In message <20050304191857.BEEC23BFE41@berkshire.machshav.com>, "Steven M. Bell
ovin" writes:

>>
>>I've noticed that placing a mini-fridge on the same power strip as my noteboo
>k
>>ac adapter causes the same thing to happen sometimes at exactly the moment 
>>when the compressor cycles on. I think this has happened under windows a few
>>times too, so perhaps it's really a hardware problem. I use optical mice, 
>>which I believe use about 100mA of current from the bus, so combinations 
>>of hardware and usb peripherals which use modest amounts of juice have simila
>r
>>problems. 
>>
>>This is just a hypothesis. I can't back it up, nor have I tested it. I can
>>think of a few reasons why it shouldn't be true. I do know that since
>>I've rearranged my power cables it happens much less freqeuntly than it used
>>to.
>
>Interesting.  I'm about to move one of the affected machines to a UPS; 
>I'll see what happens.
>

I now have two incidents where there was a failure immediately 
subsequent to a power-related event.  I haven't yet established if it's 
a line voltage problem or inductive coupling to the USB cable.  I'm 
willing to call this part of the problem solved -- the question now is 
what, if anything, NetBSD can do about it.

		--Prof. Steven M. Bellovin, http://www.cs.columbia.edu/~smb