Do you recognise one of these symptoms?
- The controller reports the array as 'failed', 'offline' or 'foreign'
- Two disks have failed in a RAID 5, or three in a RAID 6
- The rebuild stopped or ended with errors
- The controller was replaced and the array is no longer recognised
- The volume mounts but files are corrupted
Recovery by RAID level
Job sheetRAID
- Device
- RAID
- Price
- From €350, fixed quote after diagnosis
- Environment
- ISO 5 (class 100) cleanroom if the device must be opened
- Method
- Full image of the device, reconstruction on the copy, never on the original
- Turnaround
- Economy 10–15 working days · Standard 5–10 working days · Emergency Diagnosis in 24–48 hours
- Check
- List of recovered files sent before payment
- Confidentiality
- Data never passed to third parties, copies wiped after delivery
The faults we handle
- Disk failures
- Disks in an array are usually the same age and model, and they tend to fail close together. The rebuild, which reads every sector, is often the moment a second disk gives up.
- Controller failure
- A dead RAID card, a flat cache battery, a firmware update or a reset can erase the array's configuration.
- Configuration errors
- An array recreated or initialised by mistake, disks swapped, or a hot spare added to the wrong set.
- Silent corruption
- Unreadable sectors on a disk that has not yet been ejected from the array: the rebuild then writes inconsistent parity.
What not to do
- Do not force a disk back online if you are not sure it holds current data.
- Do not recreate or initialise the array in the controller.
- Do not run CHKDSK or fsck on a degraded volume.
- Label the disks with their slot numbers before removing them.
Virtual rebuild from clones
We never rebuild a damaged array 'in place'. Every disk is first imaged sector by sector, then the RAID parameters are worked out from the data itself: disk order, stripe size, parity direction and delay, start offset. The array is then assembled virtually and the file system is checked before the files are extracted.
We work with hardware controllers (Dell PERC, HPE Smart Array, LSI/Broadcom MegaRAID, Adaptec), with software RAID (Linux mdadm, Windows Storage Spaces and dynamic disks, ZFS, Btrfs) and with every NAS.
When to call us
As soon as an array is degraded and you do not have a recent backup, or if a disk shows errors during a rebuild. Stopping at the right moment is often what makes the difference between a full and a partial recovery.
RAID levels compared
| Level | Min. disks | Usable capacity | Fault tolerance | Performance | Typical use |
|---|---|---|---|---|---|
| RAID 0 | 2 | 100% | None | Excellent | Video editing, cache |
| RAID 1 | 2 | 50% | 1 disk | Good reads | Workstations, small NAS |
| RAID 5 | 3 | (n − 1) / n | 1 disk | Very good reads, average writes | NAS, file servers |
| RAID 6 | 4 | (n − 2) / n | 2 disks | Very good reads, slower writes | Large arrays, archives |
| RAID 10 | 4 | 50% | 1 per mirror | Excellent | Databases, virtual machines |
Price of RAID data recovery
From €350, fixed quote after diagnosis. The diagnosis is free and the quote is fixed: it does not change once accepted. If no data is recovered, you pay nothing. See the full price list.