Devices · array sets & servers
What ends an array is almost never the first failure. It is what follows: a rebuild driving a tired survivor through one long read until it stops, an ejected member forced back into service, disks moved between bays to see what happens. A mirror offers no shelter either, since both halves have run identical hours since installation. Arrays reach us from Maidenhead by tracked post, and we work from images alone, so nothing worsens once the set arrives.
Describe the fault to an engineer
0800 6890668
| The symptom | The likely cause | Start here |
|---|---|---|
| Foreign Configuration Found (Dell PERC) | Metadata on the members contradicts the card's own copy | Touch neither option; clone first |
| 1786 — Drive Array Recovery Needed (HP) | Cover is gone; a rebuild is either waiting or half finished | Shut it down, then image it |
| 1784 — Drive Array Drive Failure (HP) | A member of the set has died | Do not guess at a replacement disk |
| 1788 — Drive Array Reports Incorrect Drive Replacement (HP) | Members have been returned to the wrong slots | Sequence decides everything; stop |
| Virtual Drive is Degraded / Offline | The volume has lost its cover, or has disappeared entirely | Power the entire set down |
| 1720 — SMART drive detects imminent failure (HP) | A working disk is telling you it has not long left | Take a copy that day |
Nothing becomes chargeable until you have said yes to it. Each item takes a case number of its own the moment it arrives, an engineer establishes what has actually failed, and you are given a candid view of which files stand a realistic chance of returning. The single quote that follows is fixed, written down, and costs nothing.
Each member is read onto dedicated imaging hardware, including those the card had already rejected. All the work that follows happens on those copies, and nothing at any stage is written back to the disks you posted.
Stripe width, member sequence and the way parity rotates are recovered from the structure of the data itself. The array is then stood up across the images on our own equipment. Your card plays no part in it, and at no point does a rebuild run.
With the set standing, the file system, the virtual-machine containers and the databases are repaired in turn, then verified against a complete listing before anything leaves.
Nothing is billed until you have seen the recovered file list in full and told us to go ahead. What comes back arrives on new media, with the return journey at our cost, and the case stays open at this end until you have opened every file yourself.
Why single parity acquired its reputation: a consumer SATA specification permits roughly one unreadable sector for every 1014 bits read, which works out at about one in each 12.5TB. Ask a large single-parity array to rebuild and every survivor has to hand over more than that without a break, which is where rebuilds come apart. The number is the manufacturers' own rather than an engineer's invention, though how sharply it bites in practice remains a matter of argument.
Production stopped when a Friday rebuild claimed a second member and locked the live data away. Each disk was imaged separately; the stripe was reassembled from the three that read cleanly plus whatever the fourth would give up, and the line was running by Monday morning.
Send the complete set. A condemned member often holds the most recent copy of some stripes, and every block is drawn from whichever image returned it most cleanly.
It is not. Stripe width, member order and parity rotation are all legible in the data's own structure — routine analysis, not trial and error.
They can. Once the set stands, the VMDK and VHDX containers and the database stores come off first, and each is opened on the bench to prove it works rather than logged as present.
Allow four to seven working days for a set of several disks. When trading has genuinely stopped the work can be run continuously; mention that when you ring, and the intake is arranged around the deadline.
Each further power-up asks more of a drive that has already begun to fail. Open the case first and let the free diagnosis tell you what is left.