en-US/about_DMARC.help.txt

TOPIC
   about_DMARC

Independent Submission M. Kucherawy, Ed.
Request for Comments: 7489
Category: Informational E. Zwicky, Ed.
ISSN: 2070-1721 Yahoo!
                                                              March 2015


Domain-based Message Authentication, Reporting, and Conformance (DMARC)

Abstract

   Domain-based Message Authentication, Reporting, and Conformance
   (DMARC) is a scalable mechanism by which a mail-originating
   organization can express domain-level policies and preferences for
   message validation, disposition, and reporting, that a mail-receiving
   organization can use to improve mail handling.

   Originators of Internet Mail need to be able to associate reliable
   and authenticated domain identifiers with messages, communicate
   policies about messages that use those identifiers, and report about
   mail using those identifiers. These abilities have several benefits:
   Receivers can provide feedback to Domain Owners about the use of
   their domains; this feedback can provide valuable insight about the
   management of internal operations and the presence of external domain
   name abuse.

   DMARC does not produce or encourage elevated delivery privilege of
   authenticated email. DMARC is a mechanism for policy distribution
   that enables increasingly strict handling of messages that fail
   authentication checks, ranging from no action, through altered
   delivery, up to message rejection.

Status of This Memo

   This document is not an Internet Standards Track specification; it is
   published for informational purposes.

   This is a contribution to the RFC Series, independently of any other
   RFC stream. The RFC Editor has chosen to publish this document at
   its discretion and makes no statement about its value for
   implementation or deployment. Documents approved for publication by
   the RFC Editor are not a candidate for any level of Internet
   Standard; see Section 2 of RFC 5741.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   http://www.rfc-editor.org/info/rfc7489.




Kucherawy & Zwicky Informational [Page 1]
 
RFC 7489 DMARC March 2015


Copyright Notice

   Copyright (c) 2015 IETF Trust and the persons identified as the
   document authors. All rights reserved.

   This document is subject to BCP 78 and the IETF Trust's Legal
   Provisions Relating to IETF Documents
   (http://trustee.ietf.org/license-info) in effect on the date of
   publication of this document. Please review these documents
   carefully, as they describe your rights and restrictions with respect
   to this document.

Table of Contents

   1. Introduction ....................................................3
   2. Requirements ....................................................5
      2.1. High-Level Goals ...........................................5
      2.2. Out of Scope ...............................................6
      2.3. Scalability ................................................6
      2.4. Anti-Phishing ..............................................7
   3. Terminology and Definitions .....................................7
      3.1. Identifier Alignment .......................................8
      3.2. Organizational Domain .....................................11
   4. Overview .......................................................12
      4.1. Authentication Mechanisms .................................12
      4.2. Key Concepts ..............................................12
      4.3. Flow Diagram ..............................................13
   5. Use of RFC5322.From ............................................15
   6. Policy .........................................................15
      6.1. DMARC Policy Record .......................................16
      6.2. DMARC URIs ................................................16
      6.3. General Record Format .....................................17
      6.4. Formal Definition .........................................21
      6.5. Domain Owner Actions ......................................22
      6.6. Mail Receiver Actions .....................................23
      6.7. Policy Enforcement Considerations .........................27
   7. DMARC Feedback .................................................28
      7.1. Verifying External Destinations ...........................28
      7.2. Aggregate Reports .........................................30
      7.3. Failure Reports ...........................................36
   8. Minimum Implementations ........................................37
   9. Privacy Considerations .........................................38
      9.1. Data Exposure Considerations ..............................38
      9.2. Report Recipients .........................................39







Kucherawy & Zwicky Informational [Page 2]
 
RFC 7489 DMARC March 2015


   10. Other Topics ..................................................39
      10.1. Issues Specific to SPF ...................................39
      10.2. DNS Load and Caching .....................................40
      10.3. Rejecting Messages .......................................40
      10.4. Identifier Alignment Considerations ......................41
      10.5. Interoperability Issues ..................................41
   11. IANA Considerations ...........................................42
      11.1. Authentication-Results Method Registry Update ............42
      11.2. Authentication-Results Result Registry Update ............42
      11.3. Feedback Report Header Fields Registry Update ............44
      11.4. DMARC Tag Registry .......................................44
      11.5. DMARC Report Format Registry .............................45
   12. Security Considerations .......................................46
      12.1. Authentication Methods ...................................46
      12.2. Attacks on Reporting URIs ................................46
      12.3. DNS Security .............................................47
      12.4. Display Name Attacks .....................................47
      12.5. External Reporting Addresses .............................48
      12.6. Secure Protocols .........................................48
   13. References ....................................................49
      13.1. Normative References .....................................49
      13.2. Informative References ...................................50
   Appendix A. Technology Considerations .............................52
     A.1. S/MIME .....................................................52
     A.2. Method Exclusion ...........................................53
     A.3. Sender Header Field ........................................53
     A.4. Domain Existence Test ......................................54
     A.5. Issues with ADSP in Operation ..............................54
     A.6. Organizational Domain Discovery Issues .....................55
   Appendix B. Examples ..............................................56
     B.1. Identifier Alignment Examples ..............................56
     B.2. Domain Owner Example .......................................58
     B.3. Mail Receiver Example .....................................63
     B.4. Utilization of Aggregate Feedback: Example .................64
     B.5. mailto Transport Example ...................................65
   Appendix C. DMARC XML Schema ......................................66
   Acknowledgements ..................................................73
   Authors' Addresses ................................................73

1. Introduction

   The Sender Policy Framework ([SPF]) and DomainKeys Identified Mail
   ([DKIM]) provide domain-level authentication. They enable
   cooperating email receivers to detect mail authorized to use the
   domain name, which can permit differential handling. (A detailed
   discussion of the threats these systems attempt to address can be
   found in [DKIM-THREATS].) However, there has been no single widely
   accepted or publicly available mechanism to communication of



Kucherawy & Zwicky Informational [Page 3]
 
RFC 7489 DMARC March 2015


   domain-specific message-handling policies for receivers, or to
   request reporting of authentication and disposition of received mail.
   Absent the ability to obtain feedback reports, originators who have
   implemented email authentication have difficulty determining how
   effective their authentication is. As a consequence, use of
   authentication failures to filter mail typically does not succeed.

   Over time, one-on-one relationships were established between select
   senders and receivers with privately communicated means to assert
   policy and receive message traffic and authentication disposition
   reporting. Although these ad hoc practices have been generally
   successful, they require significant manual coordination between
   parties, and this model does not scale for general use on the
   Internet.

   This document defines Domain-based Message Authentication, Reporting,
   and Conformance (DMARC), a mechanism by which email operators
   leverage existing authentication and policy advertisement
   technologies to enable both message-stream feedback and enforcement
   of policies against unauthenticated email.

   DMARC allows Domain Owners and receivers to collaborate by:

   1. Providing receivers with assertions about Domain Owners' policies

   2. Providing feedback to senders so they can monitor authentication
       and judge threats

   The basic outline of DMARC is as follows:

   1. Domain Owners publish policy assertions about domains via the
       DNS.

   2. Receivers compare the RFC5322.From address in the mail to the SPF
       and DKIM results, if present, and the DMARC policy in DNS.

   3. These receivers can use these results to determine how the mail
       should be handled.

   4. The receiver sends reports to the Domain Owner or its designee
       about mail claiming to be from their domain.

   Security terms used in this document are defined in [SEC-TERMS].








Kucherawy & Zwicky Informational [Page 4]
 
RFC 7489 DMARC March 2015


   DMARC differs from previous approaches to policy advertisement (e.g.,
   [SPF] and [ADSP]) in that:

   o Authentication technologies are:

      1. decoupled from any technology-specific policy mechanisms, and

      2. used solely to establish reliable per-message domain-level
          identifiers.

   o Multiple authentication technologies are used to:

      1. reduce the impact of transient authentication errors

      2. reduce the impact of site-specific configuration errors and
          deployment gaps

      3. enable more use cases than any individual technology supports
          alone

   o Receiver-generated feedback is supported, allowing senders to
      establish confidence in authentication practices.

   o The domain name extracted from a message's RFC5322.From field is
      the primary identifier in the DMARC mechanism. This identifier is
      used in conjunction with the results of the underlying
      authentication technologies to evaluate results under DMARC.

   Experience with DMARC has revealed some issues of interoperability
   with email in general that require due consideration before
   deployment, particularly with configurations that can cause mail to
   be rejected. These are discussed in Section 10.

2. Requirements

   Specification of DMARC is guided by the following high-level goals,
   security dependencies, detailed requirements, and items that are
   documented as out of scope.

2.1. High-Level Goals

   DMARC has the following high-level goals:

   o Allow Domain Owners to assert the preferred handling of
      authentication failures, for messages purporting to have
      authorship within the domain.

   o Allow Domain Owners to verify their authentication deployment.



Kucherawy & Zwicky Informational [Page 5]
 
RFC 7489 DMARC March 2015


   o Minimize implementation complexity for both senders and receivers,
      as well as the impact on handling and delivery of legitimate
      messages.

   o Reduce the amount of successfully delivered spoofed email.

   o Work at Internet scale.

2.2. Out of Scope

   Several topics and issues are specifically out of scope for the
   initial version of this work. These include the following:

   o different treatment of messages that are not authenticated versus
      those that fail authentication;

   o evaluation of anything other than RFC5322.From;

   o multiple reporting formats;

   o publishing policy other than via the DNS;

   o reporting or otherwise evaluating other than the last-hop IP
      address;

   o attacks in the RFC5322.From field, also known as "display name"
      attacks;

   o authentication of entities other than domains, since DMARC is
      built upon SPF and DKIM, which authenticate domains; and

   o content analysis.

2.3. Scalability

   Scalability is a major 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.

   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.






Kucherawy & Zwicky Informational [Page 6]
 
RFC 7489 DMARC March 2015


2.4. Anti-Phishing

   DMARC is designed to prevent bad actors from sending mail that claims
   to come from legitimate senders, particularly senders of
   transactional email (official mail that is 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.

   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.

   DMARC does not attempt to solve all problems with spoofed or
   otherwise fraudulent email. In particular, it does not address the
   use of visually similar domain names ("cousin domains") or abuse of
   the RFC5322.From human-readable <display-name>.

3. Terminology and Definitions

   This section defines terms used in the rest of the document.

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in [KEYWORDS].

   Readers are encouraged to be familiar with the contents of
   [EMAIL-ARCH]. 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.

   The following terms are also used:

   Authenticated Identifiers: Domain-level identifiers that are
      validated using authentication technologies are referred to as
      "Authenticated Identifiers". See Section 4.1 for details about
      the supported mechanisms.

   Author Domain: The domain name of the apparent author, as extracted
      from the RFC5322.From field.




Kucherawy & Zwicky Informational [Page 7]
 
RFC 7489 DMARC March 2015


   Domain Owner: An entity or organization that owns a DNS domain. The
      term "owns" here indicates that the entity or organization being
      referenced holds the registration of that DNS domain. 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 [EMAIL-ARCH]. It can also refer
      to delegates, such as Report Receivers, when those are outside of
      their immediate management domain.

   Identifier Alignment: When the domain in the RFC5322.From address
      matches a domain validated by SPF or DKIM (or both), it has
      Identifier Alignment.

   Mail Receiver: The entity or organization that receives and
      processes email. Mail Receivers operate one or more Internet-
      facing Mail Transport Agents (MTAs).

   Organizational Domain: The domain that was registered with a domain
      name registrar. In the absence of more accurate methods,
      heuristics are used to determine this, since it is not always the
      case that the registered domain name is simply a top-level DNS
      domain plus one component (e.g., "example.com", where "com" is a
      top-level domain). The Organizational Domain is determined by
      applying the algorithm found in Section 3.2.

   Report Receiver: An operator that receives reports from another
      operator implementing the reporting mechanism described in this
      document. Such an operator might be receiving reports about its
      own messages, or reports about messages related to another
      operator. This term applies collectively to the system components
      that receive and process these reports and the organizations that
      operate them.

3.1. Identifier Alignment

   Email authentication technologies authenticate various (and
   disparate) aspects of an individual message. For example, [DKIM]
   authenticates the domain that affixed a signature to the message,
   while [SPF] can authenticate either the domain that appears in the
   RFC5321.MailFrom (MAIL FROM) portion of [SMTP] or the RFC5321.EHLO/
   HELO domain, or both. These may be different domains, and they are
   typically not visible to the end user.

   DMARC authenticates use of the RFC5322.From domain by requiring that
   it match (be aligned with) an Authenticated Identifier. The
   RFC5322.From domain was selected as the central identity of the DMARC



Kucherawy & Zwicky Informational [Page 8]
 
RFC 7489 DMARC March 2015


   mechanism because it is a required message header field and therefore
   guaranteed to be present in compliant messages, and most Mail User
   Agents (MUAs) represent the RFC5322.From field as the originator of
   the message and render some or all of this header field's content to
   end users.

   Thus, this field is the one used by end users to identify the source
   of the message and therefore is a prime target for abuse. Many
   high-profile email sources, such as email service providers, require
   that the sending agent have authenticated 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.

   Domain names in this context are to be compared in a case-insensitive
   manner, per [DNS-CASE].

   It is important to note that Identifier Alignment cannot occur with a
   message that is not valid per [MAIL], particularly one with a
   malformed, absent, or repeated RFC5322.From 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, and handling of such
   non-compliant cases is outside of the scope of this specification.
   Further discussion of this can be found in Section 6.6.1.

   Each of the underlying authentication technologies that DMARC takes
   as input yields authenticated domains as their outputs when they
   succeed. From the perspective of DMARC, each can be operated in a
   "strict" mode or a "relaxed" mode. A Domain Owner would normally
   select strict mode if it wanted Mail Receivers to apply DMARC
   processing only to messages bearing an RFC5322.From domain exactly
   matching the domains those mechanisms will verify. Relaxed mode can
   be used when the operator also wishes to affect message flows bearing
   subdomains of the verified domains.

3.1.1. DKIM-Authenticated Identifiers

   DMARC permits Identifier Alignment, based on the result of a DKIM
   authentication, to be strict or relaxed. (Note that these are not
   related to DKIM's "simple" and "relaxed" canonicalization modes.)








Kucherawy & Zwicky Informational [Page 9]
 
RFC 7489 DMARC March 2015


   In relaxed mode, 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 must be equal if
   the identifiers are to be considered aligned. In strict mode, only
   an exact match between both of the Fully Qualified Domain Names
   (FQDNs) is considered to produce Identifier Alignment.

