Rich & I managed to get draft-ietf-ipngwg-addrconf-privacy-xx.txt
revised before the cutoff. The main changes are to address the issues
raised in Rich's mail to the ipng list on 6/6/2000.  Specifically:

1) When a site has multiple active prefixes (e.g., is multihomed) it
   generates an anonymous address for each of those prefixes from the
   same random interface identifier. This reduces the number of
   solicited-node multicast addresses a node must join.

2) Public addresses are not deprecated as in -01. Instead,
   draft-ietf-ipngwg-default-addr-select-xx.txt was updated (and is in
   the publication queue) to give precedence to anonymous
   addresses. Note that this places a normative dependency from this
   document to default-addr-select.

   One advantage of this change is that the set of public addresses
   (and their associated liftimes) will be the same whether  or not
   anonymous addresses are in use.  Previously, the public address was
   deprecated (preferred lifetime set to 0) in order to ensure that it
   was not selected for outgoing communication.

3) Additional rules have been specified to ensure that when RAs update
   the lifetime of public addresses, the corresponding lifetimes of
   associated anonymous addresses are also updated. RAs can only
   shorten the lifetimes of anonymous addresses, not lengthen them. In
   the previous version of this draft, the lifetimes of anonymous
   addresses were not properly adjusted downward in some circumstances.

For those who can't wait for the official draft, an advance copy is
available at
ftp://ftp.cs.duke.edu/pub/narten/ietf/draft-ietf-ipngwg-addrconf-privacy-02.txt

Thomas & Rich

--------------------------------------------------------------------
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