DKIM overview: DomainKeys Identified Mail (DKIM) Service Overview

Network Working Group                                          T. Hansen
Request for Comments: 5585                             AT&T Laboratories
Category: Informational                                       D. Crocker
                                             Brandenburg InternetWorking
                                                         P. Hallam-Baker
                                             Default Deny Security, Inc.
                                                               July 2009


           DomainKeys Identified Mail (DKIM) Service Overview

Abstract

   This document provides an overview of the DomainKeys Identified Mail
   (DKIM) service and describes how it can fit into a messaging service.
   It also describes how DKIM relates to other IETF message signature
   technologies.  It is intended for those who are adopting, developing,
   or deploying DKIM.  DKIM allows an organization to take
   responsibility for transmitting a message, in a way that can be
   verified by a recipient.  The organization can be the author's, the
   originating sending site, an intermediary, or one of their agents.  A
   message can contain multiple signatures from the same or different
   organizations involved with the message.  DKIM defines a domain-level
   digital signature authentication framework for email, using public-
   key cryptography, with the domain name service as its key server
   technology (RFC 4871).  This permits verification of a responsible
   organization, as well as the integrity of the message contents.  DKIM
   also enables a mechanism that permits potential email signers to
   publish information about their email signing practices; this will
   permit email receivers to make additional assessments about messages.
   DKIM's authentication of email identity can assist in the global
   control of "spam" and "phishing".

Status of This Memo

   This memo provides information for the Internet community.  It does
   not specify an Internet standard of any kind.  Distribution of this
   memo is unlimited.

Copyright Notice

   Copyright (c) 2009 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 in effect on the date of
   publication of this document (http://trustee.ietf.org/license-info).
   Please review these documents carefully, as they describe your rights
   and restrictions with respect to this document.

Table of Contents

   1. Introduction ....................................................3
      1.1. DKIM's Scope ...............................................4
      1.2. Prior Work .................................................5
      1.3. Internet Mail Background ...................................6
   2. The DKIM Value Proposition ......................................6
      2.1. Identity Verification ......................................7
      2.2. Enabling Trust Assessments .................................7
      2.3. Establishing Message Validity ..............................8
   3. DKIM Goals ......................................................8
      3.1. Functional Goals ...........................................9
      3.2. Operational Goals .........................................10
   4. DKIM Function ..................................................12
      4.1. Basic Signing .............................................12
      4.2. Characteristics of a DKIM Signature .......................12
      4.3. The Selector Construct ....................................13
      4.4. Verification ..............................................13
      4.5. Sub-Domain Assessment .....................................13
   5. Service Architecture ...........................................14
      5.1. Administration and Maintenance ............................15
      5.2. Signing ...................................................16
      5.3. Verifying .................................................16
      5.4. Unverified or Unsigned Mail ...............................16
      5.5. Assessing .................................................17
      5.6. DKIM Processing within an ADMD ............................17
   6. Considerations .................................................17
      6.1. Security Considerations ...................................17
      6.2. Acknowledgements ..........................................17
   7. Informative References .........................................18
   Appendix A.  Internet Mail Background .............................20
     A.1.  Core Model ................................................20
     A.2.  Trust Boundaries ..........................................20
   Index .............................................................22

1.  Introduction

   This document provides a description of the architecture and
   functionality for DomainKeys Identified Mail (DKIM), that is, the
   core mechanism for signing and verifying messages.  It is intended
   for those who are adopting, developing, or deploying DKIM.  It will
   also be helpful for those who are considering extending DKIM, either
   into other areas of use or to support additional features.  This
   overview does not provide information on threats to DKIM or email or
   details on the protocol specifics, which can be found in [RFC4686]
   and [RFC4871], respectively.  Because the scope of this overview is
   restricted to the technical details of signing and verifying using
   DKIM, it does not explore operational issues, the details of services
   that DKIM uses, or those that, in turn, use DKIM.  Nor does it
   discuss services that build upon DKIM for enforcement of policies or
   assessments.  The document assumes a background in basic email and
   network security technology and services.

   DKIM allows an organization to take responsibility for a message in a
   way that can be verified by a recipient.  The organization can be a
   direct handler of the message, such as the author's, the originating
   sending site's, or an intermediary's along the transit path.
   However, it can also be an indirect handler, such as an independent
   service that is providing assistance to a direct handler.  DKIM
   defines a domain-level digital signature authentication framework for
   email through the use of public-key cryptography and using the domain
   name service as its key server technology [RFC4871].  It permits
   verification of the signer of a message, as well as the integrity of
   its contents.  DKIM will also provide a mechanism that permits
   potential email signers to publish information about their email
   signing practices; this will permit email receivers to make
   additional assessments of unsigned messages.  DKIM's authentication
   of email identity can assist in the global control of "spam" and
   "phishing".

   Neither this document nor DKIM attempts to provide solutions to the
   world's problems with spam, phishing, viruses, worms, joe jobs, etc.
   DKIM provides one basic tool, in what needs to be a large arsenal,
   for improving basic trust in the Internet mail service.  However, by
   itself, DKIM is not sufficient to that task and this overview does
   not pursue the issues of integrating DKIM into these larger efforts,
   beyond a simple reference within a system diagram.  Rather, it is a
   basic introduction to the technology and its use.