   To illustrate, in relaxed mode, if a validated DKIM signature
   successfully verifies with a "d=" domain of "example.com", and the
   RFC5322.From address is "alerts@news.example.com", the DKIM "d="
   domain and the RFC5322.From domain are considered to be "in
   alignment". In strict mode, this test would fail, since the "d="
   domain does not exactly match the FQDN of the address.

   However, a DKIM signature bearing a value of "d=com" would never
   allow an "in alignment" result, as "com" should appear on all public
   suffix lists (see Appendix A.6.1) and therefore cannot be an
   Organizational Domain.

   Identifier Alignment is required 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 authenticity of the Author Domain.

   Note that a single email can contain multiple DKIM signatures, and it
   is considered to be a DMARC "pass" if any DKIM signature is aligned
   and verifies.

3.1.2. SPF-Authenticated Identifiers

   DMARC permits Identifier Alignment, based on the result of an SPF
   authentication, to be strict or relaxed.

   In relaxed mode, the [SPF]-authenticated domain and RFC5322.From
   domain must have the same Organizational Domain. In strict mode,
   only an exact DNS domain match is considered to produce Identifier
   Alignment.

   Note that the RFC5321.HELO identity is not typically used in the
   context of DMARC (except when required to "fake" an otherwise null
   reverse-path), even though a "pure SPF" implementation according to
   [SPF] would check that identifier.









Kucherawy & Zwicky Informational [Page 10]
 
RFC 7489 DMARC March 2015


   For example, if a message passes an SPF check with an
   RFC5321.MailFrom domain of "cbg.bounces.example.com", and the address
   portion of the RFC5322.From field contains "payments@example.com",
   the Authenticated RFC5321.MailFrom domain identifier and the
   RFC5322.From domain are considered to be "in alignment" in relaxed
   mode, but not in strict mode.

3.1.3. Alignment and Extension Technologies

   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.

3.2. Organizational Domain

   The Organizational Domain is determined using the following
   algorithm:

   1. Acquire a "public suffix" list, 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 ".co.uk"; other TLDs
       such as ".com" appear in the IANA registry of top-level DNS
       domains. A public suffix list is the union of all of these.
       Appendix A.6.1 contains some discussion about obtaining a public
       suffix list.

   2. Break the subject DNS domain name into a set of "n" ordered
       labels. Number these labels from right to left; e.g., for
       "example.com", "com" would be label 1 and "example" would be
       label 2.

   3. 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 "x".

   4. Construct a new DNS domain name using the name that matched from
       the public suffix list and prefixing to it the "x+1"th label from
       the subject domain. This new name is the Organizational Domain.

   Thus, since "com" is an IANA-registered TLD, a subject domain of
   "a.b.c.d.example.com" would have an Organizational Domain of
   "example.com".

   The process of determining a suffix is currently a heuristic one. No
   list is guaranteed to be accurate or current.




Kucherawy & Zwicky Informational [Page 11]
 
RFC 7489 DMARC March 2015


4. Overview

   This section provides a general overview of the design and operation
   of the DMARC environment.

4.1. Authentication Mechanisms

   The following mechanisms for determining Authenticated Identifiers
   are supported in this version of DMARC:

   o [DKIM], which provides a domain-level identifier in the content of
      the "d=" tag of a validated DKIM-Signature header field.

   o [SPF], which can authenticate both the domain found in an [SMTP]
      HELO/EHLO command (the HELO identity) and the domain found in an
      SMTP MAIL command (the MAIL FROM identity). DMARC uses the result
      of SPF authentication of the MAIL FROM identity. Section 2.4 of
      [SPF] describes MAIL FROM processing for cases in which the MAIL
      command has a null path.

4.2. Key Concepts

   DMARC policies are published by the Domain Owner, and retrieved by
   the Mail Receiver during the SMTP session, via the DNS.

   DMARC's filtering function is based on whether the RFC5322.From field
   domain is aligned with (matches) an authenticated domain name from
   SPF or DKIM. When a DMARC policy is published for the domain name
   found in the RFC5322.From field, and that domain name is not
   validated through SPF or DKIM, the disposition of that message can be
   affected by that DMARC policy when delivered to a participating
   receiver.

   It is important to note that the authentication mechanisms employed
   by DMARC authenticate only a DNS domain and do not authenticate the
   local-part of any email address identifier found in a message, nor do
   they validate the legitimacy of message content.

   DMARC's feedback component involves the collection of information
   about received messages claiming to be from the Organizational Domain
   for periodic aggregate reports to the Domain Owner. The parameters
   and format for such reports are discussed in later sections of this
   document.

   A DMARC-enabled Mail Receiver might also generate per-message reports
   that contain information related to individual messages that fail SPF
   and/or DKIM. Per-message failure reports are a useful source of
   information when debugging deployments (if messages can be determined



Kucherawy & Zwicky Informational [Page 12]
 
RFC 7489 DMARC March 2015


   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 [AFRF].

   A message satisfies the DMARC checks if at least one of the supported
   authentication mechanisms:

   1. produces a "pass" result, and

   2. produces that result based on an identifier that is in alignment,
       as defined in Section 3.

4.3. Flow Diagram

    +---------------+
    | Author Domain |< . . . . . . . . . . . . . . . . . . . . . . .
    +---------------+ . . .
        | . . .
        V V V .
    +-----------+ +--------+ +----------+ +----------+ .
    | MSA |<***>| DKIM | | DKIM | | SPF | .
    | Service | | Signer | | Verifier | | Verifier | .
    +-----------+ +--------+ +----------+ +----------+ .
        | ^ ^ .
        | ************** .
        V * .
     +------+ (~~~~~~~~~~~~) +------+ * .
     | sMTA |------->( other MTAs )----->| rMTA | * .
     +------+ (~~~~~~~~~~~~) +------+ * .
                                            | * ........
                                            | * .
                                            V * .
                                     +-----------+ V V
                       +---------+ | MDA | +----------+
                       | User |<--| Filtering |<***>| DMARC |
                       | Mailbox | | Engine | | Verifier |
                       +---------+ +-----------+ +----------+

     MSA = Mail Submission Agent
     MDA = Mail Delivery Agent

   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. "sMTA" is the sending MTA, and "rMTA" is the
   receiving MTA.



Kucherawy & Zwicky Informational [Page 13]
 
RFC 7489 DMARC March 2015


   In essence, the steps are as follows:

   1. Domain Owner constructs an SPF policy and publishes it in its
        DNS database as per [SPF]. Domain Owner also configures its
        system for DKIM signing as described in [DKIM]. Finally, Domain
        Owner publishes via the DNS a DMARC message-handling policy.

   2. Author generates a message and hands the message to Domain
        Owner's designated mail submission service.

   3. Submission service passes relevant details to the DKIM signing
        module in order to generate a DKIM signature to be applied to
        the message.

   4. Submission service relays the now-signed message to its
        designated transport service for routing to its intended
        recipient(s).

   5. Message may pass through other relays but eventually arrives at
        a recipient's transport service.

   6. Recipient delivery service conducts SPF and DKIM authentication
        checks by passing the necessary data to their respective
        modules, each of which requires queries to the Author Domain's
        DNS data (when identifiers are aligned; see below).

   7. The results of these are passed to the DMARC module along with
        the Author's domain. The DMARC module attempts to retrieve a
        policy from the DNS for that domain. If none is found, the
        DMARC module determines the Organizational Domain and repeats
        the attempt to retrieve a policy from the DNS. (This is
        described in further detail in Section 6.6.3.)

   8. If a policy is found, it is combined with the Author's domain
        and the SPF and DKIM results to produce a DMARC policy result (a
        "pass" or "fail") and can optionally cause one of two kinds of
        reports to be generated (not shown).

   9. Recipient transport service either delivers the message to the
        recipient inbox or takes other local policy action based on the
        DMARC result (not shown).

   10. When requested, Recipient transport service collects data from
        the message delivery session to be used in providing feedback
        (see Section 7).






Kucherawy & Zwicky Informational [Page 14]
 
RFC 7489 DMARC March 2015


5. Use of RFC5322.From

   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.

   Several points suggest that it is the most correct and safest thing
   to do in this context:

   o Of all the identifiers that are part of the message itself, this
      is the only one guaranteed to be present.

   o 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.

   The absence of a single, properly formed RFC5322.From field renders
   the message invalid. Handling of such a message is outside of the
   scope of this specification.

   Since the sorts of mail typically protected by DMARC participants
   tend to only have single Authors, DMARC participants generally
   operate under a slightly restricted profile of RFC5322 with respect
   to the expected syntax of this field. See Section 6.6 for details.

6. Policy

   DMARC policies are published by Domain Owners and applied by Mail
   Receivers.

   A Domain Owner advertises DMARC participation of one or more of its
   domains by adding a DNS TXT record (described in Section 6.1) to
   those domains. In doing so, Domain Owners make specific requests of
   Mail Receivers regarding the disposition of messages purporting to be
   from one of the Domain Owner's domains and the provision of feedback
   about those messages.

   A Domain Owner may choose not to participate in DMARC evaluation by
   Mail Receivers. 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 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.





Kucherawy & Zwicky Informational [Page 15]
 
RFC 7489 DMARC March 2015


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

6.1. DMARC Policy Record

   Domain Owner DMARC preferences are stored as DNS TXT records in
   subdomains named "_dmarc". For example, the Domain Owner of
   "example.com" would post DMARC preferences in a TXT record at
   "_dmarc.example.com". Similarly, a Mail Receiver wishing to query
   for DMARC preferences regarding mail with an RFC5322.From domain of
   "example.com" would issue a TXT query to the DNS for the subdomain of
   "_dmarc.example.com". The DNS-located DMARC preference data will
   hereafter be called the "DMARC record".

   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, namely an
   isolated TXT record that is restricted to the DMARC context. Use of
   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.

   Per [DNS], a TXT record can comprise several "character-string"
   objects. Where this is the case, the module performing DMARC
   evaluation MUST concatenate these strings by joining together the
   objects in order and parsing the result as a single string.

6.2. DMARC URIs

   [URI] defines a generic syntax for identifying a resource. The DMARC
   mechanism uses this as the format by which a Domain Owner specifies
   the destination for the two report types that are supported.

   The place such URIs are specified (see Section 6.3) allows a list of
   these to be provided. A report is normally sent to each listed URI
   in the order provided by the Domain Owner. 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. The list of URIs is
   separated by commas (ASCII 0x2C).



Kucherawy & Zwicky Informational [Page 16]
 
RFC 7489 DMARC March 2015


   Each URI can have associated with it a maximum report size that may
   be sent to it. This is accomplished by appending an exclamation
   point (ASCII 0x21), followed by a maximum-size indication, before a
   separating comma or terminating semicolon.

   Thus, a DMARC URI is a URI within which any commas or exclamation
   points are percent-encoded per [URI], followed by an OPTIONAL
   exclamation point and a maximum-size specification, and, if there are
   additional reporting URIs in the list, a comma and the next URI.

   For example, the URI "mailto:reports@example.com!50m" would request
   that a report be sent via email to "reports@example.com" so long as
   the report payload does not exceed 50 megabytes.

   A formal definition is provided in Section 6.4.

6.3. General Record Format

   DMARC records follow the extensible "tag-value" syntax for DNS-based
   key records defined in DKIM [DKIM].

   Section 11 creates a registry for known DMARC tags and registers the
   initial set defined in this document. Only tags defined in this
   document or in later extensions, and thus added to that registry, are
   to be processed; unknown tags MUST be ignored.

   The following tags are introduced as the initial valid DMARC tags:

   adkim: (plain-text; OPTIONAL; default is "r".) Indicates whether
      strict or relaxed DKIM Identifier Alignment mode is required by
      the Domain Owner. See Section 3.1.1 for details. Valid values
      are as follows:

      r: relaxed mode

      s: strict mode

   aspf: (plain-text; OPTIONAL; default is "r".) Indicates whether
      strict or relaxed SPF Identifier Alignment mode is required by the
      Domain Owner. See Section 3.1.2 for details. Valid values are as
      follows:

      r: relaxed mode

      s: strict mode






Kucherawy & Zwicky Informational [Page 17]
 
RFC 7489 DMARC March 2015


   fo: Failure reporting options (plain-text; OPTIONAL; default is "0")
      Provides requested options for generation of failure reports.
      Report generators MAY choose to adhere to the requested options.
      This tag's content MUST be ignored if a "ruf" tag (below) is not
      also specified. The value of this tag is a colon-separated list
      of characters that indicate failure reporting options as follows:

      0: Generate a DMARC failure report if all underlying
         authentication mechanisms fail to produce an aligned "pass"
         result.

