=$KEY_COFFEE
[18:09] <superm1> kirkland, you mean the password? there's a bug open about that
[18:09] <mdz> KEY_COFFEE=152
[18:10] <mdz> superm1: no, we're working on bug 267682
[18:10] <cjwatson> that inversion of KEY_LOCK and KEY_COFFEE is suspicion
[18:10] <cjwatson> suspicious
[18:10] <superm1> ah okay
[18:10] <mdz> cjwatson: not to me, that's expected (same in the kernel)
[18:11] <cjwatson> oh, you pasted them in the opposite order from how they are in the file
[18:11] <mdz> cjwatson: two separate commands, sorry
[18:11] <kirkland> mdz: http://pastebin.ubuntu.com/50933/
[18:12] <kirkland> mdz: http://pastebin.ubuntu.com/50934/  > intrepid
[18:13]  * mdz discovers pastebinit
[18:14] <kirkland> mdz: http://pastebin.ubuntu.com/50934/
[18:14] <kirkland> mdz: http://pastebin.com/f6db34582
[18:15] <kirkland> mdz: ^ intrepid lsinput
[18:15] <mdz> kirkland: once you have the lsinput output from both, I'm interested in strace -p on hald-addon-input when you run acpi_fakekey on both hardy and intrepid
[18:16] <kirkland> mdz: lsinput from hardy: http://pastebin.com/f5b8f062c
[18:16] <mdz> so on intrepid, it's writing to event1, which is "AT Translated Set 2 keyboard"
[18:16] <mdz> same on hardy
[18:16] <mdz> so that part is the same
[18:16] <mdz> kirkland: how about hald-addon-input?
[18:17] <seb128> slangasek: ok, gnome-python-extras creates some un-installability issues in intrepid, a fixed version has been uploaded but doesn't build due to xvfb-run being broken, I've discussed it on #ubuntu-x but it's not likely it'll be fixed before monday on their side, what do you think we should do? upload a gnome-python-extras which doesn't call xvfb-run to workaround the bug? wait on monday? get somebody to fix xvfb today?
[18:17] <mdz> kirkland: root      5125  0.0  0.0   3436  1048 ?        S    Sep24   0:08 hald-addon-input: Listening on /dev/input/event1 /dev/input/event2 /dev/input/event3 /dev/input/event4 /dev/input/event5
[18:18] <slangasek> seb128: hrm, is xvfb-run needed for anything besides testing?
[18:18] <kirkland> root      6005  5883  0 12:04 ?        00:00:00 hald-addon-input: Listening on /dev/input/event1 /dev/input/event2 /dev/input/event3 /dev/input/event4 /dev/input/event5 /dev/input/event6 /dev/input/event7 /dev/input/event9
[18:18] <jcristau> mdz: presumably, since xf86-input-evdev grabs the devices, hald-addon-input won't get anything..
[18:18] <mdz> jcristau: oh?  can only one process listen to them at once?
[18:19] <munckfish> Please excuse my naive question: but do kernel filesystem modules autoload? I am trying to work out why spufs seems to need explicit mention in /etc/modules to get it to load.
[18:19] <persia> Indeed.  /dev/input/event* streams are single-access.
[18:19] <mdz> that would explain a lot
[18:19] <jcristau> persia: not necessarily
[18:19] <cjwatson> slangasek: can you approve grub-installer if it's OK by you? (and tasksel too)
[18:20] <jcristau> but xf86-input-evdev does ioctl(EVIOCGRAB)
[18:20] <mdz> jcristau: why does it do that?
[18:20] <cjwatson> slangasek: xvfb-run is fairly often used in build processes
[18:20] <slangasek> cjwatson: right, I mean in the case of gnome-python-extras - to understand what an upload which doesn't call it will do
[18:21] <cjwatson> oh, sorry, I misread
[18:21] <jcristau> mdz: to prevent key presses to go to the console, and mouse stuff to /dev/input/mice
[18:21] <jcristau> s/to go/from going/
[18:21] <seb128> slangasek: in gnome-python-extras? no, it's just used to run the testsuite
[18:21] <slangasek> seb128: right, let's go for option a) then, please
[18:21] <kirkland> mdz: hald on hardy strace: http://pastebin.com/f45596ef0
[18:22] <seb128> slangasek: alright, doing that in a minute
[18:24] <mdz> kirkland: thanks
[18:29] <lukehasnoname> I don't know if the fact the partition is encrypted makes a differeence on the write speed
[18:30] <lukehasnoname> ack wrong window, sorry.
[18:30] <mdz> kirkland,cjwatson,pitti,slangasek: new analysis in bug 267682
[18:31] <mdz> kirkland: thanks for your help
[18:31] <kirkland> mdz: no problem, shall I drop off of hold?
[18:31] <kirkland> mdz: this music is morose
[18:31] <pitti> mdz: ah, got the mail
[18:33] <mdz> kirkland: could you tell us, from the strace output, which device is receiving the event from acpi_fakekey on hardy?
