We are booking in new cases · Weekdays, 9am–5:30pm Urgent case? Call 0800 6890668
MHDR Maidenhead Data Recovery 0800 6890668 Send it in
MHDR / Specialist cases / Server system recovery

Server systems · array recovery

Server data recovery, Maidenhead. Both answers on that prompt can do damage.

A server's failure is usually not due to a problem with its storage. Usually, the issue lies within the components surrounding the disks, i.e., an inconsistent RAID card that disagrees with metadata created upon writing or a boot time prompt where you have two options (both of which can be very detrimental), or the write cache was emptied during a loss of power before it made it onto the platter. All servers come in to us from Maidenhead as bare disks; all steps we take are done on disk images; we do not require the actual RAID card, and what we send back is your volumes, database(s) and virtual machines.

Only pay if we recover your data, on most jobs Diagnosis free, then a fixed quote Get a quote, then post it in from Windsor, Slough or anywhere else

Describe the fault to an engineer
0800 6890668

Server symptoms, and what each one means.

None of these? Run the triage →
The symptomThe likely causeStart here
The PERC reports Foreign Configuration DetectedThe records of the disks and RAID card are no longer in sync and therefore the server stopped on purpose; no data has been lostDo not respond to either prompt presented to you after the reboot
1786 — Drive Array Recovery Needed (HP)A rebuild is waiting to be initiated, or has failed prior to completionPower off immediately; image all data before attempting anything else
Amber light on drives after a power outageWrite operations were interrupted and the cache contains unknown information although the disk mapping is still intact on the physical disksForce nothing back online
After rebooting your server, your data volume is missingController state or file system is damaged; all data underneath them is safeUsually recoverable
The controller has failedThe order, stripe size and parity rotation are all stored on the membersThere is no matching card required
Files are back yet guest or database will not openBeing recovered and being consistent are different states; only the second state is usefulSee the database page
Where to send it: first get a quote, by phone or with the form, then send it using a tracked, insured service to the Guildford intake laboratory, or drop it off there yourself (Monday to Friday, 9am to 5:30pm, no appointment needed). We will cover the cost of the postage back to you. From a laptop, PC or Mac, we just require the drive out of your computer; storage soldered to the board (T2 and Apple silicon Macs, soldered laptop flash) is not taken on, and if you're unsure as to type, call us first. From a server or RAID set, just send the individual drives, all clearly labelled by bay. From a NAS, send the disks, each labelled by bay, if they come out; otherwise, send the entire NAS unit. A single external drive comes whole, with its cable and power supply. If you wish to review your packaging before posting, contact an engineer who will walk you through the process over the phone. All steps involved with the postage process are listed on the postage page.

Every server array describes itself.

Dell PERC & MegaRAIDBoth write their DDF metadata (an open and published format) to the last sectors of each member. It is this record that enables a set to be reassembled from images without any need for a card at all.
The HP Smart ArrayHP used its own RIS structures at the beginning of every disk, and its parity is offset by one stripe from the conventional position. Send these structures through a stock preset and what comes out means nothing.
That Import-or-Clear promptEither can do harm if either choice is made at the wrong time. Import can lay stale metadata over live stripes when done in error, and Clear irreversibly scrubs off each member's map. Get an image of them all beforehand and the prompt loses all its bite.
Data left in the cacheThe controller card treats a write as done once it has been written into battery-backed memory, not when it reaches the disk platters. So if the power is cut part-way through a flush, the damage hides inside the file system, and that's why losing power is enough on its own to require an image.

What happens between arrival and return.

01

Booked in, then assessed at no cost Free

Nothing is chargeable until you say yes. Every item gets a case number of its own immediately upon receipt, an engineer determines what has failed, and you get a clear picture of which files have a real possibility of being recovered. The single quote that follows is fixed, written down and does not cost anything.

Diagnosis at no costOne fixed quote, in writingNothing signed
02

Freeze it exactly as it stands

Shut off the computer, record the bay number on each disk as they are removed, and do not respond to anything displayed on the screen. Afterwards, the images of each individual member are created using special equipment designed specifically for this purpose, including any the card had rejected, before anything is read as data.

Every member imaged firstEven the failed members
03

Reassemble it with no controller present

These images include the DDF (Disk Data Format) or RIS records, and provide information regarding RAID levels, stripe width, the sequence of disk placement, and if applicable, the orientation of parity on RAID 5 and RAID 6 arrays. Once these parameters are determined, the array will be reconstructed in software, and if the card has failed, it ceases to have an effect.