      1: Generate a DMARC failure report if any underlying
         authentication mechanism produced something other than an
         aligned "pass" result.

      d: Generate a DKIM failure report if the message had a signature
         that failed evaluation, regardless of its alignment. DKIM-
         specific reporting is described in [AFRF-DKIM].

      s: Generate an SPF failure report if the message failed SPF
         evaluation, regardless of its alignment. SPF-specific
         reporting is described in [AFRF-SPF].

   p: Requested Mail Receiver policy (plain-text; REQUIRED for policy
      records). Indicates the policy to be enacted by the Receiver at
      the request of the Domain Owner. Policy applies to the domain
      queried and to subdomains, unless subdomain policy is explicitly
      described using the "sp" tag. This tag is mandatory for policy
      records only, but not for third-party reporting records (see
      Section 7.1). Possible values are as follows:

      none: The Domain Owner requests no specific action be taken
         regarding delivery of messages.

      quarantine: The Domain Owner wishes to have email that fails the
         DMARC mechanism check be treated by Mail Receivers as
         suspicious. Depending on the capabilities of the Mail
         Receiver, this can mean "place into spam folder", "scrutinize
         with additional intensity", and/or "flag as suspicious".

      reject: The Domain Owner wishes for Mail Receivers to reject
         email that fails the DMARC mechanism check. Rejection SHOULD
         occur during the SMTP transaction. See Section 10.3 for some
         discussion of SMTP rejection methods and their implications.

   pct: (plain-text integer between 0 and 100, inclusive; OPTIONAL;
      default is 100). Percentage of messages from the Domain Owner's
      mail stream to which the DMARC policy is to be applied. However,



Kucherawy & Zwicky Informational [Page 18]
 
RFC 7489 DMARC March 2015


      this MUST NOT be applied to the DMARC-generated reports, all of
      which must be sent and received unhindered. The purpose of the
      "pct" tag is to allow Domain Owners to enact a slow rollout
      enforcement of the DMARC mechanism. The prospect of "all or
      nothing" is recognized as preventing many organizations from
      experimenting with strong authentication-based mechanisms. See
      Section 6.6.4 for details. Note that random selection based on
      this percentage, such as the following pseudocode, is adequate:

       if (random mod 100) < pct then
         selected = true
       else
         selected = false

   rf: Format to be used for message-specific failure reports (colon-
      separated plain-text list of values; OPTIONAL; default is "afrf").
      The value of this tag is a list of one or more report formats as
      requested by the Domain Owner to be used when a message fails both
      [SPF] and [DKIM] tests to report details of the individual
      failure. The values MUST be present in the registry of reporting
      formats defined in Section 11; a Mail Receiver observing a
      different value SHOULD ignore it or MAY ignore the entire DMARC
      record. For this version, only "afrf" (the auth-failure report
      type defined in [AFRF]) is presently supported. See Section 7.3
      for details. For interoperability, the Authentication Failure
      Reporting Format (AFRF) MUST be supported.

   ri: Interval requested between aggregate reports (plain-text 32-bit
      unsigned integer; OPTIONAL; default is 86400). Indicates a
      request to Receivers to generate aggregate reports separated by no
      more than the requested number of seconds. DMARC implementations
      MUST be able to provide daily reports and SHOULD be able to
      provide hourly reports when requested. However, anything other
      than a daily report is understood to be accommodated on a best-
      effort basis.

   rua: Addresses to which aggregate feedback is to be sent (comma-
      separated plain-text list of DMARC URIs; OPTIONAL). A comma or
      exclamation point that is part of such a DMARC URI MUST be encoded
      per Section 2.1 of [URI] so as to distinguish it from the list
      delimiter or an OPTIONAL size limit. Section 7.1 discusses
      considerations that apply when the domain name of a URI differs
      from that of the domain advertising the policy. See Section 12.5
      for additional considerations. Any valid URI can be specified. A
      Mail Receiver MUST implement support for a "mailto:" URI, i.e.,
      the ability to send a DMARC report via electronic mail. If not





Kucherawy & Zwicky Informational [Page 19]
 
RFC 7489 DMARC March 2015


      provided, Mail Receivers MUST NOT generate aggregate feedback
      reports. URIs not supported by Mail Receivers MUST be ignored.
      The aggregate feedback report format is described in Section 7.2.

   ruf: Addresses to which message-specific failure information is to
      be reported (comma-separated plain-text list of DMARC URIs;
      OPTIONAL). 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 "fo" tag
      above). The format of the message to be generated MUST follow the
      format specified for the "rf" tag. Section 7.1 discusses
      considerations that apply when the domain name of a URI differs
      from that of the domain advertising the policy. A Mail Receiver
      MUST implement support for a "mailto:" URI, i.e., the ability to
      send a DMARC report via electronic mail. If not provided, Mail
      Receivers MUST NOT generate failure reports. See Section 12.5 for
      additional considerations.

   sp: Requested Mail Receiver policy for all subdomains (plain-text;
      OPTIONAL). Indicates the policy to be enacted by the Receiver at
      the request of the Domain Owner. It applies only to subdomains of
      the domain queried and not to the domain itself. Its syntax is
      identical to that of the "p" tag defined above. If absent, the
      policy specified by the "p" tag MUST be applied for subdomains.
      Note that "sp" will be ignored for DMARC records published on
      subdomains of Organizational Domains due to the effect of the
      DMARC policy discovery mechanism described in Section 6.6.3.

   v: Version (plain-text; REQUIRED). Identifies the record retrieved
      as a DMARC record. It MUST have the value of "DMARC1". The value
      of this tag MUST match precisely; if it does not or it is absent,
      the entire retrieved record MUST be ignored. It MUST be the first
      tag in the list.

   A DMARC policy record MUST comply with the formal specification found
   in Section 6.4 in that the "v" and "p" tags MUST be present and MUST
   appear in that order. Unknown tags MUST be ignored. Syntax errors
   in the remainder of the record SHOULD be discarded in favor of
   default values (if any) or ignored outright.

   Note that given the rules of the previous paragraph, 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 "v" tag's value), but a change to any existing tags does require
   a new version of DMARC.






Kucherawy & Zwicky Informational [Page 20]
 
RFC 7489 DMARC March 2015


6.4. Formal Definition

   The formal definition of the DMARC format, using [ABNF], is as
   follows:

     dmarc-uri = URI [ "!" 1*DIGIT [ "k" / "m" / "g" / "t" ] ]
                       ; "URI" is imported from [URI]; commas (ASCII
                       ; 0x2C) and exclamation points (ASCII 0x21)
                       ; MUST be encoded; the numeric portion MUST fit
                       ; within an unsigned 64-bit integer

     dmarc-record = dmarc-version dmarc-sep
                       [dmarc-request]
                       [dmarc-sep dmarc-srequest]
                       [dmarc-sep dmarc-auri]
                       [dmarc-sep dmarc-furi]
                       [dmarc-sep dmarc-adkim]
                       [dmarc-sep dmarc-aspf]
                       [dmarc-sep dmarc-ainterval]
                       [dmarc-sep dmarc-fo]
                       [dmarc-sep dmarc-rfmt]
                       [dmarc-sep dmarc-percent]
                       [dmarc-sep]
                       ; components other than dmarc-version and
                       ; dmarc-request may appear in any order

     dmarc-version = "v" *WSP "=" *WSP %x44 %x4d %x41 %x52 %x43 %x31

     dmarc-sep = *WSP %x3b *WSP

     dmarc-request = "p" *WSP "=" *WSP
                       ( "none" / "quarantine" / "reject" )

     dmarc-srequest = "sp" *WSP "=" *WSP
                       ( "none" / "quarantine" / "reject" )

     dmarc-auri = "rua" *WSP "=" *WSP
                       dmarc-uri *(*WSP "," *WSP dmarc-uri)

     dmarc-furi = "ruf" *WSP "=" *WSP
                       dmarc-uri *(*WSP "," *WSP dmarc-uri)

     dmarc-adkim = "adkim" *WSP "=" *WSP
                       ( "r" / "s" )

     dmarc-aspf = "aspf" *WSP "=" *WSP
                       ( "r" / "s" )




Kucherawy & Zwicky Informational [Page 21]
 
RFC 7489 DMARC March 2015


     dmarc-ainterval = "ri" *WSP "=" *WSP 1*DIGIT

     dmarc-fo = "fo" *WSP "=" *WSP
                       ( "0" / "1" / "d" / "s" )
                       *(*WSP ":" *WSP ( "0" / "1" / "d" / "s" ))

     dmarc-rfmt = "rf" *WSP "=" *WSP Keyword *(*WSP ":" Keyword)
                       ; registered reporting formats only

     dmarc-percent = "pct" *WSP "=" *WSP
                       1*3DIGIT

   "Keyword" is imported from Section 4.1.2 of [SMTP].

   A size limitation in a dmarc-uri, if provided, is interpreted as a
   count of units followed by an OPTIONAL unit size ("k" for kilobytes,
   "m" for megabytes, "g" for gigabytes, "t" for terabytes). Without a
   unit, the number is presumed to be a basic byte count. Note that the
   units are considered to be powers of two; a kilobyte is 2^10, a
   megabyte is 2^20, etc.

6.5. Domain Owner Actions

   To implement the DMARC mechanism, the only action required of a
   Domain Owner is the creation of the DMARC policy record in the DNS.
   However, in order to make meaningful use of DMARC, a Domain Owner
   must at minimum either establish an address to receive reports, or
   deploy authentication technologies and ensure Identifier Alignment.
   Most Domain Owners will want to do both.

   DMARC reports will be of significant size, and the addresses that
   receive them are publicly visible, so we encourage Domain Owners to
   set up dedicated email addresses to receive and process reports, and
   to deploy abuse countermeasures on those email addresses as
   appropriate.

   Authentication technologies are discussed in [DKIM] (see also
   [DKIM-OVERVIEW] and [DKIM-DEPLOYMENT]) and [SPF].













Kucherawy & Zwicky Informational [Page 22]
 
RFC 7489 DMARC March 2015


6.6. Mail Receiver Actions

   This section describes receiver actions in the DMARC environment.

6.6.1. Extract Author Domain

   The domain in the RFC5322.From field is extracted as the domain to be
   evaluated by DMARC. If the domain is encoded with UTF-8, the domain
   name must be converted to an A-label, as described in Section 2.3 of
   [IDNA], for further processing.

   In order to be processed by DMARC, a message typically needs to
   contain exactly one RFC5322.From domain (a single From: field with a
   single domain in it). Not all messages meet this requirement, and
   handling of them is outside of the scope of this document. Typical
   exceptions, and the way they have been historically handled by DMARC
   participants, are as follows:

   o Messages with multiple RFC5322.From fields are typically rejected,
      since that form is forbidden under RFC 5322 [MAIL];

   o Messages bearing a single RFC5322.From field containing multiple
      addresses (and, thus, multiple domain names to be evaluated) are
      typically rejected because the sorts of mail normally protected by
      DMARC do not use this format;

   o Messages that have no RFC5322.From field at all are typically
      rejected, since that form is forbidden under RFC 5322 [MAIL];

   o Messages with an RFC5322.From field that contains no meaningful
      domains, such as RFC 5322 [MAIL]'s "group" syntax, are typically
      ignored.

   The case of a syntactically valid multi-valued RFC5322.From field
   presents a particular challenge. The process in this case is to
   apply the DMARC check using each of those domains found in the
   RFC5322.From field as the Author Domain and apply the most strict
   policy selected among the checks that fail.













Kucherawy & Zwicky Informational [Page 23]
 
RFC 7489 DMARC March 2015


6.6.2. Determine Handling Policy

   To arrive at a policy for an individual message, Mail Receivers MUST
   perform the following actions or their semantic equivalents.
   Steps 2-4 MAY be done in parallel, whereas steps 5 and 6 require
   input from previous steps.

   The steps are as follows:

   1. Extract the RFC5322.From domain from the message (as above).

   2. Query the DNS for a DMARC policy record. Continue if one is
       found, or terminate DMARC evaluation otherwise. See
       Section 6.6.3 for details.

   3. 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 and MUST include the
       value of the "d=" tag from each checked DKIM signature.

   4. Perform SPF validation checks. The results of this step are
       passed to the remainder of the algorithm and MUST include the
       domain name used to complete the SPF check.

   5. Conduct Identifier Alignment checks. With authentication checks
       and policy discovery performed, the Mail Receiver checks to see
       if Authenticated Identifiers fall into alignment as described in
       Section 3. If one or more of the Authenticated Identifiers align
       with the RFC5322.From domain, the message is considered to pass
       the DMARC mechanism check. All other conditions (authentication
       failures, identifier mismatches) are considered to be DMARC
       mechanism check failures.

   6. Apply policy. Emails that fail the DMARC mechanism check are
       disposed of in accordance with the discovered DMARC policy of the
       Domain Owner. See Section 6.3 for details.

   Heuristics applied in the absence of use by a Domain Owner of either
   SPF or DKIM (e.g., [Best-Guess-SPF]) SHOULD NOT be used, as it may be
   the case that the Domain Owner wishes a Message Receiver not to
   consider the results of that underlying authentication protocol at
   all.

   DMARC evaluation can only yield a "pass" 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 Receiver evaluating the message is unable to
   conclude that the DMARC mechanism had a permanent failure; they



Kucherawy & Zwicky Informational [Page 24]
 
RFC 7489 DMARC March 2015


   therefore cannot apply the advertised DMARC policy. When otherwise
   appropriate, Receivers MAY send feedback reports regarding temporary
   errors.

   Handling of messages for which SPF and/or DKIM evaluation encounter a
   permanent DNS error is left to the discretion of the Mail Receiver.

6.6.3. Policy Discovery

   As stated above, the DMARC mechanism uses DNS TXT records to
   advertise policy. Policy discovery is accomplished via a method
   similar to the method used for SPF records. This method, and the
   important differences between DMARC and SPF mechanisms, are discussed
   below.

   To balance the conflicting requirements of supporting wildcarding,
   allowing subdomain policy overrides, and limiting DNS query load, the
   following DNS lookup scheme is employed:

   1. Mail Receivers MUST query the DNS for a DMARC TXT record at the
       DNS domain matching the one found in the RFC5322.From domain in
       the message. A possibly empty set of records is returned.

   2. Records that do not start with a "v=" tag that identifies the
       current version of DMARC are discarded.

   3. If the set is now empty, the Mail Receiver MUST query the DNS for
       a DMARC TXT record at the DNS domain matching the Organizational
       Domain in place of the RFC5322.From domain in the message (if
       different). This record can contain policy to be asserted for
       subdomains of the Organizational Domain. A possibly empty set of
       records is returned.

   4. Records that do not start with a "v=" tag that identifies the
       current version of DMARC are discarded.

   5. If the remaining set contains multiple records or no records,
       policy discovery terminates and DMARC processing is not applied
       to this message.












Kucherawy & Zwicky Informational [Page 25]
 
RFC 7489 DMARC March 2015


   6. If a retrieved policy record does not contain a valid "p" tag, or
       contains an "sp" tag that is not valid, then:

       1. if a "rua" tag is present and contains at least one
           syntactically valid reporting URI, the Mail Receiver SHOULD
           act as if a record containing a valid "v" tag and "p=none"
           was retrieved, and continue processing;

       2. otherwise, the Mail Receiver applies no DMARC processing to
           this message.

   If the set produced by the mechanism above contains no DMARC policy
   record (i.e., any indication that there is no such record as opposed
   to a transient DNS error), Mail Receivers SHOULD NOT apply the DMARC
   mechanism to the message.

   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 ("fail open"), 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 ("fail
   closed").

6.6.4. Message Sampling

   If the "pct" tag is present in the policy record, the Mail Receiver
   MUST NOT enact the requested policy ("p" tag or "sp" tag") on more
   than the stated percent of the totality of affected messages.
   However, regardless of whether or not the "pct" tag is present, the
   Mail Receiver MUST include all relevant message data in any reports
   produced.

   If email is subject to the DMARC policy of "quarantine", the Mail
   Receiver SHOULD quarantine the message. If the email is not subject
   to the "quarantine" policy (due to the "pct" tag), the Mail Receiver
   SHOULD apply local message classification as normal.

