Re: ND-proxy applicability in Unmanaged [Re: WG Last Call: draft-ietf-v6ops-unmaneval-01.txt]

[Erik Nordmark <Erik.Nordmark@sun.com>]

> > Whether using stateless or stateful address configuration,
> > the ISP can send RA's without any on-link prefixes to the customer's box.
> > This means no prefix is made available to the pt-pt link between
> > the ISP's router and the customer's box. Hence the customer's box
> > can only use the configured IPv6 address.
> 
> But how does the customer's box get it's own (global) addess?  Do you 
> assume you run something like PPPv6 on the link, which assigns one 
> address and that's it?

That isn't the only choice. DHCPv6 and stateless address autoconfiguration
works as well. (There is a reason the prefix options have
a separate "on-link" flag and "addrconf" flag.)

> SEND + CGA helps a bit in the local link, between the nodes; 
> ND-proxy does not prevent that.  

AFAIK ND-proxy is incompatible with using SEND. full stop.
You can't do any proxy advertisements transparently to the hosts and
still have SEND work.

> > Could you specify the problem statement that covers "simpler setups"
> > but where  loops are somehow impossible to create by the consumer?
> 
> We are not preventing the customer from shooting him/herself in the
> foot in many other specifications either -- why is this relevant here?  

Could you please point me at an IETF standard related to routing which
can create persistent routing loops where ttl is not decremented?
I never recall seeing such a beast. Folks in the routing area seem
to have a aversion to creating persistent routing loops; even though
temporary loops occur at L3 and the damanage is limited due to
the ttl decrement.
But ndproxy explicitly doesn't decrement the ttl/hop limit!
That is why I am extremely concerned here.

  Erik