draft-ietf-dmarc-dmarcbis-29.xml

<?xml version="1.0" encoding="utf-8"?>
<!-- name="GENERATOR" content="github.com/mmarkdown/mmark Mmark Markdown Processor - mmark.miek.nl" -->
<rfc version="3" ipr="trust200902" docName="draft-ietf-dmarc-dmarcbis-29" submissionType="IETF" category="std" xml:lang="en" xmlns:xi="http://www.w3.org/2001/XInclude" obsoletes="7489, 9091" indexInclude="true" consensus="true">

<front>
<title abbrev="DMARCbis">Domain-based Message Authentication, Reporting, and Conformance (DMARC)</title><seriesInfo value="draft-ietf-dmarc-dmarcbis-29" stream="IETF" status="standard" name="Internet-Draft"></seriesInfo>
<author initials="T." surname="Herr (ed)" fullname="Todd M. Herr"><organization>Valimail</organization><address><postal><street></street>
</postal><email>todd.herr@valimail.com</email>
</address></author><author initials="J." surname="Levine (ed)" fullname="John Levine"><organization>Standcore LLC</organization><address><postal><street></street>
</postal><email>standards@standore.com</email>
</address></author><date/>
<area>Application</area>
<workgroup>DMARC</workgroup>

<abstract>
<t>This document describes the Domain-based Message Authentication,
Reporting, and Conformance (DMARC) protocol.</t>
<t>DMARC permits the owner of an email author's domain name to enable
verification of the domain's use, to indicate the Domain Owner's or
Public Suffix Operator's message handling preference regarding failed
verification, and to request reports about the use of the domain name.
Mail receiving organizations can use this information when evaluating
handling choices for incoming mail.</t>
<t>This document obsoletes RFCs 7489 and 9091.</t>
</abstract>

</front>

<middle>

<section anchor="introduction"><name>Introduction</name>
<t>RFC EDITOR: PLEASE REMOVE THE FOLLOWING PARAGRAPH BEFORE PUBLISHING:
The source for this draft is maintained on GitHub at:
<eref target="https://github.com/ietf-wg-dmarc/draft-ietf-dmarc-dmarcbis">https://github.com/ietf-wg-dmarc/draft-ietf-dmarc-dmarcbis</eref></t>
<t>Abusive email often includes unauthorized and deceptive use of a
domain name in the &quot;From&quot; header field defined in <xref target="RFC5322" sectionFormat="of" relative="#" section="3.6.2"></xref>
and referred to as RFC5322.From. The domain typically belongs to an organization
expected to be known to - and presumably trusted by - the recipient. The Sender
Policy Framework (SPF) <xref target="RFC7208"></xref> and DomainKeys Identified Mail (DKIM) <xref target="RFC6376"></xref>
protocols provide
domain-level authentication but are not directly associated with the
RFC5322.From domain. DMARC leverages these two protocols, providing a method
for Domain Owners to publish a DNS record describing the email authentication
policies for the RFC5322.From domain and to request specific handling
for messages using that domain that fail authentication checks.</t>
<t>As with SPF and DKIM, DMARC reports results as &quot;pass&quot; or &quot;fail&quot;.
To get a DMARC result of &quot;pass&quot;, a pass from either SPF or DKIM is required.
In addition, the passed domain can be &quot;aligned&quot; with the RFC5322.From domain
in one of two modes - &quot;relaxed&quot; or &quot;strict&quot;. The mode is expressed in the
domain's DMARC policy record. Domains are said to be &quot;in relaxed alignment&quot;
if they have the same &quot;Organizational Domain&quot;, which is the domain at the
top of the domain hierarchy for the RFC5322.From domain while having the
same administrative authority as the RFC5322.From domain. Domains are &quot;in
strict alignment&quot; if and only if they are identical.</t>
<t>A DMARC pass indicates only that the RFC5322.From domain has been
authenticated for that message. Authentication does not carry an
explicit or implicit value assertion about that message or about
the Domain Owner. Furthermore, a mail-receiving organization that performs
DMARC verification can choose to honor the Domain Owner's requested
message handling for authentication failures, but it is not required
to do so; it might choose different actions entirely.</t>
<t>For a mail-receiving organization supporting DMARC, a message that
passes verification is part of a message stream reliably
associated with the RFC5322.From field Domain Owner. Therefore,
reputation assessment of that stream by the mail-receiving organization
can assume the use of that domain
in the RFC5322.From field is authorized. A message that fails this verification
is not necessarily associated with the Domain Owner's domain and its
reputation.</t>
<t>DMARC policy records can also cover non-existent sub-domains below the
&quot;Organizational Domain&quot;, as well as domains at the top of the name hierarchy,
controlled by Public Suffix Operators (PSOs).</t>
<t>DMARC, in the associated <xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> and <xref target="I-D.ietf-dmarc-failure-reporting"></xref>
documents, also specifies a reporting framework. Using it, a mail-receiving
domain can generate regular reports about messages that claim to be from
a domain publishing DMARC policies, sending those reports to the address(es)
specified by the Domain Owner in the latter's DMARC policy record. Domain
Owners can use these reports, especially the aggregate reports, to identify
not only sources of mail attempting to fraudulently use their domain, but also
(and perhaps more importantly) gaps in their authentication practices. However,
as with honoring the Domain Owner's stated mail handling preference, a mail-receiving
organization supporting DMARC is under no obligation to send requested reports,
although it is recommended that they do send aggregate reports.</t>
<t>The use of DMARC creates some interoperability challenges that require due
consideration before deployment, particularly with configurations that
can cause mail to be rejected. These are discussed in <xref target="other-topics"></xref>.</t>
</section>

<section anchor="requirements"><name>Requirements</name>
<t>The following high-level goals, security dependencies, detailed requirements,
and items that are documented as out of scope guide specification of DMARC.</t>

<section anchor="high-level-goals"><name>High-Level Goals</name>
<t>DMARC has the following high-level goals:</t>

<ul>
<li><t>Allow Domain Owners and PSOs to assert their desired message handling
for authentication failures for messages purporting to have authorship
within the domain.</t>
</li>
<li><t>Allow Domain Owners and PSOs to verify their authentication deployment.</t>
</li>
<li><t>Minimize implementation complexity for both senders and receivers,
as well as the impact on handling and delivery of legitimate
messages.</t>
</li>
<li><t>Reduce the amount of successfully delivered spoofed emails.</t>
</li>
<li><t>Work at Internet scale.</t>
</li>
</ul>
</section>

<section anchor="anti-phishing"><name>Anti-Phishing</name>
<t>DMARC is designed to prevent bad actors from sending mail that claims
to come from legitimate senders, particularly transactional email
(official mail about business transactions).  One of the
primary uses of this kind of spoofed mail is phishing (enticing users
to provide information by pretending to be the legitimate service
requesting the information). Thus, DMARC is significantly informed
by ongoing efforts to enact large-scale, Internet-wide anti-phishing
measures.</t>
<t>Although DMARC can only be used to combat specific forms of
exact-domain spoofing directly, the DMARC mechanism has been found
to be useful in the creation of reliable and defensible message streams.</t>
<t>DMARC does not attempt to solve all problems with spoofed or
otherwise fraudulent emails. In particular, it does not address the
use of visually similar domain names (&quot;cousin domains&quot;) or abuse of
the RFC5322.From human-readable &lt;display-name&gt;.</t>
</section>

<section anchor="scalability"><name>Scalability</name>
<t>Scalability is a significant issue for systems that need to operate in a
system as widely deployed as current SMTP email. For this reason,
DMARC seeks to avoid the need for third parties or pre-sending
agreements between senders and receivers. This preserves the
positive aspects of the current email infrastructure.</t>
<t>Although DMARC does not introduce third-party senders (namely
external agents authorized to send on behalf of an operator) to the
email-handling flow, it also does not preclude them.  Such third
parties are free to provide services in conjunction with DMARC.</t>
</section>

<section anchor="out-of-scope"><name>Out of Scope</name>
<t>Several topics and issues are specifically out of scope of this
work. These include the following:</t>

<ul>
<li><t>Different treatment of messages that are not authenticated versus
those that fail authentication;</t>
</li>
<li><t>Evaluation of anything other than RFC5322.From header field;</t>
</li>
<li><t>Multiple reporting formats;</t>
</li>
<li><t>Publishing policy other than via the DNS;</t>
</li>
<li><t>Reporting or otherwise evaluating other than the last-hop IP
address;</t>
</li>
<li><t>Attacks in the display-name portions of the RFC5322.From header field, also known as &quot;display name&quot;
attacks;</t>
</li>
<li><t>Authentication of entities other than domains, since DMARC is
built upon SPF and DKIM, which authenticate domains; and</t>
</li>
<li><t>Content analysis.</t>
</li>
</ul>
</section>
</section>

<section anchor="terminology"><name>Terminology and Definitions</name>
<t>This section defines terms used in the rest of the document.</t>

<section anchor="conventions-used-in-this-document"><name>Conventions Used in This Document</name>
<t>The key words &quot;MUST&quot;, &quot;MUST NOT&quot;, &quot;REQUIRED&quot;, &quot;SHALL&quot;, &quot;SHALL NOT&quot;, &quot;SHOULD&quot;,
&quot;SHOULD NOT&quot;, &quot;RECOMMENDED&quot;, &quot;NOT RECOMMENDED&quot;, &quot;MAY&quot;, and &quot;OPTIONAL&quot; in this
document are to be interpreted as described in BCP 14 <xref target="RFC2119"></xref> and <xref target="RFC8174"></xref>
when, and only when, they appear in all capitals, as shown here.</t>
<t>Readers are encouraged to be familiar with the contents of
<xref target="RFC5598"></xref>.  In particular, that document defines various roles in
the messaging infrastructure that can appear the same or separate in
various contexts. For example, a Domain Owner could, via the
messaging security mechanisms on which DMARC is based, delegate the
ability to send mail as the Domain Owner to a third party with
another role. This document does not address the distinctions among
such roles; the reader is encouraged to become familiar with that
material before continuing.</t>
</section>

<section anchor="definitions"><name>Definitions</name>
<t>The following sections define the terms used in this document.</t>

<section anchor="authenticated-identifiers"><name>Authenticated Identifiers</name>
<t>Domain-level identifiers that are verified using authentication technologies
are referred to as &quot;Authenticated Identifiers&quot;. See <xref target="authentication-mechanisms"></xref>
for details about the supported mechanisms.</t>
</section>

<section anchor="author-domain"><name>Author Domain</name>
<t>The domain name of the apparent author as extracted from the RFC5322.From header field.</t>
</section>

<section anchor="domain-owner"><name>Domain Owner</name>
<t>An entity or organization that owns a DNS domain. The
term &quot;owns&quot; here indicates that the entity or organization being
referenced has control of that DNS domain, usually by holding its registration.
Domain Owners range from complex, globally distributed organizations to
service providers working on behalf of non-technical clients to
individuals responsible for maintaining personal domains. This
specification uses this term as analogous to an Administrative
Management Domain as defined in <xref target="RFC5598"></xref>. It can also refer
to delegates, such as Report Consumers when those are outside of
their immediate management domain.</t>
</section>

<section anchor="enforcement"><name>Enforcement</name>
<t>Enforcement describes a state where the Organizational Domain and
all subdomains below it are covered by a policy that is not &quot;p=none&quot;.
This means that the domain and its subdomains can only be used to send
mail that is properly authenticated, and mail using the domain name that
is unauthenticated will not reach the inbox of a mail receiver that validates
DMARC and honors the published policy.</t>
</section>

<section anchor="identifier-alignment"><name>Identifier Alignment</name>
<t>When the domain in the address in the RFC5322.From header field has the
same Organizational Domain as a domain verified by an Authenticated
Identifier, it has Identifier Alignment. (see <xref target="organizational-domain"></xref>)</t>
</section>

<section anchor="mail-receiver"><name>Mail Receiver</name>
<t>The entity or organization that receives and processes email. Mail
Receivers operate one or more Internet-facing Mail Transport Agents (MTAs).</t>
</section>

<section anchor="monitoring-mode"><name>Monitoring Mode</name>
<t>At p=none with a valid reporting address, the domain owner receives
reports that showcase authorized and unauthorized mail streams, as
well as gaps pertaining to authentication information pertaining to
both streams.</t>
</section>

<section anchor="non-existent-domains"><name>Non-existent Domains</name>
<t>For DMARC purposes, a non-existent domain is consistent with the term's meaning
as described in <xref target="RFC8020"></xref>. That is, if the response code received for a query
for a domain name is NXDOMAIN, then the domain name and all the names under it
do not exist.</t>
</section>

<section anchor="organizational-domain"><name>Organizational Domain</name>
<t>The Organizational Domain for any domain is determined by applying the algorithm
found in <xref target="organizational-domain-discovery"></xref>.</t>
</section>

<section anchor="public-suffix-domain"><name>Public Suffix Domain (PSD)</name>
<t>Some domains allow the registration of subdomains that are
&quot;owned&quot; by independent organizations.  Real-world examples of
these points are &quot;.com&quot;, &quot;.org&quot;, &quot;.us&quot;, and &quot;.co.uk&quot;.
These domains are called &quot;Public Suffix Domains (PSDs)&quot;.
For example, &quot;ietf.org&quot; is a registered domain name, and &quot;.org&quot; is its PSD.</t>
</section>

<section anchor="public-suffix-operator"><name>Public Suffix Operator (PSO)</name>
<t>A Public Suffix Operator is an organization that manages operations
within a PSD, particularly the DNS records published for names at and
under that domain name.</t>
</section>

<section anchor="pso-controlled-domain-names"><name>PSO Controlled Domain Names</name>
<t>PSO-Controlled Domain Names are names in the DNS that are managed by
a PSO. PSO-controlled Domain Names may have one label (e.g., &quot;.com&quot;) or more
(e.g., &quot;.co.uk&quot;), depending on the PSD's policy.</t>
</section>

<section anchor="report-consumer"><name>Report Consumer</name>
<t>An operator that receives reports from another operator implementing the
reporting mechanisms described in this document and/or the documents
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> and <xref target="I-D.ietf-dmarc-failure-reporting"></xref>. Such an
operator might be receiving reports about messages related to a domain
for which it is the Domain Owner or PSO or reports about messages related
to another operator's domain.  This term applies collectively to the
system components that receive and process these reports and the organizations
that operate them.</t>
</section>
</section>
</section>

<section anchor="overview-and-key-concepts"><name>Overview and Key Concepts</name>
<t>This section provides a general overview of the design and operation
of the DMARC environment.</t>

<section anchor="dmarc-basics"><name>DMARC Basics</name>
<t>DMARC permits a Domain Owner or PSO to enable verification of a domain's
use in an email message, to indicate the Domain Owner's or PSO's message
handling preference regarding failed verification, and to request reports
about use of the domain name. A domain's DMARC policy record is published
in DNS as a TXT record at the name created by prepending the label &quot;_dmarc&quot;
to the domain name and is retrieved through normal DNS queries.</t>
<t>DMARC's verification function is based on whether the RFC5322.From
domain is aligned with a domain name used in a supported authentication
mechanism, as described in <xref target="authentication-mechanisms"></xref>. When a DMARC policy
exists for the domain name found in the RFC5322.From header field, and that
domain name is not verified through an aligned supported authentication
mechanism, the handling of that message can be affected based on the
DMARC policy when delivered to a participating Mail Receiver.</t>
<t>A message satisfies the DMARC checks if at least one of the supported
authentication mechanisms:</t>

<ol>
<li><t>produces a &quot;pass&quot; result, and</t>
</li>
<li><t>produces that result based on an identifier that is in alignment,
as described in <xref target="identifier-alignment-explained"></xref>.</t>
</li>
</ol>
<t>It is important to note that the authentication mechanisms employed
by DMARC authenticate only a DNS domain. They do not authenticate the
local-part of any email address identifier found in a message, nor do
they validate the legitimacy of message content.</t>
<t>DMARC's feedback component involves the collection of information
about received messages claiming to be from the Author Domain
for periodic aggregate reports to the Domain Owner or PSO. The
parameters and format for such reports are discussed in
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref></t>
<t>A DMARC-enabled Mail Receiver might also generate per-message reports
that contain information related to individual messages that fail
authentication checks. Per-message failure reports are a useful source of
information when debugging deployments (if messages can be determined
to be legitimate even though failing authentication) or in analyzing
attacks.  The capability for such services is enabled by DMARC but
defined in other referenced material such as <xref target="RFC6591"></xref> and
<xref target="I-D.ietf-dmarc-failure-reporting"></xref></t>
</section>

<section anchor="use-of-rfc5322-from"><name>Use of RFC5322.From</name>
<t>One of the most obvious points of security scrutiny for DMARC is the
choice to focus on an identifier, namely the RFC5322.From address,
which is part of a body of data that has been trivially forged
throughout the history of email. This field is the one used by end
users to identify the source of the message, and so it has always
been a prime target for abuse through such forgery and other means.</t>
<t>Several points suggest that it is the most correct and safest thing
to do in this context:</t>

<ul>
<li><t>Of all the identifiers that are part of the message itself, this
is the only one required to be present. A message without a single,
properly formed RFC5322.From header field does not comply with <xref target="RFC5322"></xref>,
and handling such a message is outside of the scope of this specification.</t>
</li>
<li><t>It seems the best choice of an identifier on which to focus, as
most MUAs display some or all of the contents of that field in a
manner strongly suggesting those data as reflective of the true
originator of the message.</t>
</li>
<li><t>Many high-profile email sources, such as email service providers,
require that the sending agent has authenticated itself before email
can be generated. Thus, for these mailboxes, the mechanism
described in this document provides recipient end users with strong
evidence that the message was indeed originated by the agent they
associate with that mailbox, if the end user knows that these
various protections have been provided.</t>
</li>
</ul>
<t>Since the sorts of mail typically protected by DMARC participants
tend only to have single Authors, DMARC participants generally
operate under a slightly restricted profile of RFC5322 with respect
to the expected syntax of this field. See <xref target="mail-receiver-actions"></xref>
for details.</t>
</section>