   If email is subject to the DMARC policy of "reject", the Mail
   Receiver SHOULD reject the message (see Section 10.3). If the email
   is not subject to the "reject" policy (due to the "pct" tag), the
   Mail Receiver SHOULD treat the email as though the "quarantine"
   policy applies. This behavior allows Domain Owners to experiment
   with progressively stronger policies without relaxing existing
   policy.





Kucherawy & Zwicky Informational [Page 26]
 
RFC 7489 DMARC March 2015


   Mail Receivers implement "pct" via statistical mechanisms that
   achieve a close approximation to the requested percentage and provide
   a representative sample across a reporting period.

6.6.5. Store Results of DMARC Processing

   The results of Mail Receiver-based DMARC processing should be stored
   for eventual presentation back to the Domain Owner in the form of
   aggregate feedback reports. Sections 6.3 and 7.2 discuss aggregate
   feedback.

6.7. Policy Enforcement Considerations

   Mail Receivers MAY 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 "good". Mail
   Receivers are encouraged to maintain anti-abuse technologies to
   combat the possibility of DMARC-enabled criminal campaigns.

   Mail Receivers MAY choose to accept email that fails the DMARC
   mechanism check even if the Domain Owner has published a "reject"
   policy. Mail Receivers need to make a best effort not to increase
   the likelihood of accepting abusive mail if they choose not to comply
   with a Domain Owner's reject, against policy. At a minimum, addition
   of the Authentication-Results header field (see [AUTH-RESULTS]) is
   RECOMMENDED when delivery of failing mail is done. When this is
   done, the DNS domain name thus recorded MUST be encoded as an
   A-label.

   Mail Receivers are only obligated to report reject or quarantine
   policy actions in aggregate feedback reports that are due to DMARC
   policy. They are not required to report reject or quarantine actions
   that are the result of local policy. If local policy information is
   exposed, abusers can gain insight into the effectiveness and delivery
   rates of spam campaigns.

   Final disposition 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.

   DMARC-compliant Mail Receivers typically disregard any mail-handling
   directive discovered as part of an authentication mechanism (e.g.,
   Author Domain Signing Practices (ADSP), SPF) where a DMARC record is
   also discovered that specifies a policy other than "none". Deviating



Kucherawy & Zwicky Informational [Page 27]
 
RFC 7489 DMARC March 2015


   from this practice introduces inconsistency among DMARC operators in
   terms of handling of the message. However, such deviation is not
   proscribed.

   To enable Domain Owners to receive DMARC feedback without impacting
   existing mail processing, discovered policies of "p=none" SHOULD NOT
   modify existing mail disposition processing.

   Mail Receivers SHOULD also implement reporting instructions of DMARC,
   even in the absence of a request for DKIM reporting [AFRF-DKIM] or
   SPF reporting [AFRF-SPF]. Furthermore, the presence of such requests
   SHOULD NOT affect DMARC reporting.

7. DMARC Feedback

   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 better willing and able to use
   quarantine and reject policies.

7.1. Verifying External Destinations

   It is possible to specify destinations for the different reports that
   are outside the authority of the Domain Owner making the request.
   This allows domains that do not operate mail servers to request
   reports and have them go someplace that is able to receive and
   process them.

   Without checks, this would allow a bad actor to publish a DMARC
   policy record that requests that reports be sent to a victim address,
   and then send a large volume of mail that will fail both DKIM and SPF
   checks to a wide variety of destinations; the victim will in turn be
   flooded with unwanted reports. Therefore, a verification mechanism
   is included.

   When a Mail Receiver discovers a DMARC policy in the DNS, and the
   Organizational Domain at which that record was discovered is not
   identical to the Organizational Domain of the host part of the
   authority component of a [URI] specified in the "rua" or "ruf" tag,
   the following verification steps are to be taken:

   1. Extract the host portion of the authority component of the URI.
       Call this the "destination host", as it refers to a Report
       Receiver.

   2. Prepend the string "_report._dmarc".



Kucherawy & Zwicky Informational [Page 28]
 
RFC 7489 DMARC March 2015


   3. Prepend the domain name from which the policy was retrieved,
       after conversion to an A-label if needed.

   4. Query the DNS for a TXT record at the constructed name. If the
       result of this request is a temporary DNS error of some kind
       (e.g., a timeout), the Mail Receiver MAY elect to temporarily
       fail the delivery so the verification test can be repeated later.

   5. For each record returned, parse the result as a series of
       "tag=value" pairs, i.e., the same overall format as the policy
       record (see Section 6.4). In particular, the "v=DMARC1" tag is
       mandatory and MUST appear first in the list. Discard any that do
       not pass this test.

   6. If the result includes no TXT resource records that pass basic
       parsing, a positive determination of the external reporting
       relationship cannot be made; stop.

   7. If at least one TXT resource record remains in the set after
       parsing, then the external reporting arrangement was authorized
       by the Report Receiver.

   8. If a "rua" or "ruf" tag is thus discovered, replace the
       corresponding value extracted from the domain's DMARC policy
       record with the one found in this record. This permits the
       Report Receiver to override the report destination. However, to
       prevent loops or indirect abuse, the overriding URI MUST use the
       same destination host from the first step.

   For example, if a DMARC policy query for "blue.example.com" contained
   "rua=mailto:reports@red.example.net", the host extracted from the
   latter ("red.example.net") does not match "blue.example.com", so this
   procedure is enacted. A TXT query for
   "blue.example.com._report._dmarc.red.example.net" is issued. If a
   single reply comes back containing a tag of "v=DMARC1", then the
   relationship between the two is confirmed. Moreover,
   "red.example.net" has the opportunity to override the report
   destination requested by "blue.example.com" if needed.

   Where the above algorithm fails to confirm that the external
   reporting was authorized by the Report Receiver, the URI MUST be
   ignored by the Mail Receiver generating the report. Further, if the
   confirming record includes a URI whose host is again different than
   the domain publishing that override, the Mail Receiver generating the
   report MUST NOT generate a report to either the original or the
   override URI.





Kucherawy & Zwicky Informational [Page 29]
 
RFC 7489 DMARC March 2015


   A Report Receiver publishes such a record in its DNS if it wishes to
   receive reports for other domains.

   A Report Receiver that is willing to receive reports for any domain
   can use a wildcard DNS record. For example, a TXT resource record at
   "*._report._dmarc.example.com" containing at least "v=DMARC1"
   confirms that example.com is willing to receive DMARC reports for any
   domain.

   If the Report Receiver is overcome by volume, it can simply remove
   the confirming DNS record. However, due to positive caching, the
   change could take as long as the time-to-live (TTL) on the record to
   go into effect.

   A Mail Receiver might decide not to enact this procedure if, for
   example, it relies on a local list of domains for which external
   reporting addresses are permitted.

7.2. Aggregate Reports

   The DMARC aggregate feedback report is designed to provide Domain
   Owners with precise insight into:

   o authentication results,

   o corrective action that needs to be taken by Domain Owners, and

   o the effect of Domain Owner DMARC policy on email streams processed
      by Mail Receivers.

   Aggregate DMARC feedback provides visibility into real-world email
   streams that Domain Owners need to make informed decisions regarding
   the publication of DMARC policy. When Domain Owners know what
   legitimate mail they are sending, what the authentication results are
   on that mail, and what forged mail receivers are getting, they can
   make better decisions about the policies they need and the steps they
   need to take to enable those policies. When Domain Owners set
   policies appropriately and understand their effects, Mail Receivers
   can act on them confidently.

   Visibility comes in the form of daily (or more frequent) Mail
   Receiver-originated feedback reports that contain aggregate data on
   message streams relevant to the Domain Owner. This information
   includes data about messages that passed DMARC authentication as well
   as those that did not.

   The format for these reports is defined in Appendix C.




Kucherawy & Zwicky Informational [Page 30]
 
RFC 7489 DMARC March 2015


   The report SHOULD include the following data:

   o The DMARC policy discovered and applied, if any

   o The selected message disposition

   o The identifier evaluated by SPF and the SPF result, if any

   o The identifier evaluated by DKIM and the DKIM result, if any

   o For both DKIM and SPF, an indication of whether the identifier was
      in alignment

   o Data for each Domain Owner's subdomain separately from mail from
      the sender's Organizational Domain, even if there is no explicit
      subdomain policy

   o Sending and receiving domains

   o The policy requested by the Domain Owner and the policy actually
      applied (if different)

   o The number of successful authentications

   o The counts of messages based on all messages received, even if
      their delivery is ultimately blocked by other filtering agents

   Note that Domain Owners or their agents may change the published
   DMARC policy for a domain or subdomain at any time. From a Mail
   Receiver's perspective, this will occur during a reporting period and
   may be noticed during that period, at the end of that period when
   reports are generated, or during a subsequent reporting period, all
   depending on the Mail Receiver's implementation. Under these
   conditions, it is possible that a Mail Receiver could do any of the
   following:

   o generate for such a reporting period a single aggregate report
      that includes message dispositions based on the old policy, or a
      mix of the two policies, even though the report only contains a
      single "policy_published" element;

   o generate multiple reports for the same period, one for each
      published policy occurring during the reporting period;

   o generate a report whose end time occurs when the updated policy
      was detected, regardless of any requested report interval.





Kucherawy & Zwicky Informational [Page 31]
 
RFC 7489 DMARC March 2015


   Such policy changes are expected to be infrequent for any given
   domain, whereas more stringent policy monitoring requirements on the
   Mail Receiver would produce a very large burden at Internet scale.
   Therefore, it is the responsibility of report consumers and Domain
   Owners to be aware of this situation and allow for such mixed reports
   during the propagation of the new policy to Mail Receivers.

   Aggregate reports are most useful when they all cover a common time
   period. By contrast, correlation of these reports from multiple
   generators when they cover incongruent time periods is difficult or
   impossible. Report generators SHOULD, wherever possible, adhere to
   hour boundaries for the reporting period they are using. For
   example, starting a per-day report at 00:00; starting per-hour
   reports at 00:00, 01:00, 02:00; etc. Report generators using a
   24-hour report period are strongly encouraged to begin that period at
   00:00 UTC, regardless of local timezone or time of report production,
   in order to facilitate correlation.

   A Mail Receiver discovers reporting requests when it looks up a DMARC
   policy record that corresponds to an RFC5322.From domain on received
   mail. The presence of the "rua" tag specifies where to send
   feedback.

7.2.1. Transport

   Where the URI specified in a "rua" tag does not specify otherwise, a
   Mail Receiver generating a feedback report SHOULD employ a secure
   transport mechanism.

   The Mail Receiver, after preparing a report, MUST evaluate the
   provided reporting URIs in the order given. Any reporting URI that
   includes a size limitation exceeded by the generated report (after
   compression and after any encoding required by the particular
   transport mechanism) MUST NOT be used. An attempt MUST be made to
   deliver an aggregate report to every remaining URI, up to the
   Receiver's limits on supported URIs.

   If transport is not possible because the services advertised by the
   published URIs are not able to accept reports (e.g., the URI refers
   to a service that is unreachable, or all provided URIs specify size
   limits exceeded by the generated record), the Mail Receiver SHOULD
   send a short report (see Section 7.2.2) indicating that a report is
   available but could not be sent. The Mail Receiver MAY cache that
   data and try again later, or MAY discard data that could not be sent.







Kucherawy & Zwicky Informational [Page 32]
 
RFC 7489 DMARC March 2015


7.2.1.1. Email

   The message generated by the Mail Receiver MUST be a [MAIL] message
   formatted per [MIME]. The aggregate report itself MUST be included
   in one of the parts of the message. A human-readable portion MAY be
   included as a MIME part (such as a text/plain part).

   The aggregate data MUST be an XML file that SHOULD be subjected to
   GZIP compression. Declining to apply compression can cause the
   report to be too large for a receiver to process (a commonly observed
   receiver limit is ten megabytes); doing the compression increases the
   chances of acceptance of the report at some compute cost. The
   aggregate data SHOULD be present using the media type "application/
   gzip" if compressed (see [GZIP]), and "text/xml" otherwise. The
   filename is typically constructed using the following ABNF:

     filename = receiver "!" policy-domain "!" begin-timestamp
                "!" end-timestamp [ "!" unique-id ] "." extension

     unique-id = 1*(ALPHA / DIGIT)

     receiver = domain
                ; imported from [MAIL]

     policy-domain = domain

     begin-timestamp = 1*DIGIT
                       ; seconds since 00:00:00 UTC January 1, 1970
                       ; indicating start of the time range contained
                       ; in the report

     end-timestamp = 1*DIGIT
                     ; seconds since 00:00:00 UTC January 1, 1970
                     ; indicating end of the time range contained
                     ; in the report

     extension = "xml" / "xml.gz"

   The extension MUST be "xml" for a plain XML file, or "xml.gz" for an
   XML file compressed using GZIP.

   "unique-id" allows an optional unique ID generated by the Mail
   Receiver to distinguish among multiple reports generated
   simultaneously by different sources within the same Domain Owner.







Kucherawy & Zwicky Informational [Page 33]
 
RFC 7489 DMARC March 2015


   For example, this is a possible filename for the gzip file of a
   report to the Domain Owner "example.com" from the Mail Receiver
   "mail.receiver.example":

     mail.receiver.example!example.com!1013662812!1013749130.gz

   No specific MIME message structure is required. It is presumed that
   the aggregate reporting address will be equipped to extract MIME
   parts with the prescribed media type and filename and ignore the
   rest.

   Email streams carrying DMARC feedback data MUST conform to the DMARC
   mechanism, thereby resulting in an aligned "pass" (see Section 3.1).
   This practice minimizes the risk of report consumers processing
   fraudulent reports.

   The RFC5322.Subject field for individual report submissions SHOULD
   conform to the following ABNF:

     dmarc-subject = %x52.65.70.6f.72.74 1*FWS ; "Report"
                     %x44.6f.6d.61.69.6e.3a 1*FWS ; "Domain:"
                     domain-name 1*FWS ; from RFC 6376
                     %x53.75.62.6d.69.74.74.65.72.3a ; "Submitter:"
                     1*FWS domain-name 1*FWS
                     %x52.65.70.6f.72.74.2d.49.44.3a ; "Report-ID:"
                     msg-id ; from RFC 5322

   The first domain-name indicates the DNS domain name about which the
   report was generated. The second domain-name indicates the DNS
   domain name representing the Mail Receiver generating the report.
   The purpose of the Report-ID: portion of the field is to enable the
   Domain Owner to identify and ignore duplicate reports that might be
   sent by a Mail Receiver.

