Problem on DAT ?

jason andrade jason at dstc.edu.au
Fri Jun 7 19:14:00 CDT 2002


On Fri, 7 Jun 2002, Pierre POMES wrote:

> First question : It seems I only get write errors on the tape when
> dumping data from a raid container .... Can it be RAID related ? SCSI
> related ? The dat is on the internal controller of the PE2550, and the
> RAID in on a Powervault attached to a PERC3 controller...

it would seem odd.

> Second question : this question is linux only... I have some problem
> with the non rewinding device. If I run :
> 
> mt -f /dev/st0 rewind

i personally don't mix device types during these operations and prefer to
stick to one device type.. i.e use nst0 for it all.

> dump -0uf /dev/nst0 /boot
> dump -0uf /dev/nst0 /tmp
> 
> And then :
> mt -f /dev/st0 rewind
> restore -i -f /dev/nst0   -> Ok, I can see the /boot top level
> directories.
> 
> The next 'restore -i -f /dev/nst0' gives me the following error :

you cannot assume that a restore drops you to the start of the next
volume.  you should use (example) 
mt -f /dev/nst0 rewind 
restore -if /dev/nst0
mt -f /dev/nst0 rewind
mt -f /dev/nst0 fsf 1
restore -if /dev/nst0

> I am used to run usfdump/usfrestore on solaris by the same way and it
> works fine. Am I wrong wih the above commands ? Perhaps I have to set
> different parameter (density, block length, etc..) in my dump and
> restore command ?

it should really be necessary, but please try and above and see what you get.


regards,

-jason




More information about the Linux-PowerEdge mailing list