[ietf-dkim] Attempted text for x= with DSN considerations

Douglas Otis dotis at mail-abuse.org
Wed Apr 19 16:34:40 PDT 2006


On Apr 19, 2006, at 3:54 PM, <Bill.Oxley at cox.com>  
<Bill.Oxley at cox.com> wrote:

> Why would a verifyer refuse a message that had a value for x=?
>  "Verifiers MAY support checking of x= values or may refuse to
>   accept messages with the x= tag."

Some MTAs are known to forward email.  Stringent examination is  
required based upon the principle a sewer pipe must never become  
smaller.  The changed expiry text makes the x= parameter optional,  
but then excludes concerns regarding subsequent message rejection on  
that basis.

There was another minor change that used [received] time rather than  
[current] time to better define rules for various transports.  This  
was also to introduce the concept that there is a time interval  
between the received time and when the message may be forwarded.  An  
MTA can not continuously evaluate these message expiries.  When an  
MTA does not have a rate-limiting mechanism to handle in-hand expiry,  
the only safe practice would be to refuse messages that require such.

base-01:
,---
| It is beyond the scope of this specification to describe what actions
| a verifier system should make...
|
| If the verifying MTA is capable of verifying the public key of the
| signer and check the signature on the message synchronously with the
| SMTP session and such signature is missing or does not verify the MTA
| MAY reject the message with an error such as:
|
|    550 5.7.1 Unsigned messages not accepted
|
|    550 5.7.5 Message signature incorrect
'---

An effort subsequent to completion of the base draft is aimed  
specifically at providing a mechanism for the signing-domain to  
report that all messages have a valid signature.  A verifier can act  
upon that information and reject all messages purported to be from  
_that_ domain without a valid signature.  An expired signature is  
defined as _not_ being valid.

A safer practice would be to allow the receiver to evaluate the time  
the message was posted or signed.  There is a significant risk  
changing the validity of a signature on a per second basis, largely  
due to policy issues and foul play.  A slew of DSNs could be harmful  
to the reputation of the MTA.


-Doug


More information about the ietf-dkim mailing list