Hello managers,
This was my plan:
I just added a new Turbochannel dual SCSI controller (model PMAZC-AA,
fw 2.1) to our DEC 3000 (Model 800, fw 6.7, Digital Unix 3.2d). My goal was
to even out the distribution of existing SCSI devices by moving some devices
onto the additional SCSI buses provided by this new controller.
This is my problem:
Although the attached SCSI devices are correctly recognized when testing
the new Trubochannel SCSI controller from the console prompt (>>> t tc3 cnfg),
I am unable to access the devices after booting. I have made new device special
files using (what I think) are the new "rz" numbers for the SCSI devices
attached to the new Turbochannel card. There are no errors when issuing the
MAKEDEV command. However, when I try to examine the existing disklabel of these
disks, I get the error:
disklabel: /dev/rrz34a: Bad address
Does anyone know what this error means? I'm pretty sure that I've got
the rz numbers correct (please see my note below if you suspect this). Is there
some upper limit on the rz numbers in Digital Unix? Any comments are
appreciated - will summarize.
Thanks- Dave Wolinski <wolinski_at_umich.edu>
University of Michigan
--------------------------------------------------------------------------------
Note: this is how I determined the new rz numbers for using MAKEDEV:
rz number = 8 * SCSI bus number + SCSI ID
Before installing the new Turbochannel SCSI controller, we had SCSI
buses 0-3 (two internal, and two from our first PMAZC). So the _new_
Turbochannel SCSI controller provides SCSI buses 4 (port A) and
5 (port B). So, the "rz32" above was gotten by:
8 * 4 (SCSI bus) + 2 (SCSI id)
And in case you think that I have port A & B swapped, I also tried
rz40 with the same error when I try to examine the disklabel.
Received on Thu Jul 11 1996 - 18:58:31 NZST