<section anchor="authentication-mechanisms"><name>Authentication Mechanisms</name>
<t>The following mechanisms for determining Authenticated Identifiers
are supported in this version of DMARC:</t>

<ul>
<li><t>DKIM, <xref target="RFC6376"></xref>, which provides a domain-level identifier in the content of
the &quot;d=&quot; tag of a verified DKIM-Signature header field.</t>
</li>
<li><t>SPF, <xref target="RFC7208"></xref>, which can authenticate both the domain found in
an SMTP <xref target="RFC5321"></xref> HELO/EHLO command (the HELO identity) and the domain
found in an SMTP MAIL command (the MAIL FROM identity). As noted earlier,
however, DMARC relies solely on SPF authentication of the domain found in
SMTP MAIL FROM command. <xref target="RFC7208" sectionFormat="of" relative="#" section="2.4"></xref> describes MAIL FROM
processing for cases in which the MAIL command has a null path.</t>
</li>
</ul>
</section>

<section anchor="identifier-alignment-explained"><name>Identifier Alignment Explained</name>
<t>Email authentication technologies authenticate various (and
disparate) aspects of an individual message. For example, DKIM <xref target="RFC6376"></xref>
authenticates the domain that affixed a signature to the message,
while SPF <xref target="RFC7208"></xref> can authenticate either the domain that appears in the
RFC5321.MailFrom (MAIL FROM) portion of an SMTP <xref target="RFC5321"></xref> conversation or the
RFC5321.EHLO/HELO domain, or both. These domains may be different and are typically
not visible to the end user.</t>
<t>DMARC authenticates the use of the RFC5322.From domain by requiring either
that it have the same Organizational Domain as an Authenticated Identifier
(a condition known as &quot;relaxed alignment&quot;) or that it be identical to the
domain of the Authenticated Identifier (a condition known as &quot;strict
alignment&quot;). The choice of relaxed or strict alignment is left to the Domain
Owner and is expressed in the domain's DMARC policy record. Domain name
comparisons in this context are case-insensitive, per <xref target="RFC4343"></xref>.</t>
<t>It is important to note that Identifier Alignment cannot occur with a
message that is not valid per <xref target="RFC5322"></xref>, particularly one with a
malformed, absent, or repeated RFC5322.From header field, since in that case
there is no reliable way to determine a DMARC policy that applies to
the message. Accordingly, DMARC operation is predicated on the input
being a valid RFC5322 message object. For non-compliant cases, handling
is outside of the scope of this specification. Further discussion of this
can be found in <xref target="denial-of-dmarc-attacks"></xref>.</t>
<t>Each of the underlying authentication technologies that DMARC takes
as input yields authenticated domains as their outputs when they
succeed.</t>

<section anchor="dkim-identifiers"><name>DKIM-Authenticated Identifiers</name>
<t>DMARC requires that Identifier Alignment is applied to the result of a DKIM
authentication because a message can bear a valid signature from any
domain, including domains used by a mailing list or even a bad actor.
Therefore, merely bearing a valid signature is not enough to infer
the authenticity of the Author Domain.</t>
<t>DMARC requires that Identifier Alignment applied to the result of a DKIM
authentication to be strict or relaxed. (Note that these terms are
not related to DKIM's &quot;simple&quot; and &quot;relaxed&quot; canonicalization modes.)</t>
<t>In relaxed mode, the identifiers are considered to be aligned
if the Organizational Domains of both the DKIM-authenticated
signing domain (taken from the value of the d= tag in the signature)
and that of the RFC5322.From domain are equal.
In strict mode, only an exact match
between both Fully Qualified Domain Names (FQDNs) is considered to produce
Identifier Alignment.</t>
<t>To illustrate, in relaxed mode, if a verified DKIM signature
successfully verifies with a &quot;d=&quot; domain of &quot;example.com&quot;, and the
RFC5322.From address is &quot;alerts@news.example.com&quot;, the DKIM &quot;d=&quot;
domain and the RFC5322.From domain are considered to be &quot;in alignment&quot;,
because both domains have the same Organizational Domain of &quot;example.com&quot;.
In strict mode, this test would fail because the d= domain does not
exactly match the RFC5322.From domain.</t>
<t>Note that a single email can contain multiple DKIM signatures, and it
is considered to produce a DMARC &quot;pass&quot; result if any DKIM signature
is aligned and verified.</t>
</section>

<section anchor="spf-identifiers"><name>SPF-Authenticated Identifiers</name>
<t>DMARC requires that Identifier Alignment is applied to the result of an SPF
authentication. As with DKIM, Identifier Alignment can be either
strict or relaxed.</t>
<t>In relaxed mode, the identifiers are considered to be aligned if
the Organizational Domains of the SPF-authenticated
domain and RFC5322.From domain are equal.
In strict mode, only an exact match between the two FQDNs
is considered to produce Identifier Alignment.</t>
<t>For example, in relaxed mode, if a message passes an SPF check with an
RFC5321.MailFrom domain of &quot;cbg.bounces.example.com&quot;, and the address
portion of the RFC5322.From header field contains
&quot;payments@example.com&quot;, the Authenticated RFC5321.MailFrom domain
identifier and the RFC5322.From domain are considered to be &quot;in
alignment&quot; because they have the same Organizational Domain
(&quot;example.com&quot;). In strict mode, this test would fail because the
two domains are not identical.</t>
<t>Note that SPF alignment checks in DMARC rely solely
on the RFC5321.MailFrom domain. This differs from
<xref target="RFC7208" sectionFormat="of" relative="#" section="2.3"></xref>, which recommends that SPF checks be done on not only the
&quot;MAIL FROM&quot; but also on a separate check of the &quot;HELO&quot; identity.</t>
</section>

<section anchor="alignment-and-extension-technologies"><name>Alignment and Extension Technologies</name>
<t>If in the future DMARC is extended to include the use of other
authentication mechanisms, the extensions will need to allow for
domain identifier extraction so that alignment with the RFC5322.From
domain can be verified.</t>
</section>
</section>

<section anchor="flow-diagram"><name>Flow Diagram</name>

<sourcecode type="ascii-art"> +---------------+                             +--------------------+
 | Author Domain |&lt; . . . . . . . . . . . .    | Return-Path Domain |
 +---------------+                        .    +--------------------+
     |                                    .               ^
     V                                    V               .
 +-----------+     +--------+       +----------+          v
 |   MSA     |&lt;***&gt;|  DKIM  |       |   DMARC  |     +----------+
 |  Service  |     | Signer |       | Verifier |&lt;***&gt;|    SPF   |
 +-----------+     +--------+       +----------+  *  | Verifier |
     |                                    ^       *  +----------+
     |                                    *       *
     V                                    v       *
  +------+        (~~~~~~~~~~~~)      +------+    *  +----------+
  | sMTA |-------&gt;( other MTAs )-----&gt;| rMTA |    **&gt;|   DKIM   |
  +------+        (~~~~~~~~~~~~)      +------+       | Verifier |
                                         |           +----------+
                                         |                ^
                                         V                .
                                  +-----------+           .
                    +---------+   |    MDA    |           v
                    |  User   |&lt;--| Filtering |      +-----------+
                    | Mailbox |   |  Engine   |      |   DKIM    |
                    +---------+   +-----------+      |  Signing  |
                                                     | Domain(s) |
                                                     +-----------+

  MSA = Mail Submission Agent
  MDA = Mail Delivery Agent
</sourcecode>
<t>The above diagram shows a simple flow of messages through a
DMARC-aware system. Solid lines denote the actual message flow,
dotted lines involve DNS queries used to retrieve message policy
related to the supported message authentication schemes, and asterisk
lines indicate data exchange between message-handling modules and
message authentication modules. &quot;sMTA&quot; is the sending MTA, and &quot;rMTA&quot;
is the receiving MTA.</t>
<t>Put simply, when a message reaches a DMARC-aware rMTA, a DNS query
will be initiated to determine if a DMARC policy exists that applies
to the author domain. If a policy is found, the rMTA will use the results
of SPF and DKIM verification checks to determine the ultimate DMARC
authentication status. The DMARC status can then factor into the
message handling decision made by the recipient's mail system.</t>
<t>More details on specific actions for the parties involved can be
found in <xref target="domain-owner-actions"></xref> and <xref target="mail-receiver-actions"></xref>.</t>
</section>

<section anchor="dns-tree-walk"><name>DNS Tree Walk</name>
<t>The DMARC protocol defines a method for communicating information
through the publishing of records in DNS. Both the content of the
records and their location in the DNS hierarchy are used for two
purposes: policy discovery (see <xref target="dmarc-policy-discovery"></xref>) and Organizational
Domain determination (see <xref target="organizational-domain-discovery"></xref>).</t>
<t>The relevant DMARC record for these purposes is not necessarily the DMARC
policy record found in DNS at the same level as the name label for
the domain in question. Instead, some domains will inherit their
DMARC policy records from parent domains one level or more above
them in the DNS hierarchy. Similarly, the Organizational Domain
may be found at a higher level in the DNS hierarchy.</t>
<t>These records are discovered through the technique described here,
known colloquially as the &quot;DNS Tree Walk&quot;. The target of any DNS Tree
Walk is a valid DMARC policy record, but the rules defining required
content for that record depend on the reason for performing the Tree
Walk.</t>
<t>To prevent possible abuse of the DNS, a shortcut is built into the
process so that domains that have more than five labels do not result
in more than five DNS queries.</t>
<t>The generic steps for a DNS Tree Walk are as follows:</t>

<ol>
<li><t>Query the DNS for a DMARC TXT record at the appropriate starting point
for the Tree Walk. A possibly empty set of records is returned.</t>
</li>
<li><t>Records that do not start with a &quot;v=&quot; tag that identifies the current
version of DMARC are discarded. If multiple DMARC records are returned,
they are all discarded. If a single record remains and it contains a
&quot;psd=n&quot; tag, stop.</t>
</li>
<li><t>Determine the target for additional queries (if needed; see the note
in <xref target="organizational-domain-discovery"></xref>), using steps 4 through 8 below.</t>
</li>
<li><t>Break the subject DNS domain name into a set of ordered labels. Assign
the count of labels to &quot;x&quot;, and number the labels from right to left; e.g.,
for &quot;a.mail.example.com&quot;, &quot;x&quot; would be assigned the value 4, &quot;com&quot; would be
label 1, &quot;example&quot; would be label 2, &quot;mail&quot; would be label 3, and so forth.</t>
</li>
<li><t>If x &lt; 5, remove the left-most (highest-numbered) label from the subject
domain. If x &gt;= 5, remove the left-most (highest-numbered) labels from the
subject domain until 4 labels remain. The resulting DNS domain name is the
new target for the next lookup.</t>
</li>
<li><t>Query the DNS for a DMARC TXT record at the DNS domain name matching this
new target. A possibly empty set of records is returned.</t>
</li>
<li><t>Records that do not start with a &quot;v=&quot; tag that identifies the current
version of DMARC are discarded. If multiple DMARC records are returned for
a single target, they are all discarded. If a single record remains and it
contains a &quot;psd=n&quot; or &quot;psd=y&quot; tag, stop.</t>
</li>
<li><t>Determine the target for additional queries by removing a single label
from the target domain as described in step 5 and repeating steps 6 and 7
until the process stops or there are no more labels remaining.</t>
</li>
</ol>
<t>To illustrate, for a message with the arbitrary RFC5322.From domain of
&quot;a.b.c.d.e.mail.example.com&quot;, a full DNS Tree Walk would require the following
five queries to locate the policy or Organizational Domain:</t>

<ul spacing="compact">
<li>_dmarc.a.b.c.d.e.mail.example.com</li>
<li>_dmarc.e.mail.example.com</li>
<li>_dmarc.mail.example.com</li>
<li>_dmarc.example.com</li>
<li>_dmarc.com</li>
</ul>
</section>

<section anchor="dmarc-policy-discovery"><name>DMARC Policy Discovery</name>
<t>For policy discovery, a DNS Tree Walk starts at the domain found in the RFC5322.From
header of the message being evaluated. The DMARC policy to be applied to the message
will be the record found at of the following locations, listed from highest preference
to lowest:</t>

<ul spacing="compact">
<li>The RFC5322.From domain</li>
<li>The Organizational Domain (as determined by a separate DNS Tree Walk) of the
RFC5322.From domain</li>
<li>The Public Suffix Domain of the RFC5322.From domain</li>
</ul>
<t>If the DMARC policy to be applied is that of the RFC5322.From domain, then the
DMARC policy is taken from the p= tag of the record.</t>
<t>If the DMARC policy to be applied is that of either the Organizational Domain or the
Public Suffix Domain and that domain is different than the RFC5322.From domain, then
the DMARC policy is taken from the sp= tag (if any) if the RFC5322.From domain exists,
or the np= tag (if any) if the RFC5322.From domain does not exist. In the absence of
applicable sp= or np= tags, the p= tag policy is used for subdomains.</t>
<t>If a retrieved policy record does not contain a valid &quot;p&quot; tag, or contains an &quot;sp&quot; or
&quot;np&quot; tag that is not valid, then:</t>

<ul>
<li><t>If a &quot;rua&quot; tag is present and contains at least one syntactically valid reporting
URI, the Mail Receiver <bcp14>MUST</bcp14> act as if a record containing &quot;p=none&quot; was
retrieved and continue processing;</t>
</li>
<li><t>Otherwise, the Mail Receiver applies no DMARC processing to this message.</t>
</li>
</ul>
<t>If the set produced by the DNS Tree Walk contains no DMARC policy
record (i.e., any indication that there is no such record as opposed
to a transient DNS error), Mail Receivers <bcp14>MUST NOT</bcp14> apply the DMARC
mechanism to the message.</t>
<t>Handling of DNS errors when querying for the DMARC policy record is
left to the discretion of the Mail Receiver. For example, to ensure
minimal disruption of mail flow, transient errors could result in
delivery of the message (&quot;fail open&quot;), or they could result in the
message being temporarily rejected (i.e., an SMTP 4yx reply), which
invites the sending MTA to try again after the condition has possibly
cleared, allowing a definite DMARC conclusion to be reached (&quot;fail
closed&quot;).</t>
<t>Note: PSD policy is not used for Organizational Domains that have
published a DMARC policy. Specifically, this is not a mechanism to
provide feedback addresses (rua/ruf) when an Organizational Domain has
declined to do so.</t>
</section>

<section anchor="organizational-domain-discovery"><name>Organizational Domain Discovery</name>
<t>For Organizational Domain discovery, it may be necessary to perform
multiple DNS Tree Walks to determine if any two domains are in alignment.
This means that a DNS Tree Walk to discover an Organizational Domain might
start at any of the following locations:</t>

<ul spacing="compact">
<li>The domain found in the RFC5322.From header of the message being evaluated.</li>
<li>The domain found in the RFC5321.MailFrom header if there is an SPF pass result
for the message being evaluated.</li>
<li>Any DKIM d= domain if there is a DKIM pass result for that domain for the message
being evaluated.</li>
</ul>
<t>Note: There is no need to perform Tree Walk searches for Organizational Domains
under any of the following conditions:</t>

<ul spacing="compact">
<li>The RFC5322.From domain and the RFC5321.MailFrom domain (if SPF
authenticated), and/or the DKIM d= domain (if present and authenticated)
are all the same, and that domain has a DMARC record. In this case, this
common domain is treated as the Organizational Domain.</li>
<li>No applicable DMARC policy is discovered for the RFC5322.From domain during
the Tree Walk for that domain. In this case, the DMARC mechanism does not apply to
the message in question.</li>
<li>The record for the RFC5322.From domain indicates strict alignment. In this
case, a simple string comparison of the RFC5322.From domain and the
RFC5321.MailFrom domain (if SPF authenticated), and/or the DKIM d= domain
(if present and authenticated) is all that is required.</li>
</ul>
<t>To discover the Organizational Domain for a domain, perform the DNS Tree Walk
described in <xref target="dns-tree-walk"></xref> as needed for any of the domains in question.</t>
<t>For each Tree Walk that retrieved valid DMARC records, select the Organizational
Domain from the domains for which valid DMARC records were retrieved from the longest
to the shortest:</t>

<ol>
<li><t>If a valid DMARC record contains the psd= tag set to 'n' (psd=n), this is the
Organizational Domain, and the selection process is complete.</t>
</li>
<li><t>If a valid DMARC record, other than the one for the domain where the tree
walk started, contains the psd= tag set to 'y' (psd=y), the Organizational
Domain is the domain one label below this one in the DNS hierarchy, and the
selection process is complete.</t>
</li>
<li><t>Otherwise, select the record for the domain with the fewest number of labels.
This is the Organizational Domain and the selection process is complete.</t>
</li>
</ol>
<t>If this process does not determine the Organizational Domain, then the initial target
domain is the Organizational Domain.</t>
<t>For example, given the starting domain &quot;a.mail.example.com&quot;, a search
for the Organizational Domain would require a series of DNS queries for DMARC
records starting with &quot;_dmarc.a.mail.example.com&quot; and finishing with &quot;_dmarc.com&quot;.
If there are DMARC records for &quot;_dmarc.mail.example.com&quot; and
&quot;_dmarc.example.com&quot;, but not for &quot;_dmarc.a.mail.example.com&quot; or
&quot;_dmarc.com&quot;, then the Organizational Domain for this domain would be
&quot;example.com&quot;.</t>
<t>As another example, given the starting domain &quot;a.mail.example.com&quot;, if a
search for the Organizational Domain yields a DMARC record at &quot;_dmarc.mail.example.com&quot;
with the psd= tag set to 'n', then the Organizational Domain for this domain would
be &quot;mail.example.com&quot;.</t>
<t>As a last example, given the starting domain &quot;a.mail.example.com&quot;, if a
search for the Organizational Domain only yields a DMARC record at &quot;_dmarc.com&quot;
and that record contains the tag psd=y, then the Organizational Domain for
this domain would be &quot;example.com&quot;.</t>
</section>
</section>

