Re: [DNA] Issue about multiply connected to the same link and DNAbeing per-interface or not (was Re: WG LC comments ondraft-ietf-dna-goals-01.txt)

[JinHyeock Choi <jinchoe@gmail.com>]

Pekka

We are in a disagreement. 

> > DNA scheme is run per-interface because some IP configuration,
> > such as IPv6 address, is interface specific.
> 
> Well, I do see there some potentiality for using DNA related
> information across different interfaces.  Hence, I would _not_
> like to mandate DNA to be per-interface.
> 
> For example, if a host has earlier been connected twice to the
> same link through two different interfaces, and then it detects
> that one of the interfaces has undergone a link change, IMHO that
> is a strong indication that there may have been a link change
> at the other interface, too.

'mandate' may be a too strong a word but, I think, by default DNA 
will be run per-interface. Each interface should detect the identity 
of its currently attached link for itself. 

This doesn't mean each interface can't share DNA related information
among themselves. 

Even in the above example, the fact that one interface has undergone 
a link change doesn't automatically confirm that the other interface 
also has undergone a link change. As you wrote, though strong, that 
is just a hint, so the other interface needs to run DNA scheme to 
check for link change for itself.   

Moreover, in case a host has multiple interfaces, it's more likely 
each interface is of different link-layer technology. For example, 
a host would have a 802.11 interface and 802.16 interface rather than 
two 802.11 interfaces. In that case, what happened at one interface 
doesn't tell much about the status of the other interface.  

> <snip>
> 
> > However, when multiple interfaces of a host are connected to the
> > same link, it will result in few problem. Each interface may run DNA
> > scheme to check for link change and update its IP configuration in
> > case of a link change.
> 
> I agree that there shouldn't be that many new *problems* related
> to DNA.  However, I think it may be possible to use some DNA
> related information for *optimising* behaviour.  We have to
> remember that DNA is mostly about optimisation, anyway.

I agree that they may be the optimization possibility you pointed. 
However, it's not clear to me how we will reflect that in Goals.   

> > I think the expression 'multiple interfaces' in G10 is confusing.
> > I propose to change G 10 as
> >
> >    G10 Hosts in particular wireless environments may perceive routers
> >       reachable on different links.  DNA schemes should take into
> >       consideration the case where a host is attached to more than one
> >       link at the same time.
> 
> This is better.

ok. 

> > Also, as Alper pointed, we'd better explicitly mention that DNA is
> > run per-interface. I added the following in Introduction.
> >
> >    ..........................of its IP configuration, is called
> > Detecting Network
> >    Attachment (DNA).
> >
> >    DNA scheme is run per interface because some IP configuration, such
> >    as IPv6 address, is interface specific.  When a host has multiple
> >    interfaces, each interface separately check for link change to
> > verify
> >    each one's IP configuration.
> 
> As I wrote above, I don't agree with this.  I think this is unnecessary
> in the goals document.  I don't think we should include it.
> OTOH, if other's in the WG think otherwise, I won't insist.

Let me summarize my thoughts. 

In case there are multiple interfaces in a host, each interface should 
separately detect the identity of the currently attached link for itself. 

Do you disagree with this? If so, we are in a serious disagreement 
and would better find a way to resolve this. 

Best Regards

JinHyeock