Quoting Romain KUNTZ <[EMAIL PROTECTED]>:

> Hi,
>
> On 2007/05/30, at 15:41, VITOR ANDRÉ CORDEIRO SANTOS wrote:
>> To solve this problem the solution I found with Romain Kuntz was to  use the
>> OptimisticHandoff option in the HA and MR configuration. With this  
>> option the
>> MR, when it moves, doesn't have to wait for the HA BAck, and opens  
>> right away
>> the tunnel endpoint, being able to route traffic through the tunnel  to and
>> from the HA.
>
> OptimisticHandoff is clearly not the correct solution to your  
> problems. Even though you'll be able to route the traffic, the MR  
> stills think that its BU was not acknowledged and will not behave  
> correctly (eg: it will try to resend BU continuously, and will  
> certainly stop at some point).
>
> OptimisticHandoff was implemented to reduce the handover time by  
> allowing the MR to route packet before the Ack is received. Do not  
> use it as a workaround to your problems, as you'll certainly get  
> other issues in the future due to the initial problem.
>
> Your problems could be related to the mail sent rencently by Masahide 
>  Nakamura on the MIPL ML. Please check it:
> http://www.mobile-ipv6.org/pipermail/mipl/2007-May/006853.html
> I hope this will solve your issues.


We haven't a lot of time to make NEPL works, that's why we will use the 
OptimisticHandoff first.
The problem given on the above thread looks similar to our, but we 
don't understand everything and waiting to know the version of glibcc 
needed.

Thanks for your help.



-- 
Patrice Duviel
I. U. T. du Limousin



_______________________________________________
mipl mailing list
[email protected]
http://www.mobile-ipv6.org/cgi-bin/mailman/listinfo/mipl

Reply via email to