A firmware update on a RAID controller completed, rebooted, and presented an empty array. Every disk was healthy, every byte was intact, and the only thing missing was the description of how they fitted together.
← All case files · from £500 + VAT
An surveying practice lost access to its project server after a routine controller firmware update. The card came back reporting no configured arrays and offering to create a new one — which, had anyone accepted, would have ended the job before it started.
On the volume sat live project files, drawing archives and the practice’s document management database. No disk had failed. The array had simply ceased to exist as far as the controller was concerned.
no reallocated sectors of consequence, no mechanical symptoms, and full capacity reported by each.
meaning a complete reconstruction was arithmetically possible rather than merely hopeful.
the update had cleared its configuration, and this card family does not write a redundant copy back to the members in a recoverable form.
The temptation here is obvious: accept the card’s offer to create the array and hope it lands on the original layout. On many controllers creating an array initialises it, and initialisation writes — which would have overwritten the structures needed to work out the geometry in the first place.
Each disk was imaged read-only before any analysis, and all subsequent work ran against those copies. The geometry was then derived from the data itself: stripe boundaries located by looking for structural repetition across members, disk order established by testing candidate permutations against known NTFS structures, and parity rotation and offset confirmed by verifying that reconstructed stripes produced valid metadata rather than noise.
With the layout proven rather than assumed, the RAID 5 was assembled virtually and the file system mounted from the assembled volume — entirely outside the controller that had caused the problem.
Everything. Six healthy disks, intact parity and only the map missing meant a complete recovery, with folder structure and filenames preserved throughout.
The document management database was checked for structural consistency rather than simply copied, and project files were opened in their applications before the set was returned on fresh media.
A lost configuration is not lost data. When a controller forgets an array the contents are still on the disks exactly as they were — what has gone is the description of how to read them, and that can be reconstructed by analysis.
The thing that turns this into an unrecoverable job is what happens next. Don’t create, initialise or re-scan the array to see whether it comes back. Power the server down, keep the disks in bay order, and note the controller model. Server and RAID recovery starts from £500 +VAT after a free 48-hour diagnostic.
Bring the drive to our Oxford Science Park reception, or post it over — it costs nothing to learn what went wrong. You’ll have a written price from our fixed bands before any work starts.