Re: address pair exploration, flooding and state loss

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

Iljitsch van Beijnum wrote:
> On 31-mei-2005, at 19:21, Erik Nordmark wrote:
>
> > > - after a rehoming event, context loss is detected upon the  
> > > reception of any packet associated with the shim session, whether  
> > > signalling packet or data packet. for that, data packets need to  
> > > carry at least one bit that identifies them as belonging to a shim  
> > > session.
>
>
> > Makes sense to me.
>
>
> It is indeed necessary to determine that a higher layer session or  
> association belongs to a shim session. However:
>
> - That doesn't necessarily mean that this can be determined for an  
> individual packet. Another way to do this would be the "ICMP bad  
> checksum" message that I suggested a few days ago.

But as I responded, such an approach only makes sense if we are going to 
modify all the protocols which run over IP to generate such errors. And 
assuming that an implementation can do this is problematic due to "raw 
socket" APIs which allow there to be any number of IP protocols 
implemented unbeknown to the provider of the IP and shim6 layers.

> - Even if we want to determine the shim status of individual packets,  
> we don't necessarily need a bit for this. For instance, when locator- 
> only addresses and ULID addresses are disjunct this can be determined  
> from the destination address.

Correct. The above "bit" is really a conceptual piece of information 
that needs to be sent from A to B; could be in the form of
 - using a separate extension header
 - some bit stealing in the nxthdr field
 - using locators which are not allowed as ULIDs as you point out
 - or anything else which conveys the meaning

   Erik