On Wednesday 09 March 2005 04:22 pm, Tim Wilson wrote:
> Now I've seen something like this screen before. We entered the root
> password and manually fscked the partitions. They all come back clean now.
> I'm puzzled by the segfault with rcS. There is no line 243 in that file. I
> assume the actual problem is one of the initscripts that get called by rcS.

fsck is a bit fragile sometimes, especially with hardware errors (!)

> I'm also seeing weirdness with the mount command. I am manually mounting
> some partitions, but they're not showing up when I run 'mount'. I know that
> they're mounted because I can traverse the directories.

That happens when / is read-only (mount just reads /etc/mtab).  It will 
remount read-only on errors, so that may be part of it.

> Can anyone shed any light on this? I don't want to sound too panicked, but
> this is a pretty critical system for us and I need to get it back up ASAP.

Boot with KNOPPIX or some other live or rescue CD and fsck everything.    Then 
run dpkg -C or some debian command to check packages.  Anticipate 'Plan B' on 
plundering the drive and having to re-install the server.

-- 
-dave

Dave Carlson <dave at math.umn.edu>

Systems Administrator
School of Mathematics, University of Minnesota
http://www.math.umn.edu

PGP/GPG Fingerprint:
C3D0 9962 1E98 B742 132D  0E1A CE11 7C4B 5309 97A7
(visit http://www.gnupg.org for more information)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: not available
Url : http://shadowknight.real-time.com/pipermail/tclug-list/attachments/20050309/64968d71/attachment.pgp