--- Day changed

--- Log opened Tue May 11 23:59:01 2004
00:01 -!- sigjuice [~sigjuice@c-24-6-51-184.client.comcast.net] has joined #uml
00:59 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has joined #uml
01:10 -!- kkourt [~kkourt@solar.cslab.ece.ntua.gr] has quit [Ping timeout: 480 seconds]
01:13 -!- kkourt [~kkourt@solar.cslab.ece.ntua.gr] has joined #uml
01:20 -!- stbain [~Gun_w3rki@ip68-10-42-4.rn.hr.cox.net] has quit [Quit: Ni!!!!!!!!!!!]
01:22 -!- Heinachi [~krylen@ip68-14-129-139.ri.ri.cox.net] has joined #uml
01:22 < Heinachi> hrrm, 2.6.4 works mostly?
01:22 < Heinachi> what's the reason for saying that?
01:28 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has quit [Remote host closed the connection]
01:28 -!- sigjuice [~sigjuice@c-24-6-51-184.client.comcast.net] has quit [Quit: Client Exiting]
02:19 -!- rus [~rghf@jvds.demon.co.uk] has quit [Ping timeout: 480 seconds]
02:19 -!- rghf_ [~rghf@jvds.demon.co.uk] has quit [Ping timeout: 480 seconds]
02:32 -!- orospakr [~orospakr@CPE00a00cc1c1a6-CM00407b87b341.cpe.net.cable.rogers.com] has joined #uml
03:00 -!- Newsome [~sorenson@byu-gw.customer.csolutions.net] has quit [Quit: Leaving]
03:01 -!- sunny [~sunny@user-12hdttp.cable.mindspring.com] has quit [Ping timeout: 480 seconds]
03:16 -!- protocol [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has joined #uml
03:16 < protocol> wasssup
03:17 -!- _sorcer|katie [~sorcer@sorcer.sponsor.oftc.net] has quit [Ping timeout: 480 seconds]
03:18 -!- proto [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has joined #uml
03:18 -!- protocol [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has quit [Read error: Connection reset by peer]
03:18 -!- proto [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has quit [Quit: ]
03:19 -!- protocol [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has joined #uml
03:27 -!- proto [~protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has joined #uml
03:27 -!- protocol [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has quit [Read error: Connection reset by peer]
03:28 -!- dunc [~dunc@geology.thebunker.net] has joined #uml
03:33 -!- sunny [~sunny@user-12hdttp.cable.mindspring.com] has joined #uml
03:54 -!- rus [~rghf@jvds.demon.co.uk] has joined #uml
04:03 -!- sikkh [~peter@213.25.96.172] has joined #uml
04:04 -!- sikkh [~peter@213.25.96.172] has quit [Quit: ]
04:04 -!- sikkh [~peter@213.25.96.171] has joined #uml
04:07 -!- Heinachi [~krylen@ip68-14-129-139.ri.ri.cox.net] has quit [Quit: Amry of Me]
04:14 -!- DTE [~pier@host130-93.pool8019.interbusiness.it] has quit [Remote host closed the connection]
04:25 -!- DTE [~pier@host130-93.pool8019.interbusiness.it] has joined #uml
05:09 -!- knewt [~knew@p213.54.117.163.tisdip.tiscali.de] has quit [Ping timeout: 480 seconds]
05:32 -!- UKDolphin [~UKDolphin@support.thermeoneurope.com] has joined #uml
05:34 < UKDolphin> hi, I dont know if anyone can help but i am a little confused on whats needed for uml?
05:34 < UKDolphin> I have a standard 2.6 kernel, which the computer is running on.
05:35 < UKDolphin> I believe i need to apply the skas patch to it to enable uml??
05:35 < UKDolphin> then do i build another kernel with ??? as the client system
05:35 -!- cros [~cr@frn9-d9bb2f9a.pool.mediaWays.net] has quit [Ping timeout: 480 seconds]
05:40 < sikkh> skas is pretty usefull
05:41 < sikkh> imo you should add skas to host and use 2.4.24-um1 as guest system
05:41 < UKDolphin> do i need to just add the skas patch or a uml patch as well to the host?
05:41 < UKDolphin> this is where i am lost
05:42 < sikkh> only skas
05:42 < UKDolphin> ok,
05:43 < UKDolphin> then when i have built the host system i take it i build a guest kernel which has only the uml patch applied ?
05:43 < sikkh> yes
05:44 < UKDolphin> that makes more sense :) thanks
06:14 -!- nextime [~nextime@danex.i-m-c.it] has quit [Quit: nextime has no reason]
06:16 -!- nextime [~nextime@danex.i-m-c.it] has joined #uml
06:38 -!- serving [~serving@213.186.191.31] has quit [Read error: Connection reset by peer]
06:39 -!- AquaJo [~torbofh@pD95596C7.dip.t-dialin.net] has joined #uml
07:24 -!- dunc [~dunc@geology.thebunker.net] has quit [Quit: Client exiting]
07:28 -!- cros [~cr@frn9-d9bb2fbd.pool.mediaWays.net] has joined #uml
07:31 -!- dunc [~dunc@geology.thebunker.net] has joined #uml
08:02 < frink> hey
08:26 -!- knewt [~knew@p213.54.89.187.tisdip.tiscali.de] has joined #uml
08:30 -!- knewt [~knew@p213.54.89.187.tisdip.tiscali.de] has quit [Remote host closed the connection]
08:30 -!- knewt [~knew@p213.54.89.187.tisdip.tiscali.de] has joined #uml
08:34 -!- serving [~serving@213.186.191.31] has joined #uml
09:10 -!- UKDolphin [~UKDolphin@support.thermeoneurope.com] has quit [Quit: ]
09:21 -!- sigjuice [~sigjuice@c-24-6-51-184.client.comcast.net] has joined #uml
09:26 -!- orospakr [~orospakr@CPE00a00cc1c1a6-CM00407b87b341.cpe.net.cable.rogers.com] has quit [Quit: Leaving]
09:30 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has joined #uml
09:36 < proto> sup knewt man
09:52 -!- janwil [~janwil@fia48-65-100.dsl.hccnet.nl] has quit [Ping timeout: 480 seconds]
09:52 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has quit [Remote host closed the connection]
09:59 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has joined #uml
10:00 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has joined #uml
10:02 -!- UKDolphin [~UKDolphin@support.thermeoneurope.com] has joined #uml
10:02 < UKDolphin> hi
10:03 < UKDolphin> i have managed to rebuild my host kernel and am now up on that, 2.6.6 + skas i think
10:03 < UKDolphin> I have also build a guest kernel 2.6.4+uml
10:03 < UKDolphin> I have made a root image which i think may work but I dont know a) if i have done it right or b) how to start it :)
10:11 < ljlane> The HOWTO at user-mode-linux.sf.net has a lot of excellent info.
