[12:08] <debianist> i actually have very strong interest and curiousity about d-i, however i tend to find it rather confusing to start getting familiar with it, so many parts, such wealth of scattered documentation. ofter the question arise , where do I start?
[12:09] <jdub> seb128: hooray for the menu :)
[12:10] <seb128> :)
[12:11] <jdub> hrm
[12:11] <jdub> might reboot for the new kernel
[12:11] <debianist> or more specifically, how do I start?
[12:12] <jdub> debianist: (probably better to ask in the d-i channels than here)
[12:13] <debianist> ahh, i'd feel to embarassed to ask this there..actually i am already having asked it here
[12:14] <debianist> mdz : rsync really did it :) it was a download error
[12:21] <debianist> :)
[12:40] <kagou> good night
[12:57] <debianist> anybody still awake here?
[01:01] <jdub> yeah
[01:01] <jdub> this is a 24x7 channel :)
[01:03] <debianist> it's just that I saw many went to sleep
[01:03] <debianist> ;
[01:03] <debianist> strange, I have completed first stage install. I rebooted into it,
[01:05] <debianist> now when I installed I chose hebrew, I get all of the ncurses window text in greek symbols
[01:05] <debianist> looks like the encodiong wasn't set correctly
[01:06] <debianist> this is the setup stage that continues from the HD
[01:06] <debianist> I am installing under QEMU
[01:06] <debianist> could the emulated environmnet be causing this?
[01:08] <mdz> possible, but unlikely
[01:08] <mdz> send a message to the sounder mailing list
[01:09] <debianist> mdz : hey you up so late also? :)
[01:10] <SnowBoB> it's late? I just got to work
[01:10] <debianist> well, for me it's 2:11am
[01:11] <daniels> it's late? i just woke up (well, 45min ago)
[01:12] <SnowBoB> daniels, yes but for you that could be anytime :P
[01:14] <mdz> debianist: I am UTC-7
[01:16] <debianist> mdz : little bit to much for to translate to my time, where are you in ?
[01:16] <debianist> (at that hour)
[01:16] <Hrdwr_BoB> -mdz- TIME Mon Sep  6 16:16:22
[01:16] <Hrdwr_BoB> the amazing powers of IRC clients :)
[01:17] <debianist> yes
[01:17] <debianist> :)
[01:18] <debianist> I'm so tired my eyes start to fail on me :)
[01:19] <Keybuk> daniels: isn't it like 10 or 11am there though?
[01:19] <thom> hrm. i need either less books or more book shelves :/
[01:19] <Keybuk> thom: I'm at the "need more walls to put more shelves on" stage :-/
[01:20] <debianist> haha keybuk
[01:20] <elmo> I need a flatter floor
[01:20] <thom> Keybuk: yeah
[01:20] <elmo> every time I take something out of my bookshelves, it feels like the damn things about to fall down on me
[01:20] <thom> i have a spare wall, but it has about the structural integrity of a dell 
[01:21] <thom> (ie, none)
[01:22] <daniels> daniel@tycho:~% TZ=Australia/Melbourne date
[01:22] <daniels> Tue Sep  7 09:22:05 EST 2004
[01:22] <Hrdwr_BoB> that reminds me
[01:22] <daniels> thom: i think you need less cds
[01:22] <Hrdwr_BoB> I am without coffee
[01:22] <daniels> thom: i think i need more cds
[01:22] <daniels> thom: SYNERGY!
[01:22] <thom> daniels: you pay shipping, and i shall send a select collection of cds to you
[01:23] <daniels> thom: that's great, except without that whole me pay shipping thing
[01:23] <daniels> moving my housemate's car out of the drive is interesting; it's an automatic that stalls at low revs
[01:23] <thom> i'm sure you'll appreciate my exes burt bacharach cd
[01:23] <Keybuk> daniels: are you sure that's not 9pm?!  Since when do you get up in the morning?
[01:23] <thom> rofl
[01:24] <Keybuk> . o O ( contemplates a TZ=Canonical/daniels date hack )
[01:24] <daniels> something like neutral+handbrake, rev the shit out of it, handbrake off, drop it into reverse and hope like hell no-one just appeared
[01:24] <daniels> Keybuk: ever since I got back :\
[01:24] <Keybuk> really?  it didn't take me long after Oxford to get back on to KST
[01:24] <mdz> elmo: how is that powerpc build going?
[01:24] <daniels> (and then mdz promptly schedules a meeting at 2am -- maybe he's trying toget me back to a sane timezone)
[01:24] <thom> i heartily approve of 1600 GMT :-)
[01:24] <mdz> daniels: it may need to be moved on fabio's account
[01:25] <mdz> probably earlier
[01:25] <daniels> mdz: to?
[01:25] <mdz> but he hasn't gotten back to me yet
[01:25] <daniels> whoo!
[01:25] <mdz> 1400 UTC is about the practical limit for me
[01:25] <mdz> assuming I need to actually speak
[01:25] <jdub> i've had really deeply sane hours recently
[01:25] <Keybuk> thom: yeah, I'm usually up by then
[01:25] <jdub> for my local timezone
[01:25] <mdz> jdub: what hours are those?
[01:25] <daniels> mdz: 'speak'
[01:25] <daniels> mdz: it is an irc meeting, right?
[01:25] <jdub> mdz: Australia/Sydney
[01:25] <mdz> daniels: yes
[01:26] <mdz> jdub: no, I mean what hours are you usually available (UTC)
[01:26] <daniels> mdz: otherwise i'll need to find a phone card as well as red eye :) (which i needfor xorg anyway)
[01:26] <daniels> mdz: whee, cool
[01:26] <jdub> mdz: whenever, i'm used to the damage
[01:27] <Keybuk> evo is seriously lacking a Timezone field in its contacts db
[01:27] <Keybuk> that would rock, especially combined with the calendar
[01:27] <Keybuk> and To: daniels (Local time: 9.27am)
[01:27] <elmo> mdz: boring the fuck out of me is how
[01:27] <elmo> it's built 5 or 6 images and still going
[01:28] <mdz> christ
[01:28] <mdz> it's been hours
[01:28] <jdub> Keybuk: ooh, that'd be rad
[01:29] <thom> Keybuk: talk to your pet evo hacker about it :p
[01:29] <jdub> *cough*
[01:29] <Hrdwr_BoB> after you let them out of your cupboarfd
[01:29] <mdz> elmo: looks like there should be a total of 6
[01:30] <mdz> jdub: seriously; I have a list and I'll try to at least rotate the damage if not accomodate everyone
[01:31] <Keybuk> thom: he does exchange stuff mostly, iirc
[01:31] <Hrdwr_BoB> write a patch
[01:31] <daniels> mdz: sounds like t1-cyrillic is being realy stupid an povidin /usr/lib/X11 as a directoy
[01:32] <daniels> mdz: my hours are  decn approximaton of 2200->1400 UTC
[01:32] <mdz> daniels: noted
[01:32] <Keybuk> and his rates ...
[01:33] <mdz> Keybuk: negotiable
[01:33] <jdub> mdz: i'm generally awake between UTC 2100 and UTC 1400
[01:33] <thom> mdz: please send him a new keyboard
[01:33] <Keybuk> just like his affection then :)
[01:34] <mdz> jdub: thanks
[01:34] <jdub> http://www.timeanddate.com/worldclock/meeting.html
[01:34] <thom> i'm concerned that "ati" might turn into "atikthxbye" 
[01:34] <mdz> I just need pitti and alextreme to complete my collection
[01:34] <daniels> Keybuk: i'm a uni bum, i'm cheap
[01:34] <jdub> ^ totally fucking amazingly useful
[01:34] <mdz> jdub: yeah, that tells me the time, but not when people are awake
[01:34] <mdz> which is a way more useful thing
[01:34] <mdz> I wish I could program htat into the worldclock
[01:34] <jdub> mmm, it definite fluctuates
[01:34] <daniels> mdz: yeah, i hear x40s have really nice keyboards
[01:35] <daniels> mdz: might stop me typoing 'warty' as 'unstable' so much
[01:35] <mdz> daniels: I know another way to stop you from typing 'unstable'
[01:35] <jdub> $ cat /home/jdub/bin/uch
[01:35] <jdub> #!/bin/sh
[01:35] <jdub> export DEBEMAIL=jeff.waugh@canonical.com
[01:35] <jdub> exec dch -i -D warty $@
[01:35] <mdz> daniels: I could break your fingers
[01:35] <jdub> 
[01:35] <thom> i think some good pavlovian response training would cure that
[01:35] <jdub> me no type warty :)
[01:35] <jdub> hrm, breakfast
[01:36] <jdub> oh
[01:36] <jdub> so
[01:36] <daniels> looks like the problem was the usb conection, as it now works a lot better in ps/2
[01:36] <daniels> now the only problem is my cold hands
[01:36] <daniels> mdz: that also prevents me from typing 'warty'
[01:36] <jdub> mdz: how much do we trust the bluetooth daemons and stuff?
[01:36] <mdz> daniels: details
[01:36] <daniels> thom: no electrodes on my testicles
[01:36] <mdz> jdub: depends on what they want :-)
[01:36] <jdub> well, they're in DesktopSeed
[01:37] <mdz> bluetooth has an effective range of a few feet, right?
[01:37] <jdub> and we don't have any gui bits there yet
[01:37] <jdub> and i believe they 'listen by default'
[01:37] <mdz> i.e., pretty close to console access anyway
[01:37] <Hrdwr_BoB> very close
[01:37] <thom> mdz: about 10-15, iirc
[01:37] <mdz> the trouble is that people make bluetooth keyboards and what not
[01:37] <mdz> and people expect to be able to use them
[01:37] <Hrdwr_BoB> unless you play funny buggers with antennas
[01:37] <jdub> mdz: ahr, that is true
[01:37] <Mithrandir> mdz: up to a 100 for the later versions, though
[01:38] <thom> Mithrandir: 100 metres, or 100feet?
[01:38] <jdub> so can we only allow input device support or something?
[01:38] <daniels> mdz: you canget up to 100m, depends on the device
[01:38] <Mithrandir> thom: 100m
[01:38] <Mithrandir> http://www.mobileinfo.com/Bluetooth/FAQ.htm#t6
[01:38] <daniels> mdz: it's generally accepted as being 10m, tho
[01:38] <Keybuk> daniels: just put Kinnison on your testicles, it's far more ... effective
[01:38] <jdub>  4522 ?        Ss     0:00 hcid: processing events
[01:38] <jdub>  4531 ?        S<     0:00 [krfcommd] 
[01:38] <jdub>  4533 ?        Ss     0:00 /usr/sbin/sdpd
[01:38] <daniels> mdz: so you'll usually either get ~10 or 100; either way, it's enough to use a keyboard
[01:38] <daniels> Keybuk: don't give him ideas.
[01:39] <jdub> mdz: perhaps i should mail edd about it?
[01:39] <Keybuk> I don't need to, he comes up with them all by himself
[01:39] <mdz> jdub: yeah, anybody who might know
[01:39] <mdz> I have never even held a bluetooth device in my hand
[01:39] <jdub> edd: or ping you here... hi!
[01:40] <thom> mdz: uh, doesn't your T42 have bluetooth?
[01:40] <thom> or are you not counting a laptop as a device :-)
[01:41] <mdz> thom: does it?
[01:41] <daniels> i believ you can configure hcid (/etc/bluetooth/hcid.conf) to configure which classes of device it will accept with
[01:41] <daniels> but you don't want to do that per default
[01:41] <daniels> just sit on a train witha  small bluetooth keyboard and fuck shit up
[01:41] <thom> mdz: certainly my x40 does
[01:41] <mdz> lspci doesn't reveal anything that is obviously a bluetooth thingy
[01:41] <Hrdwr_BoB> maybe it's attached to usb internally or somesuch wackiness
[01:41] <Keybuk> mdz: try lsusb ... they tend to show up on there
[01:42] <Keybuk> more usefully, you configure hcid to only pair with devices if they know a SECRET PIN NUMBER :p
[01:42] <mdz> it does seem to have 4 USB controllers
[01:42] <mdz> 0000:00:1d.0 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #1 (rev 01)
[01:42] <mdz> 0000:00:1d.1 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #2 (rev 01)
[01:42] <mdz> 0000:00:1d.2 USB Controller: Intel Corp. 82801DB (ICH4) USB UHCI #3 (rev 01)
[01:42] <mdz> 0000:00:1d.7 USB Controller: Intel Corp. 82801DB (ICH4) USB2 EHCI Controller (rev 01)
[01:42] <mdz> lsusb isn't any more helpful
[01:42] <daniels> mdz: right -4 usb1.1 chanels, and one usb2 channel
[01:42] <daniels> Keybuk: yeah
[01:42] <daniels> Keybuk: whichI never did get workng properl
[01:42] <daniels> well, I did, once
[01:43] <daniels> i thnk maybe I need new batteries for his keyboard, actualy
[01:43] <thom> mdz: what happened to your perl hotplug?
[01:43] <Keybuk> daniels: I just stuck "auth enable;" in /etc/bluetooth/hcid.conf
[01:44] <Keybuk> and the pin in /etc/bluetooth/pin
[01:44] <mdz> thom: I may have, er, installed over it
[01:44] <thom> D'OH!
[01:44] <mdz> ah, no, I have a copy
[01:44] <mdz> I didn't finish it
[01:44] <mdz> you want it?
[01:44] <mdz> it handles everything except one of the bitmask fields
[01:44] <thom> hotplug irritates me everytime i boot
[01:44] <thom> stick it in arch? :-)
[01:44] <mdz> hmm, actually, this may not be a copy
[01:45] <daniels> brb wakingup hosemat and gong to stor oget new batteries
[01:45] <mdz> yeah, it's gone, I was editing it in-place
[01:46] <mdz> it was only about 7 lines of new code
[01:46] <mdz> and then ripping out all the shell awfulness
[01:46] <mdz> it took longer than it should have because I did it all in awk before learning that awk doesn't have bitwise operators (!?)
[01:46] <mdz> only gawk
[01:50] <thom> how can awk not support bitwise ops? that's weeeeird
[01:50] <thom> oh well
[01:50] <thom> it'll just have to continue to annoy me, then
[01:50] <mdz> or you could fix it :-)
[01:50] <thom> i'll wait till hoary :-)
[01:51] <thom> unless warty isn't borky enough for you right now? ;-)
[01:51] <jdub> it's not bork, thom!
[01:51] <jdub> it's CHARACTER
[01:52] <thom> character building, maybe
[01:56] <thom> isn't that why bob2 ran away screaming? ;P
[01:56] <Keybuk> I suspect so
[01:57] <Keybuk> is bad :-/  I can't even change clothes without unit testing first
[01:57] <jdub> it is always good to make sure your unit is functional and aligned correctly before getting dressed
[01:58] <Keybuk> indeed, nothing worse than heading out for the evening to meet some hottie, and discovering half way through that your unit isn't working
[01:59] <jdub> i can understand steve's point now
[01:59] <jdub> i will look upon unit tests in a different light
[02:03] <daniels> sideshow contracted unititis
[02:03] <thom> i wonder if pouring bleach into my ear will help purge my brain
[02:04] <daniels> make sure that you leave all your cds to me
[02:06] <thom> daniels: you can have Dire Straits and Jewell, too
[02:08] <thom> daniels: and i'll be sure and send you powderfinger, too. i'm sure you'd struggle to find any in .au :P
[02:09] <daniels> thom: I'll take Dire Straits, and I'll use Jewell to slice bob2 up with
[02:09] <daniels> actually, I haven't got any Powderfinger albums, off the top of my head
[02:10] <daniels> thom: oh, and your favourite DnB group are playing on Saturday
[02:10] <daniels> thom: come down on Friday, you can crash here
[02:12] <mdz> I hate perl
[02:13] <daniels> er
[02:13] <daniels> i can't file a bug with bugzilla because my mouse doesn't work right now, but could someone please file a critical bug on pfaedit
[02:13] <daniels> given that it currently only contains /usr/share/doc/pfaedit
[02:13] <daniels> assign it to me
[02:14] <justdave> Bugzilla should work pretty good in lynx/links
[02:14] <mdz> thom: so I end up with a 0 that is != another 0
[02:14] <daniels> er, nevermind
[02:15] <daniels> hm. fontforge conflicts/replaces pfaedit, but doesn't provide a pfaedit binary
[02:15] <daniels> BONG
[02:15] <daniels> mdz: 'and, for my next trick, proof that black is white!"
[02:15] <daniels> (exhibit one: michael jackson)
[02:16] <Hrdwr_BoB> daniels goes on to die at the next zebra crossing </obscure hhgtg reference>
[02:17] <mdz> $pci_class is 131072, $class_mask is 0, $class is 0, but $pci_class & $class_mask != 0
[02:18] <thom> mdz: er, sorry?
[02:18] <mdz> thom: I'm bitwise-anding something with 0 and not getting 0
[02:18] <thom> yes
[02:19] <mdz> there's probably a type fuckup in there somewhere, but damned if I can see it
[02:19] <thom> mdz: if you stick what you have in arch, i can take a look when it's not 01:20 :-)
[02:20] <thom> justdave: did the bugzilla bug i filed seem reasonable?
[02:20] <lamont> sigh.
[02:20] <lamont> mdz: checkning
[02:20] <mdz> ($vendor,$device,$subvendor,$subdevice,$class,$class_mask) = map { hex } @F;
[02:20] <mdz> $class_matches = (394240 & $class_mask == $class);
[02:20] <mdz> print STDERR "class: 394240 & $class_mask =? $class -> $class_matches";
[02:21] <mdz> prints "class: 394240 & 0 =? 0 -> 0"
[02:21] <daniels> Hrdwr_BoB: i was making the same reference
[02:21] <Hrdwr_BoB> only more obscure :)
[02:22] <daniels> yeah
[02:22] <mdz> thom: putting something in arch is harder than fixing the bug :-P
[02:22] <thom> mdz: *chuckle*
[02:22] <lamont> mdz: that looks suspiciously like perl.
[02:22] <mdz> lamont: alas, I need to write this with only essential: yes
[02:23] <lamont> mdz: that's why gcc-opt is in C. :-)
[02:23] <justdave> thom: which one was that, the assign-to-self one?
[02:23] <thom> justdave: yes
[02:24] <justdave> yeah, should be.  there's a patch for it somewhere (it's not a new request) that I was trying to dig up.
[02:26] <mdz> aaarrrrggghhhhhh
[02:26] <mdz>        Note that "&" has lower priority than relational operators, so for
[02:26] <mdz>        example the brackets are essential in a test like
[02:27] <daniels> !!
[02:27] <jdub> daniels: that would be a syntax error
[02:28] <daniels> not in brainfuck
[02:28] <jdub> the worst thing about ordering something on the interweb from overseas is that you have to wait
[02:28] <jdub> the internet is not about waiting
[02:28] <Hrdwr_BoB> yeah
[02:28] <mdz> Before: real    0m10.742s
[02:28] <Hrdwr_BoB> you need a matter transference terminal
[02:28] <mdz> After: real    0m6.375s
[02:29] <mdz> that's the full hotplug run, of which the pci bit is the only part I modified
[02:29] <jdub> mdz: yowza
[02:31] <thom> mdz: nice
[02:31] <daniels> mdz: neat
[02:31] <Keybuk> jdub: not about waiting?  clearly you haven't tried to read a site that's just made slashdot?
[02:32] <jdub> Keybuk: (i thought someone was going to make a joke about .au bandwidth...)
[02:34] <Keybuk> you have bandwidth there?
[02:36] <lamont> Keybuk: only intra-house
[02:36] <thom> ciao
[02:40] <jdub> odd rain
[02:44] <mdz> elmo: did the kernel ever finish, or did you fall asleep?
[02:47] <mdz> lamont: was linux-source-2.6.8.1 uploaded for powerpc?
[02:48] <daniels> jdub: i thought it always rained (or hailed, as the case may be) in sydney
[02:49] <elmo> mdz: it finished 5 minutes ago, just made it to the queue and is going into the archive now
[02:49] <elmo> mdz: do you need me for anything else?
[02:49] <mdz> elmo: thanks
[02:49] <mdz> elmo: nope
[02:50] <elmo> night all
[02:50] <daniels> elmo: 'night dude
[02:57] <daniels> justdave: bugzilla is broken in links, can't select a component
[02:59] <daniels> justdave: whaddyaknow, broken in lynx too
[03:02] <justdave> ah, true, the new component thing. :)  heh.
[03:02] <daniels> justdave: and broken in firefox. i'd file a bug on bugzilla about this, but y'know
[03:02] <justdave> upstream works on links, that's a local hack here.
[03:02] <justdave> what's broken about it in firefox?
[03:02] <daniels> i just get a big red box saying PICK A COMPONENT, JACKASS on links, lynx, and firefox
[03:02] <daniels> can't file a bug
[03:03] <daniels> if 'fontforge' isn't a component name, then that's a seriously unintuitivbe message
[03:03] <daniels> there is no more drop-down list of components with type-ahead
[03:03] <daniels> i miss that
[03:04] <justdave> yeah, there's no fontforge componnet
[03:04] <daniels> ok
[03:04] <daniels> could you please make one, and fix the error message? :)
[03:04] <daniels> and i really did like the box with typeahead, now you just sort of have to clutch at straws
[03:05] <justdave> it's still typeahead, you just have to type three characters before it'll start going
[03:05] <justdave> because the index file is about 150 MB if I index it to 1 character :)
[03:05] <justdave> that's the downside of making it match substrings instead of left-side
[03:06] <mdz> justdave: how big is it for 3 characters with all packages?
[03:07] <justdave> just shy of 160K
[03:08] <Kamion> elmo: argh, have you gone?
[03:08] <Kamion> I need you for byhands
[03:08] <justdave> er, just over.  it's 173K with ~1600 packages in it.
[03:09] <lamont> Can't find source for linux-non-free-modules-2.6.8.1_2.6.8.1-1
[03:09] <mdz> Kamion: he's gone :-/
[03:10] <justdave> hmm, sorry, that's 2800 packages in my sample data, now that I'm counting it
[03:10] <mdz> hmm
[03:10] <Kamion> well, I'll do what I can and finish it tomorrow
[03:10] <mdz> that means that with ~10k packages, it'd be more like a megabyte
[03:11] <justdave> daniels: fontforge is in
[03:12] <justdave> and looking at this, it should actually still work in links, you'll just have to know the package name, you won't get a list to pick from.  probably didn't work before because the component didn't exist.
[03:12] <Kamion> should've got back earlier, but badly needed to unwind by killing some aliens :)
[03:12] <mdz> Kamion: is it OK to update the seeds for 2.6.8.1?
[03:12] <justdave> the error message tell you you didn't select one instead of saying the one you selected didn't exist?  that what happened?
