cgsecurity.org

Hello,

I used an old Synology with 9 disks in raid 5 (i know it's my primary error to have a raid 5 with so much disks) and i lost 2 disks in same time, Raid failed and Synology was unabled to rebuild it.
Raid volume was used as iSCSI target for VMFS Datastore, but disk are not physically damaged or noisy, Smart data are fine.

I put all disks in Windows with a SATA Controller :



I load TestDisk and i try to recover partition on disk0 and disk1.

disk0



disk1



If i continue with "Enter" i have only one partition (same on disk0 / disk1) :



I did the test on others disks and i have same result, only one partition is displayed at this step.

Do you think i can continue with "write" or it's better to move to Ubuntu to have "mdadmin" tools too ?

Aonther solution could be to replicate partition table from a working disk (with sgdisk for example)

Thank you in advance,

Yathus

I did some test with a virtuabox VM and sgdisk did the job.

My "POC":

- VM ubuntu 22.04 with a raid1 + swap + raid5 with 4 disks (in GPT mode)
- i copied some data in raid1 and raid5 mount
- disconnect one disk and removing all partitions with another VM.
- replug "broken" disk => raid failed (1 disk offline)
- restore configuration with sgdisk -R and reboot
- raid is back without resync.

and for information testdisk print the same result, it find 3 partitions but is only able to restore last one.

Thank you for your follow-up posting.