Buying UNIX for a clone
Bill Heiser
bill at unixland.uucp
Thu May 2 08:22:39 AEST 1991
In article <1991May01.172734.10760 at nstar.rn.com> larry at nstar.rn.com (Larry Snyder) writes:
>bill at unixland.uucp (Bill Heiser) writes:
>
>>the bus and build up a daisy chain of drives. As it turns out, with
>>one combination of drives I just happen to have, I can't run SCSICNTL
>>or debug on one of the drives! To do so, I need to disconnect one of
>>the drives. This is just one example of subtle little incompatibilities.
>
>what drive gives you the problems with scsicntl?
I have the following configuration. The bus is terminated correctly
(at each end), and all drives/host adapter are configured for parity.
I experimented with various methods of providing scsi terminating power,
to no avail.
CDC Wren IV
SCSI ID 1 --|
|
Archive |
Viper 2150S |
SCSI ID 4 --| When running SCSICNTL in this configuration,
| it reports something like "drive did not respond
Adaptec | to drive size query" (or something along those
1542b HA | lines). The Toshiba drive info doesn't show up
SCSI ID 7 --| in the "select unit" menu. When running with
| either the Toshiba or Maxtor disconnected, and
| terminations/scsi ids changed as appropriate,
Toshiba | everything is fine.
MK-156FB |
SCSI ID 2 --| This configuration has been running with Esix for
| several weeks now. The only anomaly I've noticed
Maxtor | so far (aside from not being able to properly use
8760S | SCSICNTL or DEBUG) is that my UUCP *INBOUND* rates
SCSI ID 0 --| have fallen from 1250 CPS to 700 CPS !!! VERY ODD!
To answer your question, I'm not sure which drive causes the problem. It
seems to be some incompatibility problem between the Toshiba and Maxtor
drives. Someone suggested that these two drives don't implement SCSI
protocol very well, so they wouldn't be happy co-existing.
--
bill at unixland.uucp The Think_Tank BBS & Public Access Unix
...!uunet!think!unixland!bill
..!{uunet,bloom-beacon,esegue}!world!unixland!bill
508-655-3848 (2400) 508-651-8723 (9600-HST) 508-651-8733 (9600-PEP-V32)
More information about the Comp.unix.sysv386
mailing list