<section anchor="policy"><name>Policy</name>
<t>A Domain Owner or PSO advertises DMARC participation of one or more of its
domains by adding a DNS TXT record (described in <xref target="dmarc-policy-record"></xref>) to
those domains. In doing so, Domain Owners and PSOs indicate their handling
preference regarding failed authentication for email messages using their
domain in the RFC5322.From header field as well as their desire
for feedback about those messages. Mail Receivers, in turn, can take
into account the Domain Owner's stated preference when making handling
decisions about email messages that fail DMARC authentication checks.</t>
<t>A Domain Owner or PSO may choose not to participate in DMARC evaluation by
Mail Receivers simply by not publishing an appropriate DNS TXT record for
its domain(s). A Domain Owner can also choose not to have some underlying
authentication technologies apply to DMARC evaluation of its domain(s). In
this case, the Domain Owner simply declines to advertise participation in
those schemes. For example, if the results of path authorization checks
ought not to be considered as part of the overall DMARC result for a given
Author Domain, then the Domain Owner does not publish an SPF policy record
that can produce an SPF pass result.</t>
<t>A Mail Receiver implementing the DMARC mechanism gets the
Domain Owner's or PSO's published DMARC Domain Owner Assessment
Policy and uses it as an important factor in deciding how to
handle the message. Mail handling considerations based on DMARC
policy enforcement are discussed below in <xref target="policy-enforcement-considerations"></xref>.</t>

<section anchor="dmarc-policy-record"><name>DMARC Policy Record</name>
<t>Domain Owner and PSO DMARC preferences are stored as DNS TXT records in
subdomains named &quot;_dmarc&quot;. For example, the Domain Owner of
&quot;example.com&quot; would post DMARC preferences in a TXT record at
&quot;_dmarc.example.com&quot;. Similarly, a Mail Receiver wishing to query
for DMARC preferences regarding mail with an RFC5322.From domain of
&quot;example.com&quot; would issue a TXT query to the DNS for the subdomain of
&quot;_dmarc.example.com&quot;. The DNS-located DMARC preference data will
hereafter be called the &quot;DMARC record&quot;.</t>
<t>DMARC's use of the Domain Name Service is driven by DMARC's use of
domain names and the nature of the query it performs. The query
requirement matches with the DNS for obtaining simple parametric
information. It uses an established method of storing the
information associated with the target domain name, specifically an
isolated TXT record that is restricted to the DMARC context. Using
the DNS as the query service has the benefit of reusing an extremely
well-established operations, administration, and management
infrastructure, rather than creating a new one.</t>
<t>Per <xref target="RFC1035"></xref>, a TXT record can comprise several &quot;character-string&quot;
objects. Where this is the case, the module performing DMARC
evaluation <bcp14>MUST</bcp14> concatenate these strings by joining together the
objects in order and parsing the result as a single string.</t>
</section>

<section anchor="dmarc-uris"><name>DMARC URIs</name>
<t><xref target="RFC3986"></xref> defines a generic syntax for identifying a resource. The DMARC
mechanism uses this as the format by which a Domain Owner or PSO specifies
the destination for the two report types that are supported.</t>
<t>The place such URIs are specified (see <xref target="general-record-format"></xref>) allows
a list of these to be provided. The list of URIs is separated by commas
(ASCII 0x2c). A report <bcp14>SHOULD</bcp14> be sent to each listed URI provided in
the DMARC record.</t>
<t>A formal definition is provided in <xref target="formal-definition"></xref>.</t>
</section>

<section anchor="general-record-format"><name>General Record Format</name>
<t>DMARC records follow the extensible &quot;tag-value&quot; syntax for DNS-based
key records defined in DKIM <xref target="RFC6376"></xref>.</t>
<t><xref target="iana-considerations"></xref> creates a registry for known DMARC tags and
registers the initial set defined in this document. Only tags defined
in that registry are to be processed; unknown tags <bcp14>MUST</bcp14> be ignored.</t>
<t>The following tags are valid DMARC tags:</t>

<dl>
<dt>adkim:</dt>
<dd><t>(plain-text; <bcp14>OPTIONAL</bcp14>; default is &quot;r&quot;.) Indicates whether
the Domain Owner requires strict or relaxed DKIM Identifier Alignment
mode. See <xref target="dkim-identifiers"></xref> for details. Valid values are as follows:</t>

<dl spacing="compact">
<dt>r:</dt>
<dd>relaxed mode</dd>
<dt>s:</dt>
<dd>strict mode</dd>
</dl></dd>
<dt>aspf:</dt>
<dd><t>(plain-text; <bcp14>OPTIONAL</bcp14>; default is &quot;r&quot;.)  Indicates whether
the Domain Owner requires strict or relaxed SPF Identifier Alignment
mode. See <xref target="spf-identifiers"></xref> for details. Valid values are as follows:</t>

<dl spacing="compact">
<dt>r:</dt>
<dd>relaxed mode</dd>
<dt>s:</dt>
<dd>strict mode</dd>
</dl></dd>
<dt>fo:</dt>
<dd><t>Failure reporting options (plain-text; <bcp14>OPTIONAL</bcp14>; default is &quot;0&quot;)
Provides requested options for the generation of failure reports.
Report generators may choose to adhere to the requested options.
This tag's content <bcp14>MUST</bcp14> be ignored if a &quot;ruf&quot; tag (below) is not
also specified. Failure reporting options are shown below. The value
of this tag is either &quot;0&quot;, &quot;1&quot;, or a colon-separated list of the
options represented by alphabetic characters. The valid values and
their meanings are:</t>

<dl spacing="compact">
<dt>0:</dt>
<dd>Generate a DMARC failure report if all underlying authentication
mechanisms fail to produce an aligned &quot;pass&quot; result.</dd>
<dt>1:</dt>
<dd>Generate a DMARC failure report if any underlying authentication
mechanism produced something other than an aligned &quot;pass&quot; result.</dd>
<dt>d:</dt>
<dd>Generate a DKIM failure report if the message had a signature
that failed evaluation, regardless of its alignment. DKIM-specific
reporting is described in <xref target="RFC6651"></xref>.</dd>
<dt>s:</dt>
<dd>Generate an SPF failure report if the message failed SPF
evaluation, regardless of its alignment. SPF-specific
reporting is described in <xref target="RFC6652"></xref>.</dd>
</dl></dd>
<dt>np:</dt>
<dd><t>Domain Owner Assessment Policy for non-existent subdomains
(plain-text; <bcp14>OPTIONAL</bcp14>). Indicates the message handling preference
of the Domain Owner or PSO for mail using non-existent subdomains
of the domain queried but not passing DMARC verification. It applies
only to non-existent subdomains of the domain queried and not to either
existing subdomains or the domain itself. Its syntax is identical to
that of the &quot;p&quot; tag defined below. If the &quot;np&quot; tag is absent, the policy
specified by the &quot;sp&quot; tag (if the &quot;sp&quot; tag is present) or the
policy specified by the &quot;p&quot; tag, if the &quot;sp&quot; tag is not present,
<bcp14>MUST</bcp14> be applied for non-existent subdomains. Note that &quot;np&quot; will
be ignored for DMARC records published on subdomains of Organizational
Domains and PSDs due to the effect of the DMARC policy discovery
mechanism described in <xref target="dmarc-policy-discovery"></xref>.</t>
</dd>
<dt>p:</dt>
<dd><t>Domain Owner Assessment Policy (plain-text; <bcp14>RECOMMENDED</bcp14> for policy
records). Indicates the message handling preference of the Domain Owner
or PSO for mail using its domain but not passing DMARC verification.
The policy applies to the domain queried and to subdomains, unless the
subdomain policy is explicitly described using the &quot;sp&quot; or &quot;np&quot; tags.
If this tag is not present in an otherwise syntactically valid DMARC
record, then the record is treated as if it included &quot;p=none&quot; (see
<xref target="dmarc-policy-discovery"></xref>). This tag is not applicable for third-party
reporting records (see <xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> and <xref target="I-D.ietf-dmarc-failure-reporting"></xref>)
Possible values are as follows:</t>

<dl spacing="compact">
<dt>none:</dt>
<dd>The Domain Owner offers no expression of preference.</dd>
<dt>quarantine:</dt>
<dd>The Domain Owner considers such mail to be suspicious. It is possible
the mail is valid, although the failure creates a significant concern.</dd>
<dt>reject:</dt>
<dd>The Domain Owner considers all such failures to be a clear indication
that the use of the domain name is not valid. See <xref target="rejecting-messages"></xref>
for some discussion of SMTP rejection methods and their implications.</dd>
</dl></dd>
<dt>psd:</dt>
<dd><t>A flag indicating whether the domain is a PSD. (plain-text; <bcp14>OPTIONAL</bcp14>;
default is 'u'). Possible values are:</t>

<dl spacing="compact">
<dt>y:</dt>
<dd>PSOs include this tag with a value of 'y' to indicate that the domain
is a PSD. If a record containing this tag with a value of 'y' is found during
policy discovery, this information will be used to determine the Organizational
Domain and policy domain applicable to the message in question.</dd>
<dt>n:</dt>
<dd>The DMARC policy record is published for a PSD, but it is the Organizational
Domain for itself and its subdomain. There is no need to put psd=n in a DMARC
record, except in the very unusual case of a parent PSD publishing a DMARC
record without the requisite psd=y tag.</dd>
<dt>u:</dt>
<dd>The default indicates that the DMARC policy record is published for a domain
that is not a PSD. Use the mechanism described in <xref target="organizational-domain-discovery"></xref>
for determining the Organizational Domain. There is no need to explicitly publish
psd=u in a DMARC record.</dd>
</dl></dd>
<dt>rua:</dt>
<dd><t>Addresses to which aggregate feedback is to be sent (comma-separated plain-text
list of DMARC URIs; <bcp14>OPTIONAL</bcp14>). <xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> discusses
considerations that apply when the domain name of a URI differs from that of the domain
advertising the policy. See <xref target="external-report-addresses"></xref> for additional considerations. Any
valid URI can be specified. A Mail Receiver <bcp14>MUST</bcp14> implement support for a &quot;mailto:&quot;
URI, i.e., the ability to send a DMARC report via electronic mail. If the tag is not
provided, Mail Receivers <bcp14>MUST NOT</bcp14> generate aggregate feedback reports for the domain.
URIs not supported by Mail Receivers <bcp14>MUST</bcp14> be ignored. The aggregate feedback report
format is described in <xref target="I-D.ietf-dmarc-aggregate-reporting"></xref>.</t>
</dd>
<dt>ruf:</dt>
<dd><t>Addresses to which message-specific failure information is to be reported
(comma-separated plain-text list of DMARC URIs; <bcp14>OPTIONAL</bcp14>). If present, the
Domain Owner is requesting Mail Receivers to send detailed failure reports about
messages that fail the DMARC evaluation in specific ways (see the &quot;fo&quot; tag above).
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> discusses considerations that apply when
the domain name of a URI differs from that of the domain advertising the policy.
A Mail Receiver <bcp14>MUST</bcp14> implement support for a &quot;mailto:&quot; URI, i.e., the ability to
send a DMARC report via electronic mail.  If the tag is not provided, Mail Receivers
<bcp14>MUST NOT</bcp14> generate failure reports for the domain. See <xref target="external-report-addresses"></xref>
for additional considerations.</t>
</dd>
<dt>sp:</dt>
<dd><t>Domain Owner Assessment Policy for all subdomains (plain-text;
<bcp14>OPTIONAL</bcp14>). Indicates the message handling preference of the Domain Owner
or PSO for mail using an existing subdomain of the domain queried
but not passing DMARC verification. It applies only to subdomains of
the domain queried and not to the domain itself. Its syntax is
identical to that of the &quot;p&quot; tag defined above. If both the &quot;sp&quot;
tag is absent, and the &quot;np&quot; tag is either absent or not applicable,
the policy specified by the &quot;p&quot; tag <bcp14>MUST</bcp14> be applied for subdomains.
Note that &quot;sp&quot; will be ignored for DMARC records published on
subdomains of Organizational Domains due to the effect of the
DMARC policy discovery mechanism described in <xref target="dmarc-policy-discovery"></xref>.</t>
</dd>
<dt>t:</dt>
<dd><t>DMARC policy test mode (plain-text; <bcp14>OPTIONAL</bcp14>; default is 'n'). For
the RFC5322.From domain to which the DMARC record applies, the &quot;t&quot;
tag serves as a signal to the actor performing DMARC verification checks
as to whether or not the domain owner wishes the assessment policy
declared in the &quot;p=&quot;, &quot;sp=&quot;, and/or &quot;np=&quot; tags to actually be applied. This
parameter does not affect the generation of DMARC reports. Possible values
are as follows:</t>

<dl spacing="compact">
<dt>y:</dt>
<dd>A request that the actor performing the DMARC verification check not
apply the policy, but instead apply any special handling rules it might have
in place, such as rewriting the RFC5322.From header. The domain owner is
currently testing its specified DMARC assessment policy.</dd>
<dt>n:</dt>
<dd>The default is a request to apply the policy as specified to any
message that produces a DMARC &quot;fail&quot; result.</dd>
</dl></dd>
<dt>v:</dt>
<dd><t>Version (plain-text; <bcp14>REQUIRED</bcp14>).  Identifies the record retrieved
as a DMARC record. It <bcp14>MUST</bcp14> have the value of &quot;DMARC1&quot;. The value
of this tag <bcp14>MUST</bcp14> match precisely; if it does not or it is absent,
the entire retrieved record <bcp14>MUST</bcp14> be ignored. It <bcp14>MUST</bcp14> be the first
tag in the list.</t>
</dd>
</dl>
<t>A DMARC policy record <bcp14>MUST</bcp14> comply with the formal specification found
in <xref target="formal-definition"></xref> in that the &quot;v&quot; tag <bcp14>MUST</bcp14> be present and <bcp14>MUST</bcp14>
appear first. Unknown tags <bcp14>MUST</bcp14> be ignored. Syntax errors
in the remainder of the record <bcp14>MUST</bcp14> be discarded in favor of
default values (if any) or ignored outright.</t>
<t>Note that given the rules of the previous paragraph, the addition of a
new tag into the registered list of tags does not itself require a
new version of DMARC to be generated (with a corresponding change to
the &quot;v&quot; tag's value), but a change to any existing tags does require
a new version of DMARC.</t>
</section>

<section anchor="formal-definition"><name>Formal Definition</name>
<t>The formal definition of the DMARC format, using <xref target="RFC5234"></xref> and <xref target="RFC7405"></xref>,
is as follows:</t>

<artwork>  dmarc-uri     = URI
                ; &quot;URI&quot; is imported from [RFC3986]; commas 
                ; (ASCII 0x2C) and exclamation points 
                ; (ASCII 0x21) MUST be encoded

  dmarc-sep     = *WSP &quot;;&quot; *WSP

  equals        = *WSP &quot;=&quot; *WSP

  dmarc-record  = dmarc-version *(dmarc-sep dmarc-tag) [dmarc-sep] *WSP

  dmarc-tag     = 1*ALPHA equals 1*dmarc-value

  ; any printing characters but semicolon
  dmarc-value   = %x20-3A | %x3C-7E 

  dmarc-version = &quot;v&quot; equals %s&quot;DMARC1&quot; ; case sensitive

  ; specialized syntax of DMARC values
  dmarc-request = &quot;none&quot; / &quot;quarantine&quot; / &quot;reject&quot;

  dmarc-yorn    = &quot;y&quot; / &quot;n&quot;

  dmarc-psd     = &quot;y&quot; / &quot;n&quot; / &quot;u&quot;

  dmarc-rors    = &quot;r&quot; / &quot;s&quot;

  dmarc-urilist = dmarc-uri *(*WSP &quot;,&quot; *WSP dmarc-uri)

  dmarc-fo      = &quot;0&quot; / &quot;1&quot; / &quot;d&quot; / &quot;s&quot; / &quot;d:s&quot; / &quot;s:d&quot;

</artwork>
<t>&quot;Keyword&quot; is imported from <xref target="RFC5321" sectionFormat="of" relative="#" section="4.1.2"></xref>.</t>
<t>In each dmarc-tag, the dmarc-value has a syntax that depends on the tag name.
The ABNF rule for each dmarc-value is specified in the following table:</t>
<table align="left"><name>&quot;Tag Names and Values&quot;
</name>
<thead>
<tr>
<th>Tag Name</th>
<th>Value Rule</th>
</tr>
</thead>

<tbody>
<tr>
<td>p</td>
<td>dmarc-request</td>
</tr>

<tr>
<td>t</td>
<td>dmarc-yorn</td>
</tr>

<tr>
<td>psd</td>
<td>dmarc-psd</td>
</tr>

<tr>
<td>np</td>
<td>dmarc-request</td>
</tr>

<tr>
<td>sp</td>
<td>dmarc-request</td>
</tr>

<tr>
<td>adkim</td>
<td>dmarc-rors</td>
</tr>

<tr>
<td>aspf</td>
<td>dmarc-rors</td>
</tr>

<tr>
<td>rua</td>
<td>dmarc-urilist</td>
</tr>

<tr>
<td>ruf</td>
<td>dmarc-urilist</td>
</tr>

<tr>
<td>fo</td>
<td>dmarc-fo</td>
</tr>
</tbody>
</table></section>

<section anchor="domain-owner-actions"><name>Domain Owner Actions</name>
<t>This section describes Domain Owner actions to implement the
DMARC mechanism.</t>

