HP ProLiant Server Down? Don't Force a Rebuild — Image First.
UK HP ProLiant data recovery specialists since 2002. Smart Array P-series, MR-series, all RAID levels, Gen8 through Gen11. Emergency same-day collection across the UK. No-fix-no-fee.
What this means and what to do next
HP ProLiant servers — DL, ML and Apollo families — are the workhorses of UK SME and mid-market IT. They run RAID through the Smart Array controller family (P-series for SAS/SATA, MR-series for the newer Gen10 Plus and Gen11 platforms). Both controller families write proprietary metadata to each disk in the array, which means a ProLiant RAID is not portable to other hardware in the way that a software RAID is, and a failed array cannot simply be rebuilt by re-inserting the disks into another controller.
We see ProLiant recoveries in three main shapes: a single physical disk has failed in a degraded array and the rebuild aborted (the most common case), the Smart Array controller itself has failed and is reporting the array configuration as not present, or multiple disks have dropped offline in quick succession — sometimes from a shared backplane fault, sometimes from a power event. In every case the priority is the same: image the disks before any further attempt to bring the array back online.
Smart Array controllers also use a write-back cache backed by FBWC (Flash-Backed Write Cache) or BBWC. If a power loss occurs and the cache battery or supercap has failed, the cache contents are lost — and any writes that were in flight to the array become inconsistent. Resyncing such an array can corrupt filesystem metadata on disks that are otherwise healthy. The lab needs to know the cache state before any rebuild is attempted.
The four most common HP ProLiant data recovery scenarios
1. Smart Array rebuild aborted during recovery. Symptom: one disk failed in a RAID 5 or RAID 6, a replacement was inserted, the rebuild started — then aborted partway through. The array is now offline or in a non-recoverable state. Cause: a second disk has latent unreadable sectors that the controller hit during the rebuild read. Recovery: image all original disks (including the partially-rebuilt replacement) using PC-3000 with cleanroom imaging where required, then reconstruct the array virtually using the original parity. The replacement disk is treated as a parity gap, not as a source. This is the single most common reason a ProLiant array becomes 'unrecoverable in the field'.
2. Failed Smart Array controller. Symptom: server boots but reports 'No Logical Drives Present' or 'Controller Failure' at POST. The disks are healthy; the controller has died. Recovery: in many cases the array can be brought back online by replacing the controller with the same model and firmware revision, then importing the disks. We carry P-series donor controllers for this purpose and can quote remote support if the customer's IT team prefers to do the swap. If the original controller cannot be matched (older Gen6/Gen7 hardware), we image the disks and reconstruct the array in software.
3. Multi-disk failure from backplane or power event. Symptom: two or more disks marked failed simultaneously, or short bursts of disks dropping in and out of the array. The array is offline. Cause: a backplane fault, a SAS expander failure, or a power supply event has electrically disturbed the disks rather than damaged them mechanically. Recovery: image each disk independently from a known-good imager (not from the original ProLiant chassis), then reconstruct the array. Most disks in this scenario image fine — the controller's view of them as 'failed' was a transient electrical issue, not a media failure.
4. FBWC/BBWC cache loss leading to filesystem inconsistency. Symptom: array comes back online after a power event but the filesystem reports as RAW, corrupt, or partially mountable. Some files are present, some are zero-length, directory listings are inconsistent. Cause: the Smart Array's write cache lost its contents (battery or supercap failed during power loss), and writes that had been acknowledged to the OS were never flushed to disk. Recovery: filesystem-level reconstruction on top of an image of the array, with NTFS/ReFS/ext4/VMFS journal replay and orphan-file recovery as required. We will be specific about what is recoverable on the call.
How Data Clinic recovers an HP ProLiant server
Step one is a triage call with your IT team. We will ask which Smart Array controller is fitted (model number and firmware revision), the RAID level, the number of disks, which disks are marked failed, what attempts have been made so far (rebuild, controller reset, configuration import), and whether the cache battery/supercap was reporting healthy at the last boot. This call avoids common mistakes — particularly the 'just rebuild it again' instinct — and identifies whether on-site work is needed.
