Data Recovery Case File · NAS & Network Storage · Linux Filesystem, Long Dormancy
Extracted, Shelved, and Still Readable
This enquiry described something done years ago and then left. A 2TB drive that was inside a single-bay network appliance: "a few years ago it stopped working. I managed to take the drive out of the" unit, and it has sat since. Two things are worth knowing before it goes anywhere near a computer. The disk is almost certainly fine and its contents almost certainly intact — but it carries a filesystem that ordinary desktop machines cannot read, which means the first thing any Windows computer will do is offer to wipe it. And after several years on a shelf, the way it is first powered matters more than it would have done at the time.
| Media | 2TB hard drive removed from a single-bay network storage appliance — appliance failed several years previously; drive stored unpowered since extraction |
| Reported situation | Network appliance ceased functioning some years ago · drive removed from the unit by the owner · drive stored since without being read · contents required |
| Fault class | No fault attributable to the drive — appliance failure with storage retained; Linux filesystem requiring native interpretation and dormancy requiring a controlled first spin-up |
| Equipment used | Condition assessed before the first spin-up after dormancy · read write-blocked on native equipment · Linux filesystem interpreted directly · contents delivered on cross-platform media |
The decode: why it is fine, and the two things that could change that
Why the drive is probably healthy: on consumer network appliances the failure is very often the external power supply or the appliance's own board rather than the disk inside it. Somebody who removed the drive because the unit died is usually holding a perfectly good disk that simply had nothing to plug into. Nothing about the appliance failing touches what is written on the platters.
Why an ordinary computer cannot read it: network appliances run Linux and format their disks with Linux filesystems, frequently with the volume managed on top of that. Windows has no idea what any of that is, and Macs generally do not either. So connecting the disk produces not an error but something more dangerous: a disk that appears unformatted. Windows will then offer to initialise it, framed as the step required to use the disk — and accepting that writes a fresh partition table over the structures holding everything. It is the single commonest way an intact NAS disk is destroyed after being carefully saved, and it happens at the moment somebody is trying to be helpful.
What to do instead: read it on something that understands the filesystem, or on a bench where it is interpreted directly. If anyone is helping, the instruction is simple: connect it, decline every prompt, and if the only offer on screen is to initialise or format, disconnect it.
The second thing, which the years have introduced: a drive that has been unpowered for several years faces its most demanding moment at the next spin-up. Bearings stiffen and heads can adhere to platter surfaces during long storage, and the drive is not guaranteed a second attempt. So the first power-up should be spent on imaging rather than on a test — condition assessed first, then a single controlled session, rather than plugging it in casually to see whether it still works and discovering the answer the expensive way.
What removing it did not cost: worth reassuring, because people worry retrospectively. Taking a drive out of a network appliance is not like opening a sealed drive — the disk is a standard component in a plastic housing, and removing it is harmless provided it was handled reasonably and kept dry and static-free. Keeping the drive when the appliance was discarded was the right decision, and it is why this is recoverable at all.
The configuration note: single-bay units hold one disk and no redundancy, so everything is on that one platter stack. If the unit had more than one bay, every disk matters and they should be kept together and labelled by position.
On the bench
Condition was assessed before the first spin-up, since several years unpowered makes that attempt the drive's best and possibly only one, and it belongs to the imaging rather than to a test. The drive was read write-blocked on native equipment and its Linux filesystem interpreted directly rather than through any host that would have offered to initialise it — the prompt that destroys intact NAS disks. Contents were verified and delivered on cross-platform media the owner could read without any specialist software.
The outcome
The drive read on native equipment in a single controlled session, the filesystem interpreted directly and the contents delivered cross-platform. 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, for anyone holding a disk from a dead network appliance: keeping it was right, and the disk is probably healthy, because these units usually fail at the power supply or their own board rather than the drive; but it carries a Linux filesystem, so a Windows machine sees an unformatted disk and offers to initialise it, which writes over everything — decline that. And after years on a shelf, spend the first spin-up on imaging rather than on a test.
Disk taken out of a dead network drive
Don't let any computer initialise it. Network appliances run Linux and format their disks accordingly, so a Windows machine can't read the filesystem and shows you an unformatted disk — then offers to initialise it as the step needed to make it usable. Accepting that writes a new partition table over everything you were trying to save, and it's the commonest way an intact NAS disk gets destroyed after being carefully kept. If you connect it anywhere, decline every prompt, and disconnect if the only option offered is to initialise or format. Take encouragement otherwise: these units usually fail at the power supply or their own board, so the disk inside is generally fine. If it's been on a shelf for years, treat the first power-up as the one that counts — bearings stiffen and heads stick during long storage.
Don't let anything initialise it — call Newcastle Data Recovery on 0191 406 1051; condition assessed before the first spin-up, Linux filesystem read directly, contents delivered cross-platform.
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.