Re: [pcp] Comparison of PCP authentication

[Alper Yegin <[e-mail]>]

Sam said:

> So you are happy to decide PCP authentication doesn't need a PANA relay?

I really didn't understand when this "split EAP authenticator from PCP server" case came into the picture.
Unless someone has a real use case behind it, we can ignore that case.

Yoshi said:

> I am ok without supporting PANA relay for PCP authentication.
> 
> It also makes key management easier because remote transport of PCP
> key from PAA to PCP server is needed if PANA relay is supported for
> PCP authentication.
> 


Well, such a thing would have been needed irrespective of what carried EAP (PCP itself or PANA) if we were to deal with the split case.

Alper


> Yoshihiro Ohba
> 
> (2012/08/17 22:07), Sam Hartman wrote:
>> 
>>     Alper> EAP Authenticator is on the PCP server.  Hence, PAA (PANA
>>     Alper> Authentication Agent) and PCP server are on the same node.
>>     Alper> Therefore, the PAA can tell whether it's authenticating the
>>     Alper> PaC for PCP or for network access by looking at the
>>     Alper> destination port.  That's sufficient.
>> 
>> So you are happy to decide PCP authentication doesn't need a PANA relay?
>> If so, I propose we explicitly decide that.
>> 
>> It makes my channel binding question easier.
>> 
>