Re: [Autoconf] updated draft on aspects of multi-hop wireless communication

["Teco Boot" <[e-mail]>]

Inline.

|-----Oorspronkelijk bericht-----
|Van: [e-mail] [mailto:[e-mail]] Namens
|Thomas Heide Clausen
|Verzonden: dinsdag 24 februari 2009 0:21
|Aan: Charles E. Perkins
|CC: [e-mail]; Emmanuel Baccelli
|Onderwerp: Re: [Autoconf] updated draft on aspects of multi-hop wireless
|communication
|
|Hello Alex,
|
|To second Charlie here: in MANETs, routers:
|
|	o	ALWAYS have one MANET interface.
|	o	OFTEN they have other (non-MANET) interfaces
|	o	OCCASIONALLY have more than one MANET interface

I think the definition is still not clear.
Is it a requirement for a MANET Router to forward packets?
What is a MANET interface ? (I think I know, but I refer to the endless
discussion in [Autoconf])
Is the MANET a complete set of nodes, including attached hosts and
"classical routers"?

I think the term MANET is quite abstract and attempts to come up with strict
definitions tend to fail. For specifying protocols, we can come up with
strict definitions. We could say:
An OLSR Router:
 o   ALWAYS have one OLSR interface;
 o   OFTEN they have other (non-OLSR) interfaces;
 o   OCCASIONALLY have more than one OLSR interface.



|I.e. it is quite common for traffic to be forwarded over the same
|interface as the one on which it was received.

The attribute that traffic may be forwarded over the same interface as the
one on which it was received is not specific to MANET at all. Think of WLAN
AP, DSL/cable concentrators, FR/ATM/ISDN, dual-hop satcom etc. Sometimes we
use logical interfaces for links, sometimes we don't. Using logical
interfaces introduce problems utilizing layer-2 broadcast for IP broadcast /
multicast.
Please note that I do not suggest having detailed discussions on layer-2
technologies itself.


Teco.





|Thomas
|
|ps: and for the record, AFAIK all MANET wg routing protocols support
|all the situations listed above, i.e. a MANET router with a single or
|with multiple MANET interfaces and with or without other non-MANET
|interfaces.
|
|On 23Feb , 2009, at 23:01 , Charles E. Perkins wrote:
|
|>
|> Hello Alex,
|>
|> On this point I disagree.  I think it is better to show the
|> routers with a single network interface -- which is what
|> they have in real life.  This is also the reason that techniques
|> like split horizon aren't appropriate.
|>
|> Regards,
|> Charlie P.
|>
|>
|> Alexandru Petrescu wrote:
|>> Emmanuel, AUTOCONFers,
|>>
|>> Just a short clarifying point: I think it would be clarifying to
|>> illustrate the routers to have two interfaces.  Routers with only one
|>> interface aren't very typical.
|>>
|>> Just a note,
|>>
|>> Alex
|>>
|>> Emmanuel Baccelli a écrit :
|>>> Hi all,
|>>>
|>>>
|>>> following the fruitful discussions about initial version of the
|>>> document, here is an update to the draft describing aspects of
|>>> multi-hop wireless communication:
|>>>
|>>> http://www.ietf.org/internet-drafts/draft-baccelli-multi-hop-
|>>> wireless-communication-01.txt
|>>>
|>>>
|>>>
|>>> Again, the goal of this document is to identify a consensus about
|>>> this topic, and then use this to move on quicker with the rest of
|>>> the
|>>> work...
|>>>
|>>> Please review it, and provide feedback as soon as possible.
|>>>
|>>>
|>>> Cheers
|>>>
|>>> Emmanuel
|>>>
|>>>
|>>> --------------------------------------------------------------------
|>>> ----
|>>>
|>>>
|>>> _______________________________________________ Autoconf mailing
|>>> list
|>>>  [e-mail] https://www.ietf.org/mailman/listinfo/autoconf
|>>
|>> _______________________________________________
|>> Autoconf mailing list
|>> [e-mail]
|>> https://www.ietf.org/mailman/listinfo/autoconf
|>>
|>>
|>
|> _______________________________________________
|> Autoconf mailing list
|> [e-mail]
|> https://www.ietf.org/mailman/listinfo/autoconf
|
|_______________________________________________
|Autoconf mailing list
|[e-mail]
|https://www.ietf.org/mailman/listinfo/autoconf