Re: [DNA] Couple of points on draft-jinchoi-dna-soln-frame-00.txt

[JinHyeock Choi <athene@sait.samsung.co.kr>]

Dear Alper

Thanks for your feedback. 

Kindly fine my in line comments. 

> Few comments on this I-D...
> 
>    If a link change has occurred, a host assumes that its IP
>    configuration is no longer valid.  It needs new default router and IP
>    address.  If it remains at the same link, a host assumes its IP
>    configuration is still valid.
> 
> I think the delay between the time the host last used the particular
> configuration and the time it thinks the same applies to the newly
> attached link should be taken into consideration. This time is
> relatively short when we are simply comparing the pre-handover link with
> the post-handover link. But given that a configuration has a lifetime
> (based on prefix lifetime, or DHCP lease time), the same mechanism could
> be applied when the delay is higher but still bound by the lifetime.
> This might be because the host has visited some other links in between. 
> 
> When we do that, there is a problem. During the host's absence, some
> other host might have acquired the same IPv6 address. This is possible
> with stateless address autoconfiguration [low probability, can we ignore
> it?]. In theory, this should not be possible with the DHCPv6 because the
> address is reserved for the host during lease time, but I heard there
> are DHCPv4 products out there that do not observe this [maybe this would
> not be a problem for IPv6].  
> 
> Anyways, my point is that there are some additional considerations if we
> want to say I can use the same configuration on the same link as long as
> it hasn't expired.

You are right. We need to take into consideration the time factor. 
In fact originally the phrase was like below 

     address.  If it remains at the same link, a host assumes its IP 
     configuration is still valid (upon DAD).

But we suspected it might give the wrong impression that there would be 
always an extra DAD and NUD even when staying attached to the same 
link.  

In case that a host finds out that it is attached to the same link to which 
it was attached a few minutes ago, the followings need to be clarified.   
 - that the router lifetime should be honored
 - that the prefix lifetimes should be honored
 - that NUD would probably happen since the NCE is probably stale
 - it may or may not be important to redo DAD

>    Either an AP (which implements [14]) immediately sends an RA to the
>    host, or the host sends an RS to all-routers and one or more routers
>    on the link responds to the RS with an RA.
> 
> The other possibility is that the access router sends an unsolicited RA
> to the host upon detecting the new attachment. For that the access
> router needs to receive the link up, and that might require some
> additional protocol work when L2 access point is not co-located with the
> access router(s).

I see. 
 
>   2.  We assign each link a locally unique Link Identifier.  'Locally
>       unique' means, no two adjacent links have the same Link
>       Identifier.
> 
> This would have issues if the hosts ever make a long jump (skipping some
> intermediate links) for any reason.

The term 'adjacent' needs further contemplation. That's one of why I wrote that
there are complexities related to 'local uniqueness'. We may make it more clear
when we discuss the idea of 'Link Identifer'.  

Best Regards

JinHyeock