Salvaging a Centos 6 RAID 1 drive

I had a CentOS 6 server that died. The hard drives ‘should’ be fine. I’m building a Rocky Linux 9 server to replace it. Can I add one of the RAID 1 hard drives to the new server to gain access to the data on the RAID drive? The CentOS implementation used software RAID. Can I use mount it? Or is there something else I must do?

Mdadm does store metadata about the array into each member drive of the array.
Older mdadm metadata format did place the “superblock” at the end of the partition.
With that the start of the partition has filesystem data, just like regular partition.
Latest metadata formats have metadata at start and the filesystem later.

RAID 1, mirror, has (copy of) entire filesystem in each member.
Therefore, drive from older RAID 1 array should mount just like it had no RAID at all.
A drive from current RAID 1 array you have to first assemble into array and then mount;
mdadm can assemble a degraded RAID 1 array that has only one of the drives and the other “missing”.

So yes, you can mount.

Thank you. That was wjat I’d hoped.

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.