RE: [DNA] Model: how to treat "link down" events

[Mohan.Parthasarathy@nokia.com]

>
>> Is the following scenario possible ?
>>
>> 1) Host receives a link down event. Does not do anything.
>> 2) Host attaches to the new link. Link UP event happens
>>   slightly late or the event has not propagated completely
>>   through the system. Meanwhile you are able to send and
>>   receive packets.
>
>I have difficulty understanding above. How can the host send/ 
>receive packets when it is attached to a new link in which it 
>has no IP configuration information such as IP address?
>
When a host attaches to the new link, as per the solution
document, the host sends a RS with TSLLAO option to determine
whether it has changed links or not. The source address of the
RS is a link-local address which was previously in preferred
state but now in optimistic state. If it knows that it moved
to a new link, it discards old information and starts
configuring new addresses from the RA. In the above scenario,
as the host has not seen the Link UP trigger yet, it does
not send the RS, does not set the addresses to Optimistic
and just sitting there thinking it is old link. If another
node attached to the same link with the same link-local
address, both the nodes will respond to the NS and override
flag will not be cleared in both of the NAs.
Eventually right things will happen, when the RS goes out,
but in a small window, there can be this problem. This is
not a opti-DAD problem, because you send NA with override
flag cleared when you are in optimistic mode.

It looked to me that if you go to optimistic on seeing
Link Down (i understand that it is not reliable etc..), then
this problem (assuming it is one) may not happen.

-mohan

>Thanks for your kind consideration.
>
>Best Regards
>
>JinHyeock
>