eap Mailing List: Re: Identifier Wrap-around

[david [at] mitton.com (david]

Indeed.
You may even be suprised to discover that some tunneling protocols 
don't even use the same ID on one side of the tunnel that was supplied 
on the other side.

I discovered this issue when we attempted a crypto binding including 
the ID and it failed in certain circumstances.

Dave.

----Original Message----
From: Pasi.Eronen [at] nokia.com
Date: Jun 8, 2007 7:13 
To: <mahesh.dantakale [at] hotmail.com>, <eap [at] frascone.com>
Subj: Re: [eap] Identifier Wrap-around

Mahesh Dantakale wrote:
> 
> The Identifier is one ocetet field and unique for each EAP-Request.
> What happens, if the number of Requests exceeds 255 in an
> EAP-Conversation?
> 
> Will the Identifier wrap-aroud to zero after 255?  If yes, then
> wouldnt EAP regard this as duplicate packet, as the Identifier field
> in new Request packet is less than the previous Request packet ?

The identifier doesn't have to be unique for each EAP-Request:

   In order to avoid confusion between new Requests and
   retransmissions, the Identifier value chosen for each new Request
   need only be different from the previous Request, but need not be
   unique within the conversation.

and it's not necessary even an increasing sequence:

   EAP does not require the Identifier to be monotonically increasing,

So identifier sequence 0,1,0,1,0,1 would be equally acceptable
as 253,254,255,0,1,2 ...

Best regards,
Pasi