When a server won't boot, its RAID drops offline, or a host full of virtual machines goes dark, a whole business can stop with it — orders, accounts, email and client records all out of reach. The data is nearly always still recoverable, off the failed hardware. We recover Dell PowerEdge, HPE ProLiant, Lenovo and Supermicro servers for businesses across Belfast and Northern Ireland, in-house.
$ bdr diagnose /dev/server → Server: Dell PowerEdge · RAID 5 · 6 × 2 TB → Status: ARRAY OFFLINE — 2 disks failed → Client: confidential · Belfast $ bdr engineer-working → Member disks: all 6 imaged read-only → RAID 5: rebuilt off the controller → Windows Server: volumes mounted $ bdr verify → ✓ file shares — 6.1 TB → ✓ SQL databases — restored → ✓ server recovered — data back
On a downed server the instinct is to force a RAID rebuild, reinitialise the array, or reinstall — and each can turn a recoverable situation into permanent loss. A rebuild onto a failing disk is the most common cause we see. Note the disk order, stop re-seating disks, power the server down, and let us image the disks before anything overwrites them.
Servers fail across their disks, their arrays, their operating systems and the virtual machines they host. These are the situations we recover most.
We recover every make of server, on every platform, array and file system they run.
Dell PowerEdge, HPE ProLiant, Lenovo ThinkSystem, Supermicro, Fujitsu PRIMERGY, Cisco UCS and IBM System x — running Windows Server, VMware ESXi, Hyper-V and Linux, on PERC and Smart Array controllers, SAS and NVMe, across NTFS, ReFS and every RAID level, with SQL, Exchange and Active Directory.
Server recovery combines RAID reconstruction with whatever sits on top — a file system, a database, a virtual estate. We read every disk, rebuild the array virtually from read-only images, and recover your volumes, VMs and databases from that copy, never the original disks under strain.
We check each disk, identify the array, controller, operating system and any databases or VMs, and find out what failed and in what order — then send a written quote.
Any failing disk is repaired or imaged first — heads, board or firmware — so no disk is put under strain during the recovery.
Each disk is imaged read-only, and from then on all work is done on the copies, so your original disks are never altered.
We determine the true disk order, parity and layout from the images and reassemble the RAID virtually, independent of the failed controller.
From the rebuilt array we extract your volumes, virtual machines, databases and files — and repair databases or VM disks where they're damaged.
Recovered data is checked to confirm it opens and is intact, and we can show you what's come back before anything is returned.
Your data comes back on a fresh drive sized to the recovery, or via secure transfer — whichever suits you.
From a server that won't boot to an offline RAID, a downed ESXi host or a corrupt SQL database, we recover servers of every make — rebuilding the array from read-only images and extracting your volumes, VMs and databases.
Give us a few details about what went wrong and an engineer will come back to you, usually inside one working day.
We will get back to you soon. If it is urgent, call 028 9002 0144.
A free diagnostic first, then a fixed written quote before any work begins. Server recovery starts at £500 + VAT for a single-drive or mirrored server, with multi-disk RAID servers — RAID 5, 6 and 10 — from £800 and quoted per case, and no fix, no fee on most jobs — where disks need physical repair, a 50% deposit covers parts and bench time, with the balance only on success.
A representative selection of servers we've recovered across different platforms and faults — configurations and outcomes shown, customer details kept private.
Two disks had failed close together. We imaged the readable disks, rebuilt the RAID 5, and recovered the shared drives and user data in full.
The array had dropped after a power event. We rebuilt the RAID 10 from images and recovered every virtual machine on the host.
A crash had left the mailbox database in a dirty state. We rebuilt the mirror, recovered the database files, and repaired them to a mountable state.
A rebuild had corrupted the array holding the datastore. We imaged all twelve disks, reconstructed the RAID 6, and recovered the VMs.
The single boot drive had a controller fault. We revived it at board level, imaged it, and recovered the system and web data.
A volume had been removed by mistake, but the data survived on the array. We rebuilt it and recovered most of the deleted volume's contents.
Post or drop in your device for a free diagnostic, with a note on what happened — an engineer reviews it and confirms your exact quote in writing before any work begins.
First step: get the device onto our Belfast bench. Wrap it well, tuck your contact details in the box, and post it over — the diagnostic costs nothing, and you’ll have a firm written price to approve before we touch a single sector.
Posting it? A tracked, insured service is best. Dropping it off instead? You’re welcome Monday–Friday, 9am–5:30pm — please still pack the device as above.
Not ready to send anything yet? Use the form to describe the fault in your own words and one of the engineers will come back with a quote tailored to your situation.
We will get back to you soon. If it is urgent, call 028 9002 0144.
The questions we're asked most about server recovery.
Almost always. A server that won't boot or POST usually has a hardware or array fault, while the data sits intact on the disks. We recover it off the machine — rebuilding the RAID from images if needed — so you don't have to get the server itself running to reach your files.
Yes — it's one of our most common jobs. Whether the array lost too many disks, a rebuild failed, or the controller died, the data is still on the disks. We image each one and reconstruct the array virtually, independent of the failed hardware.
Yes. Once we've rebuilt the underlying array or datastore, we recover the virtual machines — VMDK, VHDX and the rest — and can repair a VM's disk or the file system inside it, and extract the databases and files within.
Yes. After rebuilding the volume the database sat on, we recover the database files and repair them where they're corrupt or in a dirty or suspect state — SQL Server MDF and LDF, Exchange databases and others — to a usable, mountable condition.
Server recovery starts at £500 + VAT for a single-drive or mirrored server, with multi-disk RAID servers from £800 and quoted per case after the free diagnostic. There's no fix, no fee on most jobs; where disks need physical repair, a deposit covers parts and bench time, with the balance only on success.
On most jobs, yes — no recovery, no fee. Where individual disks need physical repair, a deposit covers those parts and the bench time; the balance is still only charged on success. We're clear about which applies before any work begins.
All of them — Dell PowerEdge, HPE ProLiant, Lenovo ThinkSystem, Supermicro, Fujitsu, Cisco UCS and IBM System x — running Windows Server, VMware ESXi, Hyper-V or Linux, on any controller and RAID level.
Usually not — the drives are what hold your data. Send all the disks from the array, numbered by slot, and we can rebuild it here. If the fault turns out to be tied to specific hardware, we'll let you know if we need more.
Yes. We recover the volumes from a failed domain controller, including the Active Directory database and SYSVOL, so they can be restored or examined. As with any server, we rebuild the underlying storage first, then extract and repair what sits on it.
Most server recoveries take around 5 to 10 working days, depending on the array size, the number of disks, and any database or VM repair needed. The diagnostic is usually finished within 48 hours, and we can normally prioritise business-critical cases.
Drop the disks in at our Belfast office Monday to Friday, 9am to 5:30pm, or send them fully insured. Number the disks by slot, pack them so they can't knock together, and include your contact details so we can book them in and quote after the free diagnostic. For business-critical jobs, tell us and we'll prioritise.
A free diagnostic, fixed written pricing from £500 + VAT, and no fix no fee on most jobs — every make of server recovered in-house, right here in Belfast. Don't rebuild or reinitialise — power it down and send the disks in.