--- Day changed

--- Log opened Wed Jul 30 00:00:01 2003
00:38 -!- green_ [~green@64.169.161.73] has joined #uml
00:38 -!- green_ [~green@64.169.161.73] has left #uml [Client Exiting]
01:32 -!- coryb [~cory@24.124.40.190] has quit [Ping timeout: 488 seconds]
02:31 -!- mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has quit [Ping timeout: 490 seconds]
02:56 -!- mdz [~mdz@216.15.124.77] has joined #uml
03:05 -!- mdz [~mdz@216.15.124.77] has quit [Ping timeout: 490 seconds]
03:18 -!- MikeH [~mike@62.56.2.141] has quit [Ping timeout: 488 seconds]
03:55 -!- mdz [~mdz@216-15-124-77.c3-0.smr-ubr3.sbo-smr.ma.cable.rcn.com] has joined #uml
04:36 -!- QSECOFR [~rocky@67.128.240.30] has quit [Ping timeout: 488 seconds]
04:44 -!- QSECOFR [~rocky@lab.atipa.com] has joined #uml
04:48 -!- G2 [~knoppix@193.195.148.66] has joined #uml
04:49 -!- G2 [~knoppix@193.195.148.66] has quit [Client Quit]
04:49 -!- G2 [~knoppix@193.195.148.66] has joined #uml
04:56 -!- G2 [~knoppix@193.195.148.66] has quit [Quit: Client exiting]
04:58 -!- MikeH [~mike@62.56.2.150] has joined #uml
05:08 -!- G2_ [~G2@193.195.148.66] has joined #uml
05:27 -!- MikeH [~mike@62.56.2.150] has quit [Ping timeout: 488 seconds]
08:38 -!- Jon [~jon@imhotep.hursley.ibm.com] has joined #uml
08:38 < Jon> hello all
08:38 < Getty> hiho
08:38 < Jon> I have a small Q.
08:38 < Jon> I am running UML 2.4.20-um6, and I am using UML for kernel module development
08:39 < Jon> thus, I need a copy of the correct kernel source tree. I have applied the uml-6 patch to 2.4.20 sources, it went ok, but I need to generate the file 'modversions.h' which I believe is the 'make dep' stage in traditional kernels
08:39 < Jon> however, I know there are problems with the patched source to uml with assembler headers or something
08:40 < Jon> my question is: should I just ignore mismatched kernel version warnings (2.4.20 != 2.4.20-um6) or is it possible to generate a working source tree for module development, including the asm routines and modversions.h? 
08:52 < Jon> the following three commands, in 2.4.20-um6 source tree, do not generate the modversions.h:
08:52 < Jon> # make mrproper
08:52 < Jon> # make menuconfig ARCH=um
08:52 < Jon> # make dep ARCH=um
09:05 -!- Jax [~computer_@148.100.209.103] has joined #uml
09:06 < Jon> hi
09:06 < Jax> did you figure out your problem yesterday?
09:06 < Jon> not quite
09:06 < Jon> but I built my own root image and that worked better so I'm using that :) thanks
09:06 < Jon> right now I'm fighting getting a kernel source tree setup properly
09:07 < G2_> Have you heard of smoothwall???
09:07 < Jon> yeah, the distribution
09:07 < Jon> brb
09:20 -!- G2_ [~G2@193.195.148.66] has quit [Ping timeout: 501 seconds]
09:43 -!- G2_ [~G2@193.195.148.66] has joined #uml
10:55 -!- mistral [mistral@212.159.71.212] has joined #uml
12:58 -!- miah [miah@sifl.digivill.net] has joined #uml
12:58 < miah> hello
12:59 < miah> i just wonder why the docs for uml use bash -c 'echo 1 > /proc/sys/net/ipv4/ip_forward' when 'sysctl -n -w net.ipv4.ip_foward=1' is easier and doesn't require a seperate shell spawn
13:06 -!- ticallion [~ticallion@213.175.160.212] has joined #uml
13:14 -!- yacc [~chatzilla@129.34.20.17] has joined #uml
13:18 < ticallion> urls on setting up a minimal install image?! cloning maybe?!
13:19 < yacc> hey, anyone here has tried the 2.6.0-test1 uml patch? I got errors during the compile.  I don't understand why such code like "void __kernel_vsyscall" in arch/um/kernel/um_arch.c exists.
13:19 < ticallion> yacc: same prob here
13:21 < yacc> ticallion, have tried to manually fix these errors?
