Subject: Re: Booting sd0 (disk geometry versus bios geometry)
To: Jonathan Stone <jonathan@DSG.Stanford.EDU>
From: Heiko W.Rupp <hwr@pilhuhn.de>
List: port-i386
Date: 07/13/1998 11:16:42
On Fri, Jul 10, 1998 at 11:29:51AM -0700, Jonathan Stone wrote:
> Which geometry, though? The MBR partition (slice), or the BSD
> disklabel(5) used in the beginning of the BSD slice?
Oh. hard to tell, as I am no DOS guy :)
This all is on an adaptec
First it says
The current values for your disk geometry are:
real geom: 1219 cyl, 4 heads, 42 sec
BIOS geom: 1219 cyl, 4 heads, 42 sec
This is the value as reported when the kernel boots
Then I reneter the geometry as told by sysinst
You will be prompted for the geometry. Please enter the values you
want. The number of cylinders should be <= 1024 and the number of sectors
cylinders: 1024
heads: 4
sectors: 42
Then it tells me
Your disk, sd0, reports a geometry that does not address all possible sectors
on your disk. Your disk has 205561 sectors.
I have the possiblity to use the real geometry (which won't work) or
to use a fake geometry where all five possible coices also don't boot
afterwards.
---
> But what kind of mistake are you making, and what should sysinst do to
> let you catch it? Is the problem with the C/H/S geometry in the MBR,
> or with the disklabel(5) geometry?
All the C/H/S stuff. From the point on, where I stopped above all
works well (even if there are some very minor "glitches" where sysinst asks
me if I want a 120MB partition even if I only have 100MB in total)
> Is the problem, maybe, that you're trying to specify a disklabel(5)
> geometry, the geometry is passing sysinst's checks, but writing the
> BSD disklabel is failing?
I am not sure here, as sometimes I see some 'sd0: no disklabel'
passing by, but this is too fast to really see, what is going on.
--
See <a href="http://www.netbsd.org">NetBSD</a> for a multiplatform OS
## Crosspoint v08.15 ##