Geometry taken from membersOriginal controller not required
04

Prove it works, not just that it copied

The original file system is repaired on the reconstructed copy (for example: NTFS, ReFS; VMFS datastore; XFS or ext4). Next comes the ultimate test of recovery: guest machines boot successfully; databases attach correctly; shared folders open correctly.

Repairs made on images onlySeen booting and mounting
05

Opened, verified, and sent home

You don't get charged anything until you've confirmed and authorised that everything is fine after seeing the entire list of what was recovered. Any recovered items are returned via brand-new storage media and the postage costs for these are covered by us. We do not consider a project complete here until we receive confirmation that you have been able to successfully access and review every item you received back.

You approve the list firstReturned on fresh mediaThe return post is ours

What tends to show up first

  • Depending upon the manufacturer's design, either HP's RIS is located at the beginning of a disk member; or, the DDF utilised by PERC and MegaRAID is located at the end. Identification of how a disk was arranged enables a successful rebuild.
  • Even though a card may reject a disk, a member that has been dropped from an array several weeks prior may retain a more current copy of certain areas. For this reason, every sector is examined against all available images.
  • HP's non-standard parity configuration causes common preset tools to assemble an incorrect volume when attempting to rebuild the array, because the parity sits a stripe along from where a generic tool looks for it. In addition to identifying how disks are configured to operate as part of an array, HP's proprietary method requires that the reconstruction process adhere to its specifications.
  • Why we ask about the cache battery so soon: The condition in which your file system will be found is decided by whether there were any write backs stored in memory still to be written to the disks, and that is settled before we make any promises about how it will be restored.

The actual purpose of the screen options: "Import" replaces the card's working config with whatever the foreign metadata claims, and "Clear" deletes the mapping from all disks, permanently. One of these keys does one of those actions and no matter which you choose neither action can be reversed after the fact (though once an image exists, neither can take any data from you). Cloning first is not a precaution: that is the method.

From the casebook, lately.

MH · MHD-2026-6816ON RECORD ✓

A ProLiant, a dead UPS, and prompts nobody dared answer

A UPS that had run out of charge before the mains supply came back on caused the Smart Array controller to begin asking a number of very difficult questions none of the people in the practice were willing to answer, and rightly so. Each of the members was cloned first, the RIS metadata read off and the delayed parity accounted for, and thus the rebuild could proceed on the images alone, without any access to the physical media. When the practice management database was brought online again, it did so intact, current up to its last commit in the cache.

Database came up clean5 days from arrival

Before it comes to the lab.

Do these now

  • Take a picture of every screen on each controller prior to turning them off
  • Label each disk with its drive bay as you remove it
  • Get a quote before posting all of your drives (dead or alive)
  • What did your server run? SQL, Exchange, guest OS, shares?

Things best not done

  • Choosing either Import or Clear at that prompt
  • Inserting a replacement disk to initiate a rebuild
  • Upgrading the firmware of the controller at this time
  • Allowing the server to power up prior to imaging

Questions we are asked about server recovery.

Our server will not boot. Is the data gone?

Rarely. Normally what's damaged is the controller's state, the disk array metadata, or a file system that was damaged, but all of the disks themselves are still intact. Shut it down, answer none of its questions, and things stay in your favour.

'Foreign Configuration Detected' — what is it telling us?

The metadata written on the disks is conflicting with the card's own records. This will happen most often due to a board swap, putting a drive into the wrong bay, or a failing controller. The controller is being deliberate in halting. Each option on this screen can cause you to lose data; clone the disks first, and neither option can.

Do you need the original RAID card to reassemble it?

No. Each member includes its own configuration within itself (DDF metadata, an open standard used by both PERC & MegaRAID, and RIS blocks, a proprietary format devised by HP). Because reassembling takes place using bench images, there doesn't need to be an exact duplicate of any card available.

Do virtual machines and databases come back, or only files?

Both, and it is also important to note the difference. Extracting a VMDK or an MDF from the array is one process; getting the hypervisor to recognise and mount it, or the database engine to attach to it is another. On our database page we cover the latter process for databases.

More of the jobs we've closed.

Recovering RAID 5 → Striped RAID 0 → SQL & mailboxes → Media we work with →

Switched off, it keeps every chance it has.

Every additional power-up asks for even more of a drive that is already starting to fail. First open a case with us, and let the free diagnosis tell you how much remains.

0800 6890668

Ask for a free diagnosis Posting it to us