I'm seeing problems under heavy load with one of my disks on an
SS20. It is, unfortunately, my root disk, so it's difficult to
troubleshoot when this happens. The disk probes as:

sd0 at scsibus0 target 1 lun 0: <SEAGATE, ST150176LC, 0002> SCSI2 0/direct fixed
sd0: 47702 MB, 12024 cyl, 22 head, 369 sec, 512 bytes/sect x 97693755 sectors
sd0: sync (100.0ns offset 15), 8-bit (10.000MB/s) transfers, tagged queueing

When I do something strenuous, such as extracting a large tarball, the
system hangs, and the console fills up with "esp0: invalid state 6"
messages. The system is then completely hung until I reset from
PROM. After that, fsck runs normally, with no errors reported that
would suggest a hardware problem.

Posts in the archives suggest turning off tagged queueing, but I have
no idea whether that's still pertinent, nor do I know how to disable
it.

Is this likely to be a hardware problem, or software? If the latter,
can it be fixed? This is, btw, under NetBSD-1.6.1-RELEASE.

TIA,

Dan
--
Dan Debertin             | 
airboss@nodewarrior.org  | 
www.nodewarrior.org      |