[plug] Segmentation fault

Peter Wright pete at akira.apana.org.au
Sun Aug 26 12:46:33 WST 2001


On Sun, Aug 26, 2001 at 12:10:01PM +0800, Clinton Butler wrote:
> This morning we had a power failure and my Debian box was on... it was
> then turned off by lack of power. The system now refuses to run 'lilo'
> (as I have just switched the kernel back to 2.2.19_pre17 seeing as it was
> nice) and comes up with "Segmentation fualt" and puts me back to the prompt.

Um. Let me see if I've got this straight:

1. Before power failure you were running a Debian Linux box with <unspecified>
kernel which seemed to be mostly working okay?

2. You had a power failure and thus your system went down without a proper
shutdown.

3. When you rebooted you went through the usual exciting process of
watching with great interest as your disk partition(s) got fscked to make
sure they're okay.

4. You finished booting, logged in and your system seemed okay, until...

5. ...you decided to go back to using the 2.2.19_pre17 kernel instead of
the <unspecified> kernel you were using previously.

6. You went through <unspecified> steps to switch back to 2.2.19_pre17 (which
I presume included modifying /etc/lilo.conf), then you tried to run
/sbin/lilo (which reads /etc/lilo.conf and updates the boot block) and you
got a segmentation fault.


Is that pretty much it? Have I/you missed anything?

> I have no idea what this is... and it seems bad to me.

Seems bad to me too. :)

>      _.._    _    Clinton Butler

The only thing I could suggest off the top of my head without any more
information (it _might_ be helpful if you could fill in the <unspecified>
details above, especially listing _exactly_ the steps you took to switch
back to using 2.2.19_pre17, but I can't guarantee anything :) is that you
make sure you're running the latest version of lilo.

apt-get update ; apt-get install lilo

That's pretty much a stab in the dark though.

*thinks for a moment*

Okay, another stab in the dark - try changing your /etc/lilo.conf back to
the way it was before and try rerunning /sbin/lilo. The reason I suggest
that is that it's perhaps technically possible that you've made some sort
of nasty error in /etc/lilo.conf that causes lilo to crash when reading it.
Which certainly shouldn't happen, but it's worth checking out.


Pete.
-- 
http://akira.apana.org.au/~pete/

-- 
UNIX was not designed to stop you from doing stupid things, because that
would also stop you from doing clever things.
		-- Doug Gwyn



More information about the plug mailing list