The WD My Cloud unit equipped with two 6TB WD Red Plus drives was configured in RAID 1, essentially a mirror. Both drives contained an identical copy of the data – approximately 1.5 TB of family photos and work documents. This setup protects against the failure of one drive, but as it turned out, it doesn't protect against user error.
What actually happened
RAID 1 works by keeping identical data on both drives. If one drive in the array fails, the array enters a so-called degraded mode. At this point, data is still accessible from the second, healthy drive. The correct procedure is to replace the faulty drive with a new one and let the array automatically rebuild.
In this case, one of the drives started showing problems. The owner decided to format the drive, believing it would "clean it up" and the NAS would then copy data from the healthy drive onto the fresh one. However, NAS systems like those from Western Digital use specific metadata headers within the file system that define a drive's membership in the array. Formatting destroyed this information. After a restart, the NAS refused to work with the array because it evaluated the configuration as inconsistent – from its perspective, the array simply ceased to exist.
During diagnostics, we confirmed that the formatted drive was indeed empty. The other drive, which had not been formatted, had a damaged file system, but the actual user data remained intact. Both drives had a considerable number of operating hours behind them – around 40,000 – which is not unusual for drives in continuous NAS operation.
The data recovery process
The first step was creating a complete binary copy of the drive that still contained the data. A binary copy is an exact bit-for-bit replica of the entire drive, which we then work with instead of the original. This protects the original drive from any further wear or damage during the recovery process.
Next came scanning the copy and verifying data consistency. WD devices have their own specifics in how they store data and manage RAID structures, so recovery requires knowledge of these internal mechanisms. The analysis confirmed that the user data was recoverable.
After successful data recovery, we reinstalled the NAS, configured a new RAID 1 array, and transferred the recovered data back to the device. In total, we managed to recover approximately 1.5 TB of data.
When DIY NAS repair reaches its limit
The line between a successful recovery and permanent data loss is thin with NAS devices. If a drive in an array is showing errors, any attempt at software repair such as formatting or CHKDSK only puts additional strain on the drive. Worse still, every subsequent write to the drive – including rebuild attempts – can overwrite sectors containing critical file system indexes.
If your NAS stops displaying data or reports an array problem, the safest course of action is to shut the device down and leave it alone. That's exactly what our client did in the end – after the first unsuccessful formatting attempt, he didn't continue experimenting with the second drive, and that's why we had something to recover data from.
Device: WD My Cloud (WDBWAZ0120JBK)
Drives: 2× WD Red Plus 6TB (WD60EFPX) in RAID 1
Data volume: ~1.5 TB
Problem: Client formatted one disk in RAID 1, NAS stopped working, file system on the second disk damaged
Solution: Binary copy of the healthy drive → data recovery → NAS reinstallation → RAID 1 reconfiguration → data transfer back
Result: 1.5 TB of data recovered, NAS returned to operation
This case ended successfully, mainly because it was a RAID 1 array and one drive remained readable. Two practical takeaways to wrap up.
First – NAS drives need regular attention. Quality NAS drives like WD Red can last significantly longer in practice than the manufacturer specifies, but that doesn't mean you can set them up and forget about them. Log into your NAS administration regularly and check drive health. Enable automatic error email notifications so the NAS alerts you when something starts going wrong. Set up regular drive checks – for example, a quick check once a week and a full check once a month. And keep your NAS firmware up to date. These simple steps will give you an early warning before a minor issue becomes a serious problem.
Second – RAID 1 is not a backup. Mirroring protects against a single drive failure, but it won't help against accidental deletion, ransomware encryption, or indeed user error. A real backup must exist on another independent medium.
Technical terms explained
RAID 1 (mirroring): A configuration where two drives contain identical data. Every write operation is performed on both drives simultaneously. The advantage is protection against a single drive failure – if one stops working, the other contains a complete copy of the data. The disadvantage is that only half of the total capacity of two drives is usable.
Binary copy: An exact copy of the entire drive at the bit level, including empty space, system areas, and bad sectors. Unlike regular file copying, it captures the complete state of the medium and allows safe work with data without risking further damage to the original.
Degraded mode: A state where one drive has dropped out of the array configuration, but thanks to redundancy, the array remains functional. In this mode, data is not protected against further failure, and the faulty drive should be replaced and the array rebuilt as soon as possible.
