Subject: Re: DLT drives
To: Ken Hornstein <firstname.lastname@example.org>
From: Manuel Bouyer <email@example.com>
Date: 02/25/1999 10:17:22
On Feb 24, Ken Hornstein wrote
> Anyone use a DLT tape drive with NetBSD? If so, did/does it work well?
Just setup a backup server on a NetBSD 1.3.3 machine:
ncr1 at pci0 dev 11 function 0: ncr 53c810a fast10 scsi
ncr1: interrupting at irq 10
ncr1: minsync=25, maxsync=206, maxoffs=8, 16 dwords burst, normal dma fifo
ncr1: single-ended, open drain IRQ driver
ncr1: restart (scsi reset).
scsibus1 at ncr1: 8 targets
st0 at scsibus1 targ 4 lun 0: <QUANTUM, DLT7000, 204B> SCSI2 1/sequential remova
st0: st0(ncr1:4:0): 10.0 MB/s (100 ns, offset 8)
density code 0x1b, variable blocks, write-enabled
ch0 at scsibus1 targ 6 lun 0: <OVERLAND, LXB, 0331> SCSI2 8/changer removable
ch0(ncr1:6:0): 10.0 MB/s (100 ns, offset 8)
ch0: 14 slots, 1 drive, 1 picker, 0 portals
(ch0 has 15 slots, but I reserved one for the cleaning cartrige).
Software is the amanda 2.4.1p1 package I commited 2 days ago.
I didn't notice any problems yet, lthouh I didn't use it extensively.
I'm still moving clients from the Sun/Hexabyte to the NetBSD/DLT,
so it's at 20% of the tape capacity for now.
> While I'm on the subject ... anyone got any reasons to love/hate DLT
> tape drives versus any other type of tape media?
I've got lots of problem with hexabytes (can't read back tapes written
too long ago, or with another drive, but this seem very dependant on the
quality of the media: original hexatapes seems much better, but also much more
expensive :). I also used DATs, I've been hit by incompatibilites between
drives too. In both case, medias don't survive a few dozen of write cycles.
A fried of mine bougth 2 DAT changers, he never has been able to do real work
I had to set up a DLT library 3 years ago, and managed it for a year.
I never got any problems with it, and found it really fast. I've also
got good reports from other peoples I know. The tapes are expensive but
seems to survive a lot of write cycle, the capacity is also higther ...
Manuel Bouyer, LIP6, Universite Paris VI. Manuel.Bouyer@lip6.fr