ietf-dkim
[Top] [All Lists]

Re: [ietf-dkim] DKIM on envelope level

2009-10-30 14:49:33

On Oct 30, 2009, at 11:24 AM, John Levine wrote:

John mentioned CHUNKING.  The reason CHUNKING hasn't taken off is  
that ...

... the problem it was supposed to solve wasn't important.

In case it's not clear, I'm not saying that CHUNKING is a good idea.
But if for whatever reason you want the ability to abandon the
delivery between the headers and the body, CHUNKING lets you do that
without inventing anything new.

Signing the envelope strikes me as one of those things that sounds
nice, but when you try to work out the wa you'd use it in a practical
application, it turns out not to solve any interesting problems.

In particular, if the signature required an extra round trip for an
extra command, the delay would more than wipe out any speed increase.

Also, it is not my impression that the mere data bytes of mail flowing
through networks are a particular problem.  If, for example, you
pipelined DKIM checks so you could tell that you could throw away the
body of an message as it arrived rather than storing it, how much
better would an envelope check be?


Mailserver concurrency seems to be a significant problem at large  
receivers.
It's not the bytes so much as the ports multiplied by the seconds.  
Receiving the
mail and throwing it away saves you on IO bandwidth and memory, but
it still ties up that valuable delivery slot for the entire duration  
of the
message delivery.

Cheers,
   Steve

_______________________________________________
NOTE WELL: This list operates according to 
http://mipassoc.org/dkim/ietf-list-rules.html