Data Recovery Case File · Vintage & Veteran Drives · The Afternoon That Bought 33GB
Everything visible, almost nothing reachable: a veteran iMac's drive in slow collapse — why the Finder was the wrong harvester, and the four partitions imaged whole while the mechanism still had afternoons left
His enquiry read like a well-kept logbook. A 27-inch iMac from 2010, its internal drive laid out in four named partitions — Macintosh HD for the system, INT SCRATCH, STORAGE 1, STORAGE 2 — switched off for several months, then woken "to work with some important files." The waking was the horror: "the drive is failing." His preparation showed immediately — he booted from a cloned external drive he'd made, over the machine's vintage FireWire connection, and from there could see the ailing drive's partitions load "eventually" onto the desktop: "I can see all my files and even open them — very, very slowly." Then the honest arithmetic that brought him in: copying to an external "grinds to a halt after an hour or two"; a full afternoon of effort yesterday moved just 33GB of important files. At that rate, with the drive audibly worsening, the sums didn't work — and recognising that a working method can still be a losing one is exactly the judgement this archive exists to reward.
| System | 27" iMac (2010) — internal hard drive in four partitions (system, scratch, and two storage volumes) · owner equipped with his own bootable clone of the system |
| Reported symptoms | Months powered off, failing on return to service · partitions mount slowly but completely · files open at a crawl · bulk copies stall after an hour or two · one afternoon's disciplined effort: 33GB rescued |
| Fault class | Progressive read degradation — an ageing mechanism paying dearly for every sector, worsening with each session |
| Equipment used | Drive removed from the machine · hardware imagers with error-tolerant, drive-friendly sequencing · per-region prioritisation across all four partitions · full extraction and verification from the completed image |
The decode: why "slow but working" is the most expensive state a drive can be in
What the crawl really was: a healthy drive answers a read in a fingersnap. His was answering many reads only after long internal ordeals — retrying weak sectors, recalibrating, relocating what it could — each success bought with dozens of attempts. "Very, very slowly" is the outward face of a mechanism running its own emergency room for every request. The files were intact; the act of reaching them had become the illness.
Why the Finder made it worse: file-by-file copying is the most punishing possible workload for a drive in that state. It demands files in whatever order the folder tree dictates — sending the heads sprinting back and forth across a surface they can barely navigate — insists on completeness, retries failures aggressively, and stalls the whole queue behind every ordeal. His "grinds to a halt after an hour or two" is that pattern's signature: not the drive dying on schedule, but the method spending the drive's scarce good behaviour at the worst possible exchange rate. Thirty-three gigabytes per afternoon wasn't a poor effort — it was a heroic one, achieved against a method stacked the wrong way.
The clone, credited: his bootable FireWire clone deserves its paragraph. Because he'd made it, the failing drive was never asked to also run the operating system while being harvested — a burden that halves many drives' remaining life. Owners who maintain a bootable copy of their system are rare, and the habit paid him twice: a working machine to write from, and a patient spared double duty.
The recovery: one pass, sector-level, on the drive's terms
The drive came out of the iMac and onto hardware imagers, which invert every one of the Finder's cruelties: they read at sector level in drive-friendly order rather than folder order; they harvest the healthy expanse first at full speed; they skip each difficult region on first refusal and return for it later with adjusted patience; and they stop the moment the mechanism's behaviour says rest. All four partitions were captured into one complete image across the drive's remaining good spells — the storage volumes prioritised, the system and scratch areas following. From the image, on healthy hardware, everything then moved at modern speeds: his important files, the full contents of both storage partitions, the lot — with his hand-rescued 33GB reconciled in so the frightening afternoon was honoured, not wasted.
The outcome
The whole estate delivered on new media in a fraction of the time the arithmetic had promised, verified by opening files across all four partitions. Free assessment, one fixed written figure including VAT, no recovery, no fee. The sixteen-year-old drive was retired with respect — and the clone habit that served him so well was recommended onward to the machine's successor, this time in duplicate.
Old drive failing — files visible but agonisingly slow
Do the arithmetic early, as he did: if an afternoon buys gigabytes while the estate is measured in hundreds, the method is losing and the drive is paying. Stop bulk-copying through the file manager — every crawl-speed session spends mechanical life on overheads. If you rescue anything by hand, take the few most irreplaceable items only, then power down. A drive in slow collapse has a finite number of good spells left; the winning move is spending the next one on a sector-level image that reads everything once, gently, in the right order.
Stop spending its good spells — call Edinburgh Data Recovery on 0131 202 0491; imaged once, gently, 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.