I removed a disk from my NAS in Raid 5 mode for replacement with another one, and that all works OK.

Now I have a old disk WD (Red series) with 4TB capacity that I want to reuse as single disk storage.

I insert it into my USB SATA ext BOX, I connected it to my Mac, but OOPS -- now I see 2 disks of 2 TB each and not only one; the disk utility sees as 2 disks, not as 1 disk with 2 partitions.

I tried on a Linux box and on a Windows PC and it's the same strange situation on each.

How can I reformat the whole disk to 1 big 4TB partition?

This is my parted Linux box output: http://www.listenagency.com/linux.png (2 disks same model but in reality there is only 1 disk!!!)

  • Can you attach the disk directly to a SATA port rather than using the SATA-to-USB adapter? Oct 11 '17 at 6:59

It would have been easier to have cleared the RAID info using the BIOS before you removed it. Now you will have to manually remove the partition information. You can accomplish this using something like gparted or fdisk in linux or DISKPART in Windows. Delete the old partition information and then create a new partition table.

  • The disk appears as two top-level, 2TB block devices (sdg and sdh), not partitions, so it's not clear which device would have to be partitioned.
    – dhag
    Oct 6 '17 at 19:29
  • You need to clean out everything including the MBR. Maybe you could run DBAN on it in a pinch.
    – AJM
    Oct 6 '17 at 19:49

Do other 4TB disks appear as a single volume in that SATA box? I wonder if the storage device on that is responsible for this presentation, or do the two partitions match the NAS device's layout?

Take a look in /sys/block and see if the devices share the same targets as partitions should locate by numeric, and targets by alphabet.

Try something like this to wipe both volumes at the same time. It will not be faster, but the writes should buffer together to reduce seeks and should save you time (not the computer).

 dd if=/dev/zero of=- | sudo tee /dev/sdg /dev/sdh >/dev/null

Make sure that sdg and sdh are what you think they are first, check dmesg and /sys/block again to be sure. If it comes to it, plug and unplug to avoid data loss.

  • Don't use dd; use cat instead. But cat /dev/zero >/dev/sdg & cat /dev/zero > /dev/sdh & wait would be far more efficient and probably faster.
    – roaima
    Oct 10 '17 at 23:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.