> From: dlinford@myra.nebulus.net (D.Linford)
> Date: Tue, 14 May 1996 11:04:08 -0700 (PDT)

> > With the SCSI drive:
> > 
> > Same as above but root fs mounts OK
> > run install OK
> > when I am asked if I want the root fs on sd0a a say yes.
> > the system responds with "skipped" and then continues.
> > later when it tries to write to sd0a it says it can't due to
> > "INCORRECT SUPERBLOCK" and I am bounced out to the miniroot prompt.
> 
> 	We were having odd problems like this. It turns out that 
> the sequencing of partitions is important, in that the logical 
> partitions need to physically occur in sequence.

Really?  I was not aware of that.

> 	It is not documented. I guess it is one of those things 
> that most folks always do a certain way, and easily make assumptions 
> about. Our problmes were very much like yours, weird things with newfs 
> and mounting etc.. I was wanting to make it possible to leave part of 
> the drive available for another system, so we had need to be weird.

Yes, I don't think I've ever tried out-of-order partitions...

> 	Could someone tell me where to find out more about the 
> physical part of the partition table? I need to have the first two 
> sectors available for the other system to lay down its 
> partition table, if I'm to do this...

The Sun label is always in logical sector zero on the disk.
It's format is described by: arch/sun3/include/sun_disklabel.h

Gordon