<section anchor="publish-an-spf-policy-for-an-aligned-domain"><name>Publish an SPF Policy for an Aligned Domain</name>
<t>Because DMARC relies on SPF <xref target="RFC7208"></xref> and DKIM <xref target="RFC6376"></xref>, in
order to take full advantage of DMARC, a Domain Owner <bcp14>SHOULD</bcp14> first
ensure that SPF and DKIM authentication are properly configured.
As a first step, the Domain Owner <bcp14>SHOULD</bcp14> choose a domain to use as the
RFC5321.MailFrom domain (i.e., the Return-Path domain) for its mail,
one that aligns with the Author Domain, and then publish an SPF
policy in DNS for that domain. The SPF record <bcp14>SHOULD</bcp14> be constructed
at a minimum to ensure an SPF pass verdict for all known sources of
mail for the RFC5321.MailFrom domain.</t>
</section>

<section anchor="configure-sending-system-for-dkim-signing-using-an-aligned-domain"><name>Configure Sending System for DKIM Signing Using an Aligned Domain</name>
<t>While it is possible to secure a DMARC pass verdict based on only
one of SPF or DKIM, it is commonly accepted best practice to ensure that
both authentication mechanisms are in place to guard
against failure of just one of them.</t>
<t>This is particularly important because SPF will always fail in situations
where mail is sent to a forwarding address offered by a professional society,
school or other institution, where the address simply relays the message to
the recipient's current &quot;real&quot; address.  Many recipients use such addresses
and with SPF alone and not DKIM, messages sent to such users will always
produce DMARC fail.</t>
<t>The Domain Owner <bcp14>SHOULD</bcp14> choose
a DKIM-Signing domain (i.e., the d= domain in the DKIM-Signature
header) that aligns with the Author Domain.</t>
</section>

<section anchor="setup-a-mailbox-to-receive-aggregate-reports"><name>Setup a Mailbox to Receive Aggregate Reports</name>
<t>Proper consumption and analysis of DMARC aggregate reports are the
keys to any successful DMARC deployment for a Domain Owner. DMARC
aggregate reports, which are XML documents and are defined in
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref>, contain valuable data for the Domain
Owner, showing sources of mail using the Author Domain. Depending
on how mature the Domain Owner's DMARC rollout is, some of these
sources could be legitimate ones that were overlooked during the
initial deployment of SPF and/or DKIM.</t>
<t>Because the aggregate reports are XML documents, it is recommended
that they be machine-parsed, so setting up a mailbox involves more
than just the physical creation of that mailbox. Many third-party
services exist that will process DMARC aggregate reports or the
Domain Owner can create its own set of tools. No matter which
method is chosen, the ability to parse these reports and consume
the data contained in them will go a long way to ensuring a
successful deployment.</t>
</section>

<section anchor="publish-a-dmarc-policy-for-the-author-domain-and-organizational-domain"><name>Publish a DMARC Policy for the Author Domain and Organizational Domain</name>
<t>Once SPF, DKIM, and the aggregate reports mailbox are all in place,
it's time to publish a DMARC record. For best results, Domain Owners
usually start with &quot;p=none&quot;, (see <xref target="collect-and-analyze"></xref>)
with the rua tag containing a URI that references the mailbox created
in the previous step. This is commonly referred to as putting the Author
Domain into Monitoring Mode. If the Organizational Domain is different
from the Author Domain, a record also needs to be published for the
Organizational Domain.</t>
</section>

<section anchor="collect-and-analyze"><name>Collect and Analyze Reports</name>
<t>The reason for starting at &quot;p=none&quot; is to ensure that nothing's been
missed in the initial SPF and DKIM deployments. In all but the most
trivial setups, a Domain Owner can overlook a server here or be unaware
of a third party sending agreement there.  Starting at &quot;p=none&quot;, therefore,
takes advantage of DMARC's aggregate reporting function, with the Domain
Owner using the reports to audit its own mail streams' authentication
configurations.</t>
</section>

<section anchor="decide-whether-to-update-dmarc-policy"><name>Decide Whether to Update DMARC Policy</name>
<t>Once the Domain Owner is satisfied that it is properly authenticating
all of its mail, then it is time to decide if it is appropriate to
change the p= value in its DMARC record to p=quarantine or p=reject.
Depending on its cadence for sending mail, it may take many months
of consuming DMARC aggregate reports before a Domain Owner reaches
the point where it is sure that it is properly authenticating all
of its mail, and the decision on which p= value to use will depend
on its needs.</t>
<t>In making this decision it is important to understand the
interoperability issues involved and problems that can result for
mailing lists and for deliverability of legitimate mail. Those
issues are discussed in detail in <xref target="interoperability-considerations"></xref></t>
</section>
</section>

<section anchor="pso-actions"><name>PSO Actions</name>
<t>In addition to the DMARC Domain Owner actions, if a PSO publishes a
DMARC record it <bcp14>MUST</bcp14> include the psd tag (see <xref target="general-record-format"></xref>)
with a value of 'y' (&quot;psd=y&quot;).</t>
</section>

<section anchor="mail-receiver-actions"><name>Mail Receiver Actions</name>
<t>This section describes receiver actions in the DMARC environment.</t>

<section anchor="extract-author-domain"><name>Extract Author Domain</name>
<t>The domain in the RFC5322.From header field is extracted as the domain
to be evaluated by DMARC. If the domain is a U-label, the
domain name <bcp14>MUST</bcp14> be converted to an A-label, as described in Section
2.3 of <xref target="RFC5890"></xref>, for further processing.</t>
<t>If zero or more than one domain is extracted, then DMARC processing is
not possible and the process terminates. See <xref target="denial-of-dmarc-attacks"></xref>
for further discussion.</t>
</section>

<section anchor="determine-handling-policy"><name>Determine Handling Policy</name>
<t>To arrive at a policy for an individual message, Mail Receivers <bcp14>MUST</bcp14>
perform the following actions or their semantic equivalents.
Steps 2-4 <bcp14>MAY</bcp14> be done in parallel, whereas steps 5 and 6 require
input from previous steps. Further, steps 5 and 6 <bcp14>SHOULD</bcp14> only be
performed if a DMARC policy record is found in step 2.</t>
<t>The steps are as follows:</t>

<ol>
<li><t>Extract the RFC5322.From domain from the message (as above).</t>
</li>
<li><t>Query the DNS for a DMARC policy record.  Continue if one is
found, or terminate DMARC evaluation otherwise.  See
<xref target="dmarc-policy-discovery"></xref> for details.</t>
</li>
<li><t>Perform DKIM signature verification checks.  A single email could
contain multiple DKIM signatures.  The results of this step are
passed to the remainder of the algorithm, <bcp14>MUST</bcp14> include &quot;pass&quot; or
&quot;fail&quot;, and if &quot;fail&quot;, <bcp14>SHOULD</bcp14> include information about the reasons
for failure. The results <bcp14>MUST</bcp14> further include the value of the &quot;d=&quot;
and &quot;s=&quot; tags from each checked DKIM signature.</t>
</li>
<li><t>Perform SPF verification checks.  The results of this step are
passed to the remainder of the algorithm, <bcp14>MUST</bcp14> include &quot;pass&quot; or
&quot;fail&quot;, and if &quot;fail&quot;, <bcp14>SHOULD</bcp14> include information about the reasons
for failure. The results <bcp14>MUST</bcp14> further include the domain name used
to complete the SPF check.</t>
</li>
<li><t>Conduct Identifier Alignment checks.  With authentication checks
and policy discovery performed, the Mail Receiver checks to see
if Authenticated Identifiers are aligned as described in
<xref target="identifier-alignment-explained"></xref>.  If one or more of the Authenticated
Identifiers align with the RFC5322.From domain, the message is
considered to pass the DMARC mechanism check.</t>
</li>
<li><t>Apply policy, if appropriate.  Emails that fail the DMARC mechanism
check are handled in accordance with the discovered DMARC policy of the
Domain Owner and any local policy rules enforced by the Mail Receiver.
See <xref target="general-record-format"></xref> for details.</t>
</li>
</ol>
<t>DMARC evaluation can only yield a &quot;pass&quot; result after one of the
underlying authentication mechanisms passes for an aligned
identifier. If neither passes and one or both of them fail due to a
temporary error, the Mail Receiver evaluating the message cannot
conclude that the DMARC mechanism had a permanent failure; they,
therefore, cannot apply the advertised DMARC policy.  When otherwise
appropriate, Mail Receivers <bcp14>MAY</bcp14> send feedback reports regarding
temporary errors.</t>
<t>Handling of messages for which SPF and/or DKIM evaluation encounter a
permanent DNS error is left to the discretion of the Mail Receiver.</t>
</section>

<section anchor="store-results-of-dmarc-processing"><name>Store Results of DMARC Processing</name>
<t>Mail Receiver-based DMARC processing results should be stored
for eventual presentation back to the Domain Owner in the form of
aggregate feedback reports.  <xref target="general-record-format"></xref> and
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> discuss aggregate feedback.</t>
</section>

<section anchor="send-aggregate-reports"><name>Send Aggregate Reports</name>
<t>For a Domain Owner, DMARC aggregate reports provide data about all
mailstreams making use of its domain in email, to include not only
illegitimate uses but also, and perhaps more importantly, all
legitimate uses. Domain Owners can use aggregate reports to ensure
that all legitimate uses of their domain for sending email are
properly authenticated, and once they are, express a stricter message
handling preference in the p= tag in their DMARC policy records from
none to quarantine to reject, if appropriate. In turn, DMARC policy
records with p= tag values of 'quarantine' or 'reject' are higher
value signals to Mail Receivers, ones that can assist Mail Receivers
with handling decisions for a message in ways that p= tag values of
'none' cannot.</t>
<t>Given the above, to ensure maximum usefulness for DMARC across
the email ecosystem, Mail Receivers <bcp14>SHOULD</bcp14> generate and send
aggregate reports with a frequency of at least once every 24 hours.</t>
</section>
</section>

<section anchor="policy-enforcement-considerations"><name>Policy Enforcement Considerations</name>
<t>Mail Receivers <bcp14>MAY</bcp14> choose to reject or quarantine email even if email
passes the DMARC mechanism check. The DMARC mechanism does not
inform Mail Receivers whether an email stream is &quot;good&quot;; a DMARC result
of &quot;pass&quot; only means the domain in the RFC5322.From header has been
verified by the DMARC mechanism. Mail Receivers are encouraged to maintain
anti-abuse technologies to combat the possibility of DMARC-enabled criminal
campaigns.</t>
<t>Mail Receivers <bcp14>MAY</bcp14> choose to accept email that fails the DMARC
mechanism check even if the published Domain Owner Assessment Policy
is &quot;reject&quot;. In particular, because of the considerations discussed
in <xref target="RFC7960"></xref> and in <xref target="interoperability-considerations"></xref> of
this document, it is important that Mail Receivers not reject messages
solely because of a published policy of &quot;reject&quot;, but that they apply
other knowledge and analysis to avoid situations such as rejection of
legitimate messages sent in ways that DMARC cannot describe, harm to
the operation of mailing lists, and similar.</t>
<t>If they choose not to honor the published Domain Owner Assessment
Policy to improve interoperability among mail systems,
it may increase the likelihood of accepting abusive mail.
At a minimum, Mail Receivers <bcp14>SHOULD</bcp14> add the Authentication-Results
header field (see <xref target="RFC8601"></xref>), and it is <bcp14>RECOMMENDED</bcp14> when delivering
failing mail.</t>
<t>When Mail Receivers deviate from a published Domain Owner
Assessment Policy during message processing they <bcp14>SHOULD</bcp14> make
available the fact of and reason for the deviation to the Domain
Owner via feedback reporting, specifically using the
&quot;PolicyOverride&quot; feature of the aggregate report defined in
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref>.</t>
<t>The final handling of a message is always a matter of local policy.
An operator that wishes to favor DMARC policy over SPF policy, for
example, will disregard the SPF policy since enacting an
SPF-determined rejection prevents evaluation of DKIM; DKIM might
otherwise pass, satisfying the DMARC evaluation. There is a
trade-off to doing so, namely acceptance and processing of the entire
message body in exchange for the enhanced protection DMARC provides.</t>
<t>DMARC-compliant Mail Receivers typically disregard any mail-handling
directive discovered as part of an authentication mechanism (e.g.,
Author Domain Signing Practices (ADSP) <xref target="RFC5617"></xref>, SPF) where a DMARC
record is also discovered that specifies a policy other than &quot;none&quot;.
Deviating from this practice introduces inconsistency among DMARC operators
in terms of handling the message.  However, such deviation is not
proscribed.</t>
<t>To enable Domain Owners to receive DMARC feedback without impacting
existing mail processing, discovered policies of &quot;p=none&quot; <bcp14>MUST NOT</bcp14>
modify existing mail handling processes.</t>
<t>Mail Receivers <bcp14>MUST</bcp14> also implement reporting instructions of DMARC,
even in the absence of a request for DKIM reporting <xref target="RFC6651"></xref> or
SPF reporting <xref target="RFC6652"></xref>. Furthermore, the presence of such requests
<bcp14>MUST NOT</bcp14> affect DMARC reporting.</t>
</section>
</section>

<section anchor="dmarc-feedback"><name>DMARC Feedback</name>
<t>Providing Domain Owners with visibility into how Mail Receivers
implement and enforce the DMARC mechanism in the form of feedback is
critical to establishing and maintaining accurate authentication
deployments. When Domain Owners can see what effect their policies
and practices are having, they are more willing and able to use
quarantine and reject policies.</t>
<t>The details of this feedback are described in <xref target="I-D.ietf-dmarc-aggregate-reporting"></xref></t>
<t>Operational note for PSD DMARC: For PSOs, feedback for non-existent
domains is desirable and useful, just as it is for org-level DMARC
operators. See <xref target="privacy-considerations"></xref> for discussion of
Privacy Considerations for PSD DMARC.</t>
</section>

<section anchor="rfc7849-changes"><name>Changes from RFC 7489</name>
<t>This document is intended to render obsolete <xref target="RFC7489"></xref>. As one might guess,
that means there are significant differences between RFC 7489 and this
document. This section will summarize those changes.</t>

<section anchor="ietf-category"><name>IETF Category</name>
<t>RFC 7489 was not an Internet Standards Track specification; it was instead
published in the Informational Category. This document, by contrast, is intended
to be Internet Standards Track.</t>
</section>

<section anchor="changes-to-terminology-and-definitions"><name>Changes to Terminology and Definitions</name>
<t>The following changes were made to the Terminology and Definitions section.</t>

<section anchor="terms-added"><name>Terms Added</name>
<t>These terms were added:</t>

<ul spacing="compact">
<li>Enforcement</li>
<li>Monitoring Mode</li>
<li>Non-existent Domains</li>
<li>Public Suffix Domain (PSD)</li>
<li>Public Suffix Operator (PSO)</li>
<li>PSO Controlled Domain Names</li>
</ul>
</section>

<section anchor="definitions-updated"><name>Definitions Updated</name>
<t>These definitions were updated:</t>

<ul spacing="compact">
<li>Organizational Domain</li>
<li>Report Receiver (renamed to Report Consumer)</li>
</ul>
</section>
</section>

<section anchor="policy-discovery-and-organizational-domain-determination"><name>Policy Discovery and Organizational Domain Determination</name>
<t>The algorithms for DMARC policy discovery and for determining the Organizational Domain
have been changed. Specifically, reliance on the Public Suffix List (PSL) has been replaced
by a technique called a &quot;DNS Tree Walk&quot;, and the methodology for the DNS Tree Walk is explained
in detail in this document.</t>
<t>The DNS Tree Walk also incorporates PSD policy discovery, which was introduced in <xref target="RFC9091"></xref>.
<xref target="RFC9091"></xref> was an Experimental RFC, and the results of that experiment were that the RFC was
not implemented as written. Instead, this document redefines the algorithm for PSD policy
discovery, and thus obsoletes <xref target="RFC9091"></xref>.</t>
</section>

<section anchor="reporting"><name>Reporting</name>
<t>Discussion of both aggregate and failure reporting have been moved to separate documents
dedicated to the topics.</t>
</section>

<section anchor="tags"><name>Tags</name>
<t>Several tags have been added to the &quot;General Record Format&quot; section of this document since
RFC 7489 was published, and at the same time, several others were removed.</t>

<section anchor="tags-added"><name>Tags Added:</name>

<ul spacing="compact">
<li>np - Policy for non-existent domains (Imported from <xref target="RFC9091"></xref>)</li>
<li>psd - Flag indicating whether a domain is a Public Suffix Domain</li>
<li>t - Replacement for some pct tag functionality. See <xref target="removal-of-the-pct-tag"></xref> for further discussion</li>
</ul>
</section>

<section anchor="tags-removed"><name>Tags Removed:</name>

<ul spacing="compact">
<li>pct - Tag requesting application of DMARC policy to only a percentage of messages</li>
<li>rf - Tag specifying requested format of failure reports</li>
<li>ri - Tag specifying requested interval between aggregate reports</li>
</ul>
</section>
</section>

