Whatever shape the server is in, the objective is the same: getting the data off its drives. Once we have the drives in hand, each one is imaged read-only, any RAID array is built back up from these copies, and finally the file system, along with the files, virtual machines and databases on it, is recovered.
If a server fails to boot, a RAID array has dropped offline or a rebuild has stalled, stop right there. Do not rebuild, re-initialise or re-sync. Rebuilding puts new writes over existing data and can damage the very data we are trying to recover. Turn the server off, note which bay each drive was removed from, and call us. The drives themselves almost always outlive the server, and that first recovery attempt is the safest one.
Most of the server work we receive ranges from servers that can’t boot to corrupted databases: hardware failures, RAID that is offline or degraded, failed rebuilds, and corrupted virtual machines and databases.
No server is outside our range of support, whether it’s Dell, HPE, Lenovo, Supermicro or the rest, and whether it runs Hyper-V, VMware, Linux or Windows Server. Regardless of the make or the role, we get the data off the drives.
Cisco UCS, Fujitsu PRIMERGY and Supermicro; Lenovo ThinkSystem; HPE ProLiant; Dell PowerEdge · blade, rack and tower · Hyper-V, VMware ESXi, Linux and Windows Server · HP Smart Array and Dell PERC RAID · Active Directory, Exchange and SQL Server · NVMe, SATA and SAS.
Send the drives in for a free diagnostic, with the server make, RAID setup and role, and an engineer will scope them out, confirming the exact figure in writing before any work begins.
Give us information about the server (its make and model, the RAID setup it is running, what it runs and what went wrong) and an engineer will size it up and send back a written quote, usually within 48 hours.
Using dedicated hardware, every drive in the box, including the failing ones, gets a read-only image, so everything downstream runs off copies and your originals are never written to.
A drive that has failed, either mechanically or electronically, goes onto our bench for new heads or a rebuilt board, just enough to get an image off it and no more.
If the server ran RAID, we work out the disk order, stripe size and parity from the images, then build the volume back in software without the need for a controller.
We pull everything from the file system (Active Directory, SQL and Exchange databases, virtual machines, files and shares), repairing databases and VMs as needed.
Nothing goes back until the recovered volumes, databases and virtual machines have been opened and confirmed complete.
Your recovered data comes back on brand new storage sized to the server, ready to restore and drop right back into service.
Post the drives or bring them in. Diagnosis comes first and the quote next, and in most cases you only pay if we recover your data.
Tell us what happened and we’ll get back to you, usually within a working day.
We’ll be in touch very soon. If it can’t wait, call 01483 901310.
Server recovery operates on a simple tiered structure: you receive a free diagnostic and see the quote in writing before any work is started.
Examples of recent server jobs: offline arrays, failed hosts, corrupted databases and deleted volumes. Identifying details removed, and each job fully verified. Complete write-ups: an HP ProLiant RAID 5 with two bad disks and a RAID 5 that collapsed mid-rebuild.
Two disks had gone from the array. All six were imaged, the RAID 5 was rebuilt and the volume recovered.
The board was dead, but the drives themselves were good. We built the array off the drives and recovered the Hyper-V VMs.
The RAID itself was fine, but the Exchange database was damaged. The store was recovered and repaired until it mounted again.
The rebuild had written over part of the array. Every disk was imaged, the RAID 6 was reconstructed and the VMware datastore recovered.
The controller on the SSD had died. The NAND was recovered and the Linux file system rebuilt, so the site and its data could be restored.
Very little had been added on top of the data since then. The array was rebuilt and the deleted volume recovered.
Send us your drives for their free diagnostic, with the make of server, RAID configuration and role, and an engineer will assess them and confirm your exact quote in writing before any work begins.
Getting your data back starts when you send us your drives (individually labelled in bay order) or the whole server. Pack it safely, include your contact details, and after we run the free diagnostic we confirm your exact quote in writing before any work begins.
Sending it by post? We’d recommend a tracked, insured service. Dropping it in? No problem: Monday to Friday, 9am to 5:30pm. Please package the device as described above first.
Do you need a scoped quote first? If so, describe your server, RAID setup and fault on the form and one of our engineers will review it and provide you with a quote based on your individual needs.
We’ll be in touch shortly. If it’s urgent, call 01483 901310.
These are the questions we are asked most about recovering a server.
Usually, yes. Servers which will not boot most often have hardware faults or operating system issues rather than lost data. All drives are imaged and any RAID array rebuilt, then we recover the file system and everything on it from a copy.
Yes, this is core server work. We image each of the individual disks that make up the RAID set, analyse those images to work out the array’s configuration (disk order, stripe and parity), then rebuild the volume in software. This lets us recover the data even from a degraded array or a failed rebuild.
Yes. After rebuilding the array and file system, the virtual machines (VMDK, VHDX) are available for extraction just like any other files on the datastore. Where one is corrupt, we repair it so that it can be booted again.
Often, yes. Once the volume containing the database is recovered, we extract the SQL (MDF/LDF) or Exchange (EDB) database and, if it is damaged, repair it to a consistent, mountable state, ready for reattachment.
We work in tiers: a single-drive or mirrored server is from £500 + VAT, and a multi-disk RAID server (RAID 5, 6 or 10) from £800 + VAT. Every project starts with a free diagnostic and a written quote.
You will always receive a free diagnostic and a written quote first. For multi-disk, mechanical and database projects, a deposit is required against that quote, with the balance only due on a successful recovery.
All of them: Cisco UCS, Fujitsu PRIMERGY, Supermicro, Lenovo ThinkSystem, HPE ProLiant, Dell PowerEdge and bespoke builds, as well as blade, rack and tower servers. Because we recover the data directly from the drives, the make and role rarely matter.
You do not need to. Typically, you can remove just the drives from the server and send them marked with the order they go into the bays, which is all we require. If it’s easier, simply post the whole server; either is acceptable.
Yes. On a domain controller, an ESE database (NTDS.dit) on the volume holds the Active Directory information. Once the file system has been recovered, we can extract and, if required, repair the AD database plus SYSVOL.
Most are completed within three to seven working days. The length of time depends on how many drives there are, the RAID level, whether any database repair is needed and whether mechanical work is needed.
Pull out the drives and label each with its bay order prior to sending, or post the full server: both options are fine. Send it fully insured to our Guildford address, or drop it off on a weekday between 9am and 5:30pm.
Free diagnostic, tiered pricing from £500, and in most cases you only pay if we recover your data. Every server recovered off the hardware: Dell, HPE, Windows Server, VMware and beyond. Get your recovery moving today.