10:15 -!- DTE [~pier@host130-93.pool8019.interbusiness.it] has quit [Remote host closed the connection]
10:15 -!- DTE [~pier@host130-93.pool8019.interbusiness.it] has joined #uml
10:20 -!- orospakr [~orospakr@216.221.36.160] has joined #uml
10:23 -!- Newsome [~sorenson@sorenson.dsl.csolutions.net] has joined #uml
10:30 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has quit [Remote host closed the connection]
10:56 -!- Newsome [~sorenson@sorenson.dsl.csolutions.net] has left #uml [Leaving]
10:57 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has joined #uml
10:59 -!- DTE [~pier@host130-93.pool8019.interbusiness.it] has quit [Remote host closed the connection]
11:00 -!- DTE [~pier@host130-93.pool8019.interbusiness.it] has joined #uml
11:40 -!- Newsome [~sorenson@asterix.cs.byu.edu] has joined #uml
11:40 -!- schlumpf [~vx_schlum@dsl-082-083-063-060.arcor-ip.net] has quit [Read error: Connection reset by peer]
11:40 -!- orospakr_ [~orospakr@216.221.36.160] has joined #uml
11:41 -!- schlumpf [~vx_schlum@dsl-082-083-048-071.arcor-ip.net] has joined #uml
11:41 -!- orospakr [~orospakr@216.221.36.160] has quit [Ping timeout: 480 seconds]
11:50 -!- giuppi [~giuppi@213-140-22-64.fastres.net] has joined #uml
11:54 -!- jdike [~jdike@12.177.129.25] has joined #uml
11:54 < jdike> hi guys
11:54 < sikkh> hi jdike
11:54 -!- UKDolphin [~UKDolphin@support.thermeoneurope.com] has quit [Quit: ]
11:54 < jdike> sikkh: your stuff is merged, I'm cleaning up the interfaces and making it build right now
11:55 < proto> hi jman
11:55 < sikkh> cool
11:56 < frink> hey
11:56  * sikkh can't wait for -um3
11:56 < proto> hi frink
11:56 < sikkh> jdike: maybye do 2.4.26-um1?
11:56 -!- LP|WaZaR [~NNSCRIPT@ANeuilly-109-1-5-74.w81-49.abo.wanadoo.fr] has joined #uml
11:57 < frink> Anybody have debian working on 2.6 UML?
11:57 < sikkh> I use debian stable but under 2.4 UML
11:58 < frink> it breaks on 2.6
11:58 < frink> because /dev got mangled
11:58 < jdike> sikkh: I'll catch up after the next patch
11:59 < sikkh> jdike: we get new patch today?
11:59 < sikkh> btw meta_tdb works now
11:59 < sikkh> gotta still clean it up (code)
11:59 < jdike> sikkh: maybe, I've got some other stuff to do this afternoon
12:00 < jdike> sikkh: cool
12:01 < sikkh> jdike: wanna see the mess (tm) ;)
12:01 < jdike> sikkh: you can get working on meta_xattr, meta_symlinks, meta_etc, ... now
12:03 < sikkh> jdike: meta_symlinks, why?
12:04 < jdike> sikkh: putting metadata in symlinks doesn't consume inodes
12:04 < jdike> sikkh: you get really ugly dangling symlinks, though
12:04 < sikkh> symlink is an inode pointing to other inode (dir entry)
12:04 < sikkh> heh
12:04 < sikkh> at lest according to ext2 ;)
12:05 < sikkh> symlink is an inode alias but it still uses one inode
12:06 < jdike> sikkh: hmmm, yeah, just checked
12:06 < jdike> sikkh: I thought they were inlined
12:08 -!- LP|WaZaR [~NNSCRIPT@ANeuilly-109-1-5-74.w81-49.abo.wanadoo.fr] has quit [Quit: ( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de )]
12:09  * sikkh is happy to sometimes know more than his teacher :)
12:10 < jdike> hehe
12:10  * jdike watches sikkh's head swell
12:10 < sikkh> lol
12:12 -!- LP|WaZaR [~NNSCRIPT@ANeuilly-109-1-5-74.w81-49.abo.wanadoo.fr] has joined #uml
12:19 -!- sunny [~sunny@user-12hdttp.cable.mindspring.com] has quit [Quit: sunny]
12:23 < frink> is there a fedora text based package menager that fixes deps etc?
12:36 -!- orospakr_ [~orospakr@216.221.36.160] has quit [Quit: BBL.]
12:37 -!- proto- [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has joined #uml
12:37 -!- proto- [protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has quit [Quit: ]
12:37 -!- proto [~protocol@dhcp216-12-239-140.hiaw.dal.wayport.net] has quit [Read error: Connection reset by peer]
12:41 -!- Cowboy [~Cowboy@adsl-217-167-16.bgk.bellsouth.net] has quit [Remote host closed the connection]
12:47 -!- cowboy [~cowboy@adsl-217-167-16.bgk.bellsouth.net] has joined #uml
12:51 -!- G2 [~ghenry@212.159.59.85] has joined #uml
12:53 < giuppi> i'm trying to use 2.6.6 host following instruction at http://www.user-mode-linux.org/~blaisorblade/, it says to apply http://www.user-mode-linux.org/~blaisorblade/patches/uml-patch-2.6.6-01.bz2 but it doesn't apply cleanly. any idea?
