> Joshua, I've got this faint echo running around in the back of my mind
> from something over a year ago, maybe even 2, where someone was
> having trouble with an HP that was responding to multiple LUNS.  At
> the time I can't recall if the consensus was an HP bug, or a kernel
> bug.  In this gents case I think I'd name /dev/nst0 as the drive, and
> /dev/sg3 as the changer so he is using the LUN=0 device for the
> drive, and the LUN=1 for the robotics.  Only if there are weird
> results would I change it to /dev/sg2 for the robotics since that
> theoreticly would be the same as /dev/st0 at bus address 0,4,0.  And
> that would make me nervous/apprehensive/worryalot etc.

i have set the tape device to be /dev/nst0 and the robot to be /dev/sg2

Not sure why it behaves like this but i did a quick reinstall and this
occurred on a 2.2.18 kernel and now still does it on a 2.4.18 kernel on
REdHat 7.3. Therefore i'd so for a HP bug

thanks for the help it seems to be working

Tom