Re: AW: AW: [nemo] WG: new I-D on HA filtering

Hello Koojana,

At Thu, 24 Jul 2003 10:15:43 +0200,
Koojana Kuladinithi wrote:
> 
> Hi Ryuji, Thomas and all
> 
> > -----Ursprüngliche Nachricht-----
> > Von: Thomas Noel [mailto:[email protected]]
> > Gesendet: Mittwoch, 23. Juli 2003 19:04
> > An: Ryuji Wakikawa
> > Cc: Nicolas Montavont; [email protected]; [email protected];
> > [email protected]
> > Betreff: Re: AW: [nemo] WG: [mobile-ip] new I-D on HA filtering
> > 
> > Hi Ryuji and all,
> > 
> > 
> > > For example, user can configure filters to the database manually before
> > > communication. or an application may configure these by itself.
> > > If each interfaces is highly abstracted, anybody can configure
> > > filters to the database.
> > 
> > For this reason we are focused rather in how the filters are sent by the
> > MN
> > and managed by the HA.
> > 
> 
> I think since MIP is a network layer solution for mobility, shouldn't
> multiple points of attachment & flow distribution also reside at this
> level?. 

If we include flow distribution in multiple CoA support,
everybody must support this MIP oriented flow distribution. Why can't
we allow to use existing approaches such as COPS, PIB, etc.  Specially
mobile _ROUTER_ may want to use these. 

It is not good idea to force users to use only a MIP specific solution.

> What we propose in our draft is a simple extension to the data structures
> held for binding to hold flow distribution information. No such "database"
> is required as Ryuji mentioned.

Database is optional. It is required only when MN needs to divide
exact flow to multiple IFs. We did not explain anything about database
in the spec.  see my draft again.

We target on not only flow separation, but also redundancy, fault
tolerance. There are some cases when MN registers multiple
bindings as a backup binding.  In these case, MN doesn't need to have
any filters.

regards,
ryuji

> > >
> > > I am OK if there is a way to carry filters by MIP signaling, but we
> > > should allow to support the other ways. This is why I asked you to
> > > separate discussion.  I never ask you to stop discussing filter stuff:-)
> > >
> > 
> > 
> > What do you think by other ways ? Could you give us an example of other
> > way ?
> >
> 
> I too would like to know the other ways that Ryuji talks about.
>  
> Regards,
> 
> Koojana
> 
> > regards,
> > 
> > Thomas
> 
>