[Fwd: Re: [uml-devel] honeypot kernel panic]

[David C. <da...@da...>]

-------- Original Message --------
Subject: Re: [uml-devel] honeypot kernel panic
Date: Sun, 3 Nov 2002 15:53:22 +0100
From: Robin Redeker <el...@x-...>
To: David Coulson <da...@da...>, r...@en...
References: <200...@en...> 
<3DC...@da...> 
<200...@en...> 
<3DC...@da...>

 > >sure!
 > >Oh, i just discoverred another bug or maybe a symptom of the last one:
 > >------------
 > >Bringing up interface eth0:  ./ifup: line 243:   276 Segmentation 
fault
 > >ip link set ${REALDEVICE} up
 > >Failed to bring up eth0.
 > >------------
 > >All i know is, that ifup normally not segfaults ;)
 >
 > Can you run it under strace?
 >

Heh, yes of course.
But the output is about 302Mb. A bit too much to send over my
128kbit upstream ;)
I didn't work with strace much... so i don't really know whats important.
I got quite a lot of stuff like this:
--------
..
ptrace(PTRACE_POKEUSER, 3754, offsetof(struct user, u_debugreg), 0) = 0
ptrace(PTRACE_POKEUSER, 3754, offsetof(struct user, u_debugreg) + 4, 0) = 0
..
--------

 > >Sorry, i couldn't produce a backtrace :(
 > >I don't know if it helps, but i attached the uml-kernel config to this
 > >mail.
 >
 > I'd be tempted just to not bother with honeypot mode and wait until skas
 > is more stable - Honeypot is too slow to be useful under tt.
 >

Heh, OK.
I don't really needed it anyways.. i only wanted to play around with
uml a bit.

I am following the development of uml a while now, and i am happy that
it runs so well (in non-honeypot-mode ;).
And it's fun to play around with it.


cya

-- 
Robin Redeker el...@x-... http://www.x-paste.de/
irc: irc.x-paste.de:6667 #opers

-- 
David

-- 
David Coulson                                  http://davidcoulson.net/
d...@vi...                       http://journal.davidcoulson.net/