[18:33] <mdz> kirkland: is it the 'thinkpad extra buttons' device?
[18:35] <Keybuk> kees: I've yet to fully tell whether this makes our modprobe go away entirely
[18:35] <Keybuk> it just certainly makes it faster
[18:41] <pitti> bye everyone, have a nice weekend!
[18:41] <seb128> pitti: thanks, you too
[18:41] <kees> pitti: cya pitti!
[18:42] <kees> Keybuk: ah, no bootchart for it yet?
[18:42] <seb128> slangasek: I uploaded a gnome-python-extra adding a build-depends on libgl1-mesa-dri to workaround the xvfb issue, that was the faster way and I've to run now
[18:42] <Keybuk> kees: not on the Q1 where it's most notable
[18:51] <kirkland> mdz: how might i determine that?
[18:53] <mdz> kirkland: match up the file descriptor number with what's in /proc/pid/fd
[18:53] <mdz> kirkland: I'm pretty confident though, if you have other things to do, don't worry about it
[18:53] <bryce> mdz, I read your analysis on the bug.  Do you have a course of action in mind we should take?
[18:54] <pochu> slangasek: hi. Is debian/patches/01_gettext_not_xmal.patch in libgweather something that upstream could use? It looks to me like the answer is yes, but I want to be sure
[18:55] <pochu> slangasek: third entry in https://edge.launchpad.net/ubuntu/+source/libgweather
[18:55] <bryce> mdz, I recall in fedora 9 they ended up disabling evdev for keyboard handling entirely; perhaps that would be something to consider?
[18:55] <kirkland> mdz: i was just about to step out to lunch
[18:55] <liw> does there exist a library for taking a number+unit (say, size of a file in bytes) and returning a user-friendly number+unit (size of file in kB or KiB)?
[19:03] <lool> cjwatson: Hey, we were discussing enablement of automatic mode in mid images with persia; I was wondering whether it would make sense to permit enabling the automatic mode by preseeding ubiquity itself in someway (perhaps separately)?
[19:04] <cjwatson> lool: sorry, I need to finish up for today and don't have time to discuss this now; maybe mail ubuntu-installer@?
[19:05] <lool> cjwatson: Ok; this was future stuff anyway; will either mail or regrab another day
[19:05] <lool> cjwatson: Have a good evening and WE
[19:14] <seb128> pochu: I talked to vuntz about it and he was not really interested, he said we should rather ship the .xml in the language packs in ubuntu
[19:17] <liw> https://wiki.ubuntu.com/FreezeExceptionProcess says "Requests for freeze exceptions for main should be filed as bugs in launchpad against the relevant package", but then does not say that that should happen also for packages in universe -- should it? it mentions that motu-release should be subscribed for such bugs for packages in universe/multiverse, so I guess the bugs should be filed, but it'd be nice to be explicit
[19:17] <tjaalton> bryce: that was because of gnome vs. xkb deathmatch, and it's been fixed since (xkb won :)
[19:21] <slangasek> pochu: yes, it is; though I'm not sure that it's upstream-ready, I was hoping seb128 would assess that
[19:22] <pochu> slangasek: read his comment from 8 minutes ago
[19:22] <slangasek> right, I see it
[19:22] <pochu> ok
[19:22] <slangasek> that's a big "wtf" to me
[19:22] <pochu> (vuntz maintains the package upstream)
[19:22] <slangasek> someone has an unhealthy relationship with xml, it seems
[19:23] <pochu> heh
[19:23] <pochu> so I'll just fix the patch to apply cleanly again
[19:39] <\sh> slangasek: pong...from long time ago
[19:40] <slangasek> \sh: ah, I was just about to write to LP instead :)
[19:40] <slangasek> \sh: bug #271550 - this is a regression vs. hardy, and you seem to have handled the only two uploads to intrepid
[19:40] <\sh> slangasek: lemme check
[19:42] <\sh> but while I'm checking this...where did sun-java*-plugin went?
[19:42] <\sh> s/went/go/ ... my brain is on less power today somehow
[19:43] <slangasek> I see sun-java6-plugin in intrepid, so... "Nowhere"? :)
[19:43] <\sh> hmm...not me...anmd I have multiverse enabled
[19:44] <slangasek> it's present on i386 and lpia
[19:44] <slangasek> hmmmm, but I only see it when I look on drescher, not in my own apt cache
[19:44] <\sh> hmm...and on amd64 ?
[19:44] <\sh> ah
[19:44] <slangasek> oh, of course, because my own apt cache is amd64, so nevermind :P
[19:44] <\sh> I can see it on the archives too...but not in my apt-cache
[19:45] <slangasek> anyway, AFAIK it's never been available for amd64?
[19:45] <\sh> ia32-sun-java6-plugin ? because sun-java6-jre suggests it ;)
[19:45] <\sh> but even this is not there
[19:49] <\sh> slangasek: libqt4-dbus was never in ia32-libs
[19:49] <\sh> latest hardy update/security version ubuntu11 had only libqt4-core and libqt4-gui
[19:49] <slangasek> \sh: well, perhaps, but the intrepid version has /no/ qt libs :)
[19:50] <fabbione> slangasek: are we in "strong" freeze now?