1.1.  DKIM's Scope

   A person or organization has an "identity" -- that is, a
   constellation of characteristics that distinguish them from any other
   identity.  Associated with this abstraction can be a label used as a
   reference, or "identifier".  This is the distinction between a thing
   and the name of the thing.  DKIM uses a domain name as an identifier,
   to refer to the identity of a responsible person or organization.  In
   DKIM, this identifier is called the Signing Domain IDentifier (SDID)
   and is contained in the DKIM-Signature header fields "d=" tag.  Note
   that the same identity can have multiple identifiers.

   A DKIM signature can be created by a direct handler of a message,
   such as the message's author or by an intermediary.  A signature also
   can be created by an independent service that is providing assistance
   to a handler of the message.  Whoever does the signing chooses the
   SDID to be used as the basis for later assessments.  Hence, the
   reputation associated with that domain name might be an additional
   basis for evaluating whether to trust the message for delivery.  The
   owner of the SDID is declaring that they accept responsibility for
   the message and can thus be held accountable for it.

   DKIM is intended as a value-added feature for email.  Mail that is
   not signed by DKIM is handled in the same way as it was before DKIM
   was defined.  The message will be evaluated by established analysis
   and filtering techniques.  (A signing policy can provide additional
   information for that analysis and filtering.)  Over time, widespread
   DKIM adoption could permit stricter handling of messages that are not
   signed.  However, early benefits do not require this and probably do
   not warrant this.

   DKIM has a narrow scope.  It is an enabling technology, intended for
   use in the larger context of determining message legitimacy.  This
   larger context is complex, so it is easy to assume that a component
   like DKIM, which actually provides only a limited service, instead
   satisfies the broader set of requirements.

   By itself, a DKIM signature:

   o  Does not authenticate or verify the contents of the message header
      or body, such as the author From field, beyond certifying data
      integrity between the time of signing and the time of verifying.

   o  Does not offer any assertions about the behaviors of the signer.

   o  Does not prescribe any specific actions for receivers to take upon
      successful signature verification.

   o  Does not provide protection after signature verification.

   o  Does not protect against re-sending (replay of) a message that
      already has a verified signature; therefore, a transit
      intermediary or a recipient can re-post the message -- that is,
      post it as a new message -- with the original signature remaining
      verifiable, even though the new recipient(s) might be different
      from those who were originally specified by the author.

1.2.  Prior Work

   Historically, the IP Address of the system that directly sent the
   message -- that is, the previous email "hop" -- has been treated as
   an identity to use for making assessments.  For example, see
   [RFC4408], [RFC4406], and [RFC4407] for some current uses of the
   sending system's IP Address.  The IP Address is obtained via
   underlying Internet information mechanisms and is therefore trusted
   to be accurate.  Besides having some known security weaknesses, the
   use of addresses presents a number of functional and operational
   problems.  Consequently, there is a widespread desire to use an
   identifier that has better correspondence to organizational
   boundaries.  Domain names can satisfy this need.

   There have been four previous IETF Internet Mail signature standards.
   Their goals have differed from those of DKIM.  PEM and MOSS are only
   of historical interest.

   o  Privacy Enhanced Mail (PEM) was first published in 1987 [RFC0989].

   o  Pretty Good Privacy (PGP) was developed by Phil Zimmermann and
      first released in 1991.  A later version was standardized as
      OpenPGP [RFC1991] [RFC2440] [RFC3156] [RFC4880].

   o  PEM eventually transformed into MIME Object Security Services
      (MOSS) in 1995 [RFC1848].

   o  RSA Security independently developed Secure MIME (S/MIME) to
      transport a Public Key Cryptographic System (PKCS) #7 data object.
      It was standardized as [RFC3851].

   Development of both S/MIME and OpenPGP has continued.  While each has
   achieved a significant user base, neither one has achieved ubiquity
   in deployment or use.

   To the extent that other message-signing services might have been
   adapted to do the job that DKIM is designed to perform, it was felt
   that repurposing any of those would be more problematic than creating
   a separate service.  That said, DKIM only uses cryptographic
   components that have a long history, including use within some of
   those other messaging security services.

   DKIM is differentiated by its reliance on an identifier that is
   specific to DKIM use.

   DKIM also has a distinctive approach for distributing and vouching
   for keys.  It uses a key-centric, public-key management scheme,
   rather than the more typical approaches based on a certificate in the
   styles of Kohnfelder (X.509) [Kohnfelder] or Zimmermann (web of
   trust) [WebofTrust].  For DKIM, the owner of the SDID asserts the
   validity of a key, rather than having the validity of the key
   attested to by a trusted third party, often including other
   assertions, such as a quality assessment of the key's owner.  DKIM
   treats quality assessment as an independent, value-added service,
   beyond the initial work of deploying a signature verification
   service.

