Data Recovery Case File · NAS & Network Storage · Both Halves, One Damaged
NAS Dropped: One Disk Clicking in a Striped Pair
Her enquiry supplied every fact needed to answer it and asked the right question. "I've accidentally dropped my two-bay NAS and unfortunately now one of the disks — it has two 4TB SATA disks — is making a clicking noise and can't be read. Config RAID 0. What are the chances to fully recover the data, and how much will this cost?" Email contact was requested and honoured. The honest answer turns on the interaction between two facts she has already given. RAID 0 means every file is split across both disks, so the damaged one is not a spare — it holds half of everything. And clicking means mechanical damage from the fall. So the chance of full recovery is not a general estimate: it is precisely the proportion of that clicking disk which can be read after it has been serviced.
| Media | Two-bay network storage unit, striped configuration — two 4TB member disks; unit dropped; one member clicking and unreadable, the other apparently unaffected |
| Reported situation | Impact to the unit · one member producing clicking and refusing to read · array configured without redundancy · full recovery prospects and cost sought |
| Fault class | Impact damage to one member of a striped array — no redundancy available; completeness governed by how much of the damaged member can be imaged |
| Equipment used | Healthy member imaged first, write-blocked · damaged member inspected under the laminar flow bench before any power · matched donor head service · PC-3000 Express with Data Extractor prioritised imaging · stripe reconstructed offline from both images; losses mapped to files |
The decode: why the damaged disk decides everything, and what the answer depends on
What striping does to her situation: a striped array splits every file into chunks written alternately across both members — first chunk to one disk, second to the other, and so on. There is no parity and no mirror. So the clicking disk does not contain "some of the files"; it contains alternating pieces of all of them. A file cannot be reassembled from one member alone: the result is a fragment of the file followed by a gap, repeating. That is why the answer to "what are the chances of full recovery" is entirely a question about the damaged disk.
What clicking means after a drop: the head assembly is attempting to position itself and failing, almost always because the impact displaced or damaged it. That is serviceable — matched donor heads are fitted and the drive is imaged — but two things govern the outcome. Whether the fall left surface damage as well as head damage, since scored areas cannot be read by anyone. And how many times the drive has been powered since, because every click is a damaged assembly travelling across the platters. Her having stopped is the best thing she could have done.
How the honest answer is arrived at: not by estimate. The healthy member is imaged first and completely, because it is the half that is currently safe and it should be secured before anything else is attempted. The damaged member is then inspected before any power is applied, serviced, and imaged with the readable regions prioritised. Only then can the question be answered — and it is answered as a map: the stripe is reconstructed from both images, and any region the damaged disk could not yield is resolved to the specific files it affects. So instead of a percentage, she receives a list of what came back whole, what came back partial, and what did not come back. On a striped pair, a small unreadable area on one member affects a correspondingly small set of files rather than everything, which is the encouraging arithmetic here.
The thing that must not happen: the unit must not be powered up with both disks in it, and no rebuild, resync or repair should be attempted. On a striped array there is nothing to rebuild from, and letting the appliance try can overwrite the layout a reconstruction depends on. Both disks come out, labelled by bay.
On the bench
The healthy member was imaged write-blocked first — securing the half that was not at risk before anything else was touched, which is the discipline that makes a partial outcome survivable. The clicking member was not powered on arrival: it was opened under the laminar flow bench and its head assembly and platter surfaces inspected, the drop's damage located and any scoring measured, before matched donor heads were fitted. Imaging then ran on the PC-3000 Express under Data Extractor, head-mapped and prioritised, taking the readable expanse completely before approaching damaged zones. The stripe was reconstructed offline from the two images, and every unreadable region was resolved to the individual files it affected.
The outcome
Both members imaged, the stripe reconstructed from the copies, and the result delivered as a mapped list of complete, partial and lost files. Free assessment, one fixed written figure including VAT, 50% of parts and labour upfront with the balance only on successful recovery. The honest answer to her question: on a striped pair there is no redundancy, so the damaged disk holds alternating pieces of every file and the completeness of the recovery is exactly the proportion of that disk which can be read after servicing; clicking after a drop means head damage, which is serviceable, with the outcome governed by whether the surfaces were scored and by how many times it was powered afterwards; and the result should come as a per-file map rather than a percentage. Take both disks out, label the bays, and let nothing rebuild or resync.
Striped NAS with one disk damaged
Power the unit down and take both disks out, labelling which bay each came from. On a striped array there is no redundancy, so the damaged disk isn't a spare — it holds alternating pieces of every file, and nothing can be reassembled without it. Don't let the unit rebuild, resync or repair anything: there is nothing to rebuild from, and the attempt can overwrite the layout a reconstruction needs. Don't insert a replacement disk either. If one drive is clicking, stop powering it entirely — every click is a damaged head assembly crossing the platters, and the number of attempts is one of the few things you control. Ask for the healthy member to be imaged first, before any work on the damaged one, so the half that isn't at risk is secured. And ask for the result as a per-file map rather than a percentage: on a stripe, a small unreadable area affects a correspondingly small set of files.
Take both disks out and stop powering them — call Leeds Data Recovery on 0113 322 3083; healthy member secured first, damaged member serviced and imaged, losses mapped file by file.
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.