> It appears all that was done was to make the start routines not use
> SCSI_NOSLEEP when called from the strategy routine.  It is still set
> when called from an interrupt (i.e. on command completion), so if you
> have enough I/O pending to overflow the SCBs already allocated, you
> will still lose.  This is made more noticable by ccd because it can
> have the effect of queueing many requests at once to the same
> controller.

Well, I guess this is true, but in my case, two disks of the four
caused the problem while two of them worked OK.  Even the same kind
of disks fail / work, depending on the firmware version.  Maybe there's
something else that's not working right in the ahe driver?

What are the SCBs, btw?  Some limited resource of the SCSI interface
or something in the kernel or..?

  -jm