   For instance, this is a possible Subject field for a report to the
   Domain Owner "example.com" from the Mail Receiver
   "mail.receiver.example". It is line-wrapped as allowed by [MAIL]:

     Subject: Report Domain: example.com
         Submitter: mail.receiver.example
         Report-ID: <2002.02.15.1>

   This transport mechanism potentially encounters a problem when
   feedback data size exceeds maximum allowable attachment sizes for
   either the generator or the consumer. See Section 7.2.2 for further
   discussion.





Kucherawy & Zwicky Informational [Page 34]
 
RFC 7489 DMARC March 2015


7.2.1.2. Other Methods

   The specification as written allows for the addition of other
   registered URI schemes to be supported in later versions.

7.2.2. Error Reports

   When a Mail Receiver is unable to complete delivery of a report via
   any of the URIs listed by the Domain Owner, the Mail Receiver SHOULD
   generate an error message. An attempt MUST be made to send this
   report to all listed "mailto" URIs, and it MAY also be sent to any or
   all other listed URIs.

   The error report MUST be formatted per [MIME]. A text/plain part
   MUST be included that contains field-value pairs such as those found
   in Section 2 of [DSN]. The fields required, which may appear in any
   order, are as follows:

   Report-Date: A [MAIL]-formatted date expression indicating when the
      transport failure occurred.

   Report-Domain: The domain-name about which the failed report was
      generated.

   Report-ID: The Report-ID: that the report tried to use.

   Report-Size: The size, in bytes, of the report that was unable to be
      sent. This MUST represent the number of bytes that the Mail
      Receiver attempted to send. Where more than one transport system
      was attempted, the sizes may be different; in such cases, separate
      error reports MUST be generated so that this value matches the
      actual attempt that was made.

   Submitter: The domain-name representing the Mail Receiver that
      generated, but was unable to submit, the report.

   Submitting-URI: The URI(s) to which the Mail Receiver tried, but
      failed, to submit the report.

   An additional text/plain part MAY be included that gives a human-
   readable explanation of the above and MAY also include a URI that can
   be used to seek assistance.









Kucherawy & Zwicky Informational [Page 35]
 
RFC 7489 DMARC March 2015


7.3. Failure Reports

   Failure reports are normally generated and sent almost immediately
   after the Mail Receiver detects a DMARC failure. Rather than waiting
   for an aggregate report, these reports are useful for quickly
   notifying the Domain Owners when there is an authentication failure.
   Whether the failure is due to an infrastructure problem or the
   message is inauthentic, failure reports also provide more information
   about the failed message than is available in an aggregate report.

   These reports SHOULD include any URI(s) from the message that failed
   authentication. These reports SHOULD include as much of the message
   and message header as is reasonable to support the Domain Owner's
   investigation into what caused the message to fail authentication and
   track down the sender.

   When a Domain Owner requests failure reports for the purpose of
   forensic analysis, and the Mail Receiver is willing to provide such
   reports, the Mail Receiver generates and sends a message using the
   format described in [AFRF]; this document updates that reporting
   format, as described in Section 7.3.1.

   The destination(s) and nature of the reports are defined by the "ruf"
   and "fo" tags as defined in Section 6.3.

   Where multiple URIs are selected to receive failure reports, the
   report generator MUST make an attempt to deliver to each of them.

   An obvious consideration is the denial-of-service attack that can be
   perpetrated by an attacker who sends numerous messages purporting to
   be from the intended victim Domain Owner but that fail both SPF and
   DKIM; this would cause participating Mail Receivers to send failure
   reports to the Domain Owner or its delegate in potentially huge
   volumes. Accordingly, participating Mail Receivers are encouraged to
   aggregate these reports as much as is practical, using the Incidents
   field of the Abuse Reporting Format ([ARF]). Various aggregation
   techniques are possible, including the following:

   o only send a report to the first recipient of multi-recipient
      messages;

   o store reports for a period of time before sending them, allowing
      detection, collection, and reporting of like incidents;

   o apply rate limiting, such as a maximum number of reports per
      minute that will be generated (and the remainder discarded).





Kucherawy & Zwicky Informational [Page 36]
 
RFC 7489 DMARC March 2015


7.3.1. Reporting Format Update

   Operators implementing this specification also implement an augmented
   version of [AFRF] as follows:

   1. A DMARC failure report includes the following ARF header fields,
       with the indicated normative requirement levels:

       * Identity-Alignment (REQUIRED; defined below)

       * Delivery-Result (OPTIONAL)

       * DKIM-Domain, DKIM-Identity, DKIM-Selector (REQUIRED if the
          message was signed by DKIM)

       * DKIM-Canonicalized-Header, DKIM-Canonicalized-Body (OPTIONAL
          if the message was signed by DKIM)

       * SPF-DNS (REQUIRED)

   2. The "Identity-Alignment" field is defined to contain a comma-
       separated list of authentication mechanism names that produced an
       aligned identity, or the keyword "none" if none did. ABNF:

     id-align = "Identity-Alignment:" [CFWS]
                    ( "none" /
                      dmarc-method *( [CFWS] "," [CFWS] dmarc-method ) )
                    [CFWS]

     dmarc-method = ( "dkim" / "spf" )
                    ; each may appear at most once in an id-align

   3. Authentication Failure Type "dmarc" is defined, which is to be
       used when a failure report is generated because some or all of
       the authentication mechanisms failed to produce aligned
       identifiers. Note that a failure report generator MAY also
       independently produce an AFRF message for any or all of the
       underlying authentication methods.

8. Minimum Implementations

   A minimum implementation of DMARC has the following characteristics:

   o Is able to send and/or receive reports at least daily;

   o Is able to send and/or receive reports using "mailto" URIs;





Kucherawy & Zwicky Informational [Page 37]
 
RFC 7489 DMARC March 2015


   o Other than in exceptional circumstances such as resource
      exhaustion, can generate or accept a report up to ten megabytes in
      size;

   o If acting as a Mail Receiver, fully implements the provisions of
      Section 6.6.

9. Privacy Considerations

   This section discusses security issues specific to private data that
   may be included in the interactions that are part of DMARC.

9.1. Data Exposure Considerations

   Aggregate reports are limited in scope to DMARC policy and
   disposition results, to information pertaining to the underlying
   authentication mechanisms, and to the identifiers involved in DMARC
   validation.

   Failed-message reporting provides message-specific details pertaining
   to authentication failures. Individual reports can contain message
   content as well as trace header fields. Domain Owners are able to
   analyze individual reports and attempt to determine root causes of
   authentication mechanism failures, gain insight into
   misconfigurations or other problems with email and network
   infrastructure, or inspect messages for insight into abusive
   practices.

   Both report types may expose sender and recipient identifiers (e.g.,
   RFC5322.From addresses), and although the [AFRF] format used for
   failed-message reporting supports redaction, failed-message reporting
   is capable of exposing the entire message to the report recipient.

   Domain Owners requesting reports will receive information about mail
   claiming to be from them, which includes mail that was not, in fact,
   from them. Information about the final destination of mail where it
   might otherwise be obscured by intermediate systems will therefore be
   exposed.

   When message-forwarding arrangements exist, Domain Owners requesting
   reports will also receive information about mail forwarded to domains
   that were not originally part of their messages' recipient lists.
   This means that destination domains previously unknown to the Domain
   Owner may now become visible.

   Disclosure of information about the messages is being requested by
   the entity generating the email in the first place, i.e., the Domain
   Owner and not the Mail Receiver, so this may not fit squarely within



Kucherawy & Zwicky Informational [Page 38]
 
RFC 7489 DMARC March 2015


   existing privacy policy provisions. For some providers, aggregate
   reporting and failed-message reporting are viewed as a function
   similar to complaint reporting about spamming or phishing and are
   treated similarly under the privacy policy. Report generators (i.e.,
   Mail Receivers) are encouraged to review their reporting limitations
   under such policies before enabling DMARC reporting.

9.2. Report Recipients

   A DMARC record can specify that reports should be sent to an
   intermediary operating on behalf of the Domain Owner. This is done
   when the Domain Owner contracts with an entity to monitor mail
   streams for abuse and performance issues. Receipt by third parties
   of such data may or may not be permitted by the Mail Receiver's
   privacy policy, terms of use, or other similar governing document.
   Domain Owners and Mail Receivers should both review and understand if
   their own internal policies constrain the use and transmission of
   DMARC reporting.

   Some potential exists for report recipients to perform traffic
   analysis, making it possible to obtain metadata about the Receiver's
   traffic. In addition to verifying compliance with policies,
   Receivers need to consider that before sending reports to a third
   party.

10. Other Topics

   This section discusses some topics regarding choices made in the
   development of DMARC, largely to commit the history to record.

10.1. Issues Specific to SPF

   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 large 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.

   Some receiver architectures might implement SPF in advance of any
   DMARC operations. This means that a "-" prefix on a sender's SPF
   mechanism, such as "-all", could cause that rejection to go into
   effect early in handling, causing message rejection before any DMARC
   processing takes place. Operators choosing to use "-all" should be
   aware of this.






Kucherawy & Zwicky Informational [Page 39]
 
RFC 7489 DMARC March 2015


10.2. DNS Load and Caching

   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. Should Domain Owners publish a DNS record with a very
   short TTL, Mail Receivers can be provoked through the injection of
   large volumes of messages to overwhelm the Domain Owner'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.

   Conversely, long TTLs will cause records to be cached for long
   periods of time. This can cause a critical change to DMARC
   parameters advertised by a Domain Owner 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.

10.3. Rejecting Messages

   This proposal calls for rejection of a message during the SMTP
   session under certain circumstances. This is preferable to
   generation of a Delivery Status Notification ([DSN]), since
   fraudulent messages caught and rejected using DMARC would then result
   in annoying generation of such failure reports that go back to the
   RFC5321.MailFrom address.

   This synchronous rejection is typically done in one of two ways:

   o 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 notification that
      delivery failed (see Section 4.2.5 of [SMTP]).

   o A "silent discard", 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.

   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.





Kucherawy & Zwicky Informational [Page 40]
 
RFC 7489 DMARC March 2015


   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.

   In the latter case, when doing an SMTP rejection, providing a clear
   hint can be useful in resolving issues. A receiver might indicate in
   plain text the reason for the rejection by using the word "DMARC"
   somewhere in the reply text. 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. For
   example:

       550 5.7.1 Email rejected per DMARC policy for example.com

   If a Mail Receiver elects to defer delivery due to inability to
   retrieve or apply DMARC policy, this is best done with a 4xy SMTP
   reply code.

10.4. Identifier Alignment Considerations

   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.

   For example, an attacker who controls the SPF record for
   "evil.example.com" can send mail with an RFC5322.From field
   containing "foo@example.com" that can pass both authentication and
   the DMARC check against "example.com".

   The Organizational Domain administrator should be careful not to
   delegate control of subdomains if this is an issue, and to consider
   using the "strict" Identifier Alignment option if appropriate.

10.5. Interoperability Issues

   DMARC limits which end-to-end scenarios can achieve a "pass" result.

   Because DMARC relies on [SPF] and/or [DKIM] to achieve a "pass",
   their limitations also apply.





Kucherawy & Zwicky Informational [Page 41]
 
RFC 7489 DMARC March 2015


   Additional DMARC constraints occur when a message is processed by
   some Mediators, such as mailing lists. Transiting a Mediator often
   causes either the authentication to fail or Identifier Alignment to
   be lost. These transformations may conform to standards but will
   still prevent a DMARC "pass".

   In addition to Mediators, mail that is sent by authorized,
   independent third parties might not be sent with Identifier
   Alignment, also preventing a "pass" result.

   Issues specific to the use of policy mechanisms alongside DKIM are
   further discussed in [DKIM-LISTS], particularly Section 5.2.

11. IANA Considerations

   This section describes actions completed by IANA.

11.1. Authentication-Results Method Registry Update

   IANA has added the following to the "Email Authentication Methods"
   registry:

   Method: dmarc

   Defined: RFC 7489

   ptype: header

   Property: from

   Value: the domain portion of the RFC5322.From field

   Status: active

   Version: 1

11.2. Authentication-Results Result Registry Update

   IANA has added the following in the "Email Authentication Result
   Names" registry:

   Code: none

   Existing/New Code: existing

   Defined: [AUTH-RESULTS]

   Auth Method: dmarc (added)



Kucherawy & Zwicky Informational [Page 42]
 
RFC 7489 DMARC March 2015


   Meaning: No DMARC policy record was published for the aligned
      identifier, or no aligned identifier could be extracted.

   Status: active


   Code: pass

   Existing/New Code: existing

   Defined: [AUTH-RESULTS]

   Auth Method: dmarc (added)

   Meaning: A DMARC policy record was published for the aligned
      identifier, and at least one of the authentication mechanisms
      passed.

   Status: active


   Code: fail

   Existing/New Code: existing

   Defined: [AUTH-RESULTS]

   Auth Method: dmarc (added)

   Meaning: A DMARC policy record was published for the aligned
      identifier, and none of the authentication mechanisms passed.

   Status: active


   Code: temperror

   Existing/New Code: existing

   Defined: [AUTH-RESULTS]

   Auth Method: dmarc (added)

   Meaning: A temporary error occurred during DMARC evaluation. A
      later attempt might produce a final result.

   Status: active




Kucherawy & Zwicky Informational [Page 43]
 
RFC 7489 DMARC March 2015


   Code: permerror

   Existing/New Code: existing

   Defined: [AUTH-RESULTS]

   Auth Method: dmarc (added)

   Meaning: 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.

   Status: active

11.3. Feedback Report Header Fields Registry Update

   The following has been added to the "Feedback Report Header Fields"
   registry:

   Field Name: Identity-Alignment

   Description: indicates whether the message about which a report is
      being generated had any identifiers in alignment as defined in
      RFC 7489

   Multiple Appearances: No

   Related "Feedback-Type": auth-failure

   Reference: RFC 7489

   Status: current

11.4. DMARC Tag Registry

   A new registry tree called "Domain-based Message Authentication,
   Reporting, and Conformance (DMARC) Parameters" has been created.
   Within it, a new sub-registry called the "DMARC Tag Registry" has
   been created.