12:59 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has quit [Ping timeout: 480 seconds]
13:00 -!- sigjuice [~sigjuice@c-24-6-51-184.client.comcast.net] has quit [Remote host closed the connection]
13:10 -!- LP|WaZaR [~NNSCRIPT@ANeuilly-109-1-5-74.w81-49.abo.wanadoo.fr] has left #uml []
13:16 < cros> frink: debian runs fine on 2.6 uml
13:21 < jdike> woo
13:21 < jdike> it's threatening to build
13:21 < jdike> and it did
13:22 < jdike> let's see if it runs
13:29 -!- giuppi [~giuppi@213-140-22-64.fastres.net] has quit [Quit: leaving]
13:31 -!- dunc [~dunc@geology.thebunker.net] has quit [Quit: Client exiting]
13:38 < jdike> I'll be damned
13:38 < jdike> it almost works
13:39 < jdike> it can't create links, but that's the only visible bootup problem
13:43 -!- serving- [~serving@213.186.191.31] has joined #uml
13:45 < sikkh> jdike: can't create links? :|
13:45 < sikkh> jdike: it worked here
13:45 < sikkh> jdike: hardlinks or soft?
13:47 < jdike> sikkh: well, after all the extra stuff I did, that's surprisingly good
13:47 < jdike> sikkh: hardlinks
13:47 < sikkh> hehe
13:47 < sikkh> jdike: meta fs just hardlinks metadata
13:47 < sikkh> jdike: if you make a hardlin
13:47 < sikkh> k
13:47 < jdike> sikkh: yeah, it's probably some stupid mistake
13:48  * jdike goes on a bug hunt
13:48 < sikkh> :)
13:48 < sikkh> jdike: take along gdb rifle ;)
13:49  * jdike is hunting the wascawy wabbit
13:49 < sikkh> jdike: watch what's going on in meta_fs.c:create_link
13:49 < sikkh> hehe
13:49 < jdike> sikkh: looking at that now
13:49 < jdike> sikkh: it's very innocent-looking
13:50 -!- frediz [~frediz@blackbox.ahau-kin.org] has joined #uml
13:50 < sikkh> :)
13:53 < jdike> hmmm
13:53 < jdike> Cannot create link /etc/mtab~
13:53 < jdike> immediately after I watch it succeed
13:54 < sikkh> huh
14:00 < jdike> unlink was broken
14:01 < jdike> much better
14:01 < green> yeah, who needs unlink anyway? ;)
14:02 < jdike> hehe
14:02  * jdike thwaps green
14:03  * jdike starts the obligatory kernel build
14:05 < jdike> if this works, I'd say it's -um3 time
14:09 < sikkh> green: yah, not reiserfs ;)
14:09 < sikkh> green: btw the reiser guys are working on release to latest -mm, but they give us no 2.6.6 snapshot! :|
14:10 < sikkh> and nikita said we can do it ourselves (port from 2.6.6-rc3-mm2 to 2.6.6) ourselves ;(
14:11 < green> yeah, so go port it if you need it. That's how it usually works in opensource ;)
14:11  * green experiences ext3 corruptions
14:11 < sikkh> green: pls, pls help us ;)
14:13  * green fixed two reiserfs v3 bugs today already. ;)
14:14 -!- sunny [~sunny@128.238.23.97] has joined #uml
14:14 < sikkh> green: and I thought reiser3 was stable!!! :)
14:17 < green> yes. These are odd cases like if you cannot allocate memory during mount it will crash
14:19 < sikkh> jdike: working?
14:20 -!- giuppi [~giuppi@213-140-22-64.fastres.net] has joined #uml
14:20 < giuppi> hi! my guest linux hangs at "Initializing stdio console driver" what does this mean?
14:21 < sikkh> no idea, Darkman ? ;)
14:21 < giuppi> :\
14:25 < sikkh> jdike: btw how did you get the whole / onto humfs, do you have new humfsify (with file mode support)?
14:25 -!- rus [~rghf@jvds.demon.co.uk] has quit [Ping timeout: 480 seconds]
14:26 < jdike> sikkh: the mode support isn't there
14:27 < sikkh> :(((((((((((((((((((((((((((((
14:27 < jdike> sikkh: I left it out and used my existing humfs
14:27 < jdike> sikkh: easier that way
14:27 < jdike> well, the kernel build worked
14:27 < sikkh> and you say you merged my patch??? :)
14:27 < jdike> sikkh: bah
14:27 < jdike> ungrateful people
14:27 < sikkh> sheesh I thought mode support is importand stuff
14:28 < sikkh> will it get there anyway?
14:28 < jdike> sikkh: yup
14:28  * sikkh is tearing his hair out and envisioning 'merging' :|
14:28 < jdike> sikkh: just that the mode support requires bumping the humfs version number and releasing a new humfsify
14:29 < sikkh> jdike: I don't want to be one of those linux developers who never get their patches in and always have to work hard too keep their stuff up to date with vanilla :|
14:29 < jdike> sikkh: I did that for more than three years
14:29 < sikkh> jdike: then let's fscking write the humfsify
14:29 < sikkh> :(((
14:30 < jdike> sikkh: against two trees
14:30  * sikkh is depressed
14:30  * sikkh is gone to weep in lonliness
14:30  * jdike plays sad music
14:31  * giuppi looks jdike and sikkh 
14:32 -!- orospakr [~orospakr@216.221.36.160] has joined #uml
14:33 < sikkh> jdike: ok, just please tell me which fetures you drop so I can prepare and not be amazed during merge against your release?
14:34 < jdike> sikkh: the mode stuff
14:34 < sikkh> ok
14:34 < sikkh> fortunately it's not too much code
14:35 < jdike> sikkh: nope, and there are /* XXX */ comments wherever I left something out
14:35 < sikkh> ok
14:39 < giuppi> where do i take uml patch for 2.4.26? does it exist?
14:40 < sikkh> apply 2.4.24-1 UML patch
14:40 < sikkh> jdike: -um3 ready?
14:42 < jdike> sikkh: looking at the patch now
14:42 < jdike> sikkh: 3000 lines of changes
14:48 -!- frediz [~frediz@blackbox.ahau-kin.org] has quit [Quit: leaving]
14:49 < albino> giuppi: Let me know if you get that to build, I'm still having problems with that
14:49 -!- frediz [~frediz@blackbox.ahau-kin.org] has joined #uml
14:49 < giuppi> albino: what kind of problems?
