Data Recovery Case File · Portable Drives · The Hand-Me-Down Backup

Stuck at 56%, every single time: a second-hand drive, one giant archive, and the arithmetic of inherited wear

The sequence in this enquiry is one this archive wants on record, because every step felt reasonable and the sum was jeopardy. He was gifted a used external drive as a backup medium; he moved his photo collection onto it as a single large archive file; the transfer went fine, so he deleted the originals from his main drive. A month later he went to bring the archive back — and the copy slowed, stalled, and stopped at 56%. Not once: every method he tried, including attempting a disk image, met the same wall at the same place. One number, repeated identically, is not bad luck. It's a map reference.

What "always 56%" means — and why one big file made it worse

A copy that dies at a random point each time is a drive misbehaving generally; a copy that dies at the same point every time has hit geography — a patch of failed sectors at a fixed position, which every sequential read of that file must eventually visit. His drive, being second-hand, carried service history nobody could see: hours, heat cycles and wear from its previous life, with a tired region waiting at what happened to be the archive's 56% mark. Which exposes the second, quieter risk he'd taken: one giant archive file concentrates jeopardy. A thousand individual photos scattered across a drive meet a bad patch as a handful of casualties — most files never touch it. A single multi-gigabyte archive is one continuous body lying across the same landscape: any bad patch anywhere along its length halts every whole-file copy, and naive tools then treat the entire archive as one unrecoverable object — thousands of intact photos held hostage by the sectors under one stretch of it. The good news hiding in the structure: archive formats are built from independent, individually-compressed entries with a table of contents at the end, so a wounded archive is a damaged train, not a destroyed one — most carriages survive a break in the line.

The recovery

The drive was imaged on the PC-3000 the way desk-level copying never could: the cooperative expanse — everything either side of the 56% wall — banked first at pace, then the failed region itself worked in bounded, error-managed passes that recovered a further fraction of it and mapped the truly dead remainder precisely, rather than stalling forever at its edge. On the image, the archive was repaired as a structure: its table of contents read, every entry checked, the intact carriages — the overwhelming majority of the photo collection — extracted whole and verified by opening images across the set, and the casualties confined to the specific photos whose compressed bodies crossed the dead ground, named in an honest ledger rather than rounded away. Delivered unpacked, as loose verified files on new media — for the reason the whole case teaches.

Outcome

The collection substantially home from a drive that had held it hostage at a percentage sign — and three rules his sequence writes for everyone else. A second-hand drive is a drive of unknown mileage: fine as a spare copy, never as the copy — its history didn't reset when it changed hands. Don't back up as one monolithic archive: loose files, or several smaller archives, turn a bad patch from a hostage crisis into a casualty list. And the oldest rule in this archive, which his deletion of the originals briefly suspended: nothing is backed up until it exists in two places — a move is not a backup, however new the destination looks. However used, either.

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.