List Info

Thread: using MH is not a magic "re-use"




using MH is not a magic "re-use"
user name
2006-11-14 14:57:41
Alex,

Static configuration of prefixes is only one way. Routing
protocols,
Prefix Delegation and other mechanisms are also possible, in
which case
the HA does not have prior knowledge of such things.

In any case, what is it that you are proposing exactly?

George

> -----Original Message-----
> From: Alexandru Petrescu [mailto:alexandru.petrescumotorola.com]
> Sent: Tuesday, November 14, 2006 9:45 AM
> To: Monami6 WG
> Cc: Soliman, Hesham
> Subject: Re: [Monami6] using MH is not a magic
"re-use"
> 
> Tsirtsis, George wrote:
> > Having missed the MONAMIv6 meeting I am trying to
decipher the
> > discussion on the mailing list only so I might be
missing some
> > context.
> >
> > The MR already does work on behalf of the LFN and
other nodes in
> > the mobile network. If this work is done based on
static
> > configuration then, the same can be done for
moving flows.
> 
> How would HA statically configure information about
applications
> running on LFNs?  HA does statically configure
'authorization'
> information about MR's HoA and MNP.
> 
> > It is obvious that if the MR does not know what to
do, it will not
> > do anything.
> 
> MR shouldn't do anything about LFN's IP address
(shouldn't include it
> in flow bindings) unless LFN or HA auhtorized MR to do
so.
> 
> Alex
> 
> 
> _______________________________________________
> Monami6 mailing list
> Monami6ietf.org
> https:
//www1.ietf.org/mailman/listinfo/monami6

_______________________________________________
Monami6 mailing list
Monami6ietf.org
https:
//www1.ietf.org/mailman/listinfo/monami6
using MH is not a magic "re-use"
user name
2006-11-14 15:27:20
Tsirtsis, George wrote:
> Alex,
> 
> Static configuration of prefixes is only one way.
Routing
> protocols, Prefix Delegation and other mechanisms are
also
> possible, in which case the HA does not have prior
knowledge of
> such things.

(Routing protocols do not distribute in any way prefixes to
be
assigned, while DHCP does.)

How do you see DHCP assign addresses and flow preferences to
LFN such
that MR is authorized to do flow bindings for LFN's IP
address?  MR
snooping the DHCP allocation?  MR allocating to LFN?  These
are
unknowns.  HEsham seemed to say out of scope - but without
it flow
bindings on MR about LFN's IP address are not possible.

> In any case, what is it that you are proposing exactly?

Without a proper mechanism for LFN to delegate authorization
to MR
about its preferences MR has no possibility to do flow
bindings for an
LFN - in which case one can't say that flow bindings are for
MR,
they're only for MH, no NEMO.

I don't think I'm proposing any other thing than keep flow
bindings in
the context of MH only, not for MR.

I certainly don't want the MR to make a decision on its own
about
which apps on which interfaces to use, when those apps run
on LFNs.
Its LFN who's in charge of its apps.

Alex


_______________________________________________
Monami6 mailing list
Monami6ietf.org
https:
//www1.ietf.org/mailman/listinfo/monami6
using MH is not a magic "re-use"
user name
2006-11-14 15:27:20
Tsirtsis, George wrote:
> Alex,
> 
> Static configuration of prefixes is only one way.
Routing
> protocols, Prefix Delegation and other mechanisms are
also
> possible, in which case the HA does not have prior
knowledge of
> such things.

(Routing protocols do not distribute in any way prefixes to
be
assigned, while DHCP does.)

How do you see DHCP assign addresses and flow preferences to
LFN such
that MR is authorized to do flow bindings for LFN's IP
address?  MR
snooping the DHCP allocation?  MR allocating to LFN?  These
are
unknowns.  HEsham seemed to say out of scope - but without
it flow
bindings on MR about LFN's IP address are not possible.

> In any case, what is it that you are proposing exactly?

Without a proper mechanism for LFN to delegate authorization
to MR
about its preferences MR has no possibility to do flow
bindings for an
LFN - in which case one can't say that flow bindings are for
MR,
they're only for MH, no NEMO.

I don't think I'm proposing any other thing than keep flow
bindings in
the context of MH only, not for MR.

I certainly don't want the MR to make a decision on its own
about
which apps on which interfaces to use, when those apps run
on LFNs.
Its LFN who's in charge of its apps.

Alex


_______________________________________________
Monami6 mailing list
Monami6ietf.org
https:
//www1.ietf.org/mailman/listinfo/monami6
[1-3]

about | contact  Other archives ( Real Estate discussion Medical topics )