   Names of DMARC tags must be 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 [IANA-CONSIDERATIONS]. Each registration must include
   the tag name; the specification that defines it; a brief description;
   and its status, which must be one of "current", "experimental", or
   "historic". The Designated Expert needs to confirm that the provided




Kucherawy & Zwicky Informational [Page 44]
 
RFC 7489 DMARC March 2015


   specification adequately describes the new tag and clearly presents
   how it would be used within the DMARC context by Domain Owners and
   Mail Receivers.

   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.

   The initial set of entries in this registry is as follows:

    +----------+-------------+---------+------------------------------+
    | Tag Name | Reference | Status | Description |
    +----------+-------------+---------+------------------------------+
    | adkim | RFC 7489 | current | DKIM alignment mode |
    +----------+-------------+---------+------------------------------+
    | aspf | RFC 7489 | current | SPF alignment mode |
    +----------+-------------+---------+------------------------------+
    | fo | RFC 7489 | current | Failure reporting options |
    +----------+-------------+---------+------------------------------+
    | p | RFC 7489 | current | Requested handling policy |
    +----------+-------------+---------+------------------------------+
    | pct | RFC 7489 | current | Sampling rate |
    +----------+-------------+---------+------------------------------+
    | rf | RFC 7489 | current | Failure reporting format(s) |
    +----------+-------------+---------+------------------------------+
    | ri | RFC 7489 | current | Aggregate Reporting interval |
    +----------+-------------+---------+------------------------------+
    | rua | RFC 7489 | current | Reporting URI(s) for |
    | | | | aggregate data |
    +----------+-------------+---------+------------------------------+
    | ruf | RFC 7489 | current | Reporting URI(s) for |
    | | | | failure data |
    +----------+-------------+---------+------------------------------+
    | sp | RFC 7489 | current | Requested handling policy |
    | | | | for subdomains |
    +----------+-------------+---------+------------------------------+
    | v | RFC 7489 | current | Specification version |
    +----------+-------------+---------+------------------------------+

11.5. DMARC Report Format Registry

   Also within "Domain-based Message Authentication, Reporting, and
   Conformance (DMARC) Parameters", a new sub-registry called "DMARC
   Report Format Registry" has been created.

   Names of DMARC failure reporting formats must be registered with IANA
   in this registry. New entries are assigned only for values that
   satisfy the definition of Specification Required, per



Kucherawy & Zwicky Informational [Page 45]
 
RFC 7489 DMARC March 2015


   [IANA-CONSIDERATIONS]. In addition to a reference to a permanent
   specification, each registration must include the format name; a
   brief description; and its status, which must be one of "current",
   "experimental", or "historic". 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.

   The initial entry in this registry is as follows:

    +--------+-------------+---------+-----------------------------+
    | Format | Reference | Status | Description |
    | Name | | | |
    +--------+-------------+---------+-----------------------------+
    | afrf | RFC 7489 | current | Authentication Failure |
    | | | | Reporting Format (see |
    | | | | [AFRF]) |
    +--------+-------------+---------+-----------------------------+

12. Security Considerations

   This section discusses security issues and possible remediations
   (where available) for DMARC.

12.1. Authentication Methods

   Security considerations from the authentication methods used by DMARC
   are incorporated here by reference.

12.2. Attacks on Reporting URIs

   URIs published in DNS TXT records are well-understood possible
   targets for attack. Specifications such as [DNS] and [ROLES] either
   expose or cause the exposure of email addresses that could be flooded
   by an attacker, for example; MX, NS, and other records found in the
   DNS advertise potential attack destinations; common DNS names such as
   "www" plainly identify the locations at which particular services can
   be found, providing destinations for targeted denial-of-service or
   penetration attacks.

   Thus, Domain Owners will need to harden these addresses against
   various attacks, including but not limited to:

   o high-volume denial-of-service attacks;

   o deliberate construction of malformed reports intended to identify
      or exploit parsing or processing vulnerabilities;




Kucherawy & Zwicky Informational [Page 46]
 
RFC 7489 DMARC March 2015


   o 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.

12.3. DNS Security

   The DMARC mechanism and its underlying technologies (SPF, DKIM)
   depend on the security of the DNS. To reduce the risk of subversion
   of the DMARC mechanism due to DNS-based exploits, serious
   consideration should be given to the deployment of DNSSEC in parallel
   with the deployment of DMARC by both Domain Owners and Mail
   Receivers.

   Publication of data using DNSSEC is relevant to Domain Owners and
   third-party Report Receivers. DNSSEC-aware resolution is relevant to
   Mail Receivers and Report Receivers.

12.4. Display Name Attacks

   A common attack in messaging abuse is the presentation of false
   information in the display-name portion of the RFC5322.From 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.

   Generally, display name attacks are out of scope for DMARC, as
   further exploration of possible defenses against these attacks needs
   to be undertaken.

   There are a few possible mechanisms that attempt mitigation of these
   attacks, such as the following:

   o If the display name is found to include an email address (as
      specified in [MAIL]), execute the DMARC mechanism on the domain
      name found there rather than the domain name discovered
      originally. 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.,

        From: "user@example.org via Bug Tracker" <support@example.com>





Kucherawy & Zwicky Informational [Page 47]
 
RFC 7489 DMARC March 2015


   o 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.

   o In the MUA, only show the display name if the DMARC mechanism
      passes and the email address thus validated matches one found in
      the receiving user's list of known addresses.

12.5. External Reporting Addresses

   To avoid abuse by bad actors, reporting addresses generally have to
   be inside the domains about which reports are requested. In order to
   accommodate special cases such as a need to get reports about domains
   that cannot actually receive mail, Section 7.1 describes a DNS-based
   mechanism for verifying approved external reporting.

   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.

   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.

   Note that the addresses shown in the "ruf" 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 "rua" 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 [DNSSEC] is advisable if this is a concern.

   The verification mechanism presented in Section 7.1 is currently not
   mandatory ("MUST") but strongly recommended ("SHOULD"). It is
   possible that it would be elevated to a "MUST" by later security
   review.

12.6. Secure Protocols

   This document encourages use of secure transport mechanisms to
   prevent loss of private data to third parties that may be able to
   monitor such transmissions. Unencrypted mechanisms should be
   avoided.





Kucherawy & Zwicky Informational [Page 48]
 
RFC 7489 DMARC March 2015


   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.

13. References

13.1. Normative References

   [ABNF] Crocker, D., Ed., and P. Overell, "Augmented BNF for
              Syntax Specifications: ABNF", STD 68, RFC 5234,
              January 2008, <http://www.rfc-editor.org/info/rfc5234>.

   [AFRF] Fontana, H., "Authentication Failure Reporting Using the
              Abuse Reporting Format", RFC 6591, April 2012,
              <http://www.rfc-editor.org/info/rfc6591>.

   [AFRF-DKIM]
              Kucherawy, M., "Extensions to DomainKeys Identified Mail
              (DKIM) for Failure Reporting", RFC 6651, June 2012,
              <http://www.rfc-editor.org/info/rfc6651>.

   [AFRF-SPF] Kitterman, S., "Sender Policy Framework (SPF)
              Authentication Failure Reporting Using the Abuse Reporting
              Format", RFC 6652, June 2012,
              <http://www.rfc-editor.org/info/rfc6652>.

   [DKIM] Crocker, D., Ed., Hansen, T., Ed., and M. Kucherawy, Ed.,
              "DomainKeys Identified Mail (DKIM) Signatures", STD 76,
              RFC 6376, September 2011, <http://www.rfc-editor.org/
              info/rfc6376>.

   [DNS] Mockapetris, P., "Domain names - implementation and
              specification", STD 13, RFC 1035, November 1987,
              <http://www.rfc-editor.org/info/rfc1035>.

   [DNS-CASE] Eastlake 3rd, D., "Domain Name System (DNS) Case
              Insensitivity Clarification", RFC 4343, January 2006,
              <http://www.rfc-editor.org/info/rfc4343>.

   [GZIP] Levine, J., "The 'application/zlib' and 'application/gzip'
              Media Types", RFC 6713, August 2012,
              <http://www.rfc-editor.org/info/rfc6713>.

   [IDNA] Klensin, J., "Internationalized Domain Names for
              Applications (IDNA): Definitions and Document Framework",
              RFC 5890, August 2010,
              <http://www.rfc-editor.org/info/rfc5890>.




Kucherawy & Zwicky Informational [Page 49]
 
RFC 7489 DMARC March 2015


   [KEYWORDS] Bradner, S., "Key words for use in RFCs to Indicate
              Requirement Levels", BCP 14, RFC 2119, March 1997,
              <http://www.rfc-editor.org/info/rfc2119>.

   [MAIL] Resnick, P., Ed., "Internet Message Format", RFC 5322,
              October 2008, <http://www.rfc-editor.org/info/rfc5322>.

   [MIME] Freed, N. and N. Borenstein, "Multipurpose Internet Mail
              Extensions (MIME) Part One: Format of Internet Message
              Bodies", RFC 2045, November 1996,
              <http://www.rfc-editor.org/info/rfc2045>.

   [SEC-TERMS]
              Shirey, R., "Internet Security Glossary, Version 2",
              FYI 36, RFC 4949, August 2007,
              <http://www.rfc-editor.org/info/rfc4949>.

   [SMTP] Klensin, J., "Simple Mail Transfer Protocol", RFC 5321,
              October 2008, <http://www.rfc-editor.org/info/rfc5321>.

   [SPF] Kitterman, S., "Sender Policy Framework (SPF) for
              Authorizing Use of Domains in Email, Version 1", RFC 7208,
              April 2014, <http://www.rfc-editor.org/info/rfc7208>.

   [URI] Berners-Lee, T., Fielding, R., and L. Masinter, "Uniform
              Resource Identifier (URI): Generic Syntax", STD 66,
              RFC 3986, January 2005,
              <http://www.rfc-editor.org/info/rfc3986>.

13.2. Informative References

   [ADSP] Allman, E., Fenton, J., Delany, M., and J. Levine,
              "DomainKeys Identified Mail (DKIM) Author Domain Signing
              Practices (ADSP)", RFC 5617, August 2009,
              <http://www.rfc-editor.org/info/rfc5617>.

   [ARF] Shafranovich, Y., Levine, J., and M. Kucherawy, "An
              Extensible Format for Email Feedback Reports", RFC 5965,
              August 2010, <http://www.rfc-editor.org/info/rfc5965>.

   [AUTH-RESULTS]
              Kucherawy, M., "Message Header Field for Indicating
              Message Authentication Status", RFC 7001, September 2013,
              <http://www.rfc-editor.org/info/rfc7001>.







Kucherawy & Zwicky Informational [Page 50]
 
RFC 7489 DMARC March 2015


   [Best-Guess-SPF]
              Kitterman, S., "Sender Policy Framework: Best guess record
              (FAQ entry)", May 2010,
              <http://www.openspf.org/FAQ/Best_guess_record>.

   [DKIM-DEPLOYMENT]
              Hansen, T., Siegel, E., Hallam-Baker, P., and D. Crocker,
              "DomainKeys Identified Mail (DKIM) Development,
              Deployment, and Operations", RFC 5863, May 2010,
              <http://www.rfc-editor.org/info/rfc5863>.

   [DKIM-LISTS]
              Kucherawy, M., "DomainKeys Identified Mail (DKIM) and
              Mailing Lists", BCP 167, RFC 6377, September 2011,
              <http://www.rfc-editor.org/info/rfc6377>.

   [DKIM-OVERVIEW]
              Hansen, T., Crocker, D., and P. Hallam-Baker, "DomainKeys
              Identified Mail (DKIM) Service Overview", RFC 5585,
              July 2009, <http://www.rfc-editor.org/info/rfc5585>.

   [DKIM-THREATS]
              Fenton, J., "Analysis of Threats Motivating DomainKeys
              Identified Mail (DKIM)", RFC 4686, September 2006,
              <http://www.rfc-editor.org/info/rfc4686>.

   [DNSSEC] Arends, R., Austein, R., Larson, M., Massey, D., and S.
              Rose, "DNS Security Introduction and Requirements",
              RFC 4033, March 2005,
              <http://www.rfc-editor.org/info/rfc4033>.

   [DSN] Moore, K. and G. Vaudreuil, "An Extensible Message Format
              for Delivery Status Notifications", RFC 3464,
              January 2003, <http://www.rfc-editor.org/info/rfc3464>.

   [EMAIL-ARCH]
              Crocker, D., "Internet Mail Architecture", RFC 5598,
              July 2009, <http://www.rfc-editor.org/info/rfc5598>.

   [IANA-CONSIDERATIONS]
              Narten, T. and H. Alvestrand, "Guidelines for Writing an
              IANA Considerations Section in RFCs", BCP 26, RFC 5226,
              May 2008, <http://www.rfc-editor.org/info/rfc5226>.

   [ROLES] Crocker, D., "Mailbox Names for Common Services, Roles and
              Functions", RFC 2142, May 1997,
              <http://www.rfc-editor.org/info/rfc2142>.




Kucherawy & Zwicky Informational [Page 51]
 
RFC 7489 DMARC March 2015


Appendix A. Technology Considerations

   This section documents some design decisions that were made in the
   development of DMARC. Specifically, addressed here are some
   suggestions that were considered but not included in the design.
   This text is included to explain why they were considered and not
   included in this version.

A.1. S/MIME

   S/MIME, or Secure Multipurpose Internet Mail Extensions, is a
   standard for encryption and signing of MIME data in a message. This
   was suggested and considered as a third security protocol for
   authenticating the source of a message.

   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.

   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.

   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 both are deployed widely over the general
   Internet, and their adoption rates continue to be positive.

   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.














Kucherawy & Zwicky Informational [Page 52]
 
RFC 7489 DMARC March 2015


A.2. Method Exclusion

   It was suggested that DMARC include a mechanism by which a Domain
   Owner could tell Message Receivers not to attempt validation by one
   of the supported methods (e.g., "check DKIM, but not SPF").

   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 "none", which would appear
   to exclude participation by that Domain Owner.

   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 specific
   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.

   In the scenario described above, the Domain Owner has a few options:

   1. Tighten up its infrastructure to minimize the failure modes of
       the single deployed technology.

   2. Deploy the other supported authentication mechanism, to offset
       the failure modes of the first.

   3. Deploy DMARC in a reporting-only mode.

A.3. Sender Header Field

   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.

