Ruben van Staveren <ruben@luna.net> writes: 
> But anyway, something goes wrong when I use a big tape (525MB).
> 
> 
> 
> To get any activity from that tape, I issued a 'mt retension' and used
> the 1.3 alpha miniroot for the sake of /dev (should not make any difference
> with the /dev from 1.2x )
> 
> the tape hums fine, but after a long time, mt freezes somewhere in kernel mode
> and the following is printed:
> 
> st0(bzivsc0:4:0): bzivsc0: timed out [ecb 0x66f09e (flags 0x103, dleft 0,  stat 0)], <state 1, nexus 0x0, phase(c 100, p7), resid 0, msg(q 0, o 0)>
> bzivsc0: reselect from target 4 lun 0 with no nexus: sending ABORT
> bzivsc0: indentify failed

  It would appear that the timeout used for the retension operation is too
short.

  There also appears to be a problem in recovering from trying to abort an
active operation that timed out.  I saw the same behaviour with a program
I had to spin down SCSI disks.  I erroneously had too short of timeout, and
got hit with this problem on an old DEC RZ23 I was using for testing the
1.3 install/upgrade procedure.  I'm not sure if this problem is in the
machine-independent driver (ncr53c9x.c) or in the amiga interface to to
the MI driver.

Michael
---
Michael L. Hitch                        mhitch@montana.edu
Computer Consultant,  Information Technology Center
Montana State University, Bozeman, MT     USA