The array ran degraded for months, then the second drive quit halfway through the rebuild — or the controller died and took the configuration to its grave. RAID 0, 1, 5, 6 and 10 sets out of servers, workstations and NAS enclosures all over South Wales are reconstructed here — image-first, always — with urgent server data recovery for businesses worked ahead of the queue.
Free diagnostic on every raid array job. One fixed quote in writing before any work begins.
No fix, no fee all jobs except electronic and mechanical failures, chip level work, DVR and Forensic jobs. Full pricing is on the data recovery cost page.
Every raid array job starts by matching the symptoms to the fault — these fifteen cover almost everything that reaches the bench.
The array survives one loss; the rebuild then finishes off the oldest survivor — the defining disaster.
A half-run rebuild corrupts the parity it depended on; recovery maps the boundary from the images.
A dead card takes the configuration with it — the geometry is re-derived from the members alone.
Out-of-date data rejoins the set and the file system shreds itself.
A new array built over the old one — stop before initialisation completes.
Two or three members die together; each is repaired and imaged individually.
Drives 'fail' in blocks because the chassis wiring did — diagnosis before panic, always.
The array maths is fine; the NTFS/ext4/VMFS above it isn't — parser-level rebuild.
Definitions removed in error recover while the blocks beneath survive.
Dead cache batteries plus sudden power loss desync parity from data — reconciled at image level.
H710s and P420s fail holding the definition — solved from the drives, no matching card required.
One member with failed heads gets clean-bench repair before joining the imaging queue.
Matched-batch ST4000VN006s in small-business arrays age together — imaged in parallel before the next one quits.
Software arrays lose superblocks and pool metadata — reconstructed from the raw members.
A member remapping sectors daily is failing quietly; catching it at imaging time saves the set.
Rule one never bends: every member drive is captured sector by sector before any reconstruction is attempted, and the originals never spin in an array again. Geometry — member order, stripe size, rotation, offset — is solved from the images themselves, the volume is assembled virtually, and the file system is walked on top of that assembly. Stale members in a RAID 5, double failures in a RAID 6, nested 10s, controllers with amnesia: standard weekly workload.
Forcing an offline drive back in, rebuilding onto a disk that's already failing, and letting someone try 'one last thing' — those three convert recoverable arrays into write-offs more surely than any hardware fault. If the array is down: power off, label every drive with its bay, and call. Say the words 'business-critical' and the diagnostic is bumped to the front of the queue on arrival.
RAID recovery obeys a single law — image everything first — and the bench enforces it:
Member drives image simultaneously on dedicated hardware before analysis starts — no original ever rejoins a rebuild.
Runs against the images to crack the geometry — order of members, block size, rotation, offsets — then mounts the whole array virtually.
Failed members get the full hard-drive treatment first (heads, firmware, electronics) so they can be imaged alongside the healthy ones.
Parity and entropy checks confirm the solved geometry before the file system is extracted from the assembled volume.
NTFS, ReFS, ext4, XFS, BTRFS and VMFS read natively — including the virtual machine disks stored on top.
Member images are taken read-only; the drives you sent go back exactly as they came.
Hardware card, software layer or motherboard chipset — the geometry is written across the member drives, and that's where we read it from. PERC and Smart Array controllers routinely die holding the array definition, and matched-batch NAS drives (IronWolf ST4000VN006s especially) have a habit of wearing out in formation — which is exactly why members are imaged in parallel the day they arrive.
Power the server or array down before removing anything, then label every member with its slot as it comes out — reconstruction depends on knowing the original order. Post the labelled drives only; controllers and chassis stay with you, and a quick phone photo of the bays as they were is genuinely useful.
Most customers post or courier their media to us.
Sending a drive from a computer, laptop, MacBook, iMac, CCTV / DVR or server? Please remove the internal hard drive or SSD and send us just the drive — we don't provide an internal drive-removal service. We don't recover storage soldered to a motherboard (e.g. Apple Silicon Macs and some thin laptops) — only drives that can be removed and sent to us.
↓ Print the booking-in & shipping form (PDF)
Mark the package for the attention of Bristol Data Recovery and we'll call you as soon as we diagnose your media.
Not sure what to send? Call 0800 689 0668 first or use the free online diagnostic.
Free diagnostic, fixed quote, no fix no fee — start now or call the freephone.