   The DMARC development team considered this and decided not to include
   support for doing so, for the following reasons:

   1. 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 checking of
       the Sender identifier would mean applying policy to an identifier





Kucherawy & Zwicky Informational [Page 53]
 
RFC 7489 DMARC March 2015


       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.

   2. 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.

   3. 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.

A.4. Domain Existence Test

   A common practice among MTA operators, and indeed one documented in
   [ADSP], is a test to determine domain existence prior to any more
   expensive processing. This is typically done by querying the DNS for
   MX, A, or AAAA resource records for the name being evaluated and
   assuming that the domain is nonexistent if it could be determined
   that no such records were published for that domain name.

   The original pre-standardization version of this protocol included a
   mandatory check of this nature. It was ultimately removed, as the
   method's error rate was too high without substantial manual tuning
   and heuristic work. There are indeed use cases this work needs to
   address where such a method would return a negative result about a
   domain for which reporting is desired, such as a registered domain
   name that never sends legitimate mail and thus has none of these
   records present in the DNS.

A.5. Issues with ADSP in Operation

   DMARC has been characterized as a "super-ADSP" of sorts.

   Contributors to DMARC have compiled a list of issues associated with
   ADSP, gained from operational experience, that have influenced the
   direction of DMARC:

   1. 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.








Kucherawy & Zwicky Informational [Page 54]
 
RFC 7489 DMARC March 2015


   2. Nonexistent subdomains are explicitly out of scope in ADSP.
       There is nothing in ADSP that states 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).

   3. ADSP has no operational advice on when to look up the ADSP
       record.

   4. ADSP has no support for using SPF as an auxiliary mechanism to
       DKIM.

   5. ADSP has no support for a slow rollout, i.e., no way to configure
       a percentage of email on which the receiver should apply the
       policy. This is important for large-volume senders.

   6. ADSP has no explicit support for an intermediate phase where the
       receiver quarantines (e.g., sends to the recipient's "spam"
       folder) rather than rejects the email.

   7. The binding between the "From" header domain and DKIM is too
       tight for ADSP; they must match exactly.

A.6. Organizational Domain Discovery Issues

   Although protocols like ADSP are useful for "protecting" a specific
   domain name, they are not helpful at protecting subdomains. If one
   wished to protect "example.com" by requiring via ADSP that all mail
   bearing an RFC5322.From domain of "example.com" be signed, this would
   "protect" that domain; however, one could then craft an email whose
   RFC5322.From domain is "security.example.com", and ADSP would not
   provide any protection. One could use a DNS wildcard, but this can
   undesirably interfere with other DNS activity; one could add ADSP
   records as fraudulent domains are discovered, but this solution does
   not scale and is a purely reactive measure against abuse.

   The DNS does not provide a method by which the "domain of record", or
   the domain that was actually registered with a domain registrar, can
   be determined given an arbitrary domain name. Suggestions 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.

   When seeking domain-specific policy based on an arbitrary domain
   name, one could "climb the tree", dropping labels off the left end of
   the name until the root is reached or a policy is discovered, but
   then one could craft a name that has a large number of nonsense




Kucherawy & Zwicky Informational [Page 55]
 
RFC 7489 DMARC March 2015


   labels; this would cause a Mail Receiver to attempt a large number of
   queries in search of a policy record. Sending many such messages
   constitutes an amplified denial-of-service attack.

   The Organizational Domain mechanism is a necessary component to the
   goals of DMARC. The method described in Section 3.2 is far from
   perfect but serves this purpose reasonably well without adding undue
   burden or semantics to the DNS. If a method is created to do so that
   is more reliable and secure than the use of a public suffix list,
   DMARC should be amended to use that method as soon as it is generally
   available.

A.6.1. Public Suffix Lists

   A public suffix list for the purposes of determining the
   Organizational Domain can be obtained from various sources. The most
   common one is maintained by the Mozilla Foundation and made public at
   <http://publicsuffix.org>. License terms governing the use of that
   list are available at that URI.

   Note that if operators use a variety of public suffix lists,
   interoperability will be difficult or impossible to guarantee.

Appendix B. Examples

   This section illustrates both the Domain Owner side and the Mail
   Receiver side of a DMARC exchange.

B.1. Identifier Alignment Examples

   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.

B.1.1. SPF

   The following SPF examples assume that SPF produces a passing result.

   Example 1: SPF in alignment:

        MAIL FROM: <sender@example.com>

        From: sender@example.com
        Date: Fri, Feb 15 2002 16:54:30 -0800
        To: receiver@example.org
        Subject: here's a sample




Kucherawy & Zwicky Informational [Page 56]
 
RFC 7489 DMARC March 2015


   In this case, the RFC5321.MailFrom parameter and the RFC5322.From
   field have identical DNS domains. Thus, the identifiers are in
   alignment.

   Example 2: SPF in alignment (parent):

        MAIL FROM: <sender@child.example.com>

        From: sender@example.com
        Date: Fri, Feb 15 2002 16:54:30 -0800
        To: receiver@example.org
        Subject: here's a sample

   In this case, the RFC5322.From parameter includes a DNS domain that
   is a parent of the RFC5321.MailFrom domain. Thus, the identifiers
   are in alignment if relaxed SPF mode is requested by the Domain
   Owner, and not in alignment if strict SPF mode is requested.

   Example 3: SPF not in alignment:

        MAIL FROM: <sender@example.net>

        From: sender@child.example.com
        Date: Fri, Feb 15 2002 16:54:30 -0800
        To: receiver@example.org
        Subject: here's a sample

   In this case, the RFC5321.MailFrom parameter includes a DNS domain
   that is neither the same as nor a parent of the RFC5322.From domain.
   Thus, the identifiers are not in alignment.

B.1.2. DKIM

   The examples below assume that the DKIM signatures pass verification.
   Alignment cannot exist with a DKIM signature that does not verify.

   Example 1: DKIM in alignment:

        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

   In this case, the DKIM "d=" parameter and the RFC5322.From field have
   identical DNS domains. Thus, the identifiers are in alignment.





Kucherawy & Zwicky Informational [Page 57]
 
RFC 7489 DMARC March 2015


   Example 2: DKIM in alignment (parent):

        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

   In this case, the DKIM signature's "d=" parameter includes a DNS
   domain that is a parent of the RFC5322.From domain. Thus, the
   identifiers are in alignment for relaxed mode, but not for strict
   mode.

   Example 3: DKIM not in alignment:

        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

   In this case, the DKIM signature's "d=" parameter includes a DNS
   domain that is neither the same as nor a parent of the RFC5322.From
   domain. Thus, the identifiers are not in alignment.

B.2. Domain Owner Example

   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.

B.2.1. Entire Domain, Monitoring Only

   The owner of the domain "example.com" 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 receivers
   without affecting how the messages are processed, in order to:

   o Confirm that its legitimate messages are authenticating correctly

   o Verify that all authorized message sources have implemented
      authentication measures

   o Determine how many messages from other sources would be affected
      by a blocking policy





Kucherawy & Zwicky Informational [Page 58]
 
RFC 7489 DMARC March 2015


   The Domain Owner accomplishes this by constructing a policy record
   indicating that:

   o The version of DMARC being used is "DMARC1" ("v=DMARC1")

   o Receivers should not alter how they treat these messages because
      of this DMARC policy record ("p=none")

   o Aggregate feedback reports should be sent via email to the address
      "dmarc-feedback@example.com"
      ("rua=mailto:dmarc-feedback@example.com")

   o All messages from this Organizational Domain are subject to this
      policy (no "pct" tag present, so the default of 100% applies)

   The DMARC policy record might look like this when retrieved using a
   common command-line tool:

     % dig +short TXT _dmarc.example.com.
     "v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com"

   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):

     ; DMARC record for the domain example.com

     _dmarc IN TXT ( "v=DMARC1; p=none; "
                      "rua=mailto:dmarc-feedback@example.com" )

B.2.2. Entire Domain, Monitoring Only, Per-Message Reports

   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. In order to diagnose these intermittent
   problems, they wish to request per-message failure reports when
   authentication failures occur.

   Not all 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 ([AFRF]) meets the Domain Owner's needs in this scenario.








Kucherawy & Zwicky Informational [Page 59]
 
RFC 7489 DMARC March 2015


   The Domain Owner accomplishes this by adding the following to its
   policy record from Appendix B.2:

   o Per-message failure reports should be sent via email to the
      address "auth-reports@example.com"
      ("ruf=mailto:auth-reports@example.com")

   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):

     % dig +short TXT _dmarc.example.com.
     "v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com;
      ruf=mailto:auth-reports@example.com"

   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):

    ; DMARC record for the domain example.com

    _dmarc IN TXT ( "v=DMARC1; p=none; "
                     "rua=mailto:dmarc-feedback@example.com; "
                     "ruf=mailto:auth-reports@example.com" )

B.2.3. Per-Message Failure Reports Directed to Third Party

   The Domain Owner from the previous example is maintaining the same
   policy but now wishes to have a third party receive and process the
   per-message failure reports. Again, not all Receivers will honor
   this request, but those that do may implement additional checks to
   validate that the third party wishes to receive the failure reports
   for this domain.

   The Domain Owner needs to alter its policy record from Appendix B.2.2
   as follows:

   o Per-message failure reports should be sent via email to the
      address "auth-reports@thirdparty.example.net"
      ("ruf=mailto:auth-reports@thirdparty.example.net")

   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):

     % dig +short TXT _dmarc.example.com.
     "v=DMARC1; p=none; rua=mailto:dmarc-feedback@example.com;
      ruf=mailto:auth-reports@thirdparty.example.net"



Kucherawy & Zwicky Informational [Page 60]
 
RFC 7489 DMARC March 2015


   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):

     ; DMARC record for the domain example.com

     _dmarc IN TXT ( "v=DMARC1; p=none; "
                     "rua=mailto:dmarc-feedback@example.com; "
                     "ruf=mailto:auth-reports@thirdparty.example.net" )

   Because the address used in the "ruf" tag is outside the
   Organizational Domain in which this record is published, conforming
   Receivers will implement additional checks as described in
   Section 7.1 of this document. In order to pass these additional
   checks, the third party will need to publish an additional DNS record
   as follows:

   o Given the DMARC record published by the Domain Owner at
      "_dmarc.example.com", the DNS administrator for the third party
      will need to publish a TXT resource record at
      "example.com._report._dmarc.thirdparty.example.net" with the value
      "v=DMARC1".

   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):

     % dig +short TXT example.com._report._dmarc.thirdparty.example.net
     "v=DMARC1"

   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):

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

     example.com._report._dmarc IN TXT "v=DMARC1"

   Intermediaries and other third parties should refer to Section 7.1
   for the full details of this mechanism.

B.2.4. Subdomain, Sampling, and Multiple Aggregate Report URIs

   The Domain Owner has implemented SPF and DKIM in a subdomain used for
   pre-production testing of messaging services. It now wishes to
   request that participating receivers act to reject messages from this
   subdomain that fail to authenticate.



Kucherawy & Zwicky Informational [Page 61]
 
RFC 7489 DMARC March 2015


   As a first step, it will ask that a portion (1/4 in this example) of
   failing messages be quarantined, enabling examination of messages
   sent to mailboxes hosted by participating receivers. Aggregate
   feedback reports will be sent to a mailbox within the Organizational
   Domain, and to a mailbox at a third party selected and authorized to
   receive same by the Domain Owner. Aggregate reports sent to the
   third party are limited to a maximum size of ten megabytes.

   The Domain Owner will accomplish this by constructing a policy record
   indicating that:

   o The version of DMARC being used is "DMARC1" ("v=DMARC1")

   o It is applied only to this subdomain (record is published at
      "_dmarc.test.example.com" and not "_dmarc.example.com")

   o Receivers should quarantine messages from this Organizational
      Domain that fail to authenticate ("p=quarantine")

   o Aggregate feedback reports should be sent via email to the
      addresses "dmarc-feedback@example.com" and
      "example-tld-test@thirdparty.example.net", with the latter
      subjected to a maximum size limit ("rua=mailto:dmarc-feedback@
      example.com,mailto:tld-test@thirdparty.example.net!10m")

   o 25% of messages from this Organizational Domain are subject to
      action based on this policy ("pct=25")

   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):

     % dig +short TXT _dmarc.test.example.com
     "v=DMARC1; p=quarantine; rua=mailto:dmarc-feedback@example.com,
      mailto:tld-test@thirdparty.example.net!10m; pct=25"

   To publish such a record, the DNS administrator for the Domain Owner
   might create an entry like the following in the appropriate zone
   file:

     ; DMARC record for the domain example.com

     _dmarc IN TXT ( "v=DMARC1; p=quarantine; "
                       "rua=mailto:dmarc-feedback@example.com,"
                       "mailto:tld-test@thirdparty.example.net!10m; "
                       "pct=25" )





Kucherawy & Zwicky Informational [Page 62]
 
RFC 7489 DMARC March 2015


B.3. Mail Receiver Example

   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 Receivers).

B.3.1. Processing of SMTP Time

   An optimal DMARC-enabled Mail Receiver performs authentication and
   Identifier Alignment checking during the [SMTP] conversation.

   Prior to returning a final reply to the DATA command, the Mail
   Receiver's MTA has performed:

   1. An SPF check to determine an SPF-authenticated Identifier.

   2. DKIM checks that yield one or more DKIM-authenticated
       Identifiers.

   3. A DMARC policy lookup.

   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.

   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).

   For example, the following sample data is considered to be from a
   piece of email originating from the Domain Owner of "example.com":

     Author Domain: example.com
     SPF-authenticated Identifier: mail.example.com
     DKIM-authenticated Identifier: example.com
     DMARC record:
       "v=DMARC1; p=reject; aspf=r;
        rua=mailto:dmarc-feedback@example.com"

   In the above sample, both the SPF-authenticated Identifier and the
   DKIM-authenticated Identifier align with the Author Domain. The Mail
   Receiver considers the above email to pass the DMARC check, avoiding
   the "reject" policy that is to be applied to email that fails to pass
   the DMARC check.





Kucherawy & Zwicky Informational [Page 63]
 
RFC 7489 DMARC March 2015


   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 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.

   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 on with
   email processing, perhaps using the result of the DMARC check as an
   input to additional processing modules such as a domain reputation
   query.

   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.

   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 not necessary if the Domain Owner
   has not requested aggregate reports, i.e., no "rua" tag was found in
   the policy record.

B.4. Utilization of Aggregate Feedback: Example

   Aggregate feedback is consumed by Domain Owners to verify a Domain
   Owner's understanding of how the Domain Owner's 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 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.









