In your previous mail you wrote:
> PS (1): the idea is to send a magic packet to the destination
> at the known port. If only the source is natted then the peer
> address and port will be available in the encapsulated stuff.
I've observed that, even if there is traffic every few seconds,
the NAT box I have will change the UDP source port occasionally
for UDP traffic.
=> argh! This mandates a IPv6 over TCP/IPv4 too because the NAT cannot
do this for a TCP connection...
PS. The above is based on operational experience for IPv4 over UDP/IPv4
to enable IPsec tunnels on top of that.
=> and can IPsec work with such a (broken) NAT?
[EMAIL PROTECTED]
PS: something like the BSD user-PPP is more and more attractive!
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