If the recovery can be done remotely, we ship a hardware imager to your site, image all disks in parallel without the original controller in the loop, and either ship the images to our Bury lab or remote into the imager to begin reconstruction. For larger arrays or where the disks themselves have failed, we collect the server (or the disks, depending on whether the chassis is needed) and complete the work in-lab. Server-level collections across the UK are same-day for emergency cases.
Reconstruction is done with the array's original metadata where readable, or by parameter sweeping (stripe size, parity rotation, disk order) where the controller's configuration has been lost. Once the virtual array is rebuilt, the contained filesystems — usually NTFS, ReFS, VMFS, or LVM/ext4 — are repaired and the data is extracted to a fresh enterprise-grade external drive or NAS supplied to you. More on our enterprise server recovery service →.
Get a free initial diagnosis in 60 seconds
In the tool below, choose RAID/NAS/Server and the symptom that best matches your ProLiant — degraded array, failed logical drive, controller error, or unbootable. The tool routes server cases straight to our enterprise recovery team.
What our customers say
"Three years of family photos on a drive that suddenly failed. Data Clinic collected next day, kept me updated through the cleanroom work, and got everything back. Worth every penny."
"Honest, fixed-price, no-fix-no-fee. Quoted by another lab at three times the price. Recovered 100% of my files."
"Reasonable cost, clear communication, and they were straight with me about what was recoverable and what wasn't. Recommended."
Frequently asked questions
How long does HP ProLiant recovery take?
For a typical degraded-array case where the disks themselves are not mechanically damaged: 2–4 working days from receipt. For multi-disk failures with cleanroom work needed: 5–10 working days. Emergency 24-hour and 48-hour services are available for business-critical cases — call 0800 151 2207 to discuss. The first triage call is always free and gives you a realistic expectation before any disks are moved.
Can you recover ProLiant arrays without the original Smart Array controller?
Yes. Once the disks are imaged, the array can be reconstructed entirely in software using the Smart Array metadata that the controller wrote to each disk. We do not need the physical controller to recover the data — though if the controller is available and matched, it sometimes provides a faster path. For older hardware where matching controllers are no longer obtainable, software reconstruction is the standard route.
My Smart Array is showing 'Interim Recovery' — what does that mean and what should I do?
Interim Recovery means the array is operating but is no longer redundant — one disk has failed and the array is running on the remaining disks plus parity calculation. The data is accessible, but a second disk failure will take the array offline. The right next step is: (1) back up everything critical immediately, (2) call your HP support contact or call us before inserting a replacement disk. The rebuild from Interim Recovery to fully redundant is the riskiest moment in an array's life — many recoveries we see start with a rebuild aborting during this step.
Do you handle Smart Array configurations on Gen11 servers and the MR-series controllers?
Yes. The MR-series controllers (e.g. MR416i-p, MR216i-a) used in Gen10 Plus and Gen11 servers store array metadata in a different format from the older P-series, but the recovery principles are identical: image the disks, reconstruct from metadata, mount the filesystem. We have recovered from every current generation of ProLiant hardware including the most recent Apollo and DL/ML Gen11 platforms.
Is the recovery suitable for ICO or compliance reporting?
Yes. For every enterprise recovery we provide a written report covering the failure mode, the actions taken, the integrity of the recovered data, and the chain-of-custody for the disks while they were in our possession. Many of our enterprise customers use this report as part of their ICO data breach assessment or internal compliance documentation.
What does an HP ProLiant recovery cost?
Cost depends on the number of disks, the failure mode, and whether cleanroom work is required. Typical mid-market ProLiant recoveries (4–8 disks, 1–2 mechanically failed) fall between £1,200 and £4,500 including VAT. Larger arrays and emergency turnarounds are quoted individually. We give a fixed-price quote after free diagnosis. No fee if we cannot recover your data.