>> Michael Thomas wrote:
>> There is some appeal to 6to4 and 1918... it keeps
>> the problem within the cesspool of current usage
>> and doesn't try to rationalize it any further. 
>> A maze of twisty addresses, all alike...

> Tim Chown wrote:
> But as Pekka says, won't 6to4 interfaces fail to
> deliver to 2002:<RFC1918> if the network is IPv6 only,
> or doesn't use IPv4 private IPs?

I don't see why. 6to4 addresses are supposed to be completely routable within a site 
until they reach the 6to4 gateway typically placed at the edge. With that one IPv4 
address which is the gateway, you get a /48 6to4 prefix which does indeed includes 16 
bits for local subnet topology. Looking at my own network, if I wanted to use 6to4 as 
the internal private scheme, I just have to delete the following on the edge router:

| interface Tunnel6
|   description for ipv6 6to4 tunnels
|   ipv6 address 2002:D1E9:7E41::1/64
|   ipv6 traffic-filter IPV6-ACL-OUTSIDE-IN in
|   tunnel source Ethernet0/0
|   tunnel mode ipv6ip 6to4
|   tunnel path-mtu-discovery

and also delete

| ipv6 route 2002::/16 Tunnel6

And the 2002: internal routing still works with whatever IGP I choose (I have not 
tried with IS-IS). Voil�. On the host side, make sure that they pick the RA announcing 
the 6to4 prefix.


Besides, if one does not need a prefix as short as a /24, here's another good 
candidate for hijacking: 2001:0DB8::/32. I guess it makes the documentation easy to 
write :-)

Michel.


--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page:                      http://playground.sun.com/ipng
FTP archive:                      ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------

Reply via email to