Thought you might like to know about a perplexing problem I'm having
that may provide an opportunity to improve the debian boot process.
I have a Potato rescue disk that I have used on many installations. I
just got two new computers that give me a "Boot failed" message. I
verified that the disk is still good by using it on two other
computers (a 486 and a P3).
Successful boots on the computers in question:
FreeBSD 4.0 install disk
RedHat 5.something install disk
raw linux kernel dd'd to /dev/fd0
Debian 2.0 and 2.2
>From what I can tell from the OpenBSD assembly code, it is failing the
int 0x13 (function 0x02) call to read in the boot image. So mbr
passes off to biosboot and biosboot barfs trying to read the first
sector of the next stage. Sorry, but I haven't had a chance to look
at the syslinux code, so I don't know what's going on there.
I've tried numerous methods of copying the images to diskette and I've
used numerous diskettes. I've blindly ripped out hardware.
The computers are Asus P4B motherboards, running at 1.5 GHz with 1 GB
ram. The bios is Award Medallion 6.0. I've twiddled cmos till I'm
sick of it.
Is there a debug version of syslinux that would print the AH register
upon an unsuccessful int 0x13? That might be helpful.
BTW: In the mean time, is there a quick way to get a lilo-based
rescue disk. Lilo works, but I want to use Debian.