The original problem description is shown at the end of
this message. The responses I received were all over the
map, ranging from suggestions to use a different brand of
cleaning cartridge to trying different SCSI cables and/or
cards to checking for proper termination. I guess the problem
may well be attributable to any one of these causes.
Since this was one of those cases where I just needed the
tape drive working as quickly as possible, I moved it to
a machine in another room, and it worked fine. Two people -
Daniel Luechtefeld and Steve Gauthier - mentioned the RF environment
as a possible cause, so perhaps that was it. The original location
had several computers and other electronic devices (though nothing
else was experiencing problems).
for taking the time to respond ...
Original Problem Description:
One of our Exabyte 8505 tape drives was causing error messages
about the soft error rate being too high during writing to be
displayed to the console of a Sun SS/10 running Solaris 2.5.1.
(The machine has all recommended patches applied as of mid-November
of last year).
The hardware technician brought a replacement drive (refurbished),
and the same problem occurred. I've tried using a brand new tape,
and running a cleaning tape through the drive, but the same
message still occurs (although the percentage of errors sometimes
varies by one or two percent).
The technician then took our drive (and SCSI cable) back to his
shop, tested it, and was unable to reproduce the error. The main
difference was that his Sun workstation was running Solaris 2.6,
which doesn't seem like it should be significant.
The technician is willing to order yet another drive (refurbished,
since they are the only ones apparently available), but I wondered
if something else could possible cause this error. I assume the
error message is not spurious (i.e., refers to a real problem),
but I can't be sure. The error message occurs whether I perform
a tar or ufsdump.
Note that the tape drive is the only external SCSI device on the
machine, and is properly terminated. Also, no SCSI error messages -
mentioning resets or timeouts, for example - are written to the console.
If anyone has seen this before or has any suggestions, I would
appreciate hearing from you ...
This archive was generated by hypermail 2.1.2 : Fri Sep 28 2001 - 23:14:04 CDT