   Further, DKIM's key management is provided by adding information
   records to the existing Domain Name System (DNS) [RFC1034], rather
   than requiring deployment of a new query infrastructure.  This
   approach has significant operational advantages.  First, it avoids
   the considerable barrier of creating a new global infrastructure;
   hence, it leverages a global base of administrative experience and
   highly reliable distributed operation.  Second, the technical aspect
   of the DNS is already known to be efficient.  Any new service would
   have to undergo a period of gradual maturation, with potentially
   problematic early-stage behaviors.  By (re-)using the DNS, DKIM
   avoids these growing pains.

1.3.  Internet Mail Background

   The basic Internet email service has evolved extensively over its
   several decades of continuous operation.  Its modern architecture
   comprises a number of specialized components.  A discussion about
   Mail User Agents (MUAs), Mail Handling Services (MHSs), Mail Transfer
   Agents (MTAs), Mail Submission Agents (MSAs), Mail Delivery Agents
   (MDAs), Mail Service Providers (MSPs), Administrative Management
   Domains (ADMDs), Mediators, and their relationships can be found in
   Appendix A.

2.  The DKIM Value Proposition

   The nature and origins of a message often are falsely stated.  Such
   misrepresentations may be employed for legitimate or nefarious
   reasons.  DKIM provides a foundation for distinguishing legitimate
   mail, and thus a means of associating a verifiable identifier with a
   message.  Given the presence of that identifier, a receiver can make
   decisions about further handling of the message, based upon
   assessments of the identity that is associated with the identifier.

   Receivers who successfully verify a signature can use information
   about the signer as part of a program to limit spam, spoofing,
   phishing, or other undesirable behaviors.  DKIM does not, itself,
   prescribe any specific actions by the recipient; rather, it is an
   enabling technology for services that do.

   These services will typically:

   1.  Determine a verified identity as taking responsibility for the
       message, if possible.

   2.  Evaluate the trustworthiness of this/these identities.

   The role of DKIM is to perform the first of these; DKIM is an enabler
   for the second.

2.1.  Identity Verification

   Consider an attack made against an organization or against customers
   of an organization.  The name of the organization is linked to
   particular Internet domain names (identifiers).  Attackers can
   leverage using either a legitimate domain name, one without
   authorization, or a "cousin" name that is similar to one that is
   legitimate, but is not controlled by the target organization.  An
   assessment service that uses DKIM can differentiate between a domain
   (SDID) used by a known organization and a domain used by others.  As
   such, DKIM performs the positive step of identifying messages
   associated with verifiable identities, rather than the negative step
   of identifying messages with problematic use of identities.  Whether
   a verified identity belongs to a Good Actor or a Bad Actor is a
   question for later stages of assessment.

2.2.  Enabling Trust Assessments

   Email receiving services are faced with a basic decision: whether to
   accept and deliver a newly arrived message to the indicated
   recipient?  That is, does the receiving service trust that the
   message is sufficiently "safe" to be viewed?  For the modern
   Internet, most receiving services have an elaborate engine that
   formulates this quality assessment.  These engines take a variety of
   information as input to the decision, such as from reputation lists
   and accreditation services.  As the engine processes information, it
   raises or lowers its trust assessment for the message.

   In order to formulate reputation information, an accurate, stable
   identifier is needed.  Otherwise, the information might not pertain
   to the identified organization's own actions.  When using an IP
   Address, accuracy is based on the belief that the underlying Internet
   infrastructure supplies an accurate address.  When using domain-based
   reputation data, some other form of verification is needed, since it
   is not supplied independently by the infrastructure.

   DKIM satisfies this requirement by declaring a valid "responsible"
   identity -- referenced through the SDID -- about which the engine can
   make quality assessments and by using a digital signature to ensure
   that use of the identifier is authorized.  However, by itself, a
   valid DKIM signature neither lowers nor raises the level of trust
   associated with the message, but it enables other mechanisms to be
   used for doing so.

   An organization might build upon its use of DKIM by publishing
   information about its Signing Practices (SP).  This could permit
   detecting some messages that purport to be associated with a domain,
   but which are not.  As such, an SP can cause the trust assessment to
   be reduced, or leave it unchanged.

2.3.  Establishing Message Validity

   Though man-in-the-middle attacks are historically rare in email, it
   is nevertheless theoretically possible for a message to be modified
   during transit.  An interesting side effect of the cryptographic
   method used by DKIM is that it is possible to be certain that a
   signed message (or, if l= is used, the signed portion of a message)
   has not been modified between the time of signing and the time of
   verifying.  If it has been changed in any way, then the message will
   not be verified successfully with DKIM.

   As described above, this validity neither lowers nor raises the level
   of trust associated with the message.  If it was an untrustworthy
   message when initially sent, the verifier can be certain that the
   message will be equally untrustworthy upon receipt and successful
   verification.

3.  DKIM Goals

   DKIM adds an end-to-end authentication capability to the existing
   email transfer infrastructure.  That is, there can be multiple email
   relaying hops between signing and verifying.  Hence, it defines a
   mechanism that only needs to be supported by the signer and the
   verifier, rather than any of the functional components along the
   handling path.  This motivates functional goals about the
   authentication itself and operational goals about its integration
   with the rest of the Internet email service.