<section anchor="expansion-of-domain-owner-actions-section"><name>Expansion of Domain Owner Actions Section</name>
<t>This section has been expanded upon from RFC 7489.</t>
<t>RFC 7489 had just two paragraphs in its Domain Owner Actions section, and while
the content of those paragraphs was correct, it was minimalist in its approach to
providing guidance to domain owners on just how to implement DMARC.</t>
<t>This document provides much more detail and explanatory text to a domain owner,
focusing not just on what to do to implement DMARC, but also on the reasons for
each step and the repercussions of each decision.</t>
<t>In particular, this document makes explicit that domains for general-purpose
email <bcp14>MUST NOT</bcp14> deploy a DMARC policy of p=reject.</t>
</section>

<section anchor="report-generator-recommendations"><name>Report Generator Recommendations</name>
<t>In the cases where a DMARC policy record specifies multiple destinations for either aggregate
reports or failure reports, RFC 7489 stated:</t>

<artwork>  Receivers MAY impose a limit on the number of URIs to which they
  will send reports but MUST support the ability to send to at least
  two.
</artwork>
<t>This document in <xref target="dmarc-uris"></xref> says:</t>

<artwork>  A report SHOULD be sent to each listed URI provided in the DMARC 
  record.
</artwork>
</section>

<section anchor="general-editing-and-formatting"><name>General Editing and Formatting</name>
<t>A great deal of the content from RFC 7489 was preserved in this document, but much
of it was subject to either minor editing, re-ordering of sections, and/or both.</t>
</section>
</section>

<section anchor="other-topics"><name>Other Topics</name>
<t>This section discusses some topics regarding choices made in the
development of DMARC, largely to commit the history to record.</t>

<section anchor="issues-specific-to-spf"><name>Issues Specific to SPF</name>
<t>Though DMARC does not inherently change the semantics of an SPF
policy record, historically lax enforcement of such policies has led
many to publish extremely broad records containing many extensive network
ranges. Domain Owners are strongly encouraged to carefully review
their SPF records to understand which networks are authorized to send
on behalf of the Domain Owner before publishing a DMARC record.</t>
<t>Some Mail Receiver architectures might implement SPF in advance of any
DMARC operations. This means that a &quot;-&quot; prefix on a sender's SPF
mechanism, such as &quot;-all&quot;, could cause that rejection to go into
effect early in handling, causing message rejection before any DMARC
processing takes place. Operators choosing to use &quot;-all&quot; should be
aware of this.</t>
</section>

<section anchor="dns-load-and-caching"><name>DNS Load and Caching</name>
<t>DMARC policies are communicated using the DNS and therefore inherit a
number of considerations related to DNS caching. The inherent
conflict between freshness and the impact of caching on the reduction
of DNS-lookup overhead should be considered from the Mail Receiver's
point of view. If Domain Owners or PSOs publish a DNS record with a very
short TTL,  the injection of large volumes of messages could cause
Receivers to overwhelm the publisher's DNS.
Although this is not a concern specific to DMARC, the implications of
a very short TTL should be considered when publishing DMARC policies.</t>
<t>Conversely, long TTLs will cause records to be cached for long
periods. This can cause a critical change to DMARC
parameters advertised by a Domain Owner or PSO to go unnoticed for the
length of the TTL (while waiting for DNS caches to expire). Avoiding
this problem can mean shorter TTLs, with the potential problems
described above. A balance should be sought to maintain
responsiveness of DMARC preference changes while preserving the
benefits of DNS caching.</t>
</section>

<section anchor="rejecting-messages"><name>Rejecting Messages</name>
<t>This protocol calls for rejection of a message during the SMTP
session under certain circumstances. This is preferable to
generation of a Delivery Status Notification <xref target="RFC3464"></xref>, since
fraudulent messages caught and rejected using DMARC would then result
in the annoying generation of such failure reports that go back to the
RFC5321.MailFrom address.</t>
<t>This synchronous rejection is typically done in one of two ways:</t>

<ul>
<li><t>Full rejection, wherein the SMTP server issues a 5xy reply code as
an indication to the SMTP client that the transaction failed; the
SMTP client is then responsible for generating a notification that
delivery failed (see <xref target="RFC5321" sectionFormat="of" relative="#" section="4.2.5"></xref>).</t>
</li>
<li><t>A &quot;silent discard&quot;, wherein the SMTP server returns a 2xy reply
code implying to the client that delivery (or, at least, relay)
was successfully completed, but then simply discarding the message
with no further action.</t>
</li>
</ul>
<t>Each of these has a cost. For instance, a silent discard can help to
prevent backscatter, but it also effectively means that the SMTP
server has to be programmed to give a false result, which can
confound external debugging efforts.</t>
<t>Similarly, the text portion of the SMTP reply may be important to
consider. For example, when rejecting a message, revealing the
reason for the rejection might give an attacker enough information to
bypass those efforts on a later attempt, though it might also assist
a legitimate client to determine the source of some local issue that
caused the rejection.</t>
<t>In the latter case, when doing an SMTP rejection, providing a clear
hint can be useful in resolving issues. A Mail Receiver might indicate
in plain text the reason for the rejection by using the word &quot;DMARC&quot;
somewhere in the reply text. For example:</t>

<artwork>550 5.7.1 Email rejected per DMARC policy for example.com
</artwork>
<t>Many systems are able to scan the SMTP reply text to determine the nature
of the rejection. Thus, providing a machine-detectable reason for rejection
allows the problems causing rejections to be properly addressed by automated systems.</t>
<t>If a Mail Receiver elects to defer delivery due to the inability to
retrieve or apply DMARC policy, this is best done with a 4xy SMTP
reply code.</t>
</section>

<section anchor="identifier-alignment-considerations"><name>Identifier Alignment Considerations</name>
<t>The DMARC mechanism allows both DKIM and SPF-authenticated
identifiers to authenticate email on behalf of a Domain Owner and,
possibly, on behalf of different subdomains. If malicious or unaware
users can gain control of the SPF record or DKIM selector records for
a subdomain, the subdomain can be used to generate DMARC-passing
email on behalf of the Organizational Domain.</t>
<t>For example, an attacker who controls the SPF record for
&quot;evil.example.com&quot; can send mail with an RFC5322.From header field
containing &quot;foo@example.com&quot; that can pass both authentication and
the DMARC check against &quot;example.com&quot;.</t>
<t>The Organizational Domain administrator should be careful not to
delegate control of subdomains if this is an issue, and consider
using the &quot;strict&quot; Identifier Alignment option if appropriate.</t>
</section>

<section anchor="interoperability-issues"><name>Interoperability Issues</name>
<t>DMARC limits which end-to-end scenarios can achieve a &quot;pass&quot; result.</t>
<t>Because DMARC relies on SPF <xref target="RFC7208"></xref> and/or DKIM <xref target="RFC6376"></xref> to achieve
a &quot;pass&quot;, their limitations also apply.</t>
<t>Issues specific to the use of policy mechanisms alongside DKIM are
further discussed in <xref target="RFC6377"></xref>, particularly Section 5.2.</t>
<t>Mail that is sent by authorized, independent third parties might not be
sent with Identifier Alignment, also preventing a &quot;pass&quot; result.</t>
</section>

<section anchor="interoperability-considerations"><name>Interoperability Considerations</name>
<t>As discussed in &quot;Interoperability Issues between DMARC and Indirect
Email Flows&quot; <xref target="RFC7960"></xref>, use of p=reject can be incompatible with and
cause interoperability problems to indirect message flows such as
&quot;alumni forwarders&quot;, role-based email aliases, and mailing lists
across the Internet.</t>
<t>A domain that expects to send only targeted messages to account holders
- a bank, for example - could have account holders using addresses such
as jones@alumni.example.edu (an address that relays the messages to
another address with a real mailbox) or finance@association.example
(a role-based address that does similar relaying for the current head
of finance at the association).  When such mail is delivered to the
actual recipient mailbox, it will necessarily fail SPF checks, as the
incoming IP address will be that of example.edu or association.example,
and not an address authorized for the sending domain.  DKIM signatures
will generally remain valid in these relay situations.</t>
<blockquote><t>It is therefore critical that domains that publish p=reject
<bcp14>MUST NOT</bcp14> rely solely on SPF to secure a DMARC pass, and
<bcp14>MUST</bcp14> apply valid DKIM signatures to their messages.</t>
</blockquote><t>In the case of domains that have general users who send routine email,
those that publish p=reject are likely to create significant interoperability
issues. In particular, if users in p=reject domains post messages to mailing
lists on the Internet, those messages can cause operational problems for the
mailing lists and for the subscribers to those lists, as explained below and
in <xref target="RFC7960"></xref>.</t>
<blockquote><t>It is therefore critical that domains that host users who might
post messages to mailing lists <bcp14>SHOULD NOT</bcp14> publish p=reject.
Any such domains wishing to publish p=reject <bcp14>SHOULD</bcp14> first
take advantage of DMARC aggregate report data for their domain to
determine the possible impact to their users, first by publishing
p=none for at least a month, followed by publishing p=quarantine for
an equally long period of time, and comparing the message disposition
results. Domains that choose to publish p=reject <bcp14>SHOULD</bcp14> either
implement policies that their users not post to Internet mailing lists
and/or inform their users that their participation in mailing lists may
be hindered.</t>
</blockquote><t>As noted in <xref target="policy-enforcement-considerations"></xref>, receiving domains
need to apply more analysis than just DMARC evaluation in their
disposition of incoming messages.  An example of the consequences of
honoring p=reject without further anaysis is that rejecting messages
that have been relayed by a mailing list can cause your own users to
have their subscriptions to that mailing list cancelled by the
list software's automated handling of such rejections - it looks
to the list manager as though the recipient's email address is no
longer working, so the address is automatically unsubscribed.</t>
<blockquote><t>It is therefore critical that receiving domains <bcp14>MUST NOT</bcp14> reject
incoming messages solely on the basis of a p=reject policy by
the sending domain.  Receiving domains must use the DMARC
policy as part of their disposition decision, along with other
knowledge and analysis.</t>
</blockquote><t>Failure to understand and abide by these considerations can cause
legitimate, sometimes important email to be rejected, can cause
operational damage to mailing lists throughout the Internet, and
can result in trouble-desk calls and complaints from your own
employees, customers, and clients.</t>
</section>
</section>

<section anchor="iana-considerations"><name>IANA Considerations</name>
<t>This section describes actions completed by IANA.</t>

<section anchor="authentication-results-method-registry-update"><name>Authentication-Results Method Registry Update</name>
<t>IANA has added the following to the &quot;Email Authentication Methods&quot;
registry:</t>
<table align="left"><name>&quot;Authentication-Results Method Registry Update&quot;
</name>
<thead>
<tr>
<th align="left">Method</th>
<th align="left">Defined</th>
<th align="left">ptype</th>
<th align="left">Property</th>
<th align="left">Value</th>
<th align="left">Status</th>
<th align="left">Version</th>
</tr>
</thead>

<tbody>
<tr>
<td align="left">dmarc</td>
<td align="left"><xref target="RFC7489"></xref></td>
<td align="left">header</td>
<td align="left">from</td>
<td align="left">the domain portion of the RFC5322.From header field</td>
<td align="left">active</td>
<td align="left">1</td>
</tr>

<tr>
<td align="left">dmarc</td>
<td align="left"><xref target="RFC7489"></xref></td>
<td align="left">polrec</td>
<td align="left">p</td>
<td align="left">the p= value read from the discovered policy record</td>
<td align="left">active</td>
<td align="left">1</td>
</tr>

<tr>
<td align="left">dmarc</td>
<td align="left"><xref target="RFC7489"></xref></td>
<td align="left">polrec</td>
<td align="left">domain</td>
<td align="left">the domain at which the policy record was discovered, if different from the RFC5322.From domain</td>
<td align="left">active</td>
<td align="left">1</td>
</tr>
</tbody>
</table></section>

<section anchor="authentication-results-result-registry-update"><name>Authentication-Results Result Registry Update</name>
<t>IANA has added the following in the &quot;Email Authentication Result
Names&quot; registry:</t>
<table align="left"><name>&quot;Authentication-Results Result Registry Update&quot;
</name>
<thead>
<tr>
<th align="left">Code</th>
<th align="left">Existing/New Code</th>
<th align="left">Defined</th>
<th align="left">Auth Method</th>
<th align="left">Meaning</th>
<th align="left">Status</th>
</tr>
</thead>

<tbody>
<tr>
<td align="left">none</td>
<td align="left">existing</td>
<td align="left"><xref target="RFC8601"></xref></td>
<td align="left">dmarc (added)</td>
<td align="left">No DMARC policy record was published for the aligned identifier, or no aligned identifier could be extracted.</td>
<td align="left">active</td>
</tr>

<tr>
<td align="left">pass</td>
<td align="left">existing</td>
<td align="left"><xref target="RFC8601"></xref></td>
<td align="left">dmarc (added)</td>
<td align="left">A DMARC policy record was published for the aligned identifier, and at least one of the authentication mechanisms passed.</td>
<td align="left">active</td>
</tr>

<tr>
<td align="left">fail</td>
<td align="left">existing</td>
<td align="left"><xref target="RFC8601"></xref></td>
<td align="left">dmarc (added)</td>
<td align="left">A DMARC policy record was published for the aligned identifier, and none of the authentication mechanisms passed.</td>
<td align="left">active</td>
</tr>

<tr>
<td align="left">temperror</td>
<td align="left">existing</td>
<td align="left"><xref target="RFC8601"></xref></td>
<td align="left">dmarc (added)</td>
<td align="left">A temporary error occurred during DMARC evaluation. A later attempt might produce a final result.</td>
<td align="left">active</td>
</tr>

<tr>
<td align="left">permerror</td>
<td align="left">existing</td>
<td align="left"><xref target="RFC8601"></xref></td>
<td align="left">dmarc (added)</td>
<td align="left">A permanent error occurred during DMARC evaluation, such as encountering a syntactically incorrect DMARC record. A later attempt is unlikely to produce a final result.</td>
<td align="left">active</td>
</tr>
</tbody>
</table></section>

<section anchor="feedback-report-header-fields-registry-update"><name>Feedback Report Header Fields Registry Update</name>
<t>The following has been added to the &quot;Feedback Report Header Fields&quot;
registry:</t>
<t>Field Name:  Identity-Alignment</t>

<dl spacing="compact">
<dt>Description:</dt>
<dd>indicates whether the message about which a report is
being generated had any identifiers in alignment as defined in
RFC 7489</dd>
</dl>
<t>Multiple Appearances:  No</t>
<t>Related &quot;Feedback-Type&quot;:  auth-failure</t>
<t>Reference:  RFC 7489</t>
<t>Status:  current</t>
</section>

<section anchor="dmarc-tag-registry"><name>DMARC Tag Registry</name>
<t>A new registry tree called &quot;Domain-based Message Authentication,
Reporting, and Conformance (DMARC) Parameters&quot; has been created.
Within it, a new sub-registry called the &quot;DMARC Tag Registry&quot; has
been created.</t>
<t>Names of DMARC tags are registered with IANA in this new
sub-registry. New entries are assigned only for values that have
been documented in a manner that satisfies the terms of Specification
Required, per <xref target="RFC8126"></xref>. Each registration includes
the tag name; the specification that defines it; a brief description;
and its status, which is one of &quot;current&quot;, &quot;experimental&quot;, or
&quot;historic&quot;. The Designated Expert needs to confirm that the provided
specification adequately describes the new tag and clearly presents
how it would be used within the DMARC context by Domain Owners and
Mail Receivers.</t>
<t>To avoid version compatibility issues, tags added to the DMARC
specification are to avoid changing the semantics of existing records
when processed by implementations conforming to prior specifications.</t>
<t>The initial set of entries in this registry is as follows:</t>
<table align="left"><name>&quot;DMARC Tag Registry&quot;
</name>
<thead>
<tr>
<th align="left">Tag Name</th>
<th align="left">Reference</th>
<th align="left">Status</th>
<th align="left">Description</th>
</tr>
</thead>

<tbody>
<tr>
<td align="left">adkim</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">DKIM alignment mode</td>
</tr>

<tr>
<td align="left">aspf</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">SPF alignment mode</td>
</tr>

<tr>
<td align="left">fo</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">Failure reporting options</td>
</tr>

<tr>
<td align="left">np</td>
<td align="left">RFC 9091</td>
<td align="left">current</td>
<td align="left">Requested handling policy for non-existent subdomains</td>
</tr>

<tr>
<td align="left">p</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">Requested handling policy</td>
</tr>

<tr>
<td align="left">pct</td>
<td align="left">RFC 7489</td>
<td align="left">historic</td>
<td align="left">Sampling rate</td>
</tr>

<tr>
<td align="left">psd</td>
<td align="left">[this document]</td>
<td align="left">current</td>
<td align="left">Indicates whether policy record is published by a Public Suffix Domain</td>
</tr>

<tr>
<td align="left">rf</td>
<td align="left">RFC 7489</td>
<td align="left">historic</td>
<td align="left">Failure reporting format(s)</td>
</tr>

<tr>
<td align="left">ri</td>
<td align="left">RFC 7489</td>
<td align="left">historic</td>
<td align="left">Aggregate Reporting interval</td>
</tr>

<tr>
<td align="left">rua</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">Reporting URI(s) for aggregate data</td>
</tr>

<tr>
<td align="left">ruf</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">Reporting URI(s) for failure data</td>
</tr>

<tr>
<td align="left">sp</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">Requested handling policy for subdomains</td>
</tr>

<tr>
<td align="left">t</td>
<td align="left">[this document]</td>
<td align="left">current</td>
<td align="left">Test mode for the specified policy</td>
</tr>

<tr>
<td align="left">v</td>
<td align="left">RFC 7489</td>
<td align="left">current</td>
<td align="left">Specification version</td>
</tr>
</tbody>
</table></section>

