1 |
On Thursday 15 Jun 2017 15:47:03 Rich Freeman wrote: |
2 |
|
3 |
> You're comparing a 500kV breaker at a substation to a USB device? |
4 |
> |
5 |
> I'm very skeptical of the claim that any electrical effects associated |
6 |
> with unplugging a device is going to cause issues with any USB device. |
7 |
> They're basically designed to be hot swapped. |
8 |
|
9 |
The comparison was made tongue-in-cheek to illustrate the point. I don't know |
10 |
what the probabilities are but on at least two occasions I ended up with a USB |
11 |
stick which had a large number of bad blocks and partially destroyed files. |
12 |
This was a relatively young, but cheap USB stick. The I/O errors were quite |
13 |
high and over-writing the blocks with zeros did not fix the problem. I've |
14 |
come across this problem on two different USB sticks, probably because I was |
15 |
the go to guy for recovering data in an office full of MSWindows PCs. |
16 |
|
17 |
|
18 |
> I'd also buy the argument that some poorly designed USB drives could |
19 |
> end up with data loss to something other than the block being |
20 |
> immediately written, but honestly I'm skeptical that this is a |
21 |
> widespread problem. |
22 |
|
23 |
Losing the odd files because of unplugging a USB stick during a write |
24 |
operation is a regular occurrence. Ending up with large numbers of bad blocks |
25 |
is a rare phenomenon in my experience. I blamed the problem on cheap USB |
26 |
sticks not being able to cope with the enthusiastic unplugging they were |
27 |
submitted to, which caused semiconductor breakdown or damage to the USB |
28 |
controller chip. At the time I thought the wear levelling chip went sideways |
29 |
and was no longer able to re-allocate sectors. Attempts to fix the bad blocks |
30 |
were unreliable and after some time using smartctl and dd to zero badblocks I |
31 |
gave up. |
32 |
-- |
33 |
Regards, |
34 |
Mick |