jdm5548@diamond.tamu.edu (Darrell McCauley) writes:
[scsi resets with IPC]
>5. >From the Sun Field Engineer Handbook-- "SCSI bus errors may occur when
> multiple SCSI devices are used with CPU boards below 501-1689-06."
> Check the CPU board to see if this is true--if you do have a downrev
> board and you have a maintenance contract, a call to your local support
> engineer should be enough to get it replaced.
>
>I have not had a chance to check these yet (busy trying to graduate)
>and I'm waiting for the error to happen again (system load has been way
>down during the holidays).
According to Sun on IPC boards below the above mentioned
CPU-Revision (labelled on board directly above the S-Bus connectors)
the on-board routing of the SCSI Bus is somewhat long, causing
the bus not to be correctly terminated on the *Sun* side.
Hence -as the internal disk is not terminated - the bus isn't terminated at all!
Sun recommends terminating the internal disk if no other device is added.
(if only one disk is present, a bus open on one side is tolerated)
As this was not possible for me (The concerned machine was a tape server,
resulting in the Exabyte constantly resetting during backup),
Sun immediately exchanged the board.
/tatjana
-- Pierrot le fou | pierrot@cs.tu-berlin.de Tatjana Heuser | W-1000 Berlin 30 | Ettaler Str.2 | BITNET: pierrot%tubopal@DB0TUI11.BITNET (saves $$$)
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:06:33 CDT