[19:50] <slangasek> -rw-r--r-- root/root    323068 2008-04-09 18:39 ./usr/lib32/libQtDBus.so.4.3.4
[19:50] <fabbione> hard.. that was the word
[19:50] <slangasek> \sh: that's from the hardy version
[19:50] <slangasek> fabbione: yes
[19:51] <\sh> WTF!
[19:51] <\sh> they are in fetch-and-build
[19:51] <fabbione> slangasek: i'd like to give intrepid a bit more cluster love.. there is a new corosync upstream official release (3 svn commits more than what's in the archive now). openais similar situation and I'd like to push the last redhat-clsuter release i did yesterday
[19:51] <slangasek> "cluster love", how euphemistic
[19:51] <fabbione> slangasek: ehehe i know you could handle that :)
[19:52] <fabbione> slangasek: you are rm.. unless you have objections i'll upload eitehr tonight or tomorrow
[19:52] <slangasek> fabbione: well, I don't object to you uploading, but if you want them to be let out of the queue, please follow the feature freeze guidelines :-)
[19:53] <slangasek> fabbione: i.e., submit a bug documenting why these are appropriate exceptions to the feature freeze
[19:53] <\sh> slangasek: trying to bring them back...and adding libqt4-dbus
[19:53] <fabbione> slangasek: i just don't have time for all this burocracy. I could just uplaod saying that i added a patch and nobody will notice. I am asking to keep packages clean.
[19:53] <fabbione> slangasek: and i understand you have procedures.. just not enough time to follow all of them
[19:54] <fabbione> slangasek: anyway... thanks.
[19:55] <slangasek> fabbione: uploads that don't follow the freeze procedures push the bureaucracy onto the release team, who don't have gobs of spare time either; I can promise that I'll look at them if they're uploaded, but I can't promise that I'll have time to review it all myself
[19:56] <fabbione> slangasek: no worries.
[20:00] <\sh> ok...libqt4 and friends are coming back...
[20:00] <\sh> do I need a special "ack" for it nowadays?
[20:01] <\sh> but we really need a sun java plugin somehow for amd64...if we don't get it , all lovely admins with HP hardware and ilo2 are fcked
[20:03] <\sh> doko: any idea why it's missing somehow? :) it's referenced, but no installation candidate on amd64
[20:04] <slangasek> \sh: no acks needed for bugfix uploads to universe, no
[20:05] <\sh> slangasek: ok..it will land on the archives in less then 5 mins
[20:05] <\sh> s/archives/incoming/
[20:06] <directhex> \sh, i can't admin my lovely Enterprise-Grade, um, netgear
[20:06] <\sh> directhex: oh :)
[20:13] <directhex> y'know, i miss cuba. i might violate my "never return to the scene of the crime" hoiday policy
[20:14] <directhex> eep, wrong channel for such triviality. sorry.
[20:19] <_MMA_> If I have 2 licenses, that all allow derivatives, can they be combined? LGPL+CC-BY-SA for instance. This is for an image/art btw. (Im helping to get a package strait)
[20:20] <mvo> pitti: what should I check if the guest session does not work for me? is there debug output written somewhere?
[20:20] <directhex> _MMA_, the SA clause means the derivative needs to be SA too
[20:20] <mvo> pitti: I suspect its a nvidia problem, not the session itself
[20:21] <liw> mvo, I made a new system-cleaner version to fix bugs reported today... (I may have unit tests but I still let bugs get through *sigh')
[20:22] <\sh> slangasek: uploaded
[20:22] <mvo> pitti: nevermind, found it (bad nvidia bad)
[20:22] <mvo> liw: ok, on revu?
[20:22] <liw> mvo, in my PPA, and please don't upload yet, I would like confirmation from the users that the bugs are fixed
[20:22] <slangasek> \sh: thanks
[20:23] <mvo> liw: ok, thanks
[20:24] <\sh> now for something really interesting...ha nfs cluster with drbd storage failover time
[20:43] <calc> checking out the OOo human icons from bzr takes a very long time :\
[20:47] <slangasek> --lightweight?
[20:51] <calc> slangasek: seeing how that does, thanks
[20:52] <calc> still fairly slow, 2m so far for 4mb of data
[20:54] <calc> 4m 8mb
[20:55] <calc> at least its consistent in its slowness
[20:55] <\sh> crimsun: did you file a wishlist bug for leonov of integration of 5-a-day ? ,-)
[20:57] <calc> if this bzr bar is accurate i'll be getting the repo for another hour or so
[21:42] <digistyl3> TheMuso: ? :)
[22:11] <cody-somerville> slangasek, There doesn't seem to have been a livecd for Xubuntu in three days.
[22:11] <slangasek> yes, livefs builds were wedged
[22:11] <cody-somerville> Okay, thanks.
[22:12] <slangasek> respinning now