> Shall I expect a response, or is what Richard and I have posted
> enough to convince you that there are indeed some security benefits in
> using site-local address when designing a security perimeter in a
> private network?

I don't think you've made a convincing case for this.  

I generally agree that an SL address that somehow appears in
a network remote from the site in which it was used is not likely 
to be able to make it back to that site via the network's normal
routing mechanisms.  Some sort of tunnel is likely to be necessary.

On the other hand there are far more opportunities to filter an 
address that does have a global prefix (you're not limited to filtering
them at the border of the "site"), or to detect the use of such addresses 
on networks where they are supposed to be filtered, or to log such 
addresses for traffic analysis.  

I also don't think you've made a convincing case that a single, 
non-permeable security perimeter is a very useful feature of an
addressing architecture.

I do think you've made a case for using SLs in isolated networks, 
or nearly isolated networks where the only interactions between
those networks and other networks are via special-purpose 
applications.  As I said earlier I don't have a problem with using
SLs in that corner case.  I do have a problem with promoting SLs
as a security mechanism for use in general-purpose network where
off-the-shelf apps are exposed to a mixture of SLs and globals.

Keith
--------------------------------------------------------------------
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