<section anchor="dmarc-report-format-registry"><name>DMARC Report Format Registry</name>
<t>Also, within &quot;Domain-based Message Authentication, Reporting, and
Conformance (DMARC) Parameters&quot;, a new sub-registry called &quot;DMARC
Report Format Registry&quot; has been created.</t>
<t>Names of DMARC failure reporting formats are registered with IANA
in this registry. New entries are assigned only for values that
satisfy the definition of Specification Required, per
<xref target="RFC8126"></xref>.  In addition to a reference to a permanent
specification, each registration includes the format name, a
brief description, and its status, which must be one of &quot;current&quot;,
&quot;experimental&quot;, or &quot;historic&quot;. The Designated Expert needs to
confirm that the provided specification adequately describes the
report format and clearly presents how it would be used within the
DMARC context by Domain Owners and Mail Receivers.</t>
<t>The initial entry in this registry is as follows:</t>
<table align="left"><name>&quot;DMARC Report Format Registry&quot;
</name>
<thead>
<tr>
<th>Format Name</th>
<th>Reference</th>
<th>Status</th>
<th>Description</th>
</tr>
</thead>

<tbody>
<tr>
<td>afrf</td>
<td>RFC 7489</td>
<td>current</td>
<td>Authentication Failure Reporting Format (see <xref target="RFC6591"></xref>)</td>
</tr>
</tbody>
</table></section>

<section anchor="underscored-and-globally-scoped-dns-node-names-registry"><name>Underscored and Globally Scoped DNS Node Names Registry</name>
<t>Per <xref target="RFC8552"></xref>, please add the following entry to the &quot;Underscored
and Globally Scoped DNS Node Names&quot; registry:</t>
<table align="left"><name>&quot;Underscored and Globally Scoped DNS Node Names&quot; registry
</name>
<thead>
<tr>
<th>RR Type</th>
<th>_NODE NAME</th>
<th>Reference</th>
</tr>
</thead>

<tbody>
<tr>
<td>TXT</td>
<td>_dmarc</td>
<td>RFC 7489</td>
</tr>
</tbody>
</table></section>
</section>

<section anchor="privacy-considerations"><name>Privacy Considerations</name>
<t>This section discusses issues specific to private data that may be
included if DMARC reports are requested.  Issues associated with
sending aggregate reports and failure reports are addressed in
<xref target="I-D.ietf-dmarc-aggregate-reporting"></xref> and
<xref target="I-D.ietf-dmarc-failure-reporting"></xref> respectively.</t>

<section anchor="aggregate-report-considerations"><name>Aggregate Report Considerations</name>
<t>Aggregate reports may, particularly for small organizations, provide some
limited insight into email sending patterns.  As an example, in a small
organization, an aggregate report from a particular domain may be sufficient
to make the report receiver aware of sensitive personal or business
information.  If setting an rua= tag in a DMARC record, the reporting
address needs controls appropriate to the organizational requirements to
mitigate any risk associated with receiving and handling reports.</t>
<t>In the case of rua= requests for multi-organizational PSDs, additional
information leakage considerations exist.  Multi-organizational PSDs that
do not mandate DMARC use by registants risk exposure of private data of
registrant domains if they include the rua= tag in their DMARC record.</t>
</section>

<section anchor="failure-report-considerations"><name>Failure Report Considerations</name>
<t>Failure reports do provide insight into email sending patterns, including
specific users.  If requesting failure reports, data management controls
are needed to support appropriate management of this information.  The
additional detail available through failure reports (relative to aggregate
reports) can drive a need for additional controls.  As an example, a
company may be legally restricted from receiving data related to a specific
subsidiary.  Before requesting failure reports, any such data spillage risks
have to be addressed through data management controls or publishing DMARC
records for relevant sub-domains to prevent reporting on data related to
their emails.</t>
<t>Out of band agreements between failure report senders and receivers may be
required to address privacy concerns.</t>
<t>DMARC records for multi-organizational PSDs <bcp14>MUST NOT</bcp14> include the ruf= tag.</t>
</section>
</section>

<section anchor="security-considerations"><name>Security Considerations</name>
<t>This section discusses security issues and possible remediations
(where available) for DMARC.</t>

<section anchor="authentication-methods"><name>Authentication Methods</name>
<t>Security considerations from the authentication methods used by DMARC
are incorporated here by reference.</t>
</section>

<section anchor="attacks-on-reporting-uris"><name>Attacks on Reporting URIs</name>
<t>URIs published in DNS TXT records are well-understood possible
targets for attack.  Specifications such as <xref target="RFC1035"></xref> and <xref target="RFC2142"></xref> either
expose or cause the exposure of email addresses that could be flooded
by an attacker, for example. Records found in the DNS such as MX, NS,
and others advertise potential attack destinations. Common DNS names such
as &quot;www&quot; plainly identify the locations at which particular services can
be found, providing destinations for targeted denial-of-service or
penetration attacks.  This all means that Domain Owners will need to harden
these addresses against various attacks, including but not limited to:</t>

<ul>
<li><t>high-volume denial-of-service attacks;</t>
</li>
<li><t>deliberate construction of malformed reports intended to identify
or exploit parsing or processing vulnerabilities;</t>
</li>
<li><t>deliberate construction of reports containing false claims for the
Submitter or Reported-Domain fields, including the possibility of
false data from compromised but known Mail Receivers.</t>
</li>
</ul>
</section>

<section anchor="dns-security"><name>DNS Security</name>
<t>The DMARC mechanism and its underlying technologies (SPF, DKIM)
depend on the security of the DNS. Examples of how hostile parties can
have an adverse impact on DNS traffic include:</t>

<ul>
<li><t>If they can snoop on DNS traffic, they can get an idea of who is
sending mail.</t>
</li>
<li><t>If they can block outgoing or reply DNS messages, they can prevent
systems from discovering senders' DMARC policies, causing recipients
to assume p=none by default.</t>
</li>
<li><t>If they can send forged response packets, they can make aligned mail
appear unaligned or vice-versa.</t>
</li>
</ul>
<t>None of these threats are unique to DMARC, and they can be addressed using
a variety of techniques, including, but not limited to:</t>

<ul>
<li><t>Signing DNS records with DNSSEC <xref target="RFC4033"></xref>, which enables recipients to
verify the integrity of DNS data and
detect and discard forged responses.</t>
</li>
<li><t>DNS over TLS <xref target="RFC7858"></xref> or DNS over HTTPS <xref target="RFC8484"></xref> can mitigate snooping
and forged responses.</t>
</li>
</ul>
</section>

<section anchor="display-name-attacks"><name>Display Name Attacks</name>
<t>A common attack in messaging abuse is the presentation of false
information in the display-name portion of the RFC5322.From header field.
For example, it is possible for the email address in that field to be
an arbitrary address or domain name while containing a well-known
name (a person, brand, role, etc.) in the display name, intending to
fool the end user into believing that the name is used legitimately.
The attack is predicated on the notion that most common MUAs will
show the display name and not the email address when both are
available.</t>
<t>Generally, display name attacks are out of scope for DMARC, as
further exploration of possible defenses against these attacks needs
to be undertaken.</t>
<t>There are a few possible mechanisms that attempt mitigation of these
attacks, such as the following:</t>

<ul>
<li><t>If the display name includes an email address (as
specified in <xref target="RFC5322"></xref>), execute the DMARC mechanism on the domain
name found there rather than the original domain name. However, this
addresses only a very specific attack
space, and spoofers can easily circumvent it by simply not using
an email address in the display name. There are also known cases
of legitimate uses of an email address in the display name with a
domain different from the one in the address portion, e.g.,</t>
<t>From: &quot;user@example.org via Bug Tracker&quot; <eref target="mailto:support@example.com">support@example.com</eref></t>
</li>
<li><t>In the MUA, only show the display name if the DMARC mechanism
succeeds. This too is easily defeated, as an attacker could
arrange to pass the DMARC tests while fraudulently using another
domain name in the display name.</t>
</li>
<li><t>In the MUA, only show the display name if the DMARC mechanism
passes and the email address thus verified matches one found in
the receiving user's list of known addresses.</t>
</li>
</ul>
</section>

<section anchor="denial-of-dmarc-attacks"><name>Denial of DMARC Processing Attacks</name>
<t>The declaration in <xref target="extract-author-domain"></xref> and elsewhere in this document
that messages that do not contain precisely one RFC5322.From domain are
outside the scope of this document exposes an attack vector that must be
taken into consideration.</t>
<t>Because such messages are outside the scope of this document, an attacker
can craft messages with multiple RFC5322.From domains, including the spoofed
domain, in an effort to bypass DMARC validation and get the fraudulent message
to be displayed by the victim's MUA with the spoofed domain successfully shown
to the victim. In those cases where such messages are not rejected due to other
reasons (for example, many such messages would violate RFC5322's requirement that
there be precisely one From: header), care must be taken by the receiving MTA
to recognize such messages as the threats they might be and handle them
appropriately.</t>
</section>

<section anchor="external-report-addresses"><name>External Reporting Addresses</name>
<t>To avoid abuse by bad actors, reporting addresses generally have to
be inside the domains about which reports are requested.  To
accommodate special cases such as a need to get reports about domains
that cannot actually receive mail, <xref target="I-D.ietf-dmarc-aggregate-reporting" sectionFormat="of" relative="#" section="3"></xref> describes
a DNS-based mechanism for verifying approved external reporting.</t>
<t>The obvious consideration here is an increased DNS load against
domains that are claimed as external recipients. Negative caching
will mitigate this problem, but only to a limited extent, mostly
dependent on the default TTL in the domain's SOA record.</t>
<t>Where possible, external reporting is best achieved by having the
report be directed to domains that can receive mail and simply having
it automatically forwarded to the desired external destination.</t>
<t>Note that the addresses shown in the &quot;ruf&quot; tag receive more
information that might be considered private data since it is
possible for actual email content to appear in the failure reports.
The URIs identified there are thus more attractive targets for
intrusion attempts than those found in the &quot;rua&quot; tag. Moreover,
attacking the DNS of the subject domain to cause failure data to be
routed fraudulently to an attacker's systems may be an attractive
prospect. Deployment of <xref target="RFC4033"></xref> is advisable if this is a concern.</t>
</section>

<section anchor="secure-protocols"><name>Secure Protocols</name>
<t>This document encourages the use of secure transport mechanisms to
prevent the loss of private data to third parties that may be able to
monitor such transmissions. Unencrypted mechanisms should be
avoided.</t>
<t>In particular, a message that was originally encrypted or otherwise
secured might appear in a report that is not sent securely, which
could reveal private information.</t>
</section>

<section anchor="determine-org-domain-relaxed"><name>Determination of the Organizational Domain For Relaxed Alignment</name>
<t>DMARC evaluation for relaxed alignment is highly sensitive to errors in
determining the organizational domain if the RFC5322.From domain does not
have a published policy. If an incorrectly selected organizational domain is
a parent of the correct organizational domain, then relaxed alignment could
potentially allow a malicious sender to obtain DMARC PASS. This potential
exists for both the legacy <xref target="RFC7489"></xref> and current methods for determining the
organizational domain, the latter described in <xref target="organizational-domain-discovery"></xref>.</t>
<t>This issue is entirely avoided by the use of strict alignment and publishing
DMARC records for all domains/sub-domains used as RFC5322.From domain in an
organization's email.</t>
<t>For cases where strict alignment is not appropriate, this issue can be
mitigated by periodically checking the DMARC records, if any, of PSDs above
the organization's domains in the DNS tree and (for legacy <xref target="RFC7489"></xref> checking
that appropriate PSL entries remain present). If a PSD domain publishes a
DMARC record without the appropriate psd=y tag, organizational domain owners
can add psd=n to their organizational domain's DMARC record so that the PSD
record will not be incorrectly evaluated to be the organizational domain</t>
</section>
</section>

</middle>

<back>
<references><name>Normative References</name>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-dmarc-aggregate-reporting.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml-ids/reference.I-D.ietf-dmarc-failure-reporting.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.1035.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2119.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3986.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4343.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5234.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5321.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5322.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5890.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6376.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6591.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6651.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6652.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7208.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7405.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7489.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8552.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.9091.xml"/>
</references>
<references><name>Informative References</name>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2142.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.2308.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.3464.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.4033.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5598.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.5617.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.6377.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7858.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.7960.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8020.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8126.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8174.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8484.xml"/>
<xi:include href="https://xml2rfc.ietf.org/public/rfc/bibxml/reference.RFC.8601.xml"/>
</references>

<section anchor="technology-considerations"><name>Technology Considerations</name>
<t>This section documents some design decisions made in the
development of DMARC. Specifically addressed here are some
suggestions that were considered but not included in the design,
with explanatory text regarding the decision.</t>

<section anchor="s-mime"><name>S/MIME</name>
<t>S/MIME, or Secure Multipurpose Internet Mail Extensions, is a
standard for encrypting and signing MIME data in a message. This
was suggested and considered as a third security protocol for
authenticating the source of a message.</t>
<t>DMARC is focused on authentication at the domain level (i.e., the
Domain Owner taking responsibility for the message), while S/MIME is
really intended for user-to-user authentication and encryption. This
alone appears to make it a bad fit for DMARC's goals.</t>
<t>S/MIME also suffers from the heavyweight problem of Public Key
Infrastructure, which means that distribution of keys used to verify
signatures needs to be incorporated. In many instances, this alone
is a showstopper. There have been consistent promises that PKI
usability and deployment will improve, but these have yet to
materialize. DMARC can revisit this choice after those barriers are
addressed.</t>
<t>S/MIME has extensive deployment in specific market segments
(government, for example) but does not enjoy similar widespread
deployment over the general Internet, and this shows no signs of
changing. DKIM and SPF are both deployed widely over the general
Internet, and their adoption rates continue to be positive.</t>
<t>Finally, experiments have shown that including S/MIME support in the
initial version of DMARC would neither cause nor enable a substantial
increase in the accuracy of the overall mechanism.</t>
</section>

<section anchor="method-exclusion"><name>Method Exclusion</name>
<t>It was suggested that DMARC include a mechanism by which a Domain
Owner could tell Mail Receivers not to attempt verification by one
of the supported methods (e.g., &quot;check DKIM, but not SPF&quot;).</t>
<t>Specifically, consider a Domain Owner that has deployed one of the
technologies and that technology fails for some messages, but such
failures don't cause enforcement action. Deploying DMARC would cause
enforcement action for policies other than &quot;none&quot;, which would appear
to exclude participation by that Domain Owner.</t>
<t>The DMARC development team evaluated the idea of policy exception
mechanisms on several occasions and invariably concluded that there
was not a strong enough use case to include them. The target audience
for DMARC does not appear to have concerns about the failure modes of
one or the other being a barrier to DMARC's adoption.</t>
<t>In the scenario described above, the Domain Owner has a few options:</t>

<ol>
<li><t>Tighten up its infrastructure to minimize the failure modes of
the single deployed technology.</t>
</li>
<li><t>Deploy the other supported authentication mechanism, to offset
the failure modes of the first.</t>
</li>
<li><t>Deploy DMARC in a reporting-only mode.</t>
</li>
</ol>
</section>

<section anchor="sender-header-field"><name>Sender Header Field</name>
<t>It has been suggested in several message authentication efforts that
the Sender header field be checked for an identifier of interest, as
the standards indicate this as the proper way to indicate a
re-mailing of content such as through a mailing list. Most recently,
it was a protocol-level option for DomainKeys, but on evolution to
DKIM, this property was removed.</t>
<t>The DMARC development team considered this and decided not to include
support for doing so for the following reasons:</t>

<ol>
<li><t>The main user protection approach is to be concerned with what
the user sees when a message is rendered. There is no consistent
behavior among MUAs regarding what to do with the content of the
Sender field, if present. Accordingly, supporting the checking of
the Sender identifier would mean applying policy to an identifier
the end user might never actually see, which can create a vector
for attack against end users by simply forging a Sender field
containing some identifier that DMARC will like.</t>
</li>
<li><t>Although it is certainly true that this is what the Sender field
is for, its use in this way is also unreliable, making it a poor
candidate for inclusion in the DMARC evaluation algorithm.</t>
</li>
<li><t>Allowing multiple ways to discover policy introduces unacceptable
ambiguity into the DMARC evaluation algorithm in terms of which
policy is to be applied and when.</t>
</li>
</ol>
</section>

<section anchor="domain-existence-test"><name>Domain Existence Test</name>
<t>The presence of the &quot;np&quot; tag in this specification seemingly implies that
there would be an agreed-upon standard for determining a domain's existence.</t>
<t>Since the DMARC protocol is focused on email, one might think that the
definition of resolvable in <xref target="RFC5321"></xref> applies. Using that definition, only
names that resolve to MX Resource Records (RRs), A RRs, or AAAA RRs are deemed
to be resolvable and to exist in the DNS. This is also consistent with the process
documented in <xref target="RFC5617"></xref> (ADSP), and is a common practice among MTA operators to
determine whether or not to accept a mail message before performing other more
expensive processing.</t>
<t>The DMARC protocol makes no such requirement for the existence of specific DNS
RRs in order for a domain to exist; instead, if any RR exists for a domain, then
the domain exists. To use the terminology from <xref target="RFC2308"></xref>, an &quot;NXDOMAIN&quot; response
(rcode &quot;Name Error&quot;) to a DNS query means that the domain name does not exist,
while a &quot;NODATA&quot; response (rcode &quot;NOERROR&quot;) means that the given resource record
type queried for does not exist, but the domain name does.</t>
<t>Furthermore, in keeping with <xref target="RFC8020"></xref>, if a query for a name returns NXDOMAIN,
then not only does the name not exist, every name below it in the DNS hierarchy
also does not exist.</t>
</section>

<section anchor="issues-with-adsp-in-operation"><name>Issues with ADSP in Operation</name>
<t>DMARC has been characterized as a &quot;super-ADSP&quot; of sorts.</t>
<t>Contributors to DMARC have compiled a list of issues associated with
ADSP, ones gained from operational experience, that have influenced the
direction of DMARC:</t>

