> On 05 Nov 2015, at 14:01, Rafa Marin Lopez <[email protected]> wrote:
> 
> What I meant is that for security reasons we will have to refresh the keying 
> material. For example, EAP Key Management Framework recommends 8 hours as 
> default lifetime (this might be modified in this type of networks). Thus, I 
> do not think the source code will be used only once. 

Ok, I see what you mean. So, yes, if we go for the EAP approach, we would be 
using it for rekeying as well.  The problem I have there is that we will in any 
case need to use DTLS, and even maybe the object security mechanism ACE comes 
up with for application session security so I don’t think this would be optimal 
from code point of view. 


>> 
>>> EAP-AKA has two messages + EAP Success : EAP-Req/AKA (4 bytes header EAP + 
>>> 1 the type) +  EAP-Resp/AKA (4 bytes header EAP +1 the type) + EAP 
>>> success(4 bytes) = 14 overhead with respect to running “AKA" without EAP. 
>> 
>> Ok thanks. So, am I right if I say that this will trigger 3 additional 
>> frames at L2 with additional CoAP / UDP-6LoWPAN-IPv6 / 15.4 overhead?
> 
> Not sure about what you mean with 3 “additional" frames at L2. Additional to 
> what else?. You also mention additional CoAP/UDP...? Are you assuming 
> CoAP-EAP?

I was referring to the 14-byte overhead you referenced above in case we use 
CoAP for EAP transport. Is my understanding wrong? 

_______________________________________________
6tisch mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/6tisch

Reply via email to