When his WD 3TB drive gave out inside a Synology, the array flagged "bad segments" and rescued what it easily could — "The RAID recovered half the files" — leaving behind the pictures and documents that matter, "which I know are still there". What he needs is precise: the missing set recovered while "retaining the folder structure", before he retires a NAS that is "now at end of life". The half that survived was simply the half the drive could still read; the rest sits behind bad sectors the NAS was never built to fight, and preserving folder structure is exactly what separates laboratory recovery from crude carving.
| Media | Western Digital 3TB drive from a Synology NAS — surface degradation reported as bad segments, the system's own copy attempt abandoning the difficult half |
| Reported situation | Drive failing in service with bad-sector reports · NAS-level recovery salvaged roughly half the contents · photographs and documents among the missing remainder · directory organisation required intact · enclosure due for retirement once the data is safe |
| Fault class | Degraded media on a member drive — readable and unreadable regions interleaved, with the target files' blocks and the filesystem's own records scattered across both |
| Equipment used | Drive removed from the enclosure to native SATA — never worked on through the NAS · head-mapped imaging on DeepSpar Disk Imager with tight retry limits, easy regions first and bad zones revisited last · PC-3000 UDMA behind the imaging for firmware and head control · the Synology volume's Linux filesystem parsed from the image in UFS Explorer, paths and folders reconstructed · targeted extraction of the missing set, hash-verified |
A NAS meets bad sectors with the patience of an appliance, which is to say very little. Its job is serving files, so its copy attempts retry briefly and move on — "half the files" is the honest census of what surrendered easily. The abandonment says nothing about the other half being gone; it says the appliance ran out of tools, exactly where appliances do.
His conviction that the files are still there is probably correct. Bad segments arrive scattered, not as a solid wall. Between and around them, most of the surface still reads — but a single stubborn sector mid-file makes a NAS drop the whole file, so intact data gets abandoned wholesale for the sin of one difficult block. Laboratory imaging works at the sector level precisely to stop that wastage.
The imaging strategy is triage in disguise. Everything easy is taken first at full speed, banking the bulk of the surface while the drive is at its healthiest. Then the difficult regions are approached with controlled retries, per-head handling and skip-and-return ordering — spending the mechanism's remaining life on the smallest, hardest part of the map last. The half the NAS abandoned mostly lives in the first, easy category.
Folder structure is a filesystem promise, and it changes the method. Carving finds file bodies but loses names and paths — useless for a documents tree someone must navigate. Keeping the structure means recovering the volume's own records from the image and reading files through them, so everything comes back where it belonged. Synology volumes are Linux-format underneath, invisible to a desktop but entirely legible to the right tools working from a completed image.
The honest arithmetic concerns the truly unreadable residue. Sectors that never yield leave holes; a file crossing a hole comes back damaged or incomplete, and which files those are is geography, not choice. The reconstruction names them explicitly — a known list of casualties beats a vague promise, and in cases like this the list is typically short against the recovered whole.
Retiring the NAS is right, and the order matters. Nothing about the enclosure is touched until the recovered set is verified elsewhere — and the failing drive itself retires with the box, whatever fraction of it imaging redeemed.
The drive came out of the enclosure to native SATA and was imaged on DeepSpar Disk Imager with head mapping and tight retry limits — easy regions banked first, bad zones revisited last — with PC-3000 UDMA behind it for firmware and head control. The Synology volume's Linux filesystem was parsed from the completed image in UFS Explorer, the directory tree reconstructed, and the missing pictures and documents extracted along their original paths, hash-verified, with any hole-damaged files named.
Head-mapped imaging around the bad segments, the volume rebuilt from its own records, and the missing half extracted with folders intact. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode: the NAS rescued the easy half and resigned. Sector-level imaging goes back for the rest — and because the volume's own records survive on the image, your folders come home as folders, not as ten thousand renamed orphans.
Stop the enclosure's recovery attempts — appliance retries grind a degrading surface without laboratory retry control. Don't plug the bare drive into a desktop; NAS volumes look blank to Windows and the initialise prompt overwrites them. If folder organisation matters, say so up front, because it dictates filesystem-level recovery rather than carving. And keep the NAS powered down until the recovered set is verified somewhere new — retirement comes after the data, never before.
Our case files are written up from genuine enquiries our lab has handled for customers across Sheffield and South Yorkshire, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery approach our engineers apply to that fault, using the equipment listed.
Free diagnostic, fixed quote, no fix no fee — start now or call the freephone.