<ol>
<li><t>ADSP has no support for subdomains, i.e., the ADSP record for
example.com does not explicitly or implicitly apply to
subdomain.example.com. If wildcarding is not applied, then
spammers can trivially bypass ADSP by sending from a subdomain
with no ADSP record.</t>
</li>
<li><t>Nonexistent subdomains are explicitly out of scope in ADSP.
There is nothing in ADSP that states Mail Receivers should simply
reject mail from NXDOMAINs regardless of ADSP policy (which of
course allows spammers to trivially bypass ADSP by sending email
from nonexistent subdomains).</t>
</li>
<li><t>ADSP has no operational advice on when to look up the ADSP
record.</t>
</li>
<li><t>ADSP has no support for using SPF as an auxiliary mechanism to
DKIM.</t>
</li>
<li><t>ADSP has no support for a slow rollout, i.e., no way to configure
a percentage of email on which the Mail Receiver should apply the
policy.  This is important for large-volume senders.</t>
</li>
<li><t>ADSP has no explicit support for an intermediate phase where the
Mail Receiver quarantines (e.g., sends to the recipient's &quot;spam&quot;
folder) rather than rejects the email.</t>
</li>
<li><t>The binding between the &quot;From&quot; header domain and DKIM is too
tight for ADSP; they must match exactly.</t>
</li>
</ol>
</section>

<section anchor="organizational-domain-discovery-issues"><name>Organizational Domain Discovery Issues</name>
<t>An earlier informational version of the DMARC protocol <xref target="RFC7489"></xref>
noted that the DNS does not provide a method by which the &quot;domain of record&quot;,
or the domain that was actually registered with a domain registrar, can
be determined given an arbitrary domain name. That version further mentioned
suggestions that have been made that attempt to glean such information from
SOA or NS resource records, but these too are not fully reliable, as the
partitioning of the DNS is not always done at administrative boundaries.</t>
<t>That previous version posited that one could &quot;climb the tree&quot; to find the
Organizational Domain, but expressed concern that an attacker could exploit
this for a denial-of-service attack through sending a high number of messages
each with a relatively large number of nonsense labels, causing a Mail Receiver
to perform a large number of DNS queries in search of a policy record. This
version defines a method for performing a DNS Tree Walk, described in <xref target="dns-tree-walk"></xref>,
and further mitigates the risk of the denial-of-service attack by expressly limiting
the number of DNS queries to execute regardless of the number of labels in the domain
name.</t>
<t>As a matter of historical record, the method for finding the Organizational
Domain described in <xref target="RFC7489"></xref> is preserved here:</t>

<ol>
<li><t>Acquire a &quot;public suffix&quot; list (PSL), i.e., a list of DNS domain
  names reserved for registrations. Some country Top-Level Domains
  (TLDs) make specific registration requirements, e.g., the United
  Kingdom places company registrations under &quot;.co.uk&quot;; other TLDs
  such as &quot;.com&quot; appear in the IANA registry of top-level DNS
  domains. A PSL is the union of all of these.</t>
<t>A PSL can be obtained from various sources. The most common one
  is maintained by the Mozilla Foundation and made public at
  <eref target="http://publicsuffix.org">http://publicsuffix.org</eref>.  License terms governing the use of that
  list are available at that URI.</t>
<t>Note that if operators use a variety of public suffix lists,
  interoperability will be difficult or impossible to guarantee.</t>
</li>
<li><t>Break the subject DNS domain name into a set of &quot;n&quot; ordered
  labels. Number these labels from right to left; e.g., for
  &quot;example.com&quot;, &quot;com&quot; would be label 1 and &quot;example&quot; would be
  label 2.</t>
</li>
<li><t>Search the public suffix list for the name that matches the
  largest number of labels found in the subject DNS domain. Let
  that number be &quot;x&quot;.</t>
</li>
<li><t>Construct a new DNS domain name using the name that matched from
  the public suffix list and prefixing to it the &quot;x+1&quot;th label from
  the subject domain. This new name is the Organizational Domain.</t>
</li>
</ol>
<t>Thus, since &quot;com&quot; is an IANA-registered TLD, a subject domain of
   &quot;a.b.c.d.example.com&quot; would have an Organizational Domain of
   &quot;example.com&quot;.</t>
<t>The process of determining a suffix is currently a heuristic one. No
   list is guaranteed to be accurate or current.</t>
</section>

<section anchor="removal-of-the-pct-tag"><name>Removal of the &quot;pct&quot; Tag</name>
<t>An earlier informational version of the DMARC protocol <xref target="RFC7489"></xref>
included a &quot;pct&quot; tag and specified all integers from 0 to 100 inclusive
as valid values for the tag. The intent of the tag was to provide domain
owners with a method to gradually change their preferred assessment policy
(the p= tag) from 'none' to 'quarantine' or from 'quarantine' to 'reject'
by requesting the stricter treatment for just a percentage of messages
that produced DMARC results of &quot;fail&quot;.</t>
<t>Operational experience showed that the pct tag was usually not accurately
applied, unless the value specified was either &quot;0&quot; or &quot;100&quot; (the default),
and the inaccuracies with other values varied widely from implementation to
implementation. The default value was easily implemented, as it required no
special processing on the part of the Mail Receiver, while the value
of &quot;0&quot; took on unintended significance as a value used by some intermediaries
and mailbox providers as an indicator to deviate from standard handling of
the message, usually by rewriting the RFC5322.From header in an effort to
avoid DMARC failures downstream.</t>
<t>These custom actions when the pct= tag was set to &quot;0&quot; proved valuable to the
email community. In particular, header rewriting by an intermediary meant
that a Domain Owner's aggregate reports could reveal to the Domain Owner
how much of its traffic was routing through intermediaries that don't rewrite
the RFC5322.From header. It required work on the part of the Domain Owner to
compare aggregate reports from before and after the p= value was changed
and pct= was included in the DMARC policy record with a value of &quot;0&quot;, but
the data was there. Consequently, knowing how much mail was subject to
possible DMARC failure due to a lack of RFC5322.From header rewriting by
intermediaries could assist the Domain Owner in choosing whether or not
to proceed from an applied policy of p=none to p=quarantine or p=reject.
Armed with this knowledge, the Domain Owner could make an informed decision
regarding subjecting its mail traffic to possible DMARC failures based on
the Domain Owner's tolerance for such things.</t>
<t>Because of the value provided by &quot;pct=0&quot; to Domain Owners, it was logical
to keep this functionality in the protocol; at the same time, it didn't make
sense to support a tag named &quot;pct&quot; that had only two valid values. This version
of the DMARC protocol, therefore, introduces the &quot;t&quot; tag as shorthand for &quot;testing&quot;,
with the valid values of &quot;y&quot; and &quot;n&quot;, which are meant to be analogous in their
application by mailbox providers and intermediaries to the &quot;pct&quot; tag values
&quot;0&quot; and &quot;100&quot;, respectively.</t>
</section>
</section>

<section anchor="examples"><name>Examples</name>
<t>This section illustrates both the Domain Owner side and the Mail
Receiver side of a DMARC exchange.</t>

<section anchor="identifier-alignment-examples"><name>Identifier Alignment Examples</name>
<t>The following examples illustrate the DMARC mechanism's use of
Identifier Alignment. For brevity's sake, only message headers are
shown, as message bodies are not considered when conducting DMARC
checks.</t>

<section anchor="spf"><name>SPF</name>
<t>The following SPF examples assume that SPF produces a passing result.
Alignment cannot exist if SPF does not produce a passing result.</t>
<t>Example 1: SPF in alignment:</t>

<artwork>     MAIL FROM: &lt;sender@example.com&gt;

     From: sender@example.com
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: receiver@example.org
     Subject: here's a sample
</artwork>
<t>In this case, the RFC5321.MailFrom parameter and the RFC5322.From
header field have identical DNS domains. Thus, the identifiers are in
strict alignment.</t>
<t>Example 2: SPF in alignment (parent):</t>

<artwork>     MAIL FROM: &lt;sender@child.example.com&gt;

     From: sender@example.com
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: receiver@example.org
     Subject: here's a sample
</artwork>
<t>In this case, the RFC5322.From header parameter includes a DNS
domain that is a parent of the RFC5321.MailFrom domain. Thus, the
identifiers are in relaxed alignment because they both have the
same Organizational Domain (example.com).</t>
<t>Example 3: SPF not in alignment:</t>

<artwork>     MAIL FROM: &lt;sender@example.net&gt;

     From: sender@child.example.com
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: receiver@example.org
     Subject: here's a sample
</artwork>
<t>In this case, the RFC5321.MailFrom parameter includes a DNS domain
that is neither the same as, a parent of, nor a child of the
RFC5322.From domain. Thus, the identifiers are not in alignment.</t>
</section>

<section anchor="dkim"><name>DKIM</name>
<t>The examples below assume that the DKIM signatures pass verification.
Alignment cannot exist with a DKIM signature that does not verify.</t>
<t>Example 1: DKIM in alignment:</t>

<artwork>     DKIM-Signature: v=1; ...; d=example.com; ...
     From: sender@example.com
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: receiver@example.org
     Subject: here's a sample
</artwork>
<t>In this case, the DKIM &quot;d=&quot; parameter and the RFC5322.From header field have
identical DNS domains. Thus, the identifiers are in strict alignment.</t>
<t>Example 2: DKIM in alignment (parent):</t>

<artwork>     DKIM-Signature: v=1; ...; d=example.com; ...
     From: sender@child.example.com
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: receiver@example.org
     Subject: here's a sample
</artwork>
<t>In this case, the DKIM signature's &quot;d=&quot; parameter includes a DNS
domain that is a parent of the RFC5322.From domain. Thus, the
identifiers are in relaxed alignment, as they have the same
Organizational Domain (example.com).</t>
<t>Example 3: DKIM not in alignment:</t>

<artwork>     DKIM-Signature: v=1; ...; d=sample.net; ...
     From: sender@child.example.com
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: receiver@example.org
     Subject: here's a sample
</artwork>
<t>In this case, the DKIM signature's &quot;d=&quot; parameter includes a DNS
domain that is neither the same as, a parent of, nor a child of the
RFC5322.From domain. Thus, the identifiers are not in alignment.</t>
</section>
</section>

<section anchor="domain-owner-example"><name>Domain Owner Example</name>
<t>A Domain Owner that wants to use DMARC should have already deployed
and tested SPF and DKIM. The next step is to publish a DNS record
that advertises a DMARC policy for the Domain Owner's Organizational
Domain.</t>

<section anchor="entire-domain-monitoring-only"><name>Entire Domain, Monitoring Only</name>
<t>The owner of the domain &quot;example.com&quot; has deployed SPF and DKIM on
its messaging infrastructure. The owner wishes to begin using DMARC
with a policy that will solicit aggregate feedback from Mail Receivers
without affecting how the messages are processed in order to:</t>

<ul>
<li><t>Confirm that its legitimate messages are authenticating correctly</t>
</li>
<li><t>Verify that all authorized message sources have implemented
authentication measures</t>
</li>
<li><t>Determine how many messages from other sources would be affected
by a blocking policy</t>
</li>
</ul>
<t>The Domain Owner accomplishes this by constructing a policy record
indicating that:</t>

<ul>
<li><t>The version of DMARC being used is &quot;DMARC1&quot; (&quot;v=DMARC1;&quot;)</t>
</li>
<li><t>Mail Receivers should not alter how they treat these messages because
of this DMARC policy record (&quot;p=none&quot;)</t>
</li>
<li><t>Aggregate feedback reports are sent via email to the address
&quot;dmarc-feedback@example.com&quot;
(&quot;rua=<eref target="mailto:dmarc-feedback@example.com&quot;">mailto:dmarc-feedback@example.com"</eref>)</t>
</li>
<li><t>All messages from this Organizational Domain are subject to this
policy (no &quot;t&quot; tag present, so the default of &quot;n&quot; applies).</t>
</li>
</ul>
<t>The DMARC policy record might look like this when retrieved using a
common command-line tool:</t>

<artwork>  % dig +short TXT _dmarc.example.com.
  &quot;v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com&quot;
</artwork>
<t>To publish such a record, the DNS administrator for the Domain Owner
creates an entry like the following in the appropriate zone file
(following the conventional zone file format):</t>

<artwork>  ; DMARC record for the domain example.com

  _dmarc  IN TXT ( &quot;v=DMARC1; p=none; &quot;
                   &quot;rua=mailto:dmarc-feedback@example.com&quot; )
</artwork>
</section>

<section anchor="entire-domain-monitoring-only-per-message-reports"><name>Entire Domain, Monitoring Only, Per-Message Reports</name>
<t>The Domain Owner from the previous example has used the aggregate
reporting to discover some messaging systems that had not yet
implemented DKIM correctly, but they are still seeing periodic
authentication failures. To diagnose these intermittent
problems, they wish to request per-message failure reports when
authentication failures occur.</t>
<t>Not all Mail Receivers will honor such a request, but the Domain Owner
feels that any reports it does receive will be helpful enough to
justify publishing this record. The default per-message report
format (<xref target="RFC6591"></xref>) meets the Domain Owner's needs in this scenario.</t>
<t>The Domain Owner accomplishes this by adding the following to its
policy record from <xref target="entire-domain-monitoring-only"></xref>:</t>

<ul spacing="compact">
<li>Per-message failure reports are sent via email to the
address &quot;auth-reports@example.com&quot;
(&quot;ruf=<eref target="mailto:auth-reports@example.com&quot;">mailto:auth-reports@example.com"</eref>)</li>
</ul>
<t>The DMARC policy record might look like this when retrieved using a
common command-line tool (the output shown would appear on a single
line but is wrapped here for publication):</t>

<artwork>  % dig +short TXT _dmarc.example.com.
  &quot;v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com;
   ruf=mailto:auth-reports@example.com&quot;
</artwork>
<t>To publish such a record, the DNS administrator for the Domain Owner
might create an entry like the following in the appropriate zone file
(following the conventional zone file format):</t>

<artwork>  ; DMARC record for the domain example.com

  _dmarc  IN TXT ( &quot;v=DMARC1; p=none; &quot;
                    &quot;rua=mailto:dmarc-feedback@example.com; &quot;
                    &quot;ruf=mailto:auth-reports@example.com&quot; )
</artwork>
</section>

<section anchor="per-message-failure-reports-directed-to-third-party"><name>Per-Message Failure Reports Directed to Third Party</name>
<t>The Domain Owner from the previous example is maintaining the same
policy but now wishes to have a third party serve as a Report Consumer.
Again, not all Mail Receivers will honor this request, but those that
do may implement additional checks to verify that the third party wishes
to receive the failure reports for this domain.</t>
<t>The Domain Owner needs to alter its policy record from <xref target="entire-domain-monitoring-only-per-message-reports"></xref>
as follows:</t>

<ul spacing="compact">
<li>Per-message failure reports are sent via email to the
address &quot;auth-reports@thirdparty.example.net&quot;
(&quot;ruf=<eref target="mailto:auth-reports@thirdparty.example.net&quot;">mailto:auth-reports@thirdparty.example.net"</eref>)</li>
</ul>
<t>The DMARC policy record might look like this when retrieved using a
common command-line tool (the output shown would appear on a single
line but is wrapped here for publication):</t>

<artwork>  % dig +short TXT _dmarc.example.com.
  &quot;v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com;
   ruf=mailto:auth-reports@thirdparty.example.net&quot;
</artwork>
<t>To publish such a record, the DNS administrator for the Domain Owner
might create an entry like the following in the appropriate zone file
(following the conventional zone file format):</t>

<artwork>  ; DMARC record for the domain example.com

  _dmarc IN TXT ( &quot;v=DMARC1; p=none; &quot;
                  &quot;rua=mailto:dmarc-feedback@example.com; &quot;
                  &quot;ruf=mailto:auth-reports@thirdparty.example.net&quot; )
</artwork>
<t>Because the address used in the &quot;ruf&quot; tag is outside the
Organizational Domain in which this record is published, conforming
Mail Receivers will implement additional checks as described in
<xref target="I-D.ietf-dmarc-aggregate-reporting" sectionFormat="of" relative="#" section="3"></xref>. To pass these additional
checks, the Report Consumer's Domain Owner will need to publish an additional
DNS record as follows:</t>

<ul spacing="compact">
<li>Given the DMARC record published by the Domain Owner at
&quot;_dmarc.example.com&quot;, the DNS administrator for the Report Consumer
will need to publish a TXT resource record at
&quot;example.com._report._dmarc.thirdparty.example.net&quot; with the value
&quot;v=DMARC1;&quot;.</li>
</ul>
<t>The resulting DNS record might look like this when retrieved using a
common command-line tool (the output shown would appear on a single
line but is wrapped here for publication):</t>

<artwork>  % dig +short TXT example.com._report._dmarc.thirdparty.example.net
  &quot;v=DMARC1;&quot;
</artwork>
<t>To publish such a record, the DNS administrator for example.net might
create an entry like the following in the appropriate zone file
(following the conventional zone file format):</t>

<artwork>  ; zone file for thirdparty.example.net
  ; Accept DMARC failure reports on behalf of example.com

  example.com._report._dmarc   IN   TXT    &quot;v=DMARC1;&quot;
</artwork>
<t>Mediators and other third parties should refer to <xref target="I-D.ietf-dmarc-aggregate-reporting" sectionFormat="of" relative="#" section="3"></xref>
for the full details of this mechanism.</t>
</section>

