Jinmei-san, thanks for your comments.
Some replies below.
> My fundamental question here is why such closed networks need global
> IPv6 addresses (or even any IPv6 addresses) in the first place.
> Whether it's a VPN site or a closed ASP, doesn't it suffice to simply
> use private IPv4 addresses? If I were a network administrator, who is
> inherently conservative, I'd not take a risk of introducing
> possibly-unstable/immatured IPv6 equipments for purposes in which I
> don't need global connectivity.
>
> Even if I were to choose to use IPv6 for some reason, it seems I'd be
> just happy with ULAs for such purposes, and then the (revised) default
> address selection algorithm seems to suffice (and so we don't need the
> automatic configuration mechanism). I didn't understand why the
> default algorithm doesn't work here after re-reading the past message
> you attached and re-reading Stig's message. Perhaps I simply
> misunderstood the point - so could you provide a concrete example
> explaining why the default rule doesn't suffice there?
>
> BTW: I'm not necessarily against the proposal per se. I was just not
> convinced about the applicability of this mechanism *in practice*.
> So, if a majority of this wg believes we don't need convincing
> practical scenarios to adopt this proposal, I'll simply shut up.
> But if others also want such evidence, what I'd seek is:
>
> - realistic and convincing usage examples where the non default
> address selection rule is necessary. depending on the answers to
> the following points, I guess the VPN or closed ASP examples will
> suffice.
> - practical reasons why the closed network needs IPv6 to begin with.
> I've not been convinced on this point yet at all.
Do you recommend to use IPv4 addresses forever ?
IPv4 private addresses has another kind of problem even
for closed network, such as address collision.
I guess its very common to use private address in home
network, so its very likely to collide ASP's private address
and that in home network.
Another point is that when you start a new network service,
you want to use long-life technologies and not dying ones.
> - (on top of that) concrete examples showing why the (revised) default
> address selection rule doesn't work if we use ULAs for the closed
> network. (I'm probably seeking this just due to my
> misunderstanding.)
In my last e-mail, I meant that when one ASP uses more than
one ULAs or has peering with another ASP(which has ULA), a
consumer site cannot choose a correct source address by
default address selection rule.
This looks like the following.
+---+ULA1
|ASP|
+---+
|
| ULA2 ULA3
+---+ +---+
|ASP| |ASP|
+---+ +---+
| |
---------
|
+---+
|GW |
+---+
|
-----------
|
+---+
| PC|
+---+
As ULA is such a small address space(/48), it is very probable
that an ASP has multiple ULA blocks.
Additionally,
at v6ops session yesterday, 6net people showed us another
possible usage of address selection policy table. It's renumbering.
By pouring address selection policy into each end hosts,
you can easily configure address selection policy that
makes end-hosts not to use old address.
--
Arifumi Matsumoto <[EMAIL PROTECTED]>
Ubiquitous Computing Project
NTT Information Sharing Platform Laboratories
TEL: +81-422-59-3334 / FAX: +81-422-59-5652
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
