There may be a subtle other detail here Sean ..

 SD> Hello, All!

 SD> I stuck in 192MB RAM in the BBS machine and lo and 
 SD> behold, all the weird problems have been banished from 
 SD> the machine, to include the random crashes when doing 
 SD> a message area listing.

In the case of the old 1542C Adaptec SCSI controllers, it's been discovered
here that more than one DOS-VDM still, to this date of even the latest
official update kernel and FP16 has problems with strange lockups.

Your instability just might be a part of an other than 1542C controller
manifestation of this same wiglet.

As far as I can tell, the environment isn't being passed correctly during
more than one opening of a DOS-VDM and in RARE instances of even OS/2
command line sessions as a result of errors in the Lazy Write Cache!

If you want to experiment a bit, from a command line after the WPS has
loaded you can try:

     cache -lazy:off

I fought with this instability on the BBS box with MAX and BINK and all run
from batch files for many months.  Someone in the Usegroups suggested that
the error is a result of cache write failures.  The suggested the above as
well as also tinkering with the cache lazy write timing values which can
also be adjusted.

I turned the thing off with the BBS box.  No more problems except for the
49 day lockup for the kernel.  That's been fixed in a later patch after the
WR1026 last official kernel release patch.  I've not tested it on this box
with the BBS stuff in it.

I've tried a number of sent in DOSKERNEL varients to 'fix' it, but no luck.
So for months now the thing has been simply left with the write through
cache left off.


Just a thougtht.


--> Sleep well; OS/2's still awake! ;)

Mike {at} 1:117/3001

--- Maximus/2 3.01