On Dec 20, 2005, at 10:30, JINMEI Tatuya / 神明達哉 wrote:

I'm confused about the T flag for an IPv6 multicast address with a
permanent group ID and would like to solicit clarification...

According to Section 4.2 of RFC3307:

   Permanent group IDs are allocated on an Expert Review basis, in the
   range 0x40000000 to 0x7FFFFFFF.  These permanent group IDs are meant
   to be used in IPv6 multicast addresses, defined in [UNIMCAST].
(UNIMCAST=RFC3306)

I interpret this as the corresponding IPv6 multicast address being
based on a unicast network prefix as defined in RFC3306.  Thus the P
flag of the 4-bit flags field must be 1, and then the T flag must also
be 1 as specified in RFC3306:

o If P = 1, T MUST be set to 1, otherwise the setting of the T
            bit is defined in Section 2.7 of [ADDRARCH].

Meanwhile, Section 4.3 of RFC3307 reads:

   It
   should be noted that the high-order bit of the Group ID will be the
   same value as the T flag.

Doesn't this contradict the relationship between the T flag and
permanent group IDs?  In fact, the high-order bit of a permanent group
ID is 0 while the T flag must be 1 as I explained above.


Section 4.3 only deals with multicast addresses allocated dynamically.
That is, they are administered locally either by an allocation server
(as described in Section 4.3.1) or by an end-host (Section 4.3.2).

Are you considering the case where addresses with permanent group
IDs are allocated by a server?  We didn't consider that scenario.

Multicast Addresses created using permanent group IDs (Section 4.2)
does not mention the T flag mainly because addresses generated
by RFC 3306 are not permanent.  Rather, they only have a lifetime
equivalent to the lifetime of the embedded prefix.

Does that clarify things?

Regards,
Brian

Attachment: smime.p7s
Description: S/MIME cryptographic signature

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to