I believe this draft is useful. Three issues:
1) Brian Carpenter writes: > > 2.4.2. IPv4 Multicast > > > > The globally routable group 224.0.1.20 is set aside for > > experimentation. For certain experiments, the administratively > > scoped multicast groups defined in [RFC2365] may be > > useful.[[anchor10: Should there be a 'link-local' 224.0.0.x > > experiment group? --wcf]] > > To me it doesn't really seem necessary. Any link-local resource is > safe to use experimentally, subject to local administration. Unfortunately, that's not really true. Unlike in IPv6, there are no link-local IPv4 multicast addresses which aren't in the range IANA uses to assign to protocols to be hard-coded. Per RFC 2365, IANA reserves a /24 in _every_ scope for it to use for this purpose. http://www.iana.org/assignments/multicast-addresses contains the list of the link-local assignments in the top section. So I think there should indeed be a link-local 224.0.0.x experiment group. I think there is also justification for a scope-relative experiment value. 2) Section 3.4.2 mentions an experimental IPv6 group address, but it would be good to add a statement (consistent with the 2nd sentence in 2.4.2) about the use of arbitrary scoped groups with the T=1 bit set. In other words, whatever is said for IPv4 should be said for IPv6. 3) In the interest of saying what should not be done (like the statement in section 1 about hard coding values), I believe section 3.4.1 (IPv6 Unicast Addresses) should contain a statement that it is inappropriate to use RFC 3849 addresses in experiments. -Dave _______________________________________________ Int-area mailing list [email protected] https://www1.ietf.org/mailman/listinfo/int-area