<section anchor="subdomain-sampling-and-multiple-aggregate-report-uris"><name>Subdomain, Testing, and Multiple Aggregate Report URIs</name>
<t>The Domain Owner has implemented SPF and DKIM in a subdomain used for
pre-production testing of messaging services.  It now wishes to express
a handling preference for messages from this subdomain that fail to
authenticate to indicate to participating Mail Receivers that use of this
domain is not valid.</t>
<t>As a first step, it will express that it considers messages using this
subdomain that fail authentication to be suspicious. The goal here
will be to enable examination of messages sent to mailboxes hosted by
participating Mail Receivers as a method for troubleshooting any existing
authentication issues. Aggregate feedback reports will be sent to
a mailbox within the Organizational Domain, and to a mailbox at a Report
Consumer selected and authorized to receive them by the Domain Owner.</t>
<t>The Domain Owner will accomplish this by constructing a policy record
indicating that:</t>

<ul>
<li><t>The version of DMARC being used is &quot;DMARC1&quot; (&quot;v=DMARC1;&quot;)</t>
</li>
<li><t>It is applied only to this subdomain (the record is published at
&quot;_dmarc.test.example.com&quot; and not &quot;_dmarc.example.com&quot;)</t>
</li>
<li><t>Mail Receivers are advised that the Domain Owner considers messages
that fail to authenticate to be suspicious (&quot;p=quarantine&quot;)</t>
</li>
<li><t>Aggregate feedback reports are sent via email to the
addresses &quot;dmarc-feedback@example.com&quot; and
&quot;example-tld-test@thirdparty.example.net&quot;
(&quot;rua=<eref target="mailto:dmarc-feedback@example.com">mailto:dmarc-feedback@example.com</eref>,
 <eref target="mailto:tld-test@thirdparty.example.net&quot;)">mailto:tld-test@thirdparty.example.net")</eref></t>
</li>
<li><t>The Domain Owner desires only that an actor performing a DMARC
verification check apply any special handling rules it might have
in place, such as rewriting the RFC53322.From header; the Domain
Owner is testing its setup at this point and so does not want
the handling policy to be applied. (&quot;t=y&quot;)</t>
</li>
</ul>
<t>The DMARC policy record might look like this when retrieved using a
common command-line tool (the output shown would appear on a single
line but is wrapped here for publication):</t>

<artwork>  % dig +short TXT _dmarc.test.example.com
  &quot;v=DMARC1; p=quarantine; rua=mailto:dmarc-feedback@example.com,
   mailto:tld-test@thirdparty.example.net; t=y&quot;
</artwork>
<t>To publish such a record, the DNS administrator for the Domain Owner
might create an entry like the following in the appropriate zone
file:</t>

<artwork>  ; DMARC record for the domain test.example.com

  _dmarc IN  TXT  ( &quot;v=DMARC1; p=quarantine; &quot;
                    &quot;rua=mailto:dmarc-feedback@example.com,&quot;
                    &quot;mailto:tld-test@thirdparty.example.net;&quot;
                    &quot;t=y&quot; )
</artwork>
<t>Once enough time has passed to allow for collecting enough reports to
give the Domain Owner confidence that all legitimate email sent using
the subdomain is properly authenticating and passing DMARC checks, then
the Domain Owner can update the policy record to indicate that it considers
authentication failures to be a clear indication that use of the subdomain
is not valid. It would do this by altering the DNS record to advise
Mail Receivers of its position on such messages (&quot;p=reject&quot;) and removing the
testing flag (&quot;t=y&quot;).</t>
<t>After alteration, the DMARC policy record might look like this when retrieved
using a common command-line tool (the output shown would appear on a single
line but is wrapped here for publication):</t>

<artwork>  % dig +short TXT _dmarc.test.example.com
  &quot;v=DMARC1; p=reject; rua=mailto:dmarc-feedback@example.com,
   mailto:tld-test@thirdparty.example.net&quot;
</artwork>
<t>To publish such a record, the DNS administrator for the Domain Owner
might create an entry like the following in the appropriate zone
file:</t>

<artwork>  ; DMARC record for the domain test.example.com

  _dmarc IN  TXT  ( &quot;v=DMARC1; p=reject; &quot;
                    &quot;rua=mailto:dmarc-feedback@example.com,&quot;
                    &quot;mailto:tld-test@thirdparty.example.net&quot; )
</artwork>
</section>
</section>

<section anchor="mail-receiver-example"><name>Mail Receiver Example</name>
<t>A Mail Receiver that wants to use DMARC should already be checking
SPF and DKIM, and possess the ability to collect relevant information
from various email-processing stages to provide feedback to Domain
Owners (possibly via Report Consumers).</t>

<section anchor="smtp-session-example"><name>SMTP Session Example</name>
<t>An optimal DMARC-enabled Mail Receiver performs authentication and
Identifier Alignment checking during the SMTP <xref target="RFC5321"></xref> conversation.</t>
<t>Before returning a final reply to the DATA command, the Mail
Receiver's MTA has performed:</t>

<ol>
<li><t>An SPF check to determine an SPF-authenticated Identifier.</t>
</li>
<li><t>DKIM checks that yield one or more DKIM-authenticated
Identifiers.</t>
</li>
<li><t>A DMARC policy lookup.</t>
</li>
</ol>
<t>The presence of an Author Domain DMARC record indicates that the Mail
Receiver should continue with DMARC-specific processing before
returning a reply to the DATA command.</t>
<t>Given a DMARC record and the set of Authenticated Identifiers, the
Mail Receiver checks to see if the Authenticated Identifiers align
with the Author Domain (taking into consideration any strict versus
relaxed options found in the DMARC record).</t>
<t>For example, the following sample data is considered to be from a
piece of email originating from the Domain Owner of &quot;example.com&quot;:</t>

<artwork>  Author Domain: example.com
  SPF-authenticated Identifier: mail.example.com
  DKIM-authenticated Identifier: example.com
  DMARC record:
    &quot;v=DMARC1; p=reject; aspf=r;
     rua=mailto:dmarc-feedback@example.com&quot;
</artwork>
<t>In the above sample, the SPF-authenticated Identifier and the
DKIM-authenticated Identifier both align with the Author Domain. The
Mail Receiver considers the above email to pass the DMARC check, avoiding
the &quot;reject&quot; policy that is requested to be applied to email that fails
to pass the DMARC check.</t>
<t>If no Authenticated Identifiers align with the Author Domain, then
the Mail Receiver applies the DMARC-record-specified policy. However,
before this action is taken, the Mail Receiver can consult external
information to override the Domain Owner's Assessment Policy. For
example, if the Mail Receiver knows that this particular email came
from a known and trusted forwarder (that happens to break both SPF
and DKIM), then the Mail Receiver may choose to ignore the Domain
Owner's policy.</t>
<t>The Mail Receiver is now ready to reply to the DATA command. If the
DMARC check yields that the message is to be rejected, then the Mail
Receiver replies with a 5xy code to inform the sender of failure. If
the DMARC check cannot be resolved due to transient network errors,
then the Mail Receiver replies with a 4xy code to inform the sender
as to the need to reattempt delivery later. If the DMARC check
yields a passing message, then the Mail Receiver continues with
email processing, perhaps using the result of the DMARC check as an
input to additional processing modules such as a domain reputation
query.</t>
<t>Before exiting DMARC-specific processing, the Mail Receiver checks to
see if the Author Domain DMARC record requests AFRF-based reporting.
If so, then the Mail Receiver can emit an AFRF to the reporting
address supplied in the DMARC record.</t>
<t>At the exit of DMARC-specific processing, the Mail Receiver captures
(through logging or direct insertion into a data store) the result of
DMARC processing. Captured information is used to build feedback for
Domain Owner consumption. This is unnecessary if the Domain Owner
has not requested aggregate reports, i.e., no &quot;rua&quot; tag was found in
the policy record.</t>
</section>
</section>

<section anchor="treewalk-example"><name>Organizational and Policy Domain Tree Walk Examples</name>
<t>If an RFC5322.From domain has no DMARC record, a receiver uses a tree walk
to find the policy domain.</t>
<t>If the policy in a policy domain allows relaxed alignment and the SPF or
DKIM domains are different from the RFC5322.From domain, a receiver uses
a tree walk to discover the respective Organizational domains.</t>

<section anchor="simple-organizational-and-policy-example"><name>Simple Organizational and Policy Example</name>
<t>A mail receiver receives an email with:</t>

<dl spacing="compact">
<dt>RFC5322.From domain</dt>
<dd>example.com</dd>
<dt>RFC5321.MailFrom domain</dt>
<dd>example.com</dd>
<dt>DKIM signature d=</dt>
<dd>signing.example.com</dd>
</dl>
<t>In this example, _dmarc.example.com and _dmarc.signing.example.com both
have DMARC records while _dmarc.com does not. If SPF or DKIM yield pass
results, they still have to be aligned to support a DMARC pass. Since
not all domains are the same, if the alignment is relaxed then the tree
walk is performed to determine the organizational domain for each:</t>
<t>For the RFC5322.From domain, query _dmarc.example.com and _dmarc.com;
example.com is the last element of the DNS tree with a DMARC record, so
it is the organizational domain for example.com.</t>
<t>For the RFC5321.MailFrom domain, the Organizational domain already found for
example.com is example.com, so SPF is aligned.</t>
<t>For the DKIM d= domain, query _dmarc.signing.example.com, _dmarc.example.com,
and _dmarc.com. Both signing.example.com and example.com have DMARC records,
but example.com is the highest element in the tree with a DMARC record, so
example.com is the organizational domain. Since this is also the
organizational domain for the RFC5322.From domain, DKIM is aligned for relaxed
alignment.</t>
<t>Since both SPF and DKIM are aligned, they can be used to determine if the
message has a DMARC pass result. If the result is not pass, then the policy
domain's DMARC record is used to determine the appropriate policy. In this
case, since the RFC5322.From domain has a DMARC record, that is the policy
domain.</t>
</section>

<section anchor="deep-tree-walk-example"><name>Deep Tree Walk Example</name>
<t>A mail receiver receives an email with:</t>

<dl spacing="compact">
<dt>RFC5322.From domain</dt>
<dd>a.b.c.d.e.f.g.h.i.j.k.example.com</dd>
<dt>RFC5321.MailFrom domain</dt>
<dd>example.com</dd>
<dt>DKIM signature d=</dt>
<dd>signing.example.com</dd>
</dl>
<t>Both _dmarc.example.com and _dmarc.signing.example.com have DMARC records,
while _dmarc.com does not. If SPF or DKIM yield pass results, they still have
to be aligned to support a DMARC pass. Since not all domains are the same, if
the alignment is relaxed then the tree walk is performed to determine the organizational
domain for each:</t>
<t>For the RFC5322.From domain, query _dmarc.a.b.c.d.e.f.g.h.i.j.k.example.com,
skip to _dmarc.j.k.example.com, then query _dmarc.k.example.com,
 _dmarc.example.com, and _dmarc.com. None of
a.b.c.d.e.f.g.h.i.j.k.example.com, j.k.example.com, or k.example.com have
a DMARC record.</t>
<t>Since example.com is the last element of the DNS tree with a DMARC record,
it is the organizational domain for a.b.c.d.e.f.g.h.i.j.k.example.com.</t>
<t>For the RFC5321.MailFrom domain, the Organizational domain already found
for example.com is example.com. SPF is aligned.</t>
<t>For the DKIM d= domain, query _dmarc.signing.example.com, _dmarc.example.com,
and _dmarc.com. Both signing.example.com and example.com have DMARC records,
but example.com is the highest element in the tree with a DMARC record, so
example.com is the organizational domain. Since this is also the
organizational domain for the RFC5322.From domain, DKIM is aligned for relaxed
alignment.</t>
<t>Since both SPF and DKIM are aligned, they can be used to determine if the
message has a DMARC pass result. If the results for both are not pass, then
the policy domain's DMARC record is used to determine the appropriate policy.
In this case, the RFC5322.From domain does not have a DMARC record, so the
policy domain is the highest element in the DNS tree with a DMARC record,
example.com.</t>
</section>

<section anchor="example-with-a-psd-dmarc-record"><name>Example with a PSD DMARC Record</name>
<t>In rare cases, a PSD publishes a DMARC record with a psd tag, which the tree
walk must take into account.</t>
<t>A mail receiver receives an email with:</t>

<dl spacing="compact">
<dt>RFC5322.From domain</dt>
<dd>giant.bank.example</dd>
<dt>RFC5321.MailFrom domain</dt>
<dd>mail.giant.bank.example</dd>
<dt>DKIM signature d=</dt>
<dd>mail.mega.bank.example</dd>
</dl>
<t>In this case, _dmarc.bank.example has a DMARC record which includes the psd=y tag,
and _dmarc.example does not have a DMARC record.
While _dmarc.giant.bank.example has a DMARC record without a psd tag,
_dmarc.mega.bank.example and _mail.mega.bank.example have no DMARC records.</t>
<t>Since the three domains are all different, tree walks find their organization domains
to see which are aligned.</t>
<t>For the RFC5322.From domain giant.bank.example, the tree walk finds the record at _dmarc.giant.bank.example,
then the record at _dmarc.bank.example, and stops because of the psd=y flag.
The organizational domain is giant.bank.example because it is the domain below the one with psd=y.
Since the organizational domain has a DMARC record, it is also the policy domain.</t>
<t>For the RFC5321.MailFrom domain, the tree walk finds no record at _dmarc.mail.giant.bank.example,
the DMARC record at _dmarc.giant.bank.example,
then the record at _dmarc.bank.example, and stops because of the psd=y flag.
Again the organizational domain is giant.bank.example because it is the domain below the one with psd=y.
Since this is the same organizational domain as the RFC5322.From domain, SPF is aligned.</t>
<t>For the DKIM signature domain mail.mega.bank.example, the tree walk finds no records at
_dmarc.mail.mega.bank.example or _dmarc.mega.bank.example,
then finds the record at _dmarc.bank.example and stops because of the psd=y flag.
The organizational domain is mega.bank.example, so DKIM is not aligned.</t>
<t>Since SPF is aligned, it can be used to determine if the
message has a DMARC pass result.  If the result is not pass, then the policy
domain's DMARC record is used to determine the appropriate policy.</t>
</section>
</section>

<section anchor="utilization-of-aggregate-feedback-example"><name>Utilization of Aggregate Feedback: Example</name>
<t>Aggregate feedback is consumed by Domain Owners to verify their
understanding of how a given domain is being processed by the Mail
Receiver. Aggregate reporting data on emails that pass all
DMARC-supporting authentication checks is used by Domain Owners to
verify that their authentication practices remain accurate. For
example, if a third party is sending on behalf of a Domain Owner,
the Domain Owner can use aggregate report data to verify ongoing
authentication practices of the third party.</t>
<t>Data on email that only partially passes underlying authentication
checks provides visibility into problems that need to be addressed by
the Domain Owner. For example, if either SPF or DKIM fails to pass,
the Domain Owner is provided with enough information to either
directly correct the problem or understand where authentication-
breaking changes are being introduced in the email transmission path.
If authentication-breaking changes due to email transmission path
cannot be directly corrected, then the Domain Owner at least
maintains an understanding of the effect of DMARC-based policies upon
the Domain Owner's email.</t>
<t>Data on email that fails all underlying authentication checks
provides baseline visibility on how the Domain Owner's domain is
being received at the Mail Receiver. Based on this visibility, the
Domain Owner can begin deployment of authentication technologies
across uncovered email sources, if the mail that is failing the checks
was generated by or on behalf of the Domain Owner. Data regarding
failing authentication checks can also allow the Domain Owner to
come to an understanding of how its domain is being misused.</t>
</section>
</section>

<section anchor="acknowledgements" numbered="false"><name>Acknowledgements</name>
<t>This reworking of the DMARC protocol specified in <xref target="RFC7489"></xref> is the
result of contributions from many participants in the IETF Working Group
dedicated to this effort. Although the contributors are too numerous to
mention, significant contributions were made by Kurt Andersen, Laura Atkins,
Seth Blank, Alex Brotman, Dave Crocker, Douglas E. Foster, Ned Freed,
Mike Hammer, Steven M. Jones, Scott Kitterman, Murray S. Kucherawy,
Barry Leiba, Alessandro Vesely, and Tim Wicinski.</t>
<t>The authors and contributors also recognize that this document would not
have been possible without the work done by those who had a hand in producing
<xref target="RFC7489"></xref>. The Acknowledgements section from that document is preserved
in full below.</t>
</section>

<section anchor="acknowledgements-rfc7489" numbered="false"><name>Acknowledgements - RFC 7489</name>
<t>DMARC and the draft version of this document submitted to the
Independent Submission Editor were the result of lengthy efforts by
an informal industry consortium: DMARC.org (see <eref target="http://dmarc.org">http://dmarc.org</eref>).
Participating companies included Agari, American Greetings, AOL, Bank
of America, Cloudmark, Comcast, Facebook, Fidelity Investments,
Google, JPMorgan Chase &amp; Company, LinkedIn, Microsoft, Netease,
PayPal, ReturnPath, The Trusted Domain Project, and Yahoo!.  Although
the contributors and supporters are too numerous to mention, notable
individual contributions were made by J. Trent Adams, Michael Adkins,
Monica Chew, Dave Crocker, Tim Draegen, Steve Jones, Franck Martin,
Brett McDowell, and Paul Midgen. The contributors would also like to
recognize the invaluable input and guidance that was provided early
on by J.D. Falk.</t>
<t>Additional contributions within the IETF context were made by Kurt
Anderson, Michael Jack Assels, Les Barstow, Anne Bennett, Jim Fenton,
J. Gomez, Mike Jones, Scott Kitterman, Eliot Lear, John Levine,
S. Moonesamy, Rolf Sonneveld, Henry Timmes, and Stephen J. Turnbull.</t>
</section>

</back>

</rfc>