[Tinyos-devel] protocols and ActiveMessageAddress.changed

[Omprakash Gnawali]

A program has the ability to change the AM address any time. I don't
know if it makes sense to change the address after initialization.

The question is should the protocols that are distributed with T2
catch this event and reinitialize themselves. We don't do that right
now.

- om_p

On Sat, Mar 28, 2009 at 5:14 AM, Matt Welsh <mdw at eecs.harvard.edu> wrote:
> Why would an AM address change, anyway? Shouldn't most protocols treat this
> as the failure of one "node" and addition of a new "node"?
>
> On Mar 27, 2009, at 11:38 PM, Omprakash Gnawali wrote:
>
>> Should protocols such as CTP continue to work if the am address is
>> changed?
>>
>> What are the minimum conditions/events under which protocols should just
>> work?
>>
>> If that set of conditions is too large it might be too much burden.
>>
>> Here are some repercussions:
>>
>> FTSP - information who the root could be invalid. could force new
>> election.
>> CTP - have to flush the link/routing table. Other nodes might have to
>> wait for state expiration.
>> MultihopLQI - invalid parent information.
>> Drip - conflicts in <nodeid,seq>
>> etc.
>>
>> Should these protocols just boot up when ActiveMessageAddress.changed
>> is signaled?
>>
>> - om_p
>> _______________________________________________
>> Tinyos-devel mailing list
>> Tinyos-devel at millennium.berkeley.edu
>> https://www.millennium.berkeley.edu/cgi-bin/mailman/listinfo/tinyos-devel
>
>