Text 6237, 62 rader
Skriven 2006-06-07 13:35:08 av Maurice Kinal (1:140/13.1)
   Kommentar till text 6236 av Paul Rogers (1:105/360.0)
Ärende: You gotta smile
=======================
Hey Paul!

Jun 07 00:29 06, Paul Rogers wrote to Maurice Kinal:

 PR> Found it in "man lilo.conf" but I'm not sure yet about using
 PR> it in cloning.

That is all I use for that particular purpose.  I have yet to find anything
that works better for such a wide variety of media that I actually would use on
a day to day basis.  I've already decided that CD/DVD's are of little real use
to me.  I seldom like or use embedded type stuff.  Too boring for the mostpart.

 PR> I got a client's HP system today for a W98 reinstall, (Virus?)
 PR> and it couldn't find the CD it booted from.  First time I've
 PR> seen that--except for my drive that "went bad".

I've seen many CD's going south these days.  I don't trust them.  Seems to me
that even floppies are of more use other then size.  I think for installs and
backups flash media is far more reliable and much more useable then floppies,
CD's and DVD's.

 PR> Tonight I got the clone script to run through the LFS base.  I
 PR> quit there to see if I had a working system.  The BIOS said it
 PR> found a MBR, but nothing happened, not even a L.

Strange.  Sounds like the mbr might be toast.  I have one older drive that is
like that and yet it is still read/writable without errors.  It is a 40G Maxtor
that was overused and abused from repetitive Windows installs.  I think that
might be what toasted the mbr on there but can't say that with absolute
certainty but looking around the local scene I seem to be the only one who can
keep drives working over the long haul and NEVER install any version of Windows
on any of my drives.  Never have, never will.  What for?

 PR> Was /dev/hda1, the boot partition marked bootable?  Yes.

That is normal.  I've managed to succeed booting partitions that weren't marked
bootable.  I think that is a dos hangover.

 PR> "man rdev" suggests not being used in favor of lilo's parms.

Possibly.  I've managed to get away with it.

 PR> Nevertheless, the kernel had been built with a different root,
 PR> so I reset the kernel to the new device.  And checked it.
 PR> Not good enough.

Sounds like a toasted mbr.  I have a few drives kicking around like that but
have only toasted one myself (a compactflsh disk) from overtesting and abuse. 
It lasted five years of my abuse before packing it in which was pretty good I
thought.  CDRW's die in less then a year under simular abuse.

 PR> Any ideas?

Try writing lilo to the root partition instead of the mbr.  If it is marked
bootable and lilo resides on the bootable partition it might work.  However
that will complicate a cloning script but this isn't a normal situation so ...

Life is good,
Maurice

--- Msged/LNX 6.2.0
 * Origin: The Pointy Stick Society XVIII - Don't know what I want (1:140/13.1)