Hmmm... I wonder...

I'm no where near being a SCSI expert, and I don't play on on TV.  I'm
wondering if somehow the block-size issue is at the root of this.  We have
some RH servers too, but they're running at best 7.2, most are 6.3!!!  --
don't yell back!

Have you loaded a labeled tape that you know fails in drive 1, into drive
0, then I believe the command is

mt -f /dev/nst0 status

I think that will show you block sizes that tapes' labeled for.

Then, load a tape that works in drive 1 into drive 0, and do the status
command again.

I got a hunch the block sizes are different on the two tapes.

Then, it's a matter of getting the config properly set for whichever way
you wish to go.  I don't understand block sizes, or stinit.def files.  All
I do know is when we first set up Legato 6.1.3 on a RH 7.2 server,
something didn't config a stinit.def file, so our block sizes weren't what
we wanted.

When we moved up to the SuSE 8.2 and Leg 7.1, the block size thing was
automagically config'd, and now, unless I manipulate the block-size when I
need to recover from older tapes, they're totally useless.

Help?  Or continue to drive you nutz?

Keep in touch, as I think what I have is contagious and you caught it, or
vice versa.

Have a good night,

Les

Les,

       I liked your cookbook idea, so here's what I did to
find out what's wrong with my library:

       1. Loaded a blank tape into /dev/nst0.
       2. Wrote an EOF mark
       3. Labeled tape on /dev/nst0
       4. Mounted tape on /dev/nst0 -- SUCCESS
       5. Mounted tape on /dev/nst1 -- FAILED

       1. Loaded a blank tape into /dev/nst1
       2. Wrote an EOF mark
       3. Labeled tape on /dev/nst1
       4. Mounted tape on /dev/nst1 -- SUCCESS
       5. Mounted tape on /dev/nst0 -- SUCCESS

So, tapes labeled on /dev/nst0 can't be mounted on /dev/nst1,
but tapes labeled on /dev/nst1 can be mounted on /dev/nst1 and
       /dev/nst0.

Looks like something is wrong with /dev/nst0, wouldn't you say?

BTW, I'm running 7.1 on Linux (don't yell) Redhat 9.   Been running
for quite a while with no issues, by the way.



                                       Matt Temple


--
=============================================================
Matthew Temple                Tel:    617/632-2597
Director, Research Computing  Fax:    617/582-7820
Dana-Farber Cancer Institute  mht AT research.dfci.harvard DOT edu
44 Binney Street, LG300/300   http://research.dfci.harvard.edu
Boston, MA 02115              Choice is the Choice!

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=

--
Note: To sign off this list, send a "signoff networker" command via email
to listserv AT listmail.temple DOT edu or visit the list's Web site at
http://listmail.temple.edu/archives/networker.html where you can
also view and post messages to the list.
=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=*=