Subject: Re: disklabel trouble on a pmax(pmin?)
To: James O'Kane <jo2y+@cs.cmu.edu>
From: Simon Burge <simonb@telstra.com.au>
List: port-pmax
Date: 02/02/1998 00:30:53
On Sun, 01 Feb 1998 07:58:54 -0500  "James O'Kane" wrote:

> Hi,
> 	I was recently given a DECstation 2100 to play with. It had no hard drive
> in it so I added a SCSI Seagate ST3283N. I've been successful in getting a
> diskless boot to work over my network. My problem occurs when I try to use
> disklabel. First I tried 
> 
> 	disklabel /dev/rz0c -b rzboot -s bootrz
> 
> that is listed in the install manual, but that didn't work. What I finally
> got to work partially was:
> 	
> 	disklabel -w -B -b /usr/mdec/rzboot -s /usr/mdec/bootrz /dev/rz0 rz23
> 
> The rz23 is a listing in /etc/disktab. I tried my hand at writing my own
> entry to that 
> 
> 	sg|Seagate:\
>         :dt=SCSI:ty=winchester:ns#57:nt#5:nc#1689:\
>         :pa#65536:oa#0:ba#8192:fa#1024:\
>         :pb#98304:ob#65536:bb#8192:fb#1024:tb=swap:\
>         :pc#485601:oc#0:bc#8192:fc#1024:\
>         :pd#13674:od#163840:bd#8192:fd#1024:\
>         :pe#13674:oe#177514:be#8192:fe#1024:\
>         :pf#13676:of#191188:bf#8192:ff#1024:\
>         :pg#41024:og#163840:bg#8192:fg#1024:
> 
> I get the error: partition c: partition extends past the end of unit. Here
> are the specs of the drive from the seagate website.

I would guess that this is because the size of your 'c' partition
(485601) is greater than (57 * 5 * 1689 == 481365).  I tend to use
/usr/games/factor to fake a geometry for most disks - ZBR means that 57
sectors a track is just an average anyway.  factor says that (3 * 157 *
1031) is 485601, so I would try a geometry of 3 heads, 157 sectors and
1031 tracks.

BTW, you won't have these hassles if you are installing NetBSD 1.3 -
sysinst does all the work for you (although you may lose a few sectors;
in your case (485601 (real total) - 481365 (sec * head * tracks)) = 4236
due to the way sysinst calculates the disk size).

Simon.