Case file · Desktop externals · BDR-2025-0935
The Folders That Were There Are Gone.
The worst thing a backup can do (show up normally with no files anywhere on it) happened to this user, whose dedicated backup drive had thousands of photos, videos and documents backed-up
. They were unable to find/view any of the folders that were previously there
when the backup started showing up blank. Troubleshooting didn't help; now the backup itself was the emergency.
Same symptoms on your desk?
01273 964902
The decode.
If your whole folder tree disappears after you mount a drive (which is usually a sign something is broken) then it is probably your index that is broken. NTFS stores all its file records in one big table called the Master File Table. When that table is damaged (from an interrupted write, a dying area on the disk that just happens to contain those sectors, or a partition entry gone wrong), everything is hidden at once, but all of the other file data remains intact and will not disappear. Your first concern at this point should be whether or not your drive is failing. If so, your strategy for dealing with the problem has changed entirely.
Equipment on this case.
How a job runs here →| The system | What it did on the day | Why this tool |
|---|---|---|
| ACE Lab PC-3000 Express | Firmware-level health check that ruled hardware failure in or out | Speaks the vendor-specific commands that sit outside the standard ATA set |
| R-Studio Technician | Rebuilt the NTFS master file table from surviving records | Broad filesystem coverage and dependable RAID re-assembly |
| UFS Explorer Professional Recovery | Content-based recovery for files whose records were beyond repair | Stronger than most on the awkward filesystems — APFS, ReFS, XFS, ZFS, Btrfs |
At the bench.
Rule out hardware before believing it's logical
We first ran some tests at the firmware level to determine if the drive was failing physically. We ran SMART attribute checks, examined the defect list, and looked at how many blocks had been reallocated. All tests indicated that the drive was operating within mechanical specifications. This moved our attention from hardware issues to software or logical problems and allowed us to start imaging at normal speed rather than under emergency conditions.
Image anyway, then rebuild the index on the copy
Even though the drive passed all physical testing and did appear to be operational, we still created an image of the drive. Once that image was created we would have been able to perform any number of repairs to the filesystem. However, filesystem repair tools write to the drive, and any changes made by them would have been irreversible. Therefore, we worked solely with the image. Working through the image revealed significant damage to the Master File Table (MFT) across a defined range. Specifically, the damaged records contained information about where the actual file data was stored. Fortunately, a substantial proportion of these records were recoverable from the MFT's own backup copy located elsewhere on the filesystem.
Carve what the index can't account for
Files whose records were unrecoverable were retrieved by content signature instead — photos, videos and documents identified by their internal structure, then sorted by type and embedded date so the results arrived organised rather than as a numbered heap.
Where it ended up.
Thousands of photos, videos, and documents were recovered, most of them under their own name and organised in their own folders, while others were grouped by category and/or date. Because of an included lecture (a complimentary service) about replacing the one backup drive with two drives, that's what the owner did.
Related on the index.
More from Desktop externals.
Recognise your own drive in this story?
Use the same principle from all previous cases: shut off the power, and then allow the diagnostic process to proceed without cost to you before making any decisions.