Re: [Autoconf] Closing summary on consensus-call for RFC5889modifications

First note that OLSR is just an example. As a further example,
I haven't checked if AODV and/or DYMO have documented equivalent
features, but they very easily could implement them (don't
forward RREQs).

Of course if every router (or just too many routers) is selfish
the ad hoc network will fail. That's inherent in the whole concept
of an ad hoc network. OLSRv2 points this out when allowing the
concept. But this is not an autoconf or RFC 5889-to-be problem.

I don't see that a node being able to set its own willingness
to zero in any way adds to the threat. A hostile node is more
dangerous with high willingness than with low. But that's not
an authconf/RFC 5889-to-be issue, it's a routing security
problem.

Again, your last point is about routing and management. That's
not to say it and the others aren't good questions when planning
an ad hoc network (not quite an oxymoron) or for research. But
they aren't the issue here.

-- 
Christopher Dearlove
Technology Leader, Communications Group
Communications and Networks Capability
BAE Systems Advanced Technology Centre
West Hanningfield Road, Great Baddow, Chelmsford, CM2 8HN, UK
Tel: +44 1245 242194  Fax: +44 1245 242124

BAE Systems (Operations) Limited
Registered Office: Warwick House, PO Box 87,
Farnborough Aerospace Centre, Farnborough, Hants, GU14 6YU, UK
Registered in England & Wales No: 1996687

-----Original Message-----
From: reshmi r [mailto:reshmi.engg@gmail.com] 
Sent: 24 August 2010 11:08
To: Dearlove, Christopher (UK)
Cc: autoconf@ietf.org
Subject: Re: [Autoconf] Closing summary on consensus-call for RFC5889modifications

                    *** WARNING ***

  This message has originated outside your organisation,
  either from an external partner or the Global Internet. 
      Keep this in mind if you answer this message.

Hi Christopher,

Thank you for your immediate reply.

Hi All,

But i still have some questions to get it clarified.

I understood all the nodes has the capacity to act as router and some
may be ready to do it.

1. The nodes themself has the capacity to make it act as router(by
setting WILLINGNESS to non zero). But in most of the cases all nodes
are selfish and they does not want to parcipate in routing. So almost
every node will be busy in their host functions alone. so who will
take care of autoconfiguration? Imagine a network with no router!!!!!!
2. There will be more security threats if the node it self has the
capacity to configure  its  WILLINGNESS . The malicious node can
easily get in and set the WILLINGNESS to non zero and participate in
routing functions.
3. If suppose we have a method to select the hosts which can act as
router and set WILLINGNESS manually. What will be the criteria for
router selection??? Is it based on resources, traffic etc etc..

Once again thank you for your reply.

Regards,
Reshmi.T.R.
Reseach Scholar,
Ramanujan Computing center,
Anna University- Chennai
India

  Tue, Aug 24, 2010 at 2:25 PM, Dearlove, Christopher (UK)
<Chris.Dearlove@baesystems.com> wrote
> I think there's a misunderstanding here. All the way back to RFC 2501,
> a node has been defined as a router plus possible hosts. If you've got
> a wireless device that wants to participate in a MANET it needs to be
> running an ad hoc routing protocol, i.e. it's a router. It may perform
> only a limited subset of routing functions - consider for example an
> OLSR node that does not wish to be a relay, only an endpoint. It can
> do that by setting WILLINGNESS equal to zero, and if it is built only
> to take such a role it can then throw away large chunks of OLSR code
> (for example it never sends TC messages). But the node still has some
> router functions. This is the model of both RFC 2501 and 5889-to-be.
> The host can then get its addresses in any non-MANET-specific way on
> that node.
>
> --
> Christopher Dearlove
> Technology Leader, Communications Group
> Communications and Networks Capability
> BAE Systems Advanced Technology Centre
> West Hanningfield Road, Great Baddow, Chelmsford, CM2 8HN, UK
> Tel: +44 1245 242194  Fax: +44 1245 242124
>
> BAE Systems (Operations) Limited
> Registered Office: Warwick House, PO Box 87,
> Farnborough Aerospace Centre, Farnborough, Hants, GU14 6YU, UK
> Registered in England & Wales No: 1996687
>
> -----Original Message-----
> From: autoconf-bounces@ietf.org [mailto:autoconf-bounces@ietf.org] On
> Behalf Of reshmi r
> Sent: 24 August 2010 05:45
> To: autoconf@ietf.org
> Subject: Re: [Autoconf] Closing summary on consensus-call for
> RFC5889modifications
>
>
>                    *** WARNING ***
>
>  This message has originated outside your organisation,
>  either from an external partner or the Global Internet.
>      Keep this in mind if you answer this message.
>
>
> Hi Teco,
>
> Do they really mean a model for the autoconfiguration and is there is
> no role for host in autoconfiguration??........the topic goes in to
> real debate. what was the final outcome of the discussion???.
>
> Hi All,
>
> Can anyone finalise the suggested outcomes of the discussion??Do you
> all really mean that the routers only need to do the autoconfiguration
> and the nodes have no role in it??? If so how can we believe a router
> to be genuine and how can we ensure that the router will never become
> selfish??? so there should be some role in hosts to monitor the
> traffic behaviour of router and the host should be able to notify with
> some protocol mechanism. do you all really mean to change the title???
> I strongly disagree with this.
>
>
> Rgds,
> Reshmi.
>
>
> Hi Thomas,
>
> On the title change, I remember in Maastricht all accept one
> preferred the title change. On the list as well.
> There are two arguments.
> 1) it is _a_ model
> 2) the model doesn't support hosts, or at least not very well
> On the latter, there was a discussion without outcome.
>
> Regards, Teco
> _______________________________________________
> Autoconf mailing list
> Autoconf@ietf.org
> https://www.ietf.org/mailman/listinfo/autoconf
>
>
> ********************************************************************
> This email and any attachments are confidential to the intended
> recipient and may also be privileged. If you are not the intended
> recipient please delete it from your system and notify the sender.
> You should not copy it or use it for any purpose nor disclose or
> distribute its contents to any other person.
> ********************************************************************
>
>