Devices · storage boxes on the network
Switch it off, and leave it off. That single instruction saves more of these jobs than anything else we say, because the failure itself is quiet: shares go on mounting, files go on opening, and a bay can sit failed for a fortnight before anyone notices. What finishes a pool is the offer of a scrub, a repair or a rebuild — and somebody accepting it. Work on an array or a NAS starts at £500 + VAT, with the look that settles the figure free. Boxes reach us from a distribution office at Manton Wood, a Newark-on-Trent solicitor, a fabrication shop on Sheepbridge, and a cupboard under the stairs in Arnold, Nottingham.
Ask an engineer, and the first look is free
0115 8220606
| What is happening | What usually causes it | What that means for you |
|---|---|---|
| Synology says Volume Crashed | More disks down than the parity can cover | Power it off before DSM offers to mend anything |
| Synology says Storage Pool Degraded | One disk has dropped out while everything above it carries on as normal | Each hour it stays powered costs you something |
| QNAP says system volume not active, RAID unmounted | QTS can no longer mount the volume it built | Reassemble it somewhere other than the box |
| A DeadBolt notice — WARNING: Your files have been locked | Ransomware, and every file now carries a .deadbolt ending | Photograph what is on screen and alter nothing |
| Buffalo TeraStation E14: the array will not mount | Nothing wrong with the array — the box just will not bring it up | Note the E-number down, then unplug |
| Buffalo TeraStation E16: no hard disk found | That disk has stopped, or else the box cannot see it any more | Whatever it holds is usually still on it |
| Buffalo TeraStation E30: a disk the box says may be broken | One disk has been put out of the set | Say no to the rebuild |
Your device picks up a case number the day it arrives, and an engineer then establishes what has genuinely gone wrong — at no cost, and ahead of everything else. Two things come back to you together: a plain account of what can be lifted and what cannot, and one fixed figure in writing. Agree to that figure, or turn it down and pay nothing.
Imaging hardware built for the purpose clones every disk from one end to the other, weak regions included. From then on the clones take all the work. Your box sits here unpowered, so whatever it was doing is never resumed.
A single volume is usually several layers deep. Synology pairs partitions off, lays an md array over each pair and lets LVM present what results; QNAP, Netgear and Buffalo arrange the same idea their own way. Working upward from the disks, each layer is put back in software.
Once those layers stand again, the btrfs or ext4 volume above them gets repaired. Your shares reappear with the names and the folder tree you left them with. No job here is signed off before you have read the file listing.
The file list reaches you before any invoice does. Say yes and it is billed; say no and it is not — and on most jobs, if nothing comes back there is nothing to pay. Whatever comes off goes home on media bought in for your case, carriage at our end. Nothing here is closed until you have opened those files on your own machine.
What 2022 taught: around 3,700 QNAP boxes were answering from the public internet when DeadBolt's first wave went round that year, and more waves came after it. QLocker and eCh0raix had been over the same ground before it. Two things were true of every one of them: outside traffic could reach the box, and the firmware had not been touched since the carton was opened.
At two red lights the unit was still perfectly recoverable. The damage came from the rebuild started afterwards, which wrote over the metadata the box keeps for itself. Recovery meant a full copy of each disk, weak sectors and all, then reassembling the mirror block by block, taking every block from whichever disk still read it cleanly. All the shares opened exactly as before.
Hardly ever. The chassis stores none of the layout: each disk keeps its own record of the bay it sat in, what the stripe size was, and every layer built on top of it. Good disks inside a dead enclosure are among the easiest jobs that reach this bench.
Send the whole box. A NAS is wanted on the shipping form just as it is: chassis, power supply, and the disks left where they sit. Bare drives, each numbered by its bay, are how a server or a RAID set travels; a NAS does not. If yours are already out, put the bay number on each before you pack. Insure the parcel. No file is looked at here until every disk has an image.
SHR is how Synology copes with disks of different sizes. Instead of wasting the surplus, it cuts every disk into partitions that match partitions on the others, builds an md array across each of those matched sets, then hands the lot to LVM, which presents the single volume you see. Recovery means identifying every one of those layers and putting them back in order, starting at the partitions.
No. Volume Crashed describes something the box has stopped being able to do; about your data it says nothing whatever. With every drive imaged and the pool assembled in software on separate hardware, a volume DSM will not mount usually reads straight through.
The look costs nothing. What comes back to you is a file list — what opened, what did not — plus one figure to finish the job, in writing, before anything chargeable begins. Most jobs carry no fee at all unless the data comes back. Leave the drive switched off until you have that list.