Fred,

Good questions.  But could cause gigantic rat-hole :--)

> At the risk of covering old ground, one question I had is 
> whether a client must wait to receive an RA before initiating 
> stateless or stateful DHCPv6? Asked another way, can DHCPv6 
> still be used if there are no advertising routers on the link?

To get prefixes yes, and to use them the A bit should be set.  But IPv6
is stateless in the node.  The node creates the link-local and then DAD
on the network.  Though I think shutting DAD off completely in some
environments is now prudent for some users special networks.  

> 
> To an even more speculative question, if we had it all to do 
> over again would it be possible (or desireable) to design 
> things such that DHCPv6 and IPv6 ND messages could be 
> "piggybacked" within a single packet? (This begs the question 
> of whether the DHCPv6 server/relay and IPv6 router entities 
> are co-resident on the same node in the normal case.)

Two questions really.

On piggy-back I don't think so for me.  Stateless is the win for IP from
the efforts for IPv6, and permeates its technical and operational
benefits for networks, DHCP is stateful and means some server function
exists that has state.  I support the model now and the authority plane
of stateless or stateful model via the A bit and then stating to use or
not use stateful.

As far as co-locating the functions that is an implementation choice and
we should not speculate on implementation of that nature between routers
vs. hosts. vs. servers. vs clients, when building a standard on this
topic.  Also can start an anti-trust problem :--).  So I say let the
market decide on that one.  I don't think its a good idea personally
speaking strictly as individual knowing both routers and hosts as IP
stack architecture and their purpose in my view.

I think its pretty simple and backwards compatibility for ND and
addrconf is drop the o bit and keep the m bit see my previous suggestion
on this sent a day or so ago.

thanks
/jim
  
> 
> Fred
> [EMAIL PROTECTED]    
> 
> "Bound, Jim" <[EMAIL PROTECTED]> wrote:
> > > It would be really nice and handy to initiate either stateless or
> > > stateful DHCPv6 with the same message. If so, we wouldn't need
> > > the M/O bits anymore. In this case the client would 
> simply initiate
> > > a(n Information) Request message and would get all the information
> > > that are available on the link, including an address or not.
> >
> > So you don't believe that the RA in ND should be the 
> authority to use a
> > stateful model on an IPv6 link?
> >
> > /jim
> 
> _______________________________________________
> dhcwg mailing list
> [email protected]
> https://www1.ietf.org/mailman/listinfo/dhcwg
> 

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[email protected]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------

Reply via email to