Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 22 Oct 1997 15:38:14 -0400
From:      Andrew Heybey <ath@bellcore.com>
To:        scsi@freebsd.org
Subject:   Re: NCR 53c810 & DAT tape [Solved?]
Message-ID:  <199710221938.PAA00882@grapenuts.bellcore.com>

next in thread | raw e-mail | index | archive | help
Well, not really solved, but maybe a work-around.

I have the Seagate/Conner/Archive DDS-1 DAT drive (Seagate model
4320RT, identifies as "ARCHIVE Python 25501-XXX 5.45") that wedges
after writing anywhere from a couple to several hundred MB.  I poked
around on the Seagate web site and found a page that suggested that
their tape drives be run only in asynchronous mode and no faster than
5 MB/s.

So first I tried using ncrcontrol to set the drive to be asynchronous
("ncrcontrol -t 6 -s async") but that didn't help.  Then, grasping at
straws, I tried changing back to synchronous mode but limiting it to 4
MB/s ("ncrcontrol -t 6 -s sync=4").  (When initially probed, the drive
claims to do synchronous at 6.7 MB/s.)  I then managed to write a
whole 60m tape full of zeros (1.3GB) and to do four 300MB dumps in a
row without wedging (I also read back two of the dumps), so this may
have done it.

Any thoughts on whether there is any logical explanation as to why
this fix would work when just setting async mode wouldn't?  The drive
is the only device on the bus and there are terminators on both the
controller and the drive.  I haven't tried a different cable yet.

Now if I can just figure out how to accomplish the equivalent speed
limitation under Windoze NT.

andrew



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199710221938.PAA00882>