Re: Middleboxes [Was: Flow label versus Extension header - protocol itself]

[marcelo bagnulo braun <marcelo@it.uc3m.es>]

El 09/05/2005, a las 19:54, Erik Nordmark escribió:

> marcelo bagnulo braun wrote:
>
> > well, i guess that addresses could be delivered through dhcp also.
> > I mean, the middle box delivers the HBA/CGA through dhcp to the hosts 
> > and performs proxy funtions for this hosts, since it knows the 
> > associated paramenters
>
> So the DHCP server would need to be part of the middlebox, and hand 
> back the ULID to the host, while maintaining the complete HBA/CGA set.
>
> But you need to given all the addresses to the host so that there are 
> redundant addresses available for it during initial communication, 
> don't you?

yes, unless you are sure that you are communicating with a shim enabled 
peer (the peer or a middlebox in front of it) in which case you could 
use shim capabilities even for dealing with outages in the locator used 
for initial contact

> If the DHCP server only hands back one address to the host, what would 
> happen when a shim6 capable host is used? It would want all the 
> addresses/prefixes so it could do shim6 on its own.

i guess that the DHCP server should know if the requesting host is shim 
capable or not

> Seems tricky to do this.

yes

imho middleboxes don't really fit nicely in the shim arch. I mean shim 
arch is e2e by design, not suited for middle boxes i guess.

any middlebox support will be some form of hack imho

I mean, if the middle box is the one that establishes the shim context, 
then i guess we can only have one middle box, right? what happens if an 
alternative path is used and another middle box routes the packets? i 
guess that there will be issues because the middle box don't have a 
state created for the context. So, or we associate one and only one 
middle box for node (in which case there is faye sharing between the 
host and middle box, i.e. reduced fault tolerance) or we create a inter 
middle box synchronization protocol (which seems at least complex)

So, i guess that we could support middle boxes, but with limited 
capabilites and some additional effort

Regards, marcelo

>    Erik