> Blech! In SCTP we need to allow binding on arbitrary subsets of
> addresses. Our solution is to allow multiple calls to bind(). There
> is nothing special about the sets "all IPv4 addresses" and "all IPv6
> addresses" (well, there shouldn't be...).
This is a point which is not clear in the current draft (in fact, there are
many things which are not clear as Randy said...). The idea is that a v6
socket bind() to in6addr_any will associate the socket to all interface
addresses, regardless of whether they are v4 or v6. A v4 socket bind() to
INADDR_ANY is only associated with v4 addresses. There is no equivalent of v6
only bind() in the current draft (v6 socket trying to bind() to v4 INADDR_ANY
will fail).
Whether we want to have a v6 address only bind() is an issue we need to talk
about. There can be apps which support some v6 specific features and do not
want to be bothered with "accidental" v4 associations. Instead of aborting
those associations when the apps accepts them, it may be better for the OS to
reject them right away.
K. Poon.
[EMAIL PROTECTED]
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