[03:13] <lamont> mdz: linux-kernel-non-free-modules b-d's sharutils
[03:13] <mdz> lamont: yes?
[03:13] <mdz> linux-non-free-modules-2.6.8.1 (2.6.8.1-1.1) warty; urgency=low
[03:13] <mdz>   * Build-Depends: sharutils (for uudecode)
[03:13] <Kamion> mdz: crap, forgot about that for amd64 :)
[03:13] <mdz> uploaded 4 hours ago
[03:13] <Kamion> mdz: you do base, I'll do installer
[03:14] <mdz> Kamion: amd64?  base doesnt' seem to have been updated for any arches
[03:14] <Kamion> mdz: base should only need doing for powerpc in fact ...
[03:14] <lamont> mdz: cool
[03:14] <mdz> Kamion: base still says kernel-image-2.6-386
[03:14] <Kamion> mdz: base doesn't need it for i386/amd64, it uses the metapackages
[03:14] <mdz> needs s/kernel/linux/
[03:14] <Kamion> mdz: ... oh
[03:14] <Kamion> mdz: wrong metapackages, right
[03:14] <Kamion> my brain is eliding those damn things
[03:15] <lamont> mdz: linux-source uploaded for ppc
[03:15] <mdz> base done
[03:15] <mdz> lamont: yep, it's installed already
[03:15] <mdz> and I just finished test-booting it
[03:16] <lamont> ditto amd64
[03:16] <lamont> sorry I wasn't here to answer earlier
[03:17] <Kamion> installer done; will be wrong for powerpc for another few minutes
[03:19] <mdz> Kamion: updated base and supported
[03:20] <Kamion> remind me sometime to add some kind of expansion feature to germinate so that updating the installer seed is less hard work
[03:20] <Kamion> Kernel-Version: was supposed to do that, but at the moment it just restricts dependency resolution
[03:23] <mdz> Kamion: looked through all and extra; seems to be in order
[03:23] <mdz> kernel-* should probably show up in elmo's next diff for removal
[03:25] <Kamion> linux-kernel-di-powerpc-2.6 uploading
[03:25] <mdz> hmm
[03:25] <mdz> would people freak out if we stopped shipping kernel source in a .deb?
[03:25] <mdz> as I recall, one of the few good reasons for that was so that it could go on the binary CDs
[03:25] <mdz> but we don't put it on there anyway
[03:26] <mdz> hmm, I suppose it'll go on the DVD
[03:27] <mdz> Kamion: what's the location of your new germinate archive?
[03:27] <Kamion> colin.watson@canonical.com--2004/germinate--mainline--0
[03:27] <Kamion> usual place on chinstrap
[03:28] <justdave> managed to get the iBook installed by pointing it at the network archive since it wouldn't keep the cdrom mounted.
[03:28] <Kamion> ... and base-installer uploaded; I think that's all I can actually upload until the powerpc udebs are accepted, but I'll make the necessary changes to debian-installer locally
[03:28] <justdave> which went fine until I logged into gdm...  bunches of errors from X about the XKB configuration
[03:30] <Kamion> justdave: most of your bugs seemed to stem from the WEP thing; FYI, booting with DEBCONF_PRIORITY=high should work around that
[03:30] <daniels> justdave: cheers dude
[03:30] <daniels> justdave: um, i didn't get any typeahead at all
[03:30] <daniels> justdave: not in linkx, not in lynx, not in firefox
[03:31] <justdave> you won't get typeahead in the text browsers.
[03:31] <daniels> justdave: xkb errors? suck? please send me the exact errors, plus XFree86.0.log
[03:31] <justdave> works for me in firefox, just tried it
[03:31] <daniels> justdave: hm
[03:31] <Kamion> well, not most, but some
[03:34] <mdz> Kamion: does b-i have the smarts to select the tasty optimized kernel for all warty architectures?
[03:35] <mdz> oh, you just uploaded that
[03:35] <mdz> (for amd64)
[03:35] <Kamion> and powerpc
[03:35] <Kamion> well, not that it's so much optimization there as making the damn thing work
[03:36] <mdz> so should we promote a few kernels onto the CD?
[03:36] <mdz> -686 seems sensible
[03:36] <Kamion> SMP?
[03:37] <Kamion> although I can't say I've ever been terribly enthused by -power3-smp
[03:37] <jdub> for some reason, a while back, smp was chosen as the default kernel on my ppc
[03:38] <Kamion> on powerpc we could easily eat through a lot of space with smp kernels
[03:38] <Keybuk> ok, the FUCKING MOUSE just ran STRAIGHT THROUGH the trap!
[03:38] <Kamion> jdub: base-installer sometimes picks one at random if it hasn't been properly educated
[03:38] <Kamion> wouldn't worry about it, shouldn't happen from now on
[03:38] <jdub> heh
[03:38] <Keybuk> it did it while I watched it
[03:38] <mdz> GAH
[03:38] <Keybuk> danced straight through it
[03:39] <mdz> another broken init script
[03:39] <Kamion> that's at least number 4, by my reckoning
[03:39] <jdub> mdz: did nathaniel do all of the init scripts in supported, too?
[03:39] <Kamion> I think we need to conduct a full audit between now and preview release
[03:39] <Kamion> I meant to send an e-mail about that
[03:39] <mdz> jdub: I hope not
[03:39] <mdz> Kamion: I talked to mark about it on the phone today
[03:39] <jdub> mdz: which avoids bugs, but they're going to be different...
[03:39] <Kamion> he didn't do openssh-server, I was watching for that
[03:41] <mdz> jdub: yeah, and then regression-test every action (stop/start/reload/restart) :-/
[03:41] <Kamion> mdz: BTW, I guessed that misdn wasn't needed in d-i, but might've been wrong
[03:41] <Kamion> if it's needed I don't know where to put it
[03:41] <jdub> mdz: *fear*
[03:41] <mdz> Kamion: dunno, probably a question for doko
[03:41] <Kamion> I would rather that a small group of shell scripting experts went through it
[03:41] <Kamion> the diffs aren't large
[03:42] <Kamion> they need a lot of attention, but there aren't reams and reams of -+
[03:42] <jdub> so vim-tiny is in BaseSeed
[03:42] <jdub> but it wasn't installed when i did a 'custom' install the other day
[03:42] <justdave> daniels: mail sent
[03:43] <mdz> jdub: potentially because vim-tiny does not exist?
[03:43] <Kamion> tisn't in the archive, so germinate doesn't see it
[03:43] <jdub> COCK!
[03:44] <jdub> wtf?
[03:44] <Kamion> THAT EXPLAINS A LOT
[03:44] <jdub> should we switch that to vim?
[03:44] <Kamion> I'd been wondering why vim wasn't installed in the first stage
[03:44] <Kamion> oh, please do
[03:44] <mdz> there was this changelog entry in unstable
[03:44] <mdz> that said vim-tiny was going to come back with the next upload
[03:44] <Kamion> server installs without vim are crack :)
[03:44] <mdz> but the maintainer lied and it never did
[03:44] <mdz> I agree
[03:45] <jdub> ok, i'll switch
[03:45] <mdz> the last time this came up, lamont objected because vim wasn't bug-compatible enough for him :-)
[03:45] <jdub> heh
[03:45] <Kamion> I'd rather have an arguable vi than no vi; I can live with either ...
[03:45] <Kamion> (either> nvi or vim - personal preference is for vim though)
[03:46] <jdub> mmm, nvi is poo
[03:46] <jdub> and our base can handle vim
[03:46] <Kamion> shall I add that to debootstrap now, then?
[03:46] <jdub> i know!
[03:46] <jdub> let's use vim-python!
[03:46] <Kamion> *thwap*
[03:47] <jdub> i love your tough love
[03:47] <mdz> my preference is for vim as well
[03:48] <mdz> is anyone in here _not_ running the 2.6.8.1 kernel yet?
[03:48] <mdz> if so, DO IT NOW
[03:48] <lifeless> erm.
[03:48] <lifeless> next reboot.
[03:48] <Kamion> I will be as soon as we have a working installer :P
[03:48] <mdz> thank you and have a nice day
[03:48] <lifeless> what version of the 2200bg is in that kernel ?
[03:48] <Kamion> 2200bg?
[03:48] <mdz> lifeless: 0.4
[03:48] <lifeless> eek.
[03:49] <mdz> Kamion: ipw2200
[03:49] <Kamion> aha, a victim
[03:49] <mdz> do they have a new one?
[03:49] <lifeless> 0.7 is really quite essential.
[03:49] <mdz> oh, they do
[03:49] <jdub> 0.7?!
[03:49] <lifeless> new firmware too.
[03:49] <Kamion> once I add that to d-i I need people to test it
[03:49] <lifeless> development is cranking fast now.
[03:49] <Keybuk> isn't 2.6.8.1 upstream a very broken kernel ?
[03:49] <Kamion> I held off 'cos I'm fairly sure the firmware loading I attempted to add to ddetect is broken
[03:49] <jdub> they've been busy little bees
[03:49] <lifeless> I have full 54Mbits here now :}.
[03:50] <lifeless> mdz: oh there are a couple of wep patches post 0.7 you'll want as well.
[03:50] <Keybuk> mdz: what ACPI is in it?  does it have latest madwifi CVS in it?
[03:50] <Kamion> Keybuk: AIUI we've fixed the really broken bits
[03:51] <mdz> Keybuk: stock ACPI, madwifi I think is a release, not CVS
[03:51] <Keybuk> ok, I shall stick to my hand-builds then
[03:51] <mdz> Keybuk: 2.6.8.1 isn't broken for me, no
[03:51] <Keybuk> neither stock acpi or madwifi release work on the nc4010 :(
[03:52] <mdz> Keybuk: that's the sort of feedback I was looking for a week ago when I asked for testers on sounder@ :-P
[03:52] <mdz> which reminds me, I need to post another announcement now that powerpc and amd64 are there
[03:52] <Keybuk> heh, I have about a week-lag on testing warty at the moment -- I'm using this machine as a development environment, I can't keep killing it to test things <g>
[03:54] <Keybuk> once PMS is out of the way, *then* I can be your bitch :p
[04:02] <Kamion> debootstrap uploaded, adding vim and vim-common to base
[04:02] <lifeless> mdz: I'm in the same boat as keybuk.
[04:02] <lifeless> testing the distro is  alixury for me.
[04:03] <Kamion> get more machines :)
[04:03] <Keybuk> Kamion: they end up getting *used* for things
[04:06] <Kamion> mdz: going to make archive-copier the default for Sounder 8; it isn't going to get enough feedback otherwise
[04:06] <Kamion> unfortunately all of today's madness has left me without time to do the last bits of integration
[04:08] <jdub> $ uname -a
[04:08] <jdub> Linux willow 2.6.8.1-1-powerpc #1 Mon Sep 6 22:44:50 UTC 2004 ppc GNU/Linux
[04:34] <lamont> so when I install linux-image-2.6.8-1-... is it going to remove my old kernel?
[04:35] <Kamion> certainly shouldn't
[04:35] <lamont> Kamion: just being paranoid.
[04:35] <Kamion> you could watch what your package manager says :-)
[04:35] <lamont> heh
[05:20] <mdz> Kamion: sounds good
[05:26] <Kamion> mdz: why is bicyclerepair under "Germinate workarounds" in DesktopSeed?
[05:27] <mdz> Kamion: misplaced, I suppose
[05:27] <Kamion> ditto pymacs
[05:27] <lamont> wow - neat package name...
[05:27] <Kamion> neither of them make sense there as far as I can see
[05:28] <Kamion> everything in the workarounds sections should be associated with a known germinate bug, really
[05:29] <daniels> lamont: python introspection tool, iirc
[05:29] <Hrdwr_BoB> yeah it's a refactoring tool
[05:30] <lamont> daniels: yeah
[05:34] <Kamion> I think all the germinate workarounds in SupportedSeed can go away now
[05:38] <mdz> Kamion: I probably just wasn't paying attention to the different types of headings in the file when I added them
[05:38] <mdz> those categories need reworking
[05:38] <mdz> jdub: kernel treating you well?
[05:40] <jdub> mdz: haven't sprained anything yet
[05:40] <Kamion> are we going to do anything about Marco's pppoe comment at the bottom of BaseSeed?
[05:40] <jdub> but this is not a very demanding machine
[05:41] <mdz> Kamion: if that applies to the version we ship, then yes, certainly
[05:42] <mdz> I thought that was not true until later ppp versions
[05:43] <Kamion> ah, dunno
[05:43] <Kamion> not going to attempt to check at this time
[05:45] <mdz> Kamion: the changelog is convincing
[05:45] <mdz> I think it's accurate and if you're editing it anyway, go ahead and remove it
[05:45] <mdz> we can let it fall all the way to universe
[05:46] <Kamion> not editing it any more
[05:46] <Kamion> I'll do it tomorrow if nobody has beaten me to it
[05:47] <mdz> I'll just do it now
[05:47] <Kamion> I'll remove it from debootstrap tomorrow
[05:48] <mdz> ok
[05:48] <Kamion> you know, so much of apt-setup's code is there because the CD isn't mounted; gotta wonder how much easier it would be to do that in prebaseconfig
[05:48] <Kamion> before the CD gets UNmounted
[05:48] <daniels> jdub: ping?
[05:49] <jdub> PONG
[05:49] <Kamion> fabbione: ping me when you get up about apt-setup; I have an evil idea
[05:50] <mdz> Kamion: speaking of getting up, don't you need to sleep eventually so that you can do that too? :-)
[05:50] <Kamion> theoretically ...
[05:50] <Kamion> actually that was the plan nowish :)
[05:51] <mdz> not that I particularly mind if you work in my timezone :-)
[05:52] <mdz> but surely there are disadvantages
[05:54] <Kamion> oh yeah, tomorrow morning will kinda suck
[05:55] <Kamion> might take an afternoon nap to make up
[05:55] <mdz> you plan on seeing the morning? rough
[05:55] <Kamion> hence, crash now :-)
[06:14] <jdub> mdz: somehow, gpdf has not been on the supported list, due to the xpdf/gpdf thing; can i add it on? (it should be there, considering it's shipped in gnome)
[06:23] <mdz> jdub: hmm, doesn't it basically duplicate a bunch of code from xpdf?
[06:23] <mdz> I'd really prefer if we could choose one for warty
[06:25] <daniels> i thought it stuffed up in interesting ways with landscape pages, etc
[06:25] <daniels> but if it works, then it has my +1
[06:35] <daniels> oh, suck
[06:35] <daniels> justdave: um, is it possible to have the dropdown box always displayed?
[06:35] <daniels> justdave: with the list of all components if nothing has been typed
[06:37] <justdave> yes, but it'll screw performance.
[06:37] <daniels> mdz: 1056 is blocked by 1056
[06:37] <daniels> mdz: er, 1056 is blocked by 1057
[06:37] <sabdfl> jdub: is gnome-gpg no longer in supported?
[06:37] <justdave> like when you empty the box it'll take 10 or 15 seconds to fill in the list box again
[06:37] <daniels> justdave: yah, but it kinda sucks having to take a total stab in the dark
[06:37] <daniels> justdave: really that slow?
[06:37] <jdub> sabdfl: it never was; definitely not supported-worthy yet
[06:38] <jdub> mdz: ok, i'll leave it to hoary
[06:38] <justdave> it's javascript running on your local machine...
[06:38] <justdave> my machine I was testing on was 500 MHz
[06:38] <justdave> if you have a better machine it might not be so bad.
[06:39] <daniels> mine's a 2ghz athlonxp (so-called 2400+), but i can't really imagine it being that ludicrously slow
[06:39] <daniels> but there you go :)
[06:39] <justdave> I suppose there are some tricks I could pull...  like having a listbox already filled and just hide it, and swap it for the other one when you need the filled one.
[06:39] <daniels> that would be really nifty :) if kinda huge
[06:41] <justdave> it may have to wait a week or so if you can tolerate it for now, I have a number of things on my plate at the moment.
[06:42] <Keybuk> odd, when you os.fork() in Python, the interpreter follows the form
[06:42] <Keybuk> uh, fork
[06:42] <Keybuk> which is odd
[06:43] <Keybuk> ah, no, there's an exception on it which it caught
[06:49] <daniels> justdave: yeah, fo'sho
[06:52] <jdub> gar gar gar
[06:53] <sabdfl> jdub: ok, thanks
[07:07] <fabbione> morning guys
[07:07] <fabbione> Kamion: ping :-)
[07:51] <daniels> mdz: i've not done font stuff before; how easy is it to transition stuff to fontforge (worth doing for warty?)
[07:52] <mdz> daniels: I think pfaedit and fontforge are the same program with different names
[07:52] <daniels> rad
[08:03] <fabbione> mdz: changing non-free to restricted and contrib to universe will kill all the translation....
[08:03] <fabbione> specially the latter
[08:04] <fabbione> non-free can be almost sed with restricted
[08:04] <fabbione> but the contrib changes meaning completly
[08:05] <fabbione> probably we can just virtually ask for non-free/restricted when pri=low, and skip completely universe.
[08:09] <mdz> fabbione: mark doesn't want to ask any questions
[08:09] <mdz> and add restricted by default
[08:09] <mdz> and not universe
[08:11] <fabbione> mdz: i am not going to ask any question
[08:12] <mdz> fabbione: so this is only for the 'expert' install?
[08:12] <fabbione> mdz: but instead of reinventing the wheel, we can just pre-seed the debconf questions we have so that an expert user (that will have DEBCONF_PRI=low) can see the question and decide by himself
[08:12] <fabbione> mdz: correct
[08:13] <mdz> restricted and universe are different from non-free, though; we will need to change the text at least
[08:13] <fabbione> restricted matches 99% non-free
[08:13] <mdz> we will lose the translations, but we have little choice and this is only for expert mode
[08:13] <fabbione> universe and contrib are very different
[08:14] <fabbione> actually to make contrib into universe it is enough to drop one sentence
[08:14] <mdz> debian main+non-free maps to ubuntu main+restricted, but non-free doesn't map 99% to restricted
[08:14] <mdz> many things that would go in debian non-free go in ubuntu main
[08:14] <fabbione> i agree, but that's from a SC point of view...
[08:14] <mdz> ubuntu restricted will have less than 5 packages in it
[08:15] <fabbione> if we just look at the wording
[08:15] <fabbione> i think they match more than you think
[08:17] <daniels> mdz: fglrx isn't unrealistic for restricted, imo
[08:17] <daniels> given that right now it's the only thing that supports r4xx, and r3xx is now one behind the curve, so people won't unreasonably be expecting 3d on it
[08:18] <mdz> daniels: fglrx?
[08:18] <daniels> mdz: ati binary drivers
[08:18] <mdz> daniels: packaged?
[08:19] <daniels> pain in the arse to set up, but they're the only option for 3D on r3xx (9[5678] 00), and any display at all on r4xx (x[3468] 00)
[08:19] <daniels> mdz: probably packaged somewhere
[08:19] <fabbione> yes they are
[08:19] <fabbione> from an italian guy
[08:19] <fabbione> there was the URL on the wiki
[08:19] <daniels> are the packages decent?
[08:19] <fabbione> i dunno
[08:19] <fabbione> Mark is using them
[08:21] <mdz> are those the ones which crash his machine when opengl screensavers run?
[08:21] <jdub> don't bring facts into the discusson
[08:21] <fabbione> i have NO idea...
[08:22] <fabbione> when he reported the crashes, i didn't have any follow up
[08:22] <fabbione> so i really can't say
[08:23] <daniels> they were pretty stable when i tried them maybe a year back on my 9000 (r2xx)
[08:24] <mdz> don't the xfree86 drivers do 3d on those chips?
[08:24] <mdz> I thought gatos at least did
[08:25] <daniels> yes, but not tv-out
[08:25] <mdz> gatos does on r2xx
[08:25] <daniels> they also didn't do s3tc until recently (still need dodgy hacks), which was one of the things i was after
[08:25] <daniels> er, last i checked, gatos didn't
[08:25] <daniels> bear in mind this was around a year ago, but yeah
[08:26] <mdz> previously it only did tv-out with VBE calls
[08:26] <daniels> i think they had sort-of-um-maybe r200 support, certainly didn't work on my 9000 (rv250)
[08:26] <mdz> but someone did the work to figure out the right stuff since then
[08:26] <daniels> good news
[08:26] <daniels> they're merging with x.org also
[08:26] <mdz> I had a 9000 AIW at the time
[08:26] <fabbione> Template: apt-setup/universe
[08:26] <fabbione> Type: boolean
[08:26] <fabbione> Default: false
[08:26] <fabbione> _Description: Use universe software?
[08:26] <fabbione>  Some additional software has been made to work with Ubuntu. This software
[08:26] <fabbione>  is free and it is not a part of Ubuntu, but standard Ubuntu tools can be
[08:26] <mdz> oh, good
[08:26] <fabbione>  used to install it.
[08:26] <fabbione>  .
[08:26] <fabbione>  Please choose whether you want this software to be made available to you.
[08:27] <fabbione> mdz: does it look ok for you?
[08:27] <daniels> so we'll have ati.2/tv_output in x.org a week or two after the release
[08:27] <mdz> fabbione: I would say "This software is free, and while it is not a part of Ubuntu, standard Ubuntu tools can be used to install it"
[08:27] <mdz> fabbione: otherwise, looks good
[08:27] <mdz> or maybe s/while/though/
[08:28] <fabbione> while sounds ok
[08:29] <mdz> I'm not set up for that at home, so I just walk on the floor
[08:29] <fabbione> ehhe
[08:30] <fabbione> well since this template won't have translation, we can change it later on
[08:32] <Keybuk> eek
[08:32] <Keybuk> major arch-go-whoa
[08:33] <Keybuk> "CHECKSUM FILE(S) DISAGREE WITH DIRECTORY LISTING ABOUT WHAT FILES SHOULD BE PRESENT IN REVISION DIR OF ARCHIVE"
[08:33] <thom> whoa
[08:33] <Keybuk> ah
[08:33] <Keybuk> that's a *lovely* one
[08:33] <Keybuk> there was a cache rev waiting for a signature
[08:34] <Keybuk> so the .tar.gz was slowly uploading, and you can't get it until it's finished, signed and checksum.cacherev put alongside
[08:34] <Keybuk> eek
[08:34] <Keybuk> o/~ the committer and the user had a race
[08:38] <pitti> good morning everybody!
[08:44] <thom> hey dude
[08:56] <Mithrandir> moo
[09:11] <Mithrandir> mdz: 2.6.8.1-amd64-k8 seems to be happy here.
[09:12] <mdz> Mithrandir: here too, thanks
[09:12] <Mithrandir> and the bd_claim patch is applied, thanks.
[09:14] <mdz> yep
[09:14] <mdz> good night
[09:14] <Mithrandir> sleep well
[09:27] <kagou> hi
[09:27] <pitti> Hi kagou!
