Missing or corrupt files · lab log · PLY-2025-1113
It Hit the Floor and Still Sounds Right.
Partway through a backup, a shelf bracket above a desk on the Isles of Scilly gave way, and the My Book landed on the floor with its two leads still plugged in. It carried the family’s video collection entire, camcorder tapes transferred over included. Afterwards nothing about the thing looked wrong: it powers up exactly as it always did
, there are no strange noises whatsoever
, and Windows will still cheerfully offer Eject My Book
. What never appears is the drive letter.
Sounds like yours? Talk it through.
0800 6890668
Why it happens.
The drive letter comes at the end of a queue in which everything ahead of it must work: heads and motor first, then the small board that bridges USB to SATA, then the partition table, then the filesystem. Here the first two behaved — the disk reached speed and introduced itself to the computer — so the queue had stalled at number three. That is what a write cut off midway leaves behind: broken records, sound mechanism. Enclosures of this type bring a further rule with them. Decryption happens on the little board inside, so both of them arrive at the bench together and stay together. Part them and every sector you read is meaningless.
The kit this job called for.
How the job runs →| The kit | What it did here | Why we keep it |
|---|---|---|
| ACE Lab PC-3000 Express | Cleared heads, motor and platters before a scrap of logical work began | Drops the disk into its maker's service mode, out of reach over plain ATA |
| DeepSpar USB Imager | Imaged through the unit's own bridge, leaving the decryption where it was | Images at full speed over the drive's own USB bridge |
| R-Studio Technician | Repaired the partition table and the NTFS structures on the copy | Reads nearly every file system, and rebuilds RAID sets from images |
How it ran.
Prove the mechanics before believing anything
Sounding right proves nothing whatever. Spin-up current, read rates head by head, both defect lists and the reallocation counters all went onto the meters, and every figure came back where it belonged. Twenty minutes of measurement turned an alarming job into an ordinary one.
Take the copy with the board still in place
The key on a My Book belongs to that small USB board, so the copy was taken with the board still sitting between disk and imager, and it read cleanly from the first sector onward. Pull the naked disk out and read it on a SATA port instead, and what you get is a flawless copy of nothing usable — all the bytes, none of the meaning.
Mend the copy and leave the disk alone
From the copy the damage looked recent and very small: a single partition entry abandoned mid-write, in the second the unit came off the shelf. The image was where it got mended. Nothing was ever written to the drive, and the volume mounted at once, folders and timestamps in their proper places.
How it ended up.
Nothing had gone. Counted off the copy, the files tallied with what the folders claimed. From the first meter reading to the last, disk and bridge were handled as one item, and on hardware built this way that single habit does most of the work.
Pages people open after this.
More on corrupt volumes and deleted files.
Seeing the same thing?
Leave it switched off and post it in. Nothing happens until the diagnosis, which lists what can still be read and what cannot.