ADR / The hardware we handle / NAS at home or work

Devices · shared storage on a network

NAS drive recovery, London. One unit holds everything the firm has, and bay two is showing red.

Power it down, and leave it down. Nothing else on this page saves as many of these jobs, because a failed bay is a silent event: the shares stay up, the files still open, and a fortnight can pass before anyone glances at the front panel. What actually finishes a pool is the unit offering to scrub, repair or rebuild, and somebody agreeing. Any multi-disk set — NAS, array or server — is £500 + VAT upwards, and the look that fixes the figure is free. NAS boxes come in from across London: a Harley Street practice keeping its imaging on one unit in a basement, a Soho edit suite mid-grade, a Bromley surveyor, and a cupboard in a Barnet hallway.

On most jobs: no data, no bill A free look first, then one figure in writing Media posted in from Croydon, Ealing and Camden

Put it to an engineer — the first look costs nothing
0203 0868633

NAS symptoms, and how long each gives you.

Not listed? Try the finder →

One bay red, the pool degradedA member has gone and the rest are covering for it. Each hour under power leaves less to work with.Do not delay

A rebuild that stopped, or a volume the box has failedSectors a second member will not read. Change nothing, and give it no more power.Do not delay

Healthy disks in an enclosure that quitBetter than it sounds. Everything describing the array is on the drives, and none of it in the enclosure.Everyday job

The shares went after a firmware updateA damaged configuration, or a damaged container layer. Everything under it is untouched.Everyday job

You plug a member in and Windows offers to format itExpected behaviour: Windows reads none of these layouts. Say no, then stop.Everyday job

Packing it and posting it:pack it so that nothing can shift, then insure the parcel at the value of the files on it, not the price of the box, and send it tracked to our intake lab. The journey home is paid at this end. If you would sooner have an engineer go through the packing with you first, ring before you seal it. There is a full account of all of this on theguide to packing and posting.

Brand by brand, and what tends to go wrong.

Synology DSMDiskStation and RackStation under DSM, with btrfs or ext4 sitting on SHR.

QNAP QTSTS and TVS chassis, running QTS or else QuTS hero. Ransomware crews hunt for these before anything else.

Buffalo TeraStationTeraStation and LinkStation: an E-number appears on the little display and nothing moves after that.

Netgear, and WDReadyNAS and My Cloud, which as often as not hold every photograph a household owns.

What the NAS is telling you on screen.

Describe yours to us →

What is going on	The usual reason for it	Where that leaves you
Synology: the volume has been failed	Two members down where parity covers only one	Take the power off before DSM offers to repair it
Synology: storage pool degraded	A member has fallen out of the set and the shares carry on as though nothing happened	Every further hour under power narrows what is left
QNAP reports the volume inactive and the RAID unmounted	QTS has stopped being able to mount its own volume	It gets assembled again away from the enclosure
A DeadBolt notice in place of the web interface	A ransomware incident: every file has gained a .deadbolt suffix	Photograph the screen, change nothing at all
Buffalo TeraStation, E14: the array refuses to mount	The array itself is sound; the enclosure simply will not bring it up	Write the E-number down, then pull the power
Buffalo TeraStation, E16: it finds no disk	Either that drive has stopped answering or the enclosure has lost sight of it	What it holds is generally still there
Buffalo TeraStation, E30: a drive it calls faulty	A member has been dropped from the array	Say no to the rebuild offer

From the box arriving to your files going back.

Work we have closed →

Logged the day it lands, and the first look costs nothing Free

A case number goes on it the day the parcel lands, and an engineer settles what has truly failed before anything else happens — free of charge, and first in the order of work. Back to you come two things together: a straight note of what is liftable and what is not, plus one figure, fixed and written down. Accept it, or decline and owe us nothing.

Nothing to pay for lookingA single figure, put in writingNothing owed at this stage

Copies come first

Purpose-built imaging hardware clones each disk end to end, weak regions and all, and every step after that happens to the clones. Your unit sits here unpowered, so whatever it had begun is never picked up again.

Every copy made read-onlyNothing rebuilt in the chassis

Upward, one layer at a time

A volume you see as one thing is usually three or four layers deep. Synology matches partitions in pairs, puts an md array across each pair, and hands what results to LVM for presentation; QNAP, Netgear and Buffalo arrive at the same place along their own routes. Every layer is rebuilt in software, working up from the disks.