3.1.  Functional Goals

3.1.1.  Use Domain-Level Granularity for Assurance

   DKIM provides accountability at the coarse granularity of an
   organization or, perhaps, a department.  An existing construct that
   enables this granularity is the Domain Name [RFC1034].  DKIM binds a
   signing key record to a Domain Name as the SDID.  Further benefits of
   using domain names include simplifying key management, enabling
   signing by the infrastructure as opposed to the MUA, and reducing
   privacy concerns.

   Contrast this with OpenPGP and S/MIME, which associate verification
   with individual authors, using their full email addresses.

3.1.2.  Implementation Locality

   Any party, anywhere along the transit path, can implement DKIM
   signing.  Its use is not confined to particular systems, such as the
   author's MUA or the inbound boundary MTA, and there can be more than
   one signature per message.

3.1.3.  Allow Delegation of Signing to Independent Parties

   Different parties have different roles in the process of email
   exchange.  Some are easily visible to end users and others are
   primarily visible to operators of the service.  DKIM was designed to
   support signing by any of these different parties and to permit them
   to sign with any domain name that they deem appropriate (and for
   which they hold authorized signing keys).  As an example, an
   organization that creates email content often delegates portions of
   its processing or transmission to an outsourced group.  DKIM supports
   this mode of activity, in a manner that is not normally visible to
   end users.  Similarly, a reputation provider can delegate a signing
   key for a domain under the control of the provider, to be used by an
   organization for which the provider is prepared to vouch.

3.1.4.  Distinguish the Core Authentication Mechanism from Its
        Derivative Uses

   An authenticated identity can be subject to a variety of assessment
   policies, either ad hoc or standardized.  DKIM separates basic
   authentication from assessment.  The only semantics inherent to a
   DKIM signature are that the signer is asserting some kind of
   responsibility for the message.  Any interpretation of this kind of
   responsibility is the job of services building on DKIM, but the
   details are beyond the scope of that core.  One such mechanism might
   assert a relationship between the SDID and the author, as specified
   in the rfc5322.From: header field's domain identity.  Another might
   specify how to treat an unsigned message with that rfc5322.From:
   field domain.

3.1.5.  Retain Ability to Have Anonymous Email

   The ability to send a message that does not identify its author is
   considered to be a valuable quality of the current email service that
   needs to be retained.  DKIM is compatible with this goal since it
   permits authentication of the email system operator, rather than the
   content author.  If it is possible to obtain effectively anonymous
   accounts at example.com, knowing that a message definitely came from
   example.com does not threaten the anonymity of the user who authored
   it.

3.2.  Operational Goals

3.2.1.  Make Presence of Signature Transparent to Non-Supporting
        Recipients

   In order to facilitate incremental adoption, DKIM is designed to be
   transparent to recipients that do not support it.  A DKIM signature
   does not "get in the way" for such recipients.

   Contrast this with S/MIME and OpenPGP, which modify the message body.
   Hence, their presence is potentially visible to email recipients,
   whose user software needs to process the associated constructs.

3.2.2.  Treat Verification Failure the Same as No Signature Present

   DKIM must also be transparent to existing assessment mechanisms.
   Consequently, a DKIM signature verifier is to treat messages with
   signatures that fail as if they were unsigned.  Hence, the message
   will revert to normal handling, through the receiver's existing
   filtering mechanisms.  Thus, DKIM specifies that an assessing site is
   not to take a message that has a broken signature and treat it any
   differently than if the signature weren't there.

   Contrast this with OpenPGP and S/MIME, which were designed for strong
   cryptographic protection.  This included treating verification
   failure as message failure.

3.2.3.  Permit Incremental Adoption for Incremental Benefit

   DKIM can be used by any two organizations that exchange email and
   implement DKIM; it does not require adoption within the open
   Internet's email infrastructure.  In the usual manner of "network
   effects", the benefits of DKIM increase as its adoption increases.
   Although this mechanism can be used in association with independent
   assessment services, such services are not essential in order to
   obtain initial benefit.  For example, DKIM allows (possibly large)
   pairwise sets of email providers and spam filtering companies to
   distinguish mail that is associated with a known organization, versus
   mail that might deceptively purport to have the affiliation.  This in
   turn allows the development of "whitelist" schemes whereby
   authenticated mail from a known source with good reputation is
   allowed to bypass some anti-abuse filters.

   In effect, the email receiver can use their set of known
   relationships to generate their own reputation data.  This works
   particularly well for traffic between large sending providers and
   large receiving providers.  However, it also works well for any
   operator, public or private, that has mail traffic dominated by
   exchanges among a stable set of organizations.

   Management of email delivery problems currently represents a
   significant pain point for email administrators at every point on the
   mail transit path.  Administrators who have deployed DKIM
   verification have an incentive to encourage senders (who might
   subsequently complain that their email is not being delivered) to use
   DKIM signatures.

