IPv6 transition mechanism

This is an old revision of this page, as edited by 134.91.226.253 (talk) at 00:00, 23 August 2007 (removed broken link). The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

IPv6 translation mechanisms are used to allow IPv6-connected hosts to access IPv4-connected hosts. They were first defined in RFC 2765: Stateless IP/ICMP Translation, or SIIT.

SIIT

In computer networking, Stateless IP/ICMP Translation is a mechanism which translates IPv6 packets into IPv4 packets and vice-versa.

It was initially drafted in February of 2000 by E. Nordmark of Sun Microsystems and is described in RFC 2765 (Network Working Group at Internet Engineering Task Force).

Basically SIIT describes a method by which a router interprets an IPv4 header and creates a parallel IPv6 header with equivalent information and the inverse equivalent operation of converting an IPv6 header into an IPv4 header. The actual means of converting an IPv4 address to an IPv6 address or vice-versa may vary, and the means by which the routing occurs is described as unspecified in the RFC.

NAT-PT

NAT-PT, or Network Address Translation/Protocol Translation, was defined in RFC 2766. It will attempt DNS AAAA-A record translation on its own, but this makes it very difficult to implement, and few implementations are available (as of 2007).

Implementations

TRT

TRT, or Transport Relay Translation, was defined in RFC 3142. This is the most common form of SIIT, but it relies on DNS forwarding for AAAA-A record translation.

Implementations

  • faithd: BSD-based TRT
  • totd: translator for BSD & Linux
  • pTRTd: [1]; attempt at TRT for Linux.

Other Mechanisms

  • Bump-in-the-Stack / Bump-in-the-API: RFC 2767
  • Socks-based Gateway: RFC 3089
  • TCP-UDP Relay: RFC 3142

See also

External websites

References

  • ISBN 3-540-24524-3: IPv6 in Practice by Benedikt Stockebrand, 2006