A My Cloud used as the safety net for a business. Everything was copied onto it, the source was reformatted, and only afterwards did anyone try to read the files back. About a terabyte would not open.
The client had a My Cloud on the network and treated it as their backup — accessible from anywhere, no drive to carry. When their working machine started running slowly they decided to reformat it, and sensibly copied everything across first.
The reformat went ahead. Then they went to pull the files back and found roughly a terabyte of it corrupt or unreadable. By that point the original no longer existed.
Not malware, which is where the client’s mind had gone. An incomplete transfer.
Copying a terabyte to a network device over Wi-Fi takes many hours. Where the connection drops, the machine sleeps, or the transfer is interrupted and resumed, files can be written partially — and the failure mode is unhelpfully quiet.
A file’s recorded length and its actual content are separate things. The directory entry is written with the size the copy expects, and the content arrives afterwards. Interrupt that and you have an entry claiming four megabytes with rather less than four megabytes behind it. Ask the file system how big the file is and it answers correctly. Ask it to hand over the contents and the tail is missing or is whatever previously occupied those blocks.
So to anyone looking at a file browser the backup appears complete. Every filename is there, every size looks right. The gap only shows when something is opened.
That is the trap in this case and it is entirely avoidable: the backup was never verified before the source was destroyed.
My Cloud units are frequently described as using a proprietary file system that makes recovery difficult. They do not.
A My Cloud is a small Linux computer with a standard ext4 file system on an ordinary WD drive inside it. Nothing about the storage is exotic, and the appliance, its firmware and its web interface can be removed from the equation entirely once the disk is imaged.
That makes the prospects better rather than worse. ext4 is thoroughly documented, keeps a journal, and scatters backup copies of its superblock at known intervals across the volume — so damage to the structures at the front of the disk is routinely survivable using a copy from further in.
The drive was removed from the enclosure and imaged read-only. It had bad sectors of its own, which very likely contributed to some of the failed writes in the first place, so imaging ran in graduated passes with unreadable areas logged as gaps rather than allowed to stall the process.
From the image the ext4 structures were rebuilt, using an intact backup superblock where the primary was damaged and replaying the journal to bring the file system to a consistent point.
Files were then sorted into three groups rather than returned as one pile, which on this particular fault is the difference between a useful result and repeating the client’s original problem:
Complete and verified — opened, rendered or checked in their own application, and confirmed sound.
Truncated — the recoverable portion returned and clearly marked as partial, with the shortfall stated, so the client could judge whether a half-complete document was worth keeping.
Nothing usable — listed by name and path so the loss was explicit rather than discovered months later.
Handing back a large file count without that triage would have reproduced exactly the situation that caused the loss: material that looks present and is not.
About 95%. Most of what appeared corrupt was in fact complete on the disk and simply unreachable through damaged directory structures. A genuine shortfall remained where files had only ever been partially written — no recovery retrieves data that was never transferred.
The client moved to a two-copy arrangement with one copy off the premises. The more useful change was procedural.
Before wiping any source, open a sample from the copy. Not the file listing — the files. A document, a photograph, something large from the middle of the set, and something from the end of the alphabet where an interrupted transfer is most likely to have stopped. Five minutes of that would have caught this before the original was destroyed.
And a single NAS is one copy in another box on the same network, not a backup strategy. NAS recovery is from £500 +VAT, single drives from £300 +VAT.
Send the device over with a short account of what happened to it. The diagnostic costs nothing and commits you to nothing, and the figure that follows is fixed in writing before anything is opened.
Nothing can begin until the device is on the bench, which makes this the only step that needs anything from you. Pack it properly, put your details in with it, and send it over. What follows is a free diagnostic and a written figure, in that order.
Posting it? Use something tracked and insured — whatever is on the drive is worth considerably more than the postage. Bringing it in? Weekdays, 9am to 5:30pm, and it still wants packing as above for the journey.
Tell us what the device is, what it is doing, and anything already tried — an engineer will come back with a realistic view of the odds and a price band, before you commit to posting anything.
We’ll be in touch shortly. For anything urgent, call 0161 871 0788.
Usually an interrupted transfer rather than anything sinister. Large copies over a network take hours, and a dropped connection or a machine going to sleep can leave files partly written — correct name, plausible size, truncated contents. The listing looks complete.
Open things from it. Not all of it — a sample: a document, a photograph, something large from the middle of the set, and something from the end of the transfer, which is where truncation shows up first. Do that before deleting anything from the source.
No, and it is often described as harder than it is. A My Cloud is a small Linux machine with a standard ext4 file system on an ordinary drive. Once the disk is imaged the reconstruction is conventional and the appliance itself becomes irrelevant.
Only the part that arrived. Where a file was truncated mid-transfer, the portion that was written can be returned — useful for video and some documents, useless for many formats. No recovery retrieves data that never reached the device.
No. A single NAS is one copy in another box on the same network — exposed to the same power event, the same ransomware, the same theft or fire. A backup means at least two copies with one of them elsewhere.
From £500 plus VAT for NAS work, or from £300 plus VAT where it is a single disk with no array to reconstruct. Fixed in writing after a free 48-hour diagnostic.
Start with an instant online quote, or call and talk it through with us first. You'll have a clear, fixed price before any work begins.