14:50 < albino> Ones like these: /usr/lib/libc.a(elf-init.o)(.text+0x8b): In function `__libc_csu_fini':
14:50 < albino> : undefined reference to `__fini_array_start'
14:50 < albino> collect2: ld returned 1 exit status
14:50 < albino> make: *** [linux] Error 1
14:51 < albino> I get about 10 lines like the first two and then what I've just pasted
14:51 < giuppi> albino: i'm patching it right now
14:51 < giuppi> with 2.4.24-1 uml patch
14:51 < albino> giuppi: Yeah, thats the patch I applied
15:22 < Darkman> sikkh: probably no console drivers turned on in the uml config (or at least no useable)
15:22 < sikkh> giuppi: probably no console drivers turned on in the uml config (or at least no useable) ;)
15:22 < sikkh> Darkman: thx :)
15:23 < jdike> Wow
15:23 < jdike> My UML is way faster than SF's CVS server
15:23 < Darkman> jdike: thats not hard... ;)
15:23 < sikkh> hehe
15:23 < Darkman> sikkh: no prob, youre welcome ;)
15:23 < sikkh> jdike: gimmie -um3 :)
15:25 < jdike> sikkh: thanks to the CVS move, you're gonna get it faster :-)
15:25 < sikkh> why? :)
15:25 < sikkh> I hoped for a nice unified patch :)
15:26 < jdike> sikkh: well, I check stuff into cvs first
15:26 < Darkman> cause his uml is the new cvs server and so much faster then sf :)
15:26 < jdike> sikkh: and that went very fast
15:26 < sikkh> hehe
15:26 < sikkh> jdike: I hope you have local repository too in case .sf.net fails ;)
15:26 < Darkman> btw. jdike, is there a useful job a uml can do for the community?
15:26 < Darkman> (like beeing an cvs server ;)
15:33 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has joined #uml
15:33 < giuppi> sikkh: you were right thank you
15:33 < sikkh> see Darkman :)
15:33 < sikkh> thank Darkman giuppi 
15:34 < giuppi> ahh ... :)
15:34 < giuppi> thanks Darkman :-)
15:34 < Darkman> no prob
15:34 < sikkh> Darkman is the certified UML patch advisor (tm) ;)
15:34 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has quit [Quit: ]
15:35 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has joined #uml
15:37  * Darkman awaits new cert ;)
15:37 -!- schlumpf [~vx_schlum@dsl-082-083-048-071.arcor-ip.net] has quit [Remote host closed the connection]
15:38 < sikkh> Darkman: yeach we gotta bug jdike for those ;)
15:39 -!- frediz [~frediz@blackbox.ahau-kin.org] has quit [Read error: Connection reset by peer]
15:40 -!- frediz [~frediz@blackbox.ahau-kin.org] has joined #uml
15:40 < jdike> Darkman: are you a mirror yet?
15:40  * jdike doesn't remember
15:41 < Darkman> jdike: not yet cause of the maybe high traffic ;)
15:41 < sikkh> hehe
15:41 < sikkh> Darkman could use some slashdotting
15:43 < sikkh> http://prdownloads.sourceforge.net/user-mode-linux/uml-patch-2.4.24-3.bz2 there yet?
15:44 < jdike> sikkh: as of 2 seconds ago, yes
15:44 < sikkh> yeach :)
15:44 < sikkh> gotta test that jdike's AIO goodies ^_^
15:45 < sikkh> jdike: can I apply to 2.4.26?
15:45 < sikkh> I think I should be able to
15:46 < jdike> sikkh: should work as well -2
15:56  * AquaJo starts a manual syncrun on the mirror
15:57  * sikkh looks what jdike did to his precious metadata.h
16:01 < sikkh> jdike: where did hostf_* stuff go?
16:01 < AquaJo> hey, we ned a new topic :-)
16:04 < jdike> sikkh: hostfs_kern.c is now externfs.c
16:04 < jdike> sikkh: hostfs_user.c is now host_fs.c
16:05 < sikkh> jdike: ok thx
16:06 < jdike> sikkh: how's it look?
16:07 < sikkh> I'm trying to figure out merging :)
16:07 < sikkh> I'll have more idea tomorrow
16:08 -!- G2 [~ghenry@212.159.59.85] has quit [Quit: oops]
16:08 < sikkh> jdike: nice you left utility functions like host_get_mode()
16:08 < sikkh> jdike: I got them out and made into functions since the're generaly usefull
16:09 < sikkh> jdike: file_handles everywhere ;)
16:09 < sikkh> guess it's the fd reclaiming stuff
16:09  * AquaJo has finished reading the changelog
16:10 < AquaJo> is the O_Direct support in the ubd-driver too?
16:10 < sikkh> jdike: why did you remove host_get_type()?
16:10 < AquaJo> so it might save memory
16:12 < jdike> AquaJo: not yet
16:12 < jdike> AquaJo: that's one of the next things
16:12 < AquaJo> ok
16:12 < jdike> sikkh: yeah, file reclaiming everywhere
16:12 < jdike> sikkh: there was only one user of it
16:13 < sikkh> jdike: can I readd host_get_type()? it's used in meta_tdb
16:13 < jdike> sikkh: OK, sure
16:13 < jdike> sikkh: as long as there are multiple users
16:13 < sikkh> jdike: multiple code using it?
16:13 < sikkh> sure
16:14 < sikkh> it's the reason I ripped that functionality the first place
16:16 -!- frediz [~frediz@blackbox.ahau-kin.org] has quit [Quit: zouuu]
16:18 < sikkh> jdike: btw how do you merge to your cvs tree, emacs patch magic?
16:18 < jdike> sikkh: just checkins
16:18 < sikkh> jdike: I'm looking at the differences to your files and edit given two files (mine, yours) at the same time
16:19 < sikkh> jdike: in my checkout dir I have your new versions, I just 'update them' to my needs
16:19 < sikkh> jdike: and cvs diff tels me what's different from my tree
16:19 < sikkh> jdike: when I'll finish I'll just commit it
16:21 < sikkh> jdike: is my way ergonomic?