[09:28] <kagou> pitti: or mdz cn you tell seb128 that smb bug in gnome browsing is at least reolved :) ?
[09:29] <kagou> i'm at work and e ask me to test the bug quickly, and to report if it's good more quickly :)
[09:29] <kagou> thanks
[09:35] <debianist> 1GB is not enough for warty :(
[09:35] <thom> no.
[09:35] <debianist> darn, got the installation halted on "not enough space on device"
[09:36] <debianist> than I can forget of installing it on my 850MB had based alleged router machine
[09:37] <fabbione> sorry guys... how does a cdrom apt sources.list entry looks like?
[09:38] <thom> deb cdrom:[Ubuntu 4.10 _Warty Warthog_ - Unofficial i386 Binary-1 (20040906)] / unstable main
[09:38] <pitti> fabbione: apt-cdrom add
[09:39] <fabbione> thom: thanks
[09:39] <jdub> debianist: if it's a router, you can always enter 'custom' at the installer boot prompt
[09:39] <jdub> debianist: then it'll just install base
[09:39] <debianist> jdub : oh right , forgot about it. i am having to less sleep over the last couple of days. thansk
[09:39] <debianist> jdub : it was me even that asked you same question 2 days ago
[09:39] <debianist> :)
[09:40] <pitti> thom: nice :-) we both reported the not-so-eject problem at the same time :-)
[09:40] <thom> yeah
[09:40] <thom> that's pretty impressive timing :-)
[09:40] <thom> just resolved mine as a dup
[09:40] <debianist> jdub : what is pmount?
[09:41] <jdub> debianist: a special policy-based mounting tool
[09:42] <pitti> debianist: it is a wrapper around mount that allows normal users to mount removable devices
[09:42] <jdub> debianist: it's used by gnome-volume-manager in ubuntu
[09:42] <jdub> debianist: so normal users can mount and unmount removab-- thanks pitti ;)
[09:42] <pitti> debianist: this avoids mangling the fstab and allows to run hal as root
[09:42] <pitti> debianist: sorry, allows to run hal as normal user
[09:43] <debianist> pitti : hmm, but hal wasn't intended to run as root from first place, i mean by design ?
[09:43] <debianist> pitti : it was made to overcome the problem of rootless accesses to mounting
[09:43] <debianist> piiti : ?
[09:45] <pitti> debianist: it often runs as root in other distros to be able to mess up the fstab
[09:46] <pitti> debianist: but apart from that it only needs some additional groups and capabilities, so we have it run as 'hal' in Warty
[09:46] <pitti> debianist: hal is such a complex program, it shouldn't run as root
[09:46] <debianist> pitti : i am very interested in learning all the I can about it, would you point me to the proper places to start reading all about it?
[09:47] <pitti> debianist: http://wiki.no-name-yet.com/WartyWarthog_2fSecurityReports has all of my security reports
[09:47] <pitti> debianist: I tried to remove privileges from as many programs as possible
[09:48] <debianist> pitti : does it mess anything up with an already working installation of X? I have nvidia closed bin drivers on the laptop, and when installting hal and g-v-m my nvidia setup stopped working. luckily when removing the packages, everything went back to normal.
[09:49] <pitti> debianist: hmm, I cannot see how hal and gvm should influence the X graphics driver
[09:49] <pitti> debianist: gvm runs as user without any privileges, so it cannot mess up nvidia
[09:49] <debianist> pitti : well, it had all it's dependecies installed, dbus-1 also and the others.
[09:49] <pitti> debianist: hal has some more privileges, maybe it does some scans which irritate the nvidia driver
[09:50] <pitti> debianist: does the problem also occur with hal running as root?
[09:50] <debianist> pitti : strange..i was also surprised to see that, even more to realize removing the packages worked. :)
[09:51] <debianist> pitti : actually i think that is the way it's set up on sid
[09:51] <pitti> debianist: it is.
[09:51] <debianist> pitti : i didn't do anything manually, only apt-get install <pkgs>
[09:51] <pitti> debianist: I proposed the patch also to Debian, but they did not adopt it yet
[09:51] <pitti> debianist: luckily I could convince upstream to adopt the changes
[09:51] <debianist> pitti : the patch to have it purely user mode?
[09:51] <pitti> debianist: which patch?
[09:52] <debianist> "I proposed the patch also to Debian, but they did not adopt it yet"
[09:52] <fabbione> grep: /etc/environmment: No such file or directory
[09:52] <fabbione>  * Stopping GNOME Display Manager...                                                                                             [ ok ] 
[09:52] <fabbione>  * Starting GNOME Display Manager...                                                                                             [ ok ] 
[09:52] <pitti> debianist: ah, the patched hal to allow it to run also as non-root
[09:52] <debianist> yes
[09:52] <pitti> debianist: you can switch it with debconf:
[09:52] <pitti> debianist: sudo dpkg-reconfigure hal
[09:53] <pitti> debianist: default is user for Warty, root for sid
[09:53] <debianist> user in warty, as in the first regular users that the system created?
[09:53] <pitti> debianist: no, it runs as its own system user called 'hal'
[09:53] <pitti> debianist: it's all in the security report and changelog
[09:54] <pitti> debianist: do you have access to the wiki?
[09:54] <debianist> i see. much as an apache server would run and not be given root access
[09:54] <debianist> pitti : yes i do, however i tend to get lost there ;-)
[09:54] <pitti> debianist: I posted the URL above
[09:55] <debianist> oh, it did asked for user name and password ; is that the oinky one?
[09:55] <pitti> debianist: basically it's user hal + some privileged groups + some kernel capabilities (CAP_NET_ADMIN IIRC)
[09:55] <pitti> debianist: hal does not prompt for that
[09:55] <pitti> debianist: it just asks whether to run hal as user or as root
[09:56] <debianist> pitti : this would work on the sid also?
[09:56] <debianist> e.g. the dpkg-reconfigure magic
[09:57] <pitti> debianist: yes, as soon as the patch is adopted in sid
[09:58] <debianist> pitti : oh ok, ubuntu strengths are starting to show...;)
[09:58] <pitti> debianist: actually we are quite good at the moment
[09:58] <pitti> debianist: all nonnecessary suid root bits and root daemons are eliminated
[09:58] <pitti> debianist: I hope that it will get even better, I try to convince the guys to have mandatory access control by default
[09:59] <debianist> pitti : reading the secuiryt policy..yeah looks like it's going to rock
[09:59] <pitti> debianist: but this is of course Hoary
[09:59] <debianist> pitti : what do you mean by access control mandatory, which parts?
[10:01] <pitti> debianist: things like grsecurity, RSBAC and SELinux
[10:01] <thom> pitti: i think everyone agrees that in principal it's a good idea. we just have to work out which system, and how to present it to the user
[10:02] <thom> (I guess we use SELinux, since RHAT are already tackling these kinds of problems)
[10:02] <pitti> debianist: you have a global fine-grained privilege policy that cannot be overriden by users, processes and packages
[10:02] <debianist> but that can be too restricting ?
[10:02] <pitti> thom: grsecurity has the advantage that it also brings a lot of other security-enhancing patches, PaX is not the least important one :-)
[10:02] <pitti> debianist: on a server it is not too hard, but on a Desktop it will get tricky
[10:03] <pitti> debianist: I guess/hope it will become by Hoary job to deal with this
[10:03] <pitti> thom: however, SELinux is upstream in Linux, so it is a good candidate
[10:03] <debianist> yeah. we might need to conduct user session see where are the pitfalls, which parts to leave open etc
[10:03] <debianist> e.g. do an experiment with a suggested policy, see how it serves the users in simulated real life usage
[10:04] <pitti> debianist: actually normal users should not notice that there are additional ACLs
[10:04] <thom> pitti: nod. (to both points)
[10:04] <pitti> guys, I need some breakfast. Returning soon.
[10:04] <thom> it'll be interesting to see how fedora core 3 is received
[10:04] <thom> enjoy
[10:08] <fabbione> hmmm
[10:09] <fabbione> i need some suggestions...
[10:09] <fabbione> if we install from the net we can safely add different apt-lines like restricted, security & co...
[10:09] <fabbione> and base-config will test them
[10:09] <fabbione> BUT (of course there is always a but)
[10:10] <fabbione> installing from cdrom...
[10:10] <fabbione> should we add the lines before or after the test?
[10:10] <fabbione> in the former case... what if it fails to test?
[10:13] <debianist> :)
[10:14] <debianist> if i have other OSs on the sid's menu.lst , they would be added also right?
[10:14] <fabbione> yes
[10:33] <pitti> fabbione: maybe it makes sense to add the lines, but comment them out?
[10:33] <pitti> fabbione: this will make it easier for the user to enable them, but it makes no sense to have them by default if you have no network
[10:33] <fabbione> pitti: that's what i was thinking....
[10:34] <fabbione> there is also the issue of which mirror to use....
[10:34] <fabbione> but i guess we can default to no-name-yet
[10:35] <pitti> fabbione: do we actually have mirrors ATM?
[10:35] <jdub> no
[10:35] <pitti> fabbione: the user selects a region/country, so in the future we could have a mapping country -> mirror
[10:38] <fabbione> pitti: no we don't have official mirrors, but for instance if i choose the mirror manually i want that to be respected
[10:38] <fabbione> pitti: and the mapping is done already somehow
[10:39] <pitti> fabbione: you want to add a custom mirror dialog to base-config?
[10:40] <pitti> fabbione: people who are doing that can probably edit sources.list by hand, don't they?
[10:40] <pitti> fabbione: most of the users won't care about the particular mirror
[10:41] <seb128> morning
[10:41] <pitti> seb128: good morning
[10:41] <seb128> hey pitti 
[10:42] <pitti> seb128: there are still two untranslated items in my Computer menu
[10:42] <seb128> which ones ?
[10:42] <pitti> seb128: "about Ubuntu" and "Printing"
[10:42] <jdub> yo seb128 
[10:42] <seb128> oh, printing is not a panel one, it's a desktop file
[10:43] <pitti> seb128: IIRC "About Ubuntu" should belong to yesterday's translation 
[10:43] <seb128> hey hey jdub 
[10:43] <seb128> yes
[10:43] <pitti> seb128: interestingly, the bubble help of About Ubuntu is translated
[10:43] <fabbione> little flood:
[10:43] <fabbione>         # Warty change add extra commented apt-lines
[10:43] <fabbione>         if [ "$URI" = "cdrom" ] ; then
[10:43] <fabbione>                 echo "" >> ${APTETC}sources.list
[10:43] <fabbione>                 echo "# Uncomment the following line to fetch update software from the network or" >> ${APTETC}sources.list
[10:43] <fabbione>                 echo "#deb http://ftp.no-name-yet.com/no-name-yet $DIST main restricted" >> ${APTETC}sources.list
[10:43] <fabbione>         fi
[10:43] <fabbione>         echo "# uncomment the following line to be able to use more than 12000 unsupported packages." >> ${APTETC}sources.list
[10:43] <fabbione>         echo "#deb http://ftp.no-name-yet.com/no-name-yet $DIST main restricted universe" >> ${APTETC}sources.list
[10:44] <fabbione> how does it look?
[10:44] <fabbione> the latter needs s/n-n-y.com/$MIRROR
[10:45] <seb128> pitti: the string is in your po file, I'll check for the next upload
[10:45] <pitti> fabbione: nice in general. Can you please add another comment for the deb-src?
[10:45] <pitti> seb128: the "Printing" string is somewhere in the gnome cups packages?
[10:46] <pitti> seb128: we really need Rosetta, don't we?
[10:46] <fabbione> pitti: yes.. it was next in the TODO :)
[10:46] <pitti> fabbione: another hing. You add the restricted universe regardless of whether you install from CD-ROM?
[10:47] <pitti> fabbione: this line is not contained in the if URI = cdrom
[10:47] <seb128> pitti: yeah :)
[10:49] <fabbione> pitti: uh?
[10:49] <fabbione> pitti: i am not sure i undesrtand what you mean
[10:50] <pitti> fabbione: you add the first #deb line only if URI=cdrom, but you add the second #deb line unconditionally
[10:51] <fabbione> yes that's correct
[10:51] <fabbione> because universe is always disabled by default
[10:52] <fabbione> if you are netinstalling main and restricted are enabled
[10:52] <fabbione> but not universe
[10:53] <pitti> fabbione: ah, I understand.
[10:53] <fabbione> so basically the universe note will go in with cdrom|http|ftp and in case the user re-run base-config and says: "No i don't want universe"
[10:53] <fabbione> otherwise the note will not be added
[10:53] <fabbione> does it make sense?
[10:58] <pitti> fabbione: another thing: is it necessary to repeat main and restricted in the universe line?
[10:59] <pitti> fabbione: users may come to the idea to enable both, I don't know whether this hurts
[10:59] <pitti> fabbione: both = both lines
[11:04] <fabbione> pitti: we agreed to show the entire example.
[11:04] <fabbione> in any case people that will touch manually sources.list are supposed to know what they are doing
[11:07] <pitti> fabbione: okay, that makes sense
[11:09] <fabbione> hmmmm
[11:13] <sabdfl> hi all
[11:14] <pitti> hi sabdfl
[11:16] <fabbione> hi sabdfl 
[11:21] <fabbione> AHHH here it is...!
[11:21] <fabbione> the loop in apt-setup is evil
[11:22] <seb128> hey sabdfl 
[11:26] <jdub> so i'm developing a problem
[11:27] <jdub> (to complement all the other ones, haw haw, very funny)
[11:27] <jdub> in that, whenever i type 'n'
[11:27] <jdub> i end up typing 'ntu'
[11:27] <jdub> i think i can claim this on worker's compensation
[11:27] <jdub> as "crippling limitation on typing capacity"
[11:27] <Riff> jdub: you are obsessed with ISDN
[11:28] <Riff> you brain can't help typing NTU
[11:28] <jdub> is ntu an isdn thing?
[11:28] <Riff> well, not specifically, but you see the term a lot
[11:28] <Riff> a Network Termination Unit
[11:28] <jdub> ahr
[11:29] <sabdfl> seb128 hi
[11:30] <sabdfl> i have a .xscreensaver base to contribute, where can I put it? it needs to become the default *or at least the start of the discussion :-)
[11:31] <seb128> in the bug report #946 ?
[11:33] <sabdfl> seb128: ok thanks
[11:34] <thom> sabdfl: or use the wiki page referenced in the bug
[11:34] <thom> either way :-)
[11:41] <jdub> seb128: oh man
[11:41] <seb128> hey jdub :)
[11:41] <jdub> seb128: seriously arse string change bug
[11:42] <jdub> Clear the Recent Documents list?
[11:42] <jdub> If you clear the Recent Documents list, you clear the following:
[11:42] <jdub>  All items from the Actions  Recent Documents menu item.
[11:42] <jdub>  All items from the recent documents list in all applications.
[11:42] <jdub> 
[11:42] <jdub> "Actions  Recent Documents"
[11:42] <jdub> d'oh
[11:42] <seb128> oh yes
[11:43] <sabdfl> good catch
[11:43] <jdub> seb128: do you like the wifi/battery disabled icons?
[11:44] <jdub> seb128: i was playing with the mixer applet earlier, and thinking i must be stupid trying to read the code
[11:44] <jdub> seb128: and later on, BBB blogged about rewriting it because the code was so horrific ;)
[11:44] <seb128> yes, the icons are nice
[11:45] <seb128> he he
[11:45] <seb128> jdub: no feedback for the moment, I hope than the profile selection (desktop/laptop) for the panel works
[11:46] <seb128> worked fine during my tests here ...
[11:46] <jdub> seb128: oh yeah, meant to say that was meant for gnome-session first startup rather than install 8)
[11:46] <jdub> seb128: but it's fine for now :)
[11:47] <seb128> what's wrong with doing it in the postinst ?
[11:47] <seb128> that's where we register the default config
[11:48] <jdub> nothing wrong
[11:52] <seb128> jdub: have you seen the problem with nautilus and vfat drives ? All the files are +x, so nautilus just try to run them and silently fails, so nothing happens
[11:52] <seb128> that's pretty bad ...
[11:53] <ddaa> jdub: .au keyboard have a [->]  key one them or what?
[11:54] <jdub> ddaa: as in the cursor keys?
[11:54] <jdub> seb128: haven't seen that
[11:54] <ddaa> as in " Actions  Recent Documents menu item"
[11:54] <jdub> seb128: but i can imagine that not too many people test-- hrm; of course they do -> cameras
[11:54] <jdub> ddaa: copy and pasted :)
[11:56] <sabdfl> seb128: are the wifi / battery percentages disabled by default now?
[11:57] <sabdfl> thom: what's the best way to disable most of the screensavers?
[11:57] <thom> sabdfl: no idea, am justa bout to look into it
[11:57] <seb128> sabdfl: apparently no, they should ?
[11:57] <sabdfl> seb128: correct
[11:57] <sabdfl> display only the icon
[11:57] <sabdfl> mouseover should give percentage
[11:58] <seb128> ok, will do the change in the next upload
[11:58] <sabdfl> thanks
[11:58] <sabdfl> jdub: wifi applet gives strange mousover messae when wifi signal is gone
[11:58] <sabdfl> "no wireless device"
[11:59] <thom> sabdfl: i take it that the list already on the bug is that same as the one you had in mind? (for .xscreensaver)
[12:01] <jdub> sabdfl: don't like that? (it's a string change...)
[12:01] <jdub> sabdfl: no percentages is hard to do with the battery applet
[12:01] <jdub> sabdfl: unless we show the graph + icon
[12:04] <jdub> (and currently, the graph doesn't grey out nicely on disabled)
[12:05] <sabdfl> thom: not quite, i've tuned to include some newer ones
[12:05] <sabdfl> jdub: graph being the big green monster?
[12:06] <jdub> sabdfl: yes 8)
[12:06] <sabdfl> let's go with mouseover for the moment
[12:06] <jdub> sabdfl: the power state icon gives no indication of time remaining
[12:06] <sabdfl> we'll get the guy to add graph-in-battery-icon for Hoary
[12:06] <sabdfl> so users will just see a richer pic in Hoary
[12:06] <sabdfl> am trying to keep screen real-estate precious
[12:07] <sabdfl> another idea - could these things become notifications in Hoary, rather than applets?
[12:07] <jdub> possibly
[12:07] <sabdfl> also, how do I make notifications extend to the left along the panel, rather than to the right of the widget
[12:08] <jdub> it should do that already
[12:08] <jdub> when there's a new nicon, the applet should expand
[12:08] <sabdfl> hmm... ok
[12:08] <jdub> (but it doesn't contract, at this stage, from memory)
[12:08] <sabdfl> and the grippy is just to give you something to rt-click on?
[12:08] <jdub> yeah
[12:08] <sabdfl> ok
[12:08] <sabdfl> well, we work with what we have :-)
[12:08] <sabdfl> it's looking good here
[12:09] <sabdfl> even with the b&w debugging icons
[12:09] <sabdfl> is andy actually running warty?
[12:09] <jdub> i thought he was doing an install the other day
[12:12] <ddaa> Probably not the right place to speak of it, but I think the launcher/applet/notification distinction is a bit bogus. For example, if you have a gaim launcher, you end up with two gaim icons in the panel. Once again Apple hit something with their "polymorphic" icons in the doc...
[12:13] <ddaa> i.e. the gaim launcher should also play the role of the notification
[12:13] <ddaa> and solving it using an applet is (slightly) counter-intuitive.
[12:14] <jdub> ddaa: totally
[12:15] <ddaa> Similarly, the trash applet is actually a launcher+notification.
[12:15] <jdub> ddaa: that whole area needs serious improvement in gnome (and not just random guesswork)
[12:15] <sabdfl> ddaa: so ideally, gaim is just gaim. if its running, clicking that thing brings it to the focus. if it isn't, clicking it starts it up?
[12:15] <jdub> ddaa: i started writing an analysis of osx's execution model a while back
[12:16] <jdub> ddaa: which i hoped would fuel some more work on this, but it's a really huge lump of work to fix properly
[12:16] <ddaa> sabdfl: yes, that's the idea. And if it is launched, right-clicking pops the notification's menu.
[12:17] <ddaa> jdub: nice to see that upstream is thinking about it :)
[12:19] <thom> we need to make sure the notification icon is the same as the real icon for gaim, as well
[12:20] <jdub> thom: well, ddaa is talking about making them one and the same object :)
[12:21] <thom> jdub: yes, and that's a viable goal for warty :P
[12:22] <ddaa> btw, who is working on the compositing manager support for warty? :P
[12:30] <pitti> thom: did you claim yesterday that gksudo did not segfault at your machine?
[12:30] <thom> pitti: i did indeed
[12:30] <pitti> thom:  I did not hallucinate. It took me a few hours, but I know the reason
[12:31] <pitti> thom: the f**ing beast reads 256 bytes into a 16 byte buffer
[12:31] <thom> oh?
[12:31] <thom> WHAT?
[12:31] <pitti> thom: maybe you just were lucky
[12:31] <pitti> thom: you should have never get any error messages, did you?
[12:31] <pitti> thom: gksudo reads the output of sudo
[12:32] <thom> yar
[12:32] <thom> that's somewhat disturbing
[12:33] <pitti> I've got an appointment for my future health insurance, returning in some hours
[12:33] <Mithrandir> it's fucking crackful
[12:52] <debianist> meeting?
[12:52] <debianist> what about?
[12:54] <debianist> many new boot messages i don't recognize from debian are when ubuntu boots first time after install
[12:55] <debianist> clock synchronization for pool.ntp.org failed
[01:11] <debianist> ok, my ubuntu QEMU is installed.
[01:12] <debianist> I didn't get no X and gnome,
[01:12] <debianist> oh
[01:12] <debianist> i used "custom" that figures.
[01:12] <debianist> :)
[01:12] <debianist> no vim ?
[01:13] <Kamion> fixed last night
[01:13] <jdub> debianist: bug in current versions, was fixed about 10 hours ago
[01:13] <Kamion> we used to be looking for vim-tiny, which doesn't exist
[01:13] <debianist> i se
[01:13] <debianist> i see
[01:13] <fabbione> hey Kamion 
[01:13] <debianist> ok
[01:13] <Kamion> "morning" fabio
[01:13] <jdub> morning Kamion 
[01:13] <debianist> it didn't ask me no administrator password on setup
[01:13] <debianist> what is it?
[01:14] <jdub> debianist: root is disabled
[01:14] <jdub> debianist: use sudo
[01:14] <jdub> debianist: ideally, read the FAQ on the wiki :)
[01:14] <debianist> jdub : on my way ;)
[01:20] <fabbione> Kamion: i am testing the new base_config right now
[01:27] <fabbione> Kamion: base-config works on netinstall, and it should work 99.9% on cdrom install. the 0.1 is just because i couldn't test it live
[01:29] <Kamion> heh
[01:29] <Kamion> well, I'll be testing it soon enough anyway
[01:29] <fabbione> we will still miss security
[01:30] <fabbione> but i really need the archive in place for that
[01:30] <fabbione> all the other stuff is added, but commented out.. so it's no point of failure
[01:37] <debianist> from the wiki :" You can run programs as root with sudo, for example: sudo apt-get update,"
[01:37] <debianist> it doesn't work.