3.2.4.  Minimize the Amount of Required Infrastructure

   In order to allow early adopters to gain early benefit, DKIM makes no
   changes to the core Internet Mail service and, instead, can provide a
   useful benefit for any individual pair of signers and verifiers who
   are exchanging mail.  Similarly, DKIM's reliance on the Domain Name
   System greatly reduces the amount of new administrative
   infrastructure that is needed across the open Internet.

3.2.5.  Permit a Wide Range of Deployment Choices

   DKIM can be deployed at a variety of places within an organization's
   email service.  This affords flexibility in terms of who administers
   its use, as well as what traffic carries a DKIM signature.  For
   example, employing DKIM at an outbound boundary MTA will mean that it
   is administered by the organization's central IT department and that
   internal messages are not signed.

4.  DKIM Function

   DKIM has a very constrained set of capabilities, primarily targeting
   email while it is in transit from an author to a set of recipients.
   It associates verifiable information with a message, especially a
   responsible identity.  When a message does not have a valid signature
   associated with the author, a DKIM SP will permit the domain name of
   the author to be used for obtaining information about their signing
   practices.

4.1.  Basic Signing

   With the DKIM signature mechanism, a signer chooses an SDID, performs
   digital signing on the message, and adds the signature information
   using a DKIM header field.  A verifier obtains the domain name and
   the "selector" from the DKIM header field, obtains the public key
   associated with the name, and verifies the signature.

   DKIM permits any domain name to be used as the SDID, and supports
   extensible choices for various algorithms.  As is typical for
   Internet standards, there is a core set of algorithms that all
   implementations are required to support, in order to guarantee basic
   interoperability.

   DKIM permits restricting the use of a signature key to signing
   messages for particular types of services, such as only for a single
   source of email.  This is intended to be helpful when delegating
   signing authority, such as to a particular department or to a third-
   party outsourcing service.

   With DKIM, the signer explicitly lists the headers that are signed,
   such as From:, Date:, and Subject:.  By choosing the minimal set of
   headers needed, the signature is likely to be considerably more
   robust against the handling vagaries of intermediary MTAs.

4.2.  Characteristics of a DKIM Signature

   A DKIM signature applies to the message body and selected header
   fields.  The signer computes a hash of the selected header fields and
   another hash of the body.  The signer then uses a private key to
   cryptographically encode this information, along with other signing
   parameters.  Signature information is placed into DKIM-Signature:, a
   new [RFC5322] message header field.

4.3.  The Selector Construct

   The key for a signature is associated with an SDID.  That domain name
   provides the complete identity used for making assessments about the
   signer.  (The DKIM specification does not give any guidance on how to
   do an assessment.)  However, this name is not sufficient for making a
   DNS query to obtain the key needed to verify the signature.

   A single SDID can have multiple signing keys and/or multiple
   potential signers.  To support this, DKIM identifies a particular
   signature as using a combination of the SDID and an added field,
   called the "selector", specified in a separate DKIM-Signature: header
   field parameter.

   NOTE:   The semantics of the selector (if any) are strictly reserved
      to the signer and is to be treated as an opaque string by all
      other parties.  If verifiers were to employ the selector as part
      of an assessment mechanism, then there would be no remaining
      mechanism for making a transition from an old, or compromised, key
      to a new one.

4.4.  Verification

   After a message has been signed, any agent in the message transit
   path can verify the signature to determine that the owner of the SDID
   took responsibility for the message.  Message recipients can verify
   the signature by querying the DNS for the signer's domain directly,
   to retrieve the appropriate public key, and thereby confirm that the
   message was signed by a party in possession of the private key for
   the SDID.  Typically, verification will be done by an agent in the
   Administrative Management Domain (ADMD) of the message recipient.

4.5.  Sub-Domain Assessment

   Signers often need to support multiple assessments about their
   organization, such as to distinguish one type of message from
   another, or one portion of the organization from another.  To permit
   assessments that are independent, one method is for an organization
   to use different sub-domains as the SDID tag, such as
   "transaction.example.com" versus "newsletter.example.com", or
   "productA.example.com" versus "productB.example.com".  These can be
   entirely separate from the rfc5322.From header field domain.