The mdadm and LVM stack unpickedSHR partition sets matched

The file system last of all

Once those layers stand, the btrfs or ext4 file system riding on them is repaired. Shares return under the names and the folder tree you gave them. Nothing is signed off until you have been through the file listing yourself.

btrfs or ext4 volume repairedShares back under their old names

You see the file list before you pay

What was recovered is listed for you first, and only then does a bill exist. Approve the list and it is invoiced; turn it down and it is not — and where nothing has come back, most jobs carry no charge whatever. Recovered data travels home on fresh media bought in for your job, with the postage at our end. Your case is not closed until you have read those files on a machine of your own.

No charge until you accept the figureFresh media, supplied with the jobThe post home is ours

What arrives most often

SHR is nobody's secret format — mdadm underneath, LVM over it, and beneath both a set of partitions cut so they correspond across disks of unequal size. Each layer is put back on its own, in software, from images, with your enclosure unplugged on a separate bench. In all this time that repair button has saved nobody.
Degraded is a condition with a clock on it — rebuilding a RAID 5 or an SHR pool demands one clean end-to-end pass from every disk still standing. Yours came out of one carton, went in on a single afternoon, and have sat warm against each other since. Commonly one stops half-way.
The chassis counts for the least — Buffalo, Synology and the others write the array layout onto the drives, and never into the box. Of every fault that reaches this bench, a failed power supply is the one worth hoping for.
After a ransomware incident, treat the NAS as evidence — photograph the screen before anybody reboots it. A restart often clears the notice, taking the reference number with it. Our ransomware page and the forensics section take it from there.

What the DeadBolt year taught: the QNAP units caught through 2022 had two things in common, and neither was bad luck — outside traffic could reach the box, and nobody had updated the firmware since it came out of its carton. QLocker and eCh0raix had already worked the same ground. Neither point has since changed.

One job, followed all the way through.

LDN · ADR-2026-3335JOB LOGGED ✓

Two red lights on a Westminster NAS, and a rebuild begun regardless

With two bays showing red the unit was still entirely recoverable. What did the damage was the rebuild begun afterwards, which wrote across the metadata the box holds for itself. Recovery meant a complete copy of every disk, weak sectors included, then putting the mirror back together block by block, drawing each one from whichever disk had read it cleanly. Every share opened exactly as it had before.

100%of them recovered5days here, and back to you

What helps, and what harms.

Do this much first

Shown as degraded? Take the power off now
Already pulled the disks? Number each one by bay
Left alone, it should travel complete, bays untouched
Model number, please — and is the pool SHR, or ordinary RAID?

What sets us back

Saying yes to a repair or a rebuild
Starting a scrub while the pool sits degraded
Allowing Windows to initialise a member disk
Moving disks between bays to see whether that helps

Questions answered before you commit.

The enclosure is not powering on. Are the files gone too?

Very seldom. None of the layout lives in the chassis: every disk carries its own note of which bay it occupied, the stripe size, and each layer stacked above it. Sound disks in an enclosure that has stopped make one of the more straightforward jobs on this bench.

Should I take the disks out, or post the whole unit?

Post the whole unit. The shipping form wants a NAS exactly as it stands — chassis, power supply, and the disks left in their bays. Numbering bare drives by bay is what a server or an array calls for; a NAS is the exception. Should yours already be out, write the bay number on each before packing. Cover the parcel for the files, not for the box. Nothing here is read until every disk has an image.

What separates SHR from a plain RAID set?

SHR is Synology's answer to a mix of disk sizes. Rather than leave the surplus idle, it divides every disk into partitions sized to correspond with partitions on its neighbours, lays an md array over each matched set, then passes the lot to LVM, which shows you a single volume. Recovery means naming each of those layers and standing them up again in order, beginning at the partition table.

DSM has failed the volume and stopped the rebuild. Is that the end?

No. That message reports what the enclosure can no longer do for itself; about the state of your files it is silent. Once every drive is imaged and the pool is assembled in software on other hardware, a volume DSM refuses to mount will commonly read from end to end.

Begin here if yours is doing the same thing.

RAID and server jobs → Maxtor disks → Drives on USB or Thunderbolt → Begin with the symptom →

Nothing gets worse while the power is off.

Looking at it is free. Back comes a list of what opened and what did not, together with a single price to finish, set down in writing while you are still free to say no. On most jobs an invoice only follows the data. Until that list reaches you, leave the drive switched off.

0203 0868633

Get it assessed Sending it to Guildford

Drives, cards and entire machines

Whatever it is doing now

Footage, evidence and reports

What it costs, and what follows