SanDisk Dual SSD, RAID 1: Hardware Failure Causing Inaccessibility
SanDisk SDSSDA-120G RAID 1 Data Recovery
Data Recovery from Inaccessible RAID 1 Array Due to Dual Drive Hardware Failure
Client Scenario
This case involves a dual-bay storage unit configured with two SanDisk SSD Plus (SDSSDA-120G) solid-state drives in a RAID 1 (mirrored) array. RAID 1 is designed to tolerate the failure of a single drive without data loss, since the surviving mirror should hold a complete copy of the data. In this case, however, the client reported a sudden, complete loss of access to the array.
Client Profile & Challenge
- Media Type: 2 x SanDisk SSD Plus (SDSSDA-120G) SSDs, 120GB each
- The Crisis: Upon inspection, both SSDs are experiencing hardware failures, not just one. This is the exact scenario RAID 1 cannot protect against, in which 2 SSDs failed. Since no healthy mirror is available, we cannot recover through a normal logical extraction.
Technical Assessment Result
- SSD 1: Controller failure, preventing normal host communication
- SSD 2: Controller failure, preventing normal host communication
The Technical Challenge: At initial assessment, both drives were facing controller failure, with no normal host communication on either unit. A failed or malfunctioning controller sits between the host interface and the NAND flash. At this stage, we cannot identify the condition of both drives' memory chips. We had to work directly on the SSD to find out what was underneath the controller fault. We started the work only with the client’s acknowledgment.
The Recovery Methodology
Phase 1: Controller-Level Recovery
Both SSDs’ boards were removed from the enclosure and connected to dedicated recovery tools, bypassing the array controller entirely to address each drive's own controller fault.
- Drive Isolation: Separating SSD 1 and SSD 2 allowed each drive's controller issue to be worked on without one affecting the other.
- Controller Fault Resolution: The controller-level fault was successfully resolved on both drives, restoring direct read access to the NAND (memory chip where data is stored).
Phase 2: NAND-Level Assessment & Imaging
With controller communication restored on both drives, each was then assessed and imaged at the NAND level.
- SSD 1: The underlying NAND was found to be intact enough for sector-level imaging. A low-level, sector-by-sector cloning process with adjusted retry and skip parameters was run across multiple passes, recovering 80% of the drive's sectors. The remaining 20% was unreadable due to NAND cell degradation.
- SSD 2: The underlying NAND flash was found to be dead. No data could be read from the NAND.
Phase 3: Volume Parsing & File Integrity Verification
- File System Mapping: With the 80% sector image from SSD 1 in hand, the file system structure was reconstructed and parsed to map directories and file allocation metadata.
- File Extraction: Files and folders were rebuilt from the recovered sectors and checked for internal consistency before client review.
- Client Verification: Since SSD 2's NAND failure left no data to extract, the client reviewed the recovered file set from SSD 1 directly to confirm whether the data they needed are within the recoverable 80%.
Results & Data Extraction
SSD 1's controller fault was resolved, the NAND was found intact, even though it was degraded. 80% of the drive's data sectors were recovered. Upon review, the client confirmed that the specific data required for their operations was within that recovered set, and the recovered data was formally reviewed and accepted by the client.
SSD 2's controller fault was successfully resolved, but this only revealed that the NAND was dead. No data could be extracted from SSD 2.
Although full recovery was not possible due to SSD 2's NAND failure, the client was able to recover the majority of their data. This meant they did not have to reconstruct their records or workflows from scratch, significantly reducing the operational impact of the failure.
Key Issue & Conclusions
RAID 1 protects against the failure of a single drive. It does not protect against simultaneous failure across both mirror members. This case also highlights that a controller fault can mask the NAND's condition. Only after fixing the controller on both drives did it become clear that SSD 1's NAND was still alive while SSD 2's had failed. That is why it is impossible to see NAND condition at the initial assessment stage.
If your RAID 1 array has become inaccessible and you suspect more than one drive may be failing, avoid repeated power cycles or rebuild attempts. These repeated power cycles or rebuild attempts might reduce the amount of data that can still be recovered. Why? Because the original controller will further damage the compromised drives. Feel free to contact us if you need further advice.
EHDR, We Leave No Data Behind.
Professional Data Recovery Services
Secure • Confidential
SUCCESS STORIES
Over 14 years of enterprise data recovery experience