Data Recovery Case File · NAS & Network Storage · Suspect the Bay

A New Drive Showing as Failed Is Rarely a Faulty Drive

His enquiry describes a repair that made things twice as bad, in a sequence worth reading carefully. A four-bay unit with four 6TB disks: "having had a failure on disk 1, I replaced the unit with a new duplicate hard drive from the manufacturer. After connecting the drive, this then showed it had failed, which in turn made disk 2 fail. Now I am unable to access any data." A brand-new drive reporting failure within moments of being fitted is almost never a faulty drive — and what happened to disk 2 immediately afterwards is the predictable consequence rather than a second piece of bad luck.

The decode: why the new drive failed, and what happened to disk 2

Why a brand-new drive reporting failure immediately is suspicious: new drives do fail, but early failures usually appear over days of use rather than within seconds of being inserted. A drive rejected on connection points at something the drive is being connected to. The candidates are specific: a fault in that bay — worn connectors, a damaged backplane, or a failed channel on the controller; or the unit refusing the drive, because it reports a different sector size, a different firmware family, or a capacity the array does not expect.

Why that matters enormously here: if the bay is at fault, then the original disk 1 may never have been the problem either. A failing bay reports whatever is in it as failed — so the "failure" that started all of this may have been the bay all along, with a perfectly good disk 1 removed on its evidence. That single possibility changes what the surviving set looks like, and it is why every disk including the original and the replacement must be preserved and assessed rather than assumed.

What happened to disk 2, and why it was predictable: when the replacement was fitted, the unit began a rebuild — reading every sector of every remaining member, continuously, for many hours on 6TB disks. That is the heaviest load the array ever takes, applied at the moment it has no redundancy left. Disk 2 failed under it. A second failure during a rebuild is the single most likely moment for one to occur, particularly on contemporaneous disks bought and run together, and it is not an unlucky coincidence.

What must not happen now: no further rebuild, and no more reinsertion into that bay. Repeating the operation that lost disk 2 across the remaining members is the way to lose the set entirely.

What preserves the position: removing every disk — all four originals plus the replacement — labelling them by bay, and imaging each individually. That captures each while it still cooperates and allows the failure sequence to be established from the metadata each member carries, so that a stale disk is not reintroduced into a reconstruction and silently mixed with current data.

What to check on the unit itself: whether a known-good drive is recognised in that bay, tested separately from any recovery. If it is not, the appliance needs attention before any rebuilt array goes back into it.

On the bench

No further rebuild or bay reinsertion was attempted, since repeating the operation that lost the second member across the remainder is how a set is lost entirely. All members and the replacement were removed and labelled by bay — including the original disk 1, since a faulty bay reports whatever is in it as failed and that disk may never have been at fault. Each was imaged individually write-blocked on the Atola TaskForce 2, the failure sequence established from member metadata so stale members could be excluded, and the volume assembled offline from the images.

The outcome

Every disk including the original and the replacement imaged individually, the failure sequence established from 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, for anyone whose replacement drive failed on fitting: that is rarely a faulty drive. Rejection within seconds points at the bay, the backplane or the unit refusing it — which raises the possibility that your original disk was never the problem either. And the second member failing immediately afterwards was the rebuild, which is the heaviest load an array ever takes.

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.