Hello,

current text in https://tools.ietf.org/html/rfc7489#section-6 (DMARC Policy):

   Since email streams can
   be complicated (due to forwarding, existing RFC5322.From
   domain-spoofing services, etc.), Mail Receivers MAY deviate from a
   Domain Owner's published policy during message processing [... and SHOULD
   make available the fact of and reason for the deviation to the Domain
   Owner via feedback reporting, specifically using the "PolicyOverride"
   feature of the aggregate report (see Section 7.2).]

I propose this amendment to the first part of the sentance, (writen in a more 
abstract way, where the receiving site
decides on the policy.  The wording is subject to further adjustments):

* * * DMARC and Redirecting Messages

For a site, that is supposed to redirect a message with failed DMARC 
validation, to another site, if the PCT with the
policy is 100 the recommendation is not to redirect the message but reject it 
at SMTP level.  The rationale is, that
this message might be evaluated by the next hop site as Spam, while this hop 
does not consider the message as spam.  In
turn, the next hop can conclude, that this hop is sending spam.  If the next 
hop decides to apply DMARC policy reject
for the domain of the message, this hop will have to generate a bounce for the 
message, risking to be blacklisted by
some backscatters IP reputation lists.

For a site, that is supposed to redirect a message with failed DMARC 
validation, to another site, if the PCT with the
policy is between 1 and 99, the recommendation is to reject the message at SMTP 
level and not forward it further.  For
redirected messages, the PCT sampling is applied at least twice, thus there is 
a probabily that the next hop rejects the
message based on the PCT parameter, even if this hop has calculated not to 
reject the message.

It is in unknown, whether the next hop will reject or quarantine messages 
failing DMARC validation and from the sender's
perspective there is no difference, whether this hop or the next hop will 
reject the message.

The recommendations above do not fully apply, when the current hop changes the 
From: address, as if the recipient on the
next hop were a mailing list with one recipient, doing From: mungling.

It is not recommended to tell the sender that the message was delivered in the 
Junk folder of the recipient, and to
forward the message further, as the sender can get two rejections for the same 
message, from two different hops, which
is confusing.  This can happen, even if the From: address is mungled.

_______________________________________________
dmarc mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dmarc

Reply via email to