Data Recovery Case File · Formatted & Logical Faults · Names Mean Reconstruction
Lost Partition With Folder and File Names Retained
His enquiry was written by someone who knows what to ask, and the last line is the one that governs the work. A Seagate 1TB 2.5-inch laptop drive: "has unreadable sectors (a lost C partition). There are no SMART hardware issues reporting. No clicking or unusual noises. We require folder names and file names retained in the data recovery." That final requirement is not a preference to be accommodated where convenient — it determines the method. There are two fundamentally different ways to recover data from a damaged volume, and only one of them returns names and structure. Specifying it upfront, as he has, is the difference between receiving a usable dataset and receiving a hundred thousand files called recovered_00001.
| Media | Seagate 1TB 2.5-inch laptop hard drive — unreadable sectors present, primary partition lost; no adverse health reporting and no abnormal noise; folder and file names required in the delivered set |
| Reported situation | Partition no longer present · unreadable sectors encountered · drive health self-reporting as normal · mechanically quiet · structure and naming specified as a delivery requirement |
| Fault class | Partition loss with localised media defects — metadata survival determining whether structure can be rebuilt rather than carved |
| Equipment used | Imaging under per-sector timeouts with weak regions re-read on later passes (PC-3000 Express, Data Extractor) · filesystem metadata located and reconstructed on the image in preference to carving · delivered tree compared against recovered metadata; carved remainder segregated and labelled |
The decode: two methods, and what decides which you get
Method one — reconstruction. The filesystem's own structures are located and repaired: the records naming every file, recording its size and dates, and describing which folder it belongs to. Rebuild those and the volume returns as it was — folder tree, filenames, timestamps, hierarchy. This is what he is asking for, and it is always attempted first.
Method two — carving. Where the metadata cannot be recovered, files are identified by their content instead: a scan looks for the internal signatures that mark the start of a document, an image, a video, and extracts what follows. It is remarkably effective at retrieving data and completely useless at retrieving organisation, because names and folder relationships live in the metadata rather than in the files. Carved output arrives sorted by type with sequential names, and no timestamps beyond what individual formats happen to carry internally.
What decides which he gets: not the price and not the effort, but whether the metadata survived. And that is where his own observations become relevant. Unreadable sectors matter enormously depending on where they fall: filesystem metadata is concentrated in specific regions, so defects landing there cost structure while defects scattered through the data area cost individual files. So the honest answer to his requirement is: reconstruction is attempted first and thoroughly, it usually succeeds, and where it partially succeeds the result is a real tree with some files unplaced rather than an all-or-nothing outcome. What should never happen is a bench carving by default because it is quicker and delivering a flat heap.
On the healthy self-reporting: worth a note, because he mentions it as reassurance. A drive reporting no health issues while returning unreadable sectors is not a contradiction — self-monitoring tracks thresholds the manufacturer chose, and a drive can have genuine localised defects while every monitored counter sits within tolerance. Silence from that system is not a clean bill of health, and it should not be read as one. That the drive is quiet, though, is meaningful: it rules out the mechanical faults, which is consistent with a localised media problem.
The delivery detail worth insisting on: where a job produces both reconstructed and carved output, the two should be segregated and labelled rather than merged. Mixing them silently produces a tree that looks complete and a pile of unnamed extras nobody can place.
On the bench
Imaging ran first on the PC-3000 Express under Data Extractor with per-sector timeouts enforced and the weak regions revisited on later passes — because on a job specified this way, every sector recovered from the metadata region is worth disproportionately more than one recovered from the data area. Reconstruction was then attempted thoroughly on the completed image: the filesystem's structures located and repaired until the tree stood up with folder names, filenames and timestamps intact. Carving ran only for what reconstruction could not place, and its output was delivered separately and labelled rather than merged into the tree, with a written note of which files came from which route.
The outcome
The volume rebuilt from the image with folder and file names intact, and the carved remainder delivered separately and labelled. Free assessment, one fixed written figure including VAT, no recovery, no fee. The decode, for anyone specifying a recovery: naming and structure are not a delivery preference but a choice of method — reconstruction repairs the filesystem's own records and returns your tree, while carving identifies files by content and returns data stripped of names, folders and dates; which you get depends on whether the metadata survived, which in turn depends on where the unreadable sectors fell rather than how many there are. State the requirement upfront as he did, insist reconstruction is attempted before carving, and ask for any carved output to be segregated and labelled rather than merged.
Specifying a recovery where structure matters
Say at the outset that you need folder and file names retained — it changes the method rather than the price. Reconstruction repairs the filesystem's own records and gives you back your tree with names, folders and dates; carving finds files by their internal signatures and gives you the data with none of that, arriving as type-sorted folders full of sequentially numbered files. Both are legitimate, but carving is the fallback and shouldn't be the default because it's faster. Ask three things: that reconstruction is attempted first and thoroughly, that imaging prioritises the metadata regions rather than treating all sectors equally, and that any carved remainder is delivered separately and labelled rather than mixed into the tree. And don't take healthy self-monitoring as proof the drive is fine — it tracks chosen thresholds, and a drive can have real localised defects while reporting nothing.
Specify it upfront — call Glasgow Data Recovery on 0141 404 0294; imaged with metadata regions prioritised, filesystem rebuilt before any carving, carved remainder segregated and labelled.
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.