On Friday 01 Apr 2005 23:58, Jess Anderson wrote:
I'm having ongoing crashes and unrepeatable errors in
routine scripts. Memory has checked out find repeatedly,
so is not especially suspect.
My current prime suspect is the hard disk system, either
the onboard SATA controller (VIA), the disk drive itself
(Maxtor) or the OS (FC3).
I can show that data on the disk has changed since it was
written, that is, originally it was correct but now it
is wrong -- a date field that was "1998" is now "19)8".
Could I be looking at a bad block, perhaps? If I understand
the man page for e2fsck, I can do a nondestructive mapping
out of bad blocks by doing
e2fsck -c -c /dev/sda<n>
for <n> corresponding to my non-swap, non-/boot partitions
(booting from the rescue disk, of course).
Is that right? I'm open to other suggestions as to how to
track this down as definitively as possible, and if the drive
checks out what to try next.
(Because I've never had *any* hardware problems before, i.e.,
since 1986, I'm kind of a newbie about such stuff.)
Any and all help appreciated.
--
[] I realized that with hard work, the world was your oyster.
[] You could do anything you wanted to do. I learned that at
[] a young age.
[] -- Chris Evert Lloyd, 1954-
--
* Copyright 2005 Jess Anderson
*
www.jessanderson.org * anderson(a)wisc.edu
* Window Maker Themes:
www.jessanderson.org/wmthemes
Last time I kept getting unexplained errors and crashes, I eventually found a
screw dropped behind the motherboard!
Advice I've been given, is to check your power supply. I'm told that the
voltage provided by cheap power supplies can be a bit unstable, which of
course causes problems. So instead of spending £15 on a power supply I spent
£50, and bought a UPS as well.
Dave