Kucherawy & Zwicky Informational [Page 64]
 
RFC 7489 DMARC March 2015


   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.

   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. Additionally, the Domain Owner may
   come to an understanding of how its domain is being misused.

B.5. mailto Transport Example

   A DMARC record can contain a "mailto" reporting address, such as:

     mailto:dmarc-feedback@example.com

   A sample aggregate report from the Mail Receiver at
   mail.receiver.example follows:

     DKIM-Signature: v=1; ...; d=mail.receiver.example; ...
     From: dmarc-reporting@mail.receiver.example
     Date: Fri, Feb 15 2002 16:54:30 -0800
     To: dmarc-feedback@example.com
     Subject: Report Domain: example.com
         Submitter: mail.receiver.example
         Report-ID: <2002.02.15.1>
     MIME-Version: 1.0
     Content-Type: multipart/alternative;
         boundary="----=_NextPart_000_024E_01CC9B0A.AFE54C00"
     Content-Language: en-us

     This is a multipart message in MIME format.

     ------=_NextPart_000_024E_01CC9B0A.AFE54C00
     Content-Type: text/plain; charset="us-ascii"
     Content-Transfer-Encoding: 7bit







Kucherawy & Zwicky Informational [Page 65]
 
RFC 7489 DMARC March 2015


     This is an aggregate report from mail.receiver.example.

     ------=_NextPart_000_024E_01CC9B0A.AFE54C00
     Content-Type: application/gzip
     Content-Transfer-Encoding: base64
     Content-Disposition: attachment;
         filename="mail.receiver.example!example.com!
                   1013662812!1013749130.gz"

     <gzipped content of report>

     ------=_NextPart_000_024E_01CC9B0A.AFE54C00--

   Not shown in the above example is that the Mail Receiver's feedback
   should be authenticated using SPF. Also, the value of the "filename"
   MIME parameter is wrapped for printing in this specification but
   would normally appear as one continuous string.

Appendix C. DMARC XML Schema

   The following is the proposed initial schema for producing
   XML-formatted aggregate reports as described in this document.

   NOTE: Per the definition of XML, unless otherwise specified in the
   schema below, the minOccurs and maxOccurs values for each element are
   set to 1.

   <?xml version="1.0"?>
   <xs:schema xmlns:xs="http://www.w3.org/2001/XMLSchema"
     targetNamespace="http://dmarc.org/dmarc-xml/0.1">

   <!-- The time range in UTC covered by messages in this report,
        specified in seconds since epoch. -->
   <xs:complexType name="DateRangeType">
     <xs:all>
       <xs:element name="begin" type="xs:integer"/>
       <xs:element name="end" type="xs:integer"/>
     </xs:all>
   </xs:complexType>

   <!-- Report generator metadata. -->
   <xs:complexType name="ReportMetadataType">
     <xs:sequence>
       <xs:element name="org_name" type="xs:string"/>
       <xs:element name="email" type="xs:string"/>
       <xs:element name="extra_contact_info" type="xs:string"
                   minOccurs="0"/>
       <xs:element name="report_id" type="xs:string"/>



Kucherawy & Zwicky Informational [Page 66]
 
RFC 7489 DMARC March 2015


       <xs:element name="date_range" type="DateRangeType"/>
       <xs:element name="error" type="xs:string" minOccurs="0"
                   maxOccurs="unbounded"/>
     </xs:sequence>
   </xs:complexType>

   <!-- Alignment mode (relaxed or strict) for DKIM and SPF. -->
   <xs:simpleType name="AlignmentType">
     <xs:restriction base="xs:string">
       <xs:enumeration value="r"/>
       <xs:enumeration value="s"/>
     </xs:restriction>
   </xs:simpleType>

   <!-- The policy actions specified by p and sp in the
        DMARC record. -->
   <xs:simpleType name="DispositionType">
     <xs:restriction base="xs:string">
       <xs:enumeration value="none"/>
       <xs:enumeration value="quarantine"/>
       <xs:enumeration value="reject"/>
     </xs:restriction>
   </xs:simpleType>

   <!-- The DMARC policy that applied to the messages in
        this report. -->
   <xs:complexType name="PolicyPublishedType">
     <xs:all>
       <!-- The domain at which the DMARC record was found. -->
       <xs:element name="domain" type="xs:string"/>
       <!-- The DKIM alignment mode. -->
       <xs:element name="adkim" type="AlignmentType"
                   minOccurs="0"/>
       <!-- The SPF alignment mode. -->
       <xs:element name="aspf" type="AlignmentType"
                   minOccurs="0"/>
       <!-- The policy to apply to messages from the domain. -->
       <xs:element name="p" type="DispositionType"/>
       <!-- The policy to apply to messages from subdomains. -->
       <xs:element name="sp" type="DispositionType"/>
       <!-- The percent of messages to which policy applies. -->
       <xs:element name="pct" type="xs:integer"/>
       <!-- Failure reporting options in effect. -->
       <xs:element name="fo" type="xs:string"/>
     </xs:all>
   </xs:complexType>





Kucherawy & Zwicky Informational [Page 67]
 
RFC 7489 DMARC March 2015


   <!-- The DMARC-aligned authentication result. -->
   <xs:simpleType name="DMARCResultType">
     <xs:restriction base="xs:string">
       <xs:enumeration value="pass"/>
       <xs:enumeration value="fail"/>
     </xs:restriction>
   </xs:simpleType>

   <!-- Reasons that may affect DMARC disposition or execution
        thereof. -->
   <xs:simpleType name="PolicyOverrideType">
     <xs:restriction base="xs:string">
       <xs:enumeration value="forwarded"/>
       <xs:enumeration value="sampled_out"/>
       <xs:enumeration value="trusted_forwarder"/>
       <xs:enumeration value="mailing_list"/>
       <xs:enumeration value="local_policy"/>
       <xs:enumeration value="other"/>
     </xs:restriction>
   </xs:simpleType>

   <!-- How do we allow report generators to include new
        classes of override reasons if they want to be more
        specific than "other"? -->
   <xs:complexType name="PolicyOverrideReason">
     <xs:all>
       <xs:element name="type" type="PolicyOverrideType"/>
       <xs:element name="comment" type="xs:string"
                   minOccurs="0"/>
     </xs:all>
   </xs:complexType>

   <!-- Taking into account everything else in the record,
        the results of applying DMARC. -->
   <xs:complexType name="PolicyEvaluatedType">
     <xs:sequence>
       <xs:element name="disposition" type="DispositionType"/>
       <xs:element name="dkim" type="DMARCResultType"/>
       <xs:element name="spf" type="DMARCResultType"/>
       <xs:element name="reason" type="PolicyOverrideReason"
                   minOccurs="0" maxOccurs="unbounded"/>
     </xs:sequence>
   </xs:complexType>








Kucherawy & Zwicky Informational [Page 68]
 
RFC 7489 DMARC March 2015


   <!-- Credit to Roger L. Costello for IPv4 regex
        http://mailman.ic.ac.uk/pipermail/xml-dev/1999-December/
             018018.html -->
   <!-- Credit to java2s.com for IPv6 regex
        http://www.java2s.com/Code/XML/XML-Schema/
             IPv6addressesareeasiertodescribeusingasimpleregex.htm -->
   <xs:simpleType name="IPAddress">
     <xs:restriction base="xs:string">
       <xs:pattern value="((1?[0-9]?[0-9]|2[0-4][0-9]|25[0-5]).){3}
                   (1?[0-9]?[0-9]|2[0-4][0-9]|25[0-5])|
                   ([A-Fa-f0-9]{1,4}:){7}[A-Fa-f0-9]{1,4}"/>
     </xs:restriction>
   </xs:simpleType>

   <xs:complexType name="RowType">
     <xs:all>
       <!-- The connecting IP. -->
       <xs:element name="source_ip" type="IPAddress"/>
       <!-- The number of matching messages. -->
       <xs:element name="count" type="xs:integer"/>
       <!-- The DMARC disposition applying to matching
            messages. -->
       <xs:element name="policy_evaluated"
                   type="PolicyEvaluatedType"
                   minOccurs="1"/>
     </xs:all>
   </xs:complexType>

   <xs:complexType name="IdentifierType">
     <xs:all>
       <!-- The envelope recipient domain. -->
       <xs:element name="envelope_to" type="xs:string"
                   minOccurs="0"/>
       <!-- The RFC5321.MailFrom domain. -->
       <xs:element name="envelope_from" type="xs:string"
                   minOccurs="1"/>
       <!-- The RFC5322.From domain. -->
       <xs:element name="header_from" type="xs:string"
                   minOccurs="1"/>
     </xs:all>
   </xs:complexType>

   <!-- DKIM verification result, according to RFC 7001
        Section 2.6.1. -->
   <xs:simpleType name="DKIMResultType">
     <xs:restriction base="xs:string">
       <xs:enumeration value="none"/>
       <xs:enumeration value="pass"/>



Kucherawy & Zwicky Informational [Page 69]
 
RFC 7489 DMARC March 2015


       <xs:enumeration value="fail"/>
       <xs:enumeration value="policy"/>
       <xs:enumeration value="neutral"/>
       <xs:enumeration value="temperror"/>
       <xs:enumeration value="permerror"/>
     </xs:restriction>
   </xs:simpleType>

   <xs:complexType name="DKIMAuthResultType">
     <xs:all>
       <!-- The "d=" parameter in the signature. -->
       <xs:element name="domain" type="xs:string"
                   minOccurs="1"/>
       <!-- The "s=" parameter in the signature. -->
       <xs:element name="selector" type="xs:string"
                   minOccurs="0"/>
       <!-- The DKIM verification result. -->
       <xs:element name="result" type="DKIMResultType"
                   minOccurs="1"/>
       <!-- Any extra information (e.g., from
            Authentication-Results). -->
       <xs:element name="human_result" type="xs:string"
                   minOccurs="0"/>
     </xs:all>
   </xs:complexType>

   <!-- SPF domain scope. -->
   <xs:simpleType name="SPFDomainScope">
     <xs:restriction base="xs:string">
       <xs:enumeration value="helo"/>
       <xs:enumeration value="mfrom"/>
     </xs:restriction>
   </xs:simpleType>

   <!-- SPF result. -->
   <xs:simpleType name="SPFResultType">
     <xs:restriction base="xs:string">
       <xs:enumeration value="none"/>
       <xs:enumeration value="neutral"/>
       <xs:enumeration value="pass"/>
       <xs:enumeration value="fail"/>
       <xs:enumeration value="softfail"/>
       <!-- "TempError" commonly implemented as "unknown". -->
       <xs:enumeration value="temperror"/>
       <!-- "PermError" commonly implemented as "error". -->
       <xs:enumeration value="permerror"/>
     </xs:restriction>
   </xs:simpleType>



Kucherawy & Zwicky Informational [Page 70]
 
RFC 7489 DMARC March 2015


   <xs:complexType name="SPFAuthResultType">
     <xs:all>
       <!-- The checked domain. -->
       <xs:element name="domain" type="xs:string" minOccurs="1"/>
       <!-- The scope of the checked domain. -->
       <xs:element name="scope" type="SPFDomainScope" minOccurs="1"/>
       <!-- The SPF verification result. -->
       <xs:element name="result" type="SPFResultType"
                   minOccurs="1"/>
     </xs:all>
   </xs:complexType>

   <!-- This element contains DKIM and SPF results, uninterpreted
        with respect to DMARC. -->
   <xs:complexType name="AuthResultType">
     <xs:sequence>
       <!-- There may be no DKIM signatures, or multiple DKIM
            signatures. -->
       <xs:element name="dkim" type="DKIMAuthResultType"
         minOccurs="0" maxOccurs="unbounded"/>
       <!-- There will always be at least one SPF result. -->
       <xs:element name="spf" type="SPFAuthResultType" minOccurs="1"
                   maxOccurs="unbounded"/>
     </xs:sequence>
   </xs:complexType>

   <!-- This element contains all the authentication results that
        were evaluated by the receiving system for the given set of
        messages. -->
   <xs:complexType name="RecordType">
     <xs:sequence>
       <xs:element name="row" type="RowType"/>
       <xs:element name="identifiers" type="IdentifierType"/>
       <xs:element name="auth_results" type="AuthResultType"/>
     </xs:sequence>
   </xs:complexType>

   <!-- Parent -->
   <xs:element name="feedback">
     <xs:complexType>
       <xs:sequence>
         <xs:element name="version"
                     type="xs:decimal"/>
         <xs:element name="report_metadata"
                     type="ReportMetadataType"/>
         <xs:element name="policy_published"
                     type="PolicyPublishedType"/>




Kucherawy & Zwicky Informational [Page 71]
 
RFC 7489 DMARC March 2015


         <xs:element name="record" type="RecordType"
                     maxOccurs="unbounded"/>
       </xs:sequence>
     </xs:complexType>
   </xs:element>
   </xs:schema>

   Descriptions of the PolicyOverrideTypes:

   forwarded: The message was relayed via a known forwarder, or local
      heuristics identified the message as likely having been forwarded.
      There is no expectation that authentication would pass.

   local_policy: The Mail Receiver's local policy exempted the message
      from being subjected to the Domain Owner's requested policy
      action.

   mailing_list: Local heuristics determined that the message arrived
      via a mailing list, and thus authentication of the original
      message was not expected to succeed.

   other: Some policy exception not covered by the other entries in
      this list occurred. Additional detail can be found in the
      PolicyOverrideReason's "comment" field.

   sampled_out: The message was exempted from application of policy by
      the "pct" setting in the DMARC policy record.

   trusted_forwarder: Message authentication failure was anticipated by
      other evidence linking the message to a locally maintained list of
      known and trusted forwarders.

   The "version" for reports generated per this specification MUST be
   the value 1.0.

















Kucherawy & Zwicky Informational [Page 72]
 
RFC 7489 DMARC March 2015


Acknowledgements

   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 <http://dmarc.org>).
   Participating companies included Agari, American Greetings, AOL, Bank
   of America, Cloudmark, Comcast, Facebook, Fidelity Investments,
   Google, JPMorgan Chase & 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.

   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.

Authors' Addresses

   Murray S. Kucherawy (editor)

   EMail: superuser@gmail.com


   Elizabeth Zwicky (editor)
   Yahoo!

   EMail: zwicky@yahoo-inc.com



















Kucherawy & Zwicky Informational [Page 73]