Case file · NAS & RAID · BDR-2025-0928
My Cloud Drive Went Quiet, Then Strange.
The My Cloud has sat quietly on the router, hooked up to my router and I haven't touched it
, but recently it started making a strange sound
and disappeared from both the network and USB. The warranty is fine about covering the hardware, but says the data is not included.
Same symptoms on your desk?
01273 964902
The decode.
A network drive is simply an ordinary hard drive placed into a special box (the "clever" part) and those special boxes can fail in one very distinct way: if the actual hard drive in them ceases functioning, all access paths will vanish instantly. This simultaneous disappearance is how you know it isn't a network problem, because if there was a network problem your USB connection should still work. If there was a bridge failure, your network connection should still function. But losing both access routes simultaneously (and with a new noise) tells you exactly where the problem lies: with the mechanism inside the box.
Equipment on this case.
How a job runs here →| The system | What it did on the day | Why this tool |
|---|---|---|
| DeepSpar Disk Imager 4 | Head-mapped imaging once the mechanism was made stable | Head-map-aware imaging with per-head timeout, reset and power-cycle control |
| PC-3000 Data Extractor | Managed read passes with reset control on a drive that had been struggling | The imaging half of the PC-3000 platform, driven by the firmware half |
| UFS Explorer RAID Recovery | Parsed the My Cloud's Linux volume layout from the image | Reads NAS volume managers as they actually are, not as a flat array |
At the bench.
Get the drive out of the box and diagnose it as a drive
We opened the enclosure and examined the drive individually. It turns out that the strange sound was due to a mechanical failure of the head assembly (which the NAS firmware interpreted as a non-responding drive, resulting in the loss of visibility via both access routes).
Repair enough to read, then map the surfaces
We replaced the faulty components with matched parts and created a head map before doing a large read of all of the surfaces. Unfortunately, the surfaces suffered some degree of physical damage during the time the NAS continued attempting to access it while it was malfunctioning. Unattended network storage tends to continue retrying for longer periods of time than humans would continue attempting to access a drive that is known to be damaged.
Read the Linux layout, not a Windows one
After we completed the imaging, we then decoded the recovered image as it was originally constructed: as a Linux file system structure written by WD's firmware (not as a Windows NTFS disk). Once we did so, shares reappeared in their normal configuration. It is this structure that allows recovered data from a NAS device to be used instead of being merely available.
Where it ended up.
All of the data was recovered in full and copied onto new media, and then the empty enclosure was returned under warranty. The two sides of the dilemma have both been completely resolved, in the order they have to be.
Related on the index.
More from NAS & RAID.
Recognise your own drive in this story?
The same rule applies for all cases previously shown: turn off the power first. Then allow your free diagnosis to complete prior to making a decision.