Case notes · Logical & corruption · EDR-2025-2311
A Life's Work Behind a Cyclic Redundancy Check.
The 3TB Toshiba outlived her old laptop, then met its replacement badly: every big copy died on the same refusal — cyclic redundancy check
— first odd files, then whole folders. She tried a new lead, a neighbour's PC, and a free rescue program that ran all night for 4%
before she pulled the plug. At stake: every drawing since my apprenticeship
— hull lines, sail plans, the photographs of finished boats.
Sound familiar? Talk to us.
0800 6890668
In plain English.
'Cyclic redundancy check' sounds like a software complaint, and that is how it misleads. What it records is a failed delivery: the drive handed over data that no longer matched the checksum stored beside it. A scrambled filesystem on a healthy disk can do that; so can a surface losing its hold on the data — and the two need opposite treatment. Guess wrong and the price is sectors, because Windows will hammer a weak patch until the drive is spent.
The tools this job took.
See how your case runs →| System | The part it played | Why we rate it |
|---|---|---|
| DeepSpar Disk Imager 4 | Imaged the Toshiba on a strict retry budget, weak stretches noted and saved for later | Images by head map — timeouts, resets and power-cycling managed per head |
| R-Studio Technician | Did every repair on the finished image, never the drive | Handles most filesystems; dependable RAID reassembly |
| ACE Lab PC-3000 Express | Questioned the firmware first — SMART history, defect lists, the overnight session on record | Talks to a drive through the maker's own commands, not just standard ATA |
On the bench.
Ask the firmware which illness it is
Nothing was copied until the drive had been interviewed at firmware level, and its counters gave the game away: reallocated sectors climbing, pendings queued — and one long power-on session that could only be the rescue program's overnight siege. A surface in decline, then, not a filesystem in a tangle, and each calls for the other's opposite.
Spend the retries like money, which they are
The imaging rules were set before the first sector: short timeouts, a hard cap on attempts, and any read that resisted was logged and passed over instead of besieged. Stubborn regions got small, separate visits right at the end. The arithmetic is the point — winning one bad sector by force can cost the thousand good ones still waiting behind it.
Repair the copy; leave the patient switched off
Once the image existed, the Toshiba never spun again. On the copy, the news was better than feared: the folder tree had held, and the losses sat where the weak band crossed a run of scanned sheets. Each casualty was identified by name — a short written list of drawings, not a shrug — so she knew precisely what the surface had kept from her.
How it ended.
She had named the folder that mattered most — the lines plans — and that one was checked first, before any invoice existed. The archive went home on a new drive, complete bar the listed scans, most of which she can redo from paper originals still in the loft. Two copies live on, one at her son's.
More like this in the casebook.
More cases filed under Logical & corruption.
Sound like your drive?
Every case here points the same way: power the device down, take the free diagnosis, then make your decision.