Missing swap space?

[George Robbins]

In article <2304 at dali> mathisen at dali.cs.montana.edu (Jaye Mathisen) writes:
> In article <446 at gca.UUCP> beaulieu at gca.UUCP (Larry Beaulieu) writes:
> |In article <13722 at cbmvax.commodore.com>, grr at cbmvax.commodore.com (George Robbins) writes:
> |> Subject to the concern that using the a or c partition for a swap area will
> |> probably overwrite the volume label, making access to other partitions on
> |> the drive "interesting".
> | 
> |Use of partition c means using the entire drive as a swap space, so
> |'access to other partitions' is irrelevant.  I do this routinely 
> |on dataless Sun 3/80's (root file system, page/swap reside locally) 
> 
> Well, it may work that way on Sun's, but it looks like under Ultrix it don't.
> I added an RZ56 as a swap disk, and tried to swap on partition c, and it
> hung in the swapon command...  The same thing happened on a mVII, under Ultrix
> 3.1...  When I tried to swap on partition a, I was then unable to access
> the rest of the partitions. 
> 
> I think something changes between Ultrix 2.0, and 3.0 as to where information
> about the disks is stored.

Well, the thing that might make a difference is whether you're running the
default partition tables  or not.  Normally, the system will use the partition
tables written into the "disk label" (actually mushed in with the bootstrap/
first super-block with Ultrix I seem to recall).  If the partition table isn't
there, or has been corrupted, then the system will revert to the partitions
hard coded in the driver.  If they're the same, everything should still work,
delta some one-shot confusion.  If not, then zap...

-- 
George Robbins - now working for,     uucp:   {uunet|pyramid|rutgers}!cbmvax!grr
but no way officially representing:   domain: grr at cbmvax.commodore.com
Commodore, Engineering Department     phone:  215-431-9349 (only by moonlite)