HP Colorado 5gb and mt

[Joshua Lokken]

* James Long (list at museum.rain.com) wrote:
==> On Sun, Apr 13, 2003 at 01:43:35PM -0700, Joshua Lokken wrote:
==> > 
==> > The backup continues, and finishes, reporting success, and flexbackup
==> > properly stores the index of the backup, however, when I examine the
==> > tape with :
==> > 
==> > #mt fsf 1    (for example)
==> > 
==> > the tape activity is there, and I receive no errors.  The tape
==> > has no index and apparently backs up each dir to the beginning 
==> > chunk of tape, and overwrites it with the next dir!  That is not
==> > what I want.
==> 
==> I've never used flexbackup, so I can't help with the blocksize error,
==> but the symptom above suggest that you may want to test with the device
==> set to /dev/nast0.   I also use SCSI tape instead of ATAPI, but I presume
==> the /dev/nast0 device would be the same as /dev/ast0 except that /dev/ast0
==> will rewind the tape when the device is closed.  If, perchance, flexbackup
==> is closing the device when it finishes backing up one directory heirarchy,
==> the /dev/ast0 device will do a rewind upon receiving that close command,

It was, indeed.

==> and lo and behold, your second backup job will overwrite the first, and
==> the third will overwrite the second, etc.  This is conjecture, but it
==> seems to match your symptoms on this issue.
==> 
==> Since you are still playing at this point, try playing some more using
==> /dev/nast0

I did, and the backup, and subsequent verification went well!  Thanks
again, Jim.  I do still receive the blocksize error, and the device
does seem to cause a few problems, but I am able to backup properly
for now ;)

--
Joshua