> One other source of confusion might be that some implementations, when
> setting routes manually (ie. not autoconfed), instead of setting _default_
> route, prefer something like 2000::/3 so "illegal addresses" won't be
> forwarded.
They are not illegal by any means - see below.
> This is an effective default route AFAIS, but might not be one if you go
> by the definition above.
That would be a very bad idea. What if IANA and the regional registries start
handing out addresses outside of this range tomorrow? Then that implementation
would not be able to talk to nodes that get addresses from the new range.
For this reason the text in <draft-ietf-ipngwg-addr-arch-v3-06.txt>
has been clarified to state
Future specifications may redefine one or more sub-ranges of the
global unicast space for other purposes, but unless and until that
happens, implementations must treat all addresses that do not start
with any of the above-listed prefixes as global unicast addresses.
Basically implementations shouldn't have any rules about any prefixes
other than
::0/128
::1/128
0::/3 (and longer prefixes for ipv4-mapped etc)
ff80::/10
fec0::/10
ff00::/8
Do we need to make this more clear?
Erik
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
