On Nov 6, 2010, at 5:19 PM, Bob McConnell wrote:

But I still wonder if you are unique in finding this address collision,
or do others also have the same problem? If it is widespread, then it
should be solved by the people managing those devices.

Nah; one of the prominent use cases for NAT on Cisco routers is linking between two overlapping networks. (see http://www.cisco.com/en/US/tech/tk648/tk361/technologies_configuration_example09186a0080093f30.shtml

This happens when companies merge, for instance, and both of them used the same or overlapping RFC1918 networks; happens a lot with 10.0.0.0/8 and 192.168.0.0/16 (mostly in the 192.168.0.0/24 and 192.168.1.0/24), not so much in 172.16.0.0/12 (which then becomes a popular pool to NAT the overlappers to).  Judicious NAT and split DNS help solve the problem until things can get renumbered.  Large networks never do get renumbered, and NAT between enterprise networks lives on.

IPv6 includes a large block of ULA addresses to hopefully reduce collisions of this sort for non-globally-routed addresses.