Alain Durand wrote:
> 
> 100% agreed. Still, we have to pick a default.
> My point is that I'd rather like the default case to be that apps that
> require global addresses don't have anything special to do to work.

I prefer the behaviour specified by default address selection - pick the
smallest 'scope' which matches unless the application has a good reason to
do otherwise.

Stability aside, the 'smallest matching scope' rule only fails for
applications that forward addresses out-of-scope.  In this case, either:

- forward names, so that you get all in-scope options, not just the one that
worked for the forwarder.

- accept that your application is now delving into the routing space and
code accordingly.


The other way of looking at it is to observe that 'local' addresses are
proposed as a solution for when 'global' addresses are unavailable,
unstable, or otherwise untrusted.  In other words, if local addresses exist
the general assumption should be that the network wants applications to
prefer them over global ones.  I cannot think of an example for a
non-forwarding connection where this would not be true.


Note: 'prefer' is qualified by the source address selection rules, which
will (among other things) prefer (in order):

- matching scopes over non-matching
- smaller scopes over larger
- longest prefix match

In a multi-addressing environment, applications must be ready to try
multiple source-destination address pairs until one succeeds.  A source
address selection compliant gethostbyname will return the destinations in
preferred order (taking into account available source addresses).  There is
an implementation question on what order the two dimensional destination and
source space should be searched.

-- 
Andrew White                [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]
--------------------------------------------------------------------

Reply via email to