I have a 4 year old server I am using as an upgrade to an even older 9 year
old server.
I have finished migrating the older system to the newer system, using the
latest NW 6.5 SP6 overlay CDs.

The newer server has an Adaptec 3210 I2O SCSI controller, which Novell
apparantly decided to suddenly abandon about three years ago with NW6.0 SP3,
and Novell apparantly will no longer release new drivers for it even though
Adaptec continues to support it, Microsoft and everyone else supports it,
and the controller still mostly works just fine with Netware.

So anyway, the I2O controller does still work, but it does not properly
dismount volumes when the server is downed -- the server just hangs and the
volumes don't properly dismount. It is necessary to create a SHUTDOWN.NCF
that manually dismounts the volumes and unloads the I2O block storage driver
(BKSTROSM). See Novell TID 10083967

I've been using the SHUTDOWN.NCF solution since NW6.0-SP3 without any
problems, but now suddenly after doing a fresh install of NW6.5-SP6 I am
running into a new problem.

Now, when the server is downed, and the SHUTDOWN.NCF runs, I get a bizarre
series of messages that are not documented anywhere in the Novell

Unable to remove the swap file from volume 'SYS'.
Unload everything using virtual memory before dismounting this volume.
Continuing the volume dismount now will likely cause a server abend.
Dismount volume? n

Screenshot using digital camera:

When I have the "Console Monitor" running when downing the server, I see
Java is using a virtual memory space, so I added "UNLOAD JAVA" to the
SHUTDOWN.NCF, but still this message appears when I attempt to down the
server and unload BKSTROSM.

The only other memory space is the 'OS' space and I obviously cannot remove

Is there any hope here, or am I just dead in the water and going to have to
remigrate the data back off of this RAID controller to yet another server?

Is it now impossible to dismount all volumes on a server if it's necessary
to do NSS repairs?