Hello everyone,

I previously posted about problems with dump to tape. The system is an
Intel PD440FX, 128 MB memory, 8 GB main disk IDE, Eliant (Exabyte) tape
drive SCSI narrow, 100 bps Ethernet, plain video. It runs NetBSD 1.5 and
serves DNS, DHCP, SMTP. Last month, the backups started failing with pre-
mature 'end of tape detected'. I tried using the B option with a value of
7000000 (unit is 1 KB blocks) which approximates the capacity of the
Eliant (made by Exabyte, similar to 8505) when using 160 m cartridges.

The change did not solve the problem but it is easier to work with total
capacity than length and density.

Early on, I added a Python 4 mm DAT drive to the SCSI chain and started
doing backups to that drive.

Now, the Python is failing with the same error: end of tape detected.

Before this latest development, I thought that the Eliant was failing
(i.e. I have a hardware problem). It seems unlikely that the Python has
caught the same disease as the Eliant.

Here is part of the backup script I am using with the Python.

	b=32
	B=2000000
	f=/dev/nrst0
	dump 0ubBf $b $B $f /
	dump 0ubBf $b $B $f /usr
	dump 0ubBf $b $B $f /var
	dump 0ubBf $b $B $f /home
	dump 0ubBf $b $B $f /var/mail
	mt -f /dev/rst0 offline

Am I dealing with a software problem? or am I dealing with an environ-
mental problem (i.e. contamination) ?

Thank you for your assistance.

-- 
Claude Marinier, Information Technology Group    claude.marinier@dreo.dnd.ca
Defence Research Establishment Ottawa (DREO)    (613) 998-4901  FAX 998-2675
3701 Carling Avenue, Ottawa, Ontario  K1A 0Z4         http://www.dreo.dnd.ca