[01:37] <debianist> i have to set up a root password before
[01:37] <debianist> which would enable the root account
[01:37] <jdub> what happens when you type sudo apt-get update?
[01:38] <debianist> jdub : Password:_ 
[01:38] <Kamion> debianist: *your* password, not root's
[01:38] <jdub> debianist: type in *your* password
[01:38] <fabbione> debianist: enter your password
[01:38] <lucas_> hi
[01:38] <seb128> your password :p
[01:38] <Mithrandir> jdub/kamion: that should probably be explained somewhere?
[01:38] <fabbione> hey lucas_ 
[01:38] <seb128> hey lucas_ 
[01:39] <debianist> you mean regular user password?
[01:39] <fabbione> lucas_: i was just waiting for you to show up :-)
[01:39] <Mithrandir> debianist: yes
[01:39] <thom> debianist: yes
[01:39] <fabbione> debianist: yes
[01:39] <jdub> debianist: yes, yours
[01:39] <lucas_> oh, you should have mailed me
[01:39] <seb128> debianist: yes
[01:39] <fabbione> lucas_: no rush :-)
[01:39] <debianist> hey hey, was that so dumb of me to ask?
[01:39] <lucas_> I could have shown up earlier
[01:39] <lucas_> ok
[01:39] <fabbione> lucas_: i need you to install xserver-xfree86-gdb and run a nice gdb session for that segfault
[01:40] <lucas_> ok
[01:40] <mjg59> Have you managed to sort the via crash yet?
[01:40] <fabbione> do you think you can handle it yourself?
[01:40] <lucas_> will do that down, I'm too sleepy to do something else
[01:40] <lucas_> yup
[01:40] <fabbione> mjg59: not until i will have a via chipset to work on
[01:40] <fabbione> lucas_: cool
[01:40] <fabbione> lucas_: just add the info to the bug...
[01:41] <lucas_> ok
[01:41] <lucas_> rebooting, then ...
[01:41] <fabbione> thanks :-)
[01:41] <fabbione> take the time you need
[01:41] <fabbione> argh...
[01:41] <debianist> seb128 : i thought it needs the root password
[01:41] <seb128> do you have the root password ? :p
[01:42] <Mithrandir> mjg59: what via crash?
[01:42] <mjg59> Mithrandir: XFree was deeply unhappy on the via graphics chipsetted laptops
[01:42] <fabbione> Mithrandir: ddcprobe on via chipset = X dies
[01:42] <Mithrandir> mjg59: ok, via graphics chipset, I don't think I have that
[01:42] <seb128> debianist: you only have one password, the system ask for a password ... give this one :)
[01:43] <fabbione> "thank you for choosing Ubuntu"
[01:43] <fabbione> ;)
[01:43] <fabbione> another installation success
[01:44] <debianist> seb128 : is it some sudo workaround? or it's default behavior regardless the distro it's on?
[01:44] <seb128> default behaviour
[01:46] <jdub> debianist: using sudo means you never have to use your root password
[01:46] <jdub> debianist: we go one step further by disabling the root password entirely
[01:47] <lucas_> fabbione: where is that xfree86-xserver-gdb ?
[01:47] <fabbione> dbg
[01:47] <fabbione> xfree86-xserver-dbg
[01:47] <lucas_> ok
[01:47] <lucas_> sorry
[01:47] <fabbione> no problem :D
[01:48] <fabbione> lucas_: before you rebooted i was going to write: 1) take the time you need 2) the same driver is in debian since yesterday
[01:48] <debianist> jdub : sudo wraps the root process ?
[01:48] <debianist> jdub : and let a regular user exectue?
[01:48] <seb128> man sudo
[01:49] <lucas_> ok
[01:58] <lucas_> re
[01:58] <lucas_> fabbione: it just crashes, it doesn't segfault
[01:59] <lucas_> (and I've checked I have the VideoRam option)
[01:59] <fabbione> ok.. but i guess you can get a backtrace and so on..
[01:59] <lucas_> crashes=screen go black
[01:59] <lucas_> like without the VideoRAM option
[02:00] <fabbione> ah
[02:00] <fabbione> that sucks
[02:00] <lucas_> I'll try to get a log of the crash to see if I can see something
[02:01] <fabbione> ok
[02:01] <lucas_> bbl
[02:09] <Kamion> wow, amusing base-installer bug
[02:10] <Kamion> totally broke on a CD with no kernel-image-* packages, only linux-image-* ...
[02:12] <fabbione> eh?
[02:12] <Oskuro> linux-image?
[02:12] <Oskuro> is that a new name?
[02:12] <jdub> yeah
[02:12] <fabbione> Kamion: how is that possible if it worked on netinstall?
[02:12] <Oskuro> jdub: just in ubuntu or also in Debian?
[02:12] <Kamion> fabbione: netinstall had the full archive to play with, which still contains kernel-image-*
[02:12] <Kamion> Oskuro: just in Ubuntu
[02:13] <Kamion> fabbione: it won't necessarily have *used* them
[02:13] <Kamion> fabbione: but it needed apt-cache search kernel-image | grep ^kernel-image to return true :-)
[02:13] <fabbione> ahhhh
[02:13] <fabbione> ok
[02:13] <Kamion> (fixed in the upload I just made)
[02:13] <fabbione> that make sense
[02:14] <fabbione> no.. it didn't use them
[02:14] <seb128> arg, daily iso is failing
[02:14] <Kamion> seb128: what's your failure?
[02:14] <seb128> error on the kernel package during the base installation 
[02:15] <Kamion> seb128: scroll up
[02:15] <fabbione> ehehe
[02:15] <seb128> ok :)
[02:15] <fabbione> you talk about the devil... and someone gets the horn
[02:15] <fabbione> +s
[02:15] <Kamion> you can keep your horn to yourself
[02:16] <fabbione> oh it's nothing bad..
[02:16] <fabbione> it's just a common way of saying in italy
[02:16] <fabbione> it's like when you talk about the weather and that it is not raining for a while, you can be sure that in 10 minutes there will be a hurricane
[02:16] <Kamion> "speak of the devil and he shall appear" is the saying here
[02:16] <Kamion> or variants
[02:17] <fabbione> yeah
[02:17] <jdub> seb128: dude, they're calling you satan. let's steal their pants.
[02:17] <seb128> erf
[02:18] <jdub> thom: did you happen to change any of the other defaults in xscreensaver?
[02:18] <jdub> 'cos some of the defaults are pretty cracky
[02:19] <seb128> grrrrr, I need to do a fresh install to debug some stuff
[02:19] <sabdfl> thom: looks like we need to update /etc/X11/app-defaults/XScreenSaver-gl right?
[02:19] <sabdfl> can I just send you a patch?
[02:20] <fabbione> seb128: net install is your friend :-)
[02:21] <seb128> my connexion is damn slow
[02:21] <fabbione> seb128: you only need to do the base install
[02:21] <fabbione> seb128: after that you can keep going from cdrom
[02:21] <seb128> and to use the CD after that
[02:21] <Kamion> seb128: there's another approach
[02:21] <seb128> yes
[02:22] <Kamion> seb128: after "loading installer components from CD" and before the base installation, edit /var/lib/dpkg/info/base-installer.postinst
[02:22] <Kamion> seb128: look for "linux-image", and you'll find two lines like this
[02:22] <Kamion>         (chroot /target apt-cache search kernel-image | grep ^kernel-image;
[02:22] <Kamion>          chroot /target apt-cache search linux-image | grep ^linux-image) | \
[02:22] <Kamion> seb128: change them to this:
[02:23] <Kamion>         (set +e;
[02:23] <Kamion>          chroot /target apt-cache search kernel-image | grep ^kernel-image;
[02:23] <Kamion>          chroot /target apt-cache search linux-image | grep ^linux-image) | \
[02:23] <seb128> ok, I'll try this, thanks
[02:23] <fabbione> hmm i just had a dejavu
[02:23] <thom> sabdfl: already uploaded
[02:24] <fabbione> the matrix is changing something around me
[02:24] <Kamion> sorry for the mistake, it was my bug
[02:25] <thom> jdub: such as?
[02:25] <jdub> thom: all of the image manipulation and diagnostics settings should be off
[02:26] <jdub> thom: (though we could supply a bunch of piccies for choose random image, but we don't have them yet)
[02:26] <jdub> thom: we could enable power management there
[02:26] <jdub> thom: the fade to/from black could both be on, set at 2 seconds
[02:26] <Kamion> elmo: is linux-kernel-di-powerpc-2.6 in NEW, or did it fail to build?
[02:27] <elmo> kamion: nothing in NEW, I checked this morning so I guess the latter
[02:27] <elmo> I'll go find details
[02:27] <Kamion> damn, oh well
[02:27] <jdub> thom: thoughts?
[02:27] <thom> jdub: fish.
[02:27] <Kamion> I was too sleepy when I uploaded that
[02:27] <jdub> thom: plus i don't think we really answered the screen locking question
[02:27] <thom> jdub: which screen locking question
[02:27] <elmo>   Depends             : linux-image-2.6.8.1-powerpc
[02:28] <Kamion> uh, crap
[02:28] <jdub> thom: whether we enable it by default or not
[02:28] <Kamion> ok, I just suck, I'll test it this time :/
[02:28] <Gman> jdub, you want a gtk2 screen locking patch so that jamie can get pissed off at ubuntu too?
[02:28] <thom> arh
[02:28] <thom> Gman: hell yes
[02:28] <jdub> Gman: that's actually sitting in our bug tracker too
[02:28] <Gman> jdub, heh, rock
[02:29] <jdub> Gman: i might help Riff finish off the gscreensaver patches
[02:29] <Gman> awesome
[02:30] <debianist> does ubuntu have a network install servers? my cdrom is not operating ..:(
[02:30] <Kamion> it's on the wiki
[02:30] <jdub> thom: do you want me to deal with those?
[02:31] <thom> jdub: you can deal with the screen locking
[02:31] <debianist> Kamion : there is such servce?
[02:32] <jdub> thom: well, that's unanswered, i'm more interested in the other settings
[02:32] <thom> agree with image manip, diag
[02:32] <thom> agree with fade
[02:32] <thom> think i agree with power management also
[02:33] <thom> doing now
[02:33] <jdub> oh, ok
[02:33] <jdub> thanks
[02:33] <jdub> was happy to do them, just wanted to raise it :)
[02:34] <thom> well, since i have the source and was in the dir, it seemed churlish not to
[02:34] <kagou> hi
[02:48] <thom> jdub: ok, what are reasonable defaults for power management?
[02:49] <jdub> standby 15, suspend 30, off 60?
[02:49] <jdub> or 30:45:60?
[02:50] <jdub> mmm, perhaps 30:45:60, because most people's screensaver timeout will be between 10 and 30
[02:50] <thom> ack
[02:50] <thom> agreed
[02:51] <thom> we need to leave grab desktop images on for antinspect and antspotlight
[02:51] <Kamion> elmo: ok, should be fixed now
[02:52] <jdub> thom: desktop images on == information leak
[02:52] <jdub> thom: (a selection of random images would be nice)
[02:52] <thom> meh
[02:52] <thom> done
[02:55] <jdub> justdave: the bottom bar in bz looks like baby spew!
[02:55] <jdub> justdave: btw, bz.org looks tops
[02:56] <ploum> where can I find a good tuto to learn how to use signal with gstreamer ?
[02:56] <jdub> ploum: pretty sure gstreamer.net links to their docs
[02:56] <jdub> ploum: and #gstreamer is on this network, from memory
[02:57] <ploum> jdub, I'm sorry ! I wrote in the wrong tab
[02:57] <ploum> !
[02:57] <jdub> heh
[02:58] <ploum> #gstreamer is next to this tab..
[02:58] <ploum> Argh !
[02:58] <ploum> (I'm better than before. I only missed the good one for a few pixel, no more)
[02:58] <ploum> sorry for the noise
[03:12] <jdub> haha
[03:12] <jdub> dudes
[03:12] <jdub> oh man
[03:12] <jdub> http://www.biosmagazine.co.uk/images/content/prodpics/200409/i-Select-M4610_3.jpg
[03:12] <jdub> the NEC i-Select M4610
[03:13] <jdub> seen one of those recently? :)
[03:16] <Kamion> heh
[03:26] <debianist> can anybody tell if there is a boot server for installing ubuntu? I don't mean I'll set up one,'d like to know if canonical has one
[03:28] <Kamion> we don't have a netboot server (you can't do that quite so remotely); the files you need to set up your own are in the warty archive though
[03:28] <debianist> oh
[03:28] <debianist> any docs on the wiki for how to set it up?
[03:30] <debianist> btw, the meeting has started already?
[03:30] <debianist> oh 3 mour hours before the meeting
[03:42] <debianist> is there also floppy installation for ubuntu? 
[03:54] <thom> sabdfl: i'm concerned about the number of GL screensavers you have in the current list. they will look awful without acceleration...
[03:54] <elmo> yeah, we should hax0r xscreensaver to disable them if there's no dri
[03:57] <seb128> look awful and eat resources
[03:57] <debianist> in order to resize an already existing ext3 journaling fs in d-i of ubuntu i just change it's size on the partman edit partition screen?
[03:58] <fabbione> i guess sabdfl will buy 3d cards to everybody :P
[03:59] <elmo> does xscreensaver stop doing stuff when it goes into powersave mode btw?  [my monitor, gfx card and X don't get along and like to pretend DPMS doesn't exist so I can't check myself] 
[03:59] <debianist> huh?
[04:00] <sabdfl> they won't run if you remove screensaver-gl
[04:03] <thom> elmo: i believe so
[04:07] <thom> personally, i think we should just enable bouncing cow
[04:15] <thom> and on a non-jovial note, i think a fairly small selection of default screensavers, such as we have with my last upload, is reasonable
[04:17] <dieman> hah
[04:17] <dieman> bouncing cow!
[04:18] <dieman> i personally think that the apple ] [ pong mode should be the only screensaver.
[04:18] <fabbione> ROTFL
[04:21] <sabdfl> thom: list is complete now
[04:22] <sabdfl> bouncing cow included :-)
[04:22] <fabbione> i never saw it before.. it's lovely
[04:23] <sabdfl> 62 screensavers
[04:24] <sabdfl> 44 of them GL
[04:24] <sabdfl> phew
[04:24] <sabdfl> how good is evo 2 offline more?
[04:24] <sabdfl> s/more/mode/
[04:25] <thom> sabdfl: ... as awful as its online mode?
[04:26] <thom> says the committed mutt junky
[04:26] <sabdfl> mutt doesn't seem to like the number of folders I have, nor the latency of my mailserver link
[04:30] <sabdfl> why does read-edid show up as obsolete?
[04:30] <thom> ah. i have the latter fixed to some extent, the former is trickier (i only have 50 or so folders)
[04:30] <thom> because we don't use it
[04:30] <sabdfl> ok, thanks
[04:32] <Kamion> xresprobe pretty much replaced read-edid for our purposes
[04:36] <daniels> er, no
[04:36] <daniels> ddcprobe pretty much replaced read-edid for our purposes
[04:37] <daniels> if you call xresprobe with a laptop (and, if you're on powerpc, ddc must not produce any resolutions), then you need X to be installed and usable
[04:37] <daniels> if you can't guarantee that, you need to use ddc only and /usr/share/xresprobe/ddcprobe.sh will give you nice, defined output for a ddc probe
[04:38] <Kamion> daniels: the xresprobe package, I mean
[04:38] <daniels> also, disabling gl screensavers per default is pretty sane
[04:38] <daniels> i occasionally turn up home (or wake up) to a locked computer, and it's only ever occurred with long-running gl screensavers, not 2d
[04:38] <daniels> Kamion: right
[04:38] <Kamion> sabdfl: hm, you actually have more folders than me?
[04:38] <Kamion> then again I don't use mutt over IMAP
[04:38] <daniels> Kamion: well, note the above if you ever need to use xresprobe for anything installer-wise or whatever :)
[04:40] <Kamion> daniels: I plan not to touch X if I can help it, but thanks :)
[04:41] <daniels> Kamion: BIG SCARY GRAPHICAL INSTALLER WOOOOOOOOOOOOOOOO
[04:42] <daniels> sweet mother of god, our tetex setup appears to be broken now :\
[04:42] <daniels> Mithrandir: i'm melltinnnnggggg!
[04:42] <daniels> Kamion: YOU ONLY MAKE ME STRONGER
[04:42] <daniels> Kamion: I AM THE LURE OF SHINY CLICKY GOODNESS
[04:42] <Kamion>   * If /etc/apt/sources.list already contains a cdrom: URI on a new
[04:42] <Kamion>     base-config run (i.e. archive-copier ran 'apt-cdrom add'), then skip all
[04:42] <Kamion>     CD probing.
[04:42] <Kamion> bets on this working?
[04:44] <daniels> Kamion: same as england beating australia in any sporting ocntest within the next week
[04:44] <daniels> Kamion: <subliminal>shinygraphicalinstaller</subliminal>
[04:44] <Mithrandir> daniels: sure, we'll make d-i shinygraphical
[04:44] <Kamion> oh dear
[04:44] <Mithrandir> but not today
[04:45] <fabbione> Kamion: did you check if base-config will keep adding all the apt lines?
[04:45] <fabbione> Kamion: otherwise i am happy with it :-)))
[04:46] <Kamion> fabbione: shouldn't affect that piece of code
[04:46] <fabbione> Kamion: ah hold on a sec.. there is no cdrom: uri in first install
[04:46] <Kamion> there is now
[04:46] <fabbione> Kamion: there is file:///cdrom
[04:46] <Kamion> don't worry, I know what I'm doing here :)
[04:46] <fabbione> oh... did you update menu/apt-setup to match for it?
[04:46] <Kamion> archive-copier (0.0.3) warty; urgency=low
[04:46] <Kamion>   * Add prebaseconfig script to run 'apt-cdrom add' if packages were copied.
[04:46] <Kamion>  -- Colin Watson <cjwatson@canonical.com>  Tue,  7 Sep 2004 13:08:36 +0100
[04:46] <Kamion> yes
[04:47] <Kamion> the prebaseconfig script in archive-copier will replace the file:///cdrom/ entry with a cdrom: URI
[04:48] <fabbione> Kamion: rocking!
[04:49] <fabbione> too bad i have to go in like 2 minutes ago
[04:49] <fabbione> but i will test it tomorrow
[04:49] <fabbione> cya guys
[04:50] <sabdfl> does evo have a threaded view?
[04:51] <daniels> sabdfl: yah, it's under the view menu
[04:51] <daniels> might be triggered by ctrl-t
[04:51] <daniels> fabbione: seeya
[04:51] <daniels> so we're down fabio and jeff -- ho hum
[04:52] <sabdfl> oh horrors
[04:52] <sabdfl> is there a keyboard-only way to file/move a message to a folder?
[04:53] <daniels> it's not ctrl-m, is it?
[04:53] <daniels> istr coming up against this problem when i was using it locally
[04:53] <daniels> sabdfl: ctrl-shift-v
[04:58] <daniels> argh! i hate latex. so broken. :\
[04:58] <daniels> anyone here know latex much?
[04:58] <Mithrandir> how so?
[04:58] <daniels> Mithrandir: can't build t1-cyrillic from stock warty
[04:59] <daniels> Mithrandir: fmtutil complains that it can't find fmtutil.cnf, yet it's there, and running kpsewhich --format='web2c files' fmtutil.cnf, by hand works
[04:59] <daniels> seems maybe exporting all the variables breaks it?
[04:59] <Mithrandir> FTBFS from the package?
[04:59] <daniels> ftbfs by hand
[04:59] <daniels> i[e. download source, satisfy buuild-deps, debuild
[05:00] <Mithrandir> you satisfy pfa with fontforge?
[05:01] <daniels> yeah, and change the two files that run wifh pfaedit to use fontforge
[05:01] <daniels> but it fails in calling latex(1)
[05:03] <Mithrandir> daniels: how do you handle tetex-src ?
[05:04] <daniels> Mithrandir: um, apt-get install tetex-source, or whatever it was
[05:04] <daniels> i ran debuild, it complained about build-deps, so i installed them all
[05:04] <daniels> thom: i don't think seb can quite fix that ;)
[05:05] <Mithrandir> E: Couldn't find package tetex-src
[05:05] <daniels> argh, universe :\
[05:06] <seb128> thom: doing a new today's iso installation right now with your locale ...
[05:06] <Kamion> daniels: you mirror universe?
[05:06] <thom> seb128: once you experience the power of real english, you'll never want to leave :-)
[05:06] <daniels> Kamion: ya-huh
[05:07] <daniels> thom: i saw the keyboards and fled straight back, dude
[05:08] <Mithrandir> lamont: will you hate me very much for the ooo-amd64 package?
[05:08] <daniels> Mithrandir: not just lamont
[05:08] <Mithrandir> or rather, will your network connection hate me?
[05:09] <lamont> Mithrandir: does it deliver any binaries into i386?
[05:09] <daniels> Mithrandir: please tell me it's .*_amd64\.deb
[05:09] <Mithrandir> it's an amd64.deb, yes.
[05:09] <lamont> daniels: I suspect that source is large too
[05:09] <Mithrandir> but it's a big chunk of source.
[05:09] <daniels> ... 300MB of source?
[05:09] <Mithrandir> daniels: around that size, yes.
[05:09] <daniels> sweet mother of god.
[05:10] <daniels> why must you taunt me?
[05:10] <lamont> daniels: pay no attention to the cluster inside the build process...
[05:10] <thom> i should fix mine to exclude ppc now
[05:10] <daniels> lamont: ... dude.
[05:11] <daniels> lamont: does anything you have actually work as the creator intended? ;)
[05:11] <lamont> daniels: these aren't the droids you're looking for.... they can pass....
[05:11] <lamont> daniels: better to have a consistant Packages file, than part of an archive.
[05:11] <daniels> lamont: yeah, i think it's best if i let it go without thinking about it too much :P
[05:11] <daniels> bbiab
[05:11] <lamont> or rather, a Packages file that claims the mirror has files it doesn't.
[05:12] <mdz> morning
[05:12] <seb128> hey mdz 
[05:12] <lamont> (my main mirror is all binaries, minus *-dbg, universe is just a few packages)
[05:12] <Kamion> morning mdz
[05:12] <lamont> morning mdz
[05:12] <seb128> thom: ok, not for today, I can't get a warty installed for 2 days now
[05:13] <seb128> thom: will try with next iso
[05:13] <Kamion> just finishing this cdrom-detect test, then when that's built I'll upload new debian-installer, then when that's built and byhanded I'll build new isos ...
