Data Recovery Case File · SSD · The Serviceable Kind

Correct name, correct size, SMART OK, nothing readable: the SSD failure that can actually be fixed

Elsewhere in this archive sits an honest page about the SSDs nobody can currently open. This is its counterweight — the failure that looks just as final from the outside and isn't. A storage SSD holding 500–600GB of a musician's old studio sessions "gave up the ghost" without warning: macOS greeted it with the eject-or-initialise dialog, while System Report cheerfully listed it in full — correct name, correct capacity, SMART status OK — and Disk Utility could sometimes see it, greyed out, wearing its old volume name like a ghost's overcoat.

How a drive passes its physical and fails its purpose

His symptom set — perfect self-identification, zero readable data — draws the line inside an SSD better than any diagram. The controller was alive and well: answering the bus, stating its name and size, reporting SMART attributes that were, in their narrow way, honest (no failing memory, no dying electronics). What had broken was the thing SMART doesn't measure: the controller's map — the translation layer that records where every scrap of data physically lives across the memory chips, rebuilt and rewritten constantly by design, and vulnerable, on many controller families, to being caught mid-update by an unlucky disconnect or power blip. A controller with a corrupted map is a librarian with amnesia in an intact library: perfectly able to tell you the building's name and dimensions, unable to fetch a single book. Hence the greyed-out ghost in Disk Utility — the drive remembering it had a volume without being able to walk into it — and hence, crucially, the good news: the sessions themselves were never gone. The map was.

The recovery — and why this one, unlike some, could happen

Map faults are firmware faults, and the decisive question is whether the controller family speaks to anything besides itself. This one did: the drive went onto the PC-3000 in its SSD technological mode, where the controller can be addressed below the level of its broken bookkeeping. From the management data still present across the memory chips, the translation layer was reconstructed — the map redrawn from the territory's own records — until the drive could once again present its contents in order; Data Extractor then imaged the reassembled whole. The HFS+ volume mounted from the image exactly as it had last existed, and the studio archive came out intact: session folders, project files, bounces, the accumulated decade — verified by opening work across the years and delivered on new media. The honest asterisk, stated because this archive states them: not every controller family allows this. Some are open to the trade's tools, some sealed shut — which of the two you're holding is a matter of silicon, not hope, and a free assessment answers it before any money moves.

Outcome

Full recovery of a drive that had been telling everyone it was fine and showing everyone it was empty. Take the pattern home: an SSD that identifies perfectly but reads nothing has a firmware fault, not a memory fault — and firmware faults are often fixable. Decline the initialise, stop the reconnection experiments (each power-up is another chance for a half-broken map to get worse), and have the family identified. And musicians, of all people: sessions are one drive-failure from silence everywhere they live singly. Archive means two.

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.