16:21 < sikkh> I ask since you must have much more experience doing that
16:22 < jdike> sikkh: given that you want to maintain your own CVS, that's fine
16:22 < sikkh> ok, thx
16:25 < sikkh> jdike: that's intresting "struct meta_fs *meta = container_of(humfs, struct meta_fs, humfs);"
16:27 < jdike> sikkh: that's where your meta_data went
16:28 < jdike> sikkh: rather than the meta fs data being inside the humfs data, it's the other way around
16:28 < jdike> sikkh: saves some kmallocing
16:28 < sikkh> ok
16:28 < jdike> sikkh: and gets rid of a bunch of void *'s
16:35 < sikkh> jdike: what's (*init_file)(const char *path, struct inode *inode, ... for?
16:36 -!- rus [~rghf@jvds.demon.co.uk] has joined #uml
16:37 < sikkh> jdike: (*invisible)(struct humfs_file *hf);?
16:37 < jdike> sikkh: that's when an inode is created
16:37 < sikkh> inode?
16:37 < jdike> sikkh: the host file gets opened then, rather than when a UML process calls open
16:38 < jdike> sikkh: invisible is called just before the last unlink happens
16:38 < sikkh> jdike: it's for fd stuff?
16:38 < jdike> sikkh: if there was a descriptor open to it and it was closed by the reclaimer, then it had better be reopened, otherwise the file is going to disappear
16:39 < sikkh> jdike: that's init_file for, right?
16:39 -!- sunny [~sunny@128.238.23.97] has quit [Quit: sunny]
16:40 < sikkh> looks like a lot 'restructuring' of meta_tdb achead of me
16:40 < jdike> sikkh: no, that's invisible
16:41 -!- serving- [~serving@213.186.191.31] has quit [Quit: ]
16:42 < sikkh> jdike: I need to pass gcc -DSTANDALONE for tdb.c, how do I do it in Makefile?
16:42 < jdike> sikkh: what does that mean?  That it's about to be linked into something else?
16:43 < sikkh> jdike: yeach, standalone tdb mode
16:43 < jdike> sikkh: what's the other choice?
16:43 < sikkh> jdike: I wish to avoid changing tdb code, to make it easy to merge new versions
16:44 < jdike> sikkh: but what does STANDALONE mean?
16:44 < sikkh> jdike: ok I can do it in tdb.h
16:44 < jdike> sikkh: especially since you're linking it into something else
16:44 < sikkh> jdike: that tdb is not compiled into shared lib
16:44 < sikkh> jdike: using ./configure
16:44 < jdike> sikkh: ah
16:44 < jdike> sikkh: you get a .a or .o or something instead?
16:45 < jdike> sikkh: yeah, you can do it in your Makefile
16:45 < sikkh> jdike: nevermind I'll add #define to tdb.h
17:05 < sikkh> jdike: I use reiserfs v3.6 on host, should I enable "Always do synchronous disk IO for UBD"?
17:06 -!- wavefunction [~macaiba@24-63-17-200.nc-rj.rnp.br] has joined #uml
17:06 < wavefunction> g'nite.
17:08 < wavefunction> I'm trying to install crux linux as uml. My uml sources are 2.6.4. During install I get "depmod: QM_MODULES: Function not implemented". Does this have to do with the uml or the installation ?
17:10 < jdike> sikkh: no
17:10 < jdike> wavefunction: you need new modutils, just like everyone else who is running 2.5
17:10 < jdike> 2.6
17:11 < sikkh> jdike: ok, thx
17:12 < wavefunction> OK, jdike. Thanks, again ! :)
17:16 < sikkh> jdike: looking at your stuff I must admit you keep your standards high, I hope I can keep up
17:17 < jdike> sikkh: heh
17:17 < jdike> sikkh: what in particular?
17:17 < sikkh> generally, I'm lost in the new code :)
17:17 < sikkh> it'll take few days to get me up to speed (I code UML only in evenings)
17:18 < jdike> sikkh: ah
17:18 < jdike> sikkh: "I don't understand it, so it must be brilliant"
17:19 < sikkh> hehe
17:19 < sikkh> yup, kinda that feeling
17:19 < sikkh> looks more sophisticated to -um2
17:20 < jdike> sikkh: yeah, that's what I spent yesterday doing
17:20 < jdike> sikkh: I'm much happier with this code now
17:21 < sikkh> I guess so :)
17:25 < sikkh> ok, my code is still in limbo but I'll play around with jdike stuff
17:27  * jdike announces 2.4.24-3
17:31 < sikkh> Inconsistency detected by ld.so: dl-fini.c: 66: _dl_fini: Assertion `i == _dl_nloaded' failed! ;(
17:42 < sikkh> well 2.4.24-3 fails for me when used with 2.4.26
17:42 < sikkh> gotta use stock 2.4.24
17:52 < jdike> sikkh: that's on a humfs boot?
17:52 < sikkh> nope ubd
17:53 < jdike> sikkh: strange
17:53 < jdike> sikkh: that hasn't changed
17:59 -!- wavefunction [~macaiba@24-63-17-200.nc-rj.rnp.br] has left #uml [g'nite to all - time to go !]
18:09 -!- tchang [tchang@ribbit.roadtoad.net] has quit [Remote host closed the connection]
18:11 -!- schlumpf [~vx_schlum@dsl-082-083-048-071.arcor-ip.net] has joined #uml
18:15 -!- sikkh [~peter@213.25.96.171] has quit [Quit: sleep]
18:27 -!- orospakr [~orospakr@216.221.36.160] has quit [Quit: Leaving]
19:00 -!- schlumpf [~vx_schlum@dsl-082-083-048-071.arcor-ip.net] has quit [Remote host closed the connection]
19:06 -!- loko [private@c-67-165-107-230.client.comcast.net] has quit [Read error: Connection reset by peer]
19:06 -!- loko [private@c-67-165-107-230.client.comcast.net] has joined #uml
19:06 -!- Newsome [~sorenson@asterix.cs.byu.edu] has quit [Quit: Leaving]
19:26 -!- orospakr [~orospakr@CPE00a00cc1c1a6-CM00407b87b341.cpe.net.cable.rogers.com] has joined #uml
19:30 -!- Seamus [~Seamus@c-24-16-36-18.client.comcast.net] has joined #uml
19:30 < Seamus> hello!