5.  Service Architecture

   DKIM uses external service components, such as for key retrieval and
   relaying email.  This specification defines an initial set, using DNS
   and SMTP, for basic interoperability.
                                  |
                                  |- RFC5322 Message
                                  V
     +--------+    +--------------------------------+
     | Private|    |  ORIGINATING OR RELAYING ADMD  |
     | Key    +...>|  Sign Message with SDID        |
     | Store  |    +---------------+----------------+
     +--------+                    |
      (paired)                 [Internet]
     +--------+                    |                     +-----------+
     | Public |    +--------------------------------+    | Remote    |
     | Key    |    |  RELAYING OR DELIVERING ADMD   |    | Sender    |
     | Store  |    |  Message Signed?               |    | Practices |
     +----+---+    +-----+--------------------+-----+    +-----+-----+
          .              |yes                 |no              .
          .              V                    |                .
          .        +-------------+            |                .
          +.......>|  Verify     +--------+   |                .
                   |  Signature  |        |   |                .
                   +------+------+        |   |                .
                      pass|           fail|   |                .
                          V               |   |                .
                   +-------------+        |   |                .
                   |             |        |   |                .
          +.......>| Assessments |        |   |                .
          .        |             |        V   V                .
          .        +-----+--+----+      +-------+              .
          .              |  |          / Check   \<............+
          .              |  +-------->/  Signing  \
          .              |           /   Practices \<..........+
          .              |          +-------+-------+          .
          .              |                  |                  .
          .              |                  V                  .
     +----+--------+     |            +-----------+     +------+-----+
     |Reputation/  |     |            | Message   |     | Local Info |
     |Accreditation|     +----------->| Filtering |     | on Sender  |
     |Info         |                  | Engine    |     | Practices  |
     +-------------+                  +-----------+     +------------+

                    Figure 1: DKIM Service Architecture

   As shown in Figure 1, basic message processing is divided between a
   signing Administrative Management Domain (ADMD) and a verifying ADMD.
   At its simplest, this is between the originating ADMD and the
   delivering ADMD, but can involve other ADMDs in the handling path.

   signing:   Signing is performed by an authorized module within the
      signing ADMD and uses private information from the Key Store, as
      discussed below.  Within the originating ADMD, this might be
      performed by the MUA, MSA, or an MTA.

   verifying:   verifying is performed by an authorized module within
      the verifying ADMD.  Within a delivering ADMD, verifying might be
      performed by an MTA, MDA, or MUA.  The module verifies the
      signature or determines whether a particular signature was
      required.  Verifying the signature uses public information from
      the Key Store.  If the signature passes, reputation information is
      used to assess the signer and that information is passed to the
      message filtering system.  If the signature fails or there is no
      signature using the author's domain, information about signing
      practices related to the author can be retrieved remotely and/or
      locally, and that information is passed to the message filtering
      system.

   If a message has more than one valid signature, the order in which
   the signers are assessed and the interactions among the assessments
   are not defined by the DKIM specification.

5.1.  Administration and Maintenance

   A number of tables and services are used to provide external
   information.  Each of these introduces administration and maintenance
   requirements.

   Key Store:   DKIM uses public-/private-key (asymmetric) cryptography.
      The signer users a private key and the verifier uses the
      corresponding public key.  The current DKIM Signing specification
      provides for querying the Domain Names Service (DNS), to permit a
      verifier to obtain the public key.  The signing organization
      therefore needs to have a means of adding a key to the DNS, for
      every selector/SDID combination.  Further, the signing
      organization needs policies for distributing and revising keys.

   Reputation/Accreditation:   If a message contains a valid signature,
      then the verifier can evaluate the associated domain name's
      reputation, in order to determine appropriate delivery or display
      options for that message.  Quality assessment information, which
      is associated with a domain name, comes in many forms and from
      many sources.  DKIM does not define assessment services.  Its
      relevance to them is to provide a verified domain name, upon which
      assessments can be made.

   Signing Practices (SP):   Separate from determining the validity of a
      signature, and separate from assessing the reputation of the
      organization that is associated with the signed identity, there is
      an opportunity to determine any organizational practices
      concerning a domain name.  Practices can range widely.  They can
      be published by the owner of the domain or they can be maintained
      by the evaluating site.  They can pertain to the use of the domain
      name, such as whether it is used for signing messages, whether all
      mail having that domain name in the author rfc5322.From: header
      field is signed, or even whether the domain owner recommends
      discarding messages in the absence of an appropriate signature.
      The statements of practice are made at the level of a domain name,
      and are distinct from assessments made about particular messages,
      as occur in a Message Filtering Engine.  Such assessments of
      practices can provide useful input for the Message Filtering
      Engine's determination of message handling.  As practices are
      defined, each domain name owner needs to consider what information
      to publish.  The nature and degree of checking practices, if any
      are performed, is optional to the evaluating site and is strictly
      a matter of local policy.

5.2.  Signing

   Signing can be performed by a component of the ADMD that creates the
   message, and/or within any ADMD along the relay path.  The signer
   uses the appropriate private key that is associated with the SDID.

5.3.  Verifying

   Verification can be performed by any functional component along the
   relay and delivery path.  Verifiers retrieve the public key based
   upon the parameters stored in the message.

5.4.  Unverified or Unsigned Mail

   Messages lacking a valid author signature (a signature associated
   with the author of the message as opposed to a signature associated
   with an intermediary) can prompt a query for any published "signing
   practices" information, as an aid in determining whether the author
   information has been used without authorization.

5.5.  Assessing

   Figure 1 shows the verified identity as being used to assess an
   associated reputation, but it could be applied to other tasks, such
   as management tracking of mail.  Local policy guidelines may cause
   signing practices to be checked or the message may be sent directly
   to the message Filtering Engine.

