Hi all

Normally when I have a drive which contains raid metadata e.g. from fakeraid or a hardware raid card I can get rid of it with:

dmraid -r -E /dev/sdX

Or if that doesnt work, doing a simple dd if=/dev/zero of=/dev/sdX normally works

But I have 4 WD RE4 drives here and neither option seems to work. dmraid just gives me a CRC error when it reads the drive. I know the drives are fine and theyve been checked with smart + WD diagnostics.

When I try to install CentoS 5 or CentOS 6 it complains about the metadata on the drives.

The actual error on dmraid is:
Gives the same CRC error for all disks so clearly something strange the raid controller did

Any suggestions?

Thanks

Ensure that there are no active raid devices then use the zero-superblock option for each of the devices.
example
mdadm --stop /dev/md0
mdadm --zero-superblock /dev/sda1