Portable drives · written up · CDF-2025-0319
Slower Every Week, Until It Stopped Appearing.
Six weeks is how long it took. The man who minds the archive for a village near Rudry watched the drive get worse week by week — a minute to appear, then five
, then open a folder and you just get the spinner
, and in the end the light comes on and there is nothing
. The village's whole record was on it: minute books scanned back to the 1890s, the research on the war memorial, photographs collected at open days. Somebody had told him to defragment it and he had never got round to it, which turned out to be the best thing about the case.
Does that match yours? Call us.
0800 6890668
Why that happens.
Slowness here is arithmetic, not age; nothing inside is stiffening up. A sector that comes back on the second attempt costs almost nothing. A sector that comes back on the fiftieth costs a great deal, and the bill grows as the bad patch spreads. Thumbnails suffer first, because drawing a preview makes the disk read a whole file through, retries and all. And the morning it will not show up at all is that same arithmetic applied at mount time: the handshake that has to complete before your computer will show the drive now runs longer than the computer will tolerate.
What we used on this one.
Every step, in order →| The tools | What it did here | Why it helps |
|---|---|---|
| DeepSpar Disk Imager 4 | Completed the difficult passes over SATA, with timeouts tuned for the late head | Maps the heads first, works surface by surface, and keeps resets, timeouts and power in hand |
| DeepSpar USB Imager | Made the first pass through the enclosure's own USB bridge, with no operating system involved | Ordinary imaging speed, working through the USB bridge |
| UFS Explorer Professional Recovery | Built the folder tree back up from the images, then wrote it out to the new disk | Reads the volume formats that defeat most software: APFS, ReFS, XFS, ZFS, Btrfs |
How it went.
Get the numbers first, and open nothing
Nothing was opened when it arrived. A dying drive holds only so many reads, and one used on a look round is one thrown away. Measurement came first, then: the reallocated count was already high and still rising, and a per-head survey found one of the four replying late, and later each time. That head's zone lined up exactly with where the thumbnails had stopped drawing.
Take the healthy areas while they are healthy
Imaging followed confidence, not sector order. Three heads ran at full speed and had their zones clear before the day was out, which put the greater part of the archive beyond danger before the drive was pushed at all. Only then were the limits loosened — longer waits, supervised resets — and the passes moved onto the region behind six weeks of waiting.
Walk the failing head, then rebuild on a sound disk
The late head's zone was crossed at a crawl, pausing and power-cycling whenever it stalled. What came off was written to a new disk in the layout the records had always used, so it could be plugged straight in and the filing picked up where it left off.
Signing it off.
Losses were light. Back came the minute books, back came the war memorial research, and nearly every open-day photograph with them. One batch, from the very first scanning sessions, sat in the late head's zone; some of it returned damaged and some not at all, and the report names every file. Those minute books can go through a scanner again — the originals are still in the safe. Photographs that had only been lent were returned to their owners years ago.
Pages people often read next.
Filed under Pocket-size drives as well.
Is that what yours is doing?
Power it down, get it to us, and hold off on any decision until the diagnosis says what can still be read.