[05:14] <seb128> ok, please let me know
[05:14] <Kamion> with any luck I'll be able to release those
[05:14] <seb128> when it's ready
[05:14] <lulu> hey guys - please vote on the Canonical logo - see email and wiki https://www.warthogs.hbd.com/LogosAndBranding - thanks :o)
[05:15] <lamont> lulu: again? :-)
[05:16] <Mithrandir> daniels: hints.awk fails to run here.
[05:16] <lulu> oops - sorry wrong channel.. :o(
[05:20] <Kamion> mdz: if you wanted to remove pppoe, then pppoeconf will need to go too
[05:21] <mdz> Kamion: it shouldn't
[05:21] <mdz> according to md, pppoeconf can configure ppp-with-pppoe-support also
[05:21] <mdz> and the dependencies seem to reflect this
[05:21] <mdz> oh, our version is older
[05:21] <mdz> unstable: Depends: whiptail-provider | whiptail, ppp (>= 2.4.2+20040428-2) | pppoe (>= 3.0), ppp (>= 2.4.1.uus2-4)
[05:21] <mdz> wartylog: Depends: whiptail-provider | whiptail, pppoe (>= 3.0), ppp (>= 2.4.1.uus2-4)
[05:23] <lamont> Mithrandir: no OO-amd64 source even.  You may live.
[05:23] <Mithrandir> lamont: :)
[05:23] <lamont> (syncing locally, that is.)
[05:24] <Mithrandir> I haven't uploaded it yet.
[05:24] <lamont> but no content in binary-i386/Packages, yes?
[05:25] <thom> mdz: g'morning
[05:25] <Mithrandir> lamont: it's arch: amd64, all of it.
[05:26] <lamont> Mithrandir: you may live.
[05:26] <lamont> :-)
[05:26] <Mithrandir> hooray!
[05:26] <kagou> hi
[05:28] <mdz> Kamion: is this a problem for the CD build?
[05:28] <mdz> I'm going to upload a new pppoeconf
[05:28] <debianist> what happens on the "starting ubuntu" prompt?
[05:30] <Mithrandir> kernel is silent
[05:30] <debianist> has the warty-iso image changed lately? since last night?
[05:30] <debianist> (trying to figure if i'd better download a new image)
[05:30] <kagou> Mithrandir, do you mean that the silent came from "quiet" parameter in grub menu config ?
[05:30] <Kamion> mdz: not too bothered
[05:30] <Mithrandir> probably
[05:34] <debianist> why is when rebooting ubuntu, after it started stopping services it suddenly print out a bunch of  "Loading <a_service>" messages?
[05:35] <debianist> if any, it should be "Unloading <a_service>" ?
[05:36] <mdz> debianist: file a bug with the specific text that you see
[05:36] <debianist> mdz : k
[05:40] <debianist> tried to manually edit a ext3 partition, choose new size. waited 10 secs. (partition is with bas install only) come back to partitons data, see partition size unchanged. bug?
[05:40] <debianist> part size was = 939MB
[05:40] <debianist> new size requested = 539MB
[05:41] <kagou> debianist, you try with d-i 
[05:41] <kagou> ?
[05:41] <kagou> or fdisk ?
[05:42] <debianist> d-i , e.g. partman 
[05:42] <debianist> which is a frontend to parted
[05:42] <debianist> right?
[05:42] <Kamion> yes
[05:42] <Kamion> there used to be a bug like that but I thought it was gone
[05:42] <Kamion> I'll see if I can reproduce it next time I have the opportunity; in the meantime, please file a bug
[05:43] <debianist> ok, my 2 first bugs on their way. :) wooha i'm thrilled :)
[05:45] <daniels> oh dear
[05:45] <daniels> that turkish soap made everything else look fantastic in comparison
[05:49] <Kamion> the way the manual takes ages to build, you mean?
[05:49] <lamont> nah, the way it's always out of date for my mirror script, because of the daily builds.
[05:50] <Kamion> the daily builds might as well not build debian-installer-manual, if we can figure out how to suppress that; it's a waste of time and space
[05:50] <debianist> Kamion : tag it under parted, or partman ?
[05:50] <Kamion> debianist: partman
[05:50] <Mithrandir> is it 24h-daily build or 30m-daily build?
[05:50] <Kamion> Mithrandir: 24h
[05:50] <lamont> crontab daily, 24h
[05:50] <lamont> 0605 british time
[05:50] <debianist> Kamion : what shall I use in perioriy?
[05:50] <lamont> Kamion: not worth the effort, I expect
[05:50] <debianist> Kamion : what shall I use in priority?
[05:51] <Kamion> debianist: normal will do for now
[05:51] <mdz> seb128: the new gnome-vfs2 seems to work perfectly
[05:51] <seb128> cool :)
[05:51] <debianist> P2 ?
[05:51] <Kamion> don't worry about it
[05:51] <Mithrandir> poor buildd, has to get up that early
[05:51] <Kamion> if you don't know, choose the defaults
[05:51] <daniels> nine minutes ...
[05:52] <lamont> Mithrandir: that's just so that it's done before Kamion crawls out of bed, you ,now..
[05:52] <mdz> seb128: my primary remaining concern with the automounting stuff is how to handle unmounting
[05:52] <lamont> Mithrandir: as a plus, it's generally before I go to bed...
[05:52] <mdz> seb128: the unmount option under the Computer icon is too obscure; I don't think users will find it
[05:52] <Mithrandir> :)
[05:52] <Kamion> lamont: also before the daily CD image builds crawl out of bed, which IIRC are at the time they are in order to suit fabbione and me ...
[05:53] <pitti> mdz: I don't think that it is too hard to attempt umount when closing the device window
[05:53] <seb128> mdz: any idea of how to do it in a better way ?
[05:53] <mdz> pitti: should we mount with the sync option to minimize the damage when they don't unmount?
[05:53] <pitti> mdz: we already do
[05:54] <pitti> mdz: however, if they don't umount and plug it in again, the "old" device is still mounted. This might break something
[05:54] <pitti> mdz: with sync it's annoyingly slow, but that's better than waiting on umount indefinitively
[05:54] <mdz> pitti: if they unplug and replug, it will show up as a different device
[05:54] <pitti> mdz: when users see that copying lasts long, they will understand
[05:55] <mdz> it can't free the old device until it is no longer in use
[05:55] <pitti> mdz: oh, right
[05:55] <pitti> mdz: so it actually shouldn't hurt too much on VFAT filesystems
[05:55] <mdz> pitti: no; it should be the same as Windows with a floppy
[05:55] <pitti> mdz: nevertheless, a slightly more obvious umounting would be better
[05:55] <pitti> mdz: especially since you cannot just take out a CD-ROM if it's mounted :-)
[05:55] <daniels> do we have right-click unmounting (context menu from the icon of whatever it is that's mounted)?
[05:56] <pitti> mdz: okay, you can, but...
[05:56] <pitti> daniels: yes, we have
[05:56] <lamont> Mithrandir: news on e-b front?
[05:56] <daniels> rad
[05:56] <mdz> pitti: for FAT filesystems, it should be possible to attempt to unmount _after_ the user disconnects the device
[05:56] <pitti> daniels: but, that's still pretty obscure
[05:56] <Mithrandir> lamont: seb128 was going to upload it.
[05:56] <mdz> pitti: this should work fine on a FAT filesystem which was mounted with sync
[05:56] <lamont> ok.  just curious
[05:56] <pitti> mdz: so gvm should intercept that and just umount it if hal reports the umount
[05:56] <mdz> pitti: though, on some other type of filesystem, it will probably spew errors and such, but it is already too late to do anything useful
[05:57] <seb128> Mithrandir: I've uploaded epiphany some days ago
[05:57] <pitti> mdz: that's why we should do that consistently
[05:57] <mdz> seb128: it would be very nice if it could trigger an unmount when the user closes all active nautilus windows on the volume
[05:57] <pitti> mdz: we need to unmount cd-roms anyway
[05:57] <mdz> seb128: is that something which would be complex to implement?
[05:57] <debianist> mdz : can't we work out some auto unmounting scheme?
[05:58] <pitti> umount as soon as lsof does not report anything anymore?
[05:58] <seb128> mdz: not sure, nautilus keeps some monitor on files
[05:58] <lamont> seb128: epiphany-browser_1.3.8-0ubuntu3 is missing an amd64 build-dep on gcc-3.4
[05:58] <seb128>  epiphany-browser (1.3.8-0ubuntu3) warty; urgency=low
[05:58] <seb128>  .
[05:58] <seb128>    * Patch from "Tollef Fog Heen <tfheen@debian.org>" to use gcc-3.4 on amd64
[05:58] <seb128>      (Warty: #1031).
[05:58] <thom> pitti: only if you magically automount as soon as someone tries to access it again :-)
[05:58] <seb128> lamont: I've applied the patch, dunno for the rest
[05:59] <mdz> pitti,debianist: the implementation should be simple, but I am not sure in which daemon it should be placed
[05:59] <Mithrandir> *sigh*, did I forget to update the control file there as well?
[05:59] <pitti> thom: but it is still "used" if the device nautilus window is still open
[06:00] <mdz> ok, meeting starting in <1 min
[06:00] <debianist> mdz : there should be an unmounting daemon
[06:00] <pitti> mdz: 18:00 at my clock, any laudatio from you for the start? :-)
[06:00] <seb128> mdz: why not using automount/autofs if we want to do an auto umount ?
[06:00] <debianist> the meeting begins?
[06:00] <mdz> seb128: we can look at it, but I think it would be too complex for Warty
[06:01] <Mithrandir> seb128: care to add that build-dep and upload?
[06:01] <seb128> Mithrandir: ok, I'll do it
[06:01] <debianist> the idea of timeout for unmounting seems rather unattractive having tried such a setup 
[06:01] <debianist> using autofs
[06:01] <pitti> IMHO an umount attempt when the Nautilus window is closed is not the worst idea; it's natural
[06:01] <seb128> not really
[06:01] <pitti> but a warning should be printed if something still uses the device
[06:01] <debianist> can't we attempt unmount when a cdrom eject int is detected or something close?
[06:02] <seb128> I don't expect to get my cdrom mounted/unmounted each time I browse the dir
[06:02] <mdz> so, first order of business, let's review the few remaining features which we need to nail down as soon as possible to prepare for the preview release
[06:02] <pitti> seb128: why not really? the window opens on plugin automatically, it should umount at window close
[06:02] <mdz> one of those, we were already discussing: the pmount/gvm integration
[06:02] <mdz> I think the last remaining showstopper there is what to do about unmounting
[06:02] <debianist> pitti : sounds correct. however, wouldn't you like to get it back notime when need arise?
[06:03] <Kamion> it still seems to be an active discussion; perhaps put it on hold until after the meeting?
[06:03] <seb128> pitti: because my cdrom spin each tome
[06:03] <seb128> time
[06:03] <daniels> mdz: the gimp thing?
[06:03] <seb128> pitti: and that's really annoying
[06:03] <mdz> daniels: gimp thing?
[06:03] <daniels> mdz: getting the bug number now -- basically, sanitising the first startup
[06:03] <seb128> pitti: I want to keep my cd mounted
[06:03] <daniels> mdz: aha, #1012
[06:03] <mdz> Kamion: let's see if we can put our heads together and come up with a strategy for it, and then it can go offline
[06:03] <Kamion> daniels: I think that's a bug fix :-)
[06:03] <Kamion> mdz: ok
[06:04] <mdz> what's the natural action for a user to tell the system "I'm finished with this"?
[06:04] <mdz> it seems like closing the window is that
[06:04] <pitti> I agree
[06:04] <debianist> mdz : for cds, i reckon it's eject
[06:04] <mdz> but the problem is that it's impossible to get it back afterward
[06:04] <mdz> (from the desktop)
[06:04] <pitti> you have to replug
[06:04] <mdz> yes
[06:04] <pitti> for cdroms, you can do it from the computer window
[06:04] <mdz> and they have already done that once, so they know that will work
[06:04] <Kamion> pitti: that may be very difficult
[06:04] <thom> which is very unintuitive, and will generate *millions* of bug reports
[06:05] <thom> "i closed a window, and the cdrom went away!"
[06:05] <daniels> yeah
[06:05] <pitti> the cd-rom doesn't, it's in fstab
[06:05] <pitti> just the usb devices
[06:05] <debianist> isn't there a cool way to detect the eject button and unmount after it?
[06:05] <daniels> and the first thing i'd do if a nautilus window popped up when i inserted a cd would be to close it, most of the time
[06:05] <Kamion> pitti: I'm sure PowerMac CD-ROMs aren't the only devices which require software eject, though
[06:05] <seb128> mdz: closing the windows, ok ... but what about the browser mode ?
[06:05] <Kamion> if you're the type who's annoyed by random windows popping up, you will ...
[06:05] <seb128> mdz: I select an another dir in the tree
[06:05] <pitti> Kamion: ZIP drives do as well
[06:05] <mdz> debianist: yes, but it's a kernel-level thing and can't be implemented in time
[06:05] <thom> debianist: eject umounts for you.
[06:06] <mdz> we need something that we can have for the preview release
[06:06] <debianist> yes
[06:06] <Kamion> debianist: what eject button? :)
[06:06] <debianist> mdz : eject buttons that will appear on the desktop
[06:06] <debianist> Kamion : mdz has answered me 
[06:06] <debianist> Kamion : I meant the eject buttons on the CD rom itself
[06:06] <pitti> desktop would be cool. a small icon in the panel, you click on it, it's unmounted
[06:06] <mdz> well, originally, nautilus would cause an icon to appear on the desktop when a new device was mounted
[06:06] <Kamion> debianist: merely observing that not all CD drives have such a button
[06:06] <mdz> correct?
[06:06] <pitti> I think Windows has a similar icon
[06:07] <thom> mdz: *nod*
[06:07] <mdz> but we disabled that
[06:07] <debianist> Kamion : right, idea dropped
[06:07] <thom> yup
[06:07] <debianist> Kamion : :)
[06:07] <debianist> but,
[06:07] <debianist> desktop eject icons for every device automounted
[06:07] <thom> pitti: windows sticks a notification icon in for such things, to let you eject them
[06:07] <mdz> the nice thing about that was that it provided a desktop presence which corresponded exactly to the mount
[06:07] <mdz> the icon appeared when the device was mounted, and the user, whether or not they had any nautilus windows (or any other windows) open, could go there to perform an operation on it
[06:07] <pitti> thom: never seen it first hand
[06:07] <mdz> the same way that CD eject works
[06:07] <thom> and i'm not sure that's  such a bad metaphor for us - you right click, eject, and it umounts and goes away
[06:08] <thom> can we extend the eject metaphor for usb devices etc?
[06:08] <pitti> so, right click on a desktop icon, not an icon in the computer window?
[06:08] <Kamion> I find that far preferable, but I can't remember why we removed the icons
[06:08] <Mithrandir> thom: drag it to trash, like MOS used to do?
[06:08] <daniels> thom: as long as I don't have my thumbdrives shooting across the room
[06:08] <debianist> pitti : yes
[06:08] <mdz> thom: we have; there is now an 'unmount device' [jargon warning]  option in the context menu
[06:08] <thom> daniels: ;D
[06:08] <daniels> i like the right-click eject/i'm done with this metaphor
[06:08] <debianist> not in the computer window, it would get lost for new peoplel
[06:08] <mdz> thom: but the user will never find it
[06:09] <thom> mdz: if they're using it to eject a cdrom already, why not keep that?
[06:09] <pitti> they will if an icon pops up at the desktop
[06:09] <thom> and just use it for all hotpluggable stuff
[06:09] <mdz> thom: because we've buried it
[06:09] <mdz> thom: Mark did not want the icons on the desktop
[06:09] <daniels> i'll probably get punched for this, but how about a little eject button sitting, say, in the top-right of the icon that they can click?
[06:09] <debianist> thom : you mean they already use it in windows?
[06:09] <pitti> MacOS X haves icons appear on the Desktop; I like that
[06:09] <thom> mdz: disagree. the icons are in the computer. 
[06:09] <daniels> (with a small explanatory confirmation dialog to prevent accidental death)
[06:09] <debianist> mdz : "Media Manager" window
[06:09] <thom> you go to the computer, select the media, eject it
[06:09] <debianist> mdz : a foler or some sort on the desktop
[06:10] <thom> i think that's a pretty powerful metaphor
[06:10] <thom> and a fairly obvious one
[06:10] <debianist> or "Computer"->"Media"
[06:10] <mdz> thom: they didn't go to the computer in the first place to access it, and the computer icon did not react when they used the device, and they've never seen this icon under computer before...
[06:10] <mdz> debianist: currently the icons appear under Computer
[06:10] <debianist> mdz : k
[06:10] <daniels> (is it necessary to have this battle again?)
[06:11] <mdz> ok, I think we have a rough idea of the issues
[06:11] <mdz> let's take this to the mailing list
[06:11] <mdz> so that we can move on
[06:11] <pitti> @all: can we sum up the issues and discuss that separately
[06:11] <pitti> mdz: agreed; who summarizes?
[06:11] <mdz> pitti: will you summarize the issues for the mailing list and start the discussion there?
[06:11] <pitti> yes
[06:11] <mdz> ok, thanks
[06:11] <mdz> moving on, another feature that needs to be wrapped up is the init script beautification
[06:11] <debianist> (reverted back to it from "Computer")
[06:11] <mdz> npmccallum: can you give us a brief status update on that?
[06:12] <daniels> debianist: then follow up pitti's upcoming mail to the list, please
[06:12] <debianist> ok,
[06:12] <debianist> noted.
[06:12] <mdz> how many packages remain that need to be modified?
[06:12] <npmccallum> basically, the fucntions are in place.  All the packages I know of have been modified.  there are probably others that don't have scripts that I need to be made aware of
[06:12] <mdz> npmccallum: all of the packages in desktop, or all in supported?
[06:13] <npmccallum> desktop
[06:13] <daniels> we could get someone with access to the archive to find every package containing /etc/init.d/.*
[06:13] <Kamion> daniels: don't need archive access, we have Contents-*.gz files on mirrors
[06:13] <Mithrandir> as a note, it would be nice if we could remove a lot of the superflous output from random programs.
[06:13] <mdz> npmccallum: you should almost be able to install every package in supported at once
[06:13] <daniels> Kamion: ah, rad
[06:13] <mdz> there will be a few conflicts, like exim/postfix
[06:13] <pitti> the hotplug output is just endless; can we trim that somehow?
[06:14] <Mithrandir> like, if fsck only printed the volumes, not the copyright notice.
[06:14] <npmccallum> Mithrandir: any output that you see that is superflous is a bug, please file
[06:14] <Mithrandir> npmccallum: will do.
[06:14] <npmccallum> pitti: I'm going to drop the kernel log level, its the only way to rid that stuff
[06:14] <mdz> debianist: if a module is being loaded and failing, it probably should not be loaded in the first place, which is a bug
[06:14] <Kamion> I'm still concerned about some of the error handling; I'd like to have some kind of review effort to make sure all the exit codes and things are right
[06:14] <pitti> npmccallum: you could disable it temporarily for hotplug
[06:14] <debianist> mdz : noted
[06:15] <npmccallum> pitti: that is a good idea as well, I think I'd prefer that
[06:15] <mdz> Kamion: I agree
[06:15] <npmccallum> pitti: I think mdadm also spews a bunch of kernel stuff
[06:15] <Mithrandir> npmccallum: dhclient.
[06:15] <mdz> based on the number of bugs which have been uncovered already, we need to thoroughly regression-test these
[06:15] <pitti> these neverending error messages of mdadm should go away
[06:15] <thom> devmapper is the most evil one
[06:15] <Mithrandir> thom: what devmapper msg?
[06:15] <mdz> thom: that should be gone with the new kernel
[06:15] <pitti> yes, devmapper, not mdadm
[06:16] <thom> mdz: oh, cool
[06:16] <mdz> thom,pitti: shame on you for not testing the new kernel :-)
[06:16] <thom> i'll check in a bit
[06:16] <pitti> mdz: Oh, its running, I just did not observe the latest boot
[06:16] <mdz> it incorporates the bd-claim patch, which lets evms work properly
[06:16] <npmccallum> Mithrandir: I'm not sure how to handle dhclient because it actually grabs the terminal so that you can ctrl-c dhcp
[06:16] <thom> mdz: i'm not an obsessive kernel spew watcher ;-)
[06:16] <pitti> mdz: powerpc images were only available from this mornign on
[06:16] <mdz> which was the root of those problems
[06:16] <daniels> how difficult is it to take the interface down if miitool says there's no link?
[06:16] <Mithrandir> npmccallum: modify dhcpclient to not spew a lot of what it does?
[06:17] <mdz> daniels: hoary difficult
[06:17] <thom> daniels: miitool lies on some network cards
[06:17] <npmccallum> Mithrandir: I'll look at it
[06:17] <daniels> thom: gar
[06:17] <daniels> mdz: point taken
[06:17] <npmccallum> mdz: we may want to talk about NetworkManager as well on the hoary timeline :)
[06:17] <thom> npmccallum: already on the wiki
[06:17] <mdz> npmccallum: it's on the proposed feature goal list on the wiki
[06:18] <thom> i even have packages mostly done ;P
[06:18] <mdz> so, we need to review and test the init script changes
[06:18] <mdz> who besides Kamion is comfortable with shell nuances and can assist with that?
[06:18] <thom> mdz: well, i have no bugs, so i might as well :-)
[06:19] <npmccallum> Please also file bugs on anything, I was away this weekend, but I'll be bug squashing a lot this week
[06:19] <pitti> I will look into it and contact Kamion if I have some updates
[06:19] <mdz> thom, npmccallum: ok, you guys work together to review and test what has been done so far, and make sure that the ongoing work has a minimal chance of new bugs
[06:19] <daniels> (as a largely useless aside, i have ndiswrapper packages ready that i'll throw up to people.n-n-y.com tomorrowish)
[06:19] <debianist> mdz : assist with what?
[06:20] <mdz> debianist: the init script beautification work
[06:20] <daniels> mdz: if we're talking about warty features -- ati's r4xx code drop?
[06:20] <debianist> mdz : it's output, or the code?
[06:20] <Kamion> the code
[06:20] <npmccallum> debianist: both actually
[06:20] <daniels> thom: does this holy water buy me an atheros or whatever wifi card?
[06:21] <mdz> npmccallum: I think probably the easiest way to incorporate some peer review would be for you to send your diffs to thom as you go
[06:21] <npmccallum> mdz, thom: np
[06:21] <pitti> npmccallum: can you cc me as well, please?
