We’re a Recycling company based in Glasgow, Scotland and we’ve run into a bit of a dilemma that no where else seems to have either 1: run into 2: documented.
We recently had a bunch of external drives sent to us for secure data erasure, which we done with the software we use for wiping, fine, all good so far.
It seems though, between wiping them and then plugging them back in, the drives become unresponsive with this error message.
“The request could not be performed because of an I/O device error”
What we’re trying to do is discover if it’s something wrong with our drive wiping software or if it’s just the drives failing.
Most of the drives we have erased are 4TB WD My passports, but we also have 3 DiskAshur2 5TB models that it has done the same thing to every single one of them.
Well, plan B if the drive is failing is to drill a few large holes in the drive. That both EOLs it and make the data pretty much impossible to recover, although there are ways to read off some data even from that.
But, yeah, first use known good drives to see if the problem is with the drives or the setup.
We’ve not had any issues with the 5TB drives we got, the WD Models but it’s mainly the 4TB WD drives that have had this error. We’re not worried about the data being recoverable since if they completely don’t work they will be shredded anyhow, but we just wanted to see if someone has made had some issue like this before and would know how to fix it.
The software will detect the drive as working, then erase it then once we go to initialize the disk in Windows, it will come up with that error and it seems to be 40/50% of the drives we have received and it has been around maybe 60 of them so far that have had the same issue, and we’ve tried to use Linux with DD and FDisk to see if it was just a windows thing, but that had the same issue.
FWIW every time I’ve hit this in Windows it’s been a mechanical failure of the drive. Harder with enclosed 2.5s but with 3.5s I’ve consistently been able to hear it no problem. Spin up fails, excessive vibration from being out of balance, and arm stuff.
One such drive went for data recovery and the recovery service noped out on it as soon as they opened it due to damage. I never heard details but sounded like the heads had dragged over most of the platter surface.
So most likely, the drives are actually toast? Just I’ve also got three iStorage DiskAshur2’s that are doing the same thing and I can’t hear the drive clicking, or failing to spin up and it seems to be accessing it OK, but I kind of thought it might have just been mechanical failure but was just wanting to see if anyone maybe had any different ideas or suggestions on what to try?
Have you tried testing them with linux? Windows has been getting progressively more and more buggy. I noticed some weird mounting issues when I used DBAN a few months ago to wipe a drive, but had no issues mounting/reading in linux.
My guess is the drives are simply having a hardware failure, triggering a windows related bug, or something within the drives firmware is bugging out.
That would be my working assumption. It’s odd to have so many, but that’s perhaps more a question for the people sending over the drives for disposal. Like maybe they batched up all the failed ones or something. Could also be other odd stuff like the erase hitting a Passport firmware issue and the Ashurs being coincidental co-occuring failures.
If there’s a hard process requirement for assured data destruction before shredding I guess it’d be take them apart and sand off the platters or something. If this is about trying to reclaim the drives for further use it doesn’t seem worth bothering.
I have hit cases broadly similar to what you describe, with failing drives continuing to accept writes for some hours and then raising errors once reads are asked for.
It sounds like the wipe might be messing with the controller or firmware. Try testing the drives directly with SATA, not through the enclosure. That way, we can see if the drives are the problem or if it’s just the USB thingy that’s failing.
I recently had this same error after trying to wipe an Optane drive. I did it through Intel’s own software tools but it failed wiping part way through and basically locked up the drive. I can see it in the bios, I can see the proper size of the drive is there in Windows, everything looks like it should work fine. But any IO to the drive fails like the device is completely locked out.
So I guess my advice would be that with your own wiping tool, check that it actually succeeded and didnt fail at some point during the wipe. It is possible the USB controller between the drive and the PC is causing some sort of interfacing issue to the drive with the commands being sent to wipe.