Call us — 01483 901310
Mon–Fri · 9am–5:30pm · No fix, no fee
Start a free diagnostic →

Data Recovery Case File · NAS & Network Storage · Ask About the Disks

The Appliance Is a Computer, and the Data Is Not Inside It

His enquiry ends with a question that quietly points in the wrong direction. A network unit that will no longer boot: "the power supply seems fine, tested with a spare. When plugging the power cable in, it immediately spins up the fans and tries to start. I can hear noises from inside the unit too, but no power or drive indicators light up. Interested to know if you have experience fixing these uni"ts. Whether the unit can be fixed is a separate matter from whether his files can be read — and only one of those two questions needs answering first.

MediaMulti-bay network storage appliance — fans starting on power application with no indicator activity and no boot; member disks retained
Reported situationAppliance ceasing to boot · replacement power supply tested without change · fans starting immediately on power application · audible activity from within the unit · no power or drive indicators illuminating · contents required
Fault classAppliance mainboard failing to complete initialisation — member disks unaffected and readable independently of the unit
Equipment usedDisk order recorded before removal · all members removed and imaged write-blocked individually · array configuration determined from on-disk metadata · volume assembled offline from images · appliance repair treated as a separate and optional question

The decode: what the symptom says, and why it barely matters

What fans without indicators means: power reaches the unit and the fan circuit runs, which on most designs happens as soon as voltage arrives and does not wait for anything. The indicators, by contrast, are driven by the mainboard once it has started its own firmware. So fans spinning with no indicators is a board that is receiving power and not completing its initialisation — which, combined with a known-good power supply, places the fault on the mainboard itself.

Why that is a real fault and a small problem: because a network appliance is a small computer whose job is to read disks and serve them over a network. The data is not in the appliance. It is on the disks inside it, in a filesystem, exactly as it was — and those disks come out and are read directly, with the unit's condition entirely irrelevant.

So the question worth asking is not whether the unit can be repaired. It is whether the disks can be read, and the answer to that does not depend on the appliance at all. Repairing the unit is a legitimate thing to want afterwards — it is a functioning device with a fault — but it is a separate purchase, and it is not the route to the files.

Why this pattern keeps appearing: the object that failed is the unit, so people ask about the unit. Nothing in the situation prompts anybody to say the thing you want is behind four screws and does not care whether this ever powers on again. It is the same shape as a dead laptop whose drive comes out in five minutes.

What must be done carefully during removal: the disk order is recorded before anything comes out. Multi-bay units write configuration metadata to their members, and while that metadata generally identifies the arrangement independently, physical order is worth having as a cross-check rather than a reconstruction problem later.

Why the array is assembled offline rather than in a replacement unit: putting disks into a different appliance risks it initialising them as new members, and some units do that on insertion without asking clearly. Imaging every member first and assembling from the copies removes that risk entirely, and allows the assembly to be attempted repeatedly at no cost.

What not to do meanwhile: not repeatedly power-cycle the unit hoping it comes up, and not insert the disks into anything else to test them.

On the bench

Disk order was recorded before removal as a cross-check against on-disk configuration metadata. All members were removed and imaged write-blocked individually — a network appliance being a small computer whose fault has no bearing on the disks it reads, so fans running without indicators localises the failure to the unit's mainboard and not to storage. Array configuration was determined from on-disk metadata, the volume assembled offline from the images, and appliance repair treated as a separate and optional question.

The outcome

The disks removed and imaged individually, the configuration read from their own metadata and the volume assembled offline. Free assessment, one fixed written figure including VAT; where a drive has to be opened, 50% of parts and labour is payable upfront with the balance only on success — otherwise no recovery, no fee. The decode: fans start as soon as voltage arrives, while indicators wait for the mainboard's firmware — so fans without lights is a board not completing initialisation. And that matters very little, because the appliance is a computer and your files are on the disks inside it.

Network unit that powers fans and never comes up

Ask whether your disks can be read rather than whether the unit can be fixed — those are different questions and only one of them stands between you and your files. A network appliance is a small computer whose job is to read disks and serve them over a network; the data is on the disks, in a filesystem, and they come out and read directly with the unit's condition irrelevant. Fans spinning with no indicators tells you the fault is on the appliance's mainboard, since fans usually run as soon as voltage arrives while indicators wait for firmware to start. Note the disk order before removing anything, don't put them into another unit to test them, and stop power-cycling it.

Appliance dead with your data inside it?
The disks are the question — call Guildford Data Recovery on 01483 901310; order recorded before removal, every member imaged individually, configuration read from on-disk metadata and assembled offline.
Request a quote online →

Our case files are drawn from genuine enquiries received by our laboratory over the past ten years, anonymised to protect client confidentiality. Each one describes the diagnostic and recovery procedure our engineers apply to that fault, using the equipment listed.