Data Recovery Case File · NAS & RAID · The Body and the Estate
"Do you think you would be able to repair it for me?" — the honest answer split his question in two: the appliance is one job, the data is another, and the data never needed the dead board at all
He arrived with his homework already done — a four-bay Synology unit that "is not starting up," and the manufacturer's own support verdict pasted in full: the unit appears to have a fault with the main system board; it is out of warranty; they don't repair out-of-warranty units; he "may wish to find a third-party company" — though they were unable to recommend one. Then his question: "Do you think you would be able to repair it for me, how much will it cost?" The most useful thing an honest lab can do with that question is gently take it apart, because two different services are hiding inside it. Repairing the appliance — the box, the body — is an electronics job of debatable economics. Recovering the estate — everything stored on the drives inside — is a data job that does not require the body to be repaired, or even to work again, ever. The manufacturer's letter, candid as far as it went, never mentioned that distinction. It is the whole case.
| System | Four-bay Synology NAS — family and working files aboard, unit dead on its main system board per the manufacturer's own remote diagnosis |
| Reported situation | Unit not starting · manufacturer confirms board-class fault, out of warranty, no repair route offered · owner directed to third parties without a recommendation |
| Fault class | Enclosure/board failure — the storage array and its contents presumed intact on the member drives |
| Equipment used | Drive removal with bay-order preserved · per-member write-blocked imaging · virtual assembly of the NAS's array and volume structures from the images · verified extraction to new media |
The decode: where a NAS actually keeps its data — and why the dead board is a bystander
The architecture that saves him: a NAS is a small computer bolted to a set of drives, and the division of labour matters enormously on the day it dies. The computer part — the main board the manufacturer diagnosed — runs the interface and shuttles files about. But the arrangement of the data — the array layout binding the drives together, the volume structures, every file — is written onto the drives themselves. The board holds none of it. When a board dies, the estate doesn't die with it; it sits in the dark, complete, waiting for something that can read the drives directly. His unit failing to start is the failure of the librarian, not the burning of the library.
The manufacturer's letter, graded fairly: as hardware diagnostics go it was honest and even helpful — a board fault named remotely, warranty status stated plainly, third parties suggested rather than false hope offered. Its silence on the data was the silence of a hardware process, the same silence heard in the sealed-tablet genre: warranty machinery thinks in units, not in what units contain. He did exactly the right thing with that silence — he went asking further questions instead of assuming the answer.
The economics, stated before they're asked: board-level repair of a retired consumer NAS is often uneconomic against simple replacement of the enclosure — and, more to the point, it is never a precondition for the data. Which reorders his whole question: secure the estate first; then decide at leisure whether any box, old or new, deserves the repair money.
The recovery: the drives out, the array rebuilt where no board is required
The member drives came out with their bay order recorded — cheap discipline that keeps later analysis tidy — and each was imaged write-blocked, untouched beyond reading. From the images, the NAS's own array and volume structures were identified and assembled virtually: the same layout the dead board once presided over, reconstructed on the bench without it. The volume mounted read-only from that assembly; folders, names and dates presented exactly as he'd last seen them through the unit's interface; and the estate was copied out and verified.
The outcome
Everything the box had held, delivered on new media — with the drives and the unit returned as they arrived. His cost question was answered in the only form worth having: free assessment, then one fixed written figure including VAT, under no-recovery-no-fee — and a plain note that the figure was for the data, with the fate of the appliance left entirely, and affordably, up to him.
NAS dead but the drives probably fine
Don't keep power-cycling a unit the maker has already called board-faulty, and above all don't slot its drives into another NAS "to see" — a new enclosure can adopt and initialise them, converting a bystander fault into real loss. Keep the drives in bay order, powered down. Remember what redundancy actually covers: a mirror or parity set protects against a drive dying, not against the enclosure, its power supply, deletion, or ransomware. A NAS is not a backup — the estate deserves a second home the day it comes back.
The data doesn't need the box repaired — call Edinburgh Data Recovery on 0131 202 0491; drives read directly, one written figure, no recovery, no fee.
Request a quote online →
Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.