19:31 < Seamus> I have a (hopefully) simple question
19:35 < Seamus> When I run a command like: ./linux-2.4.24-pre3-djc4-1um ubd0=root_fs.fc-2-client.pristine.20040504 con0=fd:0,fd:1 con=/dev/null
19:36 < Seamus> should I then get a login prompt on stdin/out (e.g. my shell)
19:36 < Seamus> ?
19:36 < Seamus> I seem to be having poor luck in that regard.
19:36 < caker> con=null, not con=/dev/null
19:36 < caker> and make sure the uml's /etc/fstab has a getty listening on tty0 (or tty/0 for devfs)
19:37 < caker> er, that's /etc/inittab, not fstab
19:37 < Seamus> Ah.  OK.
19:37 < Seamus> yeah, I figured that :)
19:37 < Seamus> allthough, now its hanging at: 
19:37 < Seamus>                Welcome to Fedora Core
19:37 < Seamus>                 Press 'I' to enter interactive startup.
19:37 < Seamus> Configuring kernel parameters:
19:39 < Seamus> Thanks for the pointer on /etc/inittab and null, btw
19:40 < Seamus> I seem to be having more trouble getting this going than I think is normal.
19:40 < Seamus> must be my anti-linux curse :(
19:43 -!- Guy- [~korn@chardonnay.math.bme.hu] has quit [Ping timeout: 480 seconds]
19:43 < Seamus> another question- the root command line switch tells the guest kernel where to find the root disk, correct?
19:43 < Seamus> If so, what does "root=6201" mean?  I see that many places.  Is that a major and minor number?
19:43 -!- knewt [~knew@p213.54.89.187.tisdip.tiscali.de] has quit [Ping timeout: 480 seconds]
19:45 -!- knewt [~knew@p213.54.84.251.tisdip.tiscali.de] has joined #uml
19:46 < knewt> Seamus: did any of what i sent get through before my link went dead?
19:46 < Seamus> no, sorry
19:47 < knewt> 01:31 < knewt> it depends. do you have a getty running on tty0 in your inittab, or do they start at tty1?
19:47 < knewt> 01:31 < knewt> the inittab in your uml fs
19:47 < knewt> 01:32 < knewt> also, it should be con=null, not con=/dev/null
19:47 < knewt> 01:34 < knewt> if you /not/ running a getty on tty0, and you don't want to, then con0=null,fd:1 con1=fd:0,fd:1 con=null should 
19:47 < knewt>                do the trick
19:47 < knewt> 01:34 < knewt> finally, did you make sure the necessary console drivers were built into the uml kernel?
19:49 < Seamus> I'm using Bill Stearns fedora image, so I haven't checked the inittab yet- but I will. If they start at tty1, then I should get a login at whatever I point con1 to?
19:49 < Seamus> and con0 will just be the boot messages?
19:50 -!- Newsome [~sorenson@byu-gw.customer.csolutions.net] has joined #uml
19:50 < Seamus> and could you point me to which console drivers are needed?  Thats just done through the kernel config file, I presume.
19:52 -!- knewt_ [~knew@p213.54.85.205.tisdip.tiscali.de] has joined #uml
19:52 -!- knewt is now known as Guest373
19:52 -!- knewt_ is now known as knewt
19:52 < Seamus> thanks for the info/help knewt!
19:52 < Seamus> did you happen to see my next question (sorry to be a pest)
19:53 < knewt> sorry, no. i really should have learnt by now that when my link plays up i need to fully restart the router, but i still try and work around it, to my detriment
19:53 < knewt> didn't see anything after i pasted from my scrollback
19:54 < Seamus> hehe.  I'll repeat, I've been banging my head against this for a few days.
19:54 < Seamus> so its great to get some friendly advice
19:54 < Seamus> <Seamus> I'm using Bill Stearns fedora image, so I haven't checked the inittab yet- but I will. If they start at tty1, then I should get a login at whatever I point con1 to? 
19:54 < Seamus> <Seamus> and con0 will just be the boot messages? 
19:54 < Seamus> *** Newsome (~sorenson@byu-gw.customer.csolutions.net
19:55 < knewt> yep, that's right. the boot messages always go to con0. if you want you can put a getty on con0 to put a login there as well
19:55 < Seamus> ah, I get it.
19:55 < Seamus> and inside the innittab, that will be /dev/tty0
19:55 < Seamus> correct?
19:55 < knewt> as long as you're not using devfs, yes
19:56 -!- Guest373 [~knew@p213.54.84.251.tisdip.tiscali.de] has quit [Ping timeout: 480 seconds]
19:56 < Seamus> I'm setting this up as a shared hosted box for some friends, so I was planning on running conserver to the consoles to automate the uml's coming up and as backup logins for horked networks.
19:56 -!- knewt [~knew@p213.54.85.205.tisdip.tiscali.de] has quit [Quit: ]
19:56 < Seamus> and devfs does what?
19:56 -!- knewt [~knew@p213.54.85.205.tisdip.tiscali.de] has joined #uml
19:57 < knewt> gah, i don't believe this. first my link goes down twice, and then i accidentally kill irssi
19:57 < Seamus> hehe
19:57 -!- admin [~admin@dsl-128-167.aei.ca] has joined #uml
19:57 < Seamus> I find references to devs in the ubd section of the kernel swiches, but not anywhere else
19:57 < Seamus> I know nodevfs is a kernel switch- but I'm not sure what it does?
19:58 < AquaJo> ist disables devfs
19:59 < Seamus> and devfs is just /dev?
19:59 < admin> hi there. Did anyone succeeded in compiling 2.6.4 kernel
19:59 < Seamus> can you tell I'm a BSD guy?
19:59 < AquaJo> Seamus: devfs is a filesystem witch generates /dev at runtime
20:00 < Seamus> I take it in linux you can access ttys/pttys via methods other than /dev, eh?
20:00 < AquaJo> the traditional way ist to create the Files on a harddisk
20:00 < Seamus> using mknod and such.
20:00 < Seamus> I guess this just does it dynamically at boot?
20:01 < AquaJo> yes
20:01 < knewt> not just boot though
20:01 < knewt> as devices come and go
20:01 < Seamus> ah.
