> [...] I let it go for 5 hours and stopped it.
Try
dd if=/dev/hda of=/dev/hdb bs=64k
it will go faster. Also, if you happen to have two IDE ports (if this is a 1.66GHz machine, chances are it does), the copy will go faster if the two drives are on different IDE ports.
> After stopping it, the original drive was messed up!
Bizzare. Any chance you dont quite have the device names right? Also, if you are doing this on the drive level, the drive geometry should be the same. In some ways, its safer to do this on a drive partition by partition basis.
> Why? Possibilities:
> -I was making a mirror image of a brand new Windows 2000 install with the NTFS file system.
Should not matter. Data is data, at least in this case.
> -I did not mount the file systems (I did not think I needed to).
!!!!Danger Danger Will Robinson!!!!! You SHOULD NOT mount the file systems, at least not read/write. This WILL mess things up.
> -They are DMA-100 IDE drives on a 1.6GHZ computer. Maybe LNX-BBC cannot handle this hardware.
Should not matter. DMA-100 IDE drives have been supported for some time, and the processor should not be an issue either. I cant speak to the LNX-BBC distribution, however. Using a Rescue CD from any distribution should be able to handle this, however. I know Debian (potato) will (used it for almost exactly this same purpose, well, mine was "dd if=/dev/zero of=/dev/hda", but its the same idea.) and strongly suspect that RedHat, Mandrake, SUSE, etc. will handle it quite well.
--
Robert A. Klahn
rklahn@acm.org
AIM: rklahn
Yahoo Messenger: klahn
IRC:
rklahn@irc.openprojects.net
"Hope has two beautiful daughters: Anger and Courage. Anger at the way
things are, and Courage to struggle to create things as they should be." -
St. Augustine
(text/plain)