Actually this has happened since version 1.1.  The machine is
acting as an NFS server for 5 2gig seagates concatinated together
via the ccd driver.  Once a week or so this will happen:

May 30 23:27:02 bb /netbsd: sd4(bha0:5:0): timed out
May 30 23:27:02 bb /netbsd: sd3(bha0:3:0): timed out
May 30 23:27:04 bb /netbsd: sd3(bha0:3:0): timed out AGAIN
May 30 23:27:04 bb /netbsd: sd4(bha0:5:0): timed out AGAIN
May 30 23:27:11 bb /netbsd: sd4(bha0:5:0): timed out
May 30 23:27:13 bb /netbsd: sd4(bha0:5:0): timed out AGAIN
May 30 23:27:27 bb /netbsd: sd3(bha0:3:0): timed out
May 30 23:27:27 bb /netbsd: sd3(bha0:3:0): timed out
May 30 23:27:27 bb /netbsd: sd0(bha0:0:0): timed out
May 30 23:27:27 bb /netbsd: sd3(bha0:3:0): timed out
May 30 23:27:27 bb /netbsd: sd4(bha0:5:0): timed out
May 30 23:27:27 bb /netbsd: sd0(bha0:0:0): timed out
May 30 23:27:27 bb /netbsd: sd2(bha0:2:0): timed out
May 30 23:27:27 bb /netbsd: sd0(bha0:0:0): timed out
May 30 23:27:29 bb /netbsd: sd0(bha0:0:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd2(bha0:2:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd0(bha0:0:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd4(bha0:5:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd3(bha0:3:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd0(bha0:0:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd3(bha0:3:0): timed out AGAIN
May 30 23:27:29 bb /netbsd: sd3(bha0:3:0): timed out AGAIN
May 30 23:30:08 bb /netbsd: sd1(bha0:1:0): timed out
May 30 23:30:10 bb /netbsd: sd1(bha0:1:0): timed out AGAIN


Processes that attempt to access that filesystem will hang, and
as a result the nfsd's stop responding.  One of the drives (not
the same one each time) will have its busy light stuck on when
this happens.  If instead of the buslogic controler we use a ncr
controller this doesn't happen.  The buslogic controler is:

	May 30 23:46:41 bb /netbsd: bha0: model BT-946C, firmware 4.28D

Anyone else see anything similar?

						--Greg