Devices · RAID arrays and servers
Arrays rarely die of the first fault. They die of what gets done next — a rebuild dragging an already tired disk through one unbroken read until it quits, a rejected member shoved back into the set, disks shuffled between bays to see if that helps. Mirroring buys you less than people think, because both halves went in on the same day and have spun the same hours since. Everything here is done on images, so nothing gets any worse once the disks reach us from Birmingham.
Tell an engineer what it is doing
0800 6890668
| What you see | What is behind it | What to do |
|---|---|---|
| Foreign Configuration Found (Dell PERC) | What the disks say disagrees with what the card remembers | Choose neither; image the set first |
| 1786 — Drive Array Recovery Needed (HP) | Redundancy has gone, and a rebuild is pending or part done | Power it off and image it |
| 1784 — Drive Array Drive Failure (HP) | One disk in the set has gone | Do not drop in a spare and hope |
| 1788 — Drives Improperly Attached (HP) | Somebody has reseated the disks out of order | Order is everything here; stop now |
| Virtual Drive is Degraded / Offline | The volume has lost redundancy, or has vanished outright | Shut the whole lot down |
| 1720 — SMART drive detects imminent failure (HP) | A disk still in service is announcing its own failure | Copy it the same day |
Every item gets a case number of its own the day it lands. An engineer then works out what has actually failed and tells you plainly which files have a real chance of coming back and which do not. Only then does a price follow: one figure, fixed, in writing, and it costs you nothing to see it. Nothing is charged until you say yes.
Every disk goes onto purpose-built imaging kit — the ones the controller had already condemned as well. From then on the job runs entirely on those copies, and not one byte is ever written back to the disks that arrived from you.
Stripe size, disk order and the rotation of parity all come out of the data's own layout. The set is then assembled over the images on our equipment. Your controller takes no part in any of it, and nothing at any point is asked to rebuild.
Once the array stands, the file system is repaired, then the virtual-machine containers, then the databases, and the whole lot is checked against a full listing before it goes anywhere.
No invoice is raised until the full list of recovered files has sat in front of you and you have said go ahead. Your data comes back on media we buy new, posted at our expense, and the job is not closed here until every file has opened on your own machine.
Where single parity got its bad name: the datasheet for a consumer SATA disk allows about one unrecoverable sector in every 1014 bits read — roughly one per 12.5TB. Put a large single-parity array through a rebuild and the survivors between them have to hand over a good deal more than that without a stumble, which is the point at which rebuilds fall over. The figure comes from the drive makers themselves rather than from us, though engineers still argue over how hard it bites in the real world.
The rebuild took a second member with it on the Friday, and the live production data went out of reach. Nothing was powered up again. All four disks were copied to our own storage, the stripe worked out from three clean images and what the fourth still held, and the plant started on time Monday.
Yes — send every disk. The one the card threw out often carries the newest version of certain stripes, and each block is taken from whichever image gave it up most cleanly.
You have not. Stripe size, disk sequence and the way parity moves can all be read out of the data itself. That is analysis we do daily, not guesswork.
Both. With the array standing, the VMDK and VHDX files and the database stores are pulled off ahead of everything else, then each one is mounted here and opened, not simply ticked off a list.
Four to seven working days is normal for a set of several disks. If trading has genuinely stopped, we will run the case without a break — tell us that on the phone and the intake is booked to suit your deadline.
Every extra power-up takes something off a drive that is already going. Leave it switched off and let the free diagnosis say what still reads.