[ietf-dkim] Re: canonicalized null body and dkim

[Hector Santos]

Frank Ellermann wrote:
> 
> If they don't find this there's no body, that's ugly but allowed.
> It's the way how spam probes end up as "empty" mails, the spammer
> sends
 >
>     C DATA
>     S 3xx end with dot
>     C .
>     S 2xx got mail for valid address

But that means no headers too, a totally blank message which depending 
on how anal the SERVER is, may or may not accept or it will generate the 
a new header on its own based on the 2821 envelope, plus a message-id.

But in either case, it is not a DKIM message and if its a user 
submission, again, depending on the server option, it may not be 
accepted or the headers auto generated with a NULL or empty body. No 
problem here DKIM wise.

> I don't care about the canonicalization of spam probes or other
> mails without body, but the issue is no nonsense.

I'm not grasping the problem.  I think the issue is being overblown. The 
only question for me was how read or set a l=X for the SIMPLE c14n. Eric 
cleared that up for me.

---
HLS