Release 5.2 is now available

[Bob Fink]

Pedro,

At 9:40 AM -0700 9/2/97, Pedro Marques wrote:
>>>>>> "Bob" == Bob Fink <RLFink@lbl.gov> writes:
>
>    Bob> Pedro, At 10:35 AM -0700 9/1/97, Pedro Marques wrote: ...
>    >> The new addressing scheme has no relation at all with the low
>    >> 64bits of the address.
>    >>
>    >> So regardless of the changes in addressing you can still use
>    >> 48-bit 802.3 link tokens.
>
>    Bob> As I read the Aggregatable Global Unicast Address Format I-D
>
>    Bob>
>ftp://ftp.ietf.org/internet-drafts/draft-ietf-ipngwg-unicast-aggr-02.txt
>
>    Bob> the new Interface ID is part of the new addressing scheme.
>
>    Bob> Thus we should be testing the new prefix format AND the new
>    Bob> EUI-64 at the same time.
>
>Why ? Do you have any reason at all to tight the two things together ?
>They are two completely separate mechanisms for completly distinct problems.
>
>Further, operationally it makes no sense to me to try to connect the two
>issues... the 6bone is an interconection of ipv6 sites, it should not concern
>itself at all with autoconfiguration of hosts in a LAN.

Because we are testing the real implementations on both LAN and WAN.  And,
of course, because that is the standard (i.e., I-D) that we agreed to
implement and test. You had said that the new addressing scheme has no
relation to the low 64 bits, but it does in the spec.

Now I must admit I don't know what will break (if anything) if we don't
implement EUI-64 at the same time as the aggregatable prefix, but we should
try to test the real spec as much as possible.


Thanks,

Bob