Boot problem

[Craig White]

On Wed, 2006-08-02 at 08:09 -0700, Eric "Shubes" wrote:
> Let me begin by thanking Michael, Kenneth, Alan, Joseph, and Darrin for 
> their replies. We value everyone's input. Remember, what's right is 
> important, who's right is not. ;) Now on to the matter at hand.
> 
> First thing I'd do at this point (we almost did this yesterday) is 
> double check jumper settings, and try conventional 'master' and 'lone' 
> settings if CSEL was in use (as Alan and Joseph suggested).
> 
> If that doesn't solve the issue, it would be prudent to do what Darrin 
> and Michael have said: Get your data off the drive before trying much 
> (anything) else.
> 
> Given that the old drive is still accessible from linux, I would do this:
> .) get a new IDE HDD of equal or larger size (I hate to say this part, 
> at Fry's).
> .) "# dd if=/dev/hda of=/dev/hdb bs=4096" from knoppix to copy the 
> entire old drive to the new one (making darn sure of your drive letters 
> - check with qtparted first).
> .) disconnect the old drive and put it in a safe place.
> .) attempt again to make the new drive bootable (if it doesn't already).
> .) if previous step fails, build a new system on the new drive (the way 
> you really want it!), then install the old drive and copy (or dd) over 
> anything you want to keep from the old drive.
> 
----
I totally agree with this - the other thing that I probably would have
done would have been to run lilo -vv (very verbose as I recall - it's
been a long time since I used lilo) but verbose should have it report on
screen it's relative success and it is possible that by the time the
system is up and booted, that lilo or any boot block writer program is
as confused as everything else seems to be as to which is the first
drive (which strengthens the arguments to confirm ATA cables,
connections, jumpers & even disconnecting the second drive on the
system.

Craig