   A popular use of reputation information is as input to a Filtering
   Engine that decides whether to deliver -- and possibly whether to
   specially mark -- a message.  Filtering Engines have become complex
   and sophisticated.  Their details are outside of the scope of DKIM,
   other than the expectation that the verified identity produced by
   DKIM can accumulate its own reputation, and will be added to the
   varied soup of rules used by the engines.  The rules can cover signed
   messages and can deal with unsigned messages from a domain, if the
   domain has published information about its practices.

5.6.  DKIM Processing within an ADMD

   It is expected that the most common venue for a DKIM implementation
   will be within the infrastructures of the authoring organization's
   outbound service and the receiving organization's inbound service,
   such as a department or a boundary MTA.  DKIM can be implemented in
   an author's or recipient's MUA, but this is expected to be less
   typical, since it has higher administration and support costs.

   A Mediator is an MUA that receives a message and can repost a
   modified version of it, such as to a mailing list.  A DKIM signature
   can survive some types of modifications through this process.
   Furthermore, the Mediator can add its own signature.  This can be
   added by the Mediator software itself, or by any outbound component
   in the Mediator's ADMD.

6.  Considerations

6.1.  Security Considerations

   The security considerations of the DKIM protocol are described in the
   DKIM base specification [RFC4871], with [RFC4686] as their basis.

6.2.  Acknowledgements

   Many people contributed to the development of the DomainKeys
   Identified Mail and the effort of the DKIM Working Group is
   gratefully acknowledged.  In particular, we would like to thank Jim
   Fenton for his extensive feedback diligently provided on every
   version of this document.

7.  Informative References

   [Kohnfelder]  Kohnfelder, L., "Towards a Practical Public-key
                 Cryptosystem", May 1978.

   [RFC0989]     Linn, J. and IAB Privacy Task Force, "Privacy
                 enhancement for Internet electronic mail: Part I:
                 Message encipherment and authentication procedures",
                 RFC 989, February 1987.

   [RFC1034]     Mockapetris, P., "Domain names - concepts and
                 facilities", STD 13, RFC 1034, November 1987.

   [RFC1113]     Linn, J., "Privacy enhancement for Internet electronic
                 mail: Part I - message encipherment and authentication
                 procedures", RFC 1113, August 1989.

   [RFC1848]     Crocker, S., Galvin, J., Murphy, S., and N. Freed,
                 "MIME Object Security Services", RFC 1848,
                 October 1995.

   [RFC1991]     Atkins, D., Stallings, W., and P. Zimmermann, "PGP
                 Message Exchange Formats", RFC 1991, August 1996.

   [RFC2440]     Callas, J., Donnerhacke, L., Finney, H., and R. Thayer,
                 "OpenPGP Message Format", RFC 2440, November 1998.

   [RFC3156]     Elkins, M., Del Torto, D., Levien, R., and T. Roessler,
                 "MIME Security with OpenPGP", RFC 3156, August 2001.

   [RFC3851]     Ramsdell, B., "Secure/Multipurpose Internet Mail
                 Extensions (S/MIME) Version 3.1 Message Specification",
                 RFC 3851, July 2004.

   [RFC4406]     Lyon, J. and M. Wong, "Sender ID: Authenticating
                 E-Mail", RFC 4406, April 2006.

   [RFC4407]     Lyon, J., "Purported Responsible Address in E-Mail
                 Messages", RFC 4407, April 2006.

   [RFC4408]     Wong, M. and W. Schlitt, "Sender Policy Framework (SPF)
                 for Authorizing Use of Domains in E-Mail, Version 1",
                 RFC 4408, April 2006.

   [RFC4686]     Fenton, J., "Analysis of Threats Motivating DomainKeys
                 Identified Mail (DKIM)", RFC 4686, September 2006.

   [RFC4871]     Allman, E., Callas, J., Delany, M., Libbey, M., Fenton,
                 J., and M. Thomas, "DomainKeys Identified Mail (DKIM)
                 Signatures", RFC 4871, May 2007.

   [RFC4880]     Callas, J., Donnerhacke, L., Finney, H., Shaw, D., and
                 R. Thayer, "OpenPGP Message Format", RFC 4880,
                 November 2007.

   [RFC5322]     Resnick, P., Ed., "Internet Message Format", RFC 5322,
                 October 2008.

   [WebofTrust]  Network Associates, Inc. and its Affiliated Companies,
                 "How PGP works, in Introduction to Cryptography", 1999,
                 <http://www.pgpi.org/doc/pgpintro/>.

Appendix A.  Internet Mail Background

A.1.  Core Model

   Internet Mail is split between the user world, in the form of Mail
   User Agents (MUA), and the transmission world, in the form of the
   Mail Handling Service (MHS) composed of Mail Transfer Agents (MTA).
   The MHS is responsible for accepting a message from one user, the
   author, and delivering it to one or more other users, the recipients.
   This creates a virtual MUA-to-MUA exchange environment.  The first
   component of the MHS is called the Mail Submission Agent (MSA) and
   the last is called the Mail Delivery Agent (MDA).

   An email Mediator is both an inbound MDA and outbound MSA.  It takes
   delivery of a message, makes changes appropriate to its service, and
   then reposts it for further distribution.  Typically, the new message
   will retain the original rfc5322.From: header field.  A mailing list
   is a common example of a Mediator.

