We have some updates on the draft
draft-ietf-ipngwg-scoping-arch-03.txt, on which I will make a short
presentation in the Thursday ipv6 wg meeting.
Since we missed the deadline to publish a new ID and since I'm not
sure if I will have an enough slot for the presentation, please let me
post a digest of the changes. Comments to this message are, of
course, welcome.
There are two changes, both of which were based on inputs after the 03
draft:
1. revise the mobility section to clarify issues about using
site-local addresses with mobile IPv6:
+ site-local home/care-of addresses can be used when the mobile node
moves within its home site and communicate with nodes within the site.
+ in general, however, there are several issues. For example, if
the care-of address is a site-local address, an off-site mobile
node cannot communicate with the home agent to register the care-of
address. Also, if the home address is a site-local address
(without route optimization) and a correspondent node (CN) is in a
different site from the mobile node's home site, the CN cannot send
packets back to the mobile node unless it can use route
optimization.
+ to deal with such issues, we'll need an ability for mobile nodes
to tell whether a correspondent node or the home agent is in the
same site as the mobile node. However, there is currently no
standard way to implement the ability.
+ thus, we recommend to use global home/care-of addresses *whenever
possible*. (note, however, this does not mean we prohibit mobile
nodes from using site-local home/care-of addresses.)
2. revise the textual representation section
in the form of <address>%<zone_id>, the <zone_id> part is now
formatted as follows:
<scope_type>.<id_in_the_scope>
where <scope_type> is a decimal number from 0 to 15 to specify a
scope type, which corresponds to the "scop" field values of multicast
addresses, and <id_in_the_scope> is a decimal number to identify a
zone. "." is a delimiter character to separate <scope_type> from
<id_in_the_scope>.
(the reason for the delimiter character "." is because it is used in
the "normal" address representation so address parser should escape
the character. And, "." is more intuitive and more readable than
other characters (i.e. ":", "a-f0-9").)
An example: a site-local address fec0::1 on the #2 site can
represented as fec0::1%5.2 (where 5 means the site scope and 2 means
the 2nd site)
The previous "readable" representation like "link2" or "site10" were
removed accordingly. The "unreadable" numerical representation like
1342177281 (= 0x50000001) were also removed.
JINMEI, Tatuya
Communication Platform Lab.
Corporate R&D Center, Toshiba Corp.
[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]
--------------------------------------------------------------------