13:21 < ticallion> yacc: nope, gave up and setup 2.4.20 successfully
13:23 -!- rlo [~roberto@200.138.100.242] has joined #uml
13:24 < rlo> can't compile uml-patch-2.6.0-test2-1
13:25 < rlo> arch/um/kernel/um_arch.c:51: storage size of `__kernel_vsyscall' isn't known
13:25 < rlo> anyone have a clue?
13:25 < yacc> rlo, I just mentioned this problem. :-)
13:26 < rlo> eh, I just arrived
13:26 < rlo> got an answer?
13:26 < yacc> nope. Actually, we can get around this by simply comment it out. But there are a lot of other problems.
13:27 < Jax> jeff dike said in the mailing list that the test1 patch is bad (same issue), but test2 should work
13:27 < rlo> this means it's hopeless trying to compile it?
13:27 < rlo> Jax: that
13:28 < rlo> Jax: that's why I'm giving it a shot
13:36 < miah> anybody know why as soon as i bring up my tap interface i lose networking?
13:37 < ticallion> miah: r u nat'ed?!
13:39 < ticallion> everybody here uses downloaded fs images?!
13:39 < miah> the host im loading uml on does not do nat
13:39 < miah> but ip_foward does = 1 and proxy_arp does = 1
13:40 < miah> but before i even set those i lost the networking, it was as soon as i did 'ifconfig ip tun0 up'
13:40 < miah> lost ssh into the box
13:40 < ticallion> if it's nat'ed perhaps, packets r being sent out with the new host-side ip as source and that's now allowed on ur gateway
13:41 < miah> as i said, there is no nat
13:41 < ticallion> u said there wasn'
13:41 < ticallion> t nat on the box
13:41 < ticallion> u didn't say anything about the box being nat'ed or not
13:43 < miah> the box has a real ip address reachable by anybody on the internet
13:43 < miah> and the ip i assigned to uml also has a similar ip
13:44 < miah> i can ping my gateway, but nothing beyond that
13:45 < rlo> yacc: Commenting out __kernel_vsyscasll solves the compile problem, but gives a linker problema, as you said... :(
13:45  * Jax apoligizes to rlo
13:46 < Jax> seems that people on the mailing list are still having that problem
13:46 < rlo> yacc: any workaround for this???
13:46 < miah> but even so, the original ip address should still work with tap0 up even if tap0 isn't generating any traffic (which it isnt because i haven't even loaded the uml kernel yet)
13:47 < rlo> Jax: hehe
13:47 < ticallion> miah: check what's changing in your routing tables
13:51 < miah> ah, i found it
13:51 < miah> ifconfig set the wrong netmask on the tap0 ip
13:52 < miah> i'll have to remember that for when i script all this
13:56 -!- miah [miah@sifl.digivill.net] has quit [Quit: ircII EPIC4-1.0.1 -- Are we there yet?]
14:00 < attriel> the 260test2-1 patch still has the vsyscall error; no one's quite sure how anyone's gotten it to compile :o
14:00 < attriel> and there's enough of the vsyscall stuff in the headers now that commenting it out is non-trivial (and probably not really possible, at this point, either)
14:02 < rlo> The question is, anyone did compile it?
14:02 < attriel> i think jdike did, yah
14:02 -!- ticallion [~ticallion@213.175.160.212] has quit [Quit: bbl]
14:03 < rlo> Perhaps there's a magic spell he only knows :)
14:04 < attriel> i remember reading somewhere a bit about "oh, test1 was a botched patch; the test2 patch works, i even tried it to make sure this time"
14:04 < attriel> but no idea where anymore
14:04 < rlo> lkml?
14:05 < attriel> could be when i was fishingaround archives yah
14:06 < attriel> i changed the void to "unsigned long" (being the only type cast i saw) and it seems to be compiling ... still waiting for final link though
14:06 < attriel> (and i'm sure it's not a "valid" fix :o)
14:06 < rlo> good luck
14:07 < rlo> if it's a working fix it's valid for me
14:07 < attriel> heh.
14:14 < attriel> compiles and runs
14:14 < attriel> no idea what kind of "unintended side effects" it might produce though :o
14:14 < attriel> 'course, none of this actually solved my problem where UML doesn't seem to recognize anything but ubd0 :(
14:21 < yacc> attriel, what is ubd0?
14:23 < attriel> user block device (I think that's what it stands for)
14:23 < attriel> "Virtual Block Device" in the config
14:24 < attriel> trying to install slackware
14:24 < yacc> I got the kernel compiled too  (with uml patch 2.6.0-test2). But it can't recognize the ubd0 partition.
