SCSI disk to disk dump restore

[Dan Rue]

On Tue, Aug 24, 2004 at 11:12:10AM +0300, Odhiambo Washington wrote:
> * Dan Rue <drue at therub.org> [20040824 00:01]: wrote:
> > Hey Gang, 
> > 
> > I had an older scsi disk going bad, so I picked up a new disk to replace
> > it.  I did a dump | restore to move the data to the new disk, but it
> > went far slower than expected.  I'm wondering if there's an issue with
> > the different disk speeds.  
> > 
> > Old disk from dmesg:
> > da0: <QUANTUM ATLAS IV 18 WLS 0808> Fixed Direct Access SCSI-3 device
> > da0: 80.000MB/s transfers (40.000MHz, offset 31, 16bit), Tagged Queueing Enabled
> > da0: 17522MB (35885168 512 byte sectors: 255H 63S/T 2233C)
> > 
> > New disk, as reported by dmesg (it's a seagate U320):
> > da2: <SEAGATE ST336607LW 0007> Fixed Direct Access SCSI-3 device
> > da2: 80.000MB/s transfers (40.000MHz, offset 63, 16bit), Tagged Queueing Enabled
> > da2: 35003MB (71687372 512 byte sectors: 255H 63S/T 4462C)
> > 
> > The dump was 13G, and took about 2 hours.  I calculated that out to be
> > about 1.8MB/s.  Of course, that's total time I'm sure there's some dump
> > overhead (though, i didn't use -L) - but still, 2 hours for 13G?!
> 
> Well, depends! I used a dump|restore (without -L) recently and it took
> me like 5 hours to complete the dump for 28GB.
> 
> 
> > I'm thinking there was a jumper or parameter I should have set due to
> > the different disk types.  Note that I removed da0, but there's still
> > another quantum in the machine on the scsi chain.
> 
> I also had another disk on the SCSI chain.
> 
> How much time did you expect to achieve as optimum? Based on what
> factors?

Ok, well that's good enough for me.  I don't have a lot of experience
with SCSI or dump - so I thought maybe I was hitting some caveat,
especially with the mixed drive types.

Thanks for the info it sounds like my numbers were about right after
all.

drue