[06:21] <thom> sounds like a plan
[06:21] <npmccallum> pitti: sure
[06:21] <Kamion> npmccallum: I think reviewing the code is most important at the moment, to get the bugs squashed as early as possible; we can tweak the output after preview release
[06:21] <mdz> npmccallum: generating diffs for the ones which are already done might be trickier; where you still have a copy of the old package, you can use debdiff
[06:21] <mdz> Kamion: right
[06:22] <mdz> ok, moving on
[06:22] <pitti> cups admin web interface?
[06:22] <mdz> the next feature-ish thing that we're trying to wrap up is the kernel
[06:22] <pitti> the new prism2_usb drivers lack userspace support
[06:22] <mdz> this actually seems to be in rather good shape, but it is crucial that everyone start using it ASAP
[06:22] <debianist> mdz : I can overview it, see what can be left out, suggest things etc. I'd like to do it however with supervision by someone more experienced in ubuntu than me :) 
[06:23] <mdz> pitti: what kind of userspace support do they need?
[06:23] <thom> mdz: is there any chance we can get acpi_ibm and a recent ACPI drop? :-)
[06:23] <mdz> npmccallum: which?
[06:23] <pitti> linux-wlan-ng package, basically
[06:23] <thom> pitti: this is the lack of wlan-ng? right
[06:23] <pitti> mdz: iwconfig does not work with it, you need wlanctl-ng
[06:23] <mdz> thom: it's possible; let's discuss it on the list and CC herbert
[06:23] <debianist> mdz : just make sure I don't do horroble mistakes..;)
[06:24] <daniels> mdz: noting that drivers like prism54 and acx100 lack firmware, is it worth making a last-minute request to the manufacturers for a distribution license and shipping them in the non-free modules?
[06:24] <mdz> pitti: is the version in universe sufficient?
[06:24] <pitti> I'm using it, it works
[06:24] <pitti> since we don't have autosetup for these things anyway, we can as well leave it in universe
[06:24] <mdz> it doesn't look too scary; we could promote it to Supported
[06:24] <mdz> of course, those cards won't work in d-i, right?
[06:24] <pitti> I suppose
[06:24] <pitti> the modules is loaded automatically, but it's not set up
[06:25] <Kamion> mdz: well, I made an attempt to get firmware loading support going, but I doubt that it works
[06:25] <mdz> Kamion: apart from the firmware issues, is there a wlanctl-ng udeb?
[06:25] <Kamion> no
[06:25] <mdz> I don't see one
[06:25] <mdz> Kamion: I can look at firmware support with you; I have hardware I can test with
[06:25] <mdz> that shouldn't be too hard to get working
[06:26] <Kamion> good; the main tricky bit is that d-i doesn't use real hotplug, and the hotplug substitute that it uses is really just for PCMCIA
[06:26] <Kamion> I expect some interesting issues in hw-detect
[06:26] <mdz> it looks like we'll be using the existing linux-kernel-di infrastructure, since building the udebs from linux-source is going to be too complicated for our timeline
[06:26] <mdz> I don't think that's a big deal; the primary concern there for Warty is that when we do a security update, we don't need to update N packages, but only one
[06:27] <mdz> generally, security issues aren't significant in the installer environment, so that should be a special case
[06:27] <mdz> local root vulnerabilities are not really a big deal when there is only one user, on the console, and they are root :-)
[06:27] <Kamion> I think the installer updates can be done on a best-effort basis; I agree they don't seem crucial usually
[06:28] <mdz> any other questions or concerns about the kernel?
[06:28] <mdz> did I mention it needs MORE TESTING? ;-)
[06:28] <mdz> dieman: network issues?
[06:28] <dieman> (too busy to pay attenton to the meeting)
[06:28] <npmccallum> mdz: I had a wierd md bug after I installed the new kernel
[06:28] <dieman> mdz: first day of classes, network explodes.
[06:29] <npmccallum> mdz: I'll file something
[06:29] <mdz> npmccallum: ok, once it's filed, send an email to Herbert; he doesn't have a bugzilla account
[06:29] <debianist> kernel updates were from 2.6.7 to 2.6.8?
[06:29] <daniels> 2.6.8.1
[06:29] <mdz> npmccallum: Herbert Xu <herbert@gondor.apana.org.au>
[06:29] <pitti> Can everybody please test USB devices with the new kernel? My USB stick's partitions are not discovered sometimes
[06:29] <mdz> pitti: I had a problem on my desktop going from 2.6.7 to 2.6.8
[06:29] <debianist> pitti : I will
[06:30] <mdz> all of my USB devices quit working, "device not accepting new address"
[06:30] <daniels> mdz: can we please get some xfs love into the kernel? whether that's just ripping fs/xfs out of 2.6.9-rc1-mm1 (which wfm) or whatever; i'd just like to be able to use my system for an hour without an oops that sends most of my processes into D
[06:30] <pitti> when I stick it in, it sometimes says "unknown partition table"
[06:30] <mdz> the problem turned out to be that I was specifying acpi=off on the command line
[06:30] <mdz> and re-enabling ACPI fixed it up
[06:30] <daniels> mdz: (that's separate from the bd-claim issue)
[06:30] <pitti> mdz: I don't think that is a kernel bug, more a firmware bug
[06:30] <mdz> previously, it had had problems _with_ acpi, and now it has different problems _without_ it :-)
[06:30] <mdz> daniels: file bug, mail Herbert
[06:31] <daniels> pitti: 2.6.9-rc1-mm1 works fine with usb bluetooth, usb mass storage (iriver, media reader, digicam, notebook cradle thingy), keyboard/mouse, et al
[06:31] <daniels> pitti: but that's not our kernel, so ymmv
[06:31] <daniels> mdz: will do
[06:31] <pitti> daniels: I'll try
[06:31] <mdz> ok, moving on
[06:31] <pitti> daniels: but this happened also with 2.6.7
[06:31] <mdz> seb128: what remains for the desktop UI changes?
[06:31] <pitti> translations!
[06:31] <seb128> nothing AFAIK
[06:32] <mdz> seb128: ok, so Mark's requirements are met for Warty?
[06:32] <mdz> seb128: nothing else is unresolved there?
[06:32] <seb128> mdz: oh, not sure about the trash applet
[06:32] <debianist> the installer daily ships with 2.6.8.1 ?
[06:32] <debianist> currently?
[06:32] <seb128> Jeff wanted to removed it (me too)
[06:32] <seb128> not sure about Mark
[06:33] <mdz> Mark wants it in
[06:33] <seb128> it's not translated and has some issues, but it's usuable
[06:33] <tvon|x31> does anyone use the trash applet?
[06:33] <mdz> debianist: yes, it should
[06:33] <seb128> mdz: I'll add it again to the default config
[06:33] <pitti> I find two trash icons confusing
[06:33] <tvon|x31> yeah
[06:34] <pitti> one at the desktop, one in the lower panel
[06:34] <mdz> seb128: Mark said the issues were cosmetic; is that true?  I can live with it as long as it doesn't crash
[06:34] <seb128> mdz: cosmetic and translations missing yes
[06:34] <mdz> pitti: the idea was for the desktop trash icon to disappear
[06:34] <debianist> mdz : 4.10 is current enough?
[06:34] <tvon|x31> also sucks to miss the trash applet and hit the panel
[06:34] <mdz> debianist: everything is 4.10 :-)
[06:34] <tvon|x31> esp when 'trashing' images
[06:34] <seb128> but that's not a big problem, no need of text to drag stuff here and open it by clicking
[06:34] <mdz> debianist: daily/20040907
[06:34] <debianist> mdz : :)
[06:34] <mdz> tvon|x31: eek
[06:34] <tvon|x31> and its small and hard to see, so just about any icon obscures it when you are trying to drop into it
[06:35] <debianist> mdz : what about the current directory there?
[06:35] <mdz> I have to admit that I don't really use nautilus, so I haven't played with the trash applet much
[06:35] <debianist> mdz : is it falsy?
[06:35] <seb128> mdz: the current menu layout is ok for everybody ? Should be (I've changed it according to Jeff's mail 2 days ago), but I just want to check
[06:35] <pitti> in fact I just discovered the smallish icon in the panel just now
[06:35] <mdz> debianist: current should be a symlink to 20040907
[06:36] <Kamion> debianist: the 20040907 daily is broken, please leave it for a bit
[06:36] <debianist> Kamion : k
[06:36] <tvon|x31> the 'logout' stuff at the bottom of the computer menu is a big improvement
[06:37] <mdz> ok
[06:37] <daniels> the computer menu strikes me as feeling bloated now
[06:37] <seb128> mdz: perhaps some branding missing (in evolution ?) .. Jeff was supposed to check
[06:37] <mdz> daniels: it is a bit heavy; but it's not a showstopper
[06:37] <npmccallum> seb128: the menu looks great
[06:37] <daniels> whereas the applications menu looks a little too trimmed-down
[06:37] <mdz> seb128: is there a bug filed?
[06:37] <seb128> npmccallum: thanks
[06:37] <daniels> mdz: not at all, just a random peanut-gallery comment
[06:37] <tvon|x31> daniels: ya think? I like it small like that
[06:38] <daniels> i do really like run application in applications tho
[06:38] <npmccallum> Are we planning on removing the debian branding on the openoffice.org splash screens?
[06:38] <seb128> mdz: no
[06:38] <Kamion> be careful about openoffice.org, they have some interesting branding rules
[06:38] <Kamion> check their web site
[06:38] <seb128> npmccallum: Jeff was supposed to check all the branding stuff and let we know ...
[06:38] <npmccallum> seb128: ok, I just noticed it last night, just wanted to make sure it was brought up
[06:39] <mdz> seb128,npmccallum: please file bugs about that stuff so it isn't forgotten
[06:39] <seb128> ok
[06:39] <mdz> so, we need to do a security review of Warty
[06:39] <npmccallum> do @canonical.com email addresses work in bugzilla yet?
[06:40] <mdz> since we snapshotted unstable, and Debian doesn't have a rigorous procedure for tracking security issues in unstable, we need to make sure we don't have any known security issues
[06:40] <pitti> mdz: what's still missing, security-wise?
[06:40] <justdave> npmccallum: nobody's told me they do yet.
[06:40] <mdz> pitti: that :-)
[06:40] <lamont> mdz: how much is left
[06:40] <mdz> lamont: everything that wasn't already done for the sarge security review
[06:40] <lamont> right
[06:40] <justdave> elmo / thom: what's the status on that?
[06:40] <thom> chinstrap% nc fiordland 25
[06:40] <thom> fiordland.warthogs.hbd.com [82.211.81.145]  25 (smtp) : No route to host
[06:40] <mdz> we need to check that we have all of the things fixed which were discovered in the sarge security review
[06:40] <thom> heres a clue
[06:40] <elmo> I'll try and fix that tonight
[06:40] <justdave> that probably means they still don't. :)
[06:40] <pitti> mdz: ah; I will check that
[06:41] <mdz> pitti: ok, you will take responsibility for this?
[06:41] <pitti> mdz: I will
[06:41] <mdz> pitti: we also need to review Debian DSAs independently
[06:41] <mdz> pitti: and to the extent possible, CVEs
[06:41] <Kamion> might want to hit joeyh up for the script he used to start automating that
[06:42] <pitti> mdz: I will dig though the bugtraq archives
[06:42] <mdz> Kamion: we may not even have anything outstanding from his list, but we'll see
[06:42] <Kamion> sure, but I think the script was more generic
[06:42] <mdz> pitti: this seems like a huge project; there are many sources of this information and it is difficult to cross-reference against the packages included in Warty
[06:42] <Kamion> don't remember the details, there was mail on debian-release about it
[06:43] <mdz> the end result was that once he had the information about which version had the fix, he could re-run it periodically and see if that version was incorporated into sarge
[06:43] <pitti> mdz: I welcome other people to help me :-)
[06:43] <mdz> pitti: let's talk after the meeting about how to go about it
[06:43] <pitti> if anybody is willing to help, I can manage the whole issue
[06:43] <lamont> pitti: let me know what you want me doing.
[06:43] <pitti> mdz: agreed
[06:43] <mdz> lamont: thanks
[06:44] <mdz> it's going to be a lot of research, and should parallelize well
[06:44] <pitti> lamont: thanks, I will write a mail on ubuntu
[06:44] <doko> pitti: helping hand from me as well
[06:44] <mdz> you can each take a  year's worth of CVE at a time or something
[06:44] <debianist> pitti : me also :)
[06:44] <mdz> [Clint] : are you here?
[06:44] <mdz> I think I forgot to CC him on the meeting announcement
[06:45] <mdz> anyway, [Clint]  has done woody->warty upgrade testing for us, and there seem to be only minor issues remaining
[06:45] <mdz> inapporpriate debconf questions, etc.
[06:45] <pitti> mdz: plugdev membership comes to my mind
[06:45] <thom> and sudo 
[06:45] <mdz> pitti: yes, that is part of the larger question about providing the ubuntu desktop experience on upgrade
[06:45] <mdz> more so than a functional problem of the upgrade
[06:46] <pitti> right, without sudo all gnome admin stuff won't work
[06:46] <Kamion> partly that's an artifact of doing so much in base-config
[06:46] <mdz> pitti: well, they also won't have any of that installed :-)
[06:46] <Kamion> which in turn is an artifact of debconf being emphatically not a registry ... I wonder how much 'base-config new' after upgrade clears up
[06:46] <mdz> so there is a separate potential project to create a procedure and/or tool which will attempt to install the whole desktop setup
[06:47] <mdz> which would include the group memberships, standard set of applications (desktop task), etc.
[06:47] <Kamion> that sounds extremely close to base-config
[06:47] <Mithrandir> it might very well run base-config
[06:47] <mdz> it could very well
[06:47] <debianist> pitti : sudo is base right?
[06:48] <mdz> I was thinking that whatever form it took, it wouldn't require significant changes to stuff which is in Warty, and would be an external thing
[06:48] <pitti> debianist: admim
[06:48] <mdz> but if it actually used base-config, it might require changing base-config, ugh
[06:48] <debianist> pitti : hmmm..
[06:48] <pitti> debianist: admin and optional
[06:48] <debianist> need to change that than
[06:48] <Kamion> pitti: it's in warty base
[06:48] <mdz> which constrains it with regard to our release deadlines
[06:48] <Mithrandir> mdz: b-c + some upgrade-only stuff?
[06:49] <mdz> Mithrandir: my concern is that if it is going to be based on base-config, those base-config changes need to go into the release
[06:49] <Kamion> mdz: I don't think that's a big deal myself
[06:49] <mdz> anyway, let's take that offline for the meeting
[06:50] <mdz> next agenda item is the restricted component
[06:50] <mdz> which exists now, and has a package in it
[06:50] <Mithrandir> restricted =~ non-free?
[06:50] <mdz> Mithrandir: restricted <proper subset> non-free
[06:50] <pitti> which package? nvidia?
[06:50] <mdz> pitti: linux-non-free-modules (kernel stuff)
[06:51] <mdz> what we need to decide here is what we should add to it for the release
[06:51] <mdz> ati and nvidia drivers come to mind
[06:51] <daniels> nvidia +1
[06:51] <daniels> fglrx -1
[06:51] <Mithrandir> firmware stuff?
[06:51] <pitti> daniels: agreed
[06:51] <npmccallum> daniels: why no fglrx?
[06:51] <daniels> ati drivers are too much of a pain in the arse to do configuration with; it took me a good hour or so
[06:51] <mdz> Mithrandir: we agreed in Oxford that firmware stuff was OK for main
[06:51] <mdz> Mithrandir: which is one of the differences between debian non-free and warty restricted
[06:51] <Mithrandir> mdz: ok, I didn't pay attention or wasn't present, then.
[06:51] <pitti> tg3 firmware would be good
[06:52] <pitti> I've seen this card quite often
[06:52] <daniels> if anyone's smart enough to configure fglrx, then they don't need our hand-holding with the module; as much as i think it's essential, we just can't configure it near meaningfully enough out of the box to justify inclusion imo
[06:52] <mdz> firmware should be packaged with the kernel
[06:52] <mdz> we already have ipw firmware and atmel firmware in the linux-source package
[06:52] <mdz> if there's additional firmware we need which is not included in kernel.org kernels, mail herbert
[06:52] <mdz> pitti: will you take care of tg3?
[06:52] <npmccallum> daniels: I can buy that, especially because we support ati 2d fairly well
[06:52] <Mithrandir> isl3890 would be nice, if we can get permission to distribute it.
[06:52] <mdz> I have a tg3 card and it seems to work fine without firmware
[06:52] <pitti> mdz: I cannot test it currently
[06:52] <thom> daniels: there are existing debian packages for fglrx
[06:52] <Mithrandir> mdz: but not at full speed.
[06:52] <thom> and they work fine
[06:52] <doko> mdz: does this include the firmware for isdn hardware as well=
[06:53] <mdz> Mithrandir: I only have a 100mbit switch
[06:53] <npmccallum> what is happening with madwifi?
[06:53] <daniels> npmccallum: (except without 2d support for the current line of cards)
[06:53] <mdz> npmccallum: madwifi is in linux-non-free-modules, as I said above
[06:53] <pitti> mdz: if it works at 100mbps, it's probably okay, I think
[06:53] <Kamion> debian-cd needs several updates to cope with restricted; I'm looking through those now
[06:53] <Mithrandir> mdz: what about externally loaded firmware?
[06:53] <npmccallum> mdz: sorry, missed it
[06:53] <daniels> thom: sure they do, if you know how to configure it right. i just think it's too much of a configuration nightmare to bother cutifying the packaging
[06:53] <mdz> Mithrandir: all firmware gets included in the kernel package, so that it is matched to a specific version of the driver
[06:53] <mdz> Mithrandir: it gets installed in /usr/lib/hotplug and loaded by the firmware loader, but it's part of the kernel source package
[06:54] <Mithrandir> mdz: ok.
[06:54] <mdz> I think this is the sensible approach
[06:54] <mdz> Kamion: ok
[06:54] <Mithrandir> sounds fine with me.
[06:54] <mdz> so, anyone who has firmware which is redistributable, send me information about it and a URL
[06:54] <mdz> (firmware for which the driver is already in the kernel)
[06:55] <mdz> for drivers which are non-free, such as madwifi, nvidia-kernel, etc., that stuff will go in linux-restricted-modules
[06:55] <mdz> which is what linux-non-free-modules is about to be renamed to
[06:55] <mdz> currently it has madwifi in it, and fabbione is going to add the nvidia-kernel bits
[06:55] <mdz> daniels: do the ati drivers require kernel bits?
[06:55] <daniels> mdz: yes
[06:56] <mdz> daniels: currently herbert is working on linux-restricted-modules, then he will send it to fabbione to do nvidia, then he can send it to you for ati
[06:56] <mdz> daniels: send an email to fabio to let him know you want it next
[06:56] <mdz> Kamion: what shall we do about a udeb for l-r-m?
[06:57] <mdz> Kamion: pick it apart in linux-kernel-di?
[06:57] <debianist> i knwo that might be offtopic for now, but does a test install on a thin client machine with flash drive has benefit for ubuntu ?
[06:57] <mdz> debianist: sure, anything and everything
[06:57] <debianist> mdz :  k
[06:57] <mdz> daniels: what about the X portion of the ati driver?
[06:58] <Kamion> mdz: that would produce essentially restricted udebs, but in main
[06:58] <mdz> daniels: that needs to be packaged for restricted
[06:58] <debianist> the nvidia driver would be installed outofthebox as a installation goal?
[06:58] <Kamion> mdz: I was thinking a separate kernel-wedge-using thing
[06:58] <daniels> mdz: yes, yes it does
[06:58] <daniels> mdz: do you want me to do ati?
[06:58] <[Clint] > mdz: i'm vaguely here
[06:58] <mdz> Kamion: hmm, yes, that would mean linux-kernel-di would build-dep on the restricted modules, yuck
[06:58] <mdz> [Clint] : we were just talking about the upgrade tool; let's discuss it on the sounder list later?
[06:58] <Kamion> mdz: BTW, I do have another idea for cutting down the udeb problem; we can reunify linux-kernel-di-*
[06:58] <daniels> mdz: (my concerns, as stated above, were just that anyone capable of configuring it is more than capable of building it on their own, and exposing it to the world implies that it won't be quite so difficult)
[06:59] <mdz> [Clint] : (you are subscribed to the sounder list, right?)
[06:59] <mdz> Kamion: ooh, that sounds nice
[06:59] <[Clint] > mdz: yup; i can't really spare my full attention atm
[06:59] <Kamion> mdz: it might run into apt limits though, I'm sure you're familiar with that problem ...
[06:59] <mdz> daniels: what are the issues with configuratin?
[06:59] <daniels> mdz: it's a *pain*
[06:59] <mdz> Kamion: I think I bumped those limits high enough in current apt, but it's easy enough to bump them again if necessary
[06:59] <thom> mdz: there's a crazy, horrible script that munges your X config
[07:00] <Kamion> mdz: that's not the issue, think woody->warty upgrades ...
[07:00] <daniels> mdz: especially when tooling around with agpgart settings, it's often easy to get lost in myriad tweaks that you need to do
[07:00] <daniels> yeah, fglrx-config is total crack also
[07:00] <mdz> Kamion: woody->warty upgrades require an apt pre-upgrade to work well anyway
[07:00] <Kamion> (which include deb-src)
[07:00] <doko> thom: not even a script, it's a binary to obfuscate things ...
[07:00] <Kamion> mdz: the point is it can get hard to even apt-get update
[07:00] <mdz> daniels, thom: so is it unsupportable crap?
[07:00] <debianist> Kamion : woody -> warty upgrades? we want that?
[07:00] <daniels> not least because it stomps all over your x config, and puts in about twenty thousand commented-out options
[07:00] <thom> mdz: i think that's a yes
[07:00] <Kamion> debianist: er, yeah, that's why we had that long discussion above about an upgrade tool
[07:00] <daniels> mdz: as much as we need 2d and 3d, fglrx just isn't it
[07:01] <mdz> Kamion: we'll drown under that bridge when we come to it, I guess :-)
[07:01] <daniels> (r4xx 2d and 3d, r3xx 3d)
[07:01] <debianist> Kamion : sarge -> warty looked more reasonable :)
[07:01] <thom> daniels: when does the r4xx drop happen?
[07:01] <mdz> ok, let's drop fglrx
[07:01] <daniels> thom: er, the patch has already been merged into xorg about a fortnight ago
[07:01] <mdz> I believe fabbione plans to package the nvidia binary driver for restricted
[07:01] <lamont> mdz: burn that bridge when we drive off of it.
[07:02] <thom> daniels: ah
[07:02] <mdz> are there any other binary-only non-firmware blobs that should be packaged for restricted?