14:24 < yacc> I am trying different configuration.
14:25 < attriel> you got 260-test2 to compile straight up?
14:26 < yacc> no, I commented out all the code that caused trouble.  (less than 10 lines totally)
14:26 < attriel> my command line is ./linux ubd0=/home/attriel/inst_fs ubd1=/uml/root_fs ubd2r=/dev/cdrom devfs=nomount rw root=/dev/ubd0
14:26 < attriel> really?  thought i glanced and saw more than that :o
14:27 < attriel> guess i misread something :)
14:27 < attriel> 'course, i can't get past ubd0 and you can't get ubd0 even, so we've got it compiled and still not helping :o
14:47 < rlo> well, compiled and linked here, but panics on boot..VFS: Cannot open root device "ubd0" or unknown-block(0,0)
14:47 < attriel> :/
15:13 -!- yacc [~chatzilla@129.34.20.17] has quit [Remote host closed the connection]
15:18 -!- miah [miah@sifl.digivill.net] has joined #uml
15:18 < miah> ok
15:18 < miah> with uml and swap, do i have to create a seperate swapfile, or can i use the hosts swap partition ?
15:19 < miah> i probably have to create a sepearte swap file.. but i didnt see anything about it, though i could have missed it
15:22 < miah> also, is there a way to start the uml kernel without the xterms popping up?
15:22 -!- MikeH [~mike@62.56.3.225] has joined #uml
15:23 < miah> hrm, ok i think if i make the uml use a serial port it wont do xterm?
15:33 -!- rlo [~roberto@200.138.100.242] has quit [Quit: Finalizando Cliente]
15:37 < miah> hrm
15:40 < miah> lame, i set 'ssl=port:9000' and it is still launching xterms
15:40 < miah> how can i make uml not be lame and launch the damn xterms, i need to be able to start the uml out of a initscript
15:40 < miah> and it seems like if it cant launch the xterm it just hangs
15:41 < miah> software should allways fail sanely instead of leaving itself in a unusable state
15:41 < miah> because now i cant even halt the damn thing
15:52 -!- shaver [~shaver@216.138.242.5] has joined #uml
15:59 < shaver> I don't suppose anyone knows the magic to get 2.4.20+uml6 booting on the severn beta?
16:00 < shaver> I crash in memcpy very very early on, and gdb doesn't provide any context
16:08 -!- mike_ [~mike@62.56.3.246] has joined #uml
16:08 -!- MikeH [~mike@62.56.3.225] has quit [Read error: Connection reset by peer]
16:12 -!- yacc [~chatzilla@129.34.20.17] has joined #uml
16:30 -!- Jax [~computer_@148.100.209.103] has quit [Quit: going home]
16:31 -!- sizban [~sizban@66-181-105-194.cust.gwis.com] has joined #uml
16:32 < sizban> how can i access a modem attached to ttyS0 on my host from within the uml ?
16:34 < sizban> i tried ssl=tty:/dev/ttyS0, but that doesn't seem to work too well :)
16:34 -!- G2 [~G@212.159.59.85] has joined #uml
16:47 < mdz_> miah: read the HOWTO
16:51 -!- yacc [~chatzilla@129.34.20.17] has quit [Quit: ChatZilla 0.8.11 [Mozilla rv:1.2.1/20030225]]
16:58 -!- G2 [~G@212.159.59.85] has quit [Quit: Client exiting]
17:04 -!- sizban [~sizban@66-181-105-194.cust.gwis.com] has quit []
17:08 -!- shaver [~shaver@216.138.242.5] has left #uml []
17:22 -!- mike_ [~mike@62.56.3.246] has quit [Ping timeout: 501 seconds]
17:33 -!- mike_ [~mike@62.56.3.145] has joined #uml
17:41 -!- mike_ [~mike@62.56.3.145] has quit [Ping timeout: 490 seconds]
18:00 -!- mike_ [~mike@62.56.3.212] has joined #uml
18:26 -!- arthur [~arthur@64.171.188.114] has joined #uml
18:27 < arthur> where do i submit a compile bug on the uml-2.6.0-test2 patch?
18:28 < BB> uml-dev mailing list as a guess?
18:28 < arthur> arch/um/kernel/um_arch.c:51: storage size of `__kernel_vsyscall' isn't known
18:28 < arthur> has anyone compiled 2.6.0-test2 with the uml patch
18:28 < arthur> ?
18:29 < ElectricElf> arthur: I'm doing so now.
