Re: New Version Notification for draft-hui-6man-rpl-option-00.txt

Hemant,

Being able to change the option data hop-by-hop is fundamental to  
RPL's ability to verify routing information consistency.  That is why  
Section 2 of the rpl-option draft states that "the RPL option is  
expected to change en-route."

The Router Alert Option as specified today does not satisfy this  
fundamental requirement.  Section 2.1 of RFC 2711 states that "the  
option must not change en route."

--
Jonathan Hui

On May 30, 2010, at 1:46 PM, Hemant Singh (shemant) wrote:

> If RFC 2711 already defines a Router Alert Option to use, why do we  
> need
> a new option in the Hop by Hop option for roll as specified by this
> document?  I suggest we just use a new code value for roll with RFC  
> 2711
> and be done with it.  See section 2.1 of RFC 2711.  Here is a list of
> what IPv6 protocols already use the Router Alert Option.
>
> http://www.iana.org/assignments/ipv6-routeralert-values/ipv6-routeralert
> -values.xhtml
>
> RFC 2711 also already lists the security concerns with use of the  
> Router
> Alert Option.  This document is repeating the same security concern.
> Further one should look at
>
> http://tools.ietf.org/html/draft-rahman-rtg-router-alert-considerations-
> 03
>
>
> Hemant
>
> -----Original Message-----
> From: ipv6-bounces@ietf.org [mailto:ipv6-bounces@ietf.org] On Behalf  
> Of
> Philip Levis
> Sent: Saturday, May 29, 2010 1:08 PM
> To: JP Vasseur
> Cc: ipv6@ietf.org
> Subject: Re: New Version Notification for
> draft-hui-6man-rpl-option-00.txt
>
>
> On May 17, 2010, at 6:18 AM, JP Vasseur wrote:
>
>> Dear all,
>>
>> This is to re-activate our discussion during the last IETF
>> meetings. Feed-back are very welcome since we would like to move
>> forward as soon as possible, this proposed extension is indeed
>> critical for RPL to move forward.
>>
>> Many Thanks.
>>
>> JP and Jonathan.
>
> I strongly support this draft; it is a make-or-break extension,
> without which RPL will not be an effective protocol. There is
> extensive experimental evidence from many LLN deployments and
> testbeds that the mechanisms it enables (datapath validation) are
> critical.
>
> I think Jonathan has done an excellent job of narrowly defining what
> the option can do: containing it within a RPL domain addresses most
> of the major concerns with hop-by-hop options.
>
> Phil
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> ipv6@ietf.org
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> ipv6@ietf.org
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------