[07:02] <daniels> thom: maybe even a month now
[07:03] <mdz> nothing else?  so far, we have madwifi and the nvidia graphics drivers
[07:03] <debianist> mdz : what about other peripherals, digi cams, firewire xfer thingies
[07:03] <doko> what is a blob in this context
[07:03] <daniels> if it doesn't happen for warty, i might do a radeon-driver-r4xx diversion crack attack on people.n-n-y.com
[07:03] <mdz> debianist: yes, what about them? :-)  are there any binary-only drivers that you need for such hardware?
[07:04] <mdz> doko: object code
[07:04] <pitti> jdub told us about some webcams
[07:04] <daniels> pwcx, iirc
[07:04] <mdz> pitti: I think the one he was talking about was pwcx
[07:04] <mdz> and that driver was discontinued
[07:04] <doko> yes, then the avm binary capi library
[07:04] <mdz> doko: ok, will you package it?
[07:04] <pitti> wasn't there a pwc also?
[07:04] <mdz> pitti: pwc is in the kernel, and free
[07:04] <pitti> great!
[07:05] <mdz> pwcx implemented extended features of the same devices
[07:05] <Mithrandir> mdz: pwc got removed with 2.6.8, didn't it?  Or rather, is in 2.6.9 somewehere?
[07:05] <doko> mdz: ok, for linux-restricted as a module?
[07:05] <mdz> Mithrandir: it did? oh
[07:05] <mdz> doko: is it kernel or userland?
[07:05] <Mithrandir> mdz: according to lwn.net, at least,  iirc
[07:05] <lamont> I thought it was that pwc removed the hooks that pcwx needed
[07:05] <doko> mdz: kernel
[07:05] <mdz> doko: ok, then it goes in linux-restricted-modules
[07:05] <mdz> doko: herbert is working on it now, then fabio, then daniel
[07:05] <mdz> er
[07:05] <mdz> scratch daniel, we dropped ati
[07:06] <mdz> doko: so you should get linux-restricted-modules after fabio
[07:06] <debianist> mdz : creative NOMAD JukeBox player, Webcam
[07:06] <doko> mdz: fine, hope I can manage it ...
[07:06] <mdz> doko: send email to fabio asking him to pass the packages to you after he has added the nvidia driver
[07:06] <mdz> doko: if you have trouble, mail herbert
[07:06] <mdz> hopefully it should be clear from the examples
[07:06] <doko> fine
[07:06] <mdz> if not, a little documentation would be nice
[07:07] <mdz> ok
[07:07] <mdz> debianist: do you have URLs and information about the drivers?
[07:07] <mdz> debianist: if so, please send them to me via email (description of the device, URL for the driver, any other information about it you may have)
[07:07] <doko> what about inclusion of the isdnutils-base, libcapi20-3 and isdnactivecards in desktop?
[07:08] <mdz> isn't isdnutils already in base?
[07:08] <debianist> mdz : k, i'll have to recollect those, it's been long time since I tampared with it ;)
[07:08] <mdz> hmm, it's in supported
[07:08] <mdz> doko: what is isdnactivecards?
[07:08] <mdz> debianist: are these drivers that you personally use?
[07:08] <mdz> debianist: (and can help us to test)?
[07:09] <mdz> doko: isdnutils-base sounds like it should go in base
[07:09] <debianist> mdz : hmmm seems there are open source "replacments" for most of them now, I will try to work out those ones than it would releive us from adding it 
[07:09] <doko> isdnactivecards is the package to handle isdn cards using the capi interface, packaged together with firmware for active isdn cards (active: own processor on the card). maybe I should split that into to capituils and activecards.
[07:09] <debianist> mdz : to restricted , that is
[07:10] <mdz> doko: is libcapi20-3 useful on its own?
[07:10] <mdz> doko: it does not seem to be depended on by either of the other packages
[07:11] <doko> it's needed by the capi utilities in isdnactivecards.
[07:11] <mdz> debianist: ok, let me know via email if there is anything that you feel is missing that we could add to the restricted module package
[07:11] <mdz> doko: why does it not depend on it?
[07:12] <debianist> mdz : ofcourse! np
[07:12] <doko> mdz: you found a bug.
[07:12] <tvon|x31> anyone using 'tpb' for thinkpads?
[07:13] <HcE> tpb?
[07:13] <thom> tvon|x31: i have, but it's a criminally ugly hack
[07:13] <doko> tvon|x31: I couldn't get it working yesterday.
[07:13] <mdz> I'm going to add isdnutils-base to BaseSeed; Kamion, can you update debootstrap?
[07:13] <Mithrandir> HcE: thinkpad buttons
[07:13] <Kamion> mdz: sure
[07:13] <mdz> thanks
[07:13] <HcE> which "extra" buttons?
[07:13] <Mithrandir> HcE: uses some osd thingy to make volume changing and such look fancy
[07:13] <HcE> the "Access IBM" button?
[07:14] <Mithrandir> HcE: I can show you tomorrow.
[07:14] <HcE> IIIIIK
[07:14] <tvon|x31> volume controle usually...there are some others
[07:14] <Mithrandir> if you remind me
[07:14] <tvon|x31> yes, access ibm as well
[07:14] <HcE> volume buttons work here without tpb
[07:14] <mdz> doko: what about ipppd and the other binary packages?
[07:14] <mdz> doko: what is the minimum needed for ISDN networking, which should go in base?
[07:14] <tvon|x31> tpb uses xosd to display the volume
[07:14] <thom> you don't need it for any of the buttons - it just prettifies the display
[07:14] <daniels> mdz: oh, while we're on this, could we please seed bpalogin in if it isn't already?
[07:15] <thom> and it's an evil hack, reading stuff out of /dev/kmem or something iirc
[07:15] <doko> mdz: I prepare that list offline, ipppd is needed as well.
[07:15] <mdz> doko: ok
[07:15] <Mithrandir> HcE: work, yes.  But does it show it onscreen?
[07:15] <daniels> mdz: it's dfsg-free, it's a custom login client for australia's largest cable modem network (and dsl, too), and has been picked up by a few others
[07:15] <mdz> Kamion: let's hold off until we've decided on the final list
[07:15] <HcE> Mithrandir: I said they work ;)
[07:15] <tvon|x31> I'd kinda like to rewrite it, using gnome-osd or something
[07:15] <mdz> daniels: supported?
[07:15] <Kamion> mdz: I'll just remove pppoe for this release then
[07:15] <daniels> thom: yeah, we shouldn't ship anything that just randomly maps shit out of /dev/kmem ;)
[07:15] <mdz> Kamion: ok
[07:15] <daniels> mdz: in the warty sense, or the upstream sense?
[07:16] <SteveA> I have totem-xine 0.99.15.1-1.  It crashes on start.
[07:16] <mdz> daniels: they are very closely related :-)
[07:16] <Mithrandir> daniels: it doesn't use kmem, it uses some other device.
[07:16] <daniels> mdz: heh
[07:16] <mdz> SteveA: ubuntu team meeting is in progress
[07:16] <lamont> SteveA:  I have that version, runs fine here..
[07:16] <mdz> daniels: is it supportable?
[07:16] <debianist> are we already supporting dell inspiron laptops?
[07:16] <daniels> mdz: well-supported upstream, would be nice to have in shipseed at least as you kind of need it to get online here
[07:16] <mdz> daniels: if so, it can go in supported
[07:16] <SteveA> mdz: sorry -- figured that would be on ubuntu-meeting
[07:16] <daniels> mdz: yes
[07:16] <debianist> including Alps touchpad module / driver etc?
[07:16] <mdz> SteveA: long story :-)
[07:17] <daniels> debianist: aieee, alps
[07:17] <daniels> debianist: i don't know if that still needs a kernel patch or not, but last i saw, it did
[07:17] <debianist> daniels : huh?
[07:17] <daniels> the userspace support is all there tho
[07:17] <daniels> debianist: last i saw, the kernel needed a patch to meaningfully talk to alps touchpads
[07:17] <debianist> daniels : i used a modconf insertion, when it still worked :)
[07:18] <debianist> daniels : it has a kernel module
[07:18] <daniels> debianist: so it works with current warty kernels? what did you have to insert?
[07:18] <mdz> daniels: ok, I'll add it to supported and ship
[07:18] <daniels> debianist: yeah, but i'm not sure if that's in the mainline kernel yet
[07:18] <daniels> mdz: thanks dude
[07:18] <debianist> daniels : i'd test that on the warty. havn't yet done so...
[07:18] <daniels> debianist: which kernel were you using?
[07:19] <debianist> daniels : it evens works with 2.6.8.1 currentl on the sid on the inspiron
[07:19] <pitti> can we sort out this specific things out of the meeting?
[07:20] <mdz> ok
[07:20] <mdz> Kamion: a quick update on the installer is the last item on my agenda
[07:20] <Kamion> today's daily was broken, that's fixed
[07:20] <mdz> the two specific issues I wanted to talk about were the apt-setup stuff for restricted and universe, and archive-copier
[07:21] <Kamion> apt-setup has been fabbione's baby; it seems to be working on the install I just did
[07:21] <Kamion> I have commented-out lines in sources.list for main+restricted and for main+restricted+universe
[07:21] <mdz> so those changes will be in sounder 8?
[07:21] <Kamion> yep
[07:21] <Kamion> archive-copier is my current final headache for sounder 8
[07:22] <Kamion> I've almost got the integration there to make the CD totally unnecessary after the first reboot, but am working through bugs
[07:22] <mdz> archive-copier itself, or are you working out the apt-config-in-prebaseconfig bits?
[07:22] <Kamion> the latter
[07:22] <mdz> ok
[07:22] <Kamion> archive-copier itself has been working well for a while now
[07:22] <mdz> do you think sounder 8 will see the light today?
[07:22] <Kamion> yes
[07:22] <mdz> yes, it worked fine for me
[07:22] <Mithrandir> mdz: how is the amd64 installer ATM?
[07:22] <Kamion> if I can't get this to work I'll punt it
[07:22] <Kamion> (like, within the next hour or so)
[07:23] <mdz> Mithrandir: today's daily was broken, so I haven't tried it
[07:23] <mdz> I'll test sounder 8
[07:23] <mdz> Kamion: in fact, if it's ok with you, I'd like to test a sounder 8 candidate before you bless it
[07:23] <Kamion> mdz: sure
[07:23] <Mithrandir> I'm also interested in that
[07:23] <mdz> it'd be excellent to have a sounder 8 which installed on all arches
[07:23] <Kamion> any other installer issues on people's minds?
[07:24] <Kamion> nothing much else comes to mind here
[07:24] <mdz> that's all I had on my list.  anything interesting in bugzilla?
[07:24] <npmccallum> Kamion: the installer is looking great
[07:24] <Kamion> there are a couple of debconf priority issues which have come up in sounder reports
[07:24] <Kamion> WEP keys in netcfg is one of them
[07:24] <Kamion> that's currently a priority high question
[07:24] <debianist> Kamion : the partitioner bug. 
[07:25] <mdz> Kamion: is it just a matter of increasing the priority, or is it trickier than that?
[07:25] <mdz> what partitioner bug?
[07:25] <Kamion> debianist: yep, that's another open issue
[07:25] <Kamion> mdz: resizing ext3 not working; may be a parted issue, haven't diagnosed it yet
[07:25] <mdz> hmm, haven't tested that
[07:25] <Kamion> mdz: increasing the priority would fix it, but may incur Mark's wrath at an additional question :-)
[07:25] <mdz> Kamion: it would only be asked if it couldn't associate without one, no?
[07:26] <Kamion> mdz: I think it's asked regardless, but wouldn't swear to it
[07:27] <mdz> ah, I thought there was some fallback there, where it would try it with no configuration, and if that didn't work, fall back to asking for the essid
[07:27] <debianist> Kamion : also, the fact d-i advances to installation although no root filesystem has been setup. Maybe block the next stage with a flag untill a partition is created?
[07:27] <mdz> asking for the wep key at the same time sounds like a minimum of evil
[07:27] <Kamion> there are also some CD-ROM issues on my bug list that I'm not sure how to fix, and may be hardware specific
[07:28] <Kamion> oh, and all the fstab stuff; I need somebody to tell me the currently-approved state of play on that
[07:28] <Kamion> preferably in the bug so I remember
[07:28] <mdz> Kamion: is the question which devices should be added to fstab?
[07:28] <pitti> Kamion: regarding CD-ROM entries?
[07:29] <Kamion> mdz: yes
[07:29] <Kamion> pitti: and floppies
[07:29] <pitti> Kamion: we should leave them in fstab by now; floppies and CD-ROMs cannot be mounted by pmount currently
[07:29] <mdz> someone posted to the sounder list saying that their SCSI CD-ROM wasn't added
[07:30] <mdz> the current behaviour is fine for my systems
[07:30] <Kamion> bug #1040
[07:30] <pitti> Kamion: we could change pmount policy for that, but it's a bit late I think
[07:30] <pitti> Kamion: in addition, it is nice if the CD-ROM appears in the computer window even if no CD is present
[07:30] <debianist> add those to fstab,
[07:30] <debianist> leave GNOME handle the mounts
[07:30] <Kamion> please, guys, in the bug
[07:30] <debianist> sorry
[07:30] <mdz> agreed, post your comments to #1040
[07:30] <Kamion> I don't want a discussion, I just want the current approved state :-)
[07:31] <Kamion> a discussion will render me more confused, since I haven't followed all of this
[07:31] <mdz> ok, does anyone else have anything they need to discuss before we close the meeting?
[07:31] <pitti> can we agree how to disable the cups admin interface?
[07:31] <doko> planning the release party might be a bit early ;)
[07:31] <mdz> can we replace the HTML with some explanatory text?
[07:31] <pitti> the error page? should be possible
[07:31] <Mithrandir> for the next meeting, I would very much like to have a list of items to go through beforehand; it's been a bit unstructured.
[07:31] <pitti> but at that time, the user already entered his root password
[07:31] <mdz> the page you get at /admin
[07:32] <debianist> mdz : how many language checked from d-i ?
[07:32] <pitti> mdz: you can get there from everywhere
[07:32] <debianist> mdz : how many language checked for d-i ?
[07:32] <mdz> Mithrandir: oh, I think I skipped over amd64 status
[07:32] <mdz> Mithrandir: what remains on your todo list?  openoffice.org?
[07:32] <mdz> pitti: we should disable the authentication and show a placeholder page, ideally
[07:32] <Mithrandir> mdz: OOo, some mozilla stuff is still not fixed (thunderbird at least, epiphany was broken because I gave seb a broken patch)
[07:32] <mdz> which tells them where to find gnome-cups-manager
[07:33] <debianist> mdz : never mind that. I need to check this problem on real hardware also
[07:33] <pitti> mdz: you cannot disable the authentication
[07:33] <pitti> mdz: but I can change the web pages not to refer to /admin any more
[07:33] <mdz> Mithrandir: please file bugs in bugzilla so that I can keep up easily
[07:33] <Mithrandir> I need to run through the installer a few times and make sure it's works well, but apart from that, amd64 is in good shape.
[07:33] <mdz> pitti: I thought the problem was that users were accustomed to going to http://localhost:6031/admin or whatever
[07:33] <pitti> mdz: we cannot do anything about that
[07:33] <pitti> mdz: we need this interface because gnome-cups-manager uses the CGI as well
[07:34] <pitti> mdz: (anything apart from rewriting half of cups)
[07:34] <debianist> :)
[07:34] <pitti> mdz: okay, I will change the error page and remove the admin links. 
[07:34] <mdz> pitti: but gnome-cups-manager does not care what text is displayed on the page, does it?
[07:35] <pitti> mdz: I think so
[07:35] <mdz> pitti: ok, that sounds reasonable
[07:35] <Mithrandir> mdz: ok, I'll file bugs and grab them myself?
[07:35] <mdz> pitti: if the user at least gets a reasonable error when their password is rejected, that is enough for Warty
[07:35] <mdz> Mithrandir: yes, you can assign them to yourself when opening them
[07:35] <pitti> I still have an open issue: how to efficiently handle the remaining translations?
[07:35] <mdz> Mithrandir: it's just easier for me to see the status at a glance that way
[07:35] <Mithrandir> sure
[07:35] <mdz> pitti: translations of what?
[07:36] <pitti> is there a better way than to upload a new package for every translation?
[07:36] <pitti> mdz: translations of programs
[07:36] <pitti> mdz: gksudo still talks english to me
[07:36] <pitti> mdz: so does gnome-cups-admin
[07:36] <pitti> and so on...
[07:36] <mdz> if the problem is that some programs are not fully translated yet, that is a Hoary problem
[07:36] <debianist> pitti : what language do you need it ?
[07:37] <pitti> debianist: Myself I'm fine with English, but I can do the German translations
[07:37] <pitti> debianist: if I need your's, I will come back to you :-)
[07:37] <debianist> pitti : k :) thanks for that
[07:37] <mdz> pitti: we cannot afford to spend your time on it right now; we can maybe revisit it after preview
[07:37] <pitti> mdz: okay, so we defer that until we have Rosetta?
[07:38] <mdz> pitti: please file bugs
[07:38] <mdz> (severity minor)
[07:38] <pitti> will do.
[07:38] <mdz> ok, anything else?
[07:38] <mdz> trying to keep it under 2 hours here :-)
[07:39] <thom> quick, someone filibuster for 20 minutes
[07:39] <thom> i'm sure we can do it
[07:39] <mdz> I'll send out an email with the deliverables I've collected
[07:39] <debianist> mdz : yeah, some of us are dead hungry :))
[07:39] <mdz> ok, we're done
[07:39] <mdz> thanks, everyone
[07:39] <daniels> or dead tired
[07:39] <debianist> thanks mdz
[07:39] <daniels> 'night all
[07:39] <pitti> thanks, Matt
[07:39] <thom> cheers
[07:39] <pitti> daniels: sleep well
[07:40] <thom> daniels: night kiddo
[07:40] <doko> bye
[07:40] <Oskuro> nite little daniel
[07:40] <Oskuro> :)
[07:40] <Mithrandir> daniels: see you around
[07:40] <debianist> bye all
[07:40] <debianist> (that are going to sleep)
[07:40] <pitti> so where's the buffet? :-)
[07:40] <debianist> or leaving
[07:40] <debianist> haha
[07:41] <lamont> there's a buffet?
[07:41] <pitti> lamont: don't tell me there isn't!
[07:41] <debianist> I don't reckon there's too much to eat here currently, but some other folks went shopping. I hope there'd be something good for eating , although I'd rather order a pizza
[07:41] <debianist> :)
[07:43] <lamont> ummmm... pizza......
[07:43] <debianist> yes. to damn shame havn't got a nickel to spare this month, maybe next one :)
[07:43] <debianist> *too
[07:44] <debianist> Kamion : a new image is ready and fixed?
[07:44] <Kamion> debianist: no
[07:45] <debianist> Kamion : you said it was fixed on the meeting, or was it me helusinating?
[07:45] <Kamion> debianist: fixed in the archive
[07:45] <Kamion> please go have a pizza or something :)
[07:45] <doko> kamion: for a new installer, is there anything smaller to download than the whole warty iso?
[07:45] <debianist> Kamion : when would an image be ready?
[07:46] <debianist> Kamion : just though of letting it rsync while I eat :) sorry, not meant to nag you
[07:46] <Kamion> doko: rsync?
[07:47] <Kamion> debianist: I don't know.
[07:47] <Kamion> (yet)
[07:47] <Kamion> asking me every five minutes will just slow it down :)
[07:47] <debianist> hint taken.
[07:47] <doko> kamion: don't have an old archive here :(
[07:48] <lamont> -ENOSODA.  panic
[07:48] <Kamion> doko: there's also netboot ... it's in /dists/warty/main/{,daily-}installer-$ARCH/netboot/ or thereabouts
[07:48] <doko> kamion: ok, I'll look at it.
[07:52] <Kamion> aha, found my bug
[08:06] <mdz>  * Starting PCMCIA services...                                           [fail] 
[08:06] <mdz> has anyone else seen that?
[08:09] <elmo> kamion: ubuntu9's in for all arches, fyi
[08:09] <Kamion> elmo: woo, thanks
[08:09] <npmccallum> mdz: on just a normal boot?
[08:09] <mdz> npmccallum: on an upgrade
[08:10] <mdz> Preparing to replace pcmcia-cs 3.2.5-7ubuntu3 (using .../pcmcia-cs_3.2.5-7ubuntu4_powerpc.deb) ...
[08:10] <mdz> Unpacking replacement pcmcia-cs ...
[08:10] <mdz> Setting up pcmcia-cs (3.2.5-7ubuntu4) ...
[08:10] <mdz>  * Starting PCMCIA services...                                           [fail] 
[08:10] <npmccallum> mdz: did it fail to bring down the interface?
[08:10] <mdz> I didn't see a "Stopping" message
[08:11] <npmccallum> mdz: wierd
[08:11] <mdz> never mind
[08:11] <mdz> I was not logged into the system I thought I was
[08:12] <mdz> that one has no pcmcia
[08:12] <mdz> I should stop naming them all 'ubuntu'
[08:12] <npmccallum> that would do it :)
[08:12] <Mithrandir> heh :)
[08:13] <npmccallum> bbiab
[08:13] <Mithrandir> mdz: you know, if you don't feel like inventing something, ubuntu1, ubuntu2, ubuntu3, etc would work. ;)
[08:13] <Kamion> OK; dinner while everything builds, then I'll roll and test a new daily.
[08:14] <mdz> Kamion: ready to test when you are
[08:14] <Kamion> counting all the archive/buildd schedules, should be an hour or so
[08:15] <elmo> so, like, anyone got a better name for warty-p-u than -p-u?  -errata maybe? 
[08:16] <Mithrandir> RH used to call it errata, ISTR.
[08:16] <Kamion> or drop the -proposed and just call it -updates
[08:18] <elmo> kamion: clashes with debian's stable/updates on sec.d.o
[08:21] <elmo> but yeah, either works for me to be honest.. -updates or -errata
[08:22] <mdz> elmo: -updates WFM
[08:22] <lamont> dpkg: considering removing mtr-tiny in favour of mtr ...
[08:22] <lamont> dpkg: yes, will remove mtr-tiny in favour of mtr.
[08:22] <lamont> when did dpkg get polite???
[08:23] <elmo> dude, it's done that for years.. it's an iwj-ism, I think
[08:51] <lamont> hrm... jdub around yet?
[08:52] <lamont> until such time as I can force an ssh-add to happen after the login, and before all of the windows are restored...
[08:53] <Keybuk> .gnomerc ?
[08:53] <lamont> Keybuk: remember the cluefactor wrt gnome here...
[08:55] <lamont> I currently have a window popping up to do the ssh-add, but metacity likes to steal focus from it, and life is sad.
[08:55] <lamont> and all the other windows come up before I can get the passphrase love in
[08:55] <lamont> so they all want the passphrase.
[08:55] <lamont> which sucks
[08:56] <lamont> what would I put in (my nonexistant) .gnomerc?