18:30 < BB> fear the ElectricElf 
18:30 < ElectricElf> BB: *zap*? :)
18:30 < arthur> im trying to figure if im doing something wrong or if the code is broken
18:30 -!- mistral [mistral@212.159.71.212] has quit [Ping timeout: 488 seconds]
18:30 < BB> ElectricElf thats what i'm worried about ;)
18:30 < ElectricElf> BB: :)
18:30 < ElectricElf> arthur: Erk, nevermind.
18:30 < ElectricElf> arthur: I'm just compiling with SKAS, not ARCH=um
18:31 < attriel> no, far as i know everyone (except maybe jdike) has had that problem arthur
18:31 < ElectricElf> arthur: Are you using 2.6.0-test2-1?
18:31 < attriel> if you change the type to soemthign tht isn't void it might work, but no idea what kind of unfortunate side effects crop up later
18:32 < arthur> i applyed uml-patch-2.6.0-test2-1.bz2 against linux-2.6.0-test2
18:33 < attriel> yeah, i haven't heard of any successful compiles without either commenting out the bits using __kernel_vsyscall or changing it's type
18:33 < attriel> don't think the vsyscall stuff is used yet, though
18:33 < arthur> i assume that a million bug reports have already been made?
18:33 < attriel> three's been some discussion on the uml-devel list yah
18:34 < arthur> well i guess ill just hack it hard and see if it goes :)
18:35 < attriel> i changed the type and it compiles and seems to run; someone else earlier commented out the vsyscall stuff and got it to work
19:06 -!- Jax [~ben.stout@148.100.223.241] has joined #uml
19:18 < arthur> i changes the __kernel_vsyscall stuff to int and removed MTS and SCSI so far and it seems to be close to compiling
19:34 < arthur> do i need to specify the root= differentley in 2.6?    im getting "VFS: Cannot open root device "ubd0" or unknown-block(0,0)"
19:34 < arthur> (but is dose compile now :) )
19:42 < arthur> yay it boots... it is nessasary to include root=/dev/ubd0 as well ad ubd0=root_fs, is there a cleaner way of doing this?
20:01 -!- idcmp [idcmp@69.41.229.181] has joined #uml
20:01 < idcmp> is there a 2.4.21 uml patch?
20:05 -!- MikeH [~mike@62.56.3.181] has joined #uml
20:10 < arthur> 2.4.21-ac may have in already
20:11 -!- mike_ [~mike@62.56.3.212] has quit [Ping timeout: 490 seconds]
20:24 < attriel> i didn't need to do root=/dev/ubd0 :o
20:42 -!- ticallion [~ticallion@213.175.160.210] has joined #uml
20:43 < ticallion> how goes?!
21:21 < david> hello
21:22 < ticallion> hey david, how's the usOfA treating u?!
21:22 < ticallion> oh, btw, have the time to work on an iptables irc tutorials?!
21:23 < ticallion> s/rials/rial/
21:23 < david> good
21:23 < david> irc tutorial?
21:25 < ticallion> yeah, irc tut on say, freenode's iptables or something, somebody's willing to advertize it sorta, where's that link?!
21:25  * ticallion goes to look
21:25 < david> freenode's iptables?
21:25 < david> I don't get it
21:26 < ticallion> freenode's #iptables
21:26 < ticallion> sorry
21:26 < ticallion> http://www.linuks.mine.nu/irc/
21:26 < ticallion> I'm able and willing to do it on my own, but I never get around to doing it
21:27 < ticallion> I'm going to read ur articles in lxf pro when time's available
21:28 < miah> \mess
21:32 < david> ticallion: I don't have time to do an IRC thing on iptables
21:32 < ticallion> david: yeah well... 
21:49 < ticallion> honeynet support very sexy
21:49 < ticallion> s/net/pot/
21:49 -!- idcmp [idcmp@69.41.229.181] has quit [Quit: new kernel]
21:51 -!- ticallion [~ticallion@213.175.160.210] has quit [Quit: struggling with lack of space]
23:08 -!- coryb [~cory@dhcp025.ittc.ku.edu] has joined #uml
23:10 -!- coryb [~cory@dhcp025.ittc.ku.edu] has quit [Client Quit]
23:32 < Jax> evidently jdike was compiling 2.6.0-test2 from the BK tree, which might be everyone's problem
23:32 -!- Jax [~ben.stout@148.100.223.241] has quit [Quit: bed]
--- Log closed Thu Jul 31 00:00:00 2003