This is a classic example of completely misunderstanding how things work.
Here Bela explains why an OS CD can boot, but fail to install.
From: Bela Lubkin <fi...@armory.com>
Subject: Re: HP Pavilion w/Nvidia chipset and an AMD quad-core fails to install with both 5.0.7 and 6.0
Date: Tue, 9 Mar 2010 15:19:18 -0800
Pat Welch wrote:
> SATA controller is used for both the single 750 GB drive and the CD/DVD
> writer. AMD Athelon II Quad core with an Nvidia chipset. HP model# is
> Both versions of SCO 'lose' the install CD right after it tries to start
> the actual install, after getting to the point where it asks you about
> the boot CD.
> No matter what combination of defbootstr's, adapters and drivers I feed
> it, it always comes back saying the CD, which it just loaded from(!),
> was "not ready".
What you just "loaded from(!)" was the BIOS reading an El Torito boot
sector, loading an x86 assembly language boot loader, and the boot
loader loading further data using the BIOS floppy emulation code. Once
the kernel is fully loaded, the boot loader jumps to it. At that point
you need a Unix kernel driver which talks to the same hardware via
direct hardware access. Neither OSR5 nor UW7 attempt to operate storage
devices via 32-bit BIOS calls from protected mode. The point at which
the install fails is _exactly_ where you would expect it to fail if the
OS being installed didn't have a driver for the hardware. Or if you
confused it into misconfiguring the right driver.
> In 5.0.7 I change the SATA controller to IDE mode in the BIOS before
> trying to load.
> 5.0.7 'sees' the root disk OK using the 3.0 wd BTLD loaded from the USB
> I've tried everything I can think of - under 5.0.7 using the
> wd.debug="i" defbootstr option confirms the wd setup for the CD is WD0,
> id 5 as it says when I get to the screen.
That would be your problem right there. It is not in any way "ID 5".
BTW, -100 points for (1) not posting the actual boot strings you tried
and (2) not posting the actual corresponding outputs. We know that you
tried "everything you can think of", so we know absolutely for certain
that you have tried a bunch of wrong bootstrings as well as (possibly) a
right one. The debug output about "id 5" is clearly from one of the
wrong ones, but we can't tell much beyond that.
As far as I can guess from the information given, the required boot
string for OSR507 is "defbootstr link=wd". Nada mas. Did you try that?
Any strings like "Srom=wdatapi(...)" are only likely to mess it up. If
you should happen to put in the right numerical parameters, it would be
OK -- but you would only be forcing it to the values it was going to
figure out by itself anyway.
Got something to add? Send me email.