Subject: SCSI DMA timeout
To: None <port-sun3@NetBSD.ORG>
From: Neil McNeight <>
List: port-sun3
Date: 04/24/1997 14:23:29
The SCSI problem listed below seems to have migrated to the 3/60 as well.
Same tar command, same shoebox, 24M of RAM, 32M swap.

si: DMA timeout (while polling)
si0: aborting, but phase=DATA_OUT (reset)
si0: reset SCSI bus for TID=4 LUN=0
panic: ncr5380_scsi_cmd: polled request, abort failed
Stopped at      _Debugger+0x6:  unlk    a6

Any ideas?


On Wed, 9 Apr 1997, Neil McNeight wrote:

> I'm getting a similar error when backing up a large directory to tape:
> st0(si0:4:0): soft error (corrected), data = 00 00 00
> panic: done < 0; strategy broken
> Stopped at      _Debugger+0x6:  unlk    a6
> db>
> This was after using a tar -cvp /usr/local/src to back up the directory to
> tape. I don't think it's heat related since the error occured even after I
> had the system down for a few hours to clean dust out of the shoebox. I
> can also feel air moving out the bottom of the unit and it isn't
> obstructed as far as I can tell.
> All of this is happening on a 3/50 with 4M RAM and a 32M swap partition
> (I'm planning to move up to a 3/60 without repartitioning the drive).
> Any ideas? Need more information?
> -Neil

"In spite of everything, I still believe  | Neil McNeight
 that people are really good at heart."   |
                        -Anne Frank       | PGP key available