Data Recovery Case File · Portable Drives · One Device, Whole-System Symptoms
The Computer Is Fine and It Is Waiting
His enquiry describes symptoms that look like a failing computer and are not. A 3TB external drive: "when plugged in it is detected by the file manager, but when you try to access any files the progress bar fills up and lags out forever then crashes. Also, whilst it is plugged in, it stops other services working. I have tried this on a desktop and two laptops." A single unresponsive drive can make an entire machine appear broken — and the fact that it did so on three separate computers is the clearest possible confirmation of where the fault sits.
| Media | 3TB external hard drive — enumerating and presenting; read operations not completing; host services degrading while connected |
| Reported situation | Drive detected by the host file manager · file access producing an indefinitely progressing operation · file manager subsequently failing · unrelated host services affected while the drive remains connected · behaviour reproduced across three separate machines · contents required |
| Fault class | Severe read failure with the device still accepting requests — outstanding operations blocking host resources; each attempt consuming substantial margin |
| Equipment used | No further host connection attempts · Atola Insight Forensic error-rate assessment · imaged write-blocked under strict per-sector timeouts with retries capped · healthy expanse captured first and weak regions revisited · losses mapped per file |
The decode: why one drive can freeze everything around it
What the progress bar is actually showing: not progress. It is an estimate the file manager displays while it waits for a request to complete. When the drive neither answers nor refuses, there is nothing to update it with, so it fills slowly and sits there — which looks like something happening and is the opposite.
Why the drive does not simply fail the request: because it is trying. A drive that cannot read a region retries, repositions, and escalates through its internal recovery sequence before reporting an error, taking seconds per sector. A directory listing touching hundreds of unreadable sectors becomes an operation measured in hours, and nothing in the interface conveys that.
Now the part that explains the wider symptoms. A request that has been issued and not answered occupies a slot in the operating system's input queue, and it holds resources while it waits. Enough outstanding requests and the system runs short — which is why unrelated services start failing while the drive is connected. Nothing is wrong with them; they are queued behind a device that is not replying.
So the machine is not broken: it is being held. Unplug the drive and everything returns to normal within moments, which is worth doing simply to confirm it and to stop attributing the behaviour to the computers.
What three machines established: everything at the host end. The drive produced identical behaviour on a desktop and two laptops, which eliminates operating system versions, drivers, ports and configurations in one step. That is a complete elimination and there is nothing further to try.
Why each attempt is expensive: an operation that runs for an hour is an hour of a failing drive working at maximum effort, with heads crossing the affected regions repeatedly. Waiting for the bar to finish is the costliest thing available, and it is exactly what the interface encourages.
Why the recovery approach inverts the drive's own behaviour: imaging under strict per-sector timeouts caps retries deliberately — a difficult sector gets a short budget and is deferred rather than allowed to consume the session. The healthy expanse is captured quickly first, and weak regions revisited on later passes. Letting a drive retry freely recovers less and takes far longer.
On the bench
No further host connection attempts were made, an operation left running being an hour of a failing drive at maximum effort. The Atola Insight Forensic assessed error rates, and imaging ran write-blocked under strict per-sector timeouts with retries capped — a drive escalating through its internal recovery sequence takes seconds per sector and holds outstanding requests in the host's queue, which is why unrelated services degrade while it is connected. The healthy expanse was captured first and weak regions revisited, with losses resolved to individual files.
The outcome
Error rates measured, the drive imaged with retries capped and the healthy expanse captured before weak regions were revisited. 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: your computers are fine. A request that has been issued and not answered holds a slot in the system's input queue, so enough outstanding reads leave the machine short and unrelated services fail. Unplug the drive and everything returns in moments.
Drive that freezes the whole computer while connected
Unplug it — everything will return to normal within moments, and it's worth doing just to confirm your machines are fine. What's happening is that requests issued to the drive are neither answered nor refused, because a drive that can't read a region retries and escalates through its internal recovery sequence, taking seconds per sector. Each unanswered request holds a slot in the operating system's input queue, so enough of them leave the system short of resources and unrelated services start failing. The progress bar isn't showing progress either — it's an estimate displayed while waiting, with nothing arriving to update it. Don't wait for it to finish; that's the most expensive thing you can do.
Unplug it and stop — call Guildford Data Recovery on 01483 901310; error rates measured, imaged under strict per-sector timeouts with retries capped, healthy expanse captured before weak regions are revisited.
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.