20:01 < AquaJo> it does it dynaicaly at boot and runntime (for hot-plugable devices)
20:01 < Seamus> so if I attach a usb device, it dynamically gets added to /dev/foo
20:01 < AquaJo> and it has an different naming
20:01 < Seamus> rather than having to be done statically
20:01 < AquaJo> if you us devfs, yes
20:02 < admin> hi there. Did anyone succeeded in compiling 2.6.4 uml kernel?
20:02 < Seamus> is there a good reason not to run devfs in uml?
20:02 < AquaJo> I'm fine with it, and im runing more than 1 uml
20:03 < AquaJo> if you run debian, devfsd is a nice helper
20:03 < Seamus> I've built  2.6.4 uml kernel, but I don't think I'm a good one to declare it a success quite yet :)
20:03 < knewt> i don't use it anyone personally (well, not once i've rebuilt 1 last machine). udev on 2.6 is the way of the future
20:03 < knewt> s/anyone/anywhere/
20:03 < Seamus> udev is another dynamic way of doing devfs?
20:03 < knewt> udev is a userspace solution
20:03 < knewt> whereas devfs is kernelspace
20:04 < knewt> so udev keeps policy out of the kernel, as well as being much more flexible than devfs in to the bargin
20:04 < knewt> s/bargin/bargain/
20:04 < knewt> but it's a 2.6-only thing
20:04 < Seamus> ah.
20:04 < admin> anyway, can someone point me to rght place just to get it compiled?
20:05 < knewt> admin: see the topic
20:05 < Seamus> so, if I don't use devfs in the uml, I will have to mknod the appropriate ttys to get console's- correct?
20:06 < admin> Oh, sorry
20:06 < knewt> Seamus: nope, should all be there already. same device numbers are used inside uml as on normal systems
20:07 < knewt> for tty's, anyway
20:07 < Seamus> so, /dev/tty0 will be the boot console, and I put a getty running on that in order to get a login there.  But then you mentioned that was as long as I'm not using devfs...
20:08 < Seamus> would devfs put them somewhere else?
20:08 < knewt> /dev/vc/N instead of /dev/ttyN i think
20:09 < Seamus> knewt: and the command line switch to not use devfs is nodevfs correct?
20:10 < knewt> however, if you're running devfsd then aliases on the usual names will be present
20:10 < AquaJo>  devfs: /dev/vc/0
20:10 < AquaJo> or /dev/ttys/0
20:10 < Seamus> knewt: so, to verify- I should be ok with using nodevfs and then making sure getty runs on /dev/tty0
20:10 < knewt> i believe it's devfs=nomount
20:11 < knewt> however, "linux --showconfig | grep DEVFS" will tell you if your uml has it enabled or not
20:12 < knewt> what output do you get from that?
20:13 < Seamus> [darren@cappuccino uml]$ ./linux-2.4.24 --showconfig |grep DEVFS 
20:13 < Seamus> CONFIG_DEVFS_FS=y
20:13 < Seamus> CONFIG_DEVFS_MOUNT=y
20:13 < Seamus> # CONFIG_DEVFS_DEBUG is not set
20:13 < Seamus> so looks like its there
20:14 < Seamus> ok, when I run an image now, I get this:
20:14 < Seamus> Starting crond: [  OK  ] 
20:14 < Seamus> Starting xfs: [  OK  ] 
20:14 < Seamus>  
20:14 < Seamus>  
20:14 < Seamus>  
20:14 < Seamus> INIT: Id "0" respawning too fast: disabled for 5 minutes 
20:14 < knewt> yep. when you start the uml up, do you just not see a login prompt at the end of the boot, or do you see messages about inittab ?
20:14 < Seamus> INIT: Id "c" respawning too fast: disabled for 5 minutes 
20:14 < Seamus> INIT: Id "2" respawning too fast: disabled for 5 minutes 
20:14 < Seamus> INIT: Id "1" respawning too fast: disabled for 5 minutes 
20:14 < Seamus> INIT: no more processes left in this runlevel 
20:14 < Seamus>  
20:14 < Seamus> I'm thinking I need to go enable getty on /dev/tty0
20:14 < knewt> how are you starting it up?
20:14 < Seamus>  ./linux-2.4.24 ubd0=root_fs.fc-2-client.pristine.20040504 con0=fd:0,fd:1 con=null devfs=nomount
20:15 < knewt> ok, you need to leave devfs enabled, because inittab is using the devfs devices, not the normal ones
20:15 -!- admin [~admin@dsl-128-167.aei.ca] has quit [Quit: using sirc version 2.211+KSIRC/1.3.10]
20:15 < knewt> change the end to: con0=null,fd:1 con1=fd:0,fd:1 con=null
20:15 < Seamus> you happen to know Stearns default root password?
20:16 < Seamus> oh. root works
20:16 < knewt> heh :)
20:16 < Seamus> thanks!
20:16 < Seamus> finally!
20:16 < knewt> np. glad to be of help
20:17 < Seamus> so, now that I've gotten this far, a few more questions (sheepish grin)
20:17 < Seamus> :)
20:17 < Seamus> I'm trying to attach the boot consoles to a conserver-
20:17 < Seamus> or some other method to give people access to them and have them start at boot (e.g. screen)
20:18 < Seamus> but, I'm having a bit of trouble getting them to connect.
20:19 < knewt> have you seen the conserver howto?
20:19 < Seamus> yeah, I've followed it.
20:19 < Seamus> might try again, now that I'm actually able to login :)
20:19 < Seamus> when I point con0 to /dev/ptyy0 I can't get conserver to see the connection up at  /dev/ttyy0
20:20 < Seamus> I'm also confused as to how a pseudo-tty is turning into a tty.
20:20 < Seamus> e.g. why I don't configure conserver.cf to use /dev/ptty0
20:20 < knewt> afraid i've never played with that particular method of connecting consoles, so i'm not going to be much help with it
20:21 < Seamus> I'm not stuck on conserver, just something.  Does screen work well for you?
20:22 < knewt> well, lately i haven't been autoconnecting my consoles, only doing so if i need a local console.
20:23 < Seamus> how do you get them to boot then?  I thought con0 had to connect to something?
