Re: Issue: Support for IPv4 - IPv6 movements (#13)

 In your previous mail you wrote:

   2401bis already mandates support for selector sets that allow a 
   single SA to refer to a set of IP addresses and ports,

=> argh! The term "ports" here disturbs me: we are talking about
the support of multiple simultaneous *outer* addresses (not
traffic selectors aka inner addresses) in the SADB.

   which is one 
   of the major requirements for SCTP that was not available under 2401. 
   What else do you feel needs to be covered to enable SCTP to work?
   
=> in fact this is not necessary to enable SCTP to work.

Thanks

[email protected]

PS: the exact quote from RFC 3554 is:

   Similarly, SAs may have multiple associated source and destination
   addresses.  Thus an SA is identified by the extended triplet ({set of
   destination addresses}, SPI, Security Protocol).  A lookup in the
   Security Association Database (SADB) using the triplet (Destination
   Address, SPI, Security Protocol), where Destination Address is any of
   the negotiated peer addresses, MUST return the same SA.

The text becomes with a "may" so this is not a requirement. The detail
(introduce by "thus") is about the inbound processing and:
 - is fully compatible with the RFC 2401bis (lookup using the couple
   SPI/protocol) at the exception of "source/destination" terminology
   (RFC 2401bis reserves these to outer addresses and use "local/remote").
 - is easy to implement (just ignore the destination address: if the
   packet comes up to this piece of code it should have a right one :-).
Of course for outbound processing there is a peer address selection issue
and things are not simple at all.
BTW there is a glitch at the beginning of the section 4.1 of the rfc2401bis
I-D 02 and only something about the "tunnel header IP source and
destination address" (and the strange verb "purport" ???) in 4.4.2 (SAD).