Alexandru Petrescu wrote:
>
> > Appendix A of the draft, uses text from 3GPP TS 23.060, Clause 9.2.1.1
> > without any criticism.
>
> The funniest thing of the 3GPP "stateless autoconf" is that GGSN
> provides the "unique interface identifier" (an equivalent of an EUI-64)
> to the mobile as if the mobile would not know its own unique ID.
The reason for this, AFAIK, is to get rid of DAD on the GGSN side. If
the GGSN hands out IDs from a pool it can easily keep control of what
IDs are unused and DAD becomes a simple task.
On the other hand, if the GGSN was to do real DAD, it would have to scan
a probably *huge* address space to detect collisions, as the GGSN is the
only node that knows all the addresses used since the terminals never
talk directly to each other.
/Mattias
>
> OTOH, I can understand this if 3GPP considers there'll be a sort
> of "tunnel" between mobile and GGSN, like in PPP. But in that
> case they just can't state this is stateless autoconf.
>
> Alex
> --------------------------------------------------------------------
> 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]
> --------------------------------------------------------------------
--------------------------------------------------------------------
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]
--------------------------------------------------------------------