20:23 < knewt> ah, sorry. "other than the boot console, which i leave on stdin/out" should have been added there
20:24 < Seamus> ah, so you don't start multiple ones up at boot, you just use it for testing and such, I gather?
20:24 < knewt> so i just start the uml up inside a screen session, and then put all the other consoles onto pts's. Then if i need a local console i just look for a pts number in the boot log and connect to it
20:25 < knewt> i have had automated connection stuff in the past, but not got anything handy right now
20:25 < knewt> hmm, actually i do have it around, as it's running on my servers, i'd just forgotten about it as i haven't needed a local console on any of them for ages
20:26 < Seamus> so, do you use something like con1=tty:/dev/ptyz0 ?
20:26 < knewt> it's just the stuff i've been experimenting with lately i haven't been automating. the automated stuff i use was grabbed off the net anyway, from the list archives iirc
20:26 < Seamus> or con1=pty
20:27 < Seamus> OK, I'll go look.  I'm still confused about how con1=/dev/ptyy0 on the command line turns into /dev/ttyy0 for conserver/minicom etc.
20:27 < Seamus> thanks again for all the help.
20:27 < knewt> nah, i'd use con0=fd:0,fd:1 con=pts
20:27 < knewt> i think it's pts, anyway
20:27 < knewt> just let me check
20:27 < Seamus> gotcha
20:29 < Seamus> one last question before I go search around again- is there a good howto on what the different network helpers do and such- I'm just trying to set it up so that each uml uses a different address that is aliased to the hosts eth0- I can find scripts to do things, but no in depth documentation on what tunctl etc. actually do..
20:29 < knewt> yeah, looks like pts
20:29 < Seamus> cool, thanks!
20:29 < knewt> have you read user-mode-linux.sf.net/networking.html ?
20:30 < Seamus> not yet.  but I think I will :)
20:31 < Seamus> don't know how I missed that.
20:31 < knewt> user-mode-linux.sf.net is the main uml site
20:31 < Seamus> I really appreciate all the help
20:31 -!- sunny [~sunny@user-12hdttp.cable.mindspring.com] has joined #uml
20:31 < Seamus> yeah, I've looked through it.  Somehow missed the networking page.
20:31 < Seamus> that helps
20:31 < Seamus> I will stop bothering you now...  at least for a while.  This has really helped me get past a few things that were causing me to pull my hair out.
20:32 < knewt> np, glad to be of service
20:32 < Seamus> have a good one!
20:32 -!- Seamus [~Seamus@c-24-16-36-18.client.comcast.net] has quit [Quit: Trillian (http://www.ceruleanstudios.com)]
20:44 -!- giuppi [~giuppi@213-140-22-64.fastres.net] has quit [Quit: Lost terminal]
20:57 -!- loko [private@c-67-165-107-230.client.comcast.net] has quit [Ping timeout: 480 seconds]
21:03 -!- loko_ [private@c-67-165-107-230.client.comcast.net] has joined #uml
21:15 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has quit [Ping timeout: 480 seconds]
21:42 -!- sunny [~sunny@user-12hdttp.cable.mindspring.com] has quit [Quit: sunny]
21:56 -!- janwil [~janwil@cl-401.ams-01.nl.sixxs.net] has joined #uml
21:58 -!- jdike [~jdike@12.177.129.25] has quit [Quit: Leaving]
21:59 -!- cowboy [~cowboy@adsl-217-167-16.bgk.bellsouth.net] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- pirlouit [~peter@duvel.drunkcoders.com] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- vagrant [chip@r03rcbw62.desktop.umr.edu] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- lilo [levin@lilo.usercloak.oftc.net] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- ElectricElf [~david@electricelf.noc.oftc.net] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- mdz_ [~mdz@69.167.148.207] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- ichilton [~ian@cpc3-stoc3-4-0-cust188.midd.cable.ntl.com] has quit [uranium.oftc.net arion.oftc.net]
21:59 -!- lucia [~admin@37.Red-217-126-239.pooles.rima-tde.net] has quit [uranium.oftc.net arion.oftc.net]
22:00 -!- cowboy [~cowboy@adsl-217-167-16.bgk.bellsouth.net] has joined #uml
22:00 -!- pirlouit [~peter@duvel.drunkcoders.com] has joined #uml
22:00 -!- vagrant [chip@r03rcbw62.desktop.umr.edu] has joined #uml
22:00 -!- lilo [levin@lilo.usercloak.oftc.net] has joined #uml
22:00 -!- ElectricElf [~david@electricelf.noc.oftc.net] has joined #uml
22:00 -!- mdz_ [~mdz@69.167.148.207] has joined #uml
22:00 -!- ichilton [~ian@cpc3-stoc3-4-0-cust188.midd.cable.ntl.com] has joined #uml
22:00 -!- lucia [~admin@37.Red-217-126-239.pooles.rima-tde.net] has joined #uml
22:30 -!- sigjuice [~chatzilla@c-24-6-51-184.client.comcast.net] has joined #uml
22:31 -!- Brinzer [~Petruchio@acs-24-239-125-185.zoominternet.net] has joined #uml
22:31 -!- Brinzer is now known as Petruchio
22:33 < sigjuice> Hi
22:34 < Petruchio> Hello.
22:34 < sigjuice> I'm trying to run 2.4.18-17um, which is included with debian woody. I'm using root_fs_toms1.7.205
22:35 < sigjuice> I get stuck at INIT: Entering runlevel: 5 and I get messages like this after that. INIT: Id "c1" respawning too fast:
22:35 < sigjuice> the host kernel is a vanilla 2.4.25
22:41 < knewt> i'm not really here right now (need some sleep), but i'd guess that the fs is intended for devfs-enabled umls, and the debian uml isn't built that way. just a guess mind you, you should really open up the fs and take a look. seperately, however, i'd strongly urge you to upgrade to a more recent uml kernel (at least .24), as there have been bug fixes since the one you're running
22:42 -!- AquaJo [~torbofh@pD95596C7.dip.t-dialin.net] has quit [Ping timeout: 480 seconds]
23:17 -!- orospakr [~orospakr@CPE00a00cc1c1a6-CM00407b87b341.cpe.net.cable.rogers.com] has quit [Quit: Leaving]
--- Log closed Wed May 12 23:59:00 2004