Restore highly degraded RAID arrays

I have a Linux RAID6 array (mdadm) consisting of 8 disks.

2 disks failed. I replaced them and now I have 2 new empty disks.
I want to rebuild raid

But
when I started raid reconstruction, the reconstruction failed because the third disk showed a read error and was ejected from the array.< /p>

Data can be lost on the few sectors that cause read errors.

(What is there) How to recover raid array?

You can use tools like ddrescue to image the third failed drive to one of the new drives This should be an exact copy minus the bad blocks. So that it will not read errors. Then replace the third failed drive with the newly imaged drive.

Assuming there are no bad blocks Critical to RAID or file system, then you can assemble the array in degraded mode and copy as much data as possible, or try to add more drives for reconstruction/resynchronization (the latter is the more dangerous IMO).

I have a Linux RAID6 array (mdadm) consisting of 8 disks.

2 disks failed. I replaced them and now I have 2 new empty disks.
I want to rebuild raid

But
when I start raid reconstruction, the reconstruction fails because the third disk shows a read error and is ejected from the array .

Data can be lost on the few sectors that cause read errors.

(What is there) How to recover raid array?

You can use a tool like ddrescue to image the third failed drive to one of the new drives. This should be an exact copy minus the bad blocks This way it will not have read errors. Then replace the third failed drive with a newly imaged drive.

Assuming that no bad blocks are critical to the RAID or file system, then you can Assemble the array in degraded mode and copy as much data as possible, or try to add more drives for rebuild/resynchronization (the latter is the more dangerous IMO).

Leave a Comment

Your email address will not be published.