[08:56] <mdz> the right place for this stuff is in the session preferences...but you're saying that it doesn't wait for ssh-add to complete before it moves on?
[08:56] <mdz> or that ssh-add runs too late?
[08:56] <mdz> if it runs too late, there's a numeric sort order in there
[08:56] <mdz> if it doesn't wait, that seems like a bug
[08:57] <Keybuk> you're using gnome-ssh-askpass ?
[08:57] <lamont> scheduled to run early, as xterm -e ssh-add
[08:57] <lamont> Keybuk: uh, what's that...
[08:57] <Keybuk> ssh-askpass-gnome, sorry
[08:57] <Keybuk> it's an ssh-askpass UI for GNOME
[08:57] <Keybuk> so instead of an Xaw dialog for your password, you get a GNOME HIGy one
[08:58] <Keybuk> which has the advantage that it tells Metacity "THE FOCUS IS MINE" and grabs the keyboard
[08:58] <lamont> but does everything else wait?
[08:58] <Keybuk> no, but nothing takes the keyboard until it goes away
[08:59] <Keybuk> in a terminal, do: ssh-add < /dev/null
[08:59] <mdz> ssh-askpass-gnome is installed by default in ubuntu-desktop
[08:59] <Kamion> lamont: xterm -e ssh-add seems like a somewhat strange approach
[08:59] <lamont> by which time all the other xterm -e ssh sessions will have decided they want a passprhase...
[08:59] <Keybuk> yeah, I see the problem there
[08:59] <Kamion> I'd just do plain ssh-add
[08:59] <mdz> and will be used by default if no tty is available
[08:59] <Kamion> then it ought to work
[09:00] <lamont> Kamion: remember: gnome illiterate...
[09:00] <Kamion> s/work/wait/
[09:00] <Kamion> lamont: so am I, this is my plain X setup
[09:00] <Keybuk> you need to do the ssh-add before the session manager starts really
[09:00] <lamont> "No response to the SaveYourself command...."  Le huh?
[09:02] <debianist> somebody ought to change the channel's subject :)
[09:04] <Kamion> elmo: crap, I forgot to add hdparm-udeb to the cdrom initrds :-/
[09:05] <Kamion> elmo: uploading *another* debian-installer build ...
[09:06] <Oskuro> elmo: hug him!
[09:06] <mdz> elmo: what's the status of warty-security?  we're going to need to do some security updates soon, and we might as well put them through the -security infrastructure to test it out, if you'll be ready
[09:07] <mdz> thom: hmm, encountered a bug in the new xscreensaver
[09:07] <mdz> thom: ifs appears to be disabled :-P
[09:08] <lamont> my RAM arrived - gotta go fetch it this afternoon
[09:14] <seb128> hum, "An error occurred. Dammit. Error was: 270465 state kill-init at end." with the Debian BTS
[09:15] <Kamion> will fix
[09:15] <Kamion> whoa, or not. where'd the .log go?
[09:16] <Kamion> -rw-rw-r--    1 debbugs  debbugs         0 Sep  7 07:34 270465.log
[09:16] <seb128> weird
[09:16] <Kamion> we did get a scary mail earlier today
[09:18] <seb128> what sort of mail ?
[09:18] <Kamion> seb128: cron error from processall
[09:19] <Kamion> that bug doesn't appear to be in any indexes; I'm inclined to just delete it and reprocess
[09:20] <Kamion> seb128: ok, that's now bug #270528; #270465 "never existed".
[09:21] <seb128> ok
[09:21] <seb128> in fact I was just looking the "Opened/upgraded release-critical bugs" list
[09:22] <seb128> rc -= 1 :p
[09:30] <mdz> daniels: your mail is looking
[09:30] <mdz> s/looking/looping/
[09:32] <elmo> mdz: tonight
[09:32] <mdz> pitti: still here?
[09:33] <elmo> (having a life is for losers after all :P)
[09:36] <lamont> mdz: someone working on 1080?
[09:37] <elmo> mdz: oh yeah, why was I cc'ed on that?
[09:37] <mdz> lamont: you? :-)
[09:37] <mdz> elmo: ah, forgot to mention in the bug.  we had talked about whether it would be a good idea to add a global Bugs: override
[09:37] <mdz> that's part of the same issue
[09:38] <mdz> that would be one way to address the reportbug part of it
[09:38] <elmo> oh, ok
[09:38] <elmo> I think we should do that regardless
[09:38] <elmo> 'cos it's trivial and obviously correct
[09:39] <lamont> do we have an email injection method for our bugzilla?
[09:39] <elmo> hmm, assuming apt-ftparchive will overwrite existing fields, will have to check that
[09:39] <lamont> 1066 is a valid bug.  bummer.
[09:40] <lamont> not just because it means there'll be another linux-source upload...
[09:40] <lamont> mdz: who are we assigning kernel bugs to these days?
[09:41] <mdz> lamont: me, for now
[09:41] <mdz> lamont: no, but we need one
[09:42] <mdz> justdave: ?
[09:42] <lamont> mdz: I'd like to be able to burn dvd's on warty...
[09:42] <mdz> that is
[09:42] <mdz> kernel: me, for now
[09:42] <mdz> email injection method for bugzilla: no, but we need one, ->justdave
[09:43] <justdave> ah that, yes, that'd be me
[09:43] <mdz> (#1080)
[09:43] <mdz> we need a way to gateway reportbug into our bugzilla
[09:43] <lamont> mdz: OK.  First issue is that ide-scsi isn't getting loaded, hence lots of bitchiness as noted.
[09:43] <mdz> email seems like the most expedient way, since reportbug already sends a nicely-formatted email
[09:43] <mdz> lamont: ide-scsi should not be necessary anymore
[09:44] <lamont> then, once ide-scsi is modprobed into existance (and the bitchiness is gone from kern.log), it says: :-( unable to PREVENT MEDIA REMOVAL: Operation not permitted
[09:44] <lamont> or is that perms again..
[09:44] <mdz> I haven't tried to write a DVD yet under 2.6.8, but writing CDs with cdrecord works fine so far
[09:44] <mdz> I'll try one right now
[09:44] <Kamion> speaking of, fixing #923 is easy, but I'm going to leave it until sounder 8's done now
[09:46] <tvon|x31> 'sounder 8' is like an RC/beta sorta thing?
[09:46] <Kamion> yeah
[09:47] <tvon|x31> Gotcha
[09:47] <Kamion> the Sounder CD releases are testing milestones for Warty
[09:47] <tvon|x31> k
[09:47] <tvon|x31> The nightlies are snapshots of them I take it?
[09:47] <Kamion> for Hoary, they'll be Array CD releases, I'm told
[09:47] <mdz> lamont: seems to be working fine so far
[09:47] <mdz> lamont: mine is USB, so no ide-scsi madness anyway
[09:47] <lamont> will update the bug with my findings for ide...
[09:48] <Kamion> the dailies are entirely automatic snapshots, yes; they might be arbitrarily broken
[09:48] <Kamion> the Sounder CD releases usually have some kind of development milestones associated with them, they come with release notes, and they have at least been used by me to do a complete install :-)
[09:49] <mdz> lamont: try it without ide-scsi and see what happens
[09:49] <mdz> eek
[09:49] <mdz> I spoke too soon
[09:49] <mdz> it got to 4.5% and went nuts
[09:50] <mdz> er
[09:50] <mdz> well, it's spewing SCSI errors at the rate of several hundred per second
[09:50] <mdz> but the burn seems to be chugging merrily along
[09:50] <mdz> fascinating
[09:51] <lamont> bug updated, assigned to you
[09:53] <mdz> Kamion: I was just thinking
[09:53] <mdz> Kamion: apt-cdrom essentially does "find /cdrom/"
[09:53] <mdz> Kamion: maybe that could substitute for your explicit find
[09:53] <doko> mithrandir: gcc-3.4 build finished on amd64, i386, powerpc without new regressions. waiting for your ok now.
[09:56] <Mithrandir> I'm going to build m-t with it to test, ok?
[09:57] <doko> fine. amd64 binaries currently on yellow:~doko/gcc/
[09:58] <Mithrandir> is gcc-3.4_3.4.1ds1-7ubuntu5 the right version?
[10:00] <doko> no, 3.4.2-1ubuntu1
[10:01] <Kamion> mdz: probably not in cdrom-detect, since apt-cdrom isn't even close to installed yet ...
[10:03] <fabbione> hey guys
[10:03] <fabbione> was the meeting here in #ubuntu?
[10:04] <Mithrandir> yes
[10:04] <fabbione> ok
[10:04] <fabbione> thanks
[10:08] <Kamion> lamont: let me know if debian-installer fails (blocking on that at the moment)
[10:09] <Mithrandir> doko: uhm
[10:09] <Mithrandir> : tfheen@golem /tmp > g++-3.4 -o hello hello.c
[10:09] <Mithrandir> In file included from hello.c:1:
[10:09] <Mithrandir> /usr/include/stdio.h:34:21: stddef.h: No such file or directory
[10:09] <Mithrandir> (and then a lot of errors)
[10:10] <Mithrandir> m-t fails to build as well, which was what got me started
[10:10] <doko> amd64?
[10:11] <Mithrandir> yes
[10:11] <Mithrandir> from m-t:
[10:11] <Mithrandir> configure:2230: g++-3.4 -o conftest    conftest.C  1>&5
[10:11] <Mithrandir> ld: cannot open crtbegin.o: No such file or directory
[10:11] <Mithrandir> configure: failed program was:
[10:11] <Mithrandir> #line 2225 "configure"
[10:11] <Mithrandir> #include "confdefs.h"
[10:11] <Mithrandir> int main(){return(0);}
[10:12] <Mithrandir> sure you don't have to tighten some deps?
[10:12] <lamont> mdz: re 1072: yeah, ok.
[10:12] <lamont> I meant to say NOTWARTY, honest.
[10:13] <Mithrandir> doko: mea culpa, version skew
[10:13] <Mithrandir> doko: if you don't upgrade g++, but only libgcc and gcc, g++ blows up
[10:15] <doko> yes, g++-3.4_3.4.1 has a dependency on gcc-3.4 (<< 3.4.2), so that should not happen,
[10:15] <Mithrandir> well, it did.
[10:16] <thom> mdz: hey, i just do what i'm told
[10:17] <thom> :-)
[10:17] <Kamion> mdz: but yes, apt-cdrom would substitute for it in the case where you aren't using archive-copier; I didn't make any post-reboot changes along those lines
[10:20] <doko> hmm, g++-3.4_3.4.1-7ubuntuX has a Depends: gcc-3.4-base, libc6 (>= 2.3.2.ds1-4), gcc-3.4 (>= 3.4.1-10), gcc-3.4 (<< 3.4.2), libstdc++6-dev (>= 3.4.1-10)
[10:20] <fabbione> mdz: yes.. that's correct. i am going to add the nvidia bits only.
[10:21] <fabbione> daniels: i will pass the package to you after for ati
[10:21] <fabbione> no.. no ati :-)
[10:25] <fabbione> ok.. cya tomorrow guys
[10:25] <fabbione> good night everybody
[10:25] <lamont> night fabbione
[10:26] <thom> night
[10:28] <lamont> mdz: the question on dvd burning is, should failure to lock the door (EPERM, to be specific) be fatal to the burn process, or is that just caveat emptor?
[10:30] <mdz> lamont: I would say emphatically no, it should not be fatal
[10:30] <lamont> specifically, commenting out the exit()..
[10:31] <mdz> lamont: is read/write permission on the device no longer sufficient to lock the door, or something?
[10:32] <lamont> dunno, but user with rw, not root can't do it..
[10:32] <lamont> mdz: no dice.
[10:32] <lamont> About to execute 'builtin_dd if=warty-i386-1.iso of=/dev/hdc obs=32k seek=0'
[10:32] <lamont> :-( unable to PREVENT MEDIA REMOVAL: Operation not permitted
[10:32] <lamont>  /dev/hdc: "Current Write Speed" is 2.5x1385KBps.
[10:32] <lamont> :-( unable to WRITE@LBA=0h: Operation not permitted
[10:32] <lamont> :-( write failed: Operation not permitted
[10:33] <lamont> so it looks like write access to the device is no longer sufficient to write to the device.
[10:34] <mdz> clever
[10:34] <mdz> lamont: so it works as root?
[10:34] <doko> keybuk: around?
[10:34] <lamont> yes
[10:34] <lamont> providing you don't have the environment variable 'SUDO_COMMAND' in your environment
[10:34] <mdz> lamont: so is this the same issue as #1066, or really a different one?
[10:35] <lamont> this is 1066
[10:35] <mdz> lamont: yeah, I filed a bug about that
[10:35] <mdz> I think
[10:35] <mdz> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=268308
[10:35] <lamont> 1066's reporter only had the non-root-user perspective
[10:36] <mdz> Now, the funny thing here is that neither doing this as root or user
[10:36] <mdz> worked, but cdrecord with the dvd patch applied was successfull in
[10:36] <mdz> blanking the media, but not in burning afterward. Once
[10:36] <mdz> cdrecord+dvd-patch blanked the disk, i could without problem burn the
[10:36] <mdz> media with the above growisofs invocation.
[10:36] <mdz> that's from #269397
[10:36] <lamont> want me to remove the predjudice???  huh, huh, can I? can I?
[10:36] <mdz> sounds like a different problem
[10:39] <lamont> actually, 269397 looks to be the one I see after I defeat the door-lock code.
[10:39] <lamont> but still kernel bug
[10:40] <lamont> mdz: OK to upload a fix for 268308?
[10:40] <lamont> given our trend towards sudo, and all that..
[10:40] <mdz> lamont: if it's trivial (only removing code), yeah, why not
[10:42] <lamont> Kamion: debian-installer_20040801ubuntu10 succeeded x3
[10:42] <lamont> mdz: OK.  gotta go fetch the kidds
[10:42] <Kamion> cool
[10:43] <lamont> Kamion: and checking that is the only reason I'm still here... :-)
[10:45] <Kamion> elmo: can I have that byhanded? sorry to pester, I'm blocked on it
[10:58] <mdz> alextreme: ping?
[10:58] <alextreme> mdz: pong
[10:59] <alextreme> i've been lurking since the meeting started :)
[11:00] <mdz> alextreme: will it be possible for you to build a live CD which corresponds to sounder 8?  that would be excellent
[11:01] <elmo> kamion: powerpc still isn't built ..
[11:01] <elmo> the others are done though
[11:02] <Kamion> elmo: hm, lamont must've been confused
[11:02] <elmo> meh
[11:03] <elmo> 4,9,14,19,24,29,34,39,44,49,54,59* * * *        /usr/bin/buildd-uploader
[11:03] <Kamion> d'oh
[11:04] <elmo> cron.daily's running now
[11:04] <alextreme> mdz: and sounder 8 is the end of this week?
[11:04] <mdz> alextreme: sounder 8 is tonight :-)
[11:04] <alextreme> hmm, that might get kinda tight :)
[11:05] <elmo> kamion: if you're blocked on me and I've wandered off, just hate-sms me
[11:05] <Kamion> elmo: all right, thanks
[11:05] <mdz> alextreme: hmm...what is a realistic date to have an updated live CD?  is it more than one day's work?
[11:05] <Kamion> alextreme: doesn't have to be exactly in sync
[11:05] <mdz> I thought it was almost automated now
[11:06] <Kamion> alextreme: you might want to grab the archive as of a few minutes from now, though
[11:06] <alextreme> and updated live CD doesn't cost much time to make, but with only one hour left... :)
[11:07] <mdz> alextreme: well, it doesn't need to be released at the same time, the idea was just to have the same packages on it
[11:07] <mdz> or very close to it
[11:07] <alextreme> has there been a lot of changes lately?
[11:08] <mdz> alextreme: there have been a lot of changes since wartylive-v1 :-)
[11:09] <alextreme> i'll check the wiki and update my packagelist :)
[11:10] <mdz> alextreme: you should use germinate to generate the package list, so that it is consistent with ours
[11:10] <mdz> or perhaps Kamion can send you a current copy
[11:11] <alextreme> i know, but not all packages are installable or usable from the live cd, so i've had to edit it a bit
[11:12] <alextreme> seems only the base has had any changes though, desktop packagenames seem to be unaltered
[11:12] <seb128> hey SurcouF 
[11:12] <SurcouF> hi
[11:12] <SurcouF> hi seb128 
[11:12] <SurcouF> hi everyone
[11:12] <seb128> guys, SurcouF has cancelled the user creation during the installation
[11:13] <Kamion> current copies of germinate output are here: https://chinstrap.warthogs.hbd.com/~cjwatson/germinate-warty-output/
[11:13] <seb128> and he has no root password
[11:13] <Kamion> init=/bin/sh
[11:13] <seb128> and no user account
[11:13] <seb128> how he solves that ? :)
[11:13] <alextreme> k, thanks Kamion
[11:13] <SurcouF> seb128, no problem to solve
[11:13] <Kamion> or boot with single, on i386 there's probably even an entry for that on the grub boot menu
[11:13] <SurcouF> but
[11:13] <seb128> SurcouF: oh, I was thinking so :)
[11:13] <SurcouF> this is a bug
[11:13] <seb128> SurcouF: how do you log ?
[11:13] <SurcouF> from installer
[11:14] <Kamion> cancelled how?
[11:14] <SurcouF> seb128, what do you mean ?
[11:14] <SurcouF> Kamion, by select "Cancel" and hit down
[11:14] <Kamion> please file a bugzilla bug
[11:14] <SurcouF> okay
[11:14] <Kamion> component passwd (if that exists, otherwise UNKNOWN); assign to cjwatson@flatline.org.uk
[11:15] <Kamion> (i.e. me)
[11:16] <SurcouF> ok
[11:17] <Kamion> mdz: new daily is up
[11:18] <debianist> yey!
[11:18] <debianist> :)
[11:18] <debianist> btw , just installed from previous one
[11:18] <seb128> kagou: working daily ? :)
[11:18] <debianist> over my system.
[11:18] <seb128> oups
[11:19] <seb128> Kamion: working daily ? :)
[11:19] <kagou> seb128, :)
[11:19] <seb128> kagou: I hate the new xchat's completion
[11:19] <kagou> me too
[11:19] <SurcouF> seb128, like irssi ? ;-)
[11:19] <seb128> SurcouF: probably yes
[11:19] <Kamion> seb128: dunno yet
[11:20] <Kamion> rsyncing now to find out
[11:22] <alextreme> mdz: k, i've set up everything to build a new warty overnight. it still has a 2.6.7 kernel though, would that be an issue?
[11:32] <debianist> would the list be a proper place for install experience?
[11:32] <Kamion> sure
[11:32] <Kamion> there've been other installation reports there
[11:32] <debianist> however there is one thing i feel i'd better ask you here, im not sure if this is something i did wrong
[11:32] <debianist> regarding x resultion probes
[11:33] <debianist> (or the other guys for that matter)
[11:34] <Keybuk> doko: firefly break ... back now?  what's up?
[11:35] <doko> already submitted a report against dpkg, if you want to have a look ...
[11:35] <debianist> config chose the monitor biggest resulotion, however i barely can use the desktop on it. where is the gnome desktop resolution chooser?
[11:35] <Kamion> debianist: don't think the X guys are around right now, probably best to mention it in the installation report and say what graphics hardware you've got, then they'll be able to look at it
[11:36] <Kamion> ah, there's one somewhere, I don't remember where though ...
[11:36] <debianist> couldn't find any under "Computer" --> "Desktop prefs"
[11:37] <debianist> oh
[11:37] <debianist> it's under ssytem prefs
[11:37] <debianist> :)
[11:40] <Keybuk> doko: hmm, how can you fulfil that dependency?
[11:40] <Keybuk> oh, wait, sorry, misread it
[11:41] <SurcouF> Kamion, #1085
[11:42] <doko> keybuk: installing new g++-3.4, libstdc++6, libstdc++6-dev as well.
[11:43] <Kamion> SurcouF: I quite like the idea of cancelling the user addition and being sent back to set the root password, actually
[11:43] <Kamion> mdz_: what do you think?
[11:43] <SurcouF> Kamion, as you wish
[11:43] <SurcouF> just some ideas
[11:44] <Kamion> yup, thanks
[11:44] <Kamion> we'll talk about what the right thing to do is; the current situation is clearly a bug
[11:45] <SurcouF> ok
[11:45] <SurcouF> I'll report another bug against xserver-xfree86
[11:46] <Keybuk> doko: if I try with the following, is that enough of a test?
[11:46] <Keybuk>   cpp-3.4 g++-3.4 gcc-3.4 gcc-3.4-base libstdc++6 libstdc++6-dev
[11:46] <Keybuk> all 3.4.1-7 
[11:47] <doko> libgcc1 is missing, and then all these but g++-3.4,  libstdc++6 libstdc++6-dev from 3.4.2-1 (which doesn't exist yet in the archive)
[11:49] <doko> basically, it should be enough to install cpp-3.4 gcc-3.4-base libgcc1 gcc-3.4 from 3.4.1-7 and then only upgrade cpp-3.4 gcc-3.4-base libgcc1
[11:49] <Keybuk> I already have libgcc1
[11:50] <doko> do you have an amd64 platform o check on=
[11:50] <debianist> Nuvola theme missing from ubuntu :(
[11:51] <Keybuk> I have no 3.4.2 of those ... where is it?
[11:52] <debianist> i had many unmet dependencies at the package installation setup phase from the CD, after which I was sent back to aptitude to correct these. ofcourse i couldn't take note of all the packges that failed as this was blzing fast. anyone seen this?
[11:52] <doko> amd64: yellow:~doko/gcc, or else currently uploading to people.debian.org/~doko/gcc-3.4/ ...
[11:53] <Keybuk> doko: ok, ping me when the i386 debs are up
[11:54] <doko> these are debian unstable only, rebuilding the warty debs, but going to bed now.
[11:54] <Keybuk> that's ok, am on my unstable machine atm
[11:55] <mdz_> Kamion: regarding #1085?  I'll take a look
[11:56] <doko> Keybuk: I don't ping you, the upload should be finished in about one hour (it's finished if you see the treelang deb)
[11:57] <Keybuk> ah
[11:57] <Keybuk> ok
[11:57] <Keybuk> I've got it
[11:57] <Keybuk> the bug isn't what you think it is
[11:57] <Keybuk> it's just the "dpkg doesn't reverify dependencies on upgrade" bug
[11:57] <Keybuk> when you upgrade packages, it only takes their dependencies into account, it doesn't recheck any dependencies on the old package
[11:58] <mdz_> Kamion: I like the idea of falling back to setting a root password, but it seems simpler to disallow skipping that step
[11:58] <mdz_> Kamion: whatever is simplest, really
[11:58] <justdave> is there a way to make the terminal services client have real scrollbars instead of the classic xterm style ones?