Question Why some usb devices put Windows sin unresponsive state?

It shouldn't happen. I deal with USB storage devices on Windows on a regular basis and the only occasion I can think of where this has happened to me lately was with two faulty USB devices; all of the above systems (except crashing/blocking) plus lots of USB disconnects, to the point that I barely had a chance to read anything from the drive before it started messing around again. In that situation the customer had bought the USB devices from a local big-name supermarket so I doubt the drives were fake (they're a brand I often use) but I suppose it's possible.

Apart from a dodgy USB device, I get more problems with USB disconnects with a USB port that isn't mounted on the motherboard, so any that are connected to USB headers. As a result, when I'm starting an OS install I always use a board-mounted port so I don't have to spend an extra minute or two wondering if Windows setup is ever going to start.

If I were you I'd start by creating an ID for each drive (e.g. physically mark it), then keep a tally of which drives have (more?) problems. Hopefully you'll narrow it down to a particular drive/model/whatever.

Don't forget the cables. Bad cables can cause some pretty weird problems too. Expecially the ones that come with cheap enclosures.

If you look in the resource monitor and event viewer, I bet the drives that freeze the computers are causing I/O errors.

Could be cables, could be drives, could be dodgy power supplies - got a lot to narrow down. But the event viewer will show the I/O timeouts.

Now, with these dodgy drives, have you ever tried using a different operating system on them, like Linux?It would be interesting to see if the problem persists across different systems or if it's specific to Windows interaction with the drives

Klingenberg said:

Now, with these dodgy drives, have you ever tried using a different operating system on them, like Linux?It would be interesting to see if the problem persists across different systems or if it's specific to Windows interaction with the drives

Click to expand...

I used to admin a VMWare cluster with a mix of Windows and Linux VMs, and when we'd do stuff like a SAN failover or something, Windows and Linux would behave very, very differently in response to the same underlying storage burp. So, like, just be prepared for that I guess.

dave_the_nerd said:

I used to admin a VMWare cluster with a mix of Windows and Linux VMs, and when we'd do stuff like a SAN failover or something, Windows and Linux would behave very, very differently in response to the same underlying storage burp. So, like, just be prepared for that I guess.

Click to expand...

You make it sound like it'll be very very scary or something along those lines, what?

Klingenberg said:

You make it sound like it'll be very very scary or something along those lines, what?

Click to expand...

Sorry, I half-typed a full explanation, but cut it to keep the post short and to the point.

Sometimes it worked fine. (According the SAN vendor it was supposed to, after all.) But when it didn't, all the Windows VMs would crash/freeze. That was easy enough to deal with - select all and power cycle in the VM manager.

Linux VMs would keep running, but some random % of them would switch their storage mounts into read-only mode. Depending on what we were using the VM for, it might be hours, days, or weeks before we noticed an issue. You could use the mount/remount command to fix it, or just restart. We eventually configured monitors that triggered an alert if a file system was in read-only mode.

My point is just that "boot it into Linux and see if you still have the problem" isn't probably the best diagnostic, since a different OS may mask a failure, or respond in a completely different way than you might not realize has the same root cause.

My favorite weird-very-specific-storage-problem was a bad SATA cable on my home PC. Probably about 2010-ish. The only obvious symptom was that if I tried to launch Chrome, the system would freeze for about a minute solid, then carry on as if nothing happened. I put up with it for a while (I mostly used Firefox anyway) but eventually decided to poke around and see what the issue was. That was when much-younger-me learned about the Windows Event Viewer and how to use it to find storage errors.