On Tue, 21 Mar 2000, J. Buck Caldwell wrote:

> So I tried getting my VS4000/60 to newfs it's drives after disklabeling
> them correctly (thanks Chuck), and I'm getting a blowup. It actually
> happened also when I disklabel -e'ed the drive, but when I rebooted, the
> disklabel was correctly changed. The newfs appears to go according to
> plan, but when finishing, it pauses, then writes the following errors:
> 
> sd0(asc0:0:0): asc0: timed out [ecb 0x867b70d8 (flags 0x3, dleft 4000,
> stat 0)], <state 4, nexus 0x867b70d8, phase(l 97, c0, p 0), resid 400,
> msg(q 0,o 0) DMA active>
> sd0(asc0:0:0): sync negotiation disabled
> sd0(asc0:0:0): asc0: timed out [ecb 0x867b70d8 (flags 0x43, dleft 400,
> stat 0)], <state 4, nexus 0x867b70d8, phase(l 97, c 0, p 0), resid 400,
> msg(q 20,o 0) DMA active> AGAIN
> 
> After which, there's pretty much nothing left to do but HALT and reboot,
> as it doesn't respond to anything else. Well, that is, I don't know
> anything about the debugger, so I'm sure that might work, and I'd be
> happy to follow directions if anyone needs more information. It's a
> VS4000/60 with 56Megs, recent (1.4U) kernel, and the drive is a DEC
> DSP3160S. Also happens with a DEC DSP3105S.

  This same thing happens on my VS4000/VLC systems, so it appears to
be a generic problem with the VS4000 and/or the SCSI driver.

  I've been trying to track this down.  I've found that the write is
completing, but the Vax never gets the interrupt.  The NCR53C94 status
register indicates that it has an interrupt pending.  I'm not familar
enough with the VAX intrerrupt structure, or how it's implemented on
the VS4000 systems to know just what's going on.

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