   The modern Internet Mail service is marked by many independent
   operators, many different components for providing users with service
   and many other components for performing message transfer.
   Consequently, it is necessary to distinguish administrative
   boundaries that surround sets of functional components, which are
   subject to coherent operational policies.

   As elaborated on below, every MSA is a candidate for signing using
   DKIM, and every MDA is a candidate for doing DKIM verification.

A.2.  Trust Boundaries

   Operation of Internet Mail services is apportioned to different
   providers (or operators).  Each can be composed of an independent
   ADministrative Management Domain (ADMD).  An ADMD operates with an
   independent set of policies and interacts with other ADMDs according
   to differing types and amounts of trust.  Examples include an end
   user operating a desktop client that connects to an independent email
   service, a department operating a submission agent or a local Relay,
   an organization's IT group that operates enterprise Relays, and an
   ISP operating a public shared email service.

   Each of these can be configured into many combinations of
   administrative and operational relationships, with each ADMD
   potentially having a complex arrangement of functional components.
   Figure 2 depicts the relationships among ADMDs.  Perhaps the most
   salient aspect of an ADMD is the differential trust that determines
   its policies for activities within the ADMD, versus those involving
   interactions with other ADMDs.

   Basic types of ADMDs include:

      Edge:  Independent transfer services, in networks at the edge of
         the Internet Mail service.

      User:  End-user services.  These might be subsumed under an Edge
         service, such as is common for web-based email access.

      Transit:  These are Mail Service Providers (MSP) offering value-
         added capabilities for Edge ADMDs, such as aggregation and
         filtering.

   Note that Transit services are quite different from packet-level
   transit operation.  Whereas end-to-end packet transfers usually go
   through intermediate routers, email exchange across the open Internet
   often is directly between the Edge ADMDs, at the email level.

       +--------+                            +--------+    +--------+
       | ADMD#1 |                            | ADMD#3 |    | ADMD#4 |
       | ------ |                            | ------ |    | ------ |
       |        |   +----------------------->|        |    |        |
       | User   |   |                        |--Edge--+--->|--User  |
       |  |     |   |                   +--->|        |    |        |
       |  V     |   |                   |    +--------+    +--------+
       | Edge---+---+                   |
       |        |   |    +----------+   |
       +--------+   |    |  ADMD#2  |   |
                    |    |  ------  |   |
                    |    |          |   |
                    +--->|-Transit--+---+
                         |          |
                         +----------+

        Figure 2: ADministrative Management Domains (ADMD) Example

   In Figure 2, ADMD numbers 1 and 2 are candidates for doing DKIM
   signing, and ADMD numbers 2, 3, and 4 are candidates for doing DKIM
   verification.

   The distinction between Transit network and Edge network transfer
   services is primarily significant because it highlights the need for
   concern over interaction and protection between independent
   administrations.  The interactions between functional components
   within a single ADMD are subject to the policies of that domain.
   Although any pair of ADMDs can arrange for whatever policies they
   wish, Internet Mail is designed to permit inter-operation without
   prior arrangement.

   Common ADMD examples are:

         Enterprise Service Providers:

            Operators of an organization's internal data and/or mail
            services.

         Internet Service Providers:

            Operators of underlying data communication services that, in
            turn, are used by one or more Relays and Users.  It is not
            necessarily their job to perform email functions, but they
            can, instead, provide an environment in which those
            functions can be performed.

         Mail Service Providers:

            Operators of email services, such as for end users, or
            mailing lists.

Index

   A
      ADMD  6
      Administrative Management Domain  6
      assessment  7

   D
      DKIM-Signature  12-13
      DNS  6, 13-15

   I
      identifier  4-8
      identity  3-7, 9, 12
      infrastructure  5-6, 8-11, 17

   M
      Mail Delivery Agent  6
      Mail Handling Service  6
      Mail Service Provider  6
      Mail Submission Agent  6
      Mail Transfer Agent  6
      Mail User Agent  6
      MDA  6
      MHS  6
      MIME Object Security Services  5
      MOSS  5
      MSA  6
      MSP  6
      MTA  6
      MUA  6

   O
      OpenPGP  5

   P
      PEM  5
      PGP  5
      Pretty Good Privacy  5
      Privacy Enhanced Mail  5

   S
      S/MIME  5

   T
      trust  3, 7-8, 20

   V
      verification  4, 7-8, 10-11, 13, 16, 20-21

   W
      Web of Trust  6

   X
      X.509  6

Authors' Addresses

   Tony Hansen
   AT&T Laboratories
   200 Laurel Ave.
   Middletown, NJ  07748
   USA

   EMail: tony+dkimov@maillennium.att.com


   Dave Crocker
   Brandenburg InternetWorking
   675 Spruce Dr.
   Sunnyvale, CA  94086
   USA

   EMail: dcrocker@bbiw.net


   Phillip Hallam-Baker
   Default Deny Security, Inc.

   EMail: phillip@hallambaker.com