i presume folk in this wg are tracking the comments which have been received
in other wgs?

randy


From: Pavlin Radoslavov <[EMAIL PROTECTED]>
To: Randy Bush <[EMAIL PROTECTED]>
cc: mboned wg <[EMAIL PROTECTED]>
Subject: Re: v6 documents 
Date: Wed, 31 Oct 2001 01:23:17 -0800

(Sorry if a duplicate, but seems like for some reason my first
attempt few days ago wasn't propagated by the mailing list.)

> has this wg taken a look at
>   draft-ietf-ipngwg-uni-based-mcast-03.txt

Few comments:

* It seems that under this proposal the allocated multicast
  addresses per unicast prefix will always be /96, regardless of
  the size of the "network prefix".
  Is the assumption that 2^32 multicast addresses will be sufficient
  for any network, regardless of its size?
  Indeed, Section 4 says that
  "All non-significant bits of the network prefix field SHOULD be
  zero."
  I'd like to know what is the argument against allowing those
  non-significant bits to be part of the group ID.
  E.g., the alternative could be something like:
  "The group ID should be 32 bits, but if a network needs more than 2^32
   multicast addresses, it may use the non-significant bits
   from "network prefix" as well."

* Section 6
  "o Set P = 1."

  Probably this should be specific re. the T-bit as well, even
  though earlier you specify the relation between those two bits:
  E.g.: 
  o Set P = 1 and T = 1.

* Section 6

  "These settings create an SSM range of FF3x::/32 ..."

I believe this should be /96

Pavlin
--------------------------------------------------------